Measuring the volatility spill-over effects of crude oil prices on the exchange rate and stock market in Ghana

ABSTRACT

This paper examines shock and volatility spill-over effects from crude oil prices to Ghana’s exchange rate and stock market. We employ multivariate GARCH BEKK and TBEKK models using monthly data from January 1991 to December 2015. We address two main issues. First, whether oil price movements affect Ghana’s exchange rate and stock market. There are very few previous papers that consider the impact of such volatility spill-overs for Ghana and we are the first to do so in a four-variable system of equations. Second, whether any oil price effects depend on the treatment of oil prices as exogenous or endogenous. We are the first to consider specifications that treat crude oil price spill-over effects as exogenous. We find that oil prices have significant spill-over effects on the exchange rate. This result is unaffected by the treatment of oil prices as exogenous or endogenous. However, the relationship between oil prices and Ghana’s stock market depends on whether the oil price is exogenous or endogenous. The implication of these results is that internationally diversified portfolio investors in Ghana should use hedging strategies such as currency forwards, futures, and options to protect their investments from exchange rate risk emanating from oil price shocks.

KEYWORDS:

• Ghana
• exchange rate and stock markets
• oil prices
• exogeneity
• shock and volatility spill-overs
• system GARCH-TBEKK model

JEL CLASSIFICATIONS:

• C32
• F31
• F41

Notes

1 In 1982 the bilateral exchange rate of the Ghanaian currency against the US dollar was $\mathrm{c}/$2.75 per US$1. Ghana agreed to reform its exchange rate policy, to implement a flexible exchange rate regime and devalue the local currency. By 1996 the cedi declined to $\mathrm{c}/$1754 per US$1. The cedi continued to depreciate at an alarming rate for the rest of the 1990s. By December 2000 the cedi suffered its highest annual depreciation, exchanging at $\mathrm{c}/$7047 per US$1, representing a depreciation of 99% from the previous year. In 2007 the government redenominated the currency and a new currency called the Ghana cedi (GH$\mathrm{c}/$) replaced the old currency. The new currency was trading at GH$\mathrm{c}/$0.9704 per US$1 at the time of the redenomination. However, the new Ghana cedi fell steadily against the US dollar and by 2015 the exchange rate was about GH$\mathrm{c}/$3.795 per US$1.

2 Whilst the presence of structural breaks may seem likely due to the long time-span covered by our sample period we note that all our variables are specified as the growth rates that are obtained by applying the first difference to the natural logarithm of the levels data. If there are structural breaks in the form of shifts in the means of the levels data, these will be transformed into outliers after taking first differences. Hence, our use of first differenced data appears to have removed any such structural breaks that may have been present in the levels of the data. We also applied an econometric pre-test for potential structural breaks by estimating mean equation VARs that model structural breaks by including dummy variables that allow the coefficients on all variables to change in three respective periods: 2000m01, 2004m01 and 2008m01. Based upon the Schwarz information criterion (SIC), the VAR model without breaks is preferred to the three VAR models that allow breaks. That is, the system SIC for the VAR without breaks is -12.732 which is lower than the corresponding statistic for the three models allowing parameter change, being: -12.559 (2000m01), -12.593 (2004m01) and -12.612 (2008m01).

3 Shocks are the errors (the difference between actual and fitted values, ϵ_t) and volatilities the (conditional) variances (H_t). The coefficients on the lagged shocks are the ARCH coefficients, whilst the coefficients on the lagged variances/covariances are the GARCH coefficients. The ARCH and GARCH coefficients are used to describe shock spill-over and volatility spill-over respectively (e.g. see Li 2007; Li and Giles 2015; Munsunuru 2014; and Joshi 2011).

4 Because of the standard use of the transpose of A as the pre-multiplying matrix, the coefficients of the BEKK model have the opposite interpretation to usual: A(i, j) is the effect of residual i on variable j, rather than j on i.

5 Where $\mathit{R}=\left[\begin{array}{ccc}{R}_{11}& R_{12}& R_{13}\\ R_{21}& R_{22}& R_{23}\\ \begin{array}{c}{R}_{31}\\ R_{41}\end{array}& \begin{array}{c}{R}_{32}\\ R_{42}\end{array}& \begin{array}{c}{R}_{33}\\ R_{43}\end{array}\end{array}\begin{array}{c}{R}_{14}\\ R_{24}\\ \begin{array}{c}{R}_{34}\\ R_{44}\end{array}\end{array}\right]$ gives the coefficients of the VAR mean equation.

6 As a robustness check (and for comparison) we also estimated a two-variable TBEKK model that omits stock markets from the four-variable TBEKK specification – the unreported results are available from the authors on request. In this model shocks from crude oil prices do not affect the volatility of the exchange rate while oil price volatilities affect the conditional variance of the exchange rate. The latter result is consistent with the four-variable TBEKK model, however, the former result is not. Another difference in results between the two models is that in the two-variable TBEKK model, crude oil prices have no asymmetric effects on the exchange rate, whereas this relationship is significant in the four-variable TBEKK model. Here, we can argue that the interactions of the stock markets in the four-variable TBEKK model may play an important role in the asymmetric response between the crude oil prices and the Ghana cedi exchange rates.

7 The government has often prioritised economic growth at the expense of stabilizing prices and the exchange rate by lowering interest rates. However, given that the monetary authorities have struggled to stabilize the local currency for several years, the government could aim to achieve a balance between sustaining the growth rate and stabilizing the currency following oil price shocks. This is necessary because a persistent fall in the value of the local currency could result in the public’s loss of confidence in the currency which could have significant consequences on the economy.