592 Parameter Estimation in Siructurai and Reduced Form Models

mators for Functions of Parameters and Structural Coefficients of Econometric

Models,” Journal of the American Statistical Association, 74:365, March, (85-
Chapter I
193,
Zellner, Arnold and Henri Theil [1962]. hree-Stage Least Squares: Simultancous
Estimation of Simultaneous Equations, ‘onometrice, Wz), January, 54-78.

FURTHER READINGS Whether and How Much

Addison, John T. and W,Stanley Siebert [1979]. The Marketfor Labor: An Analytical
Treatment, Santa Monica, Calif: Goodyear. See especially Chapter 12, “Wage
Inflation,”pp, 414-464.
Women Work for Pay:
Barro, Robert J. [1987], Macroeconomics, Second Edition, NewYork: John Wiley and
Sons. See especially Chapter 16. “The Interplay between Nominal and Real Vari- Applications of Limited
ait Is the Evidence?,” pp. 455-461.
de Marchi, Neil and ChristopherL. Gilbert, eds. [1989]. “History and Methodology Dependent Variable
of Econometrics,” Special Issue of the Oxfard Economic Papers, 41:1, January. A
lively discussion on the historical development of econometrics with special
emphasis on simultaneous equations procedures.
Procedures
Desai, Meghnad (1976], applied Econometrics, New York: McGraw-Hill, See espe-
cially Chapter 7, “Dynamic Simultaneous Equation Models: Wages and Prices.
pp. 205-232, and Chapter 8. “Macroeconomic Models: Simulation and Policy most of the available evidence suggests that female labor
Applications,” pp. 233-268.
Ehrenberg, Ronald G. and Robert S. Smith [1985]. Modern Labor Ecomomies, Second supply, measured either as laborforce participation or as hours of
Edition, Glenview, INL: Scott, Foresman. Sce especially Chapter16, “Inflation and work, is considerably more wageand propertyincomeelasticthan
Unemployment,” pp. 524-564. matelabor supply.
Feisher, Belton and Thomas Kniesner[1984], Labor Econontics: Theory, Evidence MARK R KILLINGSWORTH (1983. p 432]
and Policy, Third Edition, Englewood Cliffs, NJ.: Prentice-Hall. See especially
Chapter 12, “Unemploymentand Wage Inflation,” pp. 462-527 “Oneunexpectedfinding is that working wives in Canada tendto
Lue: . Robert E., Jr. and Thomus J. Sargent [1981}, eds., Rational Expectations und
Econametric Practice, Minneapolis: University of Minnesota Press. A collection work fewer hours per year whenpaid moreper hour... Our
of oft-cited papers dealing with rational expectations. resulting uncompensated wageelasticities of hours of work are
McCallum, Bennett T, [1989], Monetary Economics: Theoryand Policy, New York: shownto be very similar to those reported by other researchers for
Macmillan. See especially Chapter 9, “Inflation and Unemployment: Alternative men.”
Theories,” pp. 174-200, ALICE NAKAMURA, MASAO NAKAMURA, and DALLAS CULLEN (1979. p. 787}
Sargent, Thomas J. [1987], Macroeconomic Theory, Second Editivn, Bostun: Aca-
demic Press, See especially Chapter 16, “The Phillips Curve.” pp. 438-446.
Tauchen, George, ed. [1990}, Special Issue on “Sabving Nontinear Rational Expecta- “High taxes, sometimes bydiminishing the consumptionaf the
tions Models,” Journal of Business and EconomicStatistics, 8:1, January. 1-52. taxed commodities, and sometimes by encouraging smuggling.
Acollection of eleven articles introducing and comparing alternative computa- frequently afford a smaller revenueto government than what might
tional algorithmsfor solving nonlinear rational expectations models. be drawnfrom more moderate taxes.”
Wallis, Kenneth F. [1973], Tupics itt Applied Econometrics, London: Gi ADAM SMITH (4776, BookV. p. 835)
lishing, Ltd. See especially Chapter 4, “Simultaneous Equation §
127.
_ reasonableestimates ofan aggregatelabor supply elasticity and
ofan overall marginal labor tax rateare both low enough to suggest
that broad-based cuts in labor tax rates would not increase
[government] reventa
DION FULLERTON(4982. p. 20)
593
$94 Whether and How Much Women Wark for Pay
Manycontemporary public policy debates involve issues concerning the
supply oflabor. For example, controversies surrounding welfare payments,
socialsecurity, and the income tax systemoften focus on incentives for work,
Even macroeconomic analyses of unemployment and wage rigidity often
taise microeconomic issues concerning choices between leisure and labor
supply.
Thenotion oflabor supply, however, encompasses several dimensions:
the size and demographic composition of the population, the proportion of
the working-age population that is working for pay or seeking such work
(called the laborforce parti ion rate, LFPR), the numberof hours worked
per week or per year, and the quality of labor. In this chapter we focus on
factors that affect the decisions of household members, particularly women,
on whetherto work for pay and,if so, for how many hours. The econometric
tools that we employincludevarious limited and censored dependent variable
procedures.
In the United States since World WarI] the agggregate LFPR was fairly
stable to 1968 and then trended upward. In (948, for example, the aggregate
LFPR was 58.8%, in 1968 it was 59.6%, but in 1988 it stood at 65.9%. This
trending pattern in the aggregate LFPR, however, masks two dramatic and
almostoffsetting drifts, those of males and those offemales, For example, in
1948, 1968, and 1988 the LFPR rate for females increased from 32.7%to
41.6% to 56.1%, while that for males decreased from 86.6% to 80.1% to
76.2%."
Labormarket analysts have offered numerous exptanations for growth
in the female LFPR.’ Although increases in male real wages and male real
incomes since World War II have provided opportunities for married women
10 reduce their commitment to the labor force (an income effectJ, this nega-
tive effect has apparently been more than offset by improvements in the real
wages offered women (both incomeandsubstitution effects) and by increased
availability of daycare services for children, as well as by the enhanced
employmentpossibilities that are available to women. Moreover, the greater
educational attainment of females not only has resulted in improved real
‘wages (see Chapter 5), but may also have changed women's preferences
toward obtaining meaningful work outside the home. These factors, plus
widespread growth in the availability ofbirth control, have undoubtedly con-
\ibuted to the substantial decline in birth rates since World War ll, a phe-
nomenon that has greatly facilitated increased femate labor force participa-
tion. Finally, rising productivity in the household due to labor-saving
technological progress (such as freezers, microwave ovens, self-defrosting
refrigerators, and automatic dishwashers)hasalso contributed to encouraging
women to engage in part- and full-time employment in the labor market,
Theexplanations underlying the gradual decline in the LFPR of males
are somewhatdifferent from those explaining the increased female LFPR and
are highlighted when onedisaggregates the male population by age. For male
teenagers there has been little if any change in the LFPR from World War Il
to the present, and for prime-aged males (those in the 25~54 age group) the
Whether and How Much Women Work for Pay $9
aggregate LFPR has beenrelatively stable, decliningslightly from araund 96%
or 97% in 1949 to about 94% in 1984,
Thedecline in the overall LFPR of males since World War I is due
primarily to decreases in the LFPR ofthose in the 55-64 and 65 and over age
groups. While the sharp drop in the LFPR ofmales who are 65 years ofage
and over from 47% in 1949 to 16.1% in 1984 might not be surprising, given
growth in earned benefits available from private pension and U.S.social secu-
city programs,there has also been a substantial decrease in the LFPR of males
in thetraditional preretirement years ofages 55-64; in 1949, for example, this
LFPR was 87.5%, while by 1984 i1 had fallen 18.9 percentage points to
68.6%?
Econometric research on labor supply has 2 tong history. Much ofthis
cesearch has been policy oriented, attempting to measure the eflects on labor
supply of changesin transfer programs ortheeffecis on tax revenues and labor
supply of changes in incometax rates. Although the ordinaryJeast squares
estimatorwasused in much early econometric research, several factors caused
econometricians to worry about biases resulting from the use of OLS. For
example, the domain of the dependentvariable in many regressions—hours
worked—is censored; observed hours worked are never negative, and in many
samples the numberof hours worked is zero for a large numberofindividuals.
Further, labor supply data are often truncated in that data on wages are typ-
ically unobserved for individuals who are not working.’ Finally, for those
working individuals for whom the wageis observed, issues emerge concerning
the simultaneity of wages and hours worked.
To deal with limited dependent variable, sample selectivity, censored
and truncated data, and otherstatisticalissues that arise in the empirical anat-
ysis of labor supply, within thelast two decades, econometricians have devel-
oped and implemented a hostofsophisticated econometric techniques.In this
chapter we examineseveral of the more common limited dependent variable
estimators, focusing on the analysis of labor supply, particularly the labor
force decisions of women.
Webegin in Section 11.1 with definitions of variables and 2 briefdis-
cussion of common data sources. Then in Section 11.2 we review the ect
nomic theory oflabor force participation and labor supply. Here we examine
individual and household labor supply decision making, summarize the
effects on labor supply of changes in aggregate demand, briefly discuss the
allocation of time over thelife cycle, and note implications for labor supply
ofthe fixed costs of employment.
In Section 11.3 we consider econometric issues and representative
empirical findings, using the paradigm of Mark R. Killingsworth [1983] that
distinguishes first- and second-generation studies of labor supply. We then
assess the sensitivity of empirical results to alternative econometric proce-
duresthatrely on differingstatistical and economic assumptions; this discus-
sion is drawn in large part from Thomas A, Mroz [1987]. We conclude this
section with an overview ofissues in dynamic labor supply.
Finally, in the hands on portion ofthis chapter we enable you to obtain
596 Whether and How Much Women Work for Pay
first-hand experience with implementing empirically 4 variety of estimation
techniques,including the OLS, instrumental variable, logit, probit, Tobit, and
generalized Tobit (Heckit) procedures. This is accomplished by engaging you
in a numberofexercises, each of which uses a portion of Mroz'soriginal 1976
Panel Swdy of Income Dynamicsdata set.
DEFINITIONS AND COMMONDATA SOURCES
The supply oflaboris typically defined as the amount of effort offered by a
population ofa givensize. In turn, this amount is conveniently decomposed
into four factors: (1) the percentage of the population engaged in or seeking
gainful employment, usually called the labor force participation rate; (2) the
number of hours people are willing to work per day, per week, or per year
while they are in the labor force;(3) the amountof effort that people put forth
per hour or day while at work; and (4) the level of training and skills that
workers bring to their jobs. In this chapter we focus attention primarily on
components 1 and 2 of labor supply and emphasize the lahor supply of
women.’ Not much is known yet concerning component 3;° some aspects of
component4 are discussed in Chapter 5 of this book.
In practice, researchers have used a wide variety of measures of labor
supply; a survey by Glen Cain and Harold Watts {1973a, Table 9.3], for
example,identified [8 distinct measures of labor supply. li is therefore imper-
ative that we devote some attention to measurementissues.’
Statistics on labor force participation rates (LFPR) and numberof hours
supplied are gathered by a number ofprivate and governmental organiza-
tions. In terms of LFPR in the United States, official data on mortthly labor
force, employment, and unemploymentstatistics are typically announced by
the Bureau of LaborStatistics on thefirst Friday of each month and are drawn
from the Current Population Survey.” This survey is based on interviews con-
ducted by an employee of the U.S. Bureau of the Census during the week
following the twelfth day of the previous month—called the reference week.
The sampleconsists of people in about 60,000 households in over 700 sepi
rate geographic areas. Households are chosen to reflect the principal charac
teristics of the population (e.g., age, gender, race, ethnicity, and marital sta-
tus), as taken from the most recent decennial census. The same household is
included in the survey for four consecutive months, is removed from the sur-
vey for eight months, and thenis included again for a final four consecutive
months,
As currently defined in the United States, the /abor forceis limited to
those individuals 16 years of age and older who are notinstitutionalized in
prisons, mentalinstitutions, and the like and who either are employed, are
searchingfor gainful employment, or are unemployed. Any labor supplied by
individuals under age 16 is ignored, presumably because such an amount is
small. Note that students aged 16 or over are counted as being in the labor
11.1 Definitions and Common Data Sources 597
force, and many in fact participate on a part-time or part-year basis. Separate
data are kept on the civilian and total laborforces; the latter adds resident
military personnelto thecivilian labor force. For our purposesin this chapter
the focus of the economic theory and econometric analyses of labor supply
will be confinedto the civilian portion of the laborforce.”
The employed are defined as those individuals who worked one hour or
more for wages orsalary during the reference week or did 15 hours or more
of unpaid work in a family business or farm. Individuals who were absent
from work because ofvacation,illness, inclement weather,or strikes and lock-
outs are also counted as employed but are included in the separate subcate-
gory “with a job but not at work.” The wnemployed are those who are on
layoff from a job, who have nojob but have looked for work during the pre-
ceding four weeks and were available for work during the reference week, or
who are waiting to report to 4 new job within the next 30 days.
On thebasis of questions asked by the interviewer the labor force status
of eachcivilian aged 16 or overis determined in an ati-or-nothing manner. A
person is either in or out ofthe laborforce; bydefinition, one cannot be partly
in the labor force and partly out. Thustheentire civilian noninstitutionalized
population P is decomposed into employed, £, unemployed, L/N, and out of
the labor force. O, where
Pok+ UN+0 CLL)
The laborforce LFis limited to those who are employed or unemployed,
LFoE + UN (1.2)
while the labor force participation rate, in percentages, is measuredas
LFPR = 100 (LF/P) (1.3)
The unemployment rate UR, also measured in percentages, is
UR & 100 (UN/LF) (1.4)
An importantfocusofthis chapteris the LFPR of women. Asis seen in
Table 11.1, this LFPR varies considerably by age and has increased dramat-
ically over timeforall age groups. In 1988, for example, the LFPR for women
aged 25-34 was 72.7%, more than twice the 34.9% of 33 years earlier. Sharp
increases in the LFPR of women aged 20-24 and 35~44 also occurred atter
1955 (especially after 1965), while the LFPR of women aged 45~54 increased
more than 50% from 43.8% in 1955 10 69.0% in 1988.
In terms of hours oflabor supplied, in practice most empirical labor
economists use hours of employment as a measure of hours supplied. How-
ever,this employment measure might notreflect the number of hours an indi-
vidualis willing to supply at a given wagerate. This raises severalissues. Many
employers specify standard weekly hours of work equa) to, say, 35, 37.5, or
40 hours per weck, but to someextent, employces can choose among employ-
598 Whether and How Much Women Work for Pay 11.2 Economic Theory of Labor Force Participation $99

Table 11.4 U.S. Female LaborForce vided by business firms and establishments for all manufacturing firms are
Participation, 1955-4985, collected by the U.S. Census Bureau once every five or six years (Censusof
Women Aged 20-54 Manufactures), while a smalier sample of manufacturing firms is surveyed
Year 20-24 25-34 35-44 45-54 each year (Annual Survey ofManufactures)."' Data from a samipte of all firms
and establishments, including those in nonmanufacturing sectors, are gath-
1955 45.9 34.9 4t6 43.8
ered on a monthly basis and are published in summary form by the Bureau
1960 46.1 36.0 43.4 49.8
of LaborStatistics (the Employment and Earnings Surveys). Although labor
1965 49.9 38.5 46.1 50.9
economists have long advocated that the government match data from its
1970 «$7.7 45.0 Sul 54.4 household surveys with those from establishments, such data have not yet
1975 64.1 54.6 55.8 54.6 been developed, in part because ofissues of confidentiality and budgetary
1980 69.0 65.4 65,8 59.9 considerations.”
1985 1B 9 TB 64.4
A numberof nongovernmentalinstitutions produce data on individual
i983 7: 72.7 75.2 69.0
and household labor supply. Two that are ofparticular importance to empir-
Susaver Reproduced from Duniel $. Hamermesh and Albert ical labor economists are the Center for Human Resource Research at Ohio
Rees, The Ecurumas of Work und fay. Fourth Edition, New
York: Harper & Raw, 1988, Table 1,2, (948 dats are from the State University, which conducts National Longitudina! Surveys (NLS), often
January 1989 issue of US. Hureau of Labor Statistics, Zenploy=
‘mend anut Earnings called the Parnes data, and the Survey Research Center at the University of
Michigan, which provides data from its Survey of Working Conditions. Qual-
ity of Employment Survey, and the Panel Study of Income Dynamics
ers with differing standard workweeks. Employers also vary in terms of pos {PSID).” The NLS andthe PSID have becomeincreasingly useful to empir-
sibilities for part-time and overtime work. Furthermore,firms with the same ical researchers, owing in part to the fact that both generate panet data, that
40-hourstandard workweek often offer different yearly hours of work because is, they interview the same households yearafter year. Moreover, in the PSID,
ofvarying layoff, vacation, and holiday policies. Finally, occupational choice attempts are madeto distinguish desired hours oflabor supply from hours at
has a working-hours dimension. The weekly or yearly hours of work maydif work.
fer if one chooses to becomean elementary school teacher rather than a real- Although labor force participation and hours of labor supply are the
tor, an assembly line worker rather than a funeral director, or a New York principalfocus ofthis chapter, other variables are also typically used in empir-
City lawyer rather than a small-town one. Hence, although there are con- ical studies explaining labor supply. Such variables include the wage rate,
straints, employees canstill exercise a substantial choice over hourg at work. measures of experience, the educational attainment of the individual, and
ft is also useful to distinguish hours at work from hours paid. Since nonlabor (often called property) income. Measurement ofthese variables
World War il in the United States, the wedge between hours at work and introduces additional problems of accuracy andinterpretation. For a discus-
hours paid has increased, owing to the growing importance of paid holidays, sion of measurement issues for the wage and education variables, see Section
paid vacations, and other paid leaves. The U.S. Chamber of Commerce has 5.2.A of Chapter 5 in this book; the measurement of property income and
estimated, for example, that paid time not at work as a percent of total wage other variables in the context oflabor supply is discussed by, amongothers,
and salary costs increased from 5.9% in 1955 to 7.3%, 9.4% and 10.3% in MarkKillingsworth (1983, pp. 87-100}.
1965, 1975, and 1985, respectively." In this chapter, unless otherwise speci- {t is important that difficulties in the measurement of labor supply and
fied, the measure oflabor supply used refers to hours at work, not hours puid. its determinants be borne in mind when working through the economic the-
One important measurementissue concerns the source of the hours at ory and econometric analyses surveyed in this chapter. As we shall sec, mea-
work data. The monthiy Current Population Survey asks the respondent to surementissues turn out to be very important.
list not only his or her hours at work during the reference week, but also the
hours ofworkfor all other adult household members. Inability to recall one’s
own hours the previous week or those of other household members canresult THE ECONOMIC THEORY OF LABOR FORCE
in serious measurement error. Moreover, the self-employed and many sala- PARTICIPATION AND LABOR SUPPLY
ried workers might not accurately recoltect when they were working and when
they were not. Measures of hours at work are thereforelikely to includeerrors, We now briefly review the economic theory underlying labor supply. We
Analternative governmental source of data on hours at work is based begin by examining the individuat’s participation decision and the hours sup-
‘on surveysoffirms and establishments rather than of households. Data pro- plied by the individual, given that participation. This framework is then
600 Whether and How Much Women Work for Pay 11.2 Economic Theory of Labor Force Pat

extended to encompass the household rather than the individual as a decision Goods
unit. Hypotheses concerningthe effects of demandshifts, knownas the added 6c
and discouraged worker hypotheses, are addressed. A more general treatment
ofthe allocation of time, both static and over thelife cycle, is presented, fol-
lowed bya discussion of implicationsoffixed costs of employment, By neces-
sity this review of the economic theory underlying labor supplyis brief; more
thorough anddetailed analyses are found in the labor economics textbooks
listed underthe heading Further Readings at the end of this chapter, and espe-
hy
cially in Mark Killingsworth [1983, Chapters 2, 5 and 6).
Slope = ~ P/F,
41.2.A Individual Labor Supply
‘The neoclassical model of the supply oflaboris in essence an application of
the theory of consumer behavior. The individual is assumed to allocate time
to market work and to nonmarketable activities; the latterare typically called,
erroneously, “leisure.” Utility is maximized by choosing combinations of
goods andleisure hours subject to time, price, and income constraints.’
More specifically, let the individual's preferences be represented by a
twice differentiable utility function / = U(G. L), indicating the utility (UV)
obtained from consuming alternative quantities of goods (G) andleisure (L).
{Note: Hereafter, U refers toutility, not to the unemployed; further, L islei
sure, not labor.) The marginalutilities of G and of L (AfU, = 8U/aG and
MU, = aUsaL) are both assumed to be positive, and the utility function is
concave in G and L, implying that 2U/8G", PU/IL? < 0, and eUs8GAL > HoursofLeisure L—
0. Figure 14.4 indiflerence Curves and the Budget Constraint
Along an individual's indifference curve, alternative combinations of L
and G generate the samelevelof satisfaction. In Fig. 11.1, four indifference Since indifference curves that are farther away from the origin represent
curves are drawn, Jy, /,, 42, and J;, each corresponding with successively successively higher levels of utility, the utility-maximizing individual will
greaterlevels of U. choose the highest possible indifference curve. given his or her budget
‘Two important properties ofthe indifference curve are its slope andits constraint.
shape. Theslope is derived as follows. The total differential oftheutility func- Three factors affect the budget constraint—prices. nonlabor income,
tion UG, £)is and time. First, Ict the unit price of goods be P,, and let the exogenous and
(@U/AG)dG + (@U/AL)dL = dU (15) constant wage rate be P,; thus the individual's real wage is P,/P,. Second, tet
the individual's nonlabor or property income be denoted as ¥: in terms of
Along given indifference curve, dU’ = 0. Substituting dU= 0 into Eq. (11.5) consumption goods the real amount of nonlabor income is V//P,. Third,
and rearrangingyield the slope ofindifference curves in Fig. 11,1, dG/dL, also because only a finite number of hours 7 are available in any time period,
knownasthe negative of the marginalrate of substitution ofleisure for con- leisure (nonmarket) hours £ plus hours devoted to market work H must
sumergoods and denoted here as — MRSz6. exhaust 7. that is, L + H = T. Further, labor income equals the product
PH, and thereal labor income forgone by choosing one more unitof leisure
Oe rs, (116) instead of working equals P,/P..
If it is further assumed thatthe individual spends all of his or her avail-
Since by assumption, A/U, and MU,are both positive, indifference curves able income, the above three factors imply the budget constraint
slope downward. Further, concavity implies that indifference curves are con-
vex to the origin. Hence, although it is possible to substitute £ for G and hold YoPH+V = P(T-L)+¥ = PG (1.7)
utility fixed, the greater the ratio of L to G, the greater the marginal amount where ¥, total money income,is the sum oflabor and nonlabor income. —
of L required to compensate for giving up a marginal amountof G. Equation (11.7) is often rewritten in two ways. To emphasize the notion

602) Whether and How Much Women Work for Pay
that an individual's “full income” F is spent on goods and leisure, Gary
Becker [1965] has added P,L to both sides of Eq. (11.7), obtaining
FoY¥t+PbsPiG+PLorhT+ ¥ (11.8)
In this formulation the full income budget constraint consists ofthe total
amountoftime available, T. evaluated at the constant wagerate P,. plus non-
labor income F. This full incomeF is then totally expended onleisure (PL)
and on goods (PG).
Alternatively, to facilitate graphical analysis, one can rewrite Eq. (11.7)
in termsofreal income,
tee *| ra
G -T+—|- Tek (19)
Py Py Py
When Eg.(11.9) is graphed as in Fig. #1.1, the budget line embodying the
incomeconstraint is line 8’B, with intercept equal to ((P,/P.) - T + ¥/Pe)
and slope cqual to —(P,/P,). Note that even if L = T (when all time is
devoted to feisure and none to market work), the budget line 8°8 docs not
cross the horizontal axis unless nonlabor income¥is zero.
Maximization of utility subject to the budget constraint Eq. (11.9)
involves choosing the set of G and L thatis feasible (on the budget linc) and
on the highest indifference curve that touches the budget line 2’8. In
Fig. 11.1, such a point is at P, where the slope of the indifference curve
4, (—MRS,,) equals the slope of the budget line — P,/P,. At this point the
individual purchases goods OG’. chooses leisure OH’, and supplies hours of
labor #7’T to the market.
More formally, the individual solves his or her maximizatioa problem
by maximizing U = U(G, L) subject to the budget constraint 2,G = P,
(f — L) + ¥. This is done by setting up the Lagrangian function,
W = UG. L)— MPG — PAT -— Ly - 14 (1.10)
takingfirst partial derivatives of W with respect to G and L,setting them equal
to zero, and then solving. This yields
PB
aval _ MU: = MRSiy = R aL
aujaG MU,
Notice that according to Eq. (11.11), at the point at whichutility is maxi-
mized, the AfRS,,; (the negative of which. by Eq. (11.6). equals the slope of
lhe indifference curve) is equal to the real wage rate P,/P,; (the negative of
which, by Eq, (11.9), equals the slope of the budget line 8’B).
The way in which Fig. 11.1 has been drawn allows the individual to
maximizehis or herutility at an interior solution point P where L << T and
#H > 0, that is, where the individual participates in the laborforce with non-
zero H. This need notbe thecase and in fact is critical to understanding the
laborforce participation decision. To see this, assume that the wagerate the
11.2 Economic Theoryof Labor Foree Participation 603
individual would earn were he or she to work in the markel was ?;, which is
tower than P,. Further, let the individual have the same aonlabor income 7
and preferencesas before. In such a case the budget constraint facing the indi-
vidual would be PuG = Pi(T — L) + V, which is drawn in Fig. 11.1 as the
Hatter budget line 8” 8. Now the highest indifference curve the individual
could attain, given his or her preferences and the budget constraint 8” B. is
1,. Al point # the indifference curve /, touches the budget line B’#, where L
= Tand H = 0, that is, at the point at which the individualis « nonpartici-
in the laborforce, spendingall of his or hertimein leisure (nonmarket
ies). Anyhigher indifference curveis simply unattainable toindividuals
with such preferences and budget constraints.
Points such as B represent a corner solution to the individual’s utility
maximization problem, rather than an interior solution. Note in particular
that at the corner solution 8, instead of Eq. (11.11) holding, whereMRS,, =
p./P,, the slope of the indifference curveis steeper than that of the budget
line, implying that MRS, > P,/P.. This suggests thatthe labor force partic.
ipation decision can be simply envisaged as corresponding to whether the
individual'sutility maximization problem, given budget constraints, yields a
corner solution or an interior solution. Specifically, if at the solution point,
MRS, P,{P,, then H > Qand L < T—aninterior solution occurs; but
if instead at the solution point, AARS,., > P./P,, then H 2 Qand L = T—
acornersolution obtains.
An important concept underlying the labor force participation decision
is the notion ofthe reservation wage. At point # the slope ofthe indifference
curve /,, ~MRS;,; indicates how much exira earnings the individual would
require to be induced to give up one unit ofleisure, when he or she is not
working at all, This amount of extra earnings is called the reser lion wage.
which we denote as »*.' Note that as drawn in Fig. 11.1 budget line
B’B,thereservation wage ¥* is greater than the market wage P,. that is, the
extra satisfaction from an hourof leisureis greater than the wage rate. How-
ever, if the wage rate rose so that the budgetfine rotated upward from B°B
to BB, then at some point the wage rate would exceed the reservation wage,
resulting in a positive labor force supply. Hence the condition for positive
labor force participation is simply that P, > w*.
Several important implications of this economic theory of labor force
participation are worth noting. First, for individuals with identical reservation
wages those with higher (potential) wage rates are more likely to be in the
laborfarce. Second, for individuals with identical potential wage rates those
with lower reservation wages are more likely to be the labor force. Since
“workaholics” have lower reservation wages than, say, avid hobbyists, given
identical potential wage rates, workaholics are morclikely to be in the labor
force. Similarly, womenwith very young children at home and with onlyvery
expensive daycare possibilities are likely to have higher reservation wages than
single, career-oriented women with no children; other things being equal, we
would expect the fatter to have higher laborforce participation rates. Note
604° Whether and How Much Women Work for Pay 11.2 Economic Theoryof Labor Force Participation 605

that such differences in preferences among individuals are reflected in the

shape andslope oftheir indiflerence curves. Moreover,for a given individual
the shape ofthe indifference curve may change at variouspoints during his
orherlife cycle.
Thetheoretical framework presented above can be generalized lo cases
in which only several options for hours at work are available to employees,
for example, one could only work zero, 20, or 40 hours per week. In such
situations the optimal cornersolution is the largest amountof hours of work
for which the market wageis greater than or equalto the reservation wage.
We now consider the effects on labor supply of changes in nonlabor
incomeandin the wage rate. Suppose that the nonlabor incomeofthe indi-
vidual increased from TB to TC in Fig. 11.1 but that the wage rate remained.
constantat P,. This results in a new budgetline C’C parallelto the old budget
line B’B. Relative to the initial equilibrium at P, the increase in nonlabor
income now permits the utility-maximizing individual to move to a higher
indifference curve, 7,, tangent to the new budget line C’C at point P’, where
the hours ofleisure consumed increases to OH”, the amountoflabor supplied
decreases 10 H” 7, and the amount of goods consumed increases to OG". This
outward movement from P to P’reflects a pure income effect in which Z and
G both increase in response to the increase in nonlabor income. Note, how-
ever, that the pure incomeeffect on hours oflabor supplied is negative, reflect-
ing the implicit assumption here thatleisure is a normal good.
Matters are not quite as simple when the individual experiences a HoursofLeisure £ ~ —~
change in the wage rate, since the response reflects both income and substi-
ttion effects. To see this, consider Fig. 11.2. Let the initial equilibrium be at Figure 44.2 Income andSubstitution Effects in Response to
a Changein the Wage Rate
point P, where the /y indifference curve is tangent 10 the origina! budgetfine
B’B, sesulting in OG units of goods and O# hours ofleisure cortsumed and
HT hours oflabor supplied. NowIct the wage rate increase while goodsprices, Asis shownin Fig. | !.2, the movementfrom P to P” is the compensated
preferences, and nonlabor income remain unchanged. The increase in the substitution effect: it results in reduced leisure OH” and increased labor sup-
wage rate results in an upwardly rotated new budget line 8” 8, tangent to the ply "7, reflecting the fact that leisure has become more expensive as the
higher indifference curve /, at point P’. At this new equilibrium the ntility- wage rate has increased. Once this compensated substitution effect has been
maxitaizing individual increases goods consumed from OG to OG’, decreases isolated, the pure incomeeffect(with no prices changing)is the residual move-
the amountof leisure chosen from OH to OH", and increases hours oflabor ment from P" to P’, with leisure increasing from OH” to OH" andlabor sup-
supplied to the market from HT to H‘T. plied to the market falling from H’T to H’T. Incidentally, although the
It is useful to decompose the movement from P to /” into pure income nomenclature can be confusing,it is worth noting that in muchofthelitera-
and compensated substitution effects. The compensated substitution effect is ture the total movement from P to P’ in response to an increase in the wage
defined as the response ofthe individual to a change in the wagerate holding tate is called the uncompensated or gross substitution effect. Hence the
utilityfixed. Hence the compensated substitution effect involves a movement uncompensated or gross substitution effect is the sum of the compensated
along the origina! indifference curve Jy. To show this graphically, let us substitution effect and the pure income effect.
notionally decrease the individual's nonlabor income by imposing a tax onit An important implication of the above graphical analysis is that in
such that the new notional budget line D’D has the same slope as the new response to an increase in the wage rate, the utility-maximizing individual
budget line 8” B (reflecting the higher wage rate) but is just tangent to the respondsin two different ways: The uncompensated substitution effect results
original indifference curve /, at point P”. Note that in termsof utility this in more labor supply andless leisure, while the pure incomeeffect results in
notional budget line D’D just offsets or compensates for the improvement in less labor supply and more leisure. Figure 11.2 has been drawn in such a way
earning powerthat results from the increased wagerate. that the positive compensated substitution effect on labor supply dominates
606 Whether and How Much Women Work for Pay 11.2 Economic Theoryof Labor Force Participation 607

the negative pure incomeeffect, but this need not be the case. Had we drawn 14.2.8 Household Labor Supply
the indifference curves with different shapes (but still downward sloping and
convex to the origin}, we could have obtained a new equitibrium in which the Thetheoretical framework just presented envisages an individual as making
pure incomeeffect dominated the compensated substitution effect. [t is there- labor supply decisions in isolation. in fact, however, the labor supply decision
fore important to note that even whenleisure is a normal good, on the basis is typically made in the context of decisions taken by other members of the
of economic theory, one cannot determinea priori whether an increase in the household or family within which the individual resides. Kittingsworth (1983,
wage rate, ceteris paribus, results in increased or reduced hours oflabor sup- Chapter 2] distinguishes three approaches that link family membership to
labor supply.
Dlied; this depends on the relative magnitudes of the compensated substitu-
tion and pure income effects, which in turn are empirical, not theoretical In the “male chauvinist” modelit is assumed that, when making her
issues. In termsoftraditionallabor supply curves, therefore, one needs empir- labor supply decisions, the wife views her husband's earnings as a form of
ical evidence to determine whether the labor supply curve is upward sloping Property or nonlabor income, whereas the husband decides on his labor sup-
or backward bending, ply solety on the basis of his own wage and the family's actual nonlabor
Two further remarks are worth noting. First, the above analysis of income,without reference to his wife's labor supply decisions."’ The only dif:
incomeandsubstitution effects has assumed thatinterior solutions occur both ference in this model compared to that considered in Section 11.2.4 is that
before and after the wage rate increase, thatis, it is conditional on positive in the male chauvinist framework the property income V relevant to the
labor force participation. Although it is not demonstrated here, it can be wife's labor supply decision is assumed to include the husband’slabor income
shownthat if the original solution was a corner one with no participation, as well as all nonlabor income,"
then a sufficiently large wageincrease, ceteriy paribus, could result in positive A second approach assumes the existence of a family or household
labor force participation, as could a reduction in nenlabor income, ceteris aggregate utility function U = U(G, £,, L,,..., £,), where £,is the leisure
paribus, Hence participation decisions can also be assessed in terms of income consumed bythe ith individual in the family. This family utility function is
and substitutioneffects. maximized subject to a family budget constraint."
Second, following the traditional theory of consumer demand as, for The family utility approach has becomea familiar one in labor supply
example, in R. G. D. Alien [1938] and Angus Deaton and John M. Muell- analyses, owing in part to the fact that many of the well-known comparative
bauer [1980], one can write the gross, compensated substitution and pure statics results of the individual utility maximization framework carry over
incomeeffects of a wage rate change in termsof the calculus and in elasticity with lite adaptation. An important generalization that is afforded by the
form. In particular, this results in the well-known Slutsky equation, family utility function approach is that there now are four substitutioneffects.
In particular, not only do both family members have own-substitution effects
aH
oa) aH au (the response of the ith family member's labor supply to a change in his or
BPlam” ae
aPleew * of
ay L
(142) her own wagerate), but two cross-substitution effects also occur, involving the
ith family member's labor supply response to a change in the wagerate of the
where Yis money income. Thefirst term on theright-handside of Eq. (1 1.12) jth family memberand vice versa.
is the compensated substitution effect with utitity held constant, and the sec- Because of the assumed existence of an aggregate family utility function,
ond term is the pure incomeeffect. If one multiplies the entire expression in the compensated cross-substitution effect of a change in the ith individual's
Eq. (11.1) by P,/f/ and then multiplies and divides the incomeeffect by Y, ‘wagerate on the jth member's iabor supply must equal the effect of the jth
oneobtains a Slutsky relation in elasticity form. member's wage rate on the ith member's labor supply. Further, although
these compensated cross-substitution effects must be equal in magnitude,
oH P, oH P, Hoon ¥
So aS + ses 11.13; their signs can beeither positive (indicating substitutability) or negative {com-
OP, Hi yon OP, Hlynt Y or a » plementarity). However, because the pure income effects on the two family
which in turn can be rewritten as members need not be identical, the gross (uncompensated) cross-substitution
effects are not necessarily equal. Hence the gross effect of a change in the
in, = Cue + 5° tee (14) wife's wage rate on the husband's labor supply need not equal the grosseffect
of a changein the husband’s wagerate on the wife’s labor supply.
where 5S;is the share of labor in total money income and where the g and ¢ A special case of the family utility function framework, considered in
superscripts on the substitution elasticities refer to gross and compensated detail by Malcolm S. Cohen, Samuel A. Rea, and Robert I. Lerman [1970]
changes, respectively.’* and by Orley Ashenfelter and James J. Heckman (1974), occurs when the
608 Whether and How Much Women Work for Pay
compensated cross-substitution effects equa! zero forall family members. In
this instance a change in the ith member's wage rate imparts only a pure
income effect on the labor supply of the jth individual. This implies that the
labor supply function of the jth family member depends only on his or ber
wage rate and on the sum ofall other family members’ labor incomes plus
family nonlabor income.
Becauseofits analytical simplicity, the family utility function approach
is common in empirical labor supply studies. It is worth noting, however, that
this framework has a number of drawbacks. Because the aggregate utility
function implies that the family derives utility from consumption as a whole,
the distribution ofgoods consumption does not matter; while this might make
sense for family “public goods” such as heat andlight, it is obviously ques-
tionable for “private goods” such as food, clothing, and entertainment. More-
over, the family utility function approachis silent concerning the process that
actually generates an aggregate utility function that is mutually agreeable to
all family members.
Owing in part to such considerations, a variety of alternative frame-
works have been developed thattreat labor supply within a householdsetting,
yet emphasize the individual behavior of family members. Jane Leuthold
[1968], for example, has presented a model in which individuals’ reactions to
other family members’ labor supplies and wage rates are analogous to the
Cournot or Stackelberg reaction curves of duopolists or oligopolists. As in
modern game theory, such a framework requires stringent assumptions to
ensure the existence and stability of a unique equilibrium.”
Bargaining models of family members’ behavi fe another example of
recent developments in “family economics” emph: 2 the individual util-
ities approach. As formulated by, among others, Marilyn E. Manger and Mur-
ray Brown (1979, 1980] and Marjorie B. McElroy and Mary Jean Horney
[1981], family members arrive at decisions about labor supply and consump-
lion spending through a complex process of bargaining. Without stringent
assumptions, such models seldom generate closed form solutions that are
amenable to empirical implementation, As with the aggregate utility function
approach, in these bargaining models, the total effects of changes in, say, the
husband's wage on the wife’s labor supply need not equal the total effects of
changesin the wife’s wage on the husband's labor supply.” However, unlike
y utility function framework,in the bargaining models, both thelevel
n of consumption among household members can affect the
labor supply of each member. Unfortunately, these bargaining models have
not, as yet, generated any discriminating testable restrictions.
Manitalstatus is clearly important in the modeling of houschold labor
supply. In this context, recent work by Gary Becker (1974, 5981, 1988] secks
to endogenizejointly the marital status, consumption, and labor supply deci-
sions of individuals, but at this point, no fully empirically implementable
model has been developed.
11.2 Economic Theory of Labor Furee Pas
41.2.C Effects on Labor Supply of Changesin
Aggregate Demand
‘Therelative importanceof incomeandsubstitution effects hasalso played an
important role in assessing the effects on labor supply ofcyclical changes in
aggregate demand. Here the important issue concerns the net impact on
LFPR ofchanges in aggregate demand,for example, of recessions or booms.
In the 1930s an hypothesis emerged that when the usual breadwinner became
unemployed, additional members of the household enter the labor force in
order to maintain the family’s income,the implication being that labor force
Participation rates would rise along with unemployment. This wascalled the
added-worker hypothesis.” A counter view, known as the discouraged-worker
Aypothesis, holds that when unemploymentincreases, searching for employ-
ment becomesso disheartening that some of the unemployed give up and
withdraw from the labor force, while others who would ordinarily enter the
laborforce choose not to do so.
These two hypotheses have very different implications for the measure-
ment and interpretation of the aggregate unemployment rate. According to
the added-worker hypothesis, the labor force is above the long-term trend dur-
ing recessions. implying that the unemploymentrate is overestimated in the
sense that creating one new job might reduce the number of unemployed by
two people. Onthe other hand, accordingto the discouraged-worker hypoth-
esis, the size of the aggregate laborforce varies in the same direction as aggre-
gate demandrather than in the opposite direction. Since unemploymentis
measured as the number of people who report themselves as secking work.
the discouraged-worker hypothesis suggests that unemployment is underesti-
mated, since one should add to those enumerated as unemployed another
group, the hidden unemployed, who would be searching for employment
were the search notas likely to befruitless. Which of these two hypothesesis
most valid has been viewed as having important poticy implications, since the
theories can differ sharply in implying whether increases in the measured
unemployment rate might warrant expansionary or contractionary mucro-
economic policy responses.
In terms of economic theory the added-worker hypothesis essentially
invoives a houschold incomeeffect. since it implies that the reduction in
aggregate demand lowers the income of those family members whoare in the
labor force (those who lose their jobs or are put on shorter workweeks),
thereby reducing the nonlabor income available to family members who are
not currently in the labor force. Given this reduction in nonlabor income,
reservation wages drop, the demand for leisure (a normal good) falls, and
labor supply increases,
In contrast, the discouraged-worker hypothesis involves a substitution
effect for wo reasons. First, when aggregate demandforlaborfalls, real wage
rates will often also tend to fall, making leisure hours less expensive. Second,
610 Whether and How Much Women Work for Pay 41.2 Economic Theory of Labor Force Participation 611

the probability of successfully finding a job with a given wagefalls during a work the individual's supply oflabor hours is determined jointly with con-
recession.If one defines the expected wage from job search as the product of sumption; it no longer involves just a tabor-leisure tradeoff.
the wage rate of people who have the job and the probability of finding the More formally, a model with such time consumption costs begins with
job, then both components of the expected wage from job search may fall a utility function U = U(Z,, Z;,.... Zn). where the Z, are household activ.
‘during recessionaryperiods, thereby towering theattractiveness ofjob search ities depending on both goods and time inputs.” Households are assumed to
and reducing the opportunity costofleisure hours. combine time and market goods to produce the “basic commodities” Z,, sub-
It is of course possible that both the added-worker (income)effect and Ject to household production functions Z, ZAG,.4).i= 1 2a), where
the discouraged-worker (substitution) effect coexist; the added and discour- the G, and ¢, are goods and time inputs. Beckerspecifies that the household
aged workers might well be different groups of people. Which group tends to production function is of the fixed coefficient type,
dominate is again an empirical issue. Although rescarchers such as Jacob
Z=Gfa and Z = t/b, Lt)
Mincer[1962], Shelly Lundberg, [1985] and Olivia S. Mitchell [1980] have
obtained results suggesting a significant number of added workers, many of where the 4, and 6, are units of goods and units of time per unit ofactivity,
them married women, the added-worker impact emerges only from families respectively, Since the time and money income constraints imply that
whose normal breadwinnerloses a job, and this income effect tends to occur
for only a relatively smalt portion ofthe labor force (unemployment rates in T=D,+H and TPG =P + (LI6)
the United States seldom go above 10%). By contrast, the fall in expected real : 7
wages that occurs in recessionsaffects nearly every household. and so itis not one can rewrite the full income constraint (11.8}—the income that would be
surprising that empirical researchers have generally concluded that the dis- realized if T = H—as
couraged-worker(substitution) effect is considerably larger than the added-
worker effect2> Hence, other things being equal, the labor force tends to
shrink during recessions and grow during periods of economicrecovery.” Dd PaG, + Phe =H PP+ hak (IL ET}

