American Finance Association

On the Relation between the Expected Value and the Volatility of the Nominal Excess Return
on Stocks
Author(s): Lawrence R. Glosten, Ravi Jagannathan, David E. Runkle
Source: The Journal of Finance, Vol. 48, No. 5 (Dec., 1993), pp. 1779-1801
Published by: Blackwell Publishing for the American Finance Association
Stable URL: http://www.jstor.org/stable/2329067 .
Accessed: 07/04/2011 12:15

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at .
http://www.jstor.org/action/showPublisher?publisherCode=black. .

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].

Blackwell Publishing and American Finance Association are collaborating with JSTOR to digitize, preserve
and extend access to The Journal of Finance.

http://www.jstor.org
THE JOURNAL OF FINANCE v VOL. XLVIII, NO. 5 v DECEMBER 1993

On the Relation between the Expected

Value and the Volatility of the
Nominal Excess Return on Stocks
LAWRENCE R. GLOSTEN, RAVI JAGANNATHAN,
and DAVID E. RUNKLE*

ABSTRACT
We find support for a negative relation between conditional expected monthly
return and conditional variance of monthly return, using a GARCH-Mmodel
modified by allowing (1) seasonal patterns in volatility, (2) positive and negative
innovations to returns having different impacts on conditional volatility, and (3)
nominal interest rates to predict conditionalvariance. Using the modifiedGARCH-M
model, we also show that monthly conditional volatility may not be as persistent as
was thought. Positive unanticipated returns appear to result in a downward revi-
sion of the conditional volatility whereas negative unanticipated returns result in
an upward revision of conditional volatility.

THE TRADEOFFBETWEENRISK and return has long been an important topic in

asset valuation research. Most of this research has examined the tradeoff
between risk and return among different securities within a given time
period. The intertemporal relation between risk and return has been exam-
ined by several authors-Fama and Schwert (1977), French, Schwert, and
Stambaugh (1987), Harvey (1989), Campbell and Hentschel (1992), Nelson
(1991), and Chan, Karolyi, and Stulz (1992), to name a few. This paper
extends that research.
There is general agreement that investors, within a given time period,
require a larger expected return from a security that is riskier. However,
there is no such agreement about the relation between risk and return across
time. Whether or not investors require a larger risk premium on average for
investing in a security during times when the security is more risky remains
an open question. At first blush, it may appear that rational risk-averse

* Glosten is from Columbia University, Jagannathan is from the University of Minnesota and
the Federal Reserve Bank of Minneapolis, and Runkle is from the Federal Reserve Bank of
Minneapolis and the University of Minnesota. We benefitted from discussions with Tim Boller-
slev, William Breen, Lars Hansen, Patrick Hess, David Hsieh, Ruth Judson, Narayana Kocher-
lakota, Robert McDonald,Dan Nelson, and Dan Siegel, and from comments from David Backus
and Ren6 Stulz. Ruth Judson and Joe Piepgras did many of the computations. We are especially
grateful for the insightful and detailed comments of the referee. The usual disclaimer regarding
errors applies. Part of this research was performedwhile Glosten was a Visiting Economist at
the New York Stock Exchange. The views expressed herein are those of the authors and not
necessarily those of the Federal Reserve Bank of Minneapolis, the Federal Reserve System, or
the New York Stock Exchange and its members.

1779
1780 The Journal of Finance

investors would require a relatively larger risk premium during times when
the payoff from the security is more risky. A larger risk premium may not be
required, however, because time periods which are relatively more risky
could coincide with time periods when investors are better able to bear
particular types of risk. Further, a larger risk premium may not be required
because investors may want to save relatively more during periods when the
future is more risky. If all the productive assets available for transferring
income to the future carry risk and no risk-free investment opportunities are
available, then the price of the risky asset may be bid up considerably,
thereby reducing the risk premium.' Hence a positive as well as a negative
sign for the covariance between the conditional mean and the conditional
variance of the excess return on stocks would be consistent with theory. Since
there are conflicting predictions about this aspect of the tradeoff between risk
and return, it is important to empirically characterize the nature of the
relation between the conditional mean and the conditional variance of the
excess return on stocks as a group.
The empirical literature on this topic has attempted to characterize the
nature of the linear relation between the conditional mean and the condi-
tional variance of the excess return on stocks. However, the reported findings
are conflicting. For example, Campbell and Hentschel (1992) and French,
Schwert, and Stambaugh (1987) conclude that the data are consistent with a
positive relation between conditional expected excess return and conditional
variance, whereas Fama and Schwert (1977), Campbell (1987), Pagan and
Hong (1991), Breen, Glosten, and Jagannathan (1989), Turner, Startz, and
Nelson (1989), and Nelson (1991) find a negative relation. Chan, Karolyi, and
Stulz (1992) find no significant variance effect for the United States, but
implicitly find one of the world market portfolio. Harvey (1989) provides
empirical evidence suggesting that there may be some time variation in the
relation between risk and return.
Most of the support for a zero or positive relation has come from studies
that use the standard GARCH-M model of stochastic volatility.2 Other stud-
ies, using alternative techniques, have documented a negative relation be-
tween expected return and conditional variance. In order to resolve this
conflict we examine the possibility that the standard GARCH-M model may
not be rich enough to capture the time series properties of the monthly excess
return on stocks. We consider a more general specification of the GARCH-M
model. In particular, (1) we incorporate dummy variables in the GARCH-M
model to capture seasonal effects using the procedure first suggested by
Glosten, Jagannathan, and Runkle (1988), (2) we allow for asymmetries in
the conditional variance equation, following the suggestions of Glosten,

'Abel (1988), Backus and Gregory (1992), Gennotte and Marsh (1993), and Glosten and
Jagannathan (1987) have shown that the risk premium on the market portfolio of all assets
could, in equilibrium, be lower during relatively riskier times.
2 See Bollerslev, Chou, and Kroner (1992) for an extensive survey of GARCHand GARCH-M
models in finance.
 Volatility of Stock Returns 1781

Jagannathan, and Runkle (1988), (3) we include the nominal interest rate in
the conditional variance equation, and (4) we consider the EGARCH-M
specification suggested by Nelson (1991) with the modifications mentioned in
(1) through (3) above. These models suggest a weak but statistically signifi-
cant negative relation between conditional variance and expected return.
Two of our findings are somewhat at odds with the existing literature.
First, our data provide little evidence to support the belief that the condi-
tional volatility of the monthly excess return on stocks is highly persistent,
while Nelson (1991) finds high persistence in the volatility of daily returns.
There are no theoretical reasons for the properties of the monthly and daily
returns to be the same. In particular, Nelson (1991) argues that as the
frequency at which data are sampled becomes very high, persistence should
become larger. Second, both unexpected positive and negative excess returns
on stocks change the next period's conditional volatility of the excess return
on stocks. Unexpected positive returns result in a downward revision while
unexpected negative returns result in an upward revision. In contrast, Nelson
(1991) and Engle and Ng (1993), using daily data on stock index returns, find
that large positive as well as negative unanticipated returns lead to an
upward revision in the conditional volatility, although negative shocks of
similar magnitude lead to larger revisions. Hence the time series properties
of monthly excess returns are somewhat different from those of daily returns
reported in Nelson (1991) and Engle and Ng (1993), and our results for
monthly data along with the results for daily data reported by others provide
a more complete characterization of the time series properties of stock index
returns.
The remainder of the paper proceeds as follows. Section I describes the
model that forms the basis for our empirical analysis. Section II discusses the
econometric issues involved and our estimation methods. Section III contains
the empirical results. Section IV concludes.

