Real exchange rate in commodity exporting countries
[Thesis]
Lotfi Heravi, M. Mahdi
University of Glasgow
2015
Thesis (Ph.D.)
2015
The aim of this thesis is three-fold. First, in contrast to developed exporting countries such as Australia, New Zealand and Canada, Middle East oil exporting countries are years behind achieving the prerequisites for floating exchange rate and Inflation Targeting monetary regime. On the other hand, their performance under fixed exchange rate (to the US dollar) has brought them some painful experience such as the Dutch Disease and high inflation. For a sample of five of these countries -- Qatar, Oman, Kuwait, Saudi Arabia and the UAE -- we conduct a set of counterfactual experiments. We empirically simulate government consumption expenditure, under a hypothetical peg to a nominal anchor (oil price in either the radical or moderate version) or to a basket (containing the US Dollar, Yen and the Euro) and compare this simulation with whatever exchange rate regime each country actually followed. We find that lower volatility of real oil price in local currency causes lower volatility in government expenditure and fiscal balance as a share of GDP. Hence, we face a less volatile economy. Second, we determine the equilibrium exchange rate (using BEER) of these five oil exporting countries in the Persian Gulf which depend heavily on exports of oil, natural gas and oil products. We employ a new data set for the real effective exchange rate of these countries which is updated annually and covers the period from 1980 to 2011. Given the limited length of the sample (32 years) and low power of individual country by country tests for unit root and cointegration, estimating separate equations for each country (time series) does not provide us with precise results; therefore, to increase the efficiency of the estimators, we employ panel analysis. We apply the pooled mean-group (PMG) of Pesaran et al. (1999) and four more panel estimators for a robustness check. All estimators strongly support the positive effect of real oil price on the real effective exchange rate (i.e. higher real oil price leads to appreciation of the real effective exchange rate) which is consistent with theoretical predictions and with previous studies for commodity (oil) exporting countries. The productivity deferential elasticity is 0.10 which is consistent with the results of the related literature such as the studies of MacDonald and Ricci (2004) for South Africa, and of Lee et al. (2008) for 48 countries over 1980-2004. The BEERs of Qatar, Kuwait and (to some extent) the UAE follow their real effective exchange rates. From 2000, with the increase in oil price, the BEERs appreciate while the real exchange rate of Oman and Saudi Arabia decline; therefore, the Saudi Arabian and Omani currencies get undervalued. Third, employing a new data set of Canadian commodity price indices, we revisit the Canada Bank Equation and introduce a new version with more fundamentals. We present a similar equation for Australia as one of the other developed commodity exporting countries. Using cointegration and the first differences analysis between real exchange rate and fundamentals, we investigate the SVECM and SVAR frameworks to decompose the variance of real exchange rate of Canada and Australia. In the SVECM analysis, the productivity differential and commodity price are the main contributor to the variance of the real exchange rates of Australia and Canada. For the SVAR analysis, we confirm that, as in the literature, demand shock is the dominant force in explaining the variance of real exchange rates of both countries. This result does not change even by adding the commodity price shock to the SVAR framework.