When Eq. (11.15) is substituted in Eq. (11.17), one obtains the

44.2.D The Allocation of Time: A More General
Treatment expression

The theoretical framework that we havediscussed so far assumes that all time a+ PY Zb = xy Z, (11.18)
is allocatedeither to market workorte leisure. This is clearly unrdalistic, since
much nonmarket time is spent in “producing” consumable services at home,
where
using both time and goods. For example, the preparation of meals at home
consumesboth time and food; this timeis neither leisure nor market work. mos Poa, + Pb, is... mt 11.19)
fn an important paper, Gary Becker [1965] called such activities nonmarket
work, We now briefly examinefactors that affect market labor supply when Notethat 2, is the full cost of consuming or producing activity Z,, including
the choiceset ofalternative activities is expanded to also include nonmarket both the goods cosi and forgone-carnings components.
work. If one maximizes the utility function subject to the budget constraint
Becker’sessential insight was that neither time nor consumer goodsare (11.19), one obtains from thefirst-order conditions the equilibrium relation
consumed by themselves, since time is usually not enjoyed without goods, MU,
and virtually all goods require time in order to be consumed. This implies —
MU, igedc.ym
j as
miei :
(11.20)
that consumption decisions of individuals and households are made from a
menu ofalternative activities with varying amounts of both goods and time According to Eq. (11,20), the utility-maximizing individual consumes/pro-
inputs. The total price of any consumption activity is the sum ofits market duces activities and j such that the marginalrate of substitution between any
goods price plus the value of time required to produce or consume it, Other two activitics just equals the ratio oftheir full costs, including timecosts.
things being equal, increase in the wage rate raises the relative price of To see the uscfulness of this more general (ramework, considertheeffect
time-intensive activities, induces substitution against them, and thereby ofa wage increase on an individual’s marketlabor supply. The incomeeffect
mightresult in an increase in hours worked. Note that in this broader frame- will bring about an increase in the consumption ofall normal commodities;
612) Whether and How Much Women Work for Pay
if the time-intensive nonmarket commodities are normal, the income effect
will cause less time to be devoted 10 market work and more to be allocated to
nonmarketactivities. In termsof substitution effects, however, note that the
wage increase will raise the cost of all commodities (assuming that each
requires a nonzero amountoftime) but will raise the relative cost of the more
time-intensive ones. This results in a shift away from more time-intensive
toward less time-intensive nonmarket activities and toward market labor
supply.
Within the household, for example, the increased wage mightresult in
purchases of time-saving devices such as a microwave oven and a computer-
activated water-sprinkling system. Note that although the income and substi-
tution effects on an individual’s labor supply are therefore essentially similar
in sign to those derived in the earlier, simplercase, in this more general frame-
work, insights can be gained into factors aflecting the division of nonmarket
time into more andless time-intensive activities. Further, to the extent that
technological advances result in increasing the labor productivity of home
activities, such advances could be viewed as facilitating increased market
labor supply. Hence, although this more general treatment of time probably
has principal implications for “home economics,” it also has a nontrivial
impact on market labor supply because of interactions among market and
nonmarket activities,
The most important contribution ofthe more general treatment oftime
to understanding family labor supply behavior, however, concerns household
decisions regarding family size and child-rearing. Since child-rearingis clearly
a time-intensive process, the full costs of having children depend in part on
the opportunity costs of the parents’ time. Although beyondthescope ofthis
chapter, it is worth noting that a substantialliterature exists relating fertility
rates to, among other things, wage rates of married women and incomes of
married men; not surprisingly, empirical evidence suggests that other things
being equal, married women with higher earnings potentials appear to have
fewerchildren and participate to a greater extent in the laborforce.» A num-
berofstudies have also been made of howone might valuate the contribution
of the nonmarket working time of married women, including that portion
allocated to child-rearing; on this, see, for example, Reuben Gronau [1973a.
b, c. 1977, 1986].
Muchrecent research in the labor supply literature has involved an even
more general treatment oftime, analyzing the allocation oftime overthelife
cycle.” Significantly, once one considers the longerlife cycle horizon, deci-
sions involving education, training, hours of labor supplied at different ages,
and retirement ail become interdependent and 10 a certain extent endoge-
nous, What may have been exogenous or predetermined in a static context
Row becomes in part endogenous. Someofthese issues are discussed briefly
in Chapter5, particularly in Section §.1. For our present purposes, however,
it is worth noting that not only do married women currently appear to be
having smaller families than in previous decades, but the time duration
(2.2 Economic Theory of Labor Force Pai pation 613
between children has also declined, implying that the time devoted
to child-
rearing has been compressed, while that allocated to market laborsup
ply has
expanded.
14.2.E traplications of Costs of Employment
In general, it is costly to hold a job. Afeneycosts mustbe incurred by workers,
to pay for transportation to work, purchase appropriate work clothing,
buy
meals away from home, and, in manycases, pay for
otherchild care services. Timecosts are also typicallybabysitti ng, daycare, or
incurred by workers,
since most people take timein traveling from home 10 job; further,
children
must often be transported to child care locations before the
Parent can travel
to work. There are also nonpecuniary costs to job holding, such
as dealing
withtraffic hassles or traveling through arcas where personal safety
may be at
isk, A substantial portion of these job costs are fixed in the sense
that they
are incurred regardless of whether one works ten utes or eight hours per
day: a portion ofchild care costs, however, is variable, depending on
the
amountoftime for which child care is required. We now briefly consider
the
effects on labor force participation and labor supply of changes in fixed
and
variable costs of employment.
John F. Cogan (1977, 1980b, 1981] has modeled the effects ofjob costs
on labor force participation and labor supply.** When job-holding costs
are
only lime costs and are fixed, the budget constraint line of Fig. 11.1 becomes
kinked, with a flat, horizontal segment emanating from H = 0 to,
say,
H = h’, indicating that thefixed time costof/’ contributes nothing
to earn-
ings. Asa result, eitherindividuals tend to perform no market work
at all (a
corner solution) or they enter the labor force ata significant numberof hours
per week (a “very interior” solution); hencefixed job time costs imply
that
interior solutions are Jess likely to occur near the corner point where H =
0,
One implication is that because ofthe fixed time costs of holding a job, much
of the upward-sloping portion oflabor supply curves will not be observed
in
data on hours of work.
Coganalso showsthat ifan exogenousincrease in fixed timecosts occurs
(say, an increase in traffic congestion, implying a ionger commuti
ng time),
the individual's reservation wage 1is raised, and owing to the reduced
full
incomethatis available as a result ofthe increased fixed timecost {a down-
ward parallel shift in the budget constraint, total hours available stilt
fixed at
T), hours of labor supplied H will decrease; provided that L is
a normal good,
hours in £ will also decrease, compensating in part for the increased fixed
time costs of employment.
| Costs of employment are in practice very significant and are of prime
importance in determining the labor supply behavior of married women.
For
working women, Cogan [1981] estimates that in the United States in 1967,
average annualfixed costs as a percentage of the average annual carnings
of
working women were 28.3%; furthermore,this figure increases considerably
614 Whether and How Much Women Work for Pay
with the numberof children in the household under age six and with addi-
tional years of education for the mother. Cogan also reports that for an aver-
age woman thediscontinuity in the labor supply function from H = 0 to
H > Ooccurs at about 1300 hours annually (more than half-time work), indi-
cating that when the equilibrium solution changes from a cornerto an interior
one, the result is a very substantial increase in the numberof hours worked.
This concludes our brief overview of the theoretical issues underlying
the lshor force participation and labor supply decisions of individuals and
households. We now move on to a moredetailed discussion of econometric
issues and empiricalresults, focusing particular attention on factors thataffect
female labor supply.
11.3 ECONOMETRIC ISSUES AND REPRESENTATIVE
EMPIRICAL FINDINGS
Thevast proportion of empirical studies of labor supply, beginning with the
pioneering effort of Erika Schoenberg and Paul H, Douglas [1937], has been
based on the neoclassica? analysis of choice, as outlined in the previous sec-
tion. This empirical literature has been surveyed by, among others, Mark R.
Killingsworth [1983] and Killingsworth and James J. Heckman [1986]. For
our purposesit is useful to employ the paradigm developed by Killingsworth,
in which studies are classified as first or second generation depending on the
attention paid to econometric issues and the underlying economic theory.”
11.3.A_ First Generation Studies
. Heckman, Killingsworth, and MaCurdy (1981, Table 1.1, p. 80] note, esti-
First-generation studies of labor supply are characterized by Killingsworth as
relying on ordinary least squares estimation of parameters in equations whose
functional form is chosen either arbitrarily or on the basis of specific ad hoc
considerations; in particular, the functional forms are not derived explicitly
from utility or indirect utility functions. Although dates are not precise, first-
generation research began in the 1930s and continued into the early 1970s; a
well-known first-generation study using aggregate data is by Glen Cain
[1966], while representative studies using microdata are found in the volume
edited by Glen Cain and Harold Watts (1973b].
Someexamples oflinear specifications that are commonly used in first-
generation research include
H=a+ bW+ Vite (11.21a)
! Three issues have received particularly close attention because of the
H,= 4,4 2 bW, + o¥ +6 (11.21)
=
4
A, a+ bW,+e(V+ 30 WH) +6 (11.21¢)
pier
11.3 Econometric Issues 615
where H is hours work time, JVis the real wage rate, V is per period real
property income, ¢ is a stochastic term, and the i, j subscripts refer to differing
houschold members. Equation (11.21) refers to the labor supply of a given
individual, Eq. (11.21b) to the labor supply of a given household member
allowing for nonzero intrahouschold (cross-substitution) effects on /’s labor
supply, and Eq. (11.2ic) constrains these intrahousehold substitution effects
to zero, Intercept termsin these equations are often made a function ofvar-
ious variables such as age, gender, and race,reflecting tastes for work. Further,
in many studies, transformations of the variables in Eqs. (11.21) are used,
including iogarithmsand polynomials. Finally. as Killingsworth notes, ques-
tions concerning the sources underlying the error term « are typicallyignored
in first-generation studies. Most ofthis research proceeds on the assumption
thatthe errorterm is randomlydistributed and makes nodistinction between
the measurement error or omitted regressors that might have generated «. As
weshall see,this last issue receives much more explicit treatment in second-
generation research.
in terms of empirical findings, most first-generation studies conclude
that female labor supply is considerably more sensitive to changes in wage
rates and property incomethan is male labor supply. Most also find thatJei-
sure is a normalgood for both men and women. Another commonfindingis
that an income-compensated increase in one’s own wage increases one’s labor
supply.
Although these qualitative conclusions are roughly consistent across
studies, the quantitative estimates of the labor supplyelasticities obtained
from first-generation studies of labor supply differ very dramatically. As
mates of the compensated own-wage rate laborsupplyelasticities range widely
from ~0.05 to +0.96 for males and from —0.05 to +2.00 for females; gross
own-wage rate labor supply elasticities vary from —0.45 to +0.55 for males
and from ~0.!0 to + 1.60 for females, while labor supply elasticities with
fespect to property income extendfrom 0 to —0.16 for males and from —-0.1
to —0.75 for females. Moreover. in almost ali cases the symmetry of com-
pensated wage effects among household members is rejected byfirst-genera-
tion empirical results, and at times the estimated compensated own-wage
effect on laborsupply has been positive, contrary to theory. These disappoint-
ing results (and some well-timed financial support from research funding
agencies) have fed labor researchers to examine more ctosely the theory
underlying labor supply models,as well as the econometric andstatistical pro-
cedures that have been used to obtain parameterestimates.
unsatisfactory first-gencration results, First, althoughfirst-generation empiri-
cal studies appear to have used interchangably alternative measures of labor
supply H such as labor force participation, hours worked per week, weeks
worked per year, lifetime participation, and the fraction ofa lifetime spent
working, in second-generation research,distinct theoretical models have been
616 Whether and How Much Women Workfor Pay
constructed to consider separately the participation, hours of work per week,
lifetime participation, and lifetime hours of work decisions. These distinct
theoretical models have helped in part to reconcile persistent differences in
estimates of labor supply responses obtained by fitting the same function to
data on temporally different perspectives of labor supply.
Second, empirical labor economists began focusing more attention on
functional form and econometric technique. In turn, this entailed a more
explicit treatment of unobservable variables and introduced theissue of sam-
ple selection bias. Equations such as (11.21) were now viewed more formally
as representing the outcomeofutility maximization and in that context were
found to be woefully inadequate. Specifically, with even the simplest type of
utility maximization model, the labor supply function for a single individual
can be stated as
H=at b+ ch +¢ ifandonlyif Ww > W, (11.228)
H=0 otherwise (11.226)
where H/, is the individual's reservation wage, derived by setting // in Eq.
(11.22a) equalto zero and solving for
First-generation studies ignored this distinction,typically either by fit-
ting equationssuch as Eq. {11.21a) to a random sample taken from the pop-
ulation with the labor supply of nonworkingindividuals set at zero”—here-
after called Procedure I—orby fitting the equation by OLSusing population
subsamples that consist only of working individuals—called Procedure II."
Procedure I (setting H for all nonworkers equal to zero) implicitly assumes
that Egs. (11-21) or Eq. (11.22a) holds for all values of W”, not just gor values
in excess of the reservation wage, and therefore involves a model misspecifi-
cation,yielding inconsistent estimates of the parameters. Use of Procedure I
(limiting the sample to workers) results in a nonrandom selection ofc, since
by Eq.(11.22a) an individual will be included in the estimation subsampleif
and only if H > 0, thatis, ifand only ife > —(a + 6W + cV) —J. Hence
the data on H is censored, and « must necessarily be correlated with HW and
V(unless & = c 0,or if everyone in the population works, as was virtually
thecasein certain studies of the labor supply of prime-aged males).
Since observations under Procedure If are systematically selected into
theestimation subsample according to the criterion « > —J, OLS parameter
estimates based on such subsamples do not provide consistent estimaies of
the total labor supply response parameters. This « > —J condition therefore
draws attention to corner solutions in which labor force participation {a
binary variable} is determined and interior solutions in which hours of work
are then determined, given positive LFP. As Heckman etal. [1981] empha-
size, while generated from the same preference function, the hours of work
equation and the participation equation are fundamentally very different.
Thethird issue that drew the attention of researchers who were unsat-
isfied with first-generation empirical studies wasthat of nonlinear budget con-
11.3 Econometric Issues 617
straints. Once oneintroduces into the model common incometax structures,
varioustypes ofincometransfer programs, fixed costs of work, or employers"
preferences for minimum hours of work, the simplestraight-line budget con-
straints drawn in Figs. 11.1 and 11.2 disappear and give way to complex forms
with kinks, discontinuities, gaps, and nonconvexities. Moreover, when tax
rates depend on thelevel of income,the after-tax wage rate becomes endog-
enous, and simultaneous equations problems can emerge. Mostfirst-genera-
tion work ignored these phenomenaaltogether,in spite ofthe fact that in the
late 1960s and early 1970s a great deal of policy-oriented econometric
rescarch focused on the labor supplyeffectsofalternative social benefit pro-
grams{such as the negative income tax), programsthat clearly implied com-
plex budget constraints.
These shortcomings of first-generation studies suggested numerous
opportunities for improvement. As Heckman, Killingsworth, and MaCurdy
noted,
analysis and technique go together, and first-gencration work, which
glossed over some ofthe structural aspects of labor supply decisions and
used empirical techniques that did not adequately address some ofthe
complexities of that structure, suffered from 2 numberofserious prob-
lems. The results ofsecond-generation work . . . suggest that solving these
problems makes a considerable difference for estimates oflabor supply
parameters, with correspondingly considerable implications for analysis
und policy.”
11.3.8 Second-Generation Studies
We now turn to a dicussion of second-generation studies, Wefirst consider
issues of specification and estimation, then we introducetaxes into the labor
supply model, andfinally, we briefly summarize empirical findings from sec-
ond generation research.
11.3.8.1 SPECIFICATION AND £STIMATION ISSUES A distinguishing feature
of: second-generation modelsoflabor supplyis the explicit treatment ofutility
functions and unobservable variables. As an example, consider the utility
function
Us (WUE + 0) + PL Ur + ep (1.23)
where HW’ and are real wage and real property income,total time 7" is nor-
malized to | so that // is the proportion oftime spent at market work, | ~ df
= is the proportion oftimein leisure (nonmarket aclivities), ¢ is an unob-
servable error term varying from one person to another, and the term in the
first set of square brackets is real goods consumption G, The error term ¢is
interpreted as representing interpersonaldifferences in tastes for leisure and
for goods consumption.” Note that Eq.(11.23) implies that even though two
people have identical W and V, they are permitted to derive differing utility
U from the same amountofobservable H.
618 Whether and How Much Women Work for Pay
The marginal rate of substitution Af implied by the utility function
(11.23), Af (@U/OL)/(8U/AG), can be shown to equal
(0124)
M = [BAL ~ by (WUE + 0) + MAL — + ey)
where 6 & B/(« + ). To computethe reservation wage W’, evaluate Af in
Eq. (11.24)at the point where H = Oand L = 1. Atsucha point,
(11.25)
W, = [BL — by] De + VIAL - e]
Anindividual with a given valueof ¢will provide positive labor supply if and
only if (hereafter abbreviated as iff) 1’ > HW. Setting H’> Wand rearranging
implies that H > Qiffe, > —J, where
gee and J=[(t— 6)— bY/W] (11.26)
This implies that
H>o iffe>-J (27)
H=0 iff < -J (11,28)
Further, if H > 0, then H is determined by the condition that theJ = W
in Eq. (11.24). Therefore if one sets Mf = W’ in Eq. (11.24) and solves for
H, one obtains the empirical labor supply function (given H > 0) as
He(l-b)-byW+a H>o (11.29)
More generally, one can specify a utility function U = U(G. L, e), in
which case the marginal rate of substitution function Af is M(G, L. ¢) =
MOVH + ¥. 1 — #, e)andthe reservation wage function at Hf = Ois HW, =
M(V, I, e). The individual will work iff 17 > HW, and for these individuals,
W’equals M, implying thatthe labor supply function can be derived by solv-
ing W = M(WH + V1 — H, e) for H > 0.
Although the labor supply function (11.29) looks quite similar to first-
generation equations such as Eq.(11.21), in fact the equation system (1 1.27),
(11.28), and (11.29) provides much more information and does so in an inte-
grated and coherent manner. To see this, first note that Eqs. (11.27) and
(11.28) emphasize a crucial threshold condition summarizing the labor force
participation (LFP) decision (for given V and W’, an individual will work iff
& > —4J). Second, the equation system (11.27)-(11.29) highlights the fact
that the labor supply function is in fact two functions—Eq. (11.29), which
holds if W > W, and Eq. (11.28), which holds iff #’ < 3. Third, notice
that the same observable variables, unobservable variables, and parametcrs—
namely, W, V, ¢, and b—affect both the LFP decision and the amount of
labor force supply conditionalon positive LFP. While first-generation models
overlooked these aspects of labor supply, second-generation models empha-
size Them.Let us now briefly examine someofthe principles of estimation
implied by second-generation models of jabor supply.
11.3 Economerric Issues 619
Initially, assume that measures ofthe real wage rate }Vare available for
all individuals in the population, including nonworkers. For the ah individ-
ual, from Eq. (11.26), denote values concerning tastes for work as en, 2 —¢,
andthe value of J given 5, W,, V,, and e, as J,. A common assumptionis that
¢#, has a mean ofzero, has a standard deviation of oy, and is normally dis-
tributed in the population as a whole, implying that the standardized normal
variable ¢z/¢ has a meanofzero and a variance of |.
Giventhis assumption concerning the population distribution of¢,,, the
probability that a given individual/ will workis given by
Pi works) = Pien/on) > (~Jfen)
=f Aodt = 1 ~ F(-Ffou) 11.30)
iow
where fand are the standard normal and cumulative normaldensity func-
tions, respectively. Using Eqs. (11-27) and (11.28), write thelikelihood func-
tion for a sample ofindividuals who are either working or not working as
eT]
“ - Fl-c ~ 8)" + busty -
TL A1-a - 4)+ (131)
<r
where 2 is the set of people who are working, @is the set of people who are
not working, b* = b/oy, and (1 — 6)" 3 (1 — boy.
Equation (11.31) is the standard probit equation, andits parameters &*
and (1 — 5)* can be estimated by maximizingthe samplelikelihood function
é (orits logarithm) with respect to 5* and (1 — b)*.* Given these estimates,
one can then uniquely compute the maximum likelihood estimates of a, as
/fo* + (1 — 5)*] and of b as b*/[5* + (1 — &)*}. Hypothesistests can be
implemented usingthelikelihood ratio test procedure.
An interesting feature of this probit modelis that although it provides
estimates of # and «,, and therefore provides information on parameters gov-
erning labor supply,its computation in Eg.(11.31) uses information only on
V,, W,, and whether the individual works, but does notutilize any informa-
tion on hours worked, H,, Further, the conditional expectation of F[—J/oy]
in Eq. (11.26) simply equals the probability that individual / is not working,
thatis, the probability that H; = 0. A numberofalternative procedures can
be used to incorporate the additional information provided by nonzero data
on H,,
Forthe moment, assumeagain that data on wagesare available for both
workers and nonworkers. The probability that individual { works 4, hours
and that this amount #,is positive can be written as
Pli works H, hours and H,; > 0] =
Pli works H] - P[H, > Oli works H,] = P[i works HJ + 1 = fl)
620) Whether and How Much Women Work for Pay (1.3 Econometric issues 621

where the probability density of observing individual i working exactly H, has been emphasized by Amemiya [1973], the expected
value of H for obser-
hours is fley/o)/on, Hence the likelihood function for the entire sample of vations that are above the threshold (with positive labor
supply}, now called
2 workers and 2’ nonworkers is H*. is simply X¢ plus the expected value ofthe truncated normal
error term,
€ = T] Menfoudton) « TT Ft-sion 1132) EUP) = EWU > 0) = EU[ty > — Xp) = X68 + of(zyF
(z) (1134)
eu ne
This implies that the celationship amongexpected
values ofall observations,
where ey, © H, — J,and J, [(1 — 6) — B(V,/1%,)]. Equation(11.32) is called E(H), the expected value conditional upon being above
the limit, EU"), and
a Tobitlikelihood function.” Nonlinear methods can be employed to find the probability of being above the limit, F(z), is EH) = FIZJEU
P),
estimates of 5 and o, that maximize the samplelikelihood function (1 1.32), John F. McDonald and Roben A. Moffitt [1980] have noted
thatif one
and hypotheses can be tested by using thelikelihood ratio test procedure.” differentiates this last expression with fespect to X,, where
Y, is the kth regres-
Thefirst part of the right-hand side of Eq. (11.32) refers to workers, sor, one obtains
while the secondrefers to nonworkers. Moreover,the first part is identical to
the likelihood function that is implicit in a traditional OLS regression, SEU) yay SEUPY ep y GED)
whereas the secondpart resembles the nonworking probability portion of the ax, A) ae + BUY Se (11.38)
probit likelihood function (11.31). In a sense, therefore, Tobit combines
In the labor supply context the total effect of a change
fegression and probit frameworks. Notealso that if everyone worked, then in XY, on E(H) can
therefore be decomposed into two intuitively appealing
thelast part of Eq. (11.32) would disappear, and estimation of Ea. (I 1.29) vy components: ( 1) the
change in hours worked for individuals who are already
OLS would yield unbiased estimates of 6 and s,. However, if some indi working, weighted by
the probability of working, plus (2) the change in the probabi
uals do not work, Tobit is more appropriate than OLS, since Tobit not only lity of working,
weighted by the expected value of hours for those who work.
predicts the number of hours worked for each worker (see Eq. (11.29)), but
McDonald and Moffitt go on to show thatit is relatively simple
also produces the estimated probability that an individual will not work and to com-
pute these two componentsof the totaleffect, as well as
thereby helps explain why a number of observationsare clustered at the point their relative impor-
lance. Specifically, one can use as an estimate of F(z) the
H = 0 (see Eq. (11.28). fraction of the sam-
ple abovethelimit (thatis, the samplelaborforce particip
The probit and Tobit models that we have discussed so far are based on ation rate), and one
can estimate E(#*) using Eq. (1 1.34), along with entries
a simple form of J derived from the specific utility function(1 1.23), namely, from standard nor-
mal tables evaluating F(z), parameter estimates, and data
J.= [( — 8) — b(V,/5¥)]" In a more generalcase, one might write a labor means. Also, the derivative 8E(H*)/A.X, equals 8, - 4, where
evaluated at sample
supply model A = {l — [xft2y
F(z)) — [(=)'/F(z)']}, a fraction between zero and 1° Finally
, since d#(2z)/
Ha XG t+ uy, if X6 + ty > 0 (11.33) 8X, = f\z)Bi/o, one can estimate the Proportion ofthe
total effect of a change
in X, on labor supply dueto the effect from those alread
H.=0 ifXP + tn SO, P= N.S N A”
y working simply as