I. The Relation between the Conditional Mean and the

Conditional Variance of the Excess Return on Stocks
Consider the relation between conditional variance and conditional mean
given by

t[ Xt+]= /3 ' (1)

When xt+ 1 is the excess return on the aggregate wealth portfolio, and o-t2, its
variance assessed at time t, captures most of the economic uncertainty that
agents care about, the model in (1) is the approximation to the true risk-return
relation derived by Merton (1980).
In our empirical work, we assume that (1) holds even for nominal returns.
We consider the following general model for estimation:

E[xt+, IFt] = a +P Var(xt,1 I Ft), (2)

1782 The Journal of Finance

where F, denotes the information set of agents. Campbell (1993) provides

sufficient conditions for the relation given in (2) to hold approximately in
equilibrium, where xt is the excess return on the market index portfolio.
However, p is not in general a measure of the risk aversion coefficient of the
representative agent and a is not in general equal to zero. The relation in (2)
forms the basis for our empirical work.

II. Estimating the Model

A. Econometric Issues
The parameter p in the model given by (2) cannot be estimated without
specifying how variances change over time, since Var(xt +1? Ft) is not directly
observed by the econometrician. To appreciate the difficulties involved, pro-
ject both sides of (2) on Gt, the econometrician's information set, which is a
strict subset of the agents' information set Ft. With vt-1 = Var(x tFt 1), we
get
E [x t IGt1- 1 a + 8E [vt -1 IGt- 1 (3)
Hence, we can write
xt= a + pE[vt_l IGt-] + qt

where
'lt = ut-1 + Et, ut-i = (vt-1 - E[vt- 1 Gt- 1D
and
Et = xt- E[xt IFt-1].
Since, by definition
st IF=
Et,E[E2 I t_ 1] =Vt_ 1

ut-= p[E[ IFt_1] -E[Et IG1]] .

Note that
E[-qt I Gt-l] E[ut-1 I Gt-l] E[ ut- ,Et I Gt- 1] = ?.

Therefore,

E[7 1G21] = E[u 2_1 IGt_1] + E[Etti Gt- 1]. (4)

The term on the left in equation (4) is the variance of the error in forecasting
xt based on the econometrician's information set. The first term on the right
is the variance of the measurement error, (vt1 - E[ vt 1lGt_-]), and the
second term is the expected value of the conditional variance vt- 1 based on
the econometrician's information set Gt 1. Unless the variance of the mea-
surement error is a constant, we cannot obtain a consistent estimate of P3.
This problem was first pointed out by Pagan and Ullah (1988). Also notice
that the intercept term a in equation (2) is not identifiable based on the
 Volatility of Stock Returns 1783

smaller information set available to the econometrician, since E[vt_1IGt_ ]

may involve a constant term.
To see this estimation problem more clearly, consider now the special case

E[vt- 1I Gt_ ] = bo + blzt-1

for some z E G where b1 is a row vector and z t1 is a column vector. Then

E[It2 I zt-1] = p2 Var(vt- 1 i Gt- 1) + (bo + blzt- 1).

The left side is the variance of the excess return, conditional on observing the
instrument z alone. The first term on the right side is the variance of the
measurement error 8 (vt 1 - E[ vt- 1Gt -]), and the second term is the vari-
ance of et given zt-1
There have been several approaches to the estimation of the general
econometric model in (3) through (4). One approach was suggested by Camp-
bell (1987) and assumes that Var(vt1llGt- ) is an arbitrary constant, while
Zt- is a vector of observable variables. If Var(vt 1IGt-) is a constant, then
we can test whether /8 is positive. We can estimate the regression equations
x t =Co + clZt-l + qt(5)
and
=qt2 do + d1zt_l + Vt. (6)
Since the estimated slope coefficient cl is a consistent estimate of /3bl, and
d1 provides a consistent estimate of b1, the ratio of any two corresponding
elements of cl and d1 provides a consistent estimate of /8. If zt-, is not a
scalar, then we may impose the constraint that the slope coefficients in (5)
and the slope coefficients in (6) differ only by the scale factor, /3. Such a
restriction also provides a natural test for the validity of the model specifica-
tion. We call this approach Campbell's Instrumental Variable Model. Another
approach, the GARCH-M model, assumes that Var(vtllGt_-) is identically
zero, and that zt-1 consists of innovations and variances that, while unob-
servable, can be estimated by the econometrician. A generalization of the
GARCH-M approach maintains the assumption that Var(vtllGt- ) is zero
but allows zt- 1 to consist both of observable instruments and lagged values
of estimated variances and innovations. We refer to this approach as the
Modified GARCH-M Model. Since the specification of the information set is
crucial for the Modified GARCH-M Model, we will first describe the informa-
tion set used in this study and then proceed to describe the GARCH-M
models we examine.

B. Specification of the Econometrician's Information Set

Estimating any model of the intertemporal relation between risk and
return requires taking a stand on the variables that make up the instrument
vector, zt. In our investigations we focus attention on the volatility informa-
tion in the following variables: (a) the nominal interest rate, (b) October and
1784 The Journal of Finance

January seasonal dummies, and (c) the unanticipated part of the excess
return on stocks. In what follows we provide some justification as to why we
focus our attention on these variables.
The use of nominal interest rates in conditional variance models has some
intuitive appeal. It has been well known since Fischer (1981) that the
variance of inflation increases with its level. To the extent that short-term
nominal interest rates embody expectations about inflation, they could be a
good predictor of future volatility in excess returns. Using the information
contained in nominal interest rates, Fama and Schwert (1977), Campbell
(1987), and Breen, Glosten, and Jagannathan (1989) have demonstrated that
it is possible to forecast time periods when the excess return on stocks is
relatively large and significantly less volatile. Giovannini and Jorion (1989)
and Singleton (1989) also examine the ability of nominal interest rates to
predict changes in the volatility of, respectively, foreign exchange and stock
returns.
Including deterministic seasonal dummies is motivated by the seasonal
patterns reported in Lakonishok and Smidt (1988) and Keim (1985). Table I
presents the summary statistics for the monthly excess returns on the Center
for Research in Security Prices (CRSP) value-weighted stock index portfolio
during the post-Treasury Accord period for the months of October, January,
and other calendar months.3 An apparent increase in October and January
volatility is suggested by results presented in Panels A and C.
During the period 1951:4 to 1989:12, monthly excess continuously com-
pounded returns on the CRSP value-weighted index of stocks (Panel C),
during months other than October and January, had a mean of 0.48 percent
and a standard deviation of 3.83 percent. The standard deviation of January
excess returns is 5.19 percent (i.e., 1.35 times that in other months) and the
standard deviation of October excess returns is 6.17 percent (1.61 times that
in other months). While October and January are both months of relatively
larger volatility, October, unlike January, has relatively lower excess returns
on average than other months.
There are several potential contributing explanations for the excess volatil-
ity of January excess stock returns. Relatively more news arrives in January
since most firms (almost two-thirds) use the calendar year as their fiscal
year. Such firms close their books on December 31. The annual reports are
typically, more informative since they are done more carefully and are au-
dited. Information from such reports starts leaking in during the month of
January. Further, consumer sales exhibit pronounced quarterly seasonal
patterns. This pattern arises because the fourth quarter is the important

