Applied Financial Economics, 1998, 8, 597—605

The causes of stock market volatility

in Australia
CO LM KE ARNEY and KEVI N D AL Y *
ºniversity of ¹echnology, Sydney, Australia and *Department of Economics and
Finance, ºniversity of ¼estern Sydney, Australia

The paper examines the extent to which the conditional volatility of stock market
returns in a small, internationally integrated stock market are related to the condi-
tional volatility of financial and business cycle variables. It employs a low frequency
monthly dataset for Australia including stock market returns, interest rates, inflation,
the money supply, industrial production and the current account deficit over the
period from July 1972 to January 1994. A novel feature of the analysis is the
estimation strategy employed to overcome the generated regressors problem which
pervades some recent related research. Specifically, the procedure of employing
a two-stage estimation process to first estimate conditional volatilities and then model
their interrelationships yields inefficient estimates, introduces bias into a number of
diagnostic test statistics and generates potentially invalid inferences. This problem is
overcome in the current paper by jointly estimating the equation for the conditional
volatility of stock market returns together with the equations determining the condi-
tional volatilities of all variables included in the model using the generalized least
squares (GLS) estimation procedure together with the Hendry general-to-specific
modelling strategy. Among the most important determinants of the conditional
volatility of the Australian stock market are found to be the conditional volatilities of
inflation and interest rates which are directly associated with stock market volatility,
and the conditional volatilities of industrial production, the current account deficit
and the money supply which are indirectly associated with stock market conditional
volatility. Among these variables, the strongest effect is found to be from the condi-
tional volatility of the money supply to the conditional volatility of the stock market.
By contrast, no evidence is found of volatility spillover from the foreign exchange
market to the stock market in Australia.