where X,is a vector ofregressors, 8 is the corresponding unknowncoefficient The fraction 4therefore not only provides the appropriate adjust
ment
vector, Nis the total numberofindividuals, and ty, is an independently nor- on 4, to obtain consistent estimates ofthe effects of change
s in Y; on H for
mailydistributed “tastes for work” random variable with mean zero and va those already abovethe threshold index, but also indicat
es the proportion of
ance o*, The stochastic index X,8 + uy, is observed only when it is positive the total effect due to induced changes in the behavior
of those above the
and thus results in censored samples. This index can be substituted into the threshold. Thisdistinction can be important; for exampl
e, not only is it often
probit likelihood function (11.31) or the Tobit likelihood function (11.32) in useful for policy purposes to know the effect of a Policyv
ariable such as as a
place of J, and maximum lihood estimates of # and 6 can then be specific income tax rate change on total labor supply,
but i is also pertinent
obtained by using nonlinear optimization procedures. | to be able to distinguish what portion ofthe total effect
fepresents a change
It is tempting but incorrect to interpret the Tobit ¢ coefficients as mea- in LFPR (and hence a change in unemployment rates)
irom that portion of
suring the effect on E(#,) given a change in X for individuals who are work- the effect on people who are already working.
ing, that is, QE(H,)/@X, = 8 for those H, > 0, where EC) refers to the Although the Tobit procedure utilizes information on
both LFP and
expected value operator. In the above mode!(11.33), and ignoring individual hours supplied given positive Participation, it is but one
of several procedures
subscripts i, the E(H) = XBF(z) + af{z}, where z= XBjo, fiz) is the unit that are available for estimating labor supply models. Two
variants on one
normaldensity, and F(z) is the cumulative normal distribution function. As Prominent alternative,called the selection bias-corrected regress
ion method,
622 Whether and How Much Women Work for Pay
use both regression analysis and probit techniques, but as we now see, they
do so in a manner thatis very different from Tobit.
To understandsclection bias more clearly, considera labor supply func-
tion generated from a utility-optimizing framework and having the form H,
= V6 + up, where the normally distributed t¢, have a population mean of
zero and standard deviation of¢,as in Eq. (11.33). (Often, //, or someof the
X, are logarithmic transformations.) Suppose that one estimated this equation
by OLS using data only from working individuals. As was noted in the pre-
vious section, such a procedure results in inconsistent estimates of the param-
eters in Eg. (11.33), since with such sampies, the conditional expectation of
My, is HONZErO and ity, is correlated with X.
The nature of the OLS bias has been considered by, among others,
Arthur S. Goldberger (1981] and William H. Greene [1981]. Under the
assumption that all independent variables and the dependent variable are
multivariate normalitydistributed in the population (this therefore excludes
dummyvariables as regressors), Goldberger obtains the strong result that the
OLS regression coefficients are biased downward in the sense that the OLS
coefficient vector is a scalar multiple of the “true” labor supply coefficient
vector, where thescalarlies within the 0-1 interval. Moreover, Greene shows
that to obtain consistent estimates of the true labor supply parameters in Eq.
(11.33), all one need dois divide each clementof the OLS coefficient vector
by the proportion of observations for which, in our context, // > 0.
Although Goldberger shows analytically that this remarkable result does
not hold when the multivariate normality assumptionis relaxed, Greenefinds
that with a variety of nonnormalsituations (including dummyvariable regres-
softs), this simple adjustment of OLS estimates providess surprisingly robust
approximation to maximum likelihood estimates ofthe labor supply param-
eters in Eq. (11.33). The OLS-based estimates of standard errors, however,
are inconsistent, and as Greene shows, they cannotbe easily adjusted. Hence
one possible wayofdealing with sampleselectivity is to do probit estimation
of the LFP decision based on all observations, then do OLS estimation ofthe
hours worked equation, limiting the sample to workers, and finally, obtain
consistent estimates of the hours worked parameters by multiplying the OLS
coefficient vector by the sample proportion of observations for which H > 0.
Note thattestsof significance cannot be done with this procedure, since OLS-
based estimated standard errors are unreliable.
A second variant on the sampleselectivity bias procedure involves OLS
estimation of an expanded regression equation. Recall that, using Eq. (11.34),
one can write the conditional expectation of hours worked for workers as
E(H*)= E(H|H > 0) = X68 + K (11.36)
where K,, the conditional mean of u, given H > 0,is
Kao: AAR Z)) = 0 (11.37)
}, is oftencalled the inverse of Mill's ratio, andin reliability theory it is known
as the hazard rate.Since KX, is essentially an omitted variable in the linear
11.3 Econometric Issues 623
regression equation H, = X,8 + ty, James Heckman [1976, 1979, 1980] has
suggested adding an estimateof 4, as a regressorto such an equation and then
estimating the expanded regression equation by OLS while limiting the sam-
ple to individuals for which H > 0.
Heckman suggests estimatingA, initially on the basis ofa probit regres-
sion using data from all workers and nonworkers and shows that when this
estimateofA, is appended asa regressorto the equation #/, = X,8 + ity. OLS
estimates of the labor supply parameters are consistent, However, since the
disturbance term from this expanded regression equation turns out to be het-
eroskedastic, OLS estimates of 6 are inefficient. and estimates of standard
errors are biased and inconsistent.
| Yet another alternative procedure for handling this sample selectivity
ue has been suggested by Randall J. Olsen [1980]. Olsen proposes estimat-
ing by OLS 2 linear probability model in which the dependent variable is a
0-1 dummy variable (equaling1 if Hf > 0, otherwise zero) and the regressors
are the sameas jn Eq. (11.33), then retrieving the fitted values from this OLS
Tegression (calling them /), and finally, adding (P — ) as a regressor in the
hours worked equation confined to the sample for which // > 0. Olsen shows
that under the uniform distribution this procedure provides consistent esti-
matesof the labor supply parameters in (11.33)."'
The Goldberger-Greene, Heckman, and Olsen two-step procedure:
in a sense generalizations of the Tobit technique, since the undertyingutility
function and labor supply parameters in thefirst-stage probit estimation are
not constrained to equal numerically the values that sre obtained trom the
OLS oradjusted OLS estimates in the second stage, even if the underlying
economic theory suggests that they should be identical, By contrast, the Tobit
procedure constrains these parameters to be equal in both the labur force par-
ficipation and labor supply equations. Note, however, that if one initially
specifies a more general model with discontinuities in the labor supply sched-
ule ducto, say, fixed costs of work, then the Iwo-stage estimator might well
be preferable.
Unfortunately, the Goldberger-Greene, Heckman, and Olsen two-step
structural procedures require wage data for workers and nonworkers, as does
Tobit. A major problem confronting second-generation researchers, however,
is the fact that typically wage data are available only for workers. This data
problem creates other econometric issues, to which we now turn our
attention.
A numberofalternative procedures have been employed by second-gen-
eration researchers to obtain estimates of labor supply parameters when wages
are unobserved for nonworkers. Many ofthese approaches use an additional
population wage equation of the form
W,= ZT + en, (11,38)
where Z, is a vector ofvariables observed forall workers, including, for exam-
ple, age, education, gender, region, and experience, and «,, is a mean zero
normally distributed random error term representing the effects of unob-
624 Whether and How Much Women Work for Pay
Ty
served factors such as motivation and abilities.” Killingsworth [1983] has dis-
tinguished eight estimation procedures that attempt in alternative ways to
deal with missing wage data for nonworkers.” We now summarize his
detailed discussion.
In Procedure1, wo steps are taken. First, the wage equation (11.38) is
estimated by OLS using data on workers only, and then the resulting OLS
parameters and data on Z, are used to create predicted wages H’, for both
workers and nonworkers, Second, the predicted wages are used as a repressor
in the hours worked equation H, = X,8 + ty, which is estimated by OLS
using data on all individuals and setting hours worked of nonworkers equal
to zero. As was noted carlier, this procedure produces inconsistent parameter
estimates, since it assumes that the hours worked equation applies to all pco-
pie, rather than only to those for whom H > 0.
ProcedureHf is simply a one-step procedure in which the hours worked
equation
H, = X68 + tn (1139)
is estimated by OLS using data only on workers. A variant of Procedure [1
uses predicted rather than actual wages as 3 regressor but still limits the sam-
ple for which Eq. (11.39) is estimated to individuals who work. Recalt that
since people for whom H = 0 are excluded from estimation, Procedure [I
suffers from sample selectivity and yields biased parameter estimates.
Procedure If is the Tobit estimator discussed above bul uses as a mca-
sure of the unobserved wage for nonworkers the predicted wage from Proce-
dure I. Unfortunately, this procedure results in inconsistent parameteresti-
mates, since it is highly likely that sample selectivity problems also bedevil
estimation of the wage equation (11.38), which by necessity is estimated by
using data only on workers.“ Specifically, the disturbance term ey, in Eq.
(11,38) reflecting unobserved factors thataffect wagerates (such as productiv-
ity, abilities, and motivation) is very likely correlated with the disturbance
term #,, tepresenting unobservedfactors that influence labor supply (such as
tastes for work). This implies that aw», the covariance between en, and uy,is
sonzero; alternatively, it implies that the wage variable in the hours worked
equation (11.39) is endogenous, correlated with the u,,, disturbance term. As
a result, neither use of imputed wage duta derived from OLSestimates nor
use of actual wage data (if they were somehow available for al} workers) would
yicid consistent estimates ofthe labor supply model under any of Procedures
1, II, or Ut. To be valid, each of these procedures would require that 1’, be
exogenousor, equivalently, that oy, be zero.
Procedure {V is one of several different procedures discussed by Heck-
man (1974b, 1976a}. It attempts to deal with this endogeneity issue by using
a reduced form procedure. Let the marginal rate of substitution function
derived from a utility function have the linear form
M, = X3& + tn (11.40)
11.3 Econometric issues 625
where the X* matrix is the Y matrix of Eq. (11.39) plus an additional column
vector, hours worked H,, and where the mean zero normallydistributed ¢s,
random error term represents unobservable taste factors. When expression
(11.40) is solved for that wage rate at which Hf = 0, one obtains the reserva-
tion wage equation
W, = XF0+ «, unary
where the X** matrix is the X* matrix of Eq. (11.40) with the W, and #,
vectors deleted.
Heckman begins by adopting an analytically convenient proportionality
hypothesis (consistent with linear labor supply) where hours of work //, are
proportionalto the difference between W, and Hi, whenever HY > 1, and
otherwise are zero:
H, = 6(W, — BY) = DW, — XO — beg, iP, > W, U1.42a)
H,=0 iis W, (1142b)
Theparticipation equation implied by this equation system is
Pli works) = PEW, > WW) = PIZT + en > XO + ea]
= Pl> ~4) (11.43)
where ¢p, 3 eu, — €2, and J, = Z,P ~ XT6. Since eu, and ¢g, are zero mean
normally distributed random variables, so toois tn; the varianceoftp, iS ob
= of, + oh + 2owg. The likelihood function for LFP observations from a
sample consisting of workers 2 and nonworkers 2 can then be written as
e= TP - F-sool TF adeo)l (11.44)
nr
whose estimation involves standard probit procedures.
The hours of work equation that is consistent with this framework is
most interesting. Heckman first estimates the wage determination equation
(11.38) by OLS or GLS, using data for workers only, and then directly sub-
stitutes this into Eqs.(11.424) and (11.42b), resulting in
H, = UZ) - XFe+ y, iT) >0 (1145a)
H,=0 W[-}<a i1.45b)
Hence by substituting out the endogenous wage in Eqs. (11.42) using Eq.
(11.38), Heckman obtains a seduced form equation (11.45), whose right-hand
variables are all exogenous. Three issues merit brief discussion.
First, note that certain variables in the Z matrix from the wage deter-
mination equation (11.38) might be identicalto those in the X** matrix from
the reservation wage equation (11.41), for example, age, education, gender,
race, and years of work experience. This raises issues of identification, a prob-
lem that arises whenever wages are endogenous. Heckman [1974b, 1979]
notes that a condition sufficient foridentification is that at least onc variable
626 Whether and How Much Women Warkfor Pay
included in Z must be excluded from ¥**." Second, the data in the Z and
X** matrices of the reduced form equations (11.45) are typically observed for
both workers and nonworkers, in contrast with traditional labor supply equa-
tions such as Eq, (11.29), for which data on H’ are available only for workers.
Hence it appears possible to utilize data on the entire population, not just
workers. Third, note that one could estimate Eq. (11.454) by nonlinearleast
squares (or by OLS with the parameterestimates from Eq. (11.38) directly
substituted in), using data on the entire population and setting J2 0 for
nonworkers. Killingsworth [1983] calls such a reduced form estimation
method Procedure IV.
A majorproblem with Procedure IV, unfortunately, is that if it is based
on data for all individuals with nonworkers #f set to zero,it involves a mis-
specification similar to Procedure I, since in fact Eq. (11.45a) refers only to
workers. If one instead estimates Eq. (11.45a) only for workers, sample selec-
tivity problems emerge, similar to those in Procedure {1. Hence while it is
apparently promising, Procedure IV does not solve the hours worked-wage
rate endogencity problem.
There is, however, a related two-stage estimation procedure that effec.
tively deals with this problem. In Procedure V, one takes advantageof the fact
thatép, in Eqs. (11.45a) and (11.45b) is normally distributed and that H, can
never be negative. One then specifies the Tobit likelihood function (11.32)
adjusted to accountfor the fact that the disturbance term de,is the product
of an unknown parameterto be estimated and the mean zero normallydis-
tributed random variable ¢p,. This yields
€ = TT RoeatbooVben) - [] Fi—SJon) (11.46)
where be, = H, — bJ,and J, is defined underneath Eq.(11.43).
In the second stage the parameter estimates that are obtained by maxi-
mizing the Tobit samplelikelikood function (11.46) with respect to &. I’, #,
and op, along with data on X** and Z,, are used to calculate the inverse Mills
ratio A, = fl—J/op[1 — A—J/op)], that is, one uses parameters and data
to estimate the probability that an individual with given XY and Z,will work.
This estimateof 4, is then inserted as an additionalvariable in the wage equa-
tion (11.38), whose parameters are estimated by OLS using data on workers
only. Note that with Procedure V, estimation of the panticipation-hours
worked equationsis based on a reduced form Tobit specification, while esti-
tation ofthe wagestructural equation for workers allows for sample selectiv-
ity. However, since the underlying parameters in the reduced form equation
are not constrained to equality with structural parameters of the wage equa-
tion, parameter estimates are notefficient.
In Procedure VI, also discussed in Heckman [1974b], parameters ofthe
entire equation system consisting of Eqs. (11.38) and (11.41) and the Tobit
equations(11.42a) and (11.42b) are estimated by usingfull information max-
imum likelihood (FIML)simultaneous equation techniques, with cross-equa-
11.3 Econometric Issues 627
tion parameter constraints fully imposed. This FIML specification builds on
the assumptionof joint normally distributed random disturbance terms ew, in
the wage equation (11.38) and ¢g, in the reservation equation(11.41), having
covariance gue and implying that the random variable ep, = ey, — ¢4, in the
hours equation (11.42) is also joint normally distributed with ey, having
covariance ope © oly — dwg. AS a result, the likelihood function for the wage
tate, hours of work, and participation decisions of working individuals in set
{and nonworking individuals in set 1” is
@ = [Tf
ren
Aben/bom enfow) «TE
ver
Fl-Jyo) (1.47)
where j is the joint probability density function for ¢,, and ¢,, and other terms
are as defined above.
The important distinction between the FIML (11.47) and Tobit (11.32)
likelihood functionsis therefore that while FIML treats W’ and H us simul-
tancously determined becauseof the possible correlation between ev and ¢,,
Tobit treats H’ us strictly exogenous. Further, unlike the Tobit specification
(11.45), the FIML estimation of Eq. (11.47) utilizesall available information
on Wand H for workers.
A common feature ofall the above estimation proceduresis that they
assume that labor supply falls continuously to zero as wage rates or property
incomedec! ; Se , for example, Eqs. (11.33) and (11.42a). If there are fixed
costs of employment, however, one might want to develop a procedure that
allows for discontinuous labor supply schedules, that is, one that permits the
lowest nonzero number of hours supplied by a worker to be substantially in
excess of zero. Two estimation procedures, hoth known as generalized Tobit
estimators and often called Heckit procedures, allow for such discontinuities.
In Procedure VI/ a three-stage method is implemented. Stage | is the
sameas in Procedure V, except that one now estimates parametersof a probit
{not a Tobit) likelihood function for participation, such as Eq. (11.44). By
using a probit rather than a Tobit specification, continuity ofthe labor supply
scheduleis not imposed. In the second stage of Procedure VII, one uses the
probit parameter estimates to computeprobit estimates of the inverse Mills
ratio X, = A—Jfon)/{1 ~— F(-J/ep)] and then appends A, as an additional
regressor to the wage equation (11,38) to obtain OLS selection bias-corrected
parameterestimates using data on workers only; Killingsworth [1983, p. 159}
notes that the parameter estimate on A, in this second stage should be inter-
preted as an estimate of oy», the covariance between ey, and en.
{n the third and final stage of Procedure VII, one estimates by OLS the
reduced form hours equation (1 {.45a) with the probit inverse Mills ratio vari-
able added, based on data for workers only, which, along with the siructurat
wage parameter estimates from Stage 2, yields consistent estimates of the
labor supply structural parameters. In this case the parameter estimate on A,
is interpreted as an estimate of [(ouo/e) + blewo/op)], where oyp is the
covariance between u,,, and ¢,,. Formulaefor the asymptotic variance-covari-
628 Whether and How Much Women Workfor Pay
JAMES HECKMAN
Labor Econometrician
ore than any
other econo
mist in the last
two decades, James Heck-
man has affected the
econometric modeling of
the labor force participa-
tion and labor supply de-
cisions. Heckman’s con-
tributions began with his
197} Ph.D. thesis at
Princeton, “Three Essays
on Houschold Labor Sup-
ply and the Demand for Market
Goods.” Since then, as detailed in
this chapter, Heckman’sresearch has
essentially built the core of what are
now called the “second-generation
models” of labor supply. Two distin-
guishing features of this research are
that the various estimating equations
are derived explicitly from neoclassi-
cal utility theory and that the sto-
chastic disturbance termsare an in-
tegral part of the model, rather than
being “tacked on™ at the end ofthe
derivation. Heckman's analysis of
self-selection in labor supply has pro-
duced a new methodologyforsolving
self-selection and sample selection
problems, often called the “Heckit”
procedure;it is now widely used in a
vanity ofcontexts. Although they are
not discussed in this chapter, Heck-
man has also developed and imple-
mented empirically a numberof dy-
namicdiscrete choice models, as well
as models for the analysis of du-
ration data. Someofthis
research has involved new
uses of semiparametric
and nonparametric meth-
ods.
Born in Chicago in
1944, Heckman pursued
undergraduate studies at
Colorado College, major-
ing in mathematics, earn-
ing a Phi Beta Kappa key,
and graduating summa
cum laude. He then went
on to study economics at Princeton,
and after spending a summer as Jun-
ior Economist at the Council of Eco-
nomic Advisors in Washington,
D.C., Heckman was awarded the
Ph.D.degree in 1971, studying under
Stanley Black, E. Phillip Howrey,
Harry Kelejian, Riehard Quandt,
and Albert Rees.
Heckman’s first academic
poiniment was at Columbia Univer-
sity. In 1973 he accepted a position
at the University of Chicago, and
in 1985 he was named its Henry
Schultz Professor of Economics. In
1988 he moved to Yale University,
where he nowis the A. Whitney Gris-
wold Professor of Economics. Heck-
man has been elected a Fellow of the
Econometric Society and of the
American Academy of Arts and Sci-
ences. In 1983 he was awarded the
distinguished Jobn Bates Clark
Medal by the American Economics
Association.
11.3 Econumetric Issues 629
ance matrix of estimates obtained from this three-stage procedure have been
derived by Lung-Fei Lee (1982], but consistent estimates of Standard errors
for the third-stage parameters can be obtained in a computationallyless bur-
densome mannerby using the Halbert White (1980} robust estimator,
Finally, in Provedure VIIT thefirst two Stages are identical to those in
Procedure VII, but in the third stage, instead of using the reduced form
spec-
ification (1 1.45a), one employs instrumentat variable and selectivity bias-cor-
rected techniques to estimate parameters of the structural hours equation
(11,33) for workers, More specifically, A, is added as a regressor to thefirst
equation in (11.33), and instrumental variable estimation is undertaken with
the actual wage rates of workers replaced by fitted values derived from the
second-stage selectivity bias-corrected wage equation estimates: hence the
probit A, appears as a regressor in both the wage and hours equations.
{t is worth noting thatif in fact the labor supply schedule is continuous,
parameter estimates from the Heckit multistage Procedures VII and VIII
should be identicalto those from the FIML Procedure Vi. On the other hand,
if the labor supply schedule is discontinuous owing to, say, fixed costs of
employment, then the Heckit multistage generatized Tobit estimators of Pro-
cedures VII and VIII shoulddiffer from FIML. However, unless one specifies
explicitly the nature of the optimization problem facing individuals—includ-
ing, for example, the budget constraint implicationsoffixed casts of employ-
ment, the Heckit parameterestimates are “black box”in the sense that they
arc unable to provide insights on what factors have generated the disconti-
nuities. For further discussion of such importantspecificationissues, see Kil-
lingsworth [1983, pp. 161-168], who considers theoretical and econometric
aspects of several explicit models of discontinuous labor supply.
ap-
14.3.8.2 INCORPORATING TAXES INTO LABOR SUPPLY To this point, our
discussion of labor supply has ignored the presence oftaxes, Since consump-
tion expenditures derive from after-tax income,the relevant budget constraint
sel for the analysis of individual labor supplyis the after-tax constraint, not
the before-tax one. Taxes on wage and Property income can significantly
affect labor supply, in part because taxes ofien introduce a wedge between
average and marginalafter-tax wage rates. Moreover, since much public pol-
icy discussion has focused on the labor supply effects of changing statutory
provisions ofthe incometax code,it is important that we focus attention
on
the effects of taxes on labor supply. Second-generation research on this topic
has been substantial, with noteworthy contributions made by Harvey S.
Rosen [1976], Gary Burtless and Jerry A. Hausman [1978], Hausman [1979;
1980; [981a, b; 1985], Terence J. Wales and Alan D. Woodland [1979],
James J. Heckman and Thomas E. MaCurdy (1981), and Alice Nakamura
and Masao Nakamura (1981a].
Tosee the impactof taxcs, let us assumethattheindividual receivesreal
property income V and carns a gross (before-tax) real wage of Wper hour
worked, As is shown in Fig. 11.3, in the absence oftaxes the budget constraint
630 Whether and How Much Women Work for Pay
Goods, G|
Slope =
WU = 2)
Leisure Proportion —
Figure 11.3. The Effects of Varying Marginal Tax Rates on
the Budget Constraint
is 1F, with the nonvertical portion having a slope of ~ #’. Suppose, how-
ever, that the individualis taxed on the sum of property and wage income.If
this total incomeis C, or less, let the tax rate be 100 - +, percent;if total
incomeis greater than C,, let the marginaltax rate on total income above C,
be 100- 7, percent, where +; > 7,. In Fig. 11.3 this implies thatif the individ-
ual works hours between zero and | — L, and chooses leisure amount
between L, and 1, the marginal after-tax wage rate is W(1 - 7,), while if the
individual works more than | — L, hours (chooses less £ than £,), the mar-
ginal after-tax wage rate is (1 — 7;); thus the kinked after-tax budget con-
straint is 1/,XF;. Note also that in Fig. 11.3 the vertical distance between the
two budget constraints represents the total amount of taxes paid at various
proportions of L. . sures depend on # (the marginal tax rate + varies with A = | — ZL) the
The budget constraint set in Fig. 11.3 is rather simple. It is relatively
straightforward to extend the analysis to a progressive tax structure with m
distinct and successively flatter linear segments, correspondingto increasing
marginal tax rates 7, > T,-) > * + + > 11; in such a case the budget constraint
set would be quasi-concaveto the origin. Note that with such a quasi-concave
budget constraint set, given strictly convex indifference curves, the equilib-
H1.3 Econometric Issues 631
rium point of tangency between the indifference curve and the budgetsetis
unique andoccurseither at oneof the kinksor at an interior location between
kinks.
However, if one introduces other real-world statutory tax provisions
into the analysis such as exemptions, deductibles, and transfer paymentstied
to hours worked, then the budget constraint set may include nonconcave
regions, and as a resuli, multiple tangenciesofindifference curves to the bud-
get set might occur. As Wales and Woodland [£979] and Hausman [198 1a]
emphasize,in such cases, to determine the utility-maximizing hours of work,
one must know the form of theutility function and compare utility at the
various tangencypoints. This introduces complications into the estimation of
the effects oftaxes on labor supply.
One way of simplifying the potentially complex budget constraint set
involves linearization; this has been implemented empirically by, among oth-
ers, Hall [1973] and Wales [1973]. Specifically, for an individual who works
somewhere along segment VX, when confronted with the budget set | /.XF,,
that individual could be assumed to act as if he or she faced a simple
straight-line budget constraint 1 /,XF, with reat property income equalto V,
and a real marginal wage rate of H’(1 — 1,). The effects of small changes in
property incomeor before-tax wagerates could then be analyzed by assuming
that the individual does not shift onto a different segment of the budget con-
straint set. Incidentally, in this casc the after-tax property income I, is the
amountofdisposable income that is available to the individual assuming that
H = Oand L = }. Following W. Erwin Diewert [1971], Burtless and Haus-
man [1978] call this the individual'svirtual income: hence virtual income cor-
responds with the before-tax property income V and the tax rate 7).
Similarly, for an individual who is working somewhere along segment
XF,, when confronted with the same kinked budget constraint set, that indi-
vidual might be envisaged as acting as ifhe or she faced the simple straight-
line budget constraint 1 V,XF; with a virtual income of V; and a real marginal
wagerate of W(t ~ 7;). Hence complex budget constraints can be made con-
siderably less complicated simply by linearizing them at the individual’s equi-
librium point. Note that in econometric implementation, at each observation
the measures ofafter-tax W and virtual income V used must be changed to
reflect the after-tax wage and property income corresponding with the indi-
vidual’s marginal tax rate at the equilibrium point,
Althoughthis linearized budget constraint approachis attractive, it also
introduces new econometric problems, Several issues are of particular impor
tance in empirical implementation, First, because the after-tax JV and ¥ mea-
disturbanceterm in the hours worked equation is by construction correlated
with the after-tax measures of Wand V; hence estimation by OLSprovides
inconsistent estimates of the parameters, and instrumental variable (IV) esti-
mation maybe preferable, 1V estimates have been reported by, amongothers,
Jerry A. Hausman and David A, Wise [1976] and Harvey S. Roscn [1976].
Rosen's studyis of special interest, since in it he examines tax perception.
632 Whether and How Much Women Work for Pay
Rosen findsthat he cannotreject the null hypothesis that individuals perceive
theeffects of taxes andreactto the net after-tax wage rather than to the before-
lax gross wage,
Second, and closely related to the previous point, it might be that in
response to a change in before-tax V or W, the individual would shift to an
entirely different segment on the budget constraint set. Estimation and model
predictions based on predetermined marginaltax rates would therefore be
unreliable. As might be expected, such estimation problemsincrease when the
concave budgetset is kinked rather than continuous, that is, when marginal
tax rate changes are discrete rather than continuous. Problems of estimation
become even more severe if the budget set has nonconcave and kinked
portions.
Such problems with the linearization or local approximation approach
have given rise to a variety ofalternative procedures in which the complete
‘budget set is evaluated.” In one procedure the previous no-tax estimation
framework in which the probability of H > 0 is estimated along with an hours
worked equation conditional on H > O—a single-threshold framework—is
generalized to include multiple marginal tax rates by permitting multiple
thresholds, each corresponding with a different segmentof the budget set in
which the hours worked decision entails a different marginal tax rate. This
type of modelcan be estimated by using the ordered probit function; see Ta-
Antoni
keshi Amemiya [1975, 1981] for a discussion ofthis estimator and
Zabalza [1983] and John C. Ham and Cheng Hsiao (1984) for empirical
implementations.
Yet another procedure assumes that random disturbances arise either
from individuals’ errors in optimization or from measurementerror of and
involves evaluating utility at each point ofthe individual's budget constraint
sel (including the kink points), given values ofthe relevant structural param-
eters of an assumed direct or indirect utility function. This procedure, which
is oftencalled the complete budget constraint approach (CBC).literally solves
the individual's utility maximization problem foreach set of parameter values
andis obviously very computationally intensive. Notice in particular thatfor
is
each set of parameter values, the individual's entire maximization problem
solved: estimation procceds, as has been noted by Wales and Woodland
[1980]. by maximizing the likelihood of observed hours with respect to the
entire set ofstructural parameters ofthe utility or indirectutility function.
The Wales-\WWoodland approach has been generalized by Burtless and
Hausman [1978] and by Hausman [1980] to allow the disturbance term in
the hours worked equationto originate not only from errors in optimization
orerrors in the measurement of H. but also from omitted variables that are
known by the individual but are not observable by the econometrician,
Remarkably, this more general conceptual framework can be implemented
withoutresort to the laborious Woodland-Wales maximizing procedure, pro-
vided that one makes very strong assumptions aboutthe functional form of
the hours worked equation.
11.3 Econometric Issues 633
_The above discussion therefore suggests that while the presence of taxes
considerably complicates the analysis of labor supply, empirical analysis can
proceed, provided that one is willing to make tradeoffs concerning computa-
tional complexity and flexibility of the functional form. This type oftradeoff
is, of course, a familiar quandary to the practicing econometrician.
11.3.8.3 EMPIRICAL FINDINGS We now conclude ourdiscussion of second-
generation researchin labor supply by providing a brief summary ofempiricat
findings. More detailed surveys can be found in, among others, Heckman,
Killingsworth, and MaCurdy[1981, pp. 106-112}. Kiltingsworth [1983, pp.
179-206}; and Killingsworth and Heckman [1986, pp. 185-197].
Mostofthe second-generation studies avoid the pitfalls associated with
first-generation research. In particular, second-generation analyses explicitly
distinguish the LFP function from the hours supptied fisnction, yet they also
note that both derive from a common utility-maximizing framework. Within
this framework a substantial proportion of second-generation studies treat
hours worked and the wage rate as endogenous. Finally, most of the second-
generation studies take account of theselectivity bias in estimating tabor sup-
ply functions, using one or more of the Procedures V through VII summa-
sized above,
_ A vast majority of second-generation work has been concerned either
with fixed costs and discontinuities in the labor supply function (Cogan
[5980b, 1981}, Giora Hanoch [1976, 1980], Heckman [1980], and T. Paul
Schultz (1980]) or with taxes and nonlinear budget constraints (Burtless and
Hausman [1978], Rosen [1976], and Wales and Woodland [1979]). With the
notable exception of Hausman [1980; 198ta, b; 1983] and Alice Nakamura
and Masao Nakamura [1981a}, most studies have focused on either fixed
costs or taxes but have not considered them jointly. As might be expected,
the numerous second-generation studiesdiffer not only in estimation proce-
dures. employed, but also in terms of the underlying data samples, This het-*
erogeneity makes it somewhatdifficult to provide a concise comparison and
summary of research results,
Two criteria might be used in assessing this research. First, do second-
generation results suggest important stytized facts of labor supply that ure not
apparent from first-generation studies? Second, do estimates oflabor supply
responses based on second-generation researchdiffer appreciably from those
basedon OLS estimation of first-generation equations? The answertothefirst
question appears to be “perhaps,” while the answer to the second is an
emphatic “yes.”
In termsof thefirst criterion, like thefirst-generation studies, most (but
notafi, as we shall soon see) second-generationresults generally conclude that
female labor supply is considerably more responsive to changes in propery
income and wagerates than is malelabor supply. Killingsworth (1983, p, 206}
goes even further, concluding that “female labor supply responses are some-
whatlarger than was apparentin first-generation research” (sce also his quote
634 Whether and How Much Women Work for Pay
at the beginningofthis chapter). He notes as well that second-generation stud-
ies that are based on continuous labor supply schedules (using Procedures It,
V, and VI) tend to produce greaterestimatesof female labor supply elasticities
than do second-generation studies that allow for a discontinuous labor supply
schedule (Procedures VII and VIII), Moreover, most second-generation
research explicitly considering discontinuities in the labor supply schedule
finds that such discontinuities are substantial. Estimates of the “pointofdis-
continuity”—the minimum amountof annual hours at which wages are suf-
ficiently high to overcomefixed costs—are reasonably consistent over diverse
studies. For example, Heckman [1980] and Cogan [1980b, 1981) report esti-
mates that range between only 1126 and (383 hours, while Hanoch’s lowest
estimate is 820 hours.
Unfortunately, however, like their first-generation counterparts, in
terms of wage and incomeelasticity estimates, the second-generation esti-
mates of labor supply responsesare verydiverse. In fact, noting that estimates
of the female uncompensated wageelasticity of annual hours vary from
—0.30 10 + 14.00, Killingsworth and Heckman (1986, p. 185] reach the sad-
dening conclusion that
it is not uncommonfor authors of empirical papers on labor supply to
pointto results in otherstudiessimilar to the ones they have obtained. but
...such comparisons may not always be informati ts all too casy to
find at least one otherset of results similar to almost anyset of estimates
‘one may have obtained.*
In spite ofthis diversity, and in termsofthe secondcriterion for evalu-
ating second-generation studies, on the basis of research to 1980, Heckman,
Killingsworth, and MaCurdy [1981] have argued that clasticity ‘estimates
obtained from sccond-generation studies tend to be greater—sometimes con-
siderably greater—in absolute value than those based on first-generation tech-
niques. Theycite two types of evidence to support their tentative conclusion.
First, an important Monte Carlo study by Wales and Woodland[1980] indi-
cated that, given joint normality, OLS estimates are badly biased toward zero
(though perhaps not as much in the wage equation asin the hours equation),
whereas techniques that correct for sample selection bias are much closer to
the “true” parameter values. Second, studies of female labor supply by,
amongothers, Cogan [1980a], Heckman (1976a}, and Schultz [1980], found
that labor supply elasticity estimates that were based on second-generation
estimation procedures were consistently larger than those based on first-gen-
eration OLS methods.
Notall labor econometricians agree with Heckman, Killingsworth, and
MaCurdy's assessment. In particular, already in the late 1970s, Alice Naka-
mura and Masao Nakamura [1979] reported results of a second-generation
study that found female labor supply to be basically unresponsive to changes
in wage rates, similar to the findings reported by others for males (sce their
quote at the beginning ofthis chapter); additional supporting findings were
11.3 Econometric Issues 635
presented later in Nakamura and Nakamura [1981a: 1985a,b]. This contro-
versy about whether males and females respond differentially to wage rate
changes is of considerable interest, and therefore in Section 11.3.C we will
discuss it further when we consider the Mroz [1987] anticle.
Finally, in terms of analyzing the effects of taxes on labor supply, not
much is known yet concerning the consequences andrelative merits of using
linearized (LBC) or compicte budget constraint (CBC)estimation procedures.
Kilingsworth (1983, pp. 204-205] notes that the LBC and CBC estimates
differ considerably, but since most CBC studies employed restrictive utility
function specifications or limited samples,it is not clear how muchof the
difference is due to LBC versus CBC,versus the choice ofa restrictive func-
| tional form.
The empirical results ontheeffects oftaxes on labor supply reported by
Hausman [1981a] are of particularinterest. however. Hausman incorporates
effects of a progressive incometax schedule,as well as provisions of numerous
US. transfer programsresulting in budget constraint sets with both concave
and nonconcave regions. Following the previous tradition, Hausman treats
the husbandas the primary workerin the family and specifies that in making
her labor supply decisions, the wife treats the husband's income as predeter-
mined.” Significantly, the functional form and stochastic specification
employed by Hausman constrain leisure to be a normal good. Fixed costs of
working are incorporated into the model. Hausman computes virtual
incomes correspondingto cach individual's equilibrium marginal tax rate and
employsa specification in which increases in income neverraise desired hours
of work. However, heterogeneity in preferences among individuals can result
in income effects that vary across the population. A CBC estimation proce-
dure is employed, and estimation results are compared on the basis of a com-
bined concave, nonconcave budget constraintset with results from a globally
concave approximation to a complex budget constraint set.
For husbands, Hausman finds that the uncompensated wageelasticity
is essentially zero but that the incomeeffect is negative and substantial.
Results from the concave and nonconcare estimation procedures are similar.
Given these estimated incomeeffects, Hausmanconcludes that for husbands
with averagetastes and in the $8000~$ 12.000 federal tax bracket in 1980, the
Progressive income tax structure reduces labor supply by 8%compared 10 a
No-tax situation and by 7% compared to a proportionaltax system generating
equaltax revenue.
For wives, Hausman usesresults from his carlier research that suggested
that the wives’ wage rates before taxes are not a function of hours worked;
Mmorcover, he also concludes that sampleselectivity is not a significant prob-
temin the female wage rate equation. Hausmanestimates the uncompensated
wageclasticity for the average woman workingfull time as being equal to a
rather substantial 0.995 and,iffixed costs are taken into account, at 0.906.
Theincomeelasticity is negative and substantial, reflecting in part the leisure
Rormality assumption. These results imply that progressive income taxes
 JERRY A. HAUSMAN
Modeling the Effects of Taxes on Labor Supply
erry Hausman’s con-
J tributions to the the-
ory and practice of
econometrics have been
numerous and wide rang-
ing. While students might
recognizehim bestfor the
oft-used specification test
procedure that he devel-
oped, Hausman has writ-
ten extensively in econo-
metric theory on simul-
taneous equations estima-
tion, the interpretation ofinstrumen-
talvariable estimators, estimation of
models with errors in variables, com-
putational algorithms for nonlinear
structural models, and panel data es-
timators.
Although Hausman has made
major contributions in econometric
theory, he considers himself essen-
tially an applied econometrician. He
has long been interested in public f-
nance (he currently serves as a Mem- D.Phil. in 1973; his‘dissertation was
ber of the Massachusetts Governor's entited “A Theoretical and Empiri-
Advisory Council for Revenue and
Taxation), and this interest led him
to research the effects of taxes on
jJabor supply. One of Hausman’s im-
portant econometric findings, dis-
cussed in this chapter,is that the U.S.
income tax system leads to greater
deadweight loss than had previously
been believed and that an equal rev-
enue-generating tax structure with
personal exemptions and then a pro-
portional tax would involve much
smallerefficiency losses.
In addition to his work on the ef-
fects of taxes on labor supply, Haus-
636
manhas published empir-
ical studies on the effects
of alternative rate struc-
tures on the demand for
electricity, on factors af-
fecting the purchase and
utilization of energy-using
durable goods, on the re-
lationship between pat-
ents and research and de-
velopment, and most
recently, on the effects of
competition and technical
progress in telecommunications.
Born in West Virginia in 1946,
Jerry Hausmanobtained his A.B. de-
gree (summa cum laude) from Brown.
University in 1968. Aftcr spending
twoyears with the U.S. Army Corps
of Engineers in Anchorage, Alaska,
Hausman resumed studies at Oxford
University, where he was a Marshall
Scholar. At Oxford, Hausman earned
a B.Phil. degree in 1972, and a
cal Study of Vintage Investment and
Production in Great Britain.”
Hausman’s first academic ap-
pointment was at MIT in 1972, and
aside from occasional visiting ap-
pointments, he has been there cver
since, teaching courses in economet-
tic theory, microeconomics, public
finance and a course in the industrial
organization of tclecommunications.
Hausman has beenelected a Fellow
of the Econometric Society, and in
(985 he was awarded the John Bates
Clark Medal by the American Eco-
nomic Association.
11.3 Econometric Issues 637
reduce wives’ labor supply in large part by decreasing the net after-tax wage.
Specifically, to the extent that the wife is a “secondary” worker, Hausman
hotes that she faces a “marriage tax,” since thefirst dollar of her earnings is
taxed at the rate that is applicable to the last dollar of her husband's earn-
ings.** Hausman also examinestheeffects oftaxes on labor supplyof female
headsoffamilies and reports significant sensitivity to taxes.
Overall, Hausman estimates that the progressive federal U.S.incometax
structure existing in 1980 results in an 8.6% declinein labor supply compared
to a no-tax situation and estimates that the deadweightloss as a percent of
incometax revenuesraised is 28.7%. Owing to progressivity, deadweight loss
increases sharply with market wage rates and associated marginal taxes. By
contrast, 2 proportional incometax system that generated identical tax reve-
Nues would reduce labor supply by onlyslightly more than 1%, and the aver-
age deadweight loss as a proportion of tax revenues raised would be only
7.19, Arguing that a proportional tax system with an initial exemption pro-
vides an alternative way of implementing progressivity without entailing as
large a deadweight loss, Hausman concludes that “the progressive income tax
sysiem may be a costly way to seek incomeredistribution. Other, more cost-
effective means of incomeredistribution maywell exist.”
Hausman’s findings are not universally accepted, however. For exam-
ple, in commenting on Hausman [1983]. Heckman (1983) conjectures that
Hausman’s ratherlarge wage responsiveness estimates reflect at least in part
the effects of his assumption, imposed on the data, that leisure is a normal
good.
11.3.C On the Sensitivity of Results to Alternative
Statistical and Economic Assumptions: The
Mroz Study
In reviewing second generation empirical results in the previous section we
emphasized thatthe substantial heterogeneity that is observed in the estimates
of wage and incomeelasticities could be ducin part to the fact that the various
studies differin terms of estimation method,functional form, and data base.
This suggests that an informative research project would involve undertal ing.
detailed sensitivity analyses of existing alternative second-generation estim:
tion methods but using a common dala base and functional form. That is
Precisely the approach taken by Thomas Mroz [1987], who employs data
from the 1976 Panel Study of Income Dynamics (PSID) sampleconsisting of
753 white married women aged 30-60, of whom 428 worked al some time
during 1975, Using these data, Mroz compares elasticity estimates based on
a variety ofalternative statistical and economic assumptions that were made
by authors of previous studies,
Mroz's study is of considerable interest because, although it does not
focus on new methodological developments, it yields important results con-
cerning sensitivity, and it follows up on Nakamura and Nakamura (1979;
638 Whether and How Much Women Work for Pay 11.3 Econometric Issues 639

