Challenges of Growth and Globalization in the Middle East and North Africa
GCC Countries: From Oil Dependence to Diversification
The Middle East and North Africa in a Changing Oil Market
Creating Employment in the Middle East and North Africa
Financial Development in the Middle East and North Africa
in the Middle East and North Africa
Abdelali Jbili and Vitali Kramarenko
© 2003 International Monetary Fund
Exchange rate regimes in emerging markets have been discussed in a number of Middle Eastern and North African (MENA) countries, especially in the aftermath of the Asian crisis. A number of MENA countries have made progress in liberalizing trade, opening up their financial systems, and adopting market-based monetary policy instruments. Exchange rate regimes in the region currently range from a hard peg (Djibouti) to variants of float (Iran, Egypt, and Yemen), but pegged regimes are predominant (see Table 1). The six members of the Gulf Cooperation Council (GCC; Bahrain, Kuwait, Oman, Qatar, Saudi Arabia, and the United Arab Emirates) have agreed to establish a monetary union by 2010 with a single currency pegged to the U.S. dollar.
This pamphlet reviews the exchange regimes of five emerging market countries in the region—Egypt, Jordan, Lebanon, Morocco, and Tunisia—and an oil-exporting country, Iran, to see whether they need to consider adopting more flexible arrangements as they further open their economies to trade and capital flows. In fact, two countries, Egypt and Iran, have recently adopted flexible exchange rate arrangements, but their experience is too recent to warrant meaningful conclusions. The pamphlet highlights the criteria for an exchange regime choice, presents a country-by-country assessment of existing exchange regimes, explores options for the future based on the forward-looking analysis of reform prospects in each country, raises the issue of a nominal anchor for countries with floating exchange rates, and considers the pros and cons of monetary aggregate and inflation targeting as policy anchors.
The pamphlet looks at the recent move by Egypt and Iran toward more flexible exchange rate regimes, and discusses the required steps to ensure the success of these experiences. For the other countries, the choice is less clear-cut and more of a long-term nature. In either case, transition to a more flexible exchange rate arrangement raises the question of what conditions countries would need to meet if they opted for greater flexibility, including changes to their monetary policy framework.
There has been keen interest in empirical and theoretical studies on the choice of exchange rate regimes in developing countries in the aftermath of the currency crises in Mexico (1994), Southeast Asia (1997), Russia (1998), and Brazil (1999). The debate has focused mainly on the sustainability of conventional fixed pegs—so-called soft pegs (see Box 1)—in the face of capital mobility and other shocks. Immediately after these crises, a broadening consensus evolved in developing economies that were highly integrated into the international trade and financial system around the need for either a float or a move to hard pegs. More recently, the crisis in Argentina has reduced interest in hard peg arrangements. Thus, the prevailing view now is that increased flexibility in exchange rate management would help deal with exogenous shocks, reduce the risk of bank crises, and contribute to financial stability.
The theoretical literature provides broad guidance on exchange rate regime choice. The main criterion for regime choice is to reduce the output cost (in terms of GDP) of an adjustment to exogenous shocks. Thus, the nature and the magnitude of shocks the economy is likely to face, as well as the structural characteristics of its goods, labor, and financial markets, are important considerations in choosing an exchange rate regime.
Economists distinguish between real shocks (emanating from the real side of the economy) and nominal shocks (emanating from the domestic monetary and financial system). In the MENA region, real shocks include changes in terms of trade (difference between the increase in export and import prices), variations in external demand for exports of goods and services (in particular, tourism for many countries), changes in productivity growth relative to trading partners, weather effects on agricultural output, the impact of foreign capital inflows on the supply side of the economy, and repercussions of workers’ remittances from abroad on domestic demand. Nominal shocks mainly originate from instability of money demand, which is reflected in changes in economic agents’ willingness to hold the domestic currency owing to innovations in financial instruments or swings in confidence.