I. INTRO DUCTI ON
The issue of stock market volatility has received much
attention in the finance literature. The main questions which
have been addressed include: what are the important causes
of stock market volatility? has it increased over time? has
international financial integration led to faster transmission
of volatility across international stock markets? and what
role, if any, ought regulators play in the process? Previous
researchers have examined these issues. For example, on the
causes of stock market volatility, Officer (1973) examines the
effects of volatility in business cycle variables, Black (1976)
and Christie (1982) relate stock market volatility to financial
0960—3107  1998 Routledge
leverage, Merton (1980), Poterba and Summers (1986) and
French et al. (1987) relate stock market volatility to the
volatility of expected returns, and Schwert (1989) conducts
an extensive array of tests on the macroeconomic causes of
stock market volatility over long runs of monthly data for
the United States. The issue of whether the world’s financial
and capital markets are now transmitting volatility more
quickly has been examined by Koch and Koch (1991),
Malliaris and Urrutia (1992), Chan et al. (1992) and
Rahman and Yung (1994). The related issue of whether
stock market volatility has increased over time has
been investigated by Peel et al. (1993), while Scott (1991)
and Timmermann (1993) examine the extent to which
597
598
the volatility of stock prices determines their underlying
value.
The purpose of this paper is to contribute to the literature
on the causes of stock market volatility by examining the
determinants of movements in the volatility of equity re-
turns in a small, internationally integrated stock market.
The Australian stock market serves as an especially illumi-
nating example in this regard insofar as it is increasingly
integrated with the important international stock markets
throughout Asia and the world. Previous research on the
volatility of the Australian stock market includes the work
of Brailsford and Faff (1993) and Kearns and Pagan (1993),
both of whom examined the relative explanatory power
of alternative models of conditional volatility, but neither of
whom related stock market volatility to the volatility of
financial and business cycle variables. This paper develops
and estimates a model which is capable of explaining the
financial and business cycle determinants of movements in
the conditional volatility of the Australian All Industrials
stock market index. It employs a low frequency monthly
dataset including stock market returns, interest rates, infla-
tion, the money supply, industrial production and the cur-
rent account deficit over the period July 1972 to January
1994. Among the most important determinants of Austra-
lian stock market volatility are found to be the volatility of
inflation and interest rates which are directly associated
with stock market volatility, while the volatility of industrial
production, the current account deficit and the money
supply are indirectly associated with stock market volatility.
Among these variables, the most significant determinant of
conditional volatility in the Australian stock market is the
degree of volatility which exists in the money supply. By
contrast, no evidence is found of a statistically significant
relationship between foreign exchange market volatility and
stock market volatility in Australia.
The results presented in this paper constitute new evid-
ence which is interpretable as an extension of the low
frequency analysis of Schwert (1989) who did not include
international factors such as the current account deficit and
the exchange rate in his investigation of the causes of stock
market volatility in the United States. In addition, Schwert’s
(1989) vector autoregressive model of the determination of
conditional stock market volatility, along with other re-
search which employs the same methodology, suffers from
the generated regressors problem (Pagan, 1984, 1986;
McAleer and McKenzie, 1991; Oxley and McAleer, 1993)
which yields inefficient estimates, introduces bias into
a number of diagnostic test statistics and generates poten-
tially invalid inferences. This problem is overcome in the
current paper by jointly estimating the equation for the
conditional volatility of the Australian stock market index
together with the equations determining the conditional
volatilities of all variables included in the model using the
generalized least squares (GLS) estimation procedure and
the Hendry general-to-specific estimation strategy. Related
C. Kearney and K. Daly
research by Koutoulas and Kryzanowski (1996) employs
a similar methodology to that which is employed in this
paper in order to examine the explanatory power of macro-
economic conditional volatilities in an arbitrage pricing
model of monthly Canadian stock market returns. Interest-
ingly, they find that the conditional volatilities of industrial
production and the exchange rate feature in their model
while interest rates play a more ambiguous role.
The paper is structured as follows. The next section de-
scribes the theoretical framework which motivates the em-
pirical analysis and describes the dataset which is used for
the study. Section III turns the focus of attention to the
causes of volatility in the Australian stock market. It em-
ploys the generalized least squares (GLS) estimation meth-
odology to implement the Davidian and Carroll (1987)
ARCH model of conditional volatility to model the deter-
mination of Australian stock market conditional volatility
using the Hendry general-to-specific estimation methodo-
logy (see Mizon (1995) for a recent overview). The final
section reviews the main results and draws together the
conclusions.
II . TH EORY, METHO DOLOGY AND DATA
The price of equity at any point in time is equal to the
discounted present value of expected future cash flows (in-
cluding capital gains and dividends) to shareholders:
 CG (IP , M , P , D , S )
E QG"E R>I R>I R>I R>I R>I R>I
R\ R R\ (1#R )I
I R>I
(1)
where QG is the price of asset i, CG denotes the cash flows
R
associated with it, R denotes the interest rate and E is the
R\
expectations operator. In an open economic and financial
system such as that which prevails in Australia, corporate
cash flows CG are influenced by developments in the macro-
economy such as the level of aggregate industrial produc-
tion (IP), the money supply (M) and the level of prices (P),
the external trading position of the country as measured by
the current account deficit of the balance of payments (D)
and the spot exchange rate (S) which is defined as the
domestic currency price of foreign exchange.
In moving from equity prices to returns, let qG denote the
R
actual return on asset i, and let the q' G"E [qG/I ] denote
R R R R\
its expected return conditional on the available information
set at time t!1. In addition, let pqG denote the uncondi-
R
tional standard deviation of returns on asset i, and let
pL "E [p /I ] denote its conditional counterpart. From
qG qG
R R R R\
Equation 1, the conditionally expected return on asset i is
a function, f, of the conditionally expected determinants of
the discounted cash flows:
E [qG/I ]"f +E [CG(IP , M , P , D , S ), R ]/I , (2)
R R R\ R R R R R R R R R\
Causes of stock market volatility in Australia 599
and the conditional standard deviation of returns is a func- and a (H) is another 12th-order polynomial in the lag

tion, g, of the conditional standard deviations of the deter- operator H. The measure of conditional volatility in Equa-
minants of the cash flows: tion 4 represents a generalization of the 12-month rolling
standard estimator used by Officer (1973), Fama (1976) and
E [pqG/I ]"g+E [p'., p+, p., p", p1, p0]/I , (3) Merton (1980) to measure stock market volatility, because it
R R R\ R R R R R R R R\
allows the conditional mean to vary over time in Equation
In order to empirically implement the model, we must
5 while also allowing different weights to apply to the lagged
obtain monthly estimates of the standard deviations of the
absolute unpredicted changes in stock market returns in
relevant variables. Although this can be achieved with ease
Equation 4. This measure has been used by Schwert (1989)
for the stock market and financial variables such as interest
to examine the relationship between stock market volatility
rates and exchange rates, business cycle data are not
and underlying economic volatility, and more recently by
typically available at higher frequencies than monthly.
Koutoulas and Kryzanowski (1996) to examine the role of
Accordingly, the approach adopted here is to employ the
conditional macroeconomic factors in an arbitrage pricing
methodology of Davidian and Carroll (1987). Let X denote
model. The relative merits of alternative measures of condi-
the vector of stock market and business cycle variables
tional volatility are reviewed by, inter alia, Engle (1993) and
employed in the study, i.e. X"X(Q, IP, M, P, D, S, R), let
Diebold and Lopez (1995), while Kearney (1996) elaborates
pV denote the unconditional standard deviation of these
R on the measure employed here and points to the implica-
variables, and let p̂V"E (pV/I ) denote the corresponding
R R R R\ tions for financial regulatory policies. It is similar to the
conditional standard deviations. The latter are obtained as
autoregressive conditional heteroscedasticity (ARCH)
p̂V"pV!eV from the following regression:
R R R model of Engle (1982) which, in its various forms, has been
 widely used in the finance literature. Davidian and
pV"b (H)pV# b SD #eV (4) Carroll (1987) argue that the specification in Equation
R  R K KR R
K 4 based on the absolute value of the prediction errors is
where b (H) is a 12th-order polynomial in the lag operator more robust than those based on the squared residuals in
 Equation 5.
H, the SD are monthly dummy variables to capture sea-
K The dataset for the study consists of monthly observa-
sonal variations in the means and standard deviations of the
variables, and the pV are innovations which are obtained as tions on the Australian stock market and business cycle
R variables including the interest rate on 3-month bank ac-
the absolute values of the residuals from the equation
pV"/eV / where cepted bills, the Australian—US dollar exchange rate, the
R R rate of inflation of the wholesale price index, the current
eV "*log(X) !E (* log(X) /I ) account deficit and the level of industrial production over
R R R R R\
the period July 1970—January 1994. Table 1 contains de-
 tailed descriptions of all variables used in the analysis
"*log(X) !a (H) *log(X) ! a SD (5)
R  R K KR together with their source.
K