198 1a; 1985a, b} in challenging the conventional wisdom of labor economists ate positive, the range is substantial, varying from —499 to +1462 for a $1
that the responsivencss of female labor supply to changes in wage rates and change in the wagerate; estimated incomeeffects per $1000 in 1975 dollars
incomeis larger than that of men.” We now examine the Mroz study and the are primarily negative but range from — 125 to +51. Further, there does not
associated sccond-generation empiricalliterature in more detail. appear to be any obviousrelationship between the estimation method thatis
Mroz begins by noting the very large diversity af reported estimates of employed and the magnitudeorsign ofthe estimated labor supply response.
female labor supply responses to variations in wage rates and income.” To Mroz proceeds by choosing a simple functional form for the labor sup-
makethe results of various studies comparable, Mroz evaluates the implied ply equation, one that Nicholas Stern [1986] derived explicitly from an indi-
effects of these studics on annual hours worked in 1975, assuming that the rectutility function. This labor supply equation is
wife's wage is $4.50 and her hours worked are 1500, the husband’s wage is
$7.00 and his hours are 2000, the household nonlabor incomeis $1000, and H, = dy + a, In(W) + a ¥, + aZ, + (11.48)
the labor and nonlabor marginal tax rates are 0.339 and 0.280, respectively. where H,is the ith woman's hours of work in 1975, HV, is a measure of her
Mcoz’s evidence on the wide range oflabor supply response estimates is pre- wage rate, V, is a measure of other incomereceived by the household in
sented in Table 11.2, which reproduces Table | of Mroz[!987, p. 766]. There thousands of 1975 dollars, Z, is a vector of additional control variables
it is scen that while the vast majority of uncompensated wage effect estimates (including the wife’s age, her years of schooling, the numberofchildren less
than six years old in the household, and the numberofchildren between the
ages ofsix and 18), and¢; is a stochastic disturbance term. Al a given wage
Table 11.2 MROZ's Survey of Estimates of Marrled Women's Labor Supply fate, Mroz computes the uncompensated wageeffect on hours worked for a
Responses $1 wage rate change as simply dO’, = a,/W’, and the income effect as
Estimation Method Wage Effect incomeEffect of/aV’, = a;. The corresponding uncompensated wage and incomeelastici-
ties are then computed as 3 in H./d in W, = a,/H,and dln Hfain Vi = aVs
[1973] Instrumentalvariables 29 ~16.9 H,, respectively.
‘Cogun[1980a] Tobit 865 —32,3 The 1976 PSID data sample (for year 1975) is employed by Mroz in this
Cogan (1980a] Instrumental variables 349 study because only in 1976 did the PSID survey directly interview wives in
Cogan [1980b} Tobit 632 households. Duringall other years the household head supplied information
Cogan [1980b) Fixed costs 196 aboutthe wife’s labor market experiences during the previous year. As Mroz
Cogan[1981] Fixed costs 269 notes, “one suspects that the own reporting is more accurate,andit is for this
Greenhalgh [1980) Instrumentalvariables 213 feason that many recent studies of married women’s labor supply have used
Hausman [!981a] Convex budgetsets 328 Of these data.”* Annual hours worked H, is measured as the product of the
Hausman[19813] Nonconvex budget sets 335 numberof weeks the wife worked for pay in (975 and the average number of
Hausman [1981a] Fixed costs 305 hours of work per week during the weeks she worked. The wage rate 1’, is
Heekman [1976a] Tobit 1462 measured as average hourly earnings, calculated as total labor income of the
Heckman [19763] Generalized Tobit -499 wife in 1975 divided by H,. Note that if there is a random measurement error
Heckman[1980] Generatized Tobit i401 in annual hours worked, a spurious negative correlation will emerge between.
Layard etal. [1980] Tobit 128 hours worked and this wage rate measure. Finally, the Property income mea-
Layard et al. [1980] Instrumental variables 22 sure V, is defined as the houschold’s total money income minus the wife's
Leuthold [1978] 1967estimates 14 labor income,in thousands of 1975 dollars.
Leuthold [1978] 1969estimates 45 For a “base case estimation,” Mroz reports results when parameters in
Leuthold [1978] 1971 estimates 33 Eq. (11.48) are estimated by OLS and the sampleis restricted to those who
Nakamura and Generalized Tobit -16 work(recall from Section 11.3.B.1 thatthis is Procedure It)
Nakamura (198ta)
Schultz (1980] Tobit 123 -67.0 Hy = a — 17 n( WW) — 4.2 V, — 342 KL6, ~ 115 K618, + others
Schultz [1980] Instrumentalvariables —26 19 (81) 3.2) (131) (29)
‘eter Ties effects are evaluated ot 3 wife's wage of$4.30, wife hours at £500, buaband's wage of$7.00, ustand?s hour at (11,49)
2000, huosehold nantshut income of 51000, and Tabor and rontabar raarginal tax rite of€.339and 0.280especie, The
‘Suurce effect
iacomne is per$1000
Data from ThamasinA.1975
Mrozdtiare
“The Serutvtyofan Empirical Model ofMarried Women’s HoursufWark to Economic Here KL6,is the numberofchildren underage six in the household, K618,
a SiatitenAtsumptionn” Economica. Vol. $8, No.4, 1987, TAB 1,9. 766 is the numberofchildren between ages six and 18, numbers in parentheses
od Whether and How Much Women Work for Pay
are estimated standard errors (adjusted for arbitrary formsof heteroskedastic-
ity, using the Halbert White [1980] procedure), and “others” refers to the
additional variables (wife's age and wile’s education), whose paramcleresti-
mates, along with the intercept, are nol reported by Mroz. Note that these
OLS estimates imply that both the uncompensated wageeffect and income
effect are negative (the elasticities are — 17/1500 = —09l 13 and —4,2/1500
= —0,0028, respectively) and that the presence of children in the household.
(particularly those under age six) reduces labor supply substantially, other
things being equal. The negative OLS estimated wage effect is not surprising.
since as was noted above, if there is a random measurementerrorin //,, then
a spurious negativerelationship occurs between #/, and 1, ; With the experience variables excluded as instruments the uncompensatec
With Eq. (11.49) as a basis of comparison, Mroz then examines the
effects of three variations: (1) alternative exogeneity assumptions on the
fegressors. issues that are raised in particular by Nakamura and Nakamura
[1981a]; (2) statistical controlforself-selection into thelabor force—a classic
issue; and (3) the impacts of controlling for taxes. We begin by summarizing
results that occur once Mroz controls for possible correlation between regres-
sors and the disturbance term in Eq. (11.48) through the use of an instru-
mentalvariable procedure such as two-stageleast squares QSLS).
Treating In (3¥,) as an endogenous variable that might be correlated
with the disturbance term ¢, in Eq. (11.48). Mroz uses as instruments the
wife’s reported labor market experience (the number of years the woman has
worked for pay since her eighteenth birthday), this variable squared, and seve
erat different polynomials in the wife's and husband's age and education. This
variant on Procedure II {sce Section 11.3.B.1) results in an estimated 2SLS
equation whose wage and income parameters are much different from the
OLSestimates of Eq. (11.49):
H, aa + 672 In (IK) — 64, — 283 KL6,— 85 K618, + ---
(217) (3.6) (47) (36F
(11.50)
Note that unlike the OLS estimates, here the 2SLS estimate ofthe uncom-
pensated wage response is positive and highly significant; in fact. this large
2SLS estimate on the wage coefficient is similar to some ofthe largest wage
effects reponed in Tabie 11.2. The 2SLS incomeeffect, however, is more neg-
ative than the OLS estimate, and the 2SLS estimates of theeffects of children
on labor supply, while also negative, are smaller than OLS estimates. Further,
using the clasticity formulae underneath Eq. (11.48) and evaluating deriva-
tives at H, = 1500 and V, = I (recall ¥, is measured in thousands of 1975
dollars), we see that the implied 2SLS. uncompensated wage clasticily estimate
is 672/1500 = 0.448, while the income elasticity estimate is very small,
—6.41/1500 = —0.0043.
Fellowing 3 conjecture by Nakamura and Nakamura [1981a, p. 480],
Mroz then examines whetherthe disturbance term in Eq. (11.48) reflects in
part the unobservable preferences of the individual, inctuding her attitude
IL3 Econometric issues 64
toward work. If this were true,it is likely that a variable such as the wile’
labor market experience would alse be correlated with the disturbance tern
and that as a result this variable andits square would not be legitimate instru
ments. Mroz therefore recstimates Eq. (11.48) by 2SLS but excludesthe expe
rience variable andits square as instruments. This yields an equation with ;
much smaller andstatistically insignificant estimated wageeffect:
H, = dy + 46 In (WV) — 4.4, - 337 KL6, - 112 K618, + ~~
(220) (3.3) «31 (30)
(151
wageelasticityfalls to 46/1500 0.031, and the incomeelasticity to —4.4#1,
1500 = —0.0029. Such low estimates match someof the smallest estimate:
reported in Table 11.2. The estimated effects of children on labor supply
however, are not affected as muchandin factare quite similar in the various
OLS and 2SLS specifications. These results therefore demonstrate that esti-
mates of the wage and incomeresponses are very sensitive to assumptions
concerning the endogeneity ofregressors and the choice of instruments.
To makethe choice of instrumentsless arbitrary, Mroz proceeds with 3
variant of the Durbin-Wu-Hausman-Whitespecification test.* In this context
the null hypothesisis that the experience instruments are uncorrelated with
«, and the alternative hypothesis is that they are correlated. Mroz showsthat
a test of this hypothesis amounts to testing whetherthe estimated wage coel-
ficient in the 2SLS equation (11.50) with the experience variables included as
instruments is equal to the estimated wage cocfiicient in the 2SLS equation
(11.51) with experience variables excluded as instruments. He derives the for-
mulae for the asymptotic variances and covariances and obtains an asymp-
totic normaltest statistic of 3.0, impiying that the null hypothesisis rejected
at any reasonable level of significance. Mroz therefore concludes that the
wife's previous labor market experienceis not exogenous in the labor supply
equation and that as a result the experience variables are invalid instruments,
consistent with the Nakamura and Nakamura {19813} conjecture. Interest-
ingly, Mroz undertakes similar exogeneitytests for the KL6,. K618,, and ¥,
variables andfinds that the exogeneityofthese children and nonwite income
variables cannotbe rejected at reasonable levels of significance.”
Second, as was discussed in Section !1.3.B.1.a major problem with each
of the above Procedure II estimatesis that they are based on a subsample of
working women only and therefore do not control for self-selection into the
laborforce. To examine the impactofignoringself-selection Mrozesti-
mates the labor supply equation using the Heckit multistage Procedure VINE
In particular, using the full sample of 753 married women, Mrozfirst esti-
mates a probit model of labor force participation in which the explanatory
variables include KL6, K618, J” polynomials in the wife’s age and education,
the education of the wife’s mother and father. the county's unemployment
fate, a dummy variable for large city. and the wife's previous labor market
642) Whether and How Much Women Work for Pay
experience and its square (note that the wife’s wage is not included). Of the
18 regressors, only the coefficients on KL6, V, experience, and its square are
more than twice their estimated standard error. Mroz then uses this probit
modelto estimate the inverse Mills ratio \ (see the discussion in Section
11.3.B.1, especially that on Procedures VII and VIII) and appends this vari-
able as a regressor in a wage rate determination equation in which the other
tight-hand variables are the samc as in the estimated probit equation but
where the sampleis limited to workers (428 of the 753 observations). In the
third andfinal stage, Mroz uses thefitted value from the wage determination
equation as an instrument in a 2SLS estimation of the hours worked equation
(11.48) but with the inverse Mills ratio 4, added as a regressor. It is worth
noting that since the wife’s previous labor market experience andits square
are regressors in the first two stages of this Heckit procedure, this particular
estimation assumes that these experience variables are exogenous.
The resulting estimated labor supply equation has a small positive
uncompensated wage effect and a smail positive incomeeffect, neither of
whichis statistically different from zero:
H, = ay + 122 In(W} + 3.9 V, + 53 KL6;
(225) (45) (173)
— 87 K618, — 7584, + ---
G4) a7 152)
In this case, when evaluated at the same wage and incomeas earlier, the esti-
mated uncompensated wage elasticity is but 0.081 (122/1500), and the
incomeelasticity is 0.0026 (3.9/1500). Surprisingly, the estimated coefficient
on young children KL6 is now positive (but insignificantly different from
zero), while that on K618 is still negative but considerably smailer in absolute
value than in Eq. (11.50). Since the ratio of the estimated coefficient on 4, to
its standard erroris 753/171 = 4.43, a numberthalis greaterthanthecritical
value for the normal distribution at reasonablelevels of significance, Mroz
rejects the null hypothesis that there is noself-selection bias.
However, when the same multistage Heckit estimation Procedure VII
is employed but the wife’s labor market experience andits square are omitted
as regressors in the probit and wage determination equations, a rather differ-
ent labor supply equation is obtained:
H, = a + 641n(W) ~ 1.0 V, — 183 KL6,
(227) (9.2) (408)
— 106 K618, — 294a,+ ---
(34) (467) (11.53)
Not only is the uncompensated wage effect coefficient small and insignifi-
cantly different from zero, but the incomeeffect is negative, small, and also
insignificant. Further, the estimated coefficient onA, is insignificantly different
from zero, implying that one cannotreject the null hypothesis that there is
noself-selection bias.*
11,3 Econometric tssues 643
These results suggest that whether the experience variables are assumed
to be exogenous has a decisive impact on whether theself-selection bias is
statistically significant in the labor supply equation. Totest for exogeneity of
the experience variables, Mroz uses the samespecification test procedure as
was described earlier, but with the asymptotic variance formulae adjusted to
include sample selectivity. The nui! hypothesis of exogeneity reduces again to
@ very simple test—whether the estimated wage coelticient in Eq. (11.53) is
equalto that in Eq.(11.52); since the difference in estimated coefficientsis 58
(122 — 64) and the estimated asymptotic standard error is 163, the null
hypothesis of exogeneity is not rejected.
‘On thebasis ofresults from these fourestimated equations, Mroz draws
a umberofconclusions, First, the estimated uncompensated wageeffect is
positive but typically very smail, provided that one treats the wife's experience
variables as “endogenous” and/or oneallows for sampleselectivity bias in the
wage determination andlabor supply equations. Further, under the sameesti-
mation procedures the incomeeffect is usually negative butalso rather small,
Mroz[1987, p. 794] therefore concurs with Nakamura and Nakamura
(1979; 1981a; 1985a, b] and rejects the conventional wisdom based on sec-
ond-generation studies, concluding that
the small income and wage effects found in this study provide a much
more accurate picture of the behavioral responses of working women to
variations in nontabor income and wages than those found in most pre-
viousstudies.
Suchfindings suggest as well that the modest responsiveness of married wom-
en's labor supply to changes in wage rates and incomeis not muchdifferent
from that foundin studies of the labor supply of prime-aged matried males.
Second, the sampleselectivity bias is significant only if the wife’s experience
variables are treated as exogenous. Third, the wife’s labor market experience
variables are exogenousprovided that oneallows for sampleselectivity bias,
Fourth,theresults are consistent with the view that a woman'stastes for work
are unobserved omitted variables that affect both her previous and current
labor marketparticipation and hours of work, over and abovetheir influence
on her wagerate.”
In addition 10 estimating the labor supply equation using the Heckit
multistage Procedure VII, Mroz employs the full Tobit simultaneous equa-
tion estimation Procedure VI. Recall that unlike Procedures VII and VIU,
which allow for discontinuities in the labor supply schedule dueto,say,fixed
costs of employment, the Tobit Procedure VI assumes continuity. In fact, the
Procedure V1 estimates will be equal to those from Procedures VI] and VII
whenever the labor supply schedule is continuous in wages. Mroz reports that
his Procedure VI Tobitestimates of the labor supply equation are
H, = a, + 261 In(W,) — 22.9 V, — 1035 KL6, ~ 97 K618, + - + -
(357) (4) (140) (48)
(11.54)
$44 Whether and How Much Women Work for Pay
Hence the Tobit uncompensated wage elasticity estimate is also quite small
(261/1500 = 0,174) and insignificantly different from zero, while the income
effect is significant and larger than those estimated above {—22,9/1500 =
0.015)but is stil! rather small, Further,with the Tobit specification the null
hypothesis of exogeneityof the experience variables is rejected. Mrozthen
tests for this Tobit specification (11.54) as a special case of the muitistage
Heckit Procedure VIII and in all cases rejects the null hypothesis, This leads
Mroz to conclude that the restrictions implied by the Tobit Procedure VI
must be rejected, a conclusion thatis consistent with the view that fixed costs
of employment are substantial. . | | future. Further, in dynamic models, the accumulation of nonhuman and/or
In the third phase of his study, Mroz assesses the effects of including a
measure of the marginal tax rate into the labor supply equation. Following
Hall [1973]. Mroz uses a linearization ofthe budgetset in which themarginal
afler-tax wage rate replaces the previous wage measure and virtual income is
measured as the intercept of the linearized budget set at zero hours of work;
untike Hall, however, Mroz treats both the after-tax wage rate and virtual
income as endogenous variables. Using similar estimation Procedures as
before,” Mroz finds that the estimated wage coefficient falls and in fact
becomes negative but that the magnitude of the change is at most 33 hours
andis always within 20% of one standard deviation oftheestimates obtained
without taxes. This leads Mroz [1987, p. 786] to conclude, “{n the light of the
other possible sources of bias examined in the previous sections, the influ-
encesof taxes on the estimates of the labor supply parameters appears to be
at most a second order effect.” . |
Finally. following Rosen [1976], Mroz rewrites Eq. (11.48) in after-tax
form as the expanded equation
H, = a) + a,in[Wl — 7)} + bn — 1)
+ aV,+ mYV, + aiZ,+ 6 . (11.58)
where z, is the marginal tax rate and YV, is virtual income. The null hypoth-
esis that married women optimally take taxes into consideration is that b =
ad, = 0, while the alternative hypothesis is that 6, # 0, a; # 0. One might
also consider testing the hypothesis that married women completely ignore
taxes when making labor supply decisions. This corresponds lo the null
hypothesis that 6, = —a, and b, = 0, while the alternative hypothesis is b,
# ~ay, b; # 0. Unlike Rosen, Mrozfindsthat neither of these two distinct
null hypotheses can be rejected, perhapsreflecting the relatively large standard
errors on the estimated wage coefficients and income terms. | where at least oneof the inequalitiesis strict." Thus the total number ofperi-
As can be seen, therefore, the significance of Mroz’s study is that
it highlights the empirical consequences of employing alternative assump-
tions concerning exogeneity, sample selectivity, and taxes on estimates of
the labor supply responsiveness of married women. These assumptions do
matter.
11,3 Econometric Iss eS 645
44.3.0 Issues in Dynamic Labor Supply
Oneinterpretation of Mroz's results on the importance ofthe Sabor force
experience variablesis that labor supply might better be viewed in a dynamic
life cycle context, Before concluding this chapter, therefore, we briefly over-
view the rapidly growing theoretical literature an dynamic labor supply.
Within thelast decade, considerable theoretical research has focused on
dynamic labor supply models. In dynamic models oflabor supply, economic
agents act knowing that today’s decisions will have consequences in the
human capital is modeled explicitly. Our overview ofthis literature is by
necessity brief, more detailed discussions can be found in Killingsworth
[1983, Chapter 5], Killingsworth and Heckman (1986, pp. 144-179], and,
in
particular, Pencavel [1986]. We begin by discussing models in which
wages
at each point in time are exogenous and then examine models in which
wages
are determined endogenously through human capital accumulation,
Although discussions of the timing of labor supply and human capital
accumulation duringthe life cycle uf women have a longhistory in labor eco-
nomics, one ofthe most influential early contributions on dynamic issues
is
that by Jacob Mincer [1962]. Mincer suggested that one way of thinking
through dynamicissues is simply to reinterpret the static analyses of
labor
supplyin lifetime terms, where variables such as consumption,leisure, work
at home, wages, budget constraints, and time are envisaged aslifetime vari-
ables. Note that in this view, the woman's real asset holdings should be rein-
terpreted as those occurring at the beginningof the (lifetime) time period and
not as her actual lifetime or current exogenous income,
Following Killingsworth and Heckman [1986], tet us assume that the
life cycle consists of T periods, and let us then sort single-period reat
wage
rates in descending order so that w(1) denotes the highest real wage and
w(7")
denotes the lowes. For the moment, ussume that consumptionand leisure at
different points in time are pertect substitutes. Under these conditions an
individual will work sometime duringherlife if #(1) exceeds her lifetime
res-
ervation or shadow wage rate A{RS(O), where the latter is the mutrginal rate
of
substitution evaluatedat zero lifetime haurs of work. To determine the total
numberof periods worked, compare discounted real wage rates to (he
MRS;
the individual will work exactly & time periodsif
w(k) = MRS(K) > wik + 1) (11.56)
ods worked & is a function of an unobserved taste or houschold production
variable ethat aflects lifetime utility, real initial wealth A, and the “marginal
wage" w(K), that is,
k= KDWAT, e] (1.87)
646 Whether and How Much Women Work for Pay
In this case the proportion ofall periods in the woman’s lifetime that is
devoted to work, #. is simply A = &/T. which together with Eq. (11,57)
implies that
h = Alntk), A. T, e] (10.58)
Onepractical problem with implementing empirically equations such
as Eq.(11.57) or Eq. (11.58) is that it would appear that each requires & or /t
data on labor supply over the entirelife cycle——a truly formidable obstacle.
As Mincer[1962] and Heckman [1978] have noted. however,if one abstracts
from “transitory”factors, one might assumethatthe timing of work over the
life cycle is random. If this were true, andif all individuals worked at some
point in their lives, then one could estimate parameters of Eq. (11.58) by
replacing /t (whichrefers to lifetime participation) with, say, Q, where @ now
refers to participation as of a given date; the resulting parameter estimates.
could then be used to obtain estimates ofelasticities of labor supply with
respect to (permanent) wage changes and with respect to income (here,initial
wealth).
While apparently promising, however, this approach suffers from the
practical ambiguity of determining the appropriate measure of the “marginal
wage” w(k) that could clearlydiffer from the wage prevailing as of the date
referenced by the Q variable; specifically. while data on lifetime labor supply
might not be necessary, onestill needs to be able to determine which partic-
ular wage rate—of al! the wages the individual will carn during her lifetime—
happensto be the appropriate marginal wage rate in Eq. (11.56).
Another problem with this approach, emphasized by Killingsworth and
Heckman [1986, pp. 148-149], is that using estimates of Eq. (11.58) with @
replacing / to obtain measures of substitution and incomeeffects is appro-
priate only when is strictly positive, that is, only when individuals have an
fat
interior solution 10 their lifetime labor supply optintization problem. As
Heckman (1978) notes, although appropriate data are rather scarce, there is
reason to believe that a nonnegligible portion ofthe female population never
undertakes market work for pay. Therefore this approach might not provide
usefulestimates of income and substitution effects.
Muchrecent work has focused on modeling more explicitly the factors
that affect an individual's equilibrium sequence of lifetime labor supply, lei-
sure, and consumption; the solution to thelifetime optimization problem is
often called the individual's equilibrium dynamics. A common approach is
to assumeperfect certainty and specify lifetime utility as an additively sepa-
rable utility function
r stream of anindividual's future rates from ¢ = 0 tor = T, discounted at rate
v= Vast syei(cw, LO) (11.59)
= and the permanent wage W,, that is, w, = 4() — W, AsKillingsworth and
where Tis fixed, L(f) + (4) exhausts all available time during period ¢, 5 is
the individual's subjective rate of time preference, and t[-] is the utility func-
11.3 Economettic Issues 647
tion for period 1, depending on consumption C(¢) andleisure L(2). Inciden-
tally, in this additively separable specification, C(s) and C(z’), as well as L(0)
and L 1), are not assumed to be perfect substitutes for ¢ # £5?
Lifetime utility is maximized subject to a lifetime budget constraint
r
AQ) + SO + yt FO) — PCI) = 60)
=o
where A(0)is the individual's initial asset holdings, r is the market rate of
interest, and (2), H(¢), and P(1) are the wagerate, hours of work, and price
levels in period 1, respectively.
More formally, using Lagrangian procedures, maximize Eq. (11.59)
over C(t) and H(z) subjectto the lifetime budget constraint (11.60), and inter-
pretthe Lagrangian multiplier in front of Eq. (11.60), say, A, as the marginal
Iifetime utility of initial asset holdings. Thefirst-order conditions yield an
intertemporal equilibrium plan, given a set of wage rates andprice levels P(r)
and Pt), 2 = 0,..., T. for C(t), £0), and A. Onecan ofcourse assess how
the equilibrium paths of differing individuals would vary as a result of their
holding different -1(0) or facing diverse wage rates #¥(1); such mental experi-
ments are called comparative dynamics and should be distinguished {rom the
individual's initial intertemporal equilibrium sequence,called her equilil-
riwn dynamics.
| In carrying through comparative dynamics, one mustof course be care-
ful in specifying what is being held fixed. One commonapproach,called the
Frisch 4-fixed or Frisch demand notion, examines the equilibrium timepath
effects of a changein, say, (1), treating the marginallifetimeutility ofassets
a as fixed; an alternative procedure, dubbed A-variable, allows this marginal
utility of assets to vary in response to a change in #(2)."In either case it can
be shown thatvariation in a wage at any given date may have consequences
notonly at that peint in time, but also at other dates.
; It is worth noting here that within the labor economics literature, much
ink has been spilt concerningthe effects of changes in “permanent”or “tran-
sitory” income onlabor supply overthelife cycle. Controversy has occurred,
for example, concerning whether one should include only measures of per-
manent incomein labor supply equations or whether one should also include
a measure of transitory income. Although muchofthisliterature is informal
rather than rigorous, it can be assessed within a comparative dynamics
framework.
Specifically, define permanent income W’, as the present value of the
r, and transitory income w,as the difference between the actual wage (1)
Heckman [ 1986} emphasize, empirical measures of W, have been constructed
in practice by using essentially ad hoc procedures, dependingin large part on
the nature ofthe available data.
648 Whether and Row Much Women Work for Pay
‘One body ofliterature adopts whatKillingsworth and Heckman [1986]
call the PO position, according to which hours of work and labor foree pat-
ticipation in any period ¢ depend on the permanent wage only, analogous to
Mikon Friedman’s [1957] permanent income theory of consumption. Under
the PO hypothesis, one need never include the transitory wage measure w, in
ati equation such as Eq. (11.50), since its coefficient should be zero.
An alternative to the PO position, called PT, hypothesizes that both MH,
and w(1) should appearin labor supplyequations; moreover. accordingto this
literature, the coefficient on w, should be positive and algebraically greater
than the coefficient on H,, since the latter represents the sum of a positive
substitution effect and a negative incomeeffect, whereas the formerreflects
only a positive substitution effect.”
Using comparative dynamic procedures with both A-constant and d-
variable, Killingsworth and Heckman [1986, pp. 156-158] show that con-
trary to PO,leisure and labor supply arc responsive to transitory wage differ-
ences, owing to thefact thatlife cycle asset accumulationis affected. Further,
contrary 10 PT. it is not necessarily the case that labor supply at time 7 is
positively correlated with ,. However, if one examines instead the compar-
ative dynamics of changes in exogenous income (such as the incomeof other
famity members) on an individual’s equilibrium dynamics (rather than
changes in her 4’, or w,), it turns out that labor supply depends only on per-
manent exogenous income and not on transitory exogenous income.**
Not surprisingly, Killingsworth and Heckman conclude that much of
the ambiguity in the transitory-permanent income controversyarises because
of a lack of proper theoretical foundations. As an alternative, they suggest
employing the Frisch demand framework, in which the marginal utility of
assets X constitutes a type of “permanent wage” and variations in the
observed wage #1/(4) with d constant correspondtoa type of “transitory” wage
variation. since with A constant it can be shownthat diffegences in f’(r) must
always be negatively correlated with L(1}.
Our discussion on dynamic labor supply to this point has assumed that
wages are exogenous. By contrast, and as discussed morefully in Chapter 5
ofthis book {see especially Section 5.1},there is a great deal of literature sug-
gesting that wage rates for women are endogenous.affected in large past by
imtermittent labor supply and expectations concerning child-rearing. More
specifically, since current wagerates dependin part on human capital endow-
ments, and since human capital formation is affected by educational attain-
ment and on-the-job training, the choices that women make concerning mar-
riage, child-rearing, and work affect their time paths of wagerates. As a result,
in the more complete and necessarily more complex dynamic models of
female labor supply, educational attainment, marriage, the desired number
and spacing of children, the time path of market labor supply, and the time
paths of wages and consumption are all endogenous.
In large part because of their mathematical complexity, formal theoret-
ical models of joint labor supply and wage determination have not yet yielded
U4 Additional Remarks on Econometric Analyses 649
much empirical insight into the dynamics of women’s work and wages. How-
ever, this topicis currently attracting a great dealofattention both from the-
orists and from econometricians.”
ADDITIONAL REMARKS ON ECONOMETRIC
ANALYSES OF LABOR SUPPLY
As we haveseen, the literature on factors that affect female labor supply
vides a useful example of how economic theory, measurement issues, and
econometric technique caninteract 10 enhance our understanding of impor-
tant public policy issues, But this literature has also amplyillustrated that in
Spite of enormous progress in theory, technique, and, to some extent, mea-
surement, it is still frustratinglydifficult to “nail down”precise estimates of
imponant clasticities. While critics of recent econometric practice might
interpret this difficulty as further evidence that the payofis of state-of-the art
econometric research are often meager, many economelricians who work on
labor supply remain strongly optimi: Killingsworth [1983,
example, concludes his summary ofwhat we do and don't know p.about
432], for
labor
supply by stating, “At least temporarily, then, uncertainty about. and varia-
tion in, actual magnitudes oflabor supply estimates seems larger than it used
to be.” Whether Killingsworth’s 1983 assessmentthat this large rangeofelas-
licity estimates is but a temporary condition in the history of the analysis of
labor supply remainsto be seen, .
Finally,it is worth noting that somestudents oflabor supply argue that
nereason for the large variability in labor supply elasticityestimates, which
is evident in Table 11.2, is that existing theoretical models of labor supply are
simply too crude. Although we have attempted in this chapter to discuss the
currentlabor supply literature in a relatively thorough mannerand have dem-
onstrated some ofits complexity, in fact the labor supply literature is
extremely large, muchofit still needs to be synthesized, and by necessity we
have ignored potentially important strandsofresearch. For example, we have
only skimmed theliterature on family or household d ion-making pro-
cesses, as well as the significance of hours worked quantity restrictions on
labor supply.” Further, we have entirely overlooked labor supply ues Con
cerning the quality of work, the amountofleisure time spent at work, and the
tisk and uncertainty ofjob search. In terms of econometric matters we have
considered only the relatively simple limited dependent variable models.*
and among other omissions we have not dealt with issues concerning the pos-
sibility of obtaining better econometric estimates of labor supplyelasticities
by employing data from well-designed experiments,”” nor have we examined
issues ofspecification, estimation, and inference whenthe labor Supply obse:
vations are taken from a pane! datasel.” Fortunately, readers whoare inter-
ested in pursuingthese related issues can begin by working through the tefer-
ences given here and those provided in leading labor economics textbooks,
such as thosereferenced al the end ofthis chapter.
 650 Whether and How Much Women Work for Pay
44.5 HANDS ON WITH LIMITED DEPENDENT VARIABLE
TECHNIQUES TO ASSESS THE LABOR SUPPLY OF
MARRIED WOMEN
We have now cometo the hands-on portion ofthis chapter. The goal of the
following exercises is to engage you in obtaining firsthand experience in
implementing and interpreting alternative procedures that attempt to mea-
sure factors affecting female tabor force participation and hours worked; in
particular. you will have the opportunity to carry out on your own a number
of estimation methods, including most of the second generation procedures
discussed in Section 11.3.B.1.
Note: Althoughthe exercisesof this chapter have been carefully designed
to introduce you to someofthe most important computational and interpre-
tive issues in the econometric analysis of labor supply, they are but an intro-
duction, and yourinstructor may use the data set in this chapter for other
useful and more elaborate exercises. For example, although data on the hus-
band’s hours of work and wages are provided in the MROZdata file on your
diskette, we make no use of those data in the exercises. Such data could be
useful in a model offamily labor supply. Alternatively, your instructor might
wantto provide you with a different set of data. Hence we urge you to en-
visage these exercises as a beginning step to understanding better the applic-
ability of limited dependent variable procedures to the analysis of labor
supply.
Theexercises of this chapter can be summarized as follows. Because it
is extremely important to examine data sets before becoming involved in
econometric estimation, we highly recommendthatall readers work through
Exercise 1, “Inspecting Mroz’s 1975 Panel Study of Income Dynamics Data,”
Moreover,this exercise is essential because in it you generate two dataseries
that are necessary for completion of subsequent exerciseg. In Exercise 2 we
intcoduce you to a common but inappropriate estimation procedure in which
an hours worked equation is estimated by OLS with hours of nonworkers set
to zero (Procedure I). Then in Exercise 3 we invite you to explore the female
laborforce participation decision using OLS,probit, and logit estimation pro-
cedures and to examine numericalrelationships among them. In Exercise 4
we ask you to compare the conditiona! OLS and Tobit estimators ofthe hours
worked equation (Procedures 1] and [1f) and to implement the Goldberger-
Greeneprocedure for assessing the OLS bias. Next, in Exercise 5 we have you
consideridentification issues when both the wage rate and labor supply equa-
tions are estimated by using Procedure IV. In Exercise 6 we lead you through
estimation of the hours worked equation based on theincreasingly common
Heckit multistage procedure (generalized Tobit), which Killingsworth called
Procedure VIII. In Exercise 7 we have you estimate a Heckit model in which
taxes are taken into account, and finally, in Exercise 8 we work with you in
specifying and estimating an extended Tobit model (Procedure V1).
Cercecocce
Exercises 65)
y Manyofthe data sets that are used to estimate models oflabor supply
are large,often involving over 16,000 observations, and it is simply infeasible
to provide such a large data base on your 360K data diskette. However, the
1976 PSID data set employed by Mroz [1987] in his sensitivity study is useful
and convenient, since it is rather small(it contains 753 observations on 19
variables and occupies only about 60K ofspace on yourdata diskette), andit
permits replication of some of Mroz’s most important findings.”
In the subdirectory called CHAPILDAT ofyourdata diskette you will
find a data file called MROZ.This data file contains 753 observations on mar-
ried white women aged 30-60 in 1975 for i9 variables. The first 428 obser-
vations are those for women whose hours of workin 1975 were positive, while
the final 325 observations are those for women who did not work for pay in
1975, Each 80-column row ofthis data set corresponds to one observation.
Thefirst variable, LFP. is a labor force participation dummyvariable
that equals 1 if the woman's hours of work in 1975 were positive; olherwise,
it equals zero. WHRSis the wife’s hours of work in 1975, while KL6 and
K618 indicate the numberof children in the houschold under age six and
between ages six and 18, respectively. WA is the wife’s age in years, WE is the
wife’s educational attainment in ycars of schooling, WWis the wife's 1975
average hourly earnings in 1975 dollars, and RPWG is the wife's wage
reported at the time ofthe 1976 interview,in dollars. The HHRSvariable is
the husband's hours worked in 1975, HA is his age, HE is his educational
attainment in years of schooling, and HW is his 1975 wage in 1975 dollars.
FAMINCis the family incomein 1975dollars; henceto calculate the wife's
property income, one must subtract the product of WW and WHRSfrom
FAMINC. MTRis the wife’s marginal tax rate evaluated if her hours of work
were zero. MTRis taken from published federal tax tables (it excludes state
andlocal incometaxes but includes any applicable social security benefits).
WMED isthe wife’s mother’s years of schooling, and WFED is the wite’s
father’s years of schooling. UN is the unemployment rate in the county of
residence,in percentage points, while CIT is a dummy variable that equals |
if the family lives in a large city (a Standard Metropolitan Statistical Area,
SMSA);otherwise,it equals zero. Finally, AX is the wile’s previous labor mar-
ket experience, in years. Further details on these variables are given in the
he oe file in the subdirectory CHAPII.DAT and in Mroz
EXERCISE 1: Inspecting Mroz’s 1975 Panel Study of Income
Dynamics Data
The purpose of this exercise is to help you become familiar with salient fes-
tures of the data series in the MROZdata file. While exploring this data set,
you will compute arithmetic means, standard deviations, and minimum and
652 Whether and How Much Women Work for Pay
maximum values for variables in the entire sample and for the data sorted
into various subgroups. You will also construct and save two variables that
will be used in the subsequent exercises ofthis chapter.
(a) To check whether the data are the same as those employed by Mroz
{1987, Table 111, p. 769], compute and print outthe arithmetic means
and standard deviations ofeach of the 19 variables in the MROZ data
file, using the entire sample of 753 observations. The results that you
obtain from sucha calculation should equal (for each variable named,
followed with parentheses enclosing first the mean and then the stan-
dard deviation): LFP (0.56839, 0.49563), WHRS (740.57636,
871.31422), KL6 (0.23772, 0.52396), K618 (1.35325, 1.31987), WA
(42,53785, 8.07257), WE (12.28685, 2.28025), WW (2.37457, 3.24183),
RPWG (1.84973, 2.41989), HHRS (2267.27092, 595.56665), HA
(45.12085, 8.05879), HE (12.49137, 3.02080), HW (7.48218, 4.23056),
FAMINC (23080.59495, 12190.20203), MTR (0.67886, 0.08350),
WMED (9.25100, 3.36747), WFED (8.80876, 3.57229}, UN (8.62351,
3.11493), CIT (0.64276, 0.47950), and AX (10.63081, 8.06913). Do
your results match those of Mroz? (Because ofdiffering rounding con-
ventions, your computer software program might generateslightly dif
ferent valucs. But the means and standard deviations should be very
close to those reported here.) Also compute minimum and maximum
values for each of these variables; this is a particularly useful practice,
since by doing this, one can often spot data coding errors. Are any of
your min-maxvalues “suspicious? Why or why not?
(b) Now comparethe sample of working women(thefirst 428 observations
in the MROZ data file) with those not working for pay in [975 (thefinal
325 observations). Compute arithmetic means and standard deviations
for cach of the 19 variables in these two subsamples, then print and
compare them. Mroz notes that arithmetic means in the working and
nonworking samples are quite similac for variables such as WA, WE,
K618, HA, HE, and HHRS.Doyoualso find this to be the case? How-
ever, meansin the working and nonworking samples tendto differ more
for variables such as KL6 and HW. Howdotheydiffer, and what might
this imply concerning the reservation wages of women in these two sam-
ples? Why? To the extent that the previous labor market experience of
women (AX) reflects their preferences or tastes for market work, one
might expect means of AX to differ between the working and nonwork-
ing samples. Is this the case? Interpret this difference. Are there any
otherdifferences in meansofvariablesin the two subsamples that might
affect labor force participation or hours worked? Ef so, comment on
them.
(c) In the model of labor supply estimated by Mroz,it is assumed that in
making her labor supply decisions the wife takes as given the house-
hold’s entire nonlabor income plus her husband’s labor income. Mroz
Exercises 653
calls this sum the wife’s property income and computes it as total family
income minusthe labor income earned by the wife. For the entire sam-
ple of 753 observations, computethis property incomevariable (named,
say, PRIN) as PRIN = FAMINC — (WHRS - WW).Also calculate and
print its mean and standard deviation; these should equal 20129 and
11635, respectively. Save PRIN for use in subsequent exercises ofthis
chapter.
d One ofthe variables that is often employed in empirical analyses of
laborforce participation is the wage rate. As was emphasizedearlier in
this chapter, however, the wage rateis typically not observed for women
whoare not working. Some analysts have attempted to deal with this
problem (albeit in an unsatisfactory manner—see Section 1 (.3.B.3) by
estimating a wage determination equation using data on workers only
and then using the resulting parameterestimates and characteristics of
the nonworking sample to construct fitted or predicted wages for each
ofthe nonworkers.
Restricting your sample to workers (the first 428 observations},
take the natural logarithm of the wife’s wagerate variable WW andcall
this log-transformed variable LWW. Compute and print out the mean
and standard deviation of LWW for this sample. Then, for the entire
sample of 753 observations, construct the square of the wile’s experience
variable and call it AX2, thatis, generate AX2 = AX*AX< (and, for later
use, the squareof the wife’s age, WA2 = WA*WA), Next, following the
humancapitalliterature on wage determination summarized in Chapter
5 ofthis book and using only the 428 observations from the working
sample, estimate by OLS a typical wage determination equation in
which LWWis regressed on a constant term, WA, WE,CIT, AX, and
AX2. Does this equation make sense? Why or why not? Then use the
parameterestimates from this equation and values of the WA, WE, CIT,
AX, and AX2 variables for the 325 women in the nonworking sample
1o generate the predicted orfitted log-wage for the nonworkers. Call this
fitted log-wage variable for the nonworkers FLWW, Computethe arith-
metic mean and standard deviation of FLWW for the nonworkers, and
compare them to those for LWW from the working sample. Is the dif=
ference in means substantial? How do you interpret this result? Finally,
fortheentire sample of 753 observations andfor use in subsequent exer-
cisesof this chapter, generate a variable called LWW for whichthefirst
428 observations(the working sample) LWWI = LWW andfor which
the last 325 observations (the nonworking sample) LWWI = FLWW
from above. Note that your constructed LWW variable should include
either the actual or a predicted wagefor each individual in the sample.
To ensure that you have computed the data series LWW! correctly,
compute and print out its mean and standard deviation; they should.
equal 1.10432 and 0.58268,respectively. Save the LWW data series for
use in subsequent exercises ofthis chapter,
654 Whether and How Much Women Workfor Pay
fa)
EXERCISE 2: Estimating the Hours Worked Equation Using Procedure |
The purpose ofthis exercise is to introduce you to a common, but unfortu-
nately inappropriate, procedure for estimating the hours worked equation.
Specifically, in this exercise you implement Procedure 1, in which youesti-
mate by OLS an hours worked equation using the entire sample of 753 obser-
valions and predicted wages for nonworkers and you set hours worked for
nonworkers equal to zero. Such an equationis often called a truncated nermal
regression equation. You also compute the implied responses to changes in
the wage rate and in property income, both in level and in elasticity form.
{a) Inspect the WHRSvariable (wife's hours worked) and verify that when-
ever the LFP(labor force participation) variable equals zero, the value
of WHRSis also zero. Then,using the OLS estimation procedure and
theentire set of 753 observations in the MROZ data file, regress WHRS
on a constant term and on the variables KL6, K618, WA, WE, LWW!
(constructed in part (d) of Exercise 1), and PRIN (constructed in part
(c) of Exercise 1). Do thesigns of the estimated parameters agree with
your intuition? Why or why not? Whatis the value of R*? Why might
this value be so low when Procedure I is employed?
(b) Using the above OLS parameter estimates and theelasticity formulae
underneath Eq. (11.48) evaluated at the same points as noted at the bot-
tom of Table 11.2, computetheelasticity of hours worked with respect
to wages and with respect to property income. Is the wageelasticity a
(b)
compensated or an uncompensated one? Why? Then, following Mroz’s
procedures, which are also discussed underneath Eq. (11.48), compute
the implied response of hours worked to a $1 change in the wage rate,
evaluated at the same point. How doesthis estimate compare with those
presented in Table 11.2? Finally, compute the implied response of hours
worked te a $1000 increase in property income. Camment on how this
estimate compares with those presented in Table 11.2.
(c} Although this Procedure | is simple and easy to implement, it has a
numberofserious flaws. Whatare the principal shortcomings?
EXERCISE 3: Comparing OLS, Problit, and Logit Estimates of the
LaborForce Participation Decision
Thegoal of this exercise is to help you gain experience with simple limited
dependentvariable estimation techniques by computing and then comparing
OLS, probit, and logit estimates of a typical labor force participation equa-
tion, The numerical comparison of these various estimators is based in large
part on the work of Takeshi Amemiya [1981], who has derived relationships
among them.”
Exercises 655
Econometrics textbooks typically point out that if the dependent vari-
able in an equation is a dichotomous dummy variable, and if an equa-
tionis estimated by OLS in which this dependentvariableis related lin-
carly to an intercept term, a numberof regressors, and a stochastic crror
term, then the resulting equation (often called a linear probability
model) suffers from at least two defects: (1) thefitted values are not con-
fined to the 0-1 interval, and so their interpretation as probabilities is
inappropriate; and (2) the residuals from such an equation are hetero-
skedastic. Note that in our context, LFP is such a dichotomous depen-
dent variable.
Using OLS estimation procedures and the MROZ data file for all
753 observations, estimate parameters of a linear probability model in
which LFP is related linearly to an intercept term, the LWW1 variable
constructed in part (d) of Exercise 1, KL6. K618, WA. WE, UN, CIT,
thewife's property income PRIN (constructed in part (c) of Exercise 1),
and a stochastic error term. Do signsof these OLS estimated parameters
make sense? Why or why not? Comment on the appropriateness of
using the OLS estimated standard errors to conducttests of sta 1
significance. Next retcieve and print outthe fitted values from thisesti-
mated linear probability model. For how manyobservationsare the fit-
ted values negative? For how many are they greater than 1? Why does
this complicate the interpretation of this model? What is &? in this
model? Does it have any useful interpretation? Why or why not?
Onepossible estimation procedure that is more appropriate when the
dependent variable is dichotomousis based on the assumption that the
cumulative distribution of the stochastic disturbances is the logistic;
the resulting maximum likelihood estimatoris usually called logit.”
With LFP as the dependent variable and with a constant term,
LWWI (from part (d) of Exercise 1). KL6. K618, WA, WE, UN, CIT,
and PRIN (from part (c} of Exercise 1) as eaplanatoryvariables, use the
entire sample of 753 observations in the MROZdatafile and estimate
Parameters based on a logit maximum likelihood procedure. Do the
signs of these estimated logit parameters make sense? Why or why not?
Whichof the estimated logit parameters is significantly different from
zero? interpret. Does the nonlinear logit computational algorithm in
your computer software program reach convergence quickly, after only
a few iterations(say, fewer than five)? Some computer programs provide
“goodness-of-fit” output for the estimated logit model, such as a pseudo-
R? measure or a measure indicating what percent ofthe predictions are
“correct.” Check your computer output and manual!for the interpreta-
tion of any such goodness-of-fit measures.” Finally, compare your logit
estimates to the OLS orlinear probability mode! estimates from part (a).
In particular,following Takeshi Amemiya [1981], each of the OLS slope
Parameterestimates should approximately equal 0.25 times the corre-
sponding logit slope parameter estimate. How well does Amemiya’s
656 Whether and How Much Women Work for Pay
approximation work in this sample? Further, Amemiya shows that each
of the OLS intercept and OLS dtunmyvariable imtercept terms should
approximately equal 0.25 times the correspondinglogit parameteresti-
mate, plus 0.5. Do your OLSandlogit intercept and dummy variable
intercept terms correspond well with Amemiya’s approximation?
«c) Another commonestimation procedure thatis used in the estimation of
dichotomous dependent variable models is based on the assumption
that the cumulative distribution of the disturbances is normal; this is
usually called the probit model. Likelihood functions based on the
probit model have been discussed in Section 11.3.B.1 of this chapter.
With LFP as the dependent variable and with a constant term,
LWW1 (from part (d) of Exercise 1), KL6, K618, WA, WE, UN, CIT,
and PRIN (from part (c) of Exercise 1) as explanatory variables, use the
entire sample of 753 observations in the MROZdata file and estimate
parameters based on a probit maximum likelihood procedure. Do the
signs of these estimated probit parameters make sense? Why? Which of
the estimated probit parameters is significantly different (rom zero?
Why? Does the probit computational algorithm in your computer soft-
ware program reach convergence quickly, after onlya fewiterations? Is
convergence more orless rapid than with the logit model? As in part (b),
interpret any goodness-of-fit measures that are provided as output by
your computer software program.
{d) Because the cumulative normaldistribution and thelogistic distribution
are very close to each other, in most cases thelogit and probit estimated
models will be quite similar.” Are the sample maximized log-likelinoods
in your estimated logit and probit models similar? Which is larger?
Whataboutthe signs and statisticalsignificance ofthe estimated param-
eters—are they similar? The estimatedeffect of a change in a regressor
on the probability of participating in the labor force. aP/AX,, is equal to
Pe( — P)*B;, the logit model and fUP)*B», in the probit model, where
Pis the probability of LFP, 8,; and @,, are the estimated logit and probit
coefficients, respectively, on the fh explanatory variable, and /( P) is the
cumulative normal function corresponding to P.”* Evaluate these esti-
mated derivatives for the logit and probit models, using the sample
LEPR of 425/753 = 0.568 as an estimate of P and notingthat f(?) =
{(0.568) = 0.393. At this sample mean, are the estimated effects similar
for the logit and probit models? What happens if you evaluate these
effects at thetail of the distribution, such as at P = 0.9 where f(P) =
0.175?
(e) Sincethelogistic distribution has a variation of «7/3, whereas the vari-
ation from the probit modelis usually normalized to unity, one way to
compare the logit and probit estimates is to multiply cach of thelogit
estimates by Vile & 1.73205/3.14159 = 0.5513, and then compare
these transformed logit parameters to the actual probit estimates. Ame-
miya [1981] argues, however, that a better approximation emerges if
Exercises 657
one mullplies the logit parameter estimates by 0.625 and then com-
pares these transformed logit parameters to the probit estimates. Which
of these two approximations best transforms your logit estimates to
comparable probit estimates? Why?
(f) In this exercise we have compared OLS linear probability model, logit,
and probit estimates of a labor force participation equation and have
focused on numericalrelationships amongthe estimated parameters. As
was pointed out in Section 11.3.B.1, however, there are seriousstatisti-
cal problems with each of the particular OLS,togit, and probit proce-
dures that were employed in this exercise. What are these problems?
EXERCISE 4: Relating the Tobit and Conditional OLS Estimates
The purposeofthis exercise is to engage you in the estimation and interpre-
tation of a labor supply model based on the Tobit Procedure {Lf and to enrich
your understanding of how this Tobit model relates to the conditional OLS.
frameworkofProcedure II, You will also have the opportunityto implement
empirically the analytical results of Goldberger [1981], Greene [1981], and
McDonald and Moffitt [1980], which were discussed in Section 11.3.B.1.
(a) First, estimate a conditional OLS model of labor supply (Procedure [1).
In particular,restricting your sample to the women who worked for pay
in 1975 (thefirst 428 observations in the MROZdatafile), run a regres-
sion of WHRSon constant term, KL6, K618,. WA, WE, PRIN, and
LWW(sec part (d) of Exercise 1 for a discussion of LWW and PRIN).
Compare your results to those obtained by Mroz, reproduced in Eq.
(11.49). Your standard error estimates might differ from those reported
by Mroz,since his estimates are adjusted for heteroskedasticity using the
Halbert White [1980] robust estimation procedure. If your computer
software program permits, also compute the White robust standard
errors; your results should be very close to those reported by Mroz.
(b) In Section 11.3.B.1 it was pointed out that these conditional OLS esti-
mates are biased estimates of the parameters in Eq. (11.33). Whyare
theybiased? Following Goldberger and Greene, computeconsistent esti-
mates of cach ofthe labor supply parameters using the LFP adjustment.
In particular, catculate the sample proportion of observations for which
WHRSis positive—in the MROZdatafile this is 428/753 = 0.568.
Then divide each ofthe conditional OLS parameterestimates from part
{a) by this proportion. According to Goldberger and Greene, these trans-
formed conditional OLS estimates are consistent estimatesof thelabor
supply parameters in Eq. (11.33). Are these transformed, consistentesti-
mates plausible? What can yousayregarding thestatistical significance
ofthese transformed parameters? Why?
(c} Now obtain consistent estimates using the Tobit estimation method
658 Whether and How Much Women Work for Pay
(Procedure II). In particular, using the Tobit maximum likelihood pro-
cedure {the sample likelihood function corresponding to Eq. (11.33) is
Eq. (11.32) with J, now equal to X,8 + iy,) and the same functional
form as in part (a), compute the Tobit parameterestimates and asymp-
totic standard errors, How do these estimates compare to the Goldberger
and Greene approximations from part (b)? What can you sayregarding,
thestatistical significance of these Tobit parameters?
(d) As was discussed in Section 11.3.B.1, McDonald and Moffitt have
shown (see Eq. (11.35)) that the total effect of the change in a regressor
on expected hours worked in the Tobit model can be decomposed into
two parts: the change in hours worked for those already working
weighted by the probability of workingplus the changein the probability
of working weighted by the expected value of hours worked for those
who work. Using the sample proportion of those who work (428/753 =
0.568) as an estimate of F(z), the Tobit parameter estimates from part
{c), and data evaluated at sample means,calculate 4 as outlined under-
neath Eq. (11.35). Note: For F(z) = 0.568, z = 0.175 and f(z) = 0.393.
Ofthetotal change in hours worked due to a $1 change in the wagerate,
what amountresults from changesin hours worked from those who are
already working? What amount comesfrom new entrants into the labor
force? Whatis the proportion ofthetotal effect on hours worked of a
changein any ofthe variables that derives from those women who arc
i a
already working?
Thelikelihood function (11.32) for the Tobit model indicates that the
(e)
Tobit can be envisaged as combining a probit modelof laborforce par-
ticipation with a standard regression model of hours worked for those
who work {see Section 11.3.B.1). One might therefore concludethatif
the sample were limited to those who work, Tobit estimates would be
numerically equivalent to OLS estimates. Would such a conclusion be
correct? Why or why not? Verify your intuition numerically, using data
in the MROZdata file, by estimating a Tobit model,restricting the sam-
ple to the first 428 observations(the workers) and-comparing these Tobit
estimates with the OLS estimates ofpart (a).
fn this exercise we have numerically related the Tobit and conditional
OLS estimates in a model oflabor supply. As was pointed out in Section
11.3.B.1, however,there are scriousstatistical problems with the Tobit
and conditional OLS procedures that were employed in this exercise.
Whatare these problems?
EXERCISE 5: Identifying Parameters In a Reduced Form Estimation
The purpose ofthis exercise is to explore issues of identification of the struc-
tural parameters when the reduced form Procedure [V method is imple-
mented empirically. In essence, this procedure involves first estimating a wage
Exercises 659
equation using a sample of workers only, then using the entire sample of
observations with hours of nonworkers set to zero to estimate a reduced form
hours worked equation derived from a reservation wage relationship, and
finally, identifying the structural parameters ofthe reservation wage equation
by using the reduced form and the wage equation parameterestimates. As we
shall see, the reservation wage equation is underidentified,just identified, or
overidentified depending on whetherzero, one, or more than oneregressorin
the wage equation is excluded from the reservation wage equation.
(a) Retrieve from Exercise 1, part (d), the parameterestimates from the
OLS wage determination equation in which LWW wasregressed on a
constant, WA, WE, AX, WA2, and CIT and the sample was confined
to those who worked (the first 428 observations). This corresponds to
the equation, with / subscripts deleted for simplicity,
LWW = g. + WA + g, WA2 + g, WE
+ g CIT + g, AX + ew (11.61)
where the g's are OLS parameter estimates and ey. is the OLS residual.
(b) Next specify three alternativestructuralreservation wage equations, and
then using a transform of Heckman’s proportionalityrelation (11.42),
write out their corresponding reduced form representations. Specifically,
three possible structural representations for the log of the reservation
wage (L.WR) equation, analogousto Eq. (11.41). are
Just identified:
LWR = ay + a, WA + a WA2 + a, WE + a, CIT
+ ds KL6 + a, K618 + a, PRIN + a, UN + (11.62)
Overidentified:
LWR = by + b, WA + b, WA2 + b, WE + b, KL6
+ b, K618 + 6, PRIN + 4,UN + ¢ (11.63)
Underidentified:
LWR = @ + ¢, WA + ¢, WA2 +c WE + CIT + 6, AX
+ o KL6 + ¢, K618 + & PRIN + 6 UN + ¢ (11.64)
where PRIN is the property incomevariable construcied in part {c) of
Exercise { and « is a random disturbance. Note that in Eqs. {11.62},
a 1.63), and (11.64) there are one (AX), two (AX and CIT), and zero
variables, respectively, included in the LWW equation (11.61) but
excluded from the LWR equation. For each of these three reservation
wage cquations, first employ a log-transformation of Heckman’'s pro-
portionality relation (£1.42),
H, = d(LWW, — LWR,) iff LWW, > LWR, (11.65)
H,=0 iff LWW, < LWR, (11.65b)

