Macroeconomics Canadian 5th Edition

Williamson Solutions Manual

Visit to download the full and correct content document: https://testbankdeal.com/dow
nload/macroeconomics-canadian-5th-edition-williamson-solutions-manual/
 CHAPTER 10
Credit Market Imperfections: Credit Frictions,
Financial Crises, and Social Security

KEY IDEAS IN THIS CHAPTER

1. The ideas of Chapter 9 are extended to cases where Ricardian equivalence may not
hold, and government debt can matter.

2. Two key credit market frictions are asymmetric information and limited commitment.
There is asymmetric information when participants in a particular market, or the two
parties to a particular exchange have different information. In this chapter we are
particularly interested in asymmetric information in credit markets – situations where
borrowers know more about their credit-worthiness than do lenders. Limited
commitment refers to situations where the party to a particular contract is not
committed to fulfilling the terms of that contract in the future. In loan contracts, limited
commitment means that the borrower always has the option of not repaying the loan in
the future. Lenders use collateral to offset the negative effects of limited commitment.

3. Asymmetric information and limited commitment are important for financial crises.
During a financial crisis, there is more uncertainty in credit markets, which increases
interest rate spreads and reduces lending and consumption. As well, the value of
collateral falls, which also reduces lending and consumption.

4. There are two kinds of social security programs – pay-as-you-go and fully funded.
The rationale for social security is studied, as well as the potential economic benefits.

NEW IN THE FIFTH EDITION

1. New “Macroeconomics in Action: Social Security and Incentives.”

2. New end-of-chapter problems.

3. Charts and tables have been updated to reflect new data.

TEACHING GOALS

Credit market frictions and social security may not appear to be related issues, so it is
important to stress that this chapter extends the ideas of Chapter 9, by considering instances
where credit markets are not perfect. In a world with frictionless credit markets (as
considered in Chapter 9), there would be no financial crises or social security, for example.
The first key idea is that credit market frictions are typically reflected, in our two-period
model, in a kinked budget constraint for the consumer. The chapter begins by simply
considering a kinked budget constraint, where the consumer borrowers at a higher interest
rate than he or she receives as a lender, without worrying about why that may be so.
Then, asymmetric information is introduced, to show how that leads to the kinked budget
constraint, and this leads naturally to issues related to the financial crisis, particularly the
increase in interest rate spreads, which in this instance is explained by an increase in
credit market uncertainty.

With limited commitment, we again obtain a kinked budget constraint, but now the
budget constraint shifts in an interesting way with a decrease in the value of collateral.
The drop in housing prices in the United States was a key element of the financial crisis,
and the model shows how this can be connected to a decrease in the demand for
consumption goods.

Finally, the chapter considers social security systems – pay-as-you-go and fully-funded.
The model is a simplification of an overlapping generations model, but the idea is the
same. Social security can be welfare-enhancing for everyone, so long as the population
grows at a sufficiently high rate. Fully-funded social security is harder to justify
economically, however. If this type of program is simply forced savings, then it cannot
make anyone better off, as it removes choice. However, social security can always be
justified by appealing to commitment, in that people may not save adequately if they
know that the government will always be willing to look after them old age.

CLASSROOM DISCUSSION TOPICS

Encourage students to think about the credit market frictions that exist in the world.
Individuals cannot borrow all they would like to at market interest rates; we cannot
borrow at the same interest rates at which we lend; consumers, firms, and governments
sometimes default on their debts; collateral is used in lending contracts; borrowers
sometimes have better information than do lenders about their credit-worthiness.

Students may know some of the key details of what happened during the financial crisis,
though that is quickly fading into the past. Get the students up to speed on what was
happening in credit markets in the world during the crisis, so that these details can be
related to the models studied in the chapter. Recall that interest rate spreads increased,
lending contracted, and there were credit market “freezes” in some segments of the
market. Students should be encouraged to think about the implications of this for
consumption expenditure.

