ANNEX 1B: HEIGHTENED EXCHANGE MARKET PRESSURES IN SUB-SAHARAN AFRICA

Stagnant economic growth coupled with declining oil and commodity prices have underpinned fiscal and external imbalances in countries in Sub-Saharan Africa. At the height of mounting economic pressures, countries in the region may react differently to such a terms-of-trade shock. Some countries may allow a significant but orderly depreciation of their currencies. Others with fixed exchange rate regimes may experience a large and abrupt decrease in foreign reserves. One method to assess the impact of such shocks on foreign exchange (FX) markets across countries in the region is to quantify the extent of exchange market pressure (EMP). Girton and Roper (1977) model EMP as a measure of excess demand for a currency. It was later modeled to compute pressure in FX markets under a flexible, intermediate, or fixed regime, thus enabling us to evaluate the impact of policy actions such as FX market interventions and/or interest rate adjustments.

In its simplest version, EMP is computed as the weighted sum of exchange rate depreciation and foreign reserve loss.6 The analysis exploits monthly data on the exchange rates and international reserves of 40 Sub-Saharan African countries from April to September 2016, compared with the same period in the previous year. Based on the calculated EMP, countries in Sub-Saharan Africa are classified into three groups: high-pressure, moderate-pressure, and no-pressure.7 The high-pressure group consists of countries with EMP greater than 25 percent. The moderate-pressure group is those with positive EMP but less than 25 percent. The no-pressure group is comprised of those with negative EMP, indicating no pressure of currency weakening observed during the period studied.

Table 1B.1 shows the list of countries in each group, with its dominant contributing factor. The overall average EMP of the high-pressure group is 50 percent, more than five times larger than that of the moderate-pressure group. About two-thirds of the pressure in the high-pressure group originates from foreign reserve losses. The moderate-pressure group has an EMP of 9.4 percent, with exchange rate depreciation and foreign reserve loss equally contributing to the cumulative FX pressure. Finally, the no-pressure group has an EMP of-25 percent. Small pressures coming through currency weakening are offset by foreign reserve accumulation among countries in the no-pressure group in the observed period.

When we look at the output-weighted averages of EMP for the different country groups, the results show a different picture. The high-pressure group still has an EMP of 50 percent and the contribution of exchange rate depreciation is now about two-thirds. This outcome is attributed to Nigeria's transition to a more flexible exchange rate regime in June 2016. This transition led to a weakening of the naira of about 50 percent in the following months. Angola is an example of another country that allowed its currency to undergo a significant level of depreciation when comparing April-September 2016 vis-à-vis April-September 2015. However, the Angolan kwanza has remained almost invariant since June 2015.

TABLE 1B.1: Exchange Market Pressure Classification and Contributing Factors

Note: * denotes CFA countries in the Central African Economic and Monetary Community. ** denotes CFA countries in the West African Economic and Monetary Union. EMP = exchange market pressure.

When faced with shocks, flexible exchange rate arrangements act as shock absorbers in some Sub- Saharan African countries. For instance, Mozambique has an EMP of 85 percent; the source of pressure primarily stems from exchange rate depreciation (figure 1B.1). A few countries in the moderate-pressure group, such as South Africa, Tanzania, and Rwanda, endure currency depreciation during the period.

Countries with fixed exchange rate regimes, particularly those in the Central African Economic and Monetary Community (CEMAC) and West African Economic and Monetary Union (WAEMU), had limited room for such adjustments. For example, Chad marks an EMP of 94 percent, most of which comes from loss of international reserves. Although the level of international reserves in Chad was relatively small, the country lost about 95 percent of its total reserves compared with the prior year. Two countries from the WAEMU (Benin and Mali) also endured rapid accumulation of pressure due to loss of international reserves. Although countries participating in the regional monetary unions have limited leeway to manage monetary shocks, the impacts of external shocks are not identical across the board. For example, Cameroon and Togo, both with CFA francs, increased international reserves by 13 and 77 percent, respectively.

The heterogeneity of EMP across countries participating in monetary unions could possibly be attributed to countries' dependence on natural resources, especially oil wealth. As depicted in figure 1B.2, the EMPs of six oil-rich countries in the region are on average 51.8 percent, slightly higher than the overall average of the high-pressure group. All the countries, except Gabon, are in fact in the high-pressure group, although Gabon is barely below the threshold. Finally, most countries in the no-pressure group managed to accumulate international reserves.

_______________________________________________________________________________________________

6 A more complete version of the EMP index includes not only changes in exchange rates and international reserves, but also interest rate differentials; see Aizenman and Hutchison (2012). The analysis undertaken for countries in Sub-Saharan Africa uses only exchange rates and reserves, because: (a) there is insufficient data on domestic and foreign interest rate differentials in many of the countries in the region, and (b) many countries in the region belong to a currency union and have no independent (interest-driven) monetary policy.

7 South Sudan is excluded from the analysis. Its computed EMP is about 1,291 percent.