328 LARGE AND SMALL CAP STOCKS IN EUROPEtransmitted from large to small capitalization companies but suggests that
there are differences in the dynamics of stocks prices of firms with different
market value (Conrad, Gultekin and Kaul, 1991). In addition, Ross (1989)
showed that variance changes are directly related to the rate of information
flow. Therefore, one way of studying how information spreads between
large and small firms take place is studying their volatility spillovers.
Most studies that analyse the effect of news on second moments focus
on the US, Australian and Japanese markets. They conclude that volatility
surprisesforlargefirmscanbeusedtopredictvolatilityofsmallfirmsbutnot
vice versa. Conrad, Gultekin and Kaul (1991), find that shocks to large firm
returns are important to the future dynamics of their own volatility as well
as the volatility of small firm returns in the US market. Conversely, shocks
to small firms have no impact on the behavior of the volatility of large firms.
Hendry and Sharma (1999) obtain similar results for the Australian market,
and Kroner and Ng (1998) confirm the conclusion of Conrad, Gultekin and
Kaul (1991) in a more general context.
When the dynamic relationships between volatility of large and small
firms returns are studied, it is necessary to consider asymmetric volatility
and covariance. The first one refers to the empirical evidence according to
which a negative return shock (unexpected drop in the value of the stock)
generates an increase in volatility higher than a positive return shock (unex-
pected increase in the value of the stock) of the same size (for a literature
review on asymmetric volatility see Bekaert and Wu, 2000). Asymmetric
covariance refers to the empirical evidence according to which covariance
between market and stock returns responds more after negative than after
positive market shocks.
In the financial literature, two explanations of the asymmetries in equity
markets have been put forward. The first one is based on the leverage
effect hypothesis. According to this explanation, a drop in the value of the
stock (negative return) increases financial leverage, which makes the stock
riskier and increases its volatility (Black, 1976; Christie, 1982). The second
explanation is known as the volatility feedback hypothesis. This explana-
tion maintains that the asymmetry in volatility responds to the fact that
returns could simply reflect the existence of time-varying risk premiums.
If volatility is priced an anticipated increase in volatility raises the required
return on equity, leading to an immediate stock price decline (Campbell and
Hentschel, 1992; Pindyck, 1984; French, Schewert and Stambaugh, 1987).
This hypothesis relies on two basic tenets. Firstly, volatility is persistent and
secondly, there exists a positive inter-temporal relation between expected
returns and conditional variances.
Consequently, the causality of the asymmetry in equity markets is dif-
ferent. According to the leverage effect, the return shocks lead to changes
in conditional volatility; whereas according to the volatility feedback
hypothesis return shocks are caused by changes in conditional volatility.