60 Whether and How Much Women Work for Pay
then substitute into Eqs. (11.65) for LWW the OLS estimated wage
determination equation (11.61) and for LWR one of the LWR specifi-
cations in Eqs. (11.62)-(11.64), Having done this, you should bave
obtained three alternative reduced form equations for hours worked,
each in terms of the underlying structural parameters, exogenous vari-
ables, and disturbances.
(ec) Using the entire sample of 753 observations in the MROZ data file
andsetting hours of nonworkers to zero, estimate by OLS the reduced
form equation corresponding to the just identified equation (11.62)
derived in part {b), using the g estimates from Eq. (11.61) to solve
for the @ parameters in Eq. (11.62). This procedure is often called
indirect least squares, Do these structural estimates of the a param-
eters make sense? Why or why not? What about your estimate of
d? Next, employ a more direct estimation procedure. Specifically.
using your reduced form equation corresponding to Eq. (1 1.62),
construct transformed regressors variables as the product of the
original regressors in Eq. (11.61) and their estimated g coetticients,
and then estimate by OLS the resulting equation with these trans-
formed variables as regressors. Verify that you obtain the same
structural estimates of the @ and d parameters as you did with the
indirect least squares procedure. Are the standard error estimates
from this direct procedure appropriate for doing inference? Why or
why not? . puter software programsofferthis calculation as an optional command:
(a) To understand complicationsthat arise when the reservation wage equa-
tion is over-identified,first show thatin the reduced form equation cor-
responding to Eq, (11.63) derived in part (b) 2 parameter restriction
must be imposed so that a unique estimateof d is obtained. Using the
entire sample of 753 observations in the MROZ data file and setting
hours of nonworkers to zero, impose this parameterrestriction, estimate
by OLS the constrained, overidentified reduced form equation. and
solvefor the b parameters in Eq. (11.63) and for d. Do these structural
estimates of the # and ¢ parameters make sense? Why or why not?
Derive, interpret, and then implement empirically a test for the over-
identified model (11.63) as a special case of the just identified model
11.62).
cS Show at if you were to estimate by OLS the reduced form equation
corresponding to the underidentified reservation wage equation a 1.64)
derived in part (b), there is no way that you can obtain unique estimates
of the ¢ and ¢ structural parameters. If you nonetheless estimated this
feduced form equation by OLS, what woutd the R? be relative to that
from the just identified reduced form estimation ofpart (c)? Why does
this occur?
(fy There is a serious drawback with the Procedure IV estimation you have
donein this exercise. Whatis this problem?
Exercises 661
EXERCISE 46: Implementing the Heckit Generalized Tobit Estimator
The purpose of Exercise 6 is io have you implement and interpret the Heckit
multistage Procedure VIIL. This is accomplished by having you replicate
results reported by Mroz (1987]. Youalso will compare the Heckit sample
selectivity procedure to an OLS-based method due to Olsen [1980].
(a) In the first stage of the Heckit procedure, one estimates a probit LFP
equation and retrieves from this estimated equation the inverse Mills
ratio A. Mrozfirst generates a numberof polynomial transformationsof
the wife's age, education and experience variables to be used as explan-
atory variables in the LFP equation. Forthe entice sample of753 obser-
vations in the MROZdata file, following Mroz, generate AX2 o
AX#AX, WA? & WAsWA, WE2 & WE*WE, WA3 = WA2*WA, WE3
= WE2+WE, WAWE = WA*WE, WA2WE = WA2«WE, and WAWE2
= WA*WE2. With this sample, estimate by maximum likelihood a
probit model in which LFP is the dependent variable and the explana-
tory variables include 2 constant term, Ki.6, K618, WA, WE, WA2,
WE2, WAWE, WA3, WE3, WA2WE, WAWE2, WFED, WMED,UN.
CIT, and PRIN (this last variable was calculated in part (c) of Exercise
1). From this estimated probit model, calculate the inverse Mills ratio
for each observation,save this variable, and call it INVRI. (Some com-
for others, it must be computed bybrute force, using Eq. (11.37) and
values from the normal distribution.) Now redo the probit estimation,
this time adding the experience variables AX and AX2 and calling the
corresponding inverse Mills ratio values INVR2, Commenton the sta-
listical significance of parameters estimated in these two probit equa-
tions. Note that the LWW! wagevariable is excluded as an explanatory
variable in this probit model. Since econumic theory supgests that LFP
is affected by the wage rate, why is this LWW varisble excluded? In
what sense, however, might it be included indirectly?
{b) Next, restricting your sample to those who work for pay(the first 428
observations in MROZ). estimate by OLS a wage determination eqt
tion allowing for sample selectivity and compare results to an equation
that does not take into account the sampleselectivity. In particular, fol-
lowing Mroz, let LWW be a linear function of a constant term, KL6,
K618, WA, WE, WA2, WE2, WAWE, WA3, WE3, WA2WE, WAWE2,
WMED,WFED,UN, CIT, and PRIN. Callthis set of explanatoryvari-
ables "Set A,” Estimate this equation by OLS (if your software permits,
employ the White [1980] robust standard error procedure), and com-
ment onthe signs andstatistical significance of the estimated parame-
ters. Redo this OLS estimation, addingto Set A the experience variables
AX and AX2. Interpret any changes in results. Then, with the same 42%
162 Whether and How Much Women Workfor Pay
observations, estimate by OLS two wage determination equations that
allow for sampleselectivity:first, an LWW wage determination equation
with the Set A variables and the INVR1 inverse Mills ratio variable
(from part (a)) included as regressors and second, an LWW wage deter-
mination equation with the Set A variables included, but with the AX,
AX2 experience measures and the corresponding INVR2 variable also
included. Commenton the sensitivity of the estimated parameters to
inclusion of the experience variables and to the sample sclectivily adjust-
ment.Is sample selectivity significant? (You may use large sample dis-
tribution theory and the White {1980] robust standard error method to
conduct statistical inference.)
() Finally, restricting your sample to those who workfor pay {the first 428
observations in MROZ), use the fitted values from the wage determi-
nation equation in part (b) as instruments in a sample selectivity-
adjusted two-stage least squares estimation of the hours worked equa-
tion. Specifically, following Mroz, for a base case comparison, first
employ instrumental variable (IV) estimation (with the White robust
standard error procedure if it is available in your software} of an hours
worked equation in which WHRSis the dependent variable and the
explanatory variables include a constant term, KL6, K618, WA, WE,
LWW,and PRIN;call this set of explanatory variables “Set B.” fn this
IV or 2SLS estimation, treat LWW as an endogenousvariable, and use
the Set A variables defined in part (b} to form instruments. How do your
results compare with those reported by Mroz, reproduced in Eq.
(IL51)? (Note: Mroz’s standard error estimates employ the White
tobust standard error procedure. If your sofiware does not permit this,
your OLS standard error estimates will differ slightly from those in Eq.
(11.51). Next allow for sample selectivity, but exclude the experience
variables. Specifically, using the same Set A variables plus the INVRI
inverse Mills ratio variable to form the instrument for LWW,estimate
by IV or 2SLS an hours worked equation with the Set B variables as
fegressors (using robust standard error methods if possible), but with
INVR! added as a regressor. How do your results compare with those
reported by Mroz, given in Eq. (11,53)? Is sampleselectivitysignificant?
Whyor why not? (Vote: Mroz’s standard error estimates are based on a
formula derived in his Appendix—his estimates will difler slightly from
those of the White robust standard error procedure.) Then estimate a
model by 2SLS in which the experience variables AX and AX2 are
included along with the Set A variables in thefirst-stage wage determi-
nation equation, but in which sample selectivity is not taken into
account and only the Set B variables are regressors. Compare your
tesults with those of Mroz, reproduced in Eq. (11.50). Finally, include
the experience variables AX and AX2, the Set A variables, and the
INVR2 inverse Mills ratio as variables in forming the instrument for
LWW,andthenestimate by 2SLS the hours worked equation including
as repressors the Set B variables and INVR2. Your results should con-
s 663
form closely with these reported by Mroz, reproduced in Eq. (11.52). Is
sample selectivity important when the experience variables are treated
as exogenous? Why or why not?
{d) Interpret your findings in part (c), commenting in particular on the
importance of allowing for sampleselectivity and how this sensiti is
affected by the assumption of exogeneity of the experience variables.
Mroz concludes that with his preferred specifications the uncompen-
sated wage effect for his sample is smail, as is the estimated income
effect. Do you agree? Why or why not? (You might want to compare
results here with those from other studies, reported in Table 11,2.)
{e) Now compare the Heckit inverse Mills ratio procedure to the OLS-based
linear probability model method proposed by Olsen [1980]. In particu-
lar, using a mode! either with or without the AX and AX2 experience
variables, follow the Olsen procedure outlined in Section 11.3.B.1
undemeath Eq. (11,37) and use as a regressor in the wage rate deter-
mination and hours worked equations,instead ofthe inverse Mills ratio,
thefitted probability minus | from an OLS-estimated linear probability
model. Compare these results with those obtained from the Heckit pro-
cedure. Do they differ substantially?
Tn whatsense is the estimation procedure ofthis exercise a generalized
Tobit estimator?
EXERCISE 7: Incorporating Income Faxesinto a Modelof Labor
Supply
The purpose ofthis exercise is to enable you to assess the impacts of incor-
Pporating income taxes on your estimates of the wage and income responsive-
hess of labor supply. This will involve creating virtual income and after-tax
wagerate variables that are consistent with a linearized budget constraint
(LBC) budget constraint specification, estimating by using the Heckit gener-
wen procedure, and comparing results with those reported by Mroz
1 .
@ Ourfirst task is to create several tax-related variables. For the entire
sample of 753 observations in the MROZdaia file,first create and save
a virtual property income variable defined as VPRIN = (1 —
MTR)*PRIN, where PRIN was created in part (c) of Exercise { and
MTRis the marginaltax rate variable provided in the MROZdata file.
Next, generate and save LTAX = LOG(I— MTR) andthelogarithm of
an after-tax wage variable as LTWW = LTAX + LWW1, where LWWI
is the wagevariable created in Exercise J.
(b) To implementthe Heckit multistage procedure thatallows for sample
selectivity, we must initially estimate a probit LFP equation for the
entire sample of 753 observations and compute from this estimation the
Heckman inverse Mills ratio. Following Mroz,first generate polynomial
364 Whether and How Much Women Work for Pay
transformations ofthe wife's age, education, and experience variables,
10 be used as explanatory variables in the LFP cquation. In particular,
for the entire sample of 753 observations in the MROZdata file, gen-
erate and save WA2 = WA*WA, WE2 = WE*WE, WA3 = WA2=WA,
WE3 = WE2*WE, WAWE = WA*WE, WA2WE = WA2*WE, and
WAWE2 = WA*WE2. With this sample, estimate by maximum likeli-
hood a probit model in which LFP is the dependent variable and the
explanatory variables include a constant term, KL6, K618, WA,
WE,
WA2, WE2, WAWE, WA3, WE3, WA2WE, WAWE2. WFED,
WMED,UN, CIT, and PRIN (thislast variable was calculated in part
(c) of Exercise 1), From this estimated probit model, calculate the
inverse Mills ratio for each observation,save this variable, and call it
INVR. (Some computer software programs offer this calculation
as an
optional command; for others it must be computed bybrute force using
Eq. (11.37) and values from the normal distribution.) . WHRS,is a proportion ¢of the difference between LWWand LWR iif
(c} Nowestimate a model with ae included , similart o walPre sentedi n
. (11.55), by 2SLS, Specifically, using as exogenous variables a - (11,65a) and (1).65b). Finally, analytically substitute into the hours
aot cm KL6, K618, WA, WE, WA2, WE2, WAWE, wa3, WE3,
WA2WE, WAWE2, WMED, WFED,UN, CIT, PRIN and the inverse
Mills ratio variable INVR from part (b), form an instrument for the log
of the after-tax wage rate variable LTWW (created in part (a).
With
these instruments for LTWW, estimat e by 2SLS a model in which
WHRSis a linear function of a constant term, LTWW, LTAX, KL6,
K618, WA, WE, PRIN, VPRIN, INVR, and a random disturbance
term. If your software permits, obtain the White robust standard ertor
estimates. Comment on the sign and magnitude of the estimated wage,
property income, and tax coefficients. Using the information given
underneath Eq. (11.35) and your estimated variance-covariance matrix
of the estimated coefficients, formulate and test the hypothesisthat
women optimally take income taxcs into account when making
labor
supply decisions. Then formulate and test the bypothesis tha women
entirely ignore income taxes when making labor supply decisions, Inter-
reported by
pret your test results. Do your conclusions agree with those
Mroz and by Rosen [1976]? Why or why not? | tions. Mroz reports that whenhe estimated a model similar 10 this, he
{d) Experiment with any two other plausible specifications for the probit,
wage determination, and/or hours worked equations with laxes
included, and commentonthesensitivity of your results to changes in
taxes
specification. Do Mroz’s results on the unimportance of income
appear to be robust?
EXERCISE 8: Specifying and Estimating an Extended Tobit Modei
us
In thisexercise you will specify and estimate an extended Tobitsimultanco
equations model in which paramete rs of the wage fate, reservati on wage {and
ly, hours of work), and labor force participation equations are estl-
Exercises 665
mated simultaneously by using the method of full information maximum
likelihood (FIML), with cross-equation parameter constraints imposed. As
wasnoted in Section 11.3.B.1, Killingsworth [1983] hascalled this Procedure
Vi.
(a) We begin by specifying the wage equation for workers only, where
LWWisspecified to be a linear function ofa constant term, WA, WA2,
WE, CIT, AX,and a random disturbance term ex, as in Eqs. (11.38)
and (11,61), Next we specify the functional formofthe reservation wage
equation derived from the marginalrate ofsubstitution function. In par
ticular, let the logarithm of the reservation wage equation be LWR and
tet it have the functional form as in the overidentified equation (11.63)
but denote the disturbance term in Eq. (11.63) by ex, as in Eq. (141),
Finally, following Heckman [1974b], specify that hours worked,
LWw, > LWR,andis zero iff LWW, < LWR, analogous to Eqs.
worked equations (11.65a) and (11.65b) your specifications for the
LWW and LWR equations. and note that in Eq. (11.65a) the distur-
bance term. now denoted dep,is equal to dep, = dew, — em). which is
similar to Eq. (11.45a) but with dreplacing b.
(by) Now assume that ey, and eg, are joint normally distributed with vari-
ances oj. and o} and covariance ows. This implies that in the hours
worked equation that you constructed in part (a) the disturbance term
€p, is normally distributed with variance oj, = 6% + o) — 2owe and is
joint normallydistributed with e», having covariance guy = ah ~ one.
Given these distributional assumptions, write out thelikelihood func-
tion for the entire sample of 753 observations, analogous to Eq. (11.47).
where d now replaces and othertermsare as defined by Eq. (11.47).
(c) Using the appropriate computersoftware and the 753 observations from
the MROZdatafile aswell as thelikelihood function constructed in part
(b), estimate by maximum likelihood the parameters appearing in the
extended Tobit model consisting of the wage and hours worked equa-
obtained a large positive and statistically significant coefficient on LWW
but a negative and marginally significant coefficient on PRIN. Are your
results qualitatively similar to those of Mroz?
{d) Finally, following Mroz,specify an alternative modelin which the expe-
rience variable AX is excluded from the wagerate determination equa-
tion, Asin parts (a) and (b). write out the correspondinglikelihood func-
tion, and then estimate parameters using the FIML method. Compare
yourresults to those obtained in pact (c). Mroz found that when the AX
experience variable was omitted from the wage determination equation,
the estimated FIML coefficient on LWWin the hours worked equation
was positive but much smaller than when AX was included (and statis-
tically insignificantly different from zero). However, the estimated coef
666 Whether and How Much Women Work for Pay
ficient on PRIN was negative andstatistically significant, but smaller in
absolute value than when the AX variable was included in Eq. (11.61).
Do your results conform with those of Mroz? Why or why not?
The mode!specified and estimated in this exercise is based implicitly on
an important assumption concerning the continuity of the hours
worked relationship. Whatis this assumption, and under whatpractical
considerations might it be violated? What estimation procedures are
available that do not require this strong assumption?
CHAPTER NOTES
. These data are taken from the 1989 Economic Report ofthe President, Washing-
ton, D.C.: U.S. Government Printing Office, Table B-36,p. 349.
. For a detailed discussionoffactors affecting the timeseries growth ofthe female
jabor force supply, see James P. Smith and Michael Ward [1985] and the refer-
ences cited therein.
. Reasonsfor this decline are discussed and surveyed by, amongothers, John Pen-
cavel (1986).
|. To a first approximation the effect of truncation occurs when sample data are
drawn from a subset of'alarger populationof interest (e.g., those in the laborforce
are selected from those in the population as a whole), whereas censoring occurs
whenvalues over a range of a variable are reported simply as beingless than (or
greater than) some given value(e.g., the unobserved reservation wage is less than
the market wage facing an individual). Unlike truncation, censoringis essentially
a defect in the sample data.
Fora survey ofissues concerning the labor supply of maies, see John Pencavel
”
1986}.
Veeever, see theclassic study by P. Sargant Florence (1924).
as
For a more detailed discussion of measurementissues, see, for example, Harold
‘Watts et al. [1977] and Shirley J. Smith [1983].
Detailed data on U.S, government household surveys are found in two U.S.
Bureau of LaborStatistics publications, an occasional one called Handbook of
Labor Statistics, Technical Notes, and the monthlypublication called Employ
ment and Earnings. International comparisonsof laborforce participation and
international data sources are discussed in, among others, Constance Sorrentino
11983]. The discussion that follows is based in farge part on Danie) Hamermesh
and Albert Rees [1988], Chapter 1.
Issues involved in assessing factors that affect the supply of laborto the U.S. mil-
2
itary are discussed in, among others, Cotin Ash, Bernard Udis, and Robert F.
McNown [1983].
. These figures are taken from Hamermesh and Rees {1988, Table 2.5}.
. Another,less commonly used data set is one that focuses particular attention on
the variety of sources of income.I is the U.S. Bureau of Census Survey ofIncome
and Program Participation, undertaken as a joint project with the U.S. Depart-
ment of Health and Human Services. For a discussion ofthis data set, see U.S.
Department of Commerce, Bureau of the Census [1987].
. For additionaldiscussion, see Daniel S. Hamermesh [1990].
Notes 667
. Descriptionsofthe data gathered at Ohio State are found in Parnes et al. [1970],
white those ofdata collected at the University of Michigan are presented in Mor.
gan et al, [1966, 1974].
}. Fora classic presentation of the individual labor supply model, sce H. Gregg Lewis
(1957),
}. This reservation wage notionis apparently ducto Jacob Mincer[1963]. Important
extensions and applications include those of Reuben Gronau{1973b] and James
Heckman (1974b};also sce Gronau { 1986).
. The expressions in Eqs. (11.12}-(11.14) implicitly assumethat interior solutions
occurboth before and after the small change in P,.
- This choice of nomenclature is somewhat unfortunate, since while chauvinism is
clearly undesirable, this particular model might or might not be a reasonable
description of houschold decision making.
. For examples ofthe use of the male chauvinist model, see, amongothers, Bowen
and Finegan [1965, 1969] and Hausman (1981a).
. Applicationsofthis model include Marvin Kosters [1966]: Malcolm Cohen, Sam
Rea, and Robert Lerman [1970}; Robert Hail [1973]: Orley Ashenfelter and
James Heckman [1974]; and Jerry Hausman and Paul Ruud (1984).
This duopoly type of model of labor supply has been implemented by J. S. Ash-
worth and David T. Utph (1981).
. Fora discussionof such indirect incomeeffects, see Killingsworth (1983, pp. 36-
38}.
|. The added worker hypothesis appears to have been first pointed out by Wladimir
S. Woytinsky[1940].
‘Onthis, see, for example, Mincer [1962, 1966] and Belton M. Fleisher and George
P. Rhodes, Jr. [1976].
}. Ronald G. Ehrenberg and Robert S. Smith [1985. p. 204] note that if one incor-
porated estimates of discouraged workers into official unemployment rale statis-
tics, thea the U.S. unemployment rate in 1973, for example. would have increased
from 4.9 to 5.6%. There is some evidence, however,that for a substantial aumber
ofdiscouraged workers the attachment to the laborforce is loose, and soit is not
clear how one should interpret the welfare consequences of the discouraged
workereffect; see, for example, Pau! Flaim [1984]. For this and other reasons.
considerable controversy exists concerning whether published unemployment
rates should include estimates of discouraged workers. A usefuldiscussionofthis
set of issues is found in the final report of the National Commission on Employ-
ment and UnemploymentStatistics (1979. pp. 44-49].
. The discussion that follows is based in large part on Addison and Sieben {1979
pp. 79-85).
. For a discussion of such family economics issues and additional references, see
Gary S. Becker (1981, 1988).
. A partial list of studies includes the theoretical and econometric analyses by Mare
tin Browninget al. (1985), Kim B. Clark and Lawrence H. Summers [1982], Gil-
bert R. Ghez and Gary S. Becker (1975), James J, Heckman [1978], James J.
Heckman and Thomas E. MaCurdy(1980, 1986], Richard Layard and Jacob
Mincer [1985], Thomas E, MaCurdy [1981], Robert A. Moffitt [1984}, Alice Nak-
amura and Masao Nakamura [1985a, b], and T. Paul Schultz [1980].
28, Other earlystudies cstimating the effectsoffixed costs on labor supply include
those by Giora Hanoch (1976, 1980] andJerry A. Hausman (1980).
368 Whether and How Much Women Work for Pay
29. Thisclassification scheme was actually developed earlier by James J. Heckman,
Mark R. Killingsworth, and Thomas &, MaCurdy [1981].
30. Sce, for example, Robert E. Hall (1973).
B. This procedure is foliowed by cach of the authors in Cain and Watts [1973b],
except for Hall (1973].
32. Heckman,Killingsworth, and MaCurdy [198t, pp. 83-84),
33. Whenthis type of modelis implemented using paneldata, it is often assumed that
although ¢varies across people at anygiven point in time, for any oneperson it
is fixed overtime. A discussion of econometric techniques that exploit panel data
is beyondthe scope ofthis chapter: see, however, Zvi Griliches, Bronwyn H.Hall,
and Jerry A. Hausman [1978]. Griliches and Hausman [1986}; and Cheng Hsiao
[1986].
<3 Recall that the first-order conditionsforut maximization at an interior solu-
tion implythat the marginal rate of substitution Af equals the real wage rate 1’,
that is, that the slope of the indifference curve equals the slope ofthe budgetline.
35. Probit analysis has a long history in biometrics; sce, for example, David J. Finney
[1947] and Jerome Corfield and Nathan Mantel [1950}. For early discussions in
the context of econometrics, see James Tobin {1955] and Arthur S. Goldberger
[1964, pp. 248-251].
36. Forfurtherdiscussion, see, itter afia, Tobin (1958), Amemiya {1973}, and Mad-
dala [1983, Chapter 6]. The etymology of Tobitis unclear, although Goldberger
[1964, p. 253] refers to it as Tobin's probit. Maddala (1983, p. 151] suggests that
there is presumably no connection with the prophet Tobit or the book of Tobit in
the Apocrypha. It is known that James Tobin achieved a distinguished record in
the U.S. Navy during World Waril, including anti-submarineduty, Further, Kil-
lingsworth (1983, fo. 7, p. 142) notesthat the novelist Herman Wouk wasa class-
mate of Tobin's in naval officer training school and that in his best-seller The
Woukcreated a midshipman character named “Tobit.”
. Computationalissues concerning estimation of the Tobit model have been con-
sidered by. amongothers, Richard N. Rosett and Forrest D. Nelson [1975] and
RayFair [1977].
. For a discussion of more general utitity functions and their implied consumer
demandand labor supply equations, see Arthur S. Goldberger [1987].
. McDonald and Moffitt emphasize that 4 will equal | only if X is infinity, in which
case F(z) = 1 and f(z) = 0. ‘ and Masao Nakamura [1981b; 1985a, bj.
. In practi . this is typically done byevaluating the expressions at sample means
ofthe variables and using estimated parameters.
- Forfurtherdiscussion, see Olsen [ 1980). and for a generalization to accommodate
other distributions, see Lung-Fei Lec [1982].
. For further discussion ofissues concerning the functional form ofthe wage deter
mination equation (11.38), see Chapter 5 ofthis book.
43, See Killingsworth [1983, Table 4.1, p. 15t] and Killingsworth and Heckman.
(1986, Table 2.26,p, 192] for a very brief tabular summary ofthese cight methods.
44, Oneofthe first statements of this sample selectivity problem in the context of
laborsupply is found in Reuben Gronau [1973a).
45. In practice, this factor of proportionality often applics to the differencein the nat-
ural logarithms of W, and ¥,,
46. Forfurtherdiscussion of identification issues, see Killingsworth [1983, pp. 154~
156, especially footnotes 14 and 17) and Olsen [1980].
dustin
&
Notes 669
47, There are other important problemsin analyzing thelabor supplyeffects oftaxes.
To determine the appropriate marginal after-tax wage rate, researchers typically
must employpublished statutory tax cate information. However, actual marginal
after-tax wage rates may differ substantially from statutoryrates, owing to com-
plex exemptionsand deductible expense provisions, underreporting or cunceal-
mentofincome by individuals, and/orto discretionary administrative behavior
bytax-collecting agencies. With the exception of William Gould [1979], hardly
any empirical work has been doneto date on the labor supply implicationsoftax
avoidance and tax evasion; however, for theoretical discussiuns ofthis Topi, see
Jonathan C. Baldry [1979]. Ame J. Isachsen and Steiner Strom [1980], Agoar
Sandmo {1981}, and James J. Heckman (1983). Further,it widely believed that
the available data on hours worked measure actualhours worked witherror, which
implies that the budget consiraint segmentfacing the individual might be misspec-
ified. This issue has been addressed by, among others. Burtless and Hausman
(1978), Hausman [1980], and Wates and Woodland [1980].
48. This is slightly beter than the cynical summaryoffered George Johnson (1976,
P. 107, fa. 9] concerning estimates of input substitutability in production: “In
Fecent years, economists have narrowedits value duwn to a ringe between zero
andinfinity.”
49. Wales and Woodkaind [1979] and Zabalza [1983] employ the CES direct utility
function, while Burttess and Hausman [1978] and Hausman (1980. 19819, 1983]
adopt a form that places a priori inequality (negative) restrictions on the totul
income elasticity. Ashworth and Ulph { 1981] restrict their sample to people who
work at least cight hours per week bul donot corrcct for the potent
50. Orley Ashenfelter and James J. Heckman [1974] consider a joint labor supply
mode! but do not incorporate the effects of taxation. For another attempt at
including tases in the family labor supply context, see Jerry A. Hausmanand Paul
Ruud {1984).
SI. Note that in countries such as Canada where spouses file separate income tax
returns, Unis marriage taais less of an issue. For further discussion, see Alice Nak-
amura and Masao Nakamura [198 la}.
$2. Hausman(19814, p. 63]. Also see Hausman [1986].
53. This conventional wisdom has also been persistently attacked by Alice Nakamurs.
34. A more complete survey offemale labor supply is presented in Table 2.26ofKil-
lingsworth and Heckman [1986, pp. 189-192].
55. Mroz [1987, p. 769].
56. See James Durbin [1954], De-Min Wu (1973), Jerry Hausman (1978). and Hale
bert White (198:
$7, That KL6,the number of children under age sin, and perhaps KG18, are exoge-
a
Rous is 4 somewhat surprising finding. One possibilityis that the power of the
Durbin-Wu-Hausman-White specification test may be low, Particularly when the
fit from the first-stage regressionsis low, For a discussion of such powerissues far
this test, see Nakamura and Nakamura { [985c}.
58. It is worth noting Uvat when the two experience variables are excluded fromthe
first-stage probit equation, the sample likelihood function falls from — 39% to
— 450, and ofthe 16 regressors, only KL6, and V haveestimated coefficients that
are more than twice their estimated standard errors.
70 Whether and How Much Women Work for Pay
59. Mrozalso reports thatthese findings are essentially unchanged whenthefirst-stage
probit equationis altered to reflect alternativedistributions such as thelogit or log
normal. Furthermore,tests for the exogencity of KL6, K618. and 1 uniformly
indicate nonrejectionof the null hypothesis of exogeneity.
. It is worth noting, however, thal use of the simple conditional mean or inverse
Mills ratio adjustment to account for sample selectivity is at best a “first-order
approximation” when taxes afe incorporated into the model; unless strong
assumptions are made similar to those specified by Hausman [1981a], the appro-
priate procedure is very complex and computationally cumbersome. Mroz’s
results concerningthe effects of taxes shouid therefore be viewed as lentative.
6h. Note the similarity of this frameworkto thatof progressive taxationin static labor
supply discussed in Section 11.3.B.2; in both cases the budget constraint set con-
sists of numerous segments havingdiffering slopes.
62. ion ofthis additive separability is that C(z) or L£(2) does notaffect the
marginal utility of Cir’) or £(°) for ¢ # 2". Further, if leisure is a normal good in
eachtimeperiod, then additive separability implies thatleisure times at different
points overthelife cycle must be net substitutes (holdinglifetimeutility constant).
Forfurtherdiscussion, sce Angus Deaton [1974].
63. Forfurtherdiscussion, see Martin Browning, Angus Deaton, and Margaret Irish
[1985] andthereferences cited therein.
}. See Harold Waits, Dale Poirier, and Chris Mallar [1977], among others, for 2
study implementing PO and Edward Kalachek, Wesley Mellow, and Fredric
Raines [1978]for PT.
65. Fordetails, see Killingsworth and Heckman [1986. pp. 159-161].
66. Fora generaltheoretical discussion on models ofendogenous dynamicwage deter-
mination, see Yoram Weiss [1986]. Recent econometric studies on various aspects
ofdynamic labor supply include Joseph Altonji [1986], Richard Blundell and Ian
Walker [1982]. James Heckman and Thomas MaCurdy[1980}, Richard Layard
and Jacob Mincer (1985), Thomas MaCurdy[1981], Robert Moffitt [1984], Alice
Nakamura and Masao Nakamura [1985a,b], and James P. Smith [1977, 1980].
67. The labor supply effects of such quantity restrictions have been considered by,
amongothers, Angus Deaton and John M. Muelibauer {1981}, Robert A. Moff
[1982], and John C. Ham [1982,1986].
68. For a survey of econometric problems with more general qualitative dependent
variable specifications, see, for example, Takeshi Amemiya [1981. 1984], Richard
Blundell [1987], James Heckman [1976b], G. S. Maddala [1983], and Daniel
McFadden [1984]. In the context offa laborforce participation studies, see
Jules Thecuwes [1981] for an application of the multinomiallogit specification.
69. Fordiscussions on inference from experimental data,sce, for example, Dennis J.
Aigner (1979}; Michael C. Keetcy [1981]; Michael C. Keeley, Philip K. Robins,
Roben G.Spiegelman, and Richard W. West [1978]; and Charles E. Metcalf
[1973}.
70. Paneldata issues are discussed by. amongothers, Zvi Griliches, Bronwyn H. Hall,
and Jerry A. Hausman [1978]; Gritiches and Hausman [1986]; and Cheng Hsiao
[1986).
Mm. Thanksare due to Thomas Mroz, who madethis data set available for our use.
72. Foran empirical comparison ofthe OLS,logit, and probit estimators in the con-
text of the laborforce participation decision, see Morley K. Gunderson [1980].
Gunderson [1980,p. 217] finds that, “Whenthe probability oflabor force partic-
References 671
ipation is .50, .. . , the results ofall three statistical techniques are similar. How-
ever, when the actual probability of panicipation is nearer to zero or onc. then the
probit orlogit results differ substantially from the finear probabilityfunction.”
73, Historical aspects ofthelogit specification in bioassay studies are discussed bricily
in Joseph Berkson [1951].
74. A discussion ofgoodness-of-fit measures in the context oflimited dependent vari«
able models is given in, for example, G. S. Maddala [1983. pp. 37-41, 76-77].
75, However, amongothers, Joseph Berkson [1951] has argued thatthelogit is pref
crable to the probit.
76. For a discussion andderivation, see G. S, Maddala (1983, pp. 23-24).
CHAPTER REFERENCES
Addison, John T, and W.Stanley Siebert [1979]. TheMarketfor Labor: An Analytical
Treatment, Santa Monica, Calif: Goodycar.
Aigner, Dennis J. [1979]. “A Brief Introduction to the Methodology of Optimal Exper-
imental Design,” Journal ofEconometrics, L121, September, 7
Allen, R. G. D. (1938), Afathematical Analysis for Economists, London: Mac!
Altonji, Joseph G. 11986), “Intertemporal Substitution in Labor Suppl
from Micro Data,” JournalofPolitical Economy, 94:3, Part 2. June, S176-S215.
Amemiya, Takeshi [1973]. “Regression Analysis When the Dependent Variable Is
Truncated Normal,” Econometrica, 41:6, November, 997-1017.
Amemiya, Takeshi [1975], “Qualitative Response Models,” Annats of Economic and
Suciul Measurement, 4:2, Summer, 363-372.
Amemiya, Takeshi [1981}, “Qualitative Response Models: A Survey,”Journal ofEco-
nomic Literature, 19:4, December, 1483~1536.
Amemiya, Takeshi, ed. [1984], “Censored or Truncated Regression Models,” Journal
ofEconometrics, 24:12. January/February, 1-222.
Ash, Colin, Bernard Udis, and Robert F. McNown {1983}. “Enlistments in the All-
Volunteer Force: A Military Personnel Supply Model and Its Forecasts,” tuneri-
can Economic Review, 73:1, March, 145-155.
Ashenfelter, Orley and James J. Heckman {1974}. “The Estimation of Income and
Substitution Effects in a Model of Family Labor Supply," Econometrica, 42:1,
January, 73-86.
Ashworth, J. $. and David T. Ulph (1981), “Endogenecity 1; Estimating Labor Supply
with Piecewise Linear Budget Constraints,” in Charles V. Brown, ed., Taxation
and Labor Supply, London: Allen & Unwin, pp. $3-68.
Baldry, Jonathan C. [979], “Tax Evasion and Labor Suppl Economies Letters, 3:4,
January, 53-56.
Becker, Gary S. [1965], “A Theory of the Allocation of Time Econontic Journal,
75:299, September, 493-517.
Becker, GaryS. [1974], “A Theoryof Marriage,”in Theodore W. Schultz, ed., Eco
nomies ofthe Family, Chicago: University of Chicago Press, pp. 293-344.
Becker. Gary S. [1981], 1 Treatise on the Family, Cambridge, Mass.: Harvard Uni-
versity Press.
Becker, Gary S. [1988] “Family Economics and Macro Behavior.” American Evo
nomic Review, 78: » March, 1-13.
{Whether and How Much Women Work for Pay References 673