Table 1. »ariables used and data sources

Q: Australian sharemarket All Industrial Index, sampled on close of trading on the last trading day in the month. The index is re-based to
1973 " 100. Source is Australian Stock Exchange.
IP: The monthly percentage change in the index of industrial production, rebased to 1973 " 100. Source is OECD Economic Database.
P: The monthly index of wholesale prices in Australia, re-based to 1973 " 100. Source is OECD Economic Database.
I: The monthly percentage change in the wholesale price index (P) in Australia, re-based to 1975 " 100. Source is OECD Economic
Database.
D: The current account deficit of the database of payments in Australia, in billions of Australian dollars. Source is OECD Economic
Database.
S: The spot exchange rate of the Australian—US dollar exchange rate, sampled at the end of the last trading day each month. Source is
OECD Economic Database.
R: The interest rate on 3-month bank accepted bills in Australia, sampled at the end of the last trading day each month. Source is Reserve
Source Bank of Australia.
CRASH: Dummy variable to capture the effects of the October 1987 stock market crash. It takes the value of 1 in October 1987 and 0 at
all other times.
SD: Monthly seasonal dummy variables to capture the effects of monthly variations the conditional mean and conditional standard
deviations of the business cycle and financial variables.
600
II I. MOD ELLI NG TH E C ONDITI ON AL
V OL AT I L IT Y O F TH E A U ST R A LI AN
ST OCK MA R KET
In order to appropriately account for the non-zero cross-
equation covariances which arise from the generated
regressors problem (Pagan, 1984, 1986; McAleer and
McKenzie, 1991; Oxley and McAleer, 1993), the equation
for the conditional volatility of the Australian stock market
index is estimated jointly with the equations determining
the conditional volatilities of all variables included in the
model using the generalized least squares (GLS) estimation
procedure. The GLS estimator is described by first stacking
the system to be estimated in the following form with
M equations and ¹ observations:
     
y X 2 0 b e
 "   #  (6)
y 0 2X b e
K K K K
where y is an M¹ ; 1 vector of dependent variables, X is an
M¹ ; ( + K ) matrix of explanatory variables, b is an
G G
M¹ ; ( + K ) matrix of coefficients and e is an M¹ ; 1
G G
vector of errors. In the model analysed here Equation 6 con-
tains 15 equations (i.e. M " 15) with 259 observations (i.e.,
¹ " 259) on each variable. The first seven equations are
given by Equation 5 for each variable in the X vector, the
next seven equations are given by Equation 4 for each
variable in the X vector, and the last equation is given by
p̂/"j #j (K) pL V#e (7)
R   R R
where the X vector is X(Q, IP, M, P, D, S, R) as before, and
the j (K) are polynomials of the 4th degree in the lag
 sonality, and Haugen and Jorian (1996) who have reported
operator, K. Equation 7 relates the conditional volatility of
the Australian stock market, Q, to the conditional volatility
of the Australian financial and business cycle variables in-
cluding industrial production IP, the money supply M,
inflation P, the current account deficit D, the exchange rate
S, and the interest rate R.
The system in Equation 6 incorporates the possibility of
cross-equation correlation among the error terms which
yields the following covariance matrix of errors:
 