Students should be familiar with at least the existence of social security programs in the
world, and particularly the Canada Pension Plan (CPP). A discussion could start with the
details of the CPP and how it is financed. What reasons could we think of for the
existence of social security? Is this simply income redistribution, or is there something
deeper going on here. Why would private credit markets fail to the extent that social
security might be welfare-enhancing?
OUTLINE
1. Credit Market Imperfections and Consumption
a) Kinked budget constraint
b) Tax cuts – Ricardian equivalence does not hold.

2. Asymmetric Information and the Financial Crisis

a) Banks
b) Good and bad borrowers
c) Interest rate spread and kinked budget constraint
d) Effects of an increase in the fraction of bad borrowers

3. Limited Commitment and the Financial Crisis

a) Borrowers can default on their debts.
b) To protect itself, a lender can require that the borrower post collateral.
c) The value of collateral matters – construct kinked budget constraint.
d) Effects of a decrease in the value of collateral.

4. Ricardian Equivalence, Intergenerational Redistribution, and Social Security

a) Pay-as-you-go social security
i) Population growth, constant interest rate
ii) Welfare enhancing social security if population growth is high enough
b) Fully funded social security
i) Forced saving reduces welfare.
ii) Problems with fully-funded social security

TEXTBOOK QUESTION SOLUTIONS

Problems
1. In this economy, good borrowers and lenders always pay their taxes. In the current
period, bad borrowers take out loans, so as to mimic good borrowers, and they likewise
pay their taxes, so as not to reveal themselves as bad. However, in the future period bad
borrowers do not pay their taxes. Therefore, if t and t’ are, respectively, the taxes paid
by lenders and good borrowers, the government’s present-value budget constraint is

G' N [a + (1 − a)b]t '

G+ = Nt +
1+ r 1+ r

Here, r is the interest rate that the government pays on its debt, which is the same as
the interest rate received by lenders. However, all borrowers pay the interest rate

1+ r − a
r2 = ,
a
 which includes a default premium. Now, if Ricardian equivalence holds, then given
the market interest rate, if the government changes the timing of taxes this will leave
everyone’s wealth unchanged. However, from the government’s budget constraint,
and given the interest rate on loans above, the lifetime wealth of good borrowers
would then be

ay t' ay 1 G' t ' (1 − a)b

we = −t − = − (G + )− ,
1+ r 1+ r 1+ r N 1+ r 1+ r

and the lifetime wealth of each lender is

t' 1 G' (1 − a)(1 − b)

we = y − t − = y − (G + )− t' .
1+ r N 1+ r 1+ r

Therefore, lifetime wealth of everyone cannot be invariant to changes in the timing

of taxes. Ricardian equivalence does not hold, and consumption, the real interest
rate, and welfare will change if the government reduces current taxes and increases
future taxes, holding constant government spending. We get this result because, the
higher are future taxes, the more bad borrowers have an opportunity to default on
their taxes, and this will cause a redistribution of wealth among consumers in the
economy.

2. a) The present value of government spending can at most be

G' pH
G+ = Ny * + N ,
1+ r 1+ r

where N is the number of consumers, and y* is the minimum income of any consumer
pH
in this economy. That is, is the present value of the taxes that can be paid by the
1+ r
consumer in the future, and with what can be borrowed to pay taxes in the present.
Then, if all consumers pay the same taxes, the additional income that the government
can compensate is the minimum income that any consumer has in the present.

b) A consumer’s collateral constraint can now be written

− s (1 + r )  pH − t ' ,

or

pH − t '
c  y −t +
1+ r
 c) If the collateral constraint is not binding for any consumer, then clearly Ricardian
equivalence holds just as before. However, if the collateral constraint binds for
any consumers, then Ricardian equivalence still holds. This is because, on the
right-hand side of the collateral constraint above, changing the present value of
taxes does not change the right-hand side of the constraint, so consumption can
remain unchanged, in the present and the future, even for constrained consumers.
If these consumers get a tax cut, the future taxes reduce what anyone is willing to
lend them, as they now have a higher future tax liability, and the government can
confiscate the collateral if they default.