3A careful examination of the data in Lakonishok and Smidt (1988) suggests that the monthly
seasonal patterns in volatility are unlikely to be captured adequately by treating months other
than October and January as being similar. However, our objective in this paper is limited to
showing how to model seasonals in a way different than what has been done in the literature.
Characterizing the nature of the monthly patterns in volatility is left as an exercise for the
future.
 Volatility of Stock Returns 1785

Table I
Summary Statistics for Monthly Data Recorded in the
Period 1951:4 to 1989:12
Number of Observations
Other
Oct. Jan. Months
39 38 388
Panel A. Continuously Compounded Monthly Return on the CRSP Value-weighted Index
of Status on the NYSE
Mean (xlOO) 0.25 1.77 0.91
Standard Deviation (xlOO) 6.16 5.18 3.80
Skewness -1.35 0.21 -0.44
Kurtosis 6.59 -0.41 0.64

Panel B. Continuously Compounded Monthly Return on Treasury Bills from

Ibbotson & Associates
Mean (xlOO) 0.46 0.43 0.43
Standard Deviation (xlOO) 0.26 0.22 0.26
Skewness 0.83 0.66 0.96
Kurtosis 0.54 -0.03 0.90
Panel C. Excess of the Continuously Compounded Monthly Return on the CRSP
Value-weighted Index of Status on the NYSE over That on Treasury Bills
Mean (xlOO) -0.21 1.34 0.48
Standard Deviation (xlOO) 6.17 5.19 3.83
Skewness -1.40 0.18 -0.48
Kurtosis 6.70 -0.43 0.65
Mean/variance 0.56 0.50 3.28

holiday season. Comprehensive and reliable information about consumer

spending during this period typically becomes available during January.
One has to stretch to provide an explanation for October volatility, and had
this study been done prior to October 1987, we probably would not have
singled out October. Yet, there is the perception that October has excess
volatility. Laurie Cohen, writing in a Wall Street Journal article, attributes to
Mark Twain the observation, "October is one of the peculiarly dangerous
months to speculate in stocks. The others are: July, January, September,
April, November, May, March, June, December, August, and February." Our
purpose is to illustrate a methodology for dealing with seasonal patterns
rather than to provide a detailed analysis of them. Though the seasonal
variables are statistically significant, our general conclusions from the model
estimations do not depend upon the inclusion of the seasonal variables.
Rather, their significance suggests the value of further investigations of
seasonal patterns.
We also allow positive and negative innovations to returns to have different
impacts on conditional variance. To see why this is desirable, suppose dis-
1786 The Journal of Finance

count rates are constant and have no relationship to anticipated future

volatility. Any unanticipated decrease in expected future cash flows decreases
the stock price. If the variance of the future cash flows remains the same or
does not fall proportionately to the fall in stock prices, the variance of future
cash flows per dollar of stock price will rise and future returns will be more
volatile. Hence, if most of the fluctuations in stock prices are caused by
fluctuations in expected future cash flows and the riskiness of future cash
flows does not change proportionally when investors revise their expecta-
tions, then unanticipated changes in stock prices and returns will be nega-
tively related to unanticipated changes in future volatility. Black (1976) and
Christie (1982) have suggested a different reason for the negative effect of
current returns on future variance: a decrease in today's stock price changes
a firm's capital structure by increasing leverage. This increased leverage
causes higher expected variance in the future. Both Black and Christie find
support for their predictions in the relation between expected return and
variance for individual stocks.

C. Modified GARCH-M Model of the Variance

C. 1. Model Specifications
The GARCH model assumes that the information set of investors and the
econometrician coincide. The general Modified GARCH-M model can be writ-
ten

Equation for the conditional mean:

E[xt+llGt] = t(Gt)
where 4id) is a function that describes the nature of the dependence of the
conditional mean on the elements of the information set Gt. Hence, we can
write
xt+1 = ,A(Gt) + Et+?, with E[et+llGt] = 0.

Equation for the conditional variance:

Var(xt+1lGt) Var(et+IGt) = V(Gt)

where V(.) is a function that describes the nature of the dependence of the
conditional variance on the elements of the information set Gt. It is conve-
nient to assume that the conditional variance function can be decomposed in
the following way:4
V(Gt) = fm (Gt - 1) + f(Gt \ Gt - 1)

where fm(Gt 1) is that part of the conditional variance, V(Gt), that depends
only on information known as of date t - 1, and f(Gt \ Gt_ 1) is that part of

4In the case of the EGARCH-M models, V(.) is the log of the variance.
 Volatility of Stock Returns 1787

the conditional variance that depends on the new information, Gt \ Gt1, that
becomes available at date t. Our analysis of various specifications focuses on
the function f(.).
Now consider the standard GARCH-M process suggested by Bollerslev
(1986) for stock excess return, xt, given by

Model 1:
xt = ao + a1vt-1 + ot (7)
vt1 = bo +b1v_2
+g1Eb2 (8)
where Et- [et] = 0 and Et- [E2] = vt1. The GARCH-M model specifies the
conditional mean function, pX(Gt-) = ao + a1vtv1, and the conditional vari-
ance function, V(Gt_1) = bo + b1v_2 ?g1EU1. That is, fm(Gt2) = bo +
b1v-2 and f(Gt-1 \ Gt2) = g1e21. The univariate GARCH-M model as-
sumes that the econometrician's information set consists only of the past
innovations to the excess return, xt. Hence, the only new information that
becomes available at date t - 1 is et.- 1 The model further assumes that the
function f(Gt 1 \ Gt-2) = g1ft 1. As we have argued earlier, there are a
priori reasons to suspect that this assumption may not be reasonable.
If future variance is not a function only of the squared innovation to
current return, then a simple GARCH-M model is misspecified and any
empirical results based on it alone are not reliable. In Model 2 we assume
that the impact of E2 1 on conditional variance vt1 is different when et- 1 is
positive (i.e., when the indicator or dummy variable It 1 in (9) is 1) than
when et- 1 is negative (i.e., when the indicator or dummy variable It_ in (9)
is 0). This leads to

Model 2:
vt- 1 = bo + b1vt-2 +g1et-1 +g2 t 1I_1. (9)
In Models 3 through 5 we relax the assumption that the information set,
Gt, consists only of past realizations of the excess return on the portfolio.
Including the risk-free interest rate, rft, leads to

Model 3:

Vt- = bo + b1v_2 + b2 rft + g1E21 + g2 Et1I_1. (10)

Given the results of Table I and for reasons mentioned earlier, we intro-
duce January and October seasonal dummies in the variance of stock index
excess returns. For this purpose we assume that the seasonal effects amplify
the underlying fundamental volatility (which does not by definition exhibit
any seasonal patterns) in the months of October and January by a constant
month-specific scale factor. We also assume that the fundamental volatility
next period depends only on the fundamental part of the excess return
innovation.
1788 The Journal of Finance

In particular, we assume that we can write the excess return innovation in

any calendar month as a scale multiple of some underlying fundamental
innovation that does not exhibit any seasonal patterns, as follows:
Et = (1 + A1OCTt + A2 JANt)'qt

where -qt does not exhibit any deterministic seasonal behavior. Let ht-1 =
E _1[ t ] denote the conditional variance of 't. We postulate that ht evolves
over time according to