p I p I
E(ee)")"  %
p I p I
% %%
Writing the system in matrix form as in Equation 8, we can
describe the GLS estimator as given by Equation 9:
y"Xb#e (8)
b"(X)\X)\ X)\y (9)
The gain in efficiency of this GLS estimator over OLS
estimation depends on a number of factors including the
extent of relations across the equations in the system and
the extent to which the X variables differ across equations.
C. Kearney and K. Daly
In the current context of estimating the conditional vola-
tilities of financial and economic variables, the potential
for efficiency gain by estimating the system by GLS is
considerable.
Table 2 presents the summary statistics from the OLS
estimates of the first 12 autoregressions contained in Equa-
tions 4 and 5 which are used as inputs during the GLS
estimation procedure. Looking first at the top part of the
table which summarizes the equations for the conditional
means, the equations explain between 7 and 79% of the
variation in the dependent variables, with, as expected,
the financial variables (including the stock market returns,
the interest rate and the exchange rate) occupying the lower
end of the explanatory power rankings. The equations are
all free from higher order autocorrelation as evidenced by
the Ljung—Box Q-statistics. The sums of the lagged depen-
dent coefficients (with their marginal significance levels in
brackets) are presented in the fourth column of Table 2
under the heading Sum. The columns headed by F and

F provide the F-statistics for the joint exclusion of, respec-

tively, all lagged dependants and all the seasonal variables
in each regression. The lagged dependants in the equations
for the stock market, inflation and the current account
deficit are jointly significant in their respective equations,
while the seasonal dummy variables are jointly significant in
all equations except those for inflation and the exchange
rate. With regard to the Australian stock market, the latter
finding is in conformity with other researchers such as
Gultekin and Gultekin (1983) who found statistically signifi-
cant stock market seasonality in 14 out of 17 countries
studied, Kramer (1994) who related stock market seasonal-
ity in the United States to underlying macroeconomic sea-
that the January effect has not declined in size during the
past three decades in the United States.
Looking next at the bottom part of Table 2 which
summarizes the equations for the conditional standard
deviations, the equations explain predictably less of the
variation in the dependent variables, and all are free
from higher order autocorrelation as evidenced by the
Ljung—Box Q-statistics. The lagged dependants in the equa-
tions for the stock market, industrial production, inflation,
the current account deficit and the interest rate are jointly
significant in their respective equations, while the seasonal
dummy variables are, as in the top part of Table 2, jointly
significant in the equations for all variables except inflation
and the exchange rate.
Figure 1 presents time series plots of the conditional
volatilities of all variables used in the study. Most series
display considerable time variation in their volatilities. Of
particular note is the high volatility of interest rates during
the 1980s and the stock market volatility associated with the
October 1987 crash. The summary statistics of each series
are presented in Table 3. The means of the variables show
considerable variation, ranging from a high of 0.089 for the
Causes of stock market volatility in Australia 601
Table 2. Estimation results of the ARCH models of stock market returns and business cycle variables July 1970—January 1994

Variables R SEE Q Sum F F

 
 
X " dVX ! g #mV
R H R\H KR R
H K
Q 0.21 0.047 51.05 (0.05)* !3.81 (6.05)* 35.82 (0.00)* 2.43 (0.01)*
½ 0.79 0.062 24.20 (0.93) !0.90 (0.01) 1.75 (0.06) 2.83 (0.00)*
M 0.62 0.014 30.44 (0.73) 0.33 (1.90) 1.72 (0.06) 10.39 (0.00)*
P 0.22 0.016 20.27 (0.98) 0.42 (2.89)* 4.74 (0.00)* 1.34 (0.20)
D 0.53 0.123 38.27 (0.37) !1.94 (3.65)* 15.41 (0.00)* 3.75 (0.00)*
E 0.07 0.030 20.43 (0.98) !0.02 (0.10) 1.20 (0.29) 0.47 (0.93)
R 0.21 0.081 22.77 (0.96) 0.31 (1.69) 1.74 (0.06) 2.52 (0.00)*
 
pV " aVpV # b D #eV
R H R\H K KR R
H K
Q 0.20 0.029 19.04 (0.99) 0.49 (3.70)* 4.00 (0.00)* 1.91 (0.03)*
½ 0.32 0.039 21.37 (0.97) 0.55 (4.82)* 6.20 (0.00)* 3.35 (0.00)*
M 0.10 0.009 25.85 (0.89) 0.22 (1.10) 0.94 (0.51) 2.54 (0.00)*
P 0.07 0.011 29.92 (0.75) 0.34 (2.15)* 1.02 (0.43) 1.74 (0.06)
D 0.15 0.007 22.37 (0.96) 0.42 (2.45)* 1.18 (0.30) 2.64 (0.00)*
E 0.08 0.023 26.81 (0.87) 0.33 (1.87) 1.12 (0.34) 1.68 (0.07)
R 0.17 0.054 19.95 (0.99) 0.62 (4.83)* 2.83 (0.00)* 1.97 (0.03)*