Berkson, Joseph [1951}. “Why 1 Prefer Logits to Probits," Biometrics, 7:4, December, Deaton, Angus [1974], “A Reconsiderationof the Empirical Implications of Additive
327-339. ; Preferences,” Economic Journal, 84:334, June, 338-348.
Blundell, Richard, ed. [1987], “Specifica tion Testing in Limited and Discrete Depen- Deaton, Angus and John M. Muelibauer [1980], Economics and Consumer Behavior,
dent Variable Models,” Journal of Econontet ries, 3421/2, January/F ebruary, 1 New York: Cambridge University Press.
274, Deaton, Angus and John M. Muellbauer [1981], “Functional Forms for Labor Supply
Blundell, Richard and lan Walker [1982], “Modelling the Joint Determination
of and Commodity Demands with and without Quantity Restrictions,” Economet-
Household Labor Supplics and Commodity Demands,” Economic Journal, 92, rica, 49:6, November, 1521-1532.
June, 351-364. ; Diewert, W. Erwin ({971], “Choice on Labour Markets and the Theory of the Allo-
Boskin, Michael J. [1973], “The Economics of Labor Supply,” in Glen C. CainRand and cation ofTime,”Ottawa, Ontario: Government of Canada, Department of Man-
Harold W. Watts, eds., Income Maintenance and Labor Supply, Chicago: power and Immigration, Research Branch, mimeo.
MeNally, pp. 163-181. ; Durbin, James [1954]. “Errors in Variables.” Reviewof the International Statisticat
Bowen, William G.and T. Aldrich Finegan [1965], “Labor Force Participator ion and Institute, 22:1, 23-32.
Unemployment,” in Arthur M. Ross, ed., Employment Policy andtheLab Mar- Ehrenberg, Ronald G. and Robert S. Smith [1985}, Modern Labor Economics: Theory
ket, Berkeley, Calif.: University ofCalifornia Press, pp. 115-161. ; and Public Policy, Second Edition, Glenview, Scott, Foresman.
Bowen, William G.and T, Aldrich Finegan [1969], The Economics of Lubor Force Fair, Ray C. [1977], “A Note on the Computation of the Tobit Estimator.” Keuno-
Participation, Princeton, NJ.: Princeton University Press. metrica, 41:7, October, 1723-1727.
Browning, Martin, Angus Deaton, and Margaret Irish [1985], “A Profitable Approach Finney, David J. [1947], Probit Analysis, Cambridge, England: Cambridge University
to Labor Supply and Commodity Demands over the Life-Cycle/ ,” Ecunametrica, Press
53:3, May, 503-543. Sup-
Flaim, Paul O. [1984], “Discouraged Workers: How Strong Are Their Links to the Job
Burtless, Gary and Jerry A. Hausman(1978}, “The Effect ofTaxation on Labor ofPolitica l
Market,” Monthly Labor Review, 107:8, August, 8-11.
ply: Evaluating the Gary Negative Income Tax Experimen t,” Journal Fleisher, Belton M. and George F. Rhodes, Jr. [1976], “Labor Force Participation of
.
conumy, 86:6, December, 1103-1130. Married Men and Women: A Simultaneous Modct,” Review of Economics and
Cain, Glen G. [1966], The Labor Faree Participat ion of Married Women, Chicago: Statistics, $8:6, November, 398-406,
University of Chicago Press. - Florence, P. Sargant [1924], The Economics ofFatigueand Unrest, New York: Holt.
Cain, Glen C. and Harold W. Watts (1973a], “Toward a Summary and Synthesis of Friedman, Milton [1957], A Theory ofthe Consumption Function, Princeton N.J.:
the Evidence,” in Cain and Watts, eds., Income Maintena nceand Labor Supply, Princeton University Press.
Chicago: Rand McNally, pp. 328-367. j Fullerton, Don [1982], “Onthe Possibility of an Inverse Relationship between Tax
Cain, Glen C. and Harold W. Watts, eds. [1973b}, Income Maintena nce and Labor Rates and Government Revenues,” Journal ofPublic Economics, 19:1, October,
Supply, Chicago: Rand McNally. _ 3-22.
Clark, Kim B. and Lawrence H. Summers [1982], “Labour Force Participati
ot Ghez, Gilbert R. and Gary S. Becker [1975], The Allocation ofTime and Goods over
ing and Persistence.” ReviewofEconomic Studies, 49, Supplement,825-844 $44,, theLife Cycle, New York: Columbia University Press,
Cogan, John F. [1977], Labor Supplywith Timeand Money Costsof Participat ion: Goldberger, ArthurS, [1964], Econometric Theory, New York: John Wiley and Sons.
Report No. R-2044 to the U.S, Departme nt ofHealth, Education and Welfare, Goldberger, Arthur S. (1981), “Linear Regression after Selection,” Journal ofEcono-
Santa Monica,Calif.: The Rand Corporation. | metrics, 15:3, April, 357-366.
Cogan, John F. [1980a]. “Married Women’s Lubor Supply*A Comparis onof Aiter- Goldberger, Acthur S. [1987], Functional Form and Urility: A Revi ofConsumer
native Estimation Procedures,” in James P. Smith, ed., Female Labor pp. Supply: Demand Theory, Boulder, Colo: Westview Press.
Theory and Estimation, Princeton, N.J.: Princeton University Press, 90- Gould, William [1979], “Taxes and Female Labor Supply.” unpublished Ph.D. dis-
18.
es sertation, University ofCalifornia at Los Angeles.
Cogan, John F. [1980b), “Labor Supply with Fixed Costs of Labor Market Entry,” in Greene, William H. [1981], “Onthe Asymptotic Bias of the Ordinary Least Squares
ames P. Smith, ed., Female Labor Supply: Theory and Estimatio; n, Princeton, Estimator,” Econometrica, 49:2, March, 505-513.
NJ.: Princeton University Press, pp. 327-364. Greenhalgh, Christine (1980), “Participation and Hours of Work for Murried Women
Cogan, John F. 1981}, “Fised Costs and Labor Supply,” Econometrica, ,49:4, July, in Great Britain,” Oxford EconomicPapers, 32:2, July, 296-318.
945-963. Griliches, Zvi and Jerry A. Hausman [1986}, “Ervors in Variables in Panel Data,”
Cohen, Malcolm S., Samuel A. Rea, and Robert I, Lerman [1970], “A Micro Model Journal ofEconometrics, 31:1, February, 93-118.
of Labor Supply,” Washingt on, D.C.; U.S. Bureau of LaborStati stics, BLS Staff Griliches, Zvi, Bronwyn H. Hall, and Jerry A. Hausman[1978], “Missing Data and
Paper No. 4, ; Self-Selection in Large Panels,” Anmules de INSEE, No, 30/31, April/Septem-
Cornfield, Jerome and Nathan Mantel {1950}, “Some New Aspects of the Appli ber, 137-176.
of MaximumLikelihood to the Caiculation ofthe Dosage Response Curve, Gronau,Reuben [1973a], “The Effect ofChildren on the Housewife’s Value ofTime.”
nalofthe AmericanStatistical Association, 45:250, June, 181-210. Journal ofPolitical Economy, 81:2, Part tt, March/April, $168-S199.
‘4 Whether and How Much Women Work for Pay
Gronaw, Reuben [1973p], “Home Production: A Forgotten Industry.” ReviewofEco-
nomies and Statistics, 62:5, August, 408-415, . -
Gronau, Reuben [1973 ‘The Intrafamily Allocation of Time: The Value of the
Housewives’ Time," -vnerican Econontic Review, 63:4, September, 634-65 1.
Gronau, Reuben [1977], “Leisure, Home Production, and Work—The Theoryof the
Allocation of Time Revisited.” Journal ofPolitical Economy, 85:6, December,
1099-1124, . catcd Samples: The New Jersey Income Maintenance Experiment.” tanaly of
Gronau, Reuben [1986], “Home Production: A Survey,” Chapter in Orley Ashen-
felter and Richard Layard, eds., Handbook of Labor Economics, Vol. |, New
York: Elsevier Sciences BV, pp. 273-304. . a
Gunderson, Morley K.[1980], “Probit and Logit Estimates of Labor Foree Pantici-
pation,” fndustrial Relations, 19:2, Spring, 216-220. . Heckman, James J. {1974a], “Life Cycle Consumption and Labor Supply: An Expla-
Hall, Robert E. [1973], “Wages, Income and Hours of Workin the U.S. Labor Force,
in Glen C. Cain and Harold W. Watts, eds., Income Maintenance and Labor Sup-
ple, Chicago: Rand MeNaily, pp. 102-162. ; .
Ham, John C.[1982], “Estimation of a Labour Supply Model with Censoring Due to
Unemployment and Underemployment,” Review: of Economic Studies, 49:3,
July, 335-354,
Ham, John C. [1986], “Testing Whether Unemployment Represents Intertemporal
Labour Supply Behavior,” Reviewof Economic Studies, 53:4, August, 559-578.
Ham,John C. and Cheng Hsiao [1984], “Two Stage Estimation of Structural Labor
Supply Parameters UsingInterval Data from the 1971 Canadian Census,” Journal
ofEconometries, 24:1/2. January/February, 133-158. ows ; Heckman, James J. [1978], “A Partial Survey of Recent Research an the Labor Supply
Hamermesh, Daniel S. {1990}, “Data Difficulties in Labor Economics,” in Ernst R.
Berndt and Jack E. Triplett. eds., Fifty Years ofEconomic Measurement, Chicago:
University of Chicago Press. ; metrica, ATA, Fanwary, 153-162.
Hamermesh, Daniel S. and Albert Rees (1988}, The Economics of Work and Pay,
Fourth Edition, New York: Harper & Row.
Hanoch, Giora [1980], “A Multivariate Model of Labor Supply: Methodology and
Estimation,” in James P. Smith, ed., Female Laber Supply, Princeton, NJs
Princeton University Press, pp. 249-326. | |
Hausman, Jerry A. [1978], “Specification Tests in Econometrics,” Econometrica,
46:6, November, 1251-1272.
Hausman,jerry A. [1979], “The Econometrics of Labor Supply on Convex Budget
Sets,” Economics Letters, 3:2, \7t- 174, ‘ .
Hausman, Jerry A.[1980], “The Effects of Wages, Taxcs and Fixed Costs on Women's
Labor Force Participation.” Journal ofPublic Economics, October, 161-
194.
Hausman,Jerry A. (1981a]. “Labor Supply.” in Henry J. Aaron and Joseph A. Pech-
man, eds., HowTaxes Affect Economic Behavior, Washington, D.C.: The Brook-
ings Instiution, pp. 27-72. ; ;
Hausman, Jerry A. [1981b), “Income and Payroll Tax Policy and Labor Supply.” in
Lawrence H. Meyer, ed., The Supply-Side Effects ofEconomicPolicy, St. Louis,
Mo.: Center for the Study of American Business, Washington University, 173-
202.
Hausman, Jerry A. [1983], “Stochastic Problemsin the Simulation of Labor Supply,”
in Martin S. Feldstein, ed., Behavioral Simulation Methods in Tax Policy Anal-
ysis, Chicago: University of Chicago Press for the National Burcau of Economic
Research, pp. 47-69,
References 675
Hausman, Jerry A. [1985]. “Taxes and Labor Supply, in Alan Auerbach and Martin
Feldstein, eds., Mandbuvk ofPublic Econamic New York: Elsevier Science Pub-
lishers BY, pp. 213-263.
Hausman,Jerry A. and Paul Ruud [1984], “Family Labor Supply with Tases,” Aster:
ican Economic Review, 74:2, May, 242-248.
Hausman,Jerry A, and David A. Wise [1976], “The Evaluationof Resulis fromTrun-
Economicand Social Measurement, 5:4, Fall, 391-420.
Heckman, James J. [1971], “Three Essays on the Supply of Labor and the Demand
for Market Goods.” Ph.D. dissertation, Princeton University, Deparment of
Economics.
nation of the Relationship between Income and Consumption over the Life
Cycle.” American Economic Review, 64:1, March, 188-194,
Heckmun, James J. [1974b}, “Shadow Prices, Market Wages and Labor Supply,”
Econometrica, 42:4, July, 679-694,
Heckman, James J. [1976a], “The Common Structure ofStatisti
cation, Sample Sclection and Limited Dependent Variables, and a Simple
matorfor Such Models.” -trnals ofEconomic and Sacial Measurement, 54,
475-492,
Heckman,James J., ed. [1976b}, “Special (ssue on Discrete, Qualitative and Limited
Dependent Variables,” Annals ofEconomic and Social Measurement, 5:4, Fall,
of Women,” American Economie Rev May, 200-207.
Heckman,James J. {t979], “Sample Selection Bias as a Specilication Error,” Econu-
Heckman, James J. [1980], “Sample Selection Bias as a Specification Error,” in James
P. Smith, ed., Female Labor Supply, Princeton, : Princeton University Press,
pp. 206-248.
Heckman, James J. [1983], “Comment on Hausman, Manin S. Feldstein, ed.,
Behavioral Simulation Methods in TaxPolicy Analysis, Chicago: University of
Chicago Press for the National Bureau of Economic Research, pp. 70-82,
Heckman, JamesJ. and Thomas E. MaCurdy [1930], “A Li
LaborSupply,” ReviewofEconomic Suadies, 47:1, 5:
Heckman, James J. and Thomas E, MaCurdy (1981}, “New Methodsfor Estimating
Labor Supply Functions: A Su in Ronald G. Ehrenberg, ed., Rexearch in
Labor Economics, Vol. 4, Greenwich. Conn: JAI Press, pp. 65-102
Heckman, James J. and Thomas E. MuCurdy (1986), “Labor Econometrics.” Chapter
32 in Zvi Griliches and Michael D. Intriligator, eds.. Handbouk of Econometrics,
Vol. 3, New York:Elsevier Science Publishers, pp. 1917-1977.
Heckman, James J., Mark R. Killingsworth, and Thomas E. MaCurdy (1981),
“Empirical Evidence on Static Labour Supply Models: A Surveyof Recent Devel-
opments,” in Zmira Hornstein, Joseph Grice, and Alfred Webb, eds.. The Eeu-
nomics ofthe Labour Market, London: Her Majesty's Stationery Office, pp. 73-
122.
Hsiao, Cheng [1986], Analysis of Panel Data, Cambridge, England: Cambridge Uni-
versity Press.
Isachsen, Arne J. and Steiner Steam {1980}, “The Hidden Economy: The Labor Mar-
ket and Tax Evasion,” Scandinavian Journalof Economics, 82:2, 04-311,
Whether and How Much Women Work for Pay
Johnson, George E. [1976], “Evaluating the Macroeconomic Effects of Public
Employment Programs,” in Orley Ashenfelter and James Blum, eds., Evaluating
the Labor Market Effects ofSocial Programs, Princeton, N.J.: Industrial Relations
Section, Princcton University, pp. 90-123.
Katachek, Edward D., Westey Meltow, and Fredric Raines [1978]. “The Male Labor
Supply Function Reconsidered,” Industrial and Labor Relations Review, 31:3,
April, 356-367.
Keeley, Michael C. [1981], Labor Supply and Public Policy, New York: Academic
Press.
Keeley, Michael C., Philip K. Robins, Rabert G.Spiegeiman, and Richard W. West
[1978], “The Estimation of Labor Supply Models Using Experimental Data,”
American Economic Review, 68:5, December, 873-887.
Killingsworth, Mark R. [1983], Labor Supply, New York: Cambridge University
Press.
Killingsworth, Mark R. and James J. Heckman[1986], “Female Labor Supply: A Sur-
Chapter 2 in Orley Ashenfelter and Richard Layard, eds... Handbuok of
Labor Economics, Vol. |, New York: Elsevier Science Publishers BV. pp. 103-
204,
Kosters, Marvin [1966], “Incomeand Substitution Effects in a Family Labor Supply
Model,” Santa Monica, Calif: The Rand Corporation, Report No. P3339.
Layard, Richard, Margaret Barton, and Antoni Zabalza [1980], “Married Women’s
Participation and Hours,” Economica, 47;185, February, 51-72.
Layard, Richard and Jacob Mincer, cds. [1985], “Trends in Women's Work, Educs-
tion and Family Building.” Journal ofLabor Economics, 3:1, Supplement, kan-
uary, S1-S396.
Lec, Lung-Fei [1982], “Some Approaches to the Correction of Selectivity Bias,
ReviewofEconomie Studies, 49:3, July, 355-372.
Leuthold, Jane H. (1968), “An Empirical Study of Formals income Transfers and the
site
Work Decision ofthe Poor," Journal of Human Resources, 3:3, Summer, 312-
323.
Leuthold, Jane H. [1978], “The Effect ofTaxation on the Hours Worked by Married
Women,” industrial and Labor Relations Review, 31:4, Juty, 520-526.
Lewis, H. Gregg [1957], “Hours of Work and Hours ofLeisure,” Proceedings of the
Industrial Relations Research Association, pp. 196-206.
Lundberg, Shelley [1985], “The Added WorkerEffect.” Journ ofLubor Economics,
3:1, Supplememt,January, SI1-S37.
MaCurdy, Thomas E. [1981], "An Empirical Model of Labor Supplyin a Life-Cycle
Setting,” Journal ofPolitical Economy, 89:6, December, 1059-1085.
Maddala, G. S. (1983], Limited Dependentand Qualitative Variables in Econometrics,
Cambridge, England: Cambridge University Press, Econometric Society Mono-
graphs No. 3.
Manser, Marilyn E. and Murray Brown [1979], “Bargaining Analyses of Houschold
Decisions,” in Cynthia B. Lloyd, Emily S. Andrews, and Curtis L. Gilroy, eds..
Women in the Labor Market, New York: Columbia University Press, pp. 3-26.
Manser, Marilyn E. and MurrayBrown [1980], “Marriage and Household Decision-
Making: A Bargaining Analysis," fnternational Economic Review, 21:1, February,
3-44,
McDonald, Jobn F. and Robert A. Moffitt [1980], “The Uses of Tobit Analysis,”
ReviewofEconomies and Statistics, 67:2, May, 318-321.
References 677
McElroy, Marjorie B. and Mary Jean Horney [1981], “Nash-Barguined Houschold
Decisions: Toward a Generalization of the Theory of Demand,” International
Ecanomic Review, 22:2, June, 333-349.
McFadden, Daniel (1984), “Econometric Analysis of Qualitative Response Models,”
Chapter 24 in Zvi Griliches and Michael D. Intriligator, eds., Handbook ufEcono-
metrics, Vol. 2. New York: Elsevier Science Publishers BV, pp. 1395~1457.
Metcalf, Charles E. [1973], “Making inferences from Controlted Income Maintenance
Experiments.” American Economic Review, 63:3, June, 478-583.
Mincer, Jacob |1962}, “Labor Force Participation of Married Women: A Studyof
LaborSupply.” in H. Gregg Lewis, ed., Aspects ofLabor Economics, Princeton,
N.J.: Princeton University Press, pp. 63-97.
Mincer, Jacob {1963}, “Market Prices, Opportunity Costs, and Income Elftets.” in
Carl Christ, ed.. Meusurement in Economics: Essays in Honor ofYehuda Gran-
feld, Stanford, Calif: Stanford University Press, pp. 67-82.
Mincer, Jacob [1966], “Labor Force Participation and Unemployment: A Reviewof
Recent Evidence,” in Robert A. Gordonand Margaret S. Gordon, eds., Prosperity
and Unemployment, New York: John Wiley and Sons, pp. 73-112.
Mitchell, Olivia . [1980],“Labor Force Activity of Married Womenas a Response to
Changing Jobless Rates,” Monthly Labor Review, 103:6, June.
Moffitt, Robert A. [1982], “The Tobit Model, Hours of Work and In: ional Con-
straints,” Review ofEconomics and Statistics, 64:3, August, 510-515.
Moffitt, Robert A. {1984], “Profiles of Fentifity, Labour Supply and Wages of Married
‘Women: A Compicte Life-Cycle Model.” Review: of Economic Studies, 51:2,
April, 263-278.
Moffitt, Robert A, and Ken Kehrer [1981}, “The Effect of Tax and Transfer Programs
on Labor Supply: The Evidence from the Income Maintenance Experiments,”in
Ronald G, Ehrenberg. ed. Research in Lubor Economics, Vol. 4, Greenwich,
Conn.: SAI Press, pp. 103-150.
Morgan, James N., etal. {1966], Productive Americans: A Study uf HewIndividuals
Contribute to Economie Progress, Ann Arbor, Mich.: SurveyResearch Center,
University of Michigan.
Morgan, James N. et al, [1974], Five Thousand American Famities—Patterns of Eeo-
nomic Progress, Ano. Atbor, Mich: Survey Research Center, University of
Michigan.
Mroz, Thomas A. (1987), “The Sensitivity ofan Empirical Model of Married Wom-
en's Hours of Work to Economic and Statistical Assumptions,” Econometrica,
35:4, July, 765-799,
Nakamura, Alice and Masao Nakamura (198 1a]. “A Comparison of the Labor Force
Behavior of Married Women in the United States and Canada, with Special
Attention to the Impact of income Taxes.” Econometrica, 49:2, March, 451-489.
Nakamura, Alice and Masao Nakamura (1981b], "On the Relationships among Sev-
eral Specification Error Tests Presented by Durbin, Wu and Hausman.” Econ
metrica, 49:6, November, 1583-1588.
Nakamura, Alice and Masso Nakamura [1985a}, “Dynamic Models of the Labor
Foree Behavior of Married Women Which Can Be Estimated Using Limited
AmountsofPast Information,” JournalofEconometrics, 27:3, March, 273-298.
Nakamuru, Alice and Masao Nakamura (1985b]. The Second Paycheck: 4 Sociovco-
nomic Analysis ofEarnings, Orlando, Fla: Academic Press.
Nakamura, Alice and Masao Nakamura [1985c}, “On the Performance of Tests by
 aether and How Much Women Workfor Pay
‘Wuand by Hausmanfor Detecting the Ordinary Least Squares Bias Problem.”
Journal of Econometrics, 29:3, September, 243-227.
ckamura, . Masao Nakamura, and Dallas Cullen [1979], “Job Opportunities,
the Offered Wage, and the Labor Supply of Married Women,” American Keo
nontic Rev 69:5, December, 787-805.
itional Commission on Employment and UnemploymentStatistics [1979], Cont-
ing the Labor Force, Washington, D.C.: U.S. Government Printing Office.
sen, Randall J. {1980}, “A Least Squares Correction for Selectivity Bias," Keoau-
metrica, 48:7, November, 1815-1820,
aes, Herbert S. et al. [1970]. The Pre-Retirement Years: A Longitudinal Suadyof
the Labur Market Experience ofMen, Vol. 1. Washington, D.C.: U.S, Department
of Labor, Manpower Administration,
sncavel, John [1986], “Labor Supply of Men: A Survey.” Chapter | in Orley Ash-
enfelter and Richard Layard, eds.. Handbook of Labor Economics, Val. |, New
York: Elsevier Science Publishers BV, pp.3-I01.
osen, HarveyS. [1976], “Taxes in a Labor Supply Model with Joint Wage-Hours
Determination,” Feonemetrica, 44:3, May, 485-507,
osett, Richard N. and Forrest D. Nelson [1975], “Estimation of the Two-Limit
Probit Regression Model,” Econvmvtrica, 43:1, January, 141-146,
andmo, Agnar {1981}, “Income Tax Evasion, Labour Supply and the Equity-Eth-
ciency Tradcoff.”* Journal ufPublic Economics, 6:1, February, 37-54,
choenberg, Erika and Paul H. Douglas [ [937], “Studies in the Supply Curve of Labor.
The Relation between Average Earnings in American Cities and the Proportion
Seeking Employment,” Journal ofPolitical Economy, 45:1, February, 45-62.
chult, T. Paul [1980], “Estimating Labor Supply Functions for Married Women,
in James P. Smith, ed., Female Labor Supply. Princeton N_J.; Princeton Uni
sity Press, pp. 25-89.
mith, Adam [1776], The S¥ealth of Nations, London: M. Dent & Sons. Reprinted
in 1937 as the Cannan edition by The Modern Library, Random House, New
York.
mith, James P. [1977], “Assets, Savings and Labor Supply.” Economic Inquiry. 15:4.
October, 551-373.
«mith, James P. [1980]. “Assets and Labor Supply,” James P. Smith, ed., Female
LaborSupply. Princeton N.J.: Princeton University Press, pp. 166-205.
imith, James P. and Michact Ward [1985] ‘ime-Series Growth in the Female Labor
Force,” Journalof Labor Economic |. January, Supplement, S59-S90_
imith, Shirley J. 1983], “Estimating Annual Hours of Labor Force Activity.”
Monthly Labor Review, (06:2, 13-22.
jorrentino, Constance {1983}, “International Comparisons of Labor Force Participu-
tion, 1960-1981," Afonthiy Labor Review, 106:2, 23-36.
stern, Nicholas [1986], “Onthe Specification of Labor Supply Functions.” in Richard
Blundell and tan Walker, eds., Unemployment, Search and Labour Supply, Cam-
bridge, England: Cambridge University Press, 143-189,
Theeuwes, Jules [1981], “Family Labour Force Participation: Multinomial Logit Esti-
mates,” Applied Economics, 13:4, December, 481-498.
Tobin, James [1955], “The Application of Multivariate Probit Analysis to Economic
SurveyData," Cowles Foundation Discussion Paper 1.
Tobin, James [1958], “Estimation of Relationships for Limited Dependent Variables,”
Econometrica, 26:1, January, 24-36.
USS. Departmentof Commerce. Bureauof the Census [1987], Surveyof Income and
Further Readings 679
ProgramPanicipation: Users’ Guide, Washington, D.C: Customer Services, Data
UserServices Division.
Wales, Terence J. [1973], “Estimation of a Labor Supply Curve for .
Business Proprietors,” International Econumic Review, 14:1, February, 69-80.
Wales, Terence J. and Alan D. Woodland [[977}, "Estimation ofthe Allocation of
Timefor Work,Leisure and Housework,” Econometrica, 45:1, January, 115-132.
Wales. Terence J, and Alan D. Woodland [1979], “Labour Supply and Progressive
Taxes,” Review ofEconomic Studies, 46:1, January, 83-95.
Wates, Terence J. and Alin D, Woodland [1980]. “Sample Select and the Esti-
mation of Labor Supply Functions,” faternational EconomicRen 2, June,
437-468.
Watts, Harold, Dale Poirier, and Charles Mallar [1977], “Sample, Variables and Con-
cepts Used in the Analysis,” in Harold Watts and Albert Rees, eds,, The New
Jersey Income Maintenance Experiment: Labor Supply Response, Vol. 2, New
York: Academic Press, pp. 33-56.
Weiss, Yorum [1986}, “The Determinantsof Life Cycle Earnings: A Survey.” Chapter
11 in Orley Ashenfelter and Richard Layard, eds... Hunelbouk of Labor Economics,
Vol. 1, NewYork: Elsevier Science Publishers BV,pp. 603-640.
White, Halbert J., [1980], “A Heteroskedasticity-Consisteat Covariance Matrix Esti-
mator and a Direct Test for Heteroskedasticity.” Heonumetricu, 48:4, May, 817
838. :
White, Halbert J, [1982], “Maximum Likelihood Estimation of Misspecified Models,”
Econometrica, 50:2, March, 483-499,
Woytinsky. Wladimir S. [1940], Additionat Workers and the Volume of Unemplay-
ment in the Depression, Washington, D.C.: Social Science Research Council.
Wu, De-Min [1973], “Tests of Independence between Stochastic Regressors and Dis-
turbances,” Econometrica, 41:4, July, 733-750.
Zabalza, Antoni[1983], “The CES Utility Function, Nonlinear Budget Constraints
and Labour Supply: Results on Female Participation and Hours.” Ecrnumic Jour-
nat, 93:370, June, 312-320.
FURTHER READINGS (All are textbooksin labor economics)
Addison, John T. and W. Stanley Siebert [1979], TheMarketfor Labor: An anabytical
Treatment, Santa Monica, Calif: Goodyear.
Ehrenberg, Ronald G. and Robert S. Smith [1985], Modern Labor Eeunumics: Theory
und Pablic Policy, Second Edition, Glenview, IIL: Scott, Foresman, ‘
Fleisher, Belton and Thomas Kniesner [1984], Labur Economics: Theory, Evidence
andPolicy, Third Edition, Englewood Ctiils, N.J.: Prentice Hall.
Gunderson, Mosley and W. Craig Riddell {1988}, Labor Market Economics: Theory,
Evidence ‘and Policyin Canada, Second Edition, Toronto: McGraw-Hill Ryerson,
Ad.
Hamermesh, Daniel S. and Albert Rees [1988], The Economics of Work und Pay.
Fourth Edition, NewYork: Harper & Row.
MeConnell. Campbell R. and Stantey L. Brac [1986], Contemporary Labor Econom.
ies, New York: McGraw-Hill.
Reynolds, Lloyd G., Stanley H. Masters, and Colletta H. Moser {1987], Economics of
Labor. Englewood Cliffs, N.J.: Prentice Hall.
its continued from copyright neering and Econometric Interpretations of
Energy-Capitat Complementarity.” Ameri-
ral Seties No. 73, NY: Columbia Uni- can Economic Review, 69:3, June 1979, p.
y Press for the National Bureau of Eco- 351. Reprinted with permission, Fried-
¢ Research, 1961, p. 35. Reprinted with man, “The Role ofMonetary Policy.” Anter.
issiun. Griliches: Technology, Edu icun EconomieReview, 5B:1, Macch 1968, p.
dand Productivity, NY: Basil Blackwell, UL Reprinted with permission. Okun,
1988, p. 120. Reprinted with per- “Poswar Macroeconomic Performance,” in
Martin S. Feldstein, ed., The American Econ.
on. Forester, oll, The Information
nology Review, Cambridge, MA. omy in Transition, Chicago: The University Author Index
MITPress, 1985, Reprinted with of Chicago Press. 1980, p. 166. Reprinted
ission. Machlup, Journut of Political with permission. Solow, “What We Know
omy, 82:4, July/August, 1974, p. 892, and Don't KnowAbout Inflation.”” Technol-
inted with permission. Halll and Tay- ogy Review, 81:3, December 1978/Jan- Almen,Shirley. 237,
dtacrocconomics: Theory, Petformance. wary 1979, p. 31. Reprinted with permission Aaker, David, 407, 441, Ashley, Richard, 356f,
from Technology Review, copyright 1978, 4a7 288, 299 357, 374, 383, 398,
. 237, Reprinted with Lucas, Jr. and Sargent, “After Kenyesian Aaron, Henry J., 674 Altonji, Joseph G., 539, 427, 438f, 4408,
liches, “Commems on Macroeconomics,” in After the Phillips Abel, Andrew, 260, 584, 670f, 671 441
tor/Kendrick, eds... Fron Curve: Persistence of High Inflation and 296f, 297f, 298, Alworth, Julian, 302 Ashworth, 5. S., 667f,
ofQuantitative Econumics, Vol. Il, Am- High Unemployment, BOSTON: Federal 299 Amel, Dean F., 146f, 6698
im; North-Holland, 1974, p. 335. Re> Reserve Bank ofBoston, ConferenceSeries, Abernathy, William J., 146 Assmus, Gert, 392
xd with permission, Koopmans: No, 19, p. 69. Gordon, “The Theory of 676, 96f, 98 Amemiya, Takeshi, Atkinson,Scott E., 484,
ited from “Report ofthe Modeling Re- Domestic Inflation.” american Economic 621, 632, 654-656, Sol
¢ Group, 1978, with permission from Review, 67:1, March 1977, p. 132, Reprinted Abowd,John M., 210f,
670f, 671 Atrostic, B. K., 211f.
ativnal Academy ofSciences, Washing- with permission. Killingsworth, Labor 214, 522, 5810, 584
DC. Reprinted by permission of Supply, NY: Cambridge University Press, Abraham, Katherine, Anderson, Richard G.. 214
RICAN SALESMAN, © TheNational 1983, p. 432. Reprinted with permission. 176, 214, 218, 219 475, 501 Auerbach,Alan J.,
ich Bureau, Inc., 424 North Third St, Nakamura, Nakamura, and Cullen, “Job Ackley, Gardner, 363, Anderson, Theodore 297f, 299, 675
ngton, lowa 52601-5224. Clarke, Opportunities, the Offered Wage, and the 437f, 441 W.. 186 Angarten,Stan, 196,
nometric Measurement ofthe Duration Labor Supply of Married Women,” stmert- Adelman,Irma, 144f, Ando,Albert, 269, 145f, 146
Svertising Effects on Sales.” Journal of” can Econumics Review, 69:5, Devernber 146 297f, 299 Ayanian, Robert, 4401,
eting Research, 13:4, Novem! 1979, p. 787, Reprinted with permission. Andress, Frank J., 966. 44t
. p. 355. Reprinted with permission. Fullerton, “On the Possibility of an Inverse Addison, John T., 211f,
98
talensee, The Econumics ofAdvertising, Relationship between Tax Rates and Gav- 214, 222, 592,
Andrews. Emily S., 676 Bailey, Elizabeth E.,
jorth-Holland, 1972, p. 38. ernment Revenues,” Journal uf Public Evo- 667£, 671, 679
mied. with permission. Fuss, Mc numnies, 19:1, October 1982, p. 20. Reprinted Ager, MargaretE., 505 Andrisani, Paul, 189, 500f, 501
en and Mundlak, “A Survey of Fune- with permission. Jack Johnston, Ecouv- Ahl, David H., 186 216 Bailey, Martin J., 504
1 Formsin the Economics Analysis of metic Methods, Ye, © 1984, p. 335. Re» Aigner. Dennis J., 182, Anton,James J., 96f, 98 Baily, Martin N., 587
action,” in Fuss/MeFadden, eds. Prom printed with permission of McGraw-Hifl, 214, 310, 354f, Arbeiter, Larry. 407 Bain, Joseph, 415, 441
-mEconomics: A Ducal Approach to The- Inc. Rudiger Dornbusch and Stanley 356, 357, 439f, Archibald, G. Baldry, Jonathan C.,
ind Applications, Amsterdam: North- Fischer, Macrovcunomics, 4Je, © 1987, p. 441, 447, 4996, Christopher, $81f, 669f, 671
ind, 1978, p. 266. Reprinted with 315. Reprinted with permistion of McGraw. 503, 670f, 671 585 Barnett, Wiliam A.,
ission, Berndt and Wood, “Engi- Hill, Ine. Aitchison, John, 144f, Arndt, Johan, 414, 446 299, 501f, 501
TO CREDITS: 146 Arnold, Stephen J., 407, Barquin-Stolleman.
4: courtesy of Guy Orcutt. Page 5: tesy of Robert Hall Page 258: counesy of Akerlof, George A., 437, 441 Joan A., 146
ice V. Wilkes, David J. Wheeler, and James Tobin. Page 329: courtesy of Hea- 219, 212f 214 Arrow, Kenneth J.. 157. Barro, Robert J., 582f.
+y Gill, Preparation ofProgramsjor ant drik Houthakker. Page 334: photo of Alberts, William, 96f, 214, 365, 444, 452, 585, 592
cpnie Digital Computer. © V9S1, ce Daniel McFadden: courtesy ofTito Simboli. 97f, 98 50% Bartik, Timothy, £30,146
1 1985; Adgison-Wesley Publishing Page 395: courtesy of Richurd Schmnalensee. Albion, Mark S., 440£, Ash, Colin, 666f, 671 Barton, Margaret, 676
any. Reprinted with permission ofthe Page 406: courtesy of Harry V. Roberts. 443 Ashenfelter, Orley, 179, Basmann, Robert L.,
cher. Page 24: courtesy of Harry M. Page 459: courtesy of W. Erwin Diewen, Alchian, Armen, 96f, 210f, 2436, 214, 371, 441, 583f, 585
owitz. Page 38: courtesy ofSteven A. Page 440: courtesy of Dale Jongenson. 98 215, 216, 218, 221, Bass, Frank M., 380,
Page 82: courtesy of Mare Nerlove. Page 537: courtesy of Rubert E. Lucas, Je, 222, $39, 585, 607, 391, 43Bf, 441. 442
Page 546: courtesy of Ray C. Fair. Page Aliber, Robert Z., 586
115: courtesyof Zvi Griliches. Page 667f, 6698, 671, Beach, Charles M...
zourtesy of Gregory Chow. Page 628: courtesy of James Heekman. Page Allen, R.G. D., 606,
674, 676, 678, 679 356f, 357
sourtesy of Jacob Mincer. Page 185: 636: courtesyof Jerry A. Hausman. 67
sy of Alan Blinder, Page 246: cour-
681
 Author Index Author Index 683