Model 4:
2
ht-, = bo + blht2 ?g1'q-1 g2 lIt_1; (11)

or, when the risk-free rate is included,

Model 5:
ht-, = bo + blht2 + b2rft + g, 1l ?g2 j_I_1. (12)
Notice that Model 1 is obtained from Model 5 by imposing the restriction that
A1 = A2 = b2 = 92 = 0. Similarly, Models 2, 3, and 4 can be considered as
restricted versions of Model 5.
Our approach to modelling seasonals is different from the one used by
Baillie and Bollerslev (1989). In our specification, we assume that we can
deseasonalize the excess return innovation, E, to get -q.The realized value of
the deseasonalized innovation, , influences the conditional variance of the
distribution from which the deseasonalized innovation for the next period is
drawn from. In contrast, in Baillie and Bollerslev (1989), the seasonal part of
the innovation to this period's return affects the variance of the deseasonal-
ized innovation next period.
Because inference in GARCH-M models depends on the correct specifica-
tion of the information set and the validity of the functions used to represent
the conditional mean and the conditional variance, we estimate three addi-
tional models to check our specification. First, we check for nonlinearity in
the mean equation by adding v1'2f to Model 2 and Model 4. These models are
then called Model 6 and Model 7. If the coefficient on vt- 1' is significantly
different from zero, that difference is evidence of misspecification.
In the above models, there are a priori reasons to suspect that the coeffi-
cient g2 as well as g1 -+ g2 are negative, since empirical evidence suggests
that a positive innovation to stock return is associated with a decrease in
return volatility. However, if g1 + g2 is negative, conditional variance can
potentially become negative for some realization of e. Hence we also follow
the suggestions of Engle (1982) and Nelson (1991) and consider the exponen-
tial form for the law of motion for.conditional variance, as given below:

Model 2-L:

log(ht -1) = b0 + b1 log(h t- 2) + glit- ht2 + g2t- 1It- 1/ h2. (13)

 Volatility of Stock Returns 1789

Following Nelson (1991), we use t- 1/ ht2 instead of functions of 'q2_1 in

equation (13) to minimize the impact of extremely large realizations in
absolute value so that the stochastic process for ht will be well behaved.5
Model 1-L is the same as Model 2-L but with g2 restricted to be zero.
Since we also want to test whether the risk-free rate, rft, helps predict
conditional variance using the log specification, we also estimate

Model 3-L:

log(ht-1) = b0 + b llog(ht-2) + b2rft +?glt-1/ ht2+g2tt-I i/t- h

(14)

Models 4-L and 5-L add deterministic seasonals to the variance equation of
Models 2-L and 3-L in the manner adopted for the level specification. Two
additional models were estimated to test the specification of the EGARCH-M
model. Model 6-L adds v1/ 2 to the mean equation for Model 4-L.
C.2. Estimation and Inference and Diagnostic Tests
We estimate all models discussed in this section by maximizing the log-
likelihood function for the model, assuming that Et is conditionally normally
distributed. Even if this assumption is incorrect, as long as the conditional
means and variances are correctly specified, the quasi-maximum likelihood
estimates will be consistent and asymptotically normal, as pointed out by
Glosten, Jagannathan, and Runkle (1988) and Bollerslev and Wooldridge
(1992). All our inference is based on robust standard errors from the quasi-
maximum likelihood estimation, employing the procedures described in
Bollerslev and Wooldridge (1992) and Glosten, Jagannathan, and Runkle
(1988). We compute robust standard errors using two-sided numerical deriva-
tives.6
We also use a variety of diagnostic tests to determine whether various
aspects of our different models are correctly specified. First, we examine
whether the residuals of the estimated models display excess skewness and
kurtosis. Properly specified GARCH-M and EGARCH-M models should be
able to significantly reduce the excess skewness and kurtosis evident in
nominal excess returns. We test for excess skewness and kurtosis, under the

5 Potential negative values for the constructed conditional variances are not the only possible
reason for using the log specification. It may also be true that the log model simply models the
true conditional variance better than the level model. For more on this issue, see Engle and Ng
(1993).
6 Since we use dummy variables which take the value of one or zero, it may appear as though
we may be violating the differentiability assumptions underlying the derivation of the robust
standard errors. Note, however, that since the dummy variables are multiplied by the corre-
sponding squared innovations, the differentiability conditions will be satisfied for the modified
GARCH-Mmodels we consider. Although the differentiability conditions will be violated for the
modified versions of Nelson's E-GARCHmodel we consider, this is unlikely to be an issue since
points at which the differentiability assumptions are not satisfied will occur with zero probabil-
ity, and the numerical derivatives we compute are always bounded.
1790 The Journal of Finance

null hypothesis that the errors are drawn from a conditional normal distribu-
tion. These tests have been previously applied to GARCH-M models by
Campbell and Hentschel (1992).
Second, we examine whether the squared standardized residuals from the
estimated models, (Et/ t2, are independent and identically distributed.
We use the three tests proposed by Engle and Ng (1993): the Sign Bias Test,
the Negative Size Bias Test, and the Positive Size Bias Test as well as a joint
test of all three.
In the Sign Bias Test, the squared standardized residuals are regressed on
a constant and a dummy variable, denoted St, that takes a value of one if
Et-l is negative and zero otherwise. The Sign Bias Test Statistic is the
t-statistic for the coefficient on S -. This test shows whether positive and
negative innovations affect future volatility differently from the prediction of
the model.
In the Negative Size Bias Test, the squared standardized residuals are
regressed on a constant and St Et 1- The Negative Size Bias Test Statistic is
the t-statistic for the coefficient on St Et 1- This test shows whether larger
negative innovations are correlated with larger biases in predicted volatility.
In the Positive Size Bias Test, the squared standardized residuals are
regressed on a constant and St Et- where St = 1 - St. The Positive Size
Bias Test Statistic is the t-statistic for the coefficient on St Et-1. This test
shows whether larger positive innovations are correlated with larger biases
in predicted volatility.
There is one additional comparison that we make among the models,
although it is not formally a diagnostic test. Because the parameterization of
the models differs so much, it is hard to compare the amount of persistence in
variance that these models predict. One way to compare persistence in
variance across models is to regress ht on a constant and ht 1. We report the
slope coefficient and its standard error (it is one over the square root of the
number of observations) of the regression for each model.

III. Empirical Results

Our objective is to examine the role of model specification in determining
the estimated relation between risk and return. The discussion above sug-
gests that we can estimate this relation using either Campbell's Instrumental
Variable Model or a variety of Modified GARCH-M and EGARCH-M models.
While our focus is on the latter, we first present results using the first
apprcach and find that Campbell's general conclusions are replicated in our
data. We then present the results for the various GARCH-M models.

A. Campbell's Instrumental Variable Model

Table II provides the empirical results obtained when the CRSP value-
weighted index of stocks on the New York Stock Exchange (NYSE) is used as
 Volatility of Stock Returns 1791

Table II
Monthly Risk-Return Relation on the CRSP Value-weighted
Index of NYSE Equities: Campbell's Instrumental Variable
Approach, 1951:4 to 1989:12
The variable, xt, is the differential between the continuously compounded monthly return on the
CRSP value-weighted index of equities on the NYSE and rft, the continuously compounded
monthly return on Treasury bills from Ibbotson & Associates. The variable JANt takes the value
one in January and zero otherwise, and OCTt takes the value one in October and zero otherwise.
The t-statistics are computed using the procedures in Hansen (1982) which allows for conditional
heteroskedasticity. The reported t-statistics are for 20 lags.