The table presents the regression summary results for Equations 2 and 3 in the text, regressed on each variable in the table. SEE denotes the
standard error of the regression and Q denotes the Ljung—Box test for higher order autocorrelation. Sum denotes the sums of coefficients of
the lagged dependent variables in each regression equation. F denotes the F-statistics for the joint exclusion of all lagged dependants in

each regression. F denotes the F-statistics for the joint exclusion of all the seasonal coefficients in each regression. Marginal significance

levels for these test statistics are presented in the brackets.

current account deficit (D) to a low of 0.010 for the money
supply (M), with the intermediate rankings being in order
from highest to lowest, interest rate (R), industrial produc-
tion (IP), stock market (Q) and inflation (P). These rankings
are also mostly replicated in terms of the variances. There is
evidence of non-normality as indicated by the marginal
significance levels of the skewness of most variables and of
the Kurtosis of some variables. The two right-hand columns
of Table 3 provide the minimum and maximum values of
the conditional volatilities along with their dates in
brackets. As expected, the stock market crash features as
being the most volatile month during the study period for
the stock market variable, and the early and mid-1980s
feature as being the period of most volatile interest rates.
As a precursor to the final model specification, the gener-
ated conditional volatilities were tested for their order of
integration in order to avoid any possible spurious regres-
sion relationships. Although inspection of Fig. 1 suggests
that the variables are all stationary, it is advisable to check
the correctness of this supposition. Given the non-normality
of some variables, Table 4 presents the results of unit
root tests for each variables using alternatively the
Dickey—Fuller (1979) and the Phillips—Perron (1988) testing
methodologies. As expected, each test strongly indicates
stationarity of every variable employed in the model.
Since financial theory does not contain predictions about
the form of the lag structure which may be appropriate in
this analysis, the model was estimated by GLS along the
lines of the general-to-specific modelling strategy (Mizon,
1995). This involved first estimating the system described in
Equation 6 with the general specified lag structure for the
conditional volatility of the stock market described in Equa-
tion 7, and then sequentially restricting the latter equation
by excluding its statistically insignificant components. The
final form of the equation for the conditional volatility of
the stock market is presented in Equation 10 and the
estimation results are presented in Table 5.
p̂/"j #j Crash#j p̂/ #j p̂/ #j p̂'.
R    R\  R\  R\
#j p̂+#j p̂.#j p̂. #j p̂"
 R  R  R\  R\
#j p̂" #j p̂##j p̂0 #e (10)
 R\  R  R\ R
Looking first at the model’s overall performance as present-
ed by the equation diagnostics in the bottom part of Table 5,
the R statistic indicates that it explains an average of 35%
of the variation in the conditional variance of the Australian
stock market, with the standard error of the estimate being
equal to one-third of the mean of the dependent variable.
This compares well with related work of, for example, Stew-
ard (1993). The regression F-statistic is highly significant,
indicating that the included variables are jointly statistically
significant determinants of the conditional volatility of the
Australian stock market. The Durbin—Watson (D¼) and
602
Fig. 1. Conditional volatilities of financial and business cycle vari-
ables
Kolmogorov—Smirnov (K!S) statistics indicate that the
model is free from first and higher order autocorrelation.
The ARCH test for heteroscedasticity indicates that this is
not a problem, which is intuitive insofar as we are working
with the conditional volatilities of all variables. Finally,
three versions of the Chow test for structural stability are
C. Kearney and K. Daly
included in Table 5; CHO¼ splits the sample in half
(from 1974:1—1984:12 and from 1985:1—1994:1), CHO¼
splits it into 1/3 and 2/3 (from 1974:1—1981:12 and from
1982:1—1994:1), while CHO¼ splits it into 2/3 and 1/3
(from 1974:1—1987:12 and from 1988:1—1994:1), respectively.
These test statistics confirm that the model’s structure re-
mains stable over time.
Looking next at the individual coefficient estimates of the
lagged dependent variable, the j and j coefficients indi-
 
cate that the model incorporates significant lagged depen-
dent terms which implies that volatility in the Australian
stock market responds dynamically to variations in the
volatility of the financial and business cycle variables. The
equation also contains a positively signed and statistically
significant constant term with a value of j " 0.035 and

associated t-statistic of 6.07 which indicates that stock mar-
ket volatility in Australia tends to proceed independently
from the other influences included in the model. The dummy
variable for the October 1987 stock market crash is
also positively signed as expected with a coefficient of
j " 0.063, and it is strongly statistically significant with

a t-statistic of 5.89.
Turning now to examine the effects of the conditional
volatility of the business cycle variables on the conditional
volatility of the Australian stock market, all variables are
statistically significant at the 5% level of significance except
for the conditional volatility of the exchange rate. Increases
in the conditional volatility of interest rates and inflation are
associated with higher stock market conditional volatility,
while increases in the conditional volatility of industrial
production, the current account deficit and the money
supply are associated with lower stock market conditional
volatility. More specifically, the positive j coefficient of