Jeauvais, Edward, 358 Blattenberges, Gail R., Brown, James A. C., Chamberiain, Gary, Cobb, Charles, 68, 451, Day. George S., 96f,
Becker, Gary, 152, 155, 315, 354f, 355f, 198, 1440, 146, 167, 215 4986, 503 97f, 98, 99
156, 169, 170, £80, 359 356f, 357 Chambertin, Edward Cochrane, Donald, 5 Daymont, Thomas,
181, 211f, 2126, Blaug, Mark, 212f, 215 Brown, Murray, 505, H., 368, 442 Coen, Robert M., 296f, 189, 216
214, 216, 602, 610, Blinder, Alan, 156, 676 Chambers, Robert G., 297f, 300. 583f, Deaton, Angus, 606,
667f, 671, 673 182-184, 188, 191, Brown, Philip, 56f, $7 S00f, 502, 506 585 670f, 672, 673
Begg, David K. H., 549, 203-208, 211f, 215 Brown, Randall S.. 484. Chapman, L. Duane, Cogan, John F., 613, de Leeuw, Frank, 236,
585 Bloch, Fasrelt, 178, 202, 502 3558, 358 633, 634, 638, 672 300
Behrman,Jere R., 189, 215 Browning, Martin, 667f. Chen,Carl R., 560, 57 Cohen, Malcoim S., Deloitte, Haskins and
215 Bloch, Harry, 441f, 442 670f, 672 607,6674, 672 Sells, 263, 300
Ben-Porath, Yoram, Bloom,David, 182, 215 Brue, Stanley L.. 180f, Cole, Rosanne, 124, de Marchi, Neil, 550,
210f, 215 Blum, James, 676 218, 222, 679 126f, 138~139, 585, 592
Bentzel, Ragnar, 577, Blume, Marshall E., Brunner. Karl, 303, 296f. 299, 452, S01 143, 1448, t45f, Desai, Meghnad, 295f,
585 55f, 56 447, 588. 590 Chiang, Judy S. Wang. 146 300, S80f, 585, 592
Ben-Ur, Joseph, 438f, Blundell, Richard, 670f, Buchanan, Norman, 482, 502 Comanor,William S.. Dhalta, Norman K.,
446 672, 678 365, 442 Chirinko, Robert S., 4408, 4411, 442 439F, 442
Bergman,Barbara R.. Bohi, Douglas R.. 328, Bultez, Alain V.. 437f, 298f, 299 Comins, KathyA., Dhrymes, Phoebus,
223 S31, 344, 357 43RE, 442, 444 Chiswick, Barry. 187, 228f, 296f, 301 145f, 147, 296f,
Bergson, Abraham, 453, Bojanek, Robert, 415, Burdett, Kenneth, 212f, 2110, 2138, 214, Constuner Reports, 300, 3558, 357
301 445 215 ans 4408, 442 Diamond, Peter A..
Berkson,Joseph, 671f, Borden, Neil H., 363, Burmeister, Edwin, 55f, Chow, Gregory C., 56f, Conway, Delores, 191, 5811, 585
672 380, 424, 442 57, 58 57, 97f, 98, 104, 215 Diewen, W. Erwin,
Berndt, Ernst R., 146f, Boskin, MichaelJ., 638, Burridge,Peter, 580f, 118, 119, 136-137, Cooper, Guy M., 55f, 100, 304, 453,
146, 217, 297f, 672 585 139-142, 1456, 58 459-460, 4991,
299, 301, 311f, Boston Consulting Burstein. Meyer L., 146, $49, 585. Cooper,J. Phillip, 288, S00f, 501, 503,
355f, 357, 449, Group, 60, 96f, 98 1446, 146 Chow, Simeon,440f, 300 631, 673
450, 454, 456, 472, Bowen, Howard R., 216 Burtless, Gary, 355f, 442 Corcoran, Mary, 184, Dillon,John, 453, 503
479, 484, 486, 487, Bowen,William G., 357, 631-633. Christ, Carl F., 54f, 57, 188, 215 Dolan, Robert J. 96f,
SOOF, SIF, 501, 6670, 672 669F, 672 99, 505, 677 Corio, Marie, 3556, 357 98
$02, 504, 505, 06, Bowles, Samuel, 213f, Butler, John £., 951, 98 Christensen, Laurits R., Cornfield, Jerome, 668f, Donner, Arthur, $16,
674 215 Butler, Richard, 186, 81-84, 90-92, 98, 672 586
Betancourt, RogerR., Box, George E. P., 145f, 215 247, 299, 354f, Court, Andrew T., 110, Dorfman, Robert, 365,
298f, 299, 354f, 146, 299, 320, 357 357, 453, 480, 482, 144f 147 442
356f, 357 Boyer, Kenneth D., Cable, John, 439f, 442 484, 499f, 502 Cowing, Thomas G.. Dornbusch, Rudiger,
Bischoff, Charles W., 4411, 442 Cain, Glen G., 182, Chu, Shih-Fan, 499f, 81, 98, 482, 503 224, 300, 5831, 86
250, 253. 286, 293, Bradford, David F., 214, 215, $96, 614, 503 Cowles, Alfred W. II, Douglas, Paul H., 68,
2981, 299 297f, 299 8686, 672, 674 Ciccolo, John H.. Jr. 20. $4f, 57 450, 451, 4981,
Bitros, George C., 296f, Brealey, Richard, 54f, Card, David, 539, 585 2971, 300 Cowling, Keith, 439f 503, 614, 678
299 558, 56f. 57 Carliner, Geoffrey. 187, Clark, J. Maurice, 233, Cox, David R., 1456, Douglass, George K.,
Black, Fisher, 38, 55f, Brechling, Frank P. R., 25 296f, 300 146 2121, 216
56 S80f, 585 Carman, David, 407, Clark, Kim B., $22, Cragg, John G., 59 Dréze, Jacques, 499f,
Blanchard, Olivier J.. Breusch, TrevorS., 568, 441 585, 6671, 672 Culten, Dallas, $93, 506
260, 296f, 2978, S84f, 585 Cartwright, David W., Clark, Peter K., 298f, 678 Dubin, Jeff, 334
298, 299, S81f, Brinner, Roger E., 5810, 125, 126f, 144f, 300 Dubrul, Stephen M.,
S836, 585 585 146 Clarke, Darral G., 361, Dasgupta, Partha, 96f, 144f, 147
Blank, David M., 393, Brown, Charles, 210f, Caves, Douglas W., 387, 389-391, 398, 98 Duesenberry, James,
398, 4376, 439f, 218 354f, 357, 482, 502 415, 4398, 440f, Davies, Nicholas, 439f, 239, 296f, 298f,
442 Brown, Charles V., 67! Ceef, Christopher, 18F 442 300, 586
684 Author Index Author Index 685

Dutberger, Ellen R., Farris, Paul W., 440f, French, Kenacth R.. Goett, Andrew A., 3566, Griffiths, William E.. Hamitton, David. 97,
145f, 146f, 146, 443 56f, 57 357 57, 99, 148, 218 99, 135, 148
147 Feldstein, Martin S., Friedlacnder, Ann F.. Goldberger, Arthur S., Griliches, Zvi, 97f, 99, Hanoch, Giora, 170,
Duncan, Greg, 184, 269, 296f, 301, 481, 482, SO0E, 97f, 99, 1440, 147, 102, 115-116, ITF, 217, 476,
188, 210f, 215, 216 590, 674, 675 50t, 502, 505 191, 192, 216, 144f, 1456, 1468, 500F, 503, 633,
Durbin, James M., 282, Ferber, Marianne A., Friedman, James W., 298f, 301, 3551, 146, 147, 148, 149, 667f, 674
296f, 300, 347, 192, 216 A37f, 443 356f, 358, 4381, 166, 167, 217, 224, Hansen, Alvin, 363,
357, 438f, 442, Ferguson, James M., Friedman, Milton, 210f, 443, 499f, $03, 236, 237, 264, 301. 437f
568, 582f, 584f, 4a0f, 443 216, 314, 357, 507, 550, 5531, 535. 304, 305, 3551, Hansen, Bent, S80f, 587
586, 669f, 673 Field, Barry C., 502, 521, 586, 648, 673 574, $80f, 5831, 358, 387, 396, 443, Hansen, Lars P., 298f,
Dutton, John M., 95f, 504 Frisch, Ragnar, 452, 587, 588, 622, 657, 504, 587, 668f, 301, $01f, $03,
98 Filer, Randall K., 189, 503 668f, 673 670f, 673, 675, 677 564, 5830, 587
216 Fromm, Gary, 297, Gollop, Frank M., 506 Gronau, Reuben, 2136, Hansen, W. Lee, 169,
Eastlack, Joseph O., Finegan, T. Aldrich, 300 Gordon, Margaret S., 217, 612, 667f, 207
4401, 442 6676, 672 Fudenberg, Drew, 96f, 677 668F, 673, 674 Hanser, Philip, 359
Eatwell, John, 445 Finney, David J., 668f, 9 Gordon, Robert A., 677 Grunfeld, Yehuda, 239, Hanssens, Dominique
Eckstcin, Otto, 586, 588 Fullerton, Don, 297f, Gordon, Rober J., 301 M,, 4408, 443
Ehrenberg, Ronald G., Fischer, Stanley, 224, 299, 301, 302, 593, 143f, 1458, 147, Guilkey, David, 482, Hanushek, Eric A..
213f, 216, 220, 300, 582f, 586 673 149, 507, 540, 503 2 217
222, 592, 672, 675, Fisher, Franklin M., Fuss, Melvyn A., 2976, SBIE, S820 S83, Gujarati, DamodarN.. Harberger, Arnold C.,
677, 679 56f, 57, 1436, 145f, 299, 449, 501f, 587 56f, 57 308, 504
tein, Albert, 20, 57 146f, 147, 312, 502, $03, 504, 506 Gorman,John A., 228f, Gunderson, Morley K.. Harman,Harry H.,
Eisner, Robert, 252, 354f, 355f, 357, 295f, 301 213f, 217, 222, 1456, 148
269, 296f, 2971, 371, 442, 577, Gabor, Andre, 354f, Gorman, William M.. 6701, 674, 679 Harper, Michael! J.,
300 SB2f, 586 144f, 147 2978, 301
Ekelund, Robert B., Jr., Fisher, Irving, 516, Galbraith, John Gould, John P.. 298f, Haavelmo, Trygve, 243, Hart, Peter, $90
393, 443 S80f, 586 Kenneth, 363, 301 297f, 301. 584f, Hartman, RaymondS.,
England. Paula, 213f, Flaim,Paul, 667f, 673 437f£, 443 Gould, William, 669f, 587 328, 3556, 358
216 Fleisher, Belton M., Gallant, A. Ronald, 96f, 673 Hall, Bronwyn H., 6681, Harvey, Andrew C.,
Epple, Dennis, 130, 147 222, 592, 6676, 99, 500f, 503 Gramm,William P., 670f, 673 289, 298f, 302,
Epstein, Roy J., 19f, 673, 679 Garber, Peter M., 549, 393, 443 Hall, Robert E., 146f, 304, 359
550, 583f, 586 Flemming, John S., 586 Granger, Clive W.3., 148, 153, 217, 224, Haugh, Larry D., 382.
Estes, William K., 95f, 269, 301 Gardner, David, 99 289, 301, 359, 374, 245, 246, 247, 445
98 Flood, Robert P., 549 Geisel, Martin S., 355f, 383, 398, 427, 2966, 297f, 301, Haugh, William D.,
Evans, MichaelK., 236, Florence, P. Sargant. 359 439f, 441, 443 S81f, 582f, $87, 212¢
301 666f, 673 Ghemawat, Pankaj, 79, Green, Carole A., 192, 631, 644,6671. Hausmun, Jerry A.,
Fomby, Thomas B.. 80f, 80, 97f, 99 216 668f, 674 355f, 357, 4388,
Fair, Ray C., 266, 292, 97f, 98 Ghez,Gilbert R.. 156, Greene, William H.. Hallett, A. J. Hughes. 443, 499f, 503,
301, 304, 4411, Foot, David K., 296f, 216, 667f, 673 81-84, 90-92, 98, S83f, 587 550, 565, SOF.
443, 535, 545, 546, 301 Gilbert, Christopher L., 622, 657, 673 Halvorsen, Robert, 325. 587, $88, 629,
549, $52, 583F, Forester, Tom, 102, 147 515, 550, 580f, Greenhalgh, Christine, 358, 484, 501 631-638,6671,
586, 668f, 672 Frankel, Marvin, 4991, 5B5, $86, 592 638, 673 Ham,Joho C., $39, 668F, 6691, 670f,
Fama, Eugene F., $50. 503 Gilroy, Curtis L.. 676 Greenwood, Joen, 1456, 587, 632, 6708, 672, 673, 674, 675
56f, 57, 59 Franklin, Benjamin, Gintis, Herbert, 213f 1468, 147 674 Hausman, Leonard J.,
Farber, Henry S.. 176, 150 Gloricux, G., 269, 302 Gregory, Allan W., Hamermesh, Daniel, 215
179, 2146, 218, Fraumeni, Barbara, Goddeeris, John H., S00f, 503 172, 214f, 217, Hax, Amoldo.60, 96f,
216 486, 500f, 504, 506 2108, 216 Grenier, Gilles, 187, 222, 5798, $87, 9
Farley, John U., 392, Freeman, Richard B.. Godfrey, Leslie G., 568, 216 SORE, 6668, 674. Hayashi, Fumio, 260,
439f, 441,443 171, 181, 216 S84F, $86 Grice, Joseph, 675 679 297f, 302
Author Index Author Index 687

Haydel, Juanita M.. Holmiund, Bertil, 2108, Jenkins, Gwilym M., Keim, Donald B., 56f, Knox, A. D., 296f, 303 Langford, David, 18f
56f, 57 216 289, 299, 320, 357 37 Koopmans, Tjalling C.. Laslavic, Thomas J..
Heady, Earl, 453, 303 Holt, Charles C., 581%, Jensen, Michaet, 38, Kelejian, Harry H., 18f, 306, 358, S84f, S6f. 57
Heck, Jean Louis, 57 588 55f, 56 296f, 2986, 299, 588 Lau, Lawrence J., 453,
Heckman, JamesJ., Holtmann, Alphonse, Jeuland, Abel P., 96f, 3560, 357 Kopcke, Richard W., 480, 4991, SO0f,
157, 186, 191, 482, 503 98 Kelly. Michael, 439f, 226, 237, 240, 241, 502, 564
2116, 212f, 215, Holzer, Harry, 2136, Jhally, Sut, 4370, 444 442 254, 261. 270, 279. Lau, Sheila, 32, 59
217, 607, 614-617, 217 Johnson, George E., Kemmis, Robyn, 581f, 284-287, 291, Layard, Richard G.,
623-629,633, 634, Hood, William C., 584f, 179, 2116, 2138, 585 2958, 2971, 303 215, 216, 218, 221,
637, 638, 645-648, 588 214, 217, 6698, Kendrick, David A., Kosters, Marvin, 667f, 222, 505, 638,
665, 667f, 668. Horney, Mary Jean, 676 301, 503 676 6671, 6708, 674,
6696. 6708, 671. 677 Johnson, Samuel, 362 Keynes, John M., 305 Kotler, Philip, 437f, 676, 678, 679
675, 676 Hornstein, Zmira, 675 Johnson,Stanley R.. Khaled, Mohammed S., 444, 487 Lazear, Edward, 156,
Heien, Dale M., 144f, Horsky, Dan, 440f, 443 978, 98 5008, 501 Koyck, Leendert M.. 176, 186, 218
148 Hough, Douglas E.. Johnson, Thomas, 182, Kiefer, Nichotas M., 148, 303, 355f, Lec, Lung-Fei, 629,
Helliwell, John F., 269. 2126 218 217 2126, 218 358, 4391, 444, 591 668f, 676
302, 305 Houston, Franklin S.. Johnston, J., 100, 449, Kiker, Billy F., 2108, 218 Krinsky, Itzhak, SOOf, Lee, Tsoung-Chao, 57,
Helpman, Elhanan, 96f, 4396, 443, 447 504 Killingsworth, Mark ., 504 99, £48, 218
9 Houthakker, Hendrik Jones, Deborah J., 505 157, 182, 213%, Krishnamurthi, Lehmann,Donald R..
Helvacian, Nurhan, S., 2-3, 5-6, 196, Jones, F, L., 2136 217 215, 216, 218, 223, Lakshman, 440f. 392, 439, 441, 443
146-147 144f, 148, 212f, Jorgenson, Dale W., 593, 595, 599, 600, 444 Leibowitz, Arlecn, 211f,
Henderson, J. Stephen, 217, 308, 310, 328. 124f, 147, 149, 607, 614, 615, 617, Krueger, Alan B, 2116, 28
384f, 358 329-330, 356f, 237, 242, 247, 250, 624, 626-627, 629, 212f, 218 Leiss, William. 437f.
Hendry, David F.. 305, 387, 358, 439f, 296f, 297F, 2986, 633-635, 645-649, Krugman, Pau! R., 96f, 444
583f, 588 443, 501 299, 301, 302, 453, 665, 6674, 668F, 99 Lenin, Viadimir I., 306
Herriges, Joseph A., Hsiao, Cheng, 668f, 479-481, 486, 6698, 670f, 675, Kuh, Edwin E., 239, Leone, Robert P., 391,
354f, 357 670f, 674, 675 499f, SO0f, SOI, 676 303 439f, 441, 444
Hesse, Dieter M.. 484, Huang. Chi-fu, 59 502, 503, 504, 506, King, Mervyn A., 297f, Kulatilaka, Nalin, 475, Lerman,RobertI., 607,
50! Hubbard, Elbert 582f, 588 302 504 667f. 672
Hickman,Bert G., (Green), | Joskow, Paul L., 96f, 99 Kirshner, Daniel. 3556, Kuskin, Mark, 178, Lerner, Abba P.. 437f,
296f, 302, 5836, Hudson, Edward A., Jovanovic, Boyan, 176, 358 202, 215 444
585 SOOF, 504 2426, 217, 219 Klaassen, Leo H.. 148, Kuznets, Simon, 2108, Leuthold.Jane H., 638,
Hicks, Sir John R., 227, Hulten, Charles R., Judge, George G.. 56, 591 216 676
2959, 302 231, 296f, 299, 302 57, 97f, 99, 145f, Kleidon,Allan W., 56f. Levy, Haim, 58
Hill, M. N., 216, 223 Hunter, Jerald, 2976, 148, 214f, 218 37 Labrune, Francis, 358 Levy, Robert A.. 558,
Hill, R. Carter, 57, 976, 298f, 302 Kein, Benjamin, 410, Lafontaine, Francine, 58
98, 99, 148, 218 Huss, William R., 349, Kalachek, Edward, 432, 446 500f, 504 Lewis, H. Gregg, 176,
Hirsch, Wemer, 96f, 99. 354f, 358 670f, 676 Klein, Lawrence R., Laidler, David E. W., 178, 179, 202,
Hirschberg, Joseph G., Kaldor, Nicholas, 365, VBF, 296f, 303, 510, 580f, 585, 588 2U3Ef, 218, 6671,
310, 356, 357 Ibbotson, Roger, 55f, 57 444 550, 575. 576, Lakonishok, Josef, 56f, 676, 677
Hirschey, Mark, 441f, Intriligator, Michael D., Kalish, Shlomo, 440f, S79F, SBOE, 583f. 57 Lieberman, Marvin B.,
443 301, 304, 305, 503, S8df, 588 Lambin, Jean-Jacques, 96f, 99
Hirschleifer, Jack, 212f, 504, 587, 675, 677 Kamalich, Richard, Kline, Stephen, 437f, 438f, 444, 447 Lilien, Gary, 447
217 Irish, Margaret, 670f, 175, 191, 218 444 Lancaster, Kelvin, 149 Lindsay, Robert, 302
Hirschmann,Wilfred Kaysen, Carl, 146f, 147, Kmenta, Jan, 437, 440f, Lancaster, Kent M., Lintner, John, 550, 58
B., 96f, 99 Isachsen, Arne 3., 669f, 312, 3546, 355f, 444, 4991, 504, 44, 442 Lipsey, Richard G..
Hodge, James H.. 147 675 357 305, 5836, 589 Lancaster. Tony, 2126, 518, 524, $26.
Hogben, Lancelot, 329 Keeley, Michael C., Kaiesner, Thomas J. 218 S8If, 588
Holland, Daniel M., James, S. F., 19F 670f, 676 204f, 218, 222, Landau, Ralph, 124f. Little, John D. C., 414,
297, 302 Jefferson, Thomas, 361 Kehver, Ken, 677 592, 679 147, 149, 302 415, 448
8 AuthorIndex Author Index 689

Litzenberger, Robert Malkiel, Burton G., 20, McGowan,John J., Modigiliani, Franco, Myers, John G., 447 Newbold, Paul, 439f,
HL, 59 550, 566, 58, 59 145£, 1461, 147 S550, 58, 269, 2978. Myers, Kenneth H.. 442
Liviatan, Nissan, 395, Mailar, Chris, 670f, 679 McGuinness, Tony, 299, 303, $39, 4371, 444 Newman,Peter, 445
444 Mankiw, N. Gregory, 439f, 442 S82f, 589 Myers, Stewart C., $4f, Nicholls, William H..
Lloyd, Cynthia B., 2134, $39, 589 McNown,Robert F., Moff, Robert A., 621, 358, 566, 57, 297F, 424, 445
258, 223, 676 Mann, Henry B., 58éf, 666f, 671 657, 6676, 668f, 302 Nicholson, Walter, 96f,
Lorenz, Wilhelm. 2118. 589 Medoff, James, 176, 6701, 676, 677 100
221 Manser, Marilyn E., 219 Montgomery, David B.. Nadisi, M. ishag, 252, Nickell, Stephen J.,
Lorie, James, 406 676 Mellow, Wesley, 670f, 96f, 97f, 98, 99, 297F, 298f, 300, 296f, 303
Lovell, C. A. Knox, Manser, Martin H., 18f 676 439f, 440f, 444 302, 485, 505 Niemi, Beth T.. 213f,
482, 503 Mantel, Nathan, 668f, Melrose, Kendrick B., Montmarquette, Naert, Philippe A.. 218, 223
Lucas, Robert E., Jr., 672 380, 444 Claude, 391, 501 37, 438F, 442, Norsworthy,
270, 298f, 303, Marden, William D., Meltzer, Aian N., 303, Morgan, Chris, 18f 444 Randolph, 500f,
455, 504, 507, 509, 2126 218 447, 588, 590 Morgan, James N.. Nakamura, Alice, 593, 505
527, 530, 531, 534, Markowitz, Harry M., Menill, Lynch,Pierce, 210f, 219, 667f, 677 629, 633-635, 637, Norton,Peter, 19f
536, $37, 549, 23, 24, 55f, 58, 59 Fenner and Smith, Moriguchi, C., 391, 444 638, 640, 641, 643,
582f, 588, $89, 592 Marsh, Terry A... 56f, 39 Morimune, Kimio, 6671, 669f. 670f. Oaxaca. Ronald, 182,
Lundberg, Shelly, 610, ST Messer. Karen, 208, A38f, 444 677, 678 188, 191, 203-208,
676 219 Nakamura, Masao, 593, 20

waist
Marshall,Alfred, 450 Morley, J. Kentfield, 20
Lutkepohl, Helmut, 57, Martin, Stephen, 441f, Metcalf, Charles B.. Moroney, John R.. 357, 629, 633-635, 637, Oberhofer, Walter,
99, 148, 218 444 670, 677 359, 502 638, 640, 641, 643, 499f, 505, 5831,
Lydall, Harold, 210f, Mason, William, 166, Meyer, John R., 239, Morison, Catherine J.. 6671, 6691. 670f. 589
218 217 303 484, 485, SOIT, 677, 678 O'Connor, Charles J.,
Massy, William F., 371, Meyer, Lawrence H., 502, 504, 505, 506 Nathan,Felicia, 211f. 2101, 224
MacBeth, James D.. 44a 674 Moser, Colletta H., 222, 219 Ofek, Haim, 174, 189,
556, 57 Masters, Stanley H., Middleton, Kenneth A., 679 National Bureau of 219
Machlup, Fritz, 150, 218 222, 679 96F, 99 Mount, TimothyD., Economic Ohta, Makoto, 130.
MacKinnon, James G.. McCallum, Bennett T.. Milgate, Murray. 445 3550, 358 Research. 102, 148 146F, 148
356f, 357 345, $49, 561, 562. Miller, Merton H., Mroz, Thomas A., 595, Navasky,Victor, 18f Oi, Walter, 159, 210f,
MaCurdy, Thomas, 563, 5B1f, S82, 2976, 303 635, 637, 638-644, Naylor, Thomas H., 210 219
179, 218, 615, 617, 589, 592 Miller, P. B., 414, 445 650, 652, 661, 663, 96f, 9 Okun, Arthur M., 507,
629, 633, 634, McCann, John M., 387, Mills, Gordon, 590 6691, 677 Nefici, Salih, 582f, 589 540, 541, 583f.
6678, 6686, 6708, 442 Milne, A. A.. 116 Muellbauer, John M.. Nelson,Charles R.. 56f, 589, 590
675, 616 McCollum, James F., Mincer, Jacob, 152, 606, 670f. 673 58, 289, 303, 308, Olsen, Randall J.. 623,
Madansky, Albert, 584f, $16, 586 * 155, 163, 168, 169, Mueller, M. G., 300 310, 3141, 350, 661, 663, 6680, 678
589 McConnell, Campbell 172, 174, 176, 188, Muller, Eitan, 414, 444 352, 353, 354f. Olson, Kenneth, t
Maddala, G. S., 549, R.. 180, 218, 393, 189, 196, 208, Mundlak,Yair. 391, 359, 360 Orcutt, Guy H., 3-5.
553K, SB4P, 589, 446, 679 206, 24208 2138, 444, 449, 503 Nelson,Forrest 0.. 18f, 19f
668F, 6708, 6711. MeDonald, John F., 219, 610, 645, 646, Murphy, Kevin M.. 668f, 678 Oswald, Andrew J.,
676 96F, 99, 621, 657, 667f, 670F. 676, 172, 187. 219, 410, Nelson, Randy A., SOIL. 179, 219
Mahajan, Vijay, 414, 668f, 676 677 432, 446 505 Otuteye. Eben, 440f,
444 McElroy, Marjorie B., Minhas, Bagichs, 452, Murray, Michael P., Nerlove, Mare, 81-84, 445
Majd, Saman, 966, 99 55f, $8, 471, 504, 501 354f, 358 87-94, 99, 100, Our,Tae H., 407, 436,
Majluf, Nicholas, 60, 677 Mishkin, Frederic, 549, Musgrave,John C.. 237, 303, 365, 401, 441
96f, 99 McFadden, Daniel, 334, 589 228f, 296, 301 417, a37f, 444,
Makridakis, Spyros G., 355F, 3568, 357, Mitchell, Bridger M., Mussa, Michael, 298f, 303 454, 457, 505, Packard, Vance, 437f,
386f, 358, 359 358, 449, 499F, 328, 358 Muth, John F.. 268, 583f, 589 445
Malinvaud, Edmond, SOOF, 503, 504, Mitchell, Olivia S., 610, 298f, 303. 319, Neumark, David, 213f, Padilla, Arthur H..
345, 589 306, 670f, 677 677 358, 536, 589 219 2146, 218
 Author Index Author Index 691