Model A

Mean Equation: xt = c0 + cl OCTt + c2 JANt + c3rft + Et

Variance Equation: Es = do + d1 OCTt + d2 JANt + d3rft + t

do X 104 X 104 d2 X 104 d3 co x 100 c1 X 100 c2 X lOO c3

d,
Coefficient 7.25 21.68 13.11 0.18 1.48 -0.63 0.86 -2.31
t-Statistic 2.70 1.35 2.25 2.74 5.21 -0.56 1.01 -3.42

Model B

Restricted Mean Equation: xt = c0 + fd1 OCTt + fd2 JANt + fd3rft + Et

Variance Equation: E2 = do + d, OCTt + d2 JANt + d3rft + t

do x 104 d1 X 104 d2 X 104 d3 co X 100 p ChiSQ(2) p-Value

Coefficient 8.25 9.23 4.19 0.15 1.44 - 12.75 3.22 0.20
t-Statistic 3.49 0.87 1.42 2.90 5.02 -2.43

the stock index portfolio. We limit attention to 1951:4 to 1989:12, which is the
post-Treasury Accord period. The estimated value of the slope coefficient for
the risk-free rate in equation (5) for expected excess return is -2.31 (t =
- 3.42.). The estimated value of the slope coefficient for the risk-free rate in
the variance equation given by (6) is 0.18 (t = 2.74).
Note that the residuals in equations (5) and (6) can be serially correlated,
since the econometrician's information set may be strictly smaller than that
of economic agents. Therefore, the t-statistics were computed using the
procedures suggested by Newey and West (1987). Since there is substantial
persistence in the residuals of equation (6), we report the t-statistics corre-
sponding to a lag length of 20. In this sample, the t-statistics decrease as the
number of lags increases, but stabilize at about 10 lags.
We also estimate the model by imposing the constraint that the slope
coefficients in equation (5) be scalar multiples of the slope coefficients in
equation (6). The estimated value of the scalar, /3, is - 12.75 (t = -2.43).
With this restriction are two over-identifying restrictions. The null hypothe-
sis that the over-identifying restrictions are not binding lead to a chi-square
(D.F. = 2) value of 3.22 with an associated p-value of 0.20. Hence, based on
these results, we cannot reject the hypothesis that there is a negative relation
1792 The Journal of Finance

between the conditional mean and conditional variance of the excess return
on stocks.
The natural question that arises at this stage is why the findings reported
by French et al. (1987) for the standard GARCH-M model are different from
the conclusions in this section. We address this issue in the next section.

B. Modified GARCH-M Models

Tables III and IV present the estimates for Models 1 through 7. A compari-
son of Model 1 and Model 2 illustrates the restrictive nature of the standard
GARCH-M specification for the conditional variance equation. Model 1 pre-
sents the results for a standard GARCH-M model. Both positive and negative
innovations to excess returns result in upward revisions of the conditional
variance (g1 is positive). Also, time periods with relatively large variances
are associated, on average, with relatively larger returns (a1 is positive).
However, the association is weak and not statistically significant at conven-
tional levels.
These relations change as soon as positive and negative unanticipated
returns are allowed to have different effects on conditional variance. Model 2
allows for such a difference by estimating the parameter g2. A simple
specification test comparing Model 1 and Model 2 shows that the standard
GARCH-M model is too restrictive. If the parameters that the two models
share are compared using a generalized specification test, computed using
robust standard errors, the value of the test statistic is 12.829.7 Under the
null hypothesis that Model 1 is correctly specified, this test statistic should be
asymptotically distributed as a X5 random variable. Thus, we can reject the
null hypothesis at the 5 percent level. Note that in Model 2 an unexpected
negative return sharply increases conditional variance of the next period
excess return, while an unexpected positive return decreases conditional
variance. Model 1 does not allow for such a possibility.
There is another important difference between Model 1 and Model 2. Table
IV shows that for Model 2, the estimated persistence of variance from one
period to the next, as measured by the first-order autoregressive coefficient
for ht, is smaller than it is for Model 1.
Even though Model 2 seems less restrictive than Model 1, there are two
reasons that we should. not be satisfied with it. First, the robust standard
errors suggest that the coefficient g2 is imprecisely estimated. In fact, it is
not significantly different from zero. Second, Model 2 does little better than
Model 1 in all of the diagnostic tests.
Models 3, 4, and 5 attempt to solve the deficiencies in Model 2 by including
the effect of the risk-free interest rate and deterministic seasonals on condi-
tional variance. Each of these models results in statistically significant
asymmetry in the conditional variance equation (i.e., g2 is not equal to zero).

7For more on generalized specification tests, see Newey (1985).

 Volatility-of Stock Returns 1793

Table III
Monthly Risk-Return Relation of the CRSP Value-weighted
Index of NYSE Equities: Modified GARCH-M, 1951:4 to 1989:12
(Model Estimates)
With xt the differential between the continuously compounded monthly return on the CRSP
value-weighted index of equities on the NYSE and rft, the continuously compounded monthly
return on Treasury bills from Ibbotson & Associates, the models are defined by

xt = ao + a1vt-1 + a2v 2l + Et; Et=(1 + A1QCT - A2JAN)-qt; vt_1 = Vart- GEt);

ht_ 1 = Vart - 11t); It- 1= 1 if qt- 1 > 0, and 0 otherwise;

ht-, = bo + blht-2 + b2rft +?g 1 t- +?g2qtr_tIt-1;

where OCT takes the value one in October and zero otherwise and JAN takes the value one in
January and zero otherwise. Robust t-statistics (in brackets) are calculated using the procedure
in Bollerslev and Wooldridge (1992).

Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7

ao - 0.453 1.064 1.850 1.071 1.854 5.730 7.782
(xlOO) [-0.575] [1.947] [4.232] [2.398] [4.419] [2.064] [2.289]
a1 5.926 - 2.843 - 7.625 -3.165 - 8.019 16.893 30.349
[1.307] [-0.878] [-2.621] [- 1.131] [-2.828] [1.163] [1.652]
bo 0.016 0.074 0.035 0.026 0.030 0.059 0.018
(xlOO) [2.058] [1.653] [1.595] [2.467] [2.477] [1.902] [3.295]
b,10.842 0.483 0.334 0.769 0.506 0.623 0.848
[16.758] [1.572] [0.824] [8.801] [2.927] [2.830] [17.299]
b2 0.159 0.078
[1.433] [2.221]
91 0.070 0.257 0.188 0.153 0.177 0.161 0.121
[2.541] [1.709] [2.109] [2.590] [2.268] [1.502] [2.734]
92 - 0.340 -0.248 -0.227 -0.252 -0.267 - 0.207
[- 2.270] [- 2.602] [-3.570] [- 3.272] [-2.683] [- 3.749]
A1 0.677 0.454 0.606
[3.638] [3.120] [3.600]
A2 0.269 1.254 0.306
[1.816] [1.795] [1.936]
a2 - 1.983 - 3.123
[- 1.576] [- 1.947]
Log
likelihood 1248.202 1252.917 1266.288 1268.422 1276.518 1254.276 1271.824