0.078 (with its t-statistic of 2.42) indicates that the condi-
tional volatility of interest rates impacts positively upon the
conditional volatility of the stock market with a one-month
lag. The negative j coefficient of !0.541 (with its t-statistic

of 2.00) and the positive j coefficient of 0.853 (with its

t-statistic of 3.09) together imply that the conditional vola-
tility of inflation has a net positive impact on the condi-
tional volatility of the stock market with a one-month lag.
By way of contrast, the negative j coefficient of !0.088

(with t-statistic of 3.11) indicates that the conditional volatil-
ity of the stock market varies indirectly with the conditional
volatility of industrial production with a three-month lag.
As the negative j coefficient of !1.093 (with its t-statistic

of 4.41) indicates, the same qualitative relationship pertains
to the effect of changes in the conditional volatility of the
money supply on the conditional volatility of the stock
market, although in this case the strength of the relationship
is greater and the impact is immediate (i.e. within one
month). Indeed, the magnitude of the j coefficient together

with its associated t-statistic indicates that, apart from the
stock market’s recent volatility history and the dummy
variable for the October 1987 stock market crash, the most
Causes of stock market volatility in Australia 603
Table 3. Descriptive statistics for the conditional volatilities of variables used in the analysis

Minimum Maximum
Variable Mean Variance Skewness Kurtosis value value

Q 0.033 (0.00) 2.01\ 1.68 (0.00) 6.78 (0.00) 0.002 (88 : 8) 0.121 (87 : 12)
½ 0.039 (0.00) 6.66\ 1.48 (0.00) 3.32 (0.00) 0.001 (90 : 3) 0.164 (80 : 6)
M 0.010 (0.00) 9.23\ 0.50 (0.00) 0.06 (0.85) 0.002 (82 : 2) 0.018 (91 : 1)
P 0.011 (0.00) 9.19\ 0.62 (0.00) 0.37 (0.24) 0.005 (76 : 11) 0.023 (86 : 9)
D 0.089 (0.00) 9.40\ 0.79 (0.00) 0.50 (0.11) 0.035 (77 : 10) 0.201 (93 : 12)
E 0.019 (0.00) 4.21\ 0.52 (0.00) 0.45 (0.16) 0.002 (77 : 3) 0.040 (77 : 1)
R 0.053 (0.00) 5.55\ 0.84 (0.00) 0.99 (0.00) 0.012 (85 : 6) 0.148 (83 : 4)

The descriptive statistics are all calculated over the period 1973:8—1994:1. The figures in parentheses following the mean, skewness and
kurtosis are the marginal significance levels for zero. The figures in parentheses following the minimum and maximum values provide the
dates.

Table 4. ºnit root tests of stock markets and business cycle Table 5. G¸S estimation of the ARCH model of conditional volatil-
variables ity of the Australian stock market

Variable Dickey—Fuller test Phillips—Perron test Explanatory Model Estimated

variable coefficient coefficient t-statistic
Q !179.76 !211.69
½ !135.43 !133.60 Constant j 0.035 6.07