Pagan, Adrian R., 305 Pierce, David A., 382, Rasche, Robert, 269, Ross, Stephen A., 37, Schmalensec, Richard, Silverstein, Gerald,
Palda, Kristian S., 380, 4398, 445 299 38, 52, 55f, $7, 58 361, 365, 368-372, 228f, 296f, 301
390, 402, 438f, Pindyck, Robert S., 96f, Rasmussen, Arne, 365, Rotemberg, Julio R., 374, 376, 380, 383, Simon, Hermann, 414,
440, 445 98F, 99, 100, 356f, 445 484, 505, 539, 393-396, 398, 410, 446
Palm, Franz, 264, 297f, 359, 484, 505, Rea, Samuel A., 607, 583f, 589, 590 415, 427, 429, 430, Simon,Julian L., 370,
304 583f, 590 667f, 672 Rothenberg. ThomasJ., A37E, 438F, 4391, 414, 4381, 4398,
Papademos, Lucas, Plourde, Andre, 356f, Reder, Melvin W.. 467, 505 440F, 445 AAIE, 446
$82f, 589 359 214f, 220 Rothschild, Michael, Schmidt, Peter, 179, Simon, Leonard S.,
Park, Joon Y., 50Of, Pogue, Gerald A., 55f, Rees, Albert, 172, 214, 2968. 298f. 301, 213f, 220 SOF, 443
505 38 217, 222, $79, 403 Schneider, Lynne, 410, Sims, Christopher A..
Park, Rolla E., 358 Poirier, Dale, 670f, 679 587, 598L, 666f, Rowe, Robert D., 391, AIL, 432, 446 2976, 304, 4391,
Park, Soo-Bin, 584f, Polachek, Solomon, 674, 679 445 Schoenberg, E. H.. 380, 4401, 446, 582,
391 174, 175, 188, 191, Rehn, Gésta, 580, 587 Rowley, John C. R., 410, 446 390
Parkin, J. Michael, 524, 211, 213f, 214f, Reimers, Cordelia, 187, 295f, 303 Schoenberg, Erika, 614, Singer, Burton. 212f,
526, S80f, 581f, 207, 218, 219 220 Roy, A. D., 165, 220 678 217
588 Pollak, Robert A., 143f, Reinganum, Marc R., Ruback, Richard S., Scholes, Myron, 38, Singleton, Kenneth J.,
Parnes, Herbert S., 148, 189, 215 58 S6f, 58 55f, 56 2981, 301, 564,
2108, 219, 667f, Pollay, Richard W.. Renahack, Robert, 315, Rubinfeld, Daniel L.. Schudson, Michael, $831, 587
678 404, 445, 448 3548, 3558, 359 986, 100, 356f, 359 437f, 446 Sinquefield, Rex, S5f.
Parsons, Leonard J., Porter, Michael E., 96f, Reynolds, Lloyd G., Russell, R. Robert, 96f, Schultz, George P., 586 37
380, 438f, 440f, 100 679 100 Schulz,Henry, 405, Smeeding, Timothy,
442, 443, 448 Poterba, James M., Rhodes, George F., Jr., Ruud,Paul, 667£, 6691, 446 241f, 220
Patterson, J. Wayne, 2971, 303, 581f, 590 667f, 673 675 Schultz, Randall L., |. Seymour, 56f. 57
96f, 100 Prais, S. J., 196. 356f, Riddell, W. Craig, 222, Ryan, David, 356f, 359 4391, 440f, 443, , Adam, 150, 152.
Pazderka, Bohumir, 357 679 444, 448 210F, 220, 593, 678
407, 436, 441 Prucha,IngmarR., 485, Riley, John G.. 210f Salinger, Michacl A., Schultz, T. Paul, 633, ‘Smith, James P.. 186,
Pechman, Joscph A., 505, 578, 590 Rizzuto, Ronald, 211. 297f, 303 634, 638. 667f, 678 187, 2136,
674 Psacharopoulos. 220 Samuelson,Paul A., Schultz. Theodore W.. 666. 670,
Peck, Stephen C., 308, George, 171, I71f, Robb, A.Leslie, 500f, 97f, 100, 498f, 505, #52, 220, 671 674, 675, 678
310, 311f, 350, 212f, 220 504 517, 518, $80f, $90 Scarle, Allan D., 96f, Smith, Robert S., 210£.
352, 353, 354f, Puig, Carlos, 355f, 358 Roberts, Harry V., 191, Sandmo, Agnar, 669, 100 221, 222, $92,
359, 360 Pulley, Lawrence, 358 215, 405-407, 678 Selickacrts, Willy, 302 6671, 673, 679
Pencavel, John, 79, Purdy, David L., 385 440f, 445 Sargan, J. Denis, 305, Shapiro, Carl, 222, 220 Smith,Shirley J., 666f.
218, 220, 645, Robins, Philip K., 670f, 4998, 505, 590 Sharpe, William F., 40,
666f, 678 Quandt, Richard, 380, 67% Sargent, Thomas J., S46, 55f, 561, SB Smith, V. Kerry, 81, 98
Perkins, J. W., S84f, A38f, 445 Roester. Theodore W., 2986, 303, SOIf, Sheffrin, Steven M., Snetsinger, Douglas W..
585 Quigley, John M., 211f, 393, 446 503, 507, 549, 549, 591 407, 436, 441
Perloff, Jeffrey M., 540, 217 Roll, Richard, 55f, 566, S8If, 5821, 589, Shell, Karl, 143f, 147, Solnik, Bruno, 55f, $8
590 57, 58 590, 592 371, 443 Solow, Rober M., 363.
Perry, George L., 581, Raines, Fredric, 670f, Rose, Nancy L.,. 96f, 99 Sarnat, Marshall, 58 Shoven, John B., 504 446, 452, 301. 507,
5820, 583f, 590 676 Rosen, Harvey S., 629, Sasieni, Maurice W.. Sicktes, Robin C., 482, 517, 518, S80f,
Petly, Sir William, 152 Raj, S. P., 440%, 444 633, 644, 66d, 678 414, 445 503 S81f, 590, 591
Phelps, Edmund S., Rao, Ambar G., 414, Rosen, Sherwin, 130, Savin, N. Eugene, 479, Siebert, Calvin D., 297f, Sorrentino, Constance,
521, $89, 590 440f, 442, 445 148, 2108. 211f, SO0f, 502 666f, 678
Phillips. A. William, Rao, Ram C., 440f, 445 2U2F, 214, 216, Sawa, Takamitsu, 584f, Siebert, W. Stanley, Spady, Richard H., 481.
512, 513, 515, Rapping, Leonard J., 220, 372, 445 591 2Ibf, 214, 222, 505
579f, 580f, 590 96f, 100, 309, 527, Rosenberg. Nathan, 101 Scarf, Herbert E., 504 592, 667f, 671, 679 Spann, Robert, 358
Phillips, Peter C. B., 530, 531, 534, Rosett, Richard N., Scherer, F. Michael, Silk, Alvin J., 415, 439f, Spence, A. Michael, 60,
SO0F, 505, 545, 590 582f, 588 668F, 678 956, 100 AAOf, 442, 444, 445 96f, 100, 157, 221
92 Author index Author Index 693

Spiegelman, Robert G., Taylor, John B., 224, Toyama, Nate. 359 Wachtel, Paul, 2118, Wayne, Kenneth, 671, Winer, Russel S., 439,
670f, 676 301, 359, 510, $41, Treadway, ArthurB., 220 96f, 98 447
Spitzer, John 3. 145f, $49, 582F, 583f, 270, 298f, 304 Wachter, Michacl L.. Webb, Aifted, 675 Winfrey, Robert, 227,
146f, 148 586, $87, 591 Tretheway. Michael W., 540, 582f, 590, 591 Weil, Robman, 406 296f, 304
Stafford, Frank P,, 211f, Taylor, Lester D., 315, 482, 502 ‘Wagner, Joachim, 21 If, Weinberg, Charles B., Wise, David A., 210F.
204f, 217, 221 321, 323, 328, 343, Treynor, Jack L., 55f. 221 391, 392, 447 222, 631, 675
Steiner, Peter O.. 365. 354f, 355f, 359, 59 ‘Wagner, Wayne,32. 59 Weisbrod, Burton, 240f, Witteveen, Hendrik J.
442 397, 439f, 443, 446 Triggs, C. M., 439%, 442 Wakeman,Frederic, 148, 591
Stern, Nicholas, 639, Telser, Lester G., 365, Triptett, Jack E., 120, 437%, 447 Weiserbs, Daniel, 397, Wold, Herman, 577,
678 380, 384, 410, 411. 123, 124f, 128, Wald, Abraham, 584f 439%, 446 585
Stevens, Joseph C,, 438f, 4408, 446 1446, 1456, 148, Wales, Terence J., 210f, Weiss, Doyle L., 391, Wolf, Barbara L., 211,
4381, 446 Teng, Sinn-Tsair, 440f, 149, 217, 219, 220, 211f, 221, 355f, 392, 439f, 443, 447 221
Stigler, George J., 440f, 446 614 359, 500f, 503, Weiss, Yoram, 156, Wolfe, JN. 443
446 Thaler, Richard H.. 56f, Trivedi, Pravin K.. 629, 631-634, 157, 215, 222, Wolpin. Kenneth, 210f,
Stigler, Stephen M., 149 58, 59 295f, 303 669F, 679 6701, 679 22
Stiglitz, Joseph,96f, 98, ‘Theeuwes,Jules, 670f. Turnovsky, Steven J., Walker, lan, 670f, 672, Welch,Finis, 171f, 172, Wormer. N. Keith, 96f,
2108, 2L1f, 2126, 678 269, 299, 582f, 391 678 186, 187, 2121, 100
220, 221 Theil, Henri, 97f, 100, Twain, Mark, 20, 59 Wail, Kent D., 55f. 57 213f, 219, 221, Woo, Chi-Keung. 354f,
Stoker, Thomas M., 298f, 304, 355, ‘Tyrrell, Timothy J., Wallis, Kenneth F., 222 359
S796, 591 359, 391, 446, 3550, 358 295f, 298f, 304, Wenz, Kenneth L.. 96f, Wuod. David O.. 297F,
Stone, Richard, 116, 5530, SB4f, 591, 4400, 447, 525, 99 301, 449, 430, 472,
144f, 148 392 Udis, Bernard, 666f, 671 527, 549, 580f, West, Richard W., 670f, 486, 487, SOIL.
Strauss, Lewis L., 306, Thomas, Annie, 95f, 98 Uhler, Robert G., 311f, S81f, 585, 591, 676 502
359 Thomas, Howard, 99 353, 354f, 359 592 ‘Weyant, John P., 360 Woodland, Alan D.,
Strauss, Robert, 179, Thomas, R. L.. 295f, Ulph, David T., 667f, Walsh, John R., 210f, Wheelwright, Steven C.. 355K, 359, S0OF.
220 304 669f, 671 221 356f, 358, 359 506, 629, 631-634,
Stromm,Steiner, 669f, Thompson, Gerald L., Usher, Dan, 302, 304 Walters, Alan A., 101, Whitaker, John K., 669f. 679
675 440f, 446 Uzawa, Hirofumi, 499f, 4998, 506 590 Wouk, Herman,
Strowz, Robert H.. 270, Thurow, Lester, 170, 506 Walwick, Nancy J., White, Halbert, 53, 59, est
300 182, 2128, 2138, 515, 591 208, 209, 219, 222, Woytinsky, Wladimir.
Summers, Anita A.. 2148, 221 Vandaele, Walter, 356f, Ward, Michacl, 666f, 299, 356f, 359. 6671, 679
2016 221 ‘Thursby, Jerry G., 475, 359 678 629, 640, 657. 661, Wu, De-Min, 4381, 447,
Summers, Lawrence H., 501 Varian, Hal R.. 96f, Watkins, G. Campbell, 662. 6691, 679 584f, $91, 669F.
2U1f, 2N2E 218, Thurstone, L. L.. 95f, 100 501f, 502 White, Kenneth J., 679
297f, 304, 522, 100 Veall, Michael R., SOOf, Watson, Mark W., 580f Svor, 504 Wykoff, Frank, 231.
539, 580f, SBE Tidwell, Paul W., 1456, Watson, Thomas J.. 1, Wildasin, David, 297f, 296f, 302
585, 589, 590, 591, 146 Verdon, Walter A., 393, 18f 304 Wyplosz, Charles, 297f,
6671, 672 Tinbergen, Jan, 144f, 446, ‘Wats, Harold W.. 135, Wilkes, Maurice V.. 299
Swan, Craig, 2951, 304 148, 2981, 304, Verleger, Philip K., Jr., 149, 596, 614, it
Swinton, David, 2134, $50, 579f, 583f, 355f, 359 666F, 6636, 6701, Wilkinson, Maurice, Yao, Dennis A.. 96f,
221 50 Verma, Vinod K., 372. 672, 674, 679 96f, 100 98
Tirole, Jean. 96f. 99, 447 ‘Waugh,Frederick V.. Willis, Robert, 162, Yelle, Louis E.. 95f.
Taubman, Paul, 166, 100 Vernon, John M., 96f, 104, 106-110, 167, 171f, 172, 100
189, 2108, 211f, Tobin, James M., 55f, 99 132-135, 142, 208, 210f, 212f, Yellen, Janet L., 211f,
215, 221 59, 256, 258-259, Viscusi, W. Kip. 216f, 144f, 145£, 1466, 222 212f, 214, 222
Tauchen, George, 304, 582f, 591, 221 149, 401, 417, Wilson, Thomas A.. Yoshikawa, Hiroshi,
592 668F, 678 von Furstenberg, 4378, 444 4401, 441f, 442 260, 297f, 304
Taylor, Jim, 580f, Toevs, Alden L., 500f, George M., 297f, Waverman, Leonard, Windal, Pierre, 391, Youmans, Kenwood C.,
591 505 304 SOIL, $02 392, 439, 447 213f, 217
1 Author Index
Zabalza, Antoni, 632, Zeliner, Arnold, 214f, 506, 522, S8tf,
669F, 676, 222, 264, 297f, S84f, 584, 591, ]
679 304, 355f, 359, 592 q
Zeides, Stephen P., 539. 391, 438f, 4391, Zimmenman,Martin
Subject Index
591 441, 447, 4998, L., 96f, 100
ies:
distribution of, 161
as omitted variables, 166-
167
Accelerator model of
investment:
alternative distributed lags
for, 234-237
empirical implementations,
237-239, 281-283,
291-292
flexible, 234
naive, 233
theory, 233-234
Adaptive expectations, See
Expectations
Added worker hypothesis, See
Labor supply
Adding-upconditions ofshare
‘equations, 470-474,
478-479, 491-492
Adjusiment cost models, 269—
270, 484-485
Advertising:
as barrierto entry, 415,
brand loyalty model of, 388,
392, 423-424
consistent market share
specifications, 383-385
to counteract cyclical
TMacrocconomic
fluctuations, 363
cumulative effect on sales,
387-390, 401-404
current elfects models of,
387-389, 423
distributed lag models of,
385-389, 423
duration levels, 389-393
effects on aggregate
consumption, 393-400,
427-429
effects ofban on cigarette
consumption, 409-414,
432-436
expenditures, 362, 373
frequency. 374
Granger causality with sates,
380-383, 427-429
importance ofcopy quality,
407-409, 436-437
as information, 372
Fingering effects models of,
386-393, 423,
measurementof, 373-375
optimal advertising-sales
ratios, 366-371
role ia changing consumers’
tastes, 371-372
rote ofregression analysis in
analyzing, 416
scale economics in, 415-416
shape of advertising
response function, $14~
als
simultancity with sales,
375-380, 417-420,
429-432
stock of, 435
temporal aggregation and
data interval bias, 389~
393
theoretical determinants of
advertising outlay, 365-
at
theoretical effects on sales,
31-373
Arbitrage pricing model, 37-
38, 52-53
AREMOS/PC, 16, 17
ASCII code, 7
Autocorrelation:
Breusch-Godireytest for,
$67-570
Cochrane-Oreutt estimation
procedure, 5, 54, 92,
93, 283, 345, 348, 356f,
388, 558
Durbin-Watsontest for, 54,
92, 347, 497
Durbin’s m- and h-test
procedure, 235, 347,
534, 567-568
estimation with lagged
dependentvariable.
281-283, 294-295,
520, 423- 424, $52
Hitdeeth-Lu estimation
procedure, 54, 92, 93,
253, 282-283, 345,
347, 388, 424, 558
local-giobat problems, 348
maximum likelihood
estimation procedure,
345, 348
Prais-Winsten estimation
procedure, 356f
with simmultaneity, 291-292,
432, 534-535, $52,
586, $58, 561-565. 569
in systemsof equations,
476-479, 497
testing with lagged
dependentvariable,
235, 281-283, 294~
295, 347, 404, $34-
$35, 556, 567-570
vector, 477-479
Autoregressive integrated
moving average
(ARIMA)models, 268,
289-291, 353, 382,
427-829
695
 396 Subject Index
Auniliary regression equation,
16-77, 322
BASSSTAT.16, 17
Bayesian procedures, 39
Bertrandoligopoly model,
69
Beta value:
definition, 27, 33
different in January, 50-51
estimation of, 34-41
exemstudies, 48-50
for gold, 45
stability, 35-36, 48
Blinder, Atan, [85-186
BMDP/PC,16, 17
Boston Consulting Group, 60,
960
Box-Coxtransformation, 128,
142-143, 211f, 231
Box-Jenkins ARIMA models,
268, 289-291, 319-
320, 352~353, 382-
383, 390-400, 564
Box-Pierce Q statistic, 290,
Sox-Tidwell transformation,
128
Brand loyalty model, See
Advertising
‘CADET, 246
Capital asset pricing model:
beta value, 27, 33-34
estimation of, 34~41
sisk-return linear
telation: F=32
as special case of APM, 52-
33
test of, 37-39
use in constructing
portfolios, 47
Capital stock:
constant exponential decay,
228-232
detetioration and
depreciation, 230-232
gr055, 228
mortalitydistributions, 227
net, 228
one-hoss shay deterioration,
perpetual inventory
method, 228
Cash flow model of
investment, 239-242,
284-286, 5St
Causality, See Granger
causality
Censored data, 595, 616, 6666
‘Center for Research on
Securities’ Prices, 37
CHANCE,12, 19f
Chow test, 48, $1, 93-94,
139-141, 201-208,
525-526
Cigarette advertising ban,
effects of, 409-414,
432-436
Cobb-Dougias production
function, 68-75, 248-
256, 266, 286-287,
450-452, 488-489
Cochrane-Orcutt procedure,
See Autocorretation
Coefficientofdetermination:
Partial, 108
relationship to R?, 108-110
separate, 109
Compensated substitution
effect, 604, 65-008
Computer commands and
operations:
BASIC,10
CD(change directory), 9
COPY, 8-9
DISKCOPY,8
DOS(disk operating
system), 8
EDLIN,10
FORMAT,6
MD(inake directory), 9
operating system, &
pathnames, 8-9
RENAME. 9
texteditor, 10
TYPE,10
word processors, 10
XCOPY, &
‘Computers in econometrics:
briefhistory, 2-6,
Pioneering studies, 340
Conoco takeaver, 48-50
Constant elasticity of
substitution (CES)
production function,
252-254, 292-294,
452-457, 488-489, $30
Consumption function, $51~
555, 566, 370-572, 576
Cost function:
briefhistory of functional
formsfor, 457-458
choice between production
and cost function, 457-
460
general, 63
generalized Leontief, 460-
465
multiproduct, 481-482
Iransiog. 81, 470-475, 483
variable cost function, 482-
434
Cournotoligopoly model, 369
Cowles Commission, 2, 21
Cross-correlograms, 399, 428-
any
Current effects model, See
Advertising
DATA-FIT, 16, 17
Data interval bias, 389-393
Dirichlet distribution, 500f
Discrimination,effects on
wages:
by gender, [8B-190, 205
by race, 181-187, 203-205
theory and measurement
issues, 180-184
Discouraged worker
hypothesis. See Labor
supply
Distributed lags:
accelerator model, 234-237,
281-283
. advertising-sales models,
385-389, 401-408
Almonlag, 236, 284-289
with autocorrelation, 292-
294, 316
cash flow model, 241-242,
284-286
specification, 236
models, 316-320
geometric, 235, 317, 386
inverted V-Ing, 236
neoclassical model of
invesiment, 249-256
polynomial distributed lag,
236-237, 284-289
rational tag, 237, 582f
with simuhtancity, 291-292
Tobin's ¢ model, 287-289
Double-in-ten formuis, 307,
339, 354f
Duality theory, 65
Dummyvariables, 48-50, 51,
TEI-114, 136-14:
151, 161-166,19:
Durationintervals, 389-393,
421-422
Durbin-Watson test statistic:
test for first order
autocorrelation, 54
use with lagged dependent
variables, 235, 281~
283, 386,
Economies ofscale:
definition, 63
empirical evidence, 81~83
general, 61
imponance of, 61
and shape of average cost
‘curve, 64
six-tenths rute, 61
ECONOMIST
WORKSTATION,16,
7 Error correction process, 529
EDSAC Computer, 5-6, 356
Education,rate of return to:
empirical evidence, 169-
172
stability over time, (71-172
Elasticity ofsubstitution:
for generalized Leantiet
function, 464465,
492-494
Hicks-Alten partial in
multiple input case,
463
production function with
constant, 482-457
short: and long-run, 484
for anslog function, 475,
492-494
two-input case, 452
Electricity demand:
autocorrefationin models
of, 346-348
data guality for analysis of,
34-315
demand charges, 310
double-in-ten formula, 307,
339
empiricat evidence on, 328
335
energy charges. 310
with equipmentstocks
included explicitly,
312-315
with equipment stocks
included indirectly,
315-320
essential facts, 308-312
forecasting, 310, 344-346,
349-353
functional form for models.
of, 326-328
heteroskedastic errors in
modeis of, 34)
marginal and intramarginal
prices, 321-324
multicblock rate schedules,
309, 321-326
short-run model, 312-313
simuhtaneity ofprice and
quantity, 324-326,
38
Typicai Etectric Bills data,
325-326
Enteringdata, 13-15
ET, 16, 17
Event studies, 48-30
Expectations:
adaptive, 267-269, 317~
320, 529-530, $38-
539, 544, 564
rational, 268-269, 319-320,
484-485, 509, $38-
539, 541, $44, $49,
556, 561-568, $79
Experience curves, See
Learningcurves
Feedback, 382
Final form, 399, 428-429
Forecast
advertising-sales
relationships, 398-400,
433
double exponential
smoothing procedure,
350
dynamic, 271, 274-275
Subject Index 697
electricity demand, 307~
308, 344-346, 349-352
exponential growth,
Procedure, 349
‘ex post dynamic, 274-276,
Holt’s two-parameter
exponential smoothing
provedure, 351
investment in fixed business
capital, 271-276, 294-
295
single exponential
smoothing procedure,
380
static, 271
timeseries procedures,
unit costs with learning
curve effects, 94-95
Fourier functionall form, 482
Ful income, 602
Full information maximum
likelihood, 463, 485,
545, 552-554, 556,
573-875,578, 626-
627, 665
Functional form:
Bor-Cox procedure, 128,
142-143, 2116
choice of, 127-128, 161~
162, 326-328
flexible production and cost
models, 482
GAUSS,16, 17
Genera? equilibrium models,
479-481
Generatized Box-Cox
functional form, 482
Generalized Leontief
functional form, 327,
460-455, 478, 489-498
Generalized Tobit model,
Limited dependent
variable models
Geometry mean,
computation of, 194
Gord, 45
Government spendingimpact
muhtiplier, $24
Granger causality, 380-383,
427-429
Gross output, 488

98 Subject Index
iross substitutioneffect, 605- transformation of data to
606 ensure
homoskedasticity, 137,
Jausman specificationtest, 328, 341-52
379-380, 409, 419,
430-432,565-567, 641 , White test
Jazard rate, 622 for, 53, 208-209
leckit model, See Limited Houthakker-Taylor state
dependent variable adjustment model,
models 397-398
Yedonic prices: Humancapital theory:
for advertising copy quality, equalizing differences. 152-
409 153
‘measurementissues in
empirical
wy implementation, 159-
applied to asparagus, 105- 160
110, 132-135, 142 on-the-job training as
143 investment, 155-157
applied to automobiles, general training, 155
110-116 specific training, 155-
applied to computer 156, 178
industry, 118-126, role ofabilities, 155
136-143 schooling as investment,
briefhistory, 105-116 ESI-158
chained, 141-142 serecning as an alternative
incorporated by U.S, hypothesis, 157-158
Bureau of Economic Hypothesis testing:
Analysis, 125 individual tests, 36, 78, 86~
make effects, 129 87, 113, 137
in models of cost and joint tests, 78-79, 86-87,
production, 481—482 T13-EE4, 139-141
as outcome ofsupply- Lagrange multiplier test,
demand relationships, 465-467, 494-496
117118, 129-131 likelihood ratio test, 465-
parameter stability, 12 467, 494-496, 554,
131 $75, 579
test for constantrate of nonlinear, 97f, 475, 534
price change, 139 overidentifying restrictions,
with unobserved attributes, $54, $57, 873-575, $79
128-129 relation to A? vatue, 468-
leteroskedasticity: 469, 519, SBIE
in electricity demand Wald test, 465-867, 494-
equations, 328, 34 496
in labor force particip:
equations, 655 Identification, 376-378, 419-
obtained when using 420, 479, $24-525,
averaged data, 127 552, 556, $70-573,
robust standard errors with, 625-626, 658-660
53, 208, 209, 341, 629, Indirect least squares, 377~
640, 657, 661, 662 379, S71, 660
Info World, 12, 19
functions, 208-209 Information set, $43, 561-565
Subject Index 699
erative three-stage feast measurementissues, $96- Tobit, 596, 620-624, 626~
Instrumental variable squares (13SLS) 599 627, 629, 643, 644,
imation: estimator, 463, 474, second generation studies 650, 657-658
consistency with rational 484, 485, 492, $82, of, 617-637 Limited information
eapectations $53, $56, 573 and taxes, 617, 629-633, maximum likelihoot
hypothesis, 267-268, Iterative Zellner-fficient 644, 663-664, 6697 estimator,552
561-565 (IZEF)estimator, 463, theory, dynamic tabor in prices,
general, 316, 319, 378-379, 474, 488-496, 556 supply, 645-649
395~396, 484-985, theory, household labor Linear probability model, 655
556, 569, 664 Jacobian, 576-578 supply. 607-609 Lingering effects models, Sev
Interaction terms, 163, 165, theory, individual labor Advertising
197-199, 200 Klein's Modet 1 of the US. supply, 600-606 Logit model, See Limited
Intertemporal utility function, economy, 550-558, See also Unemployment dependent variable
527-528, 539-540, $65~567, 572-578 Lagrange multiplier test models
646-648 Kolmogorov-Smimovtest for procedure, 465-467, Lydiametries, 364, 404
Inverse Mills ratio, See normality, 54, 195 474-475, 494-496
sampleselectivity Koyck transformation, 235, Lagrangianfunction, 65, 602, MacUser, 12, 19f
Investment in fixed business 267, 318, 386, 529 MacWeek, 12, 19f
capital: Learning curves: MacWorld, 12, 19f
accelerator model, 234-239, Laborforee, measurementof, autocorrelation, 92 Marginal rate ofsubstitution,
281-283, 295-292 See Unemployment, ‘empirical evidence, 79-80 600-605, 618, 624,
adjustment cost models, labor supply estimation of, 85 645
269-270 Laborforce participation rate, and experience curves, 66
autocorrelation, 293-294 1. 5594-595, forecasting units costs with,
cash flow model, 239-24 94-95
284-286, 551 general, 13, 61-62 estimation, 463, 473-
changes in business added worker hypothesis, integration with Cobb- 475, 494-497, $73-579
ventaries, 226 609-610 Douglas function, 71- Mcanvalue function, 453
‘empirical comparison of allocation oftime 18 Meta-analysis, 392-393
models, 270-277, 294- framework, 610-613 measurementissues, 75-76 MicroTSP, 15, 17
295 backward bending function, ‘omitted variable bias, 76-78 Minflex Laurentfunctional
forecasting, 294-295 606 overview, 66-68 form, 482
gross, net and replacement, comparative dynamics, 647 slope, 67, 80, 85 Minimum chi-square
definitions of, 229 complete budget constraint Lemer index of monopoly estimator, 462-463
moving average errors in, approach to, 632, 635 power Minimumdistance estimator,
267-268 discouraged worker Likelihood test 545,547
neoclassical model, 242~ hypothesis, 609-610, procedure, 465-467. Minimum expected loss
256, 286-287 667F 474-475, 494-496, estimator, 584f
non-static expectations, effects from cyclical $54, 575, 579, 619, 620 MINITAB,16, 17
© 268-269 aggregate demand, LIMDEP, 16, 17 Mode! T Fords, 66-65
putty-clay model, 250-254, 609-610 Limited dependent variable Money illusion, 536, 538
286-287 equilibrium dynamics. models: Monte Carlo techniques, 475,
eesidential construction, 637 general, $9S 573
226 first generation studies of, generalized Tobit, 596, Movingaverage error models,
simuttancity issues, 266— 614-617 627-629, 650, 661- 20, 413, 427
267, 291-292 Frisch demand notion 664, 665
time series/autoregressive implications ofcosts of Heckit, $96, 627-629,651, 64,
model, 263-265, 283- employment, 613-614, 642-644, 650, 661-664 Multicollinearity, 136, 319,
284, 289-291 627-629, 666 logit, 333, 596, 650, 654- 339
Tobin's q model, 256-263, linearized budget constraint, 657 Multifactor productivity
287-289 631, 63: Brobit, 596, 619-620, 623, growth, 485-486, 497-
Iterative minimum distance male chauvinist model, 607, 625, 627, 629, 641, 498
estimator, $45, $47- 667t 642, 650, 654-658, 668 MYSTAT,16
549, 579
00 Subject Index
datural rate of
unemployment, 522
Acoctassical model of
investment:
‘empirical implementation,
248-256, 286-287
theory, 243-245
user cost ofcapital, 245—
248
Neo-Keynesian models, $39,
S41
Nested logit model, 333
Net marginat revenue
product, 366
New classical school, 538
Non-aceelerating inflation rate
of unemployment, 540
Nonhomothetic functions, 469
Nonlinearleast squares, 377,
418, 434, 626
Normality:
chi-square goodness of fit
teat, 54, 195
Kolomogorov-Smirnov test
procedure, 54, 195
‘Okun's Law, $40
Omitted variable bias:
auxiliary regression, 76-77,
322, 344
in clectricity demand
context, 321-324, 343-
a 105
Jearning curves, 76-78, 86
retums to schooling, 166-
167
On-the-job usining. rate of
relum to:
affected by intermittent
labor force
participation, 174
empiricat evidence, 172-
196
‘Overidentifying restrictions.
test for, $54, 337, 573~
575, 579
Panel data, 167, 649, 668f
Partial adjustment hypothesis,
315, 320, 346-348, 408
PC-GIVE, 16, 17
PC Magazine, 12, 190
PC-TSP, 15, 17, 18, 348
PC Week, 12, 196
PDQ, 16. 17
Permanent income measure,
314, 647-648
Phiblips curve:
carly history, 312-517
expectations augmented,
precursors to, 516-517
stability, 517-527
underlying theory, 512-513,
518-520
Pinkham,Lydia E. Medicine
Company, 402, 420-
424
Pooled crass-section, time
series estimation, 436—
4ayT
Ponfolio:
aggressive, 14
defensive, 34
diversification, 23-27, 32
efficient, 31
optimal, 27
Potentiat output, 522, $41
Price elasticities, 464-165,
Price indexes:
chained priceindexes, 141-
142
quality change via matched
model procedure, 101—
See also Hedonic prive
indexes
Probit models, Sec Limited
dependent variable
models
Production function:
briefhistoricat review of
nal forms for,
choice between production
and cost function, 457--
460
‘Cobb-Douglas, 68-75, 248-
256, 450-452
constantelasticity of
substitution (CES)
function, 452-457
definition, 63
integration with learning
curves, 71-75,
PSTAT, 16, 17
R? measure:
changes in to conduct
hyportiesis tests, 519.
Seif
high value docs not imply
valid model, 499
interpretation in CAPM
framework, 40, 47
interpretation in linear
probability model, 655
relation to correlation
coefficient, 131-135
relation to partial and
separate coefficients of
determination, 108-
409, 131-135
in reverse regressions, 456,
489
in simple and mukipte
regressions, 86-87,
131-135
in systems of equations,
468-469, 496
value in regression ofactual
on fitted y. 135
Rate of retum:
definition, 21-22
marginal return, 26
marginal variance, 26
risk-free, 22
risk premium, 22
to schooling, 161-163
Rational expectations, See
Expectations
Rational predictor, 319
RATS, 15, 17
Reduced form model:
to compute indrect feast
squares estimates, 571
equivalence to structural
estimates, 573-575
gencral, 376-380, 533, 535,
$84, $45, 555-555, 556
tase in forecasting or
simulation, 526-527
use in fabor supply models,
624-626, 658-660
Reservation wage, 603, 613,
616, G18, 625, 645
Residential End-Use Energy
Planning System
(REEPS), 332-335
Returns to scale:
for Cobb-Douglas function,69
definition, 63
empirical evidence, 80-83
estimation of, 87-94
and shape ofaverage cost
for translog function, 476
Reverse regression, 45, 134,
191-192,455-456
Risk:
definition, 22
linearrelationship with
return, 27-32
market, 32
and portfolio diversification,
2-27
premium, 22
tiskefree return, 22
specif, 32
systematic, 32
unique, 32
unsystematic, 32
Root mean squared error:
to evaluate forecasts, 272-
277, 291, 294-298
Sales:
effects from advertising,
371-373, 393-400
Granger causality with
advertising, 380-383
See also Advertising
Sample selectivity:
general, SBIf, 595
inverse Milts ratio, 191,
622, 627, 642, 661~
664, 670F
in labor supply equations,
616-629, 642-644,
661-666
in wage rate equations,
190
SAS/STAT,16, 17
Saturation model, 313
SCA SYSTEM, 16, 18
Screening hypothesis of
education, 157-158
Scemingly unrelated
estimator, 462-263,
S71, S74
Setection biss-corrected
tegression, 621-623
Shadow values, 484
Share models:
adding-up conditions for,
470-471, 491-492
autocorrelationin, 477-479
consistent specification in
advertising context,
383-385, 424-426
in cost functions, 470-474,
491-492,
SHAZAM, 15, 18
Shephard's Lemma, 461, 370,
983, 485
Simultaneous equations
estimation:
in advertising-sales models,
378-380, 429-432
alternative estimators, $50-
558, $70-$73
with autocorrelation, 291-
292, 569
in cost and production
models, 474-475
in electricity demand
in investment models, 291~
292
Singular equation systems,
345, 472-474, 478-
479, 491-492
Slutsky equation, 606
SORITEC,15, 18
Sorting data. 195, 652-653
SPSS/PC +, 16, 18
SST,16, 18
STATA, 16, 18
STATGRAHICS,16, (8
Statistical software programs:
data, VL
data terminator, U1
header, 11
overview, 15-17
reviews of, 12
vendors for, 17-18
STATPRO,16, 18
STATVIEW 512+, 16, 18
‘Structural equation modet,
375, 380, $45, 558,
570, $73-875
Structurally recursive models,
556, 575-578
Successiveleast squares, 556,
317
Subject Index 701
Symmetric generalized
‘McFadden functions!
form, 482
SYSTAT,16, 17
Taylor's series expansions, $53
Technical progress:
biased, 498
disembodied, 485, 397
embodied, 486
and the measurement of
multifactor
productivity growth,
485486, 497198
and shifts in average cost
curve, 64-65
Temporal aggregation, 389~
393, 420-422
Theil’s incquality coefficient,
295
‘Three Mite Island, 48-50
Three-stage least squares, 463,
474-475, 485, 492,
$45, 552, 553, 556,
564, 570, 573
‘Time SeriesAutoregressive
investment models:
empitical implementation,
24-265, 283-284,
249-291
theoretical foundations,
263-264
Tobin's g model of
investment:
average vs. marginal g, 259-
261
empirical implementation,
259-263, 280, 287-289
theory, 257-259
Tobit model, See Limited
dependent variable
models
Transfer function, 428-829
Transitory income, 647-648
Translogfunction, 81, 327.
469-476, 480, 489-398
Truncated data, $95, 666F
Truncated normal repression
equation, 654
‘TSP, 15, $48, $53, 555f
‘Two-stage least squares:
in advertising-sales models,
379-340, 398, 417-
420, 432-433
 eo

702 Subject Index

Two-stage least squares (cont) Fefation to suycessiye least effects of cyclical variations,
with autgcorrelation, 432, Squares estimator, 577 192
534-535 efficiency wages, 175, 192
consistency with rational Uncompensated substitution empirical evidence on, 167~
‘expectations effect, 605-608, 639- 179
hypothesis, 267-268, G44, 654, 662 estimating experience-
561-565 Unemployment, measurement carnings profiles, 199--
in cost of production of, 530-512, $31-532, 200
function content, 463, oo7t human capital theory, 152+
474 User cost ofcapit
difference from two-step definition, 245 sample selectivity issues, 191
ordinary east squares expected capital gains term, sercening hypothesis, 157-
estimation, 420, 5$8— 248 188
561 measurement of, 245-248 union-nonunion relative
general, $52 wages, 176-179, 196—
in investment models, 291- Value-added output, 188 197, 201-203
2 Virtual income, 631, 663 Wald test procedure. 465-467,
in labor supply models, 474, 194-196
640 Wage rates, determinants off WeberFechner law of
in macro-cconomenic differences forsingle and psychophysics, 372
models, 545, 552-554, married women, 197
556-567, 570-573 198 Zelingr-cfficient estimator,
in Phillips curve context, discrimination as source, 462-463, 474, 488—
525, $33-535 180-190 492, 545, S71, 578