3. This problem combines asymmetric information with limited commitment in the

loan market. Fraction a of lenders are good borrowers who have collateral which will
have a value pH in the future period, and fraction 1-a are bad borrowers who have
collateral that will be valueless. Since there is limited commitment, the bad borrowers
will all default. Therefore, just as in the asymmetric information model, all of the bad
borrowers behave in the same way as the good borrowers.

a) There is a bank the pays an interest rate r1 on deposits and lends at the interest
rate r2. Just as in the asymmetric information model, in equilibrium

1 + r1
r2 = −1 .
a

The collateral constraint for the consumer is then

−(1 + r2 ) s  pH ,

Which can be rewritten as

pH
c  y −t + .
1 + r2

As well, if s > 0, then a consumer faces the interest rate r1, and if s < 0, then
he or she faces the interest rate r2. Thus, the budget constraint has two kinks, as
depicted by ABDE in Figure 10.1.

b) If a decreases, then r2 increases, and the collateral constraint tightens, so that

the budget constraint shifts to ABFG in Figure 10.1. In the figure, we show
a consumer for whom the collateral constraint binds. For this consumer,
consumption in the current period falls, consumption in the future period stays
the same, and savings increases. If the consumer had been a borrower for whom
the collateral constraint does not bind, then the analysis would be identical to
what was done for the asymmetric information model in Figure 10.3 of the
chapter. Consumption would decline in the current period, future consumption
could increase or decrease (depending on income and substitution effects), and
savings would increase. For a lender, there will be no effect.
 Figure 10.1

4. Taking one unit of income away from unconstrained consumers in taxes, who behave
according to the standard theory, reduces their current consumption by less than one
unit, as these consumers will save less as a result. However, if we give one unit of
income to the constrained consumers, who behave as in Figure 10.5, by cutting their
taxes, then their current consumption will increase by one unit. Thus, the government
can leave its tax revenue unchanged, while shifting the tax burden from constrained to
unconstrained consumers, this will increase the demand for consumption goods.

5. If each good borrower chooses to borrow L, then the cost of deposits for the bank is
(1+r1)L. The payoff on each loan to a good borrower is (1+r2)L, and bad borrowers
always default, but the bank can have each borrower post A as collateral, so the
payoff to the bank on a loan to a bad borrower is the smaller of A or (1+r2)L. Suppose
first that A< (1+r2)L. Then, the expected profit for the bank to making a loan is zero
in equilibrium, or

−(1 + r1 ) L + a (1 + r2 ) L + (1 − a ) A = 0

Then, if we solve the above equation for r2, we obtain

A
1 + r1 − (1 − a )
r2 = L
a

Therefore, the larger is A, the smaller is r2, the loan interest rate. The larger the
quantity of collateral available, the larger the payoff the bank receives when a bad
borrower defaults. This increases bank profits, but in equilibrium expected profits are
zero for the bank, borrowers benefit from having a higher quantity of collateral with
which to secure loans. If A is sufficiently large, i.e. A  (1+r2)L, then the bank’s
payoff on a loan to a bad borrower is (1+r2)L, and so r1 = r2. Thus, if A is large
enough then there is sufficient collateral to eliminate the credit market friction.
6. a) When the program is first instituted, the current old receive b in benefits and pay
nothing. The effect on the current old is as in Figure 10.6 in the text. The current
young receive b in benefits when they are old. This effect is also captured by the
shift from BA to FD in the text’s Figure 10.6. The current young also lend bN to
the government in period T and receive (1 + r )bN in principal and interest when
they are old. In per capita terms, these amounts are bN/(1 + n) N = b /(1 + n)
and (1 + r )bN /(1 + n) N = (1 + r )b /(1 + n) respectively. However, this borrowing and
lending is represented in Figure 10.6 as movements along the budget line. Unless
there is a change in the real interest rate, there is no additional shift in the budget
line. Therefore, both these generations unambiguously benefit from the program.