M !298.61 !290.51 Dum87 j 0.063 5.89

P !139.18 !143.73 p̂/ j 0.123 2.24
D R\ 
!206.29 !239.43 p̂/ j 0.326 5.92
E !216.13 !187.84 R\ 
p̂'. j !0.088 3.11
R !130.03 !140.46 R\ 
p̂+ j !1.093 4.41
R 
The tests are conducted over the period from 1974:1—1994:1 for the p̂. j !0.541 2.00
R 
levels, and 1974:2—1994:1 for the conditional volatilities. The criti- p̂. j 0.853 3.09
cal values are !2.57 for both tests. See Dickey and Fuller (1979), R\ 
p̂" j !0.064 2.53
Fuller (1976), Phillips (1987) and Phillips and Perron (1988). R\ 
p̂" j !0.088 3.46
R\ 
p̂# j 0.164 1.40
R 
p̂0 j 0.078 2.42
R\ 
significant determinant of volatility in the Australian stock Equation diagnostics
Estimation period 1974:1—1994:1
market is the degree of volatility which exists in the money Observations, df 241, 229
supply. R 0.35
Turning to the effects of the conditional volatility of the SEE 0.012
current account deficit on the conditional volatility of the SSR 0.031
stock market, the negative j coefficient of !0.064 (with its F(11, 229) 11.80 (0.00)
 D¼ 2.21
t-statistic of 2.53) and the greater negative j coefficient
 K—S 0.12 (0.10)
of !0.088 (with its t-statistic of 3.46) together indicate ARCH 0.32 (0.57)
an indirect relationship with a lag of up to two months. CHO¼ 1.17 (0.32)
Perhaps one of the most interesting findings to emerge is CHO¼ 1.24 (0.28)
that there is no evidence of volatility spillover from the CHO¼ 1.74 (0.09)
foreign exchange market to the stock market in this low SEE and SSR denote the standard error of the estimate and the
frequency analysis of the Australian data. Although the sum of squared residuals. F(11, 129) denotes the regression
j coefficient is positive and equal to 0.164, its t-statistic of F-statistic. D¼ and K—S denote the Durbin—Watson statistic and
 the Kolmogorov—Smirnov test for higher order autocorrelation.
1.40 indicates that the relationship is not statistically signifi-
cant. This finding is consistent with the work of Ratner The ARCH and CHO¼ tests are for heteroscedasticity and struc-
tural stability. Concerning the latter; CHO¼, CHO¼ and
(1993) who employed the bivariate cointegration methodo- CHO¼ split the sample in half, into 1/3 and 2/3, and into 2/3 and
logy in order to examine whether the exchange rate impacts 1/3, respectively.
upon monthly stock prices in the United States, and con-
cluded negatively before suggesting that any such rela-
tionship is spurious if the markets operate efficiently. Our Ratner (1993) in a low frequency context. Whether this
analysis, which focuses on the conditional volatility rather finding would be replicated in higher frequency studies
than on the levels of market returns, confirms the findings of remains on the research agenda.
604
IV . S UMM A RY AND C ONCLU SION S
The purpose of this paper has been to contribute to the
literature on the causes of stock market volatility by exam-
ining the determinants of movements in the volatility of
equity returns in a small, internationally integrated stock
market. The Australian stock market serves as an interest-
ing example in this regard insofar as it is increasingly inte-
grated with the important international stock markets
throughout Asia and the world. The paper developed and
estimated a model which is capable of explaining move-
ments in the conditional volatility of the Australian All
Industrials stock market index. It employed a low frequency
monthly dataset including stock market returns, interest
rates, inflation, the money supply, industrial production and
the current account deficit over the period from July 1972 to
January 1994.
The results presented in this paper constitute new evid-
ence which is interpretable as an extension of the low
frequency analysis of Schwert (1989) who did not include
international factors such as the current account deficit and
the exchange rate in his investigation of the causes of stock
market volatility in the United States. In addition, a novel
feature of the analysis presented here is the estimation
strategy employed to overcome the generated regressors
problem (Pagan, 1984, 1986; McAleer and McKenzie, 1991;
Oxley and McAleer, 1993) which yields inefficient estimates,
introduces bias into a number of diagnostic test statistics
and generates potentially invalid inferences. This problem is
overcome in the current paper by jointly estimating the
equation for the conditional volatility of the stock market
returns together with the equations determining the condi-
tional volatilities of all variables included in the model using
the generalized least squares (GLS) estimation procedure
and the Hendry general-to-specific estimation strategy.
Among the most important determinants of the condi-
tional volatility of the Australian stock market are found to
be the conditional volatilities of inflation and interest rates
which are directly associated with stock market volatility,
and the conditional volatilities of industrial production, the
current account deficit and the money supply which are
indirectly associated with stock market conditional volatility.
Among these variables the strongest effect is found to be from
the conditional volatility of the money supply to the condi-
tional volatility of the stock market. By contrast, no evidence
is found of a statistically significant relationship between
the conditional volatility of the foreign exchange market and
the conditional volatility of the stock market in Australia.
RE F ER E N C E S
