|
Bond is a fixed interest financial asset issued by governments, companies, banks, public
utilities and other large entities, while stock, also known as equity or a share, is a portion of
the ownership of a corporation. Among broadly defined asset classes, investors like to allocate
their scarce capital resources in stocks and bonds. A common belief among investors is that
the expected return of stock depends on the expected return of bond and additional risk
premium in relation to the bond. Over the years, dynamic economic environment may lead
to changes in risk premium of bond market. Therefore, investors are likely to come in and out
between these two markets until they achieve an equilibrium stage in their investment.
From the standard finance perspective, these two markets are interconnected and they
are sensitive to the changes of interest rate and inflation rate. However, the relationship
between these two markets is ambiguous. As stated by Barsky (1989), the stock and bond
prices may or may not move together depending on the degree of risk-aversion of investors.
In financial market, stocks and bonds are considered to be close substitution for balancing
the portfolio of assets, especially short-term bonds. Short-term bonds and common stocks
are similar in terms of average holding period, liquidity risk and default risk. Therefore, these
two markets are supposed to move simultaneously (Rahman and Mustafa, 1997).
If stocks and bonds depend on each other, dynamic allocation of fund from one market to
another is possible to reap abnormally high return. The importance of relationship between
the yield of stocks and bonds has led to extensive studies on this issue. Most of the studies
examined the relationship between bond and stock markets through correlation, while little focus was put on its cointegration. While correlation test gauges the degree of co-movement
between the bond and stock markets, cointegration test identifies whether there is a tendency
for the two markets to move together with time towards a long-run equilibrium state. From
an econometric perspective, Constant Correlation-Generalized AutoRegressive Conditional
Heteroskedasticity(CC-GARCH) of Bollerslev (1990), the BEKK GARCH of Engle and
Kroner (1995) and the Dynamic Conditional Correlation GARCH (DCC-GARCH) of Engle
and Sheppard (2001) are the prominent contributions that address correlations between
stocks and bonds.
In the international market, research on the relationship between international stock
index and bond index includes that of Solnik et al. (1996) and Lim et al. (1998). Solnik et al.
(1996) tested the correlation between stock and bond markets of the major foreign markets
(Germany, France, the UK, Switzerland and Japan) with the US market. They also included
the volatility in foreign exchange rate between the two countries as an explanatory variable.
They found bond and stock markets to be non-synchronized and concluded that movements
in international stock market correlations did not closely follow movements in the
international bond market correlations or vice versa. However, the result of Lim et al. (1998)
is in contrast to that of Solnik et al. (1996). In the study by Lim et al. (1998), using Engle and
Granger (1987) bivariate cointegration test, the stock and bond indices are found to be
cointegrated to the first order for the first sub-period (November 1988 to May 1991). If the
international bond markets and international stock markets are cointegrated, this implies
that markets are inefficient. This also suggests that a long-run relationship exists between
bond market returns and stock market returns. They concluded that international markets
were more inefficient during the first half than during the second half of the study period.
|