In Model 3, the risk-free rate is included as an explanatory variable in the

conditional variance equation in Model 2. Note that a1 and g2 both become
statistically significant. The coefficient on the risk-free rate itself, b2, is
positive, but not significant, and may be imprecisely measured. Table IV
shows that excess skewness and kurtosis remain quite severe after the
risk-free rate is included.
1794 The Journal of Finance

Table IV
Monthly Risk-Return Relation on the CRSP Value-weighted
Index of NYSE Equities: Modified GARCH-M, 1951:4 to 1989:12
(Diagnostic Tests)
Skewness and Kurtosis are the estimated skewness and kurtosis of the estimated standardized
residuals from the mean equation. The Sign bias, Negative size bias, Positive size bias, and Joint
tests are those suggested by Engle and Ng (1993). We report the slope coefficient and t-statistic
from the regression of the squared standardized residual on (respectively) (1) an indicator
variable which takes the value one if the residual is negative and zero otherwise, (2) the product
of this indicator variable and the residual, and (3) the product of the residual and an indicator
variable that takes the value one if the residual is positive and zero otherwise. The AR(1)
coefficient is the slope coefficient from the regression of the fitted deseasonalized variance at
time t on the fitted deseasonalized variance at time t - 1.

Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7

Skewness -0.781 -0.701 -0.463 - 0.455 - 0.289 - 0.709 - 0.478
t-Statistic - 6.855 - 6.157 - 4.065 - 3.992 - 2.532 - 6.225 - 4.190
Kurtosis 3.177 3.359 1.918 0.927 0.557 3.527 1.154
t-Statistic 13.880 14.675 8.380 4.052 2.431 15.408 5.040
Sign bias 0.745 0.988 0.776 0.710 0.429 0.841 0.673
t-Statistic 2.280 2.992 2.756 2.932 1.895 2.515 2.613
Negative size bias 0.013 8.117 5.758 5.539 3.861 4.243 5.477
t-Statistic 0.0028 1.612 1.329 1.483 1.052 0.837 1.389
Positive size bias -3.866 2.138 1.342 0.015 - 1.220 1.988 -0.826
t-Statistic -0.596 0.345 0.256 0.0035 - 0.314 0.316 -0.166
Joint test 16.797 11.854 10.851 13.050 6.459 9.415 11.997
Significance level 0.0008 0.0079 0.013 0.0045 0.091 0.024 0.0074
AR(1) Coefflcient
on deseasonalized
conditional variance
(std. error = 0.0464) 0.897 0.494 0.609 0.265 0.374 0.677 0.355

In Model 4, deterministic seasonals are added to Model 2. A Wald test that

the October and January seasonal effects are jointly significant (i.e., the null
that A1 = A2 = 0) is estimated to be 6.38. Under the null, this statistic should
be asymptotically distributed as a X22 random variable. Thus, we can reject
the hypothesis that A1 = A2 = 0 at the 5 percent level. There are three other
important characteristics of this model worth noting. First, with the inclusion
of deterministic seasonals, both g1 and g2 are statistically significant (as
they were in Model 3). Second, this method of modelling seasonals in variance
greatly reduces the excess kurtosis in the residuals. Finally, note that the
amount of persistence in the -conditional variance, as measured by the
first-order autoregressive coefficient for ht, is much smaller than in any of
the previous models.
Model 5 adds both the risk-free rate and deterministic seasonals to Model
2. As in Model 4, both g1 and g2 are significantly different from zero. Unlike
 Volatility of Stock Returns 1795

in Model 3, b2, the coefficient on the risk-free rate, is significantly positive.

Note that the serial correlation in the estimated conditional variances is
much smaller than the standard GARCH-M model, Model 1.
Table IV shows that the level of excess skewness and kurtosis have been
significantly reduced-although the null hypothesis of no excess skewness or
kurtosis can be rejected at the 5 percent level. Model 5 is also the first model
that does not fail the Sign Bias Test at the 5 percent level. In addition, it is
the first model not to fail the joint test of sign bias, negative size bias, and
positive size bias at the 5 percent level. These diagnostics suggest that Model
5 is the most satisfactory model considered thus far.
Despite its success, Model 5 is still quite fragile. We attempted to check the
robustness of the specification by adding v l/~2 to the mean equation. Even
with great effort, we were not able to get the parameter estimates from that
model to converge. Models 6 and 7 show the effects of adding vt12l to the
mean equation in Models 2 and 4, respectively. In neither case was the
coefficient on v 1/2l statistically significant. Note also that the estimated
coefficients in the conditional variance equation are relatively close in all of
these models.
We therefore come to the following five conclusions from our examination of
the seven different GARCH-M specifications.
1. The relation between conditional mean and conditional variance is
negative and statistically significant;
2. the risk-free rate contains information about future volatility, within the
Modified GARCH-M framework;
3. the October and January seasonals in volatility are statistically signifi-
cant;
4. conditional volatility of the monthly excess return is not highly persis-
tent; and
5. negative residuals are associated with an increase in variance, while
positive residuals are associated with a slight decrease in variance.
With the exception of the third conclusion, these results hold with or without
the inclusion of seasonals.
Because even the best of these models displays excess skewness and
kurtosis, we also estimated different EGARCH-M models. The estimates for
Models 1-L and 6-L are shown in Tables V and VI. Models 1-L and 2-L are
based on the EGARCH-M model proposed by Nelson (1991). However, the
results in these models, using monthly data, are quite different from those
found by Nelson. Note that g2, the coefficient detecting asymmetry in the
conditional variance equation, has a very small t-statistic. Table V shows
that both Model 1-L and Model 2-L display excess skewness and kurtosis, and
that both fail the Sign Bias Test. Note also that the first-order serial
correlation of the h 's in both models is quite low.
Unlike in the level models, the coefficient on g2 remains insignificant,
regardless of which additional variables are included in the conditional
variance equation. As a result, we do not report these regressions. Instead,
1796 The Journal of Finance

Table V
Monthly Risk-Return Relation on the CRSP Value-weighted
Index of NYSE Equities: Modified EGARCH-M, 1951:4 to
1989:12 (Model Estimates)
With xt the differential between the continuously compounded monthly return on the CRSP
value-weighted index of equities on the NYSE and rft, the continuously compounded monthly
return oni Treasury bills from Ibbotson & Associates, the models are defined by

= aO + a1vt-1 + a2vl'i
=t + Et; Et=(1 + AlOCT + A2JAN)rqt; vt-1=Vart- (Et);
ht_ 1 = Vart - 61t); Ht- 1= log(ht- 1); It- l = 1 if r-t- l > 0, and 0 otherwise;

Ht_ = bo + bjHtH2 + b2rft +?g(t11ht-2) +g2(,qt- / 2)Iti ;

where OCT takes the value one in October and zero otherwise, and JAN takes the value one in
January and zero otherwise. Robust t-statistics (in brackets) are calculated using the procedure
in Bollerslev and Wooldridge (1992).