Black, F. (1976) Studies of stock price volatility changes, in Pro-
ceedings of the 1976 Meetings of the Business and Economics
Statistics Section, American Statistical Association, pp. 177—81.
C. Kearney and K. Daly
Brailsford, T. J. and Faff, R. W. (1993) Modelling Australian stock
market volatility, Australian Journal of Management, 18,
109—32.
Chan, K. C., Karolyi, G. A. and Stulz, R. M. (1992) Global
financial markets and the risk and premium on US equity,
Journal of Financial Economics, 32, 137—67.
Christie, A. A. (1982) The stochastic behaviour of common stock
variances: value, leverage and interest rate effects, Journal of
Financial Economics, 10, 407—32.
Davidian, M. and Carroll, R. J. (1987) Variance function estimation,
Journal of the American Statistical Association, 82, 1079—91.
Dickey, D. and Fuller, W. A. (1979) Distribution of the estimators
for time series regressions with a unit root, Journal of the
American Statistical Association, 74, 427—31.
Diebold, F. X. and Lopez, J. A. (1995) Modeling volatility dynamics,
in Macroeconometrics; Developments, ¹ensions and Prospects,
ed. K. D. Hoover, Kluwer Academic Publishers, Boston.
Engle, R. F. (1982) Autoregressive conditional heteroscedasticity
with estimates of the variance of UK inflation, Econometrica,
50, 987—1008.
Engle, R. F. (1993) Statistical models for financial volatility, Finan-
cial Analysts Journal, 49, 72—8.
Fama, E. F. (1976) Forward rates as predictors of future rates,
Journal of Financial Economics, 3(4), 351—77.
Fama, E. F. (1990) Stock returns, expected returns and real activ-
ity, Journal of Finance, 45, 1089—1108.
French, K. R., Schwert, G. W. and Staumbaugh, R. F. (1987)
Expected stock returns and volatility, Journal of Financial
Economics, 19, 3—29.
Fuller, W. A. (1976) Introduction to Statistical ¹ime Series, Wiley,
New York.
Gultekin, M. and Gultekin, B. (1983) Stock market seasonality:
international evidence, Journal of Financial Economics, 12,
469—82.
Haugen, R. A. and Jorian, P. (1996) The January effect: still there
after all these years, Financial Analysts Journal, January—
February, 27—31.
Kearney, C. (1996) Volatility and risk in integrated financial sys-
tems: measurement, transmission and policy implications,
in Managing Risk in »olatile Financial Markets, ed. D. A.
Sargent, Kluwer Academic Press.
Kearns, P. and Pagan, A. (1993) Australian stock market volatility:
1875—1987, Economic Record, 69, 205, 163—78.
Kramer, C. (1994) Macroeconomic seasonality and the January
effect, ¹he Journal of Finance, 49, 1883—91.
Koch, P. D. and Koch, T. W. (1991) Evolution in dynamic linkages
across daily national stock indices, Journal of International
Money and Finance, 10, 231—51.
Koutoulas, G. and Kryzanowski, L. (1996) Macrofactor condi-
tional volatilities, time-varying risk premia and stock return
behaviour, ¹he Financial Review, 31, 169—95.
Malliaris, A. G. and Urrutia, J. L. (1992) The international crash of
October 1987: causality tests, Journal of Financial and Quant-
itative Analysis, 27, 353—64.
McAleer, M. and McKenzie, C. R. (1991) When are two step
estimators efficient?, Econometric Reviews, 10, 235—52.
Merton, R. C. (1980) On estimating the expected return on the
market: an exploratory investigation, Journal of Financial
Economics, 8, 323—61.
Mizon, G. (1995) Progressive modelling of macroeconomic time
series: the LSE methodology, in Macroeconometrics; Develop-
ments, ¹ensions and Prospects, ed. K. D. Hoover, Kluwer
Academic Publishers, Boston.
Officer, R. R. (1973) The variability of the market factor of the New
York Stock Exchange, Journal of Business, 46, 434—53.
Causes of stock market volatility in Australia
Oxley, L. and McAleer, M. (1993) Econometric issues in macroeco-
nomic models with generated regressors, Journal of Economic
Surveys, 7, 1—40.
Pagan, A. (1984) Econometric issues in the analysis of regressions
with generated regressors, International Economic Review, 25,
221—47.
Pagan, A. (1986) Two stage and related estimators and their
applications, Review of Economic Studies, 53, 517—38.
Peel, D. A., Pope, P. F. and Yadav, P. K. (1993) Deregulation and
the volatility of UK stock prices, Journal of Business Finance
and Accounting, 20, 359—72.
Phillips, P. C. B. (1987) Time series regressions with a unit root,
Econometrica, 55, 277—301.
Phillips, P. C. B. and Perron, P. (1988) Testing for a unit root in
time series regressions, Biometrika, 65, 335—46.
Poterba, J. M. and Summers, L. H. (1986) The persistence of
volatility and stock market fluctuations, American Economic
Review, 76, 1142—51.
605
Rahman, H. and Yung, K. (1994) Atlantic and Pacific stock mar-
kets — correlation and volatility transmission, Global Finance
Journal, 5, 103—19.
Ratner, M. (1993) A cointegration test of the impact of foreign
exchange rates on US stock market prices, Global Finance
Journal, 4, 93—101.
Schwert, G. W. (1989) Why does stock market volatility change
over time? Journal of Finance, 54, 1115—51.
Scott, L. O. (1991) Financial market volatility: a survey, Interna-
tional Monetary Fund Staff Papers, 38, 582—625.
Steward, F. (1993) Extracting significant innovations from pub-
lished sources in Great Britain, in New Concepts in Innovation
Output Measurement, eds A. Kleinknecht and D. Bain, Mac-
millan Press, London, pp. 138—52.
Timmermann, A. G. (1993) How learning in financial markets
generates excess volatility and predictability in stock prices,
Quarterly Journal of Economics, 108, 1135—45.