b) Once the program is running, it is identical to the pay-as-you-go system in the

text. This program benefits a typical cohort as long as n  r , as is depicted in
textbook Figure 10.7. A special circumstance applies to the cohort born in
period T + 1 . These individuals each receive a benefit per capita of b /(1 + r ) in
present value terms. However, they pay taxes to support two generations’ worth
of benefits. They pay taxes to retire the principal and interest on debt incurred in
period T. The per capita share of principal and interest on their grandparents’
benefits is equal to (1 + r )b /(1 + n)2 . The per capita share of their parents’ benefits is
equal to b /(1 + n) . This generation can only benefit if:

(1 + r ) (1 + r )2
1 + .
(1 + n) (1 + n)2

This requirement is obviously more stringent than n  r .

7. An individual consumer chooses c and c’ to satisfy the lifetime budget constraint

c' y '+ b
c(1 + s) + = y+ . (1)
1+ r 1+ r

In equilibrium, taxes on the young must finance social security benefits for the old, so

N ' sc = Nb,

or

(1 + n) sc = b. (2)

Then, substituting for b in equation (1), using (2), and simplifying, we can determine
that, in equilibrium, c and c’ must also satisfy

 ( r − n) s  c ' y'
c 1 +  + = y+ . (3)
 1+ r  1+ r 1+ r
First, consider the case where r  n . Then, in Figure 10.2, without the social security
program, the consumer chooses point D on budget constraint AB. But with the social
security program, the consumption bundle chosen by the consumer must satisfy (3), i.e.
it lies on AF in Figure 10.2. But point D is strictly preferred by the consumer to any
point on AF, so the social security program cannot make consumers better off if r  n .
Then, consider the case where r  n . It is possible that we could have a situation as in
Figure 10.3, where the consumer chooses G in the absence of the program, and H with
the social security program so that, again, the consumer is worse off with social
security. In the figure, the consumer’s budget constraint (1) is given by EF, and the
constraint (3) is given by AD, while AB is the budget constraint without social security.
However, it is also possible to have a situation as in Figure 10.4, where, again, AF is
constraint (1), AD is constraint (3), and the budget constraint without social security is
AB. In this case the consumer chooses G without social security, and H when the
program is in place. The consumer is better off with social security than without it.

Figure 10.2

Figure 10.3
 Figure 10.4

The critical difference in this problem from the basic model is that the tax on the
young is a distorting tax, which results in a “friction” in the program. There is a
welfare loss simply from the way the tax is collected. Just as with lump-sum taxes,
social security cannot improve things for everyone unless n > r, but it is possible that
n > r and social security cannot improve everyone’s welfare, because the tax
increases the price of current consumption relative to future consumption.

8. a) Under this change in government policy, the government debt issued in period T
is

DT = Nb ,

Then, in period T+1, the interest and principal on the debt will be Nb(1+r), and
the quantity of new debt issued will be N’b. Therefore, the total tax paid by the
old consumers in period T+1 is

N ' t = Nb (1 + r ) − N ' b ,

where t is the tax paid by each old consumer in period T+1. Therefore, solving the
above equation for t and simplifying, we get

b( r − n )
t=  0.
1+ n
 Therefore, since every period from T+1 on will look the same, the lifetime wealth
of any consumer born in periods T+1 on will now be

y' n−r
we = y + − .
1+ r 1+ n

But note that this is identical to the lifetime wealth these consumers would have if
the pay-as-you-go social security system had stayed in place. Thus, it makes no
difference, given this financing scheme, whether the government gets rid of the
social security system or not.

b) If the pay-as-you-go system were replaced by a fully-funded system, this cannot

make consumers any better off. Just as in the analysis of this chapter, if a fully-
funded system is put in place, this at best has no effect, and at worst constrains the
savings of consumers and makes them worse off.