Model 1-L Model 2-L Model 3-L Model 4-L Model 5-L Model 6-L

ao 1.195 1.199 1.604 1.097 1.536 3.890

(xlOO) [2.932] [2.290] [4.176] [2.657] [4.158] [1.231]
a1 - 3.947 -4.022 - 6.387 -3.486 - 6.119 12.157
[- 1.532] [ - 1.537] [ - 2.483] [ - 1.305] [- 2.426] [0.707]

a2 - 1.366
[-0.908]
bo - 5.583 -5.567 - 5.728 - 5.035 - 5.102 - 4.479
(xlOO) [- 6.476] [ - 6.4721 [ - 5.719] [ - 4.513] [- 4.390] [- 2.654]

b, 0.133 0.133 0.183 0.235 0.281 0.321

[0.999] [0.992] [1.271] [1.386] [1.701] [1.248]
b2 99.600 81.002
[3.624] [3.127]
91 - 0.456 -0.427 - 0.383 -0.378 - 0.338 - 0.355
[- 6.448] [ -3.599] [- 5.1951 [- 5.249] [- 4.811] [- 3.738]
92 --0.052
[ -0.274]
Al 0.426 0.349 0.451
[2.573] [2.392] [2.579]
A2 0.318 0.299 0.346
[2.027] [2.035] [2.017]
Log
likelihood 1262.937 1262.274 1272.081 1270.970 1279.081 1271.538

we try to address the deficiencies in Model 1-L in Models 3-L, 4-L, and 5-L by
including the effect of the risk-free interest rate and deterministic seasonals
on conditional variance. Note that Models 3-L through 5-L impose the restric-
tion that g2 = 0.
Model 3-L adds the risk-free rate to the conditional variance equation in
Model 1-L. In contrast to Model 3, the coefficient on the risk-free rate has a
 Volatility of Stock Returns 1797

Table VI
Monthly Risk-Return Relation on the CRSP Value-weighted
Index of NYSE Equities: Modified EGARCH-M, 1951:4 to
1989:12 (Diagnostic Tests)
Skewness and Kurtosis are the estimated skewness and kurtosis of the estimated standardized
residuals from the mean equation. The Sign bias, Negative size bias, Positive size bias, and Joint
tests are those suggested by Engle and Ng (1993). We report the slope coefficient and t-statistic
of the regression of the squared standardized residual on (respectively) (1) an indicator variable
which takes the value one if the residual is negative and zero otherwise, (2) the product of this
indicator variable and the residual, and (3) the product of the residual and an indicator variable
that takes the value one if the residual is positive and zero otherwise. The AR(1) coefficient is the
slope coefficient in the regression of the fitted deseasonalized variance at time t on the fitted
deseasonalized variance at time t - 1.

Model 1-L Model 2-L Model 3-L Model 4-L Model 5-L Model 6-L
Skewness - 0.484 - 0.491 - 0.400 - 0.416 - 0.337 - 0.436
t-Statistic - 4.249 - 4.306 - 3.506 - 3.654 - 2.954 - 3.825
Kurtosis 1.372 1.409 1.095 0.620 0.441 0.722
t-Statistic 5.995 6.157 4.786 2.710 1.925 3.153
Sign bias 0.555 0.577 0.364 0.507 0.409 0.549
t-Statistic 2.109 2.180 1.446 2.183 1.833 2.311
Negative size bias 4.111 3.794 3.244 3.676 3.575 4.342
t-Statistic 1.004 0.919 0.816 1.018 1.011 1.192
Positive size bias 7.469 8.257 3.054 5.982 3.449 4.910
t-Statistic 1.535 1.693 0.660 1.409 0.861 1.103
Joint test 4.632 5.074 2.139 4.813 2.407 5.408
Significance level 0.201 0.166 0.544 0.186 0.333 0.144
AR(1) Coefficient on
deseasonalized
conditional
variance (std.
error = 0.0464) 0.078 0.089 0.430 0.062 0.335 0.102

large t-statistic, even without deterministic seasonals. However, excess skew-

ness and kurtosis are still a problem in this model. Note that there is no
significant sign bias in this model. In fact, none of the Engle-Ng tests show
any evidence of misspecification in this model.
Model 4-L adds* deterministic seasonals to Model 1-L, using the same
method adopted for the level models. A Wald test that the October and
January seasonal effects are jointly significant (i.e., the test of the null that
A1 = A2 = 0) is estimated to be 7.66. Under the null, this statistic should be
asymptotically distributed as a X22 random variable. Thus, we can reject the
hypothesis that A1 = A2 = 0 at the 5 percent level. Although the amount of
excess kurtosis is much lower for Model 4-L than for any of the previous
models, we can still reject the hypothesis of no excess kurtosis at the 5
percent level. Note also, that Model 4-L appears to have sign bias.
1798 The Journal of Finance

Model 5-L adds both the risk-free interest rate and deterministic seasonals
to Model 1-L. The coefficients on all of those terms are statistically signifi-
cant. The Wald test statistic for the hypothesis that A1 = A2 = 0 is 7.31, while
the test statistic for the hypothesis that A1 = A2 = b2 = 0 is 15.53. Thus, we
can reject both hypotheses at the 5 percent level. Table V shows that we
cannot reject the hypothesis that there is no excess kurtosis in the estimated
residuals from Model 5-L. Model 5-L also shows no signs of sign bias,
negative size bias, or positive size bias.
Since Model 5-L passed more of the diagnostic tests than any other model,
it is our preferred specification. As a further check, we estimated Model 6-L,
by adding v 1 to the conditional mean equation. The coefficient on v1V2 is
not statistically significant, and the results for the conditional variance
equation are qualitatively the same as for Model 5-L. However, the diagnostic
tests show that Model 6-L performs worse than Model 5-L in some important
ways. Model 6-L has a statistically significant amount of excess kurtosis, and
it fails the Sign Bias Test. This suggests there is little evidence of misspecifi-
cation in Model 5-L, and that the model should be the preferred specification.
Note also that the first-order serial correlation of ht in Model 5-L is still
relatively low at 0.3358.8
Given the essentially exploratory nature of our seasonal analysis, it is
important to note that the coefficient estimates for Model 3-L (without
seasonals) and Model 5-L are very close. Thus, general conclusions about the
nature of stochastic volatility are invariant to the inclusion and exclusion of
the January and October seasonals. Notice also that the conclusions derived
from Model 5 (the "levels" model with the risk-free rate and seasonals) and
Model 5-L are the same.
Since the finding of low persistence of conditional variance is so different
from results reported in the literature (except for Campbell and Hentschel
1992), it needs some explanation. At this point, we can only speculate.
Perhaps there are regimes in which variance is relatively persistent, but
there are frequent and relatively unpredictable regime shifts.9 Thus, the data
are characterized by both persistence and random changes in variance. This
explanation is suggested by the fact that the likelihood function of Model 2-L
has two local maxima (we report the global maximum results). The local (not
global) maximum is characterized by variance estimates that are highly
persistent, but produces residuals that exhibit substantial skewness and
kurtosis. It is possible that the two local maxima are merely an artifact of the
relatively small postwar sample. On the other hand, the likelihood function

8
We also tested the robustness of our conclusions in two other ways. First, we estimated
Model 5-L with a sample ending in December 1986 to see whether the October 1987 stock market
crash had an undue influence on our estimates. Second, we estimated Model 5-L using equally
weighted returns. Both sets of results were qualitatively similar to those for Model 5-L.
9 Models of "regime shifting" have been examined in Hamilton and Susmel (1992), and Cai
(1993).
 Volatility of Stock Returns 1799

may be suggesting that there are two ways to fit the data, and the fit with
lower persistence is slightly better.

IV. Conclusion
There is a positive but insignificant relation between the conditional mean
and conditional volatility of the excess return on stocks when the standard
GARCH-M framework is used to model the stochastic volatility of stock
returns. On the other hand, Campbell's Instrumental Variable Model esti-
mates a negative relation between conditional mean and conditional volatil-
ity. In this paper we empirically show that the standard GARCH-M model is
misspecified and alternative specifications provide a reconciliation between
these two results. When the model is modified to allow positive and negative
unanticipated returns to have different impacts on the conditional variance,
we find a negative relation between the conditional mean and the conditional
variance of the excess return on stocks. This relation becomes stronger and
statistically significant when conditional variance is allowed to have deter-
ministic monthly seasonals and to depend on the nominally risk-free interest
rate. Hence our results are consistent with the negative relation between
volatility and expected return reported in Fama and Schwert (1977), Camp-
bell (1987), Breen, Glosten, and Jagannathan (1989), and Harvey (1991). We
show that our conclusions do not change when we use Nelson's EGARCH-M
model modified to include the risk-free rate or seasonals or both.
We also find that the time series properties of monthly excess returns are
substantially different from the reported properties of daily excess returns.
First, persistence of conditional variance in excess returns is quite low in
monthly data while Nelson (1991) finds persistence high in daily data.
Second, positive and negative unexpected returns have vastly different effects
on future conditional variance; the expected impact of a positive unexpected
return is negative. In contrast, Nelson (1991) and Engle and Ng (1993) find
different effects for positive and negative unexpected returns, but both lead to
variance increases.

REFERENCES
Abel, Andrew B., 1988, Stock prices under time-varying dividend risk: An exact solution in an
infinite-horizon general equilibrium model, Journal of Monetary Economics 22, 375-393.
Backus, David K., and Allan Gregory, 1992, Theoretical relations between risk premiums and
conditional variances, Working paper, New York University.
Baillie, Richard T., and Tim Bollerslev, 1989, The message in daily exchange rates: A condi-
tional-variance tale, Journal of Business and Economic Statistics 7, 297-305.
Black, Fischer, 1976, Studies of stock price volatility changes, Proceedings of the 1976 Meetings
of the American Statistical Association, Business and Economics Statistics Section, pp.
177-181.
Bollerslev, Tim, 1986, Generalized autoregressive conditional heteroskedasticity, Journal of
Econometrics 31, 307-327.
, Ray Y. Chou, and Kenneth F. Kroner, 1992, ARCH modeling in finance: A review of the
theory and empirical evidence, Journal of Econometrics 52, 5-59.
1800 The Journal of Finance

Bollerslev, Tim, and Jeffrey M. Wooldridge, 1992, Quasi-maximum likelihood estimation and
inference in dynamic models with time-varying covariances, Econometric Reviews 11,
143-172.
Breen, William, Lawrence R. Glosten, and Ravi Jagannathan, 1989, Economic significance of
predictable variations in stock index returns, Journal of Finance 44, 1177-1189.
Cai, J., 1993, A Markovmodel of switching-regimeARCH, Unpublished manuscript, Northwest-
ern University.
Campbell,John Y., 1987, Stock returns and the term structure, Journal of Financial Economics
18, 373-399.
, 1993, Intertemporal asset pricing without consumption data, American Economic
Review 83, 487-512.
and Ludger Hentschel, 1992, No news is good news: An asymmetric model of changing
volatility in stock returns, Journal of Financial Economics 31, 281-318.
Chan, K. C., G. A. Karolyi, Rene M. Stulz, 1992, Global financial markets and the risk premium
on U.S. equity, Journal of Financial Economics 32, 137-167.
Christie, Andrew A., 1982, The stochastic behavior of common stock variances: Value, leverage,
and interest rate effects, Journal of Financial Economics 10, 407-432.
Cohen, Laurie, 1992, That frightening month is here again, Wall Street Journal, October6.
Engle, Robert F., 1982, Autoregressive conditional heteroscedasticity with estimates of the
variance of United Kingdom inflation, Econometrica 50, 987-1007.
and Victor K. Ng, 1993, Measuring and testing the impact of news on volatility, Journal
of Finance 48, 1749-1778.
Fama, Eugene F., and G. William Schwert, 1977, Asset returns and inflation, Journal of
Financial Economics 5, 115-146.
Fischer, Stanley, 1981, Relative shocks, relative price variability, and inflation, Brookings
Papers on Economic Activity 2, 381-431.
French, Kenneth R., G. William Schwert, and Robert F. Stambaugh, 1987, Expected stock
returns and volatility, Journal of Financial Economics 19, 3-29.
Gennotte, Gerard, and Terry Marsh, 1993, Variations in economic uncertainty and risk premi-
ums on capital assets, European Economic Review (Symposium on Finance issue), Forth-
coming.
Giovannini, Alberto, and Phillippe Jorion, 1989, The time variation of risk and return in the
foreign exchange and stock markets, Journal of Finance 44, 307-325.
Glosten, Lawrence, and Ravi Jagannathan, 1987, Money, real activity and security prices,
Unpublished manuscript, University of Minnesota.
and David Runkle, 1988, On the relationship between the expected value and the
volatility of the nominal excess return on stocks, Research Department WorkingPaper 505,
Federal Reserve Bank of Minneapolis.
Hamilton, J., and R. Susmel, 1992, Autoregressive conditional heteroskedasticity and changes in
regime, Unpublished manuscript, University of California at San Diego.
Harvey, Campbell R., 1989, Time-varying conditional covariances in tests of asset pricing
models, Journal of Financial Economics 24, 289-317.
, 1991, The specification of conditional expectations, Unpublished manuscript, Duke
University.
Keim, Donald B., 1985, Dividend yields and stock returns: Implications of abnormal January
returhs, Journal of Financial Economics 14, 473-489.
Lakonishok, Josef, and Seymour Smidt, 1988, Are seasonal anomalies real: A ninety year
perspective, The Review of Financial Studies 1, 403-425.
Merton, Robert C., 1980, On estimating the expected return on the market: An exploratory
investigation, Journal of Financial Economics 8, 323-361.
Nelson, Daniel B., 1991, Conditional heteroskedasticity in asset returns: A new approach,
Econometrica 59, 347-370.
Newey, Whitney K., 1985, Maximum likelihood specification testing and conditional moment
tests, Econometrica 53, 1047-70.
 Volatility of Stock Returns 1801

and Kenneth D. West, 1987, A simple, positive semi-definite, heteroskedasticity and

autocorrelationconsistent covariance matrix, Econometrica 55, 703-708.
Pagan, Adrian R., and Y. S. Hong, 1991, Nonparametric estimation and the risk premium, in
William Barnett, James Powell, and George Tauchen, eds.: Nonparametric and Semipara-
metric Methods in Econometrics and Statistics (Cambridge University Press, Cambridge),
pp. 51-75.
and Aman Ullah, 1988, The econometricanalysis of models with risk terms, Journal of
Applied Econometrics 3, 87-105.
Singleton, Kenneth J., 1989, Disentangling the effects of noise and aggregate economic distur-
bances on daily stock price volatility, Unpublished manuscript, Stanford University.
Turner, Christopher M., Richard Startz, and Charles R. Nelson, 1989, A Markov model of
heteroskedasticity, risk, and learning in the stock market, Journal of Financial Economics
25, 3-22.