All six countries are subject to real shocks to varying degrees, with
Egypt, Iran, and Morocco being the most vulnerable to terms of trade shocks
(see Table 2). The countries under review are also
vulnerable to regional geopolitical disturbances, which are difficult
to anticipate or measure. In addition, Egypt, Lebanon, Jordan, and, more
recently, Iran are dependent on capital flows, while in the case of Jordan
and Morocco, changes in workers remittances can produce shocks.
Figure 1. Comprehensive Index of Financial Development in the MENA Region: Comparing High, Medium, and Low Development Countries, Scale 0–10
Sources: IMF, International Financial Statistics
and World Economic Outlook; World Bank, World Development Indicators;
and IMF staff Estimates
With fixed exchange rates and flexible wages and prices, an economy can adjust to real and nominal shocks, as well as to any combination thereof, without incurring large output costs. Indeed, as prices and wages adjust to new market clearing levels, there is no need for the exchange rate to move to modify relative prices. Hong Kong SAR is usually quoted as a role model for countries with fixed peg regimes because of its flexible labor and goods markets. However, the sustainability of the fixed peg should not be taken for granted even if an economy has the required degree of flexibility. Only hard pegs, often in the form of currency boards, can withstand real and nominal exogenous shocks when supported by fiscal discipline and little discretion in monetary policy. In contrast to currency boards, some discretion in monetary policy under conventional fixed pegs is still possible under capital controls in the short run, but these pegs are vulnerable to crises, particularly if they are not supported by fiscal discipline.
A high degree of flexibility of prices and wages is rare among developing countries, and MENA countries are no exception to this rule. All six countries covered in this analysis maintain price controls to a varying degree, contributing to stickiness of their prices. Moreover, labor market indicators for Egypt, Morocco, and Tunisia point to the presence of rigidities (Table 2), while anecdotal evidence for Iran and Lebanon also suggests that their labor markets are quite rigid. Jordan seems to have some labor market flexibility, albeit less than Hong Kong SAR. For these reasons, we will focus on the mechanisms of adjustment to shocks under sticky wages and prices for three particular cases of vulnerability to (1) real shocks, (2) nominal shocks, and (3) a combination of both.
Under sticky nominal wages and prices, countries that are subject to real shocks would benefit from exchange rate flexibility. In the face of negative real shocks, a depreciation of the exchange rate would help reduce real wages and ensure expenditure switching from more expensive foreign goods to relatively cheaper domestically produced ones, thereby maintaining employment and output. A credible nominal anchor—for example, a monetary aggregate or an inflation target—as well as the absence of fiscal dominance of monetary policy are key to a successful implementation of a floating exchange regime. Experience has shown that in the absence of a policy framework consistent with price stability, a floating exchange rate can be easily transformed into a “free falling exchange rate,” in particular under a liberalized capital account.
Countries that are subject to nominal shocks would be better off with a fixed exchange rate regime. Nominal shocks to money demand would be absorbed through purchases and sales of foreign exchange by the central bank in the foreign exchange market, which would directly affect the amount of high-powered money in circulation and maintain the fixity of the exchange rate. Under this adjustment mechanism, there is no need for wage or domestic price movements to absorb the shock, other things being equal.
If there is a combination of real and nominal shocks, the more that real shocks predominate over the nominal shocks, the more exchange rate flexibility is needed. In such circumstances, many economists advocate some form of exchange rate management with a credible monetary policy rule. Intermediate regimes are also justified on the grounds that developing countries would have difficulties in adopting policy anchors other than the exchange rate; that excessive exchange rate volatility has a negative impact on investment and growth; and that dollarized economies cannot afford large devaluations because of their negative effects on the liabilities of the corporate sector, households, and the government.
Empirical evidence suggests that developing countries often do not maintain freely floating exchange regimes, even when the regime is officially described as such. Many officially announced floating regimes are de facto intermediate regimes. Based on the IMF de facto classification of exchange regimes, only 20 percent of the IMF member countries have an independently floating exchange regime (see Box 1). This is not surprising, given the fact that most developing countries face both real and nominal shocks. Emerging countries seem to be “floating with a lifejacket.”
What happens when a country makes a suboptimal choice of an exchange regime or follows policies that do not support the selected exchange regime? In the short term, interest rate spreads over major partner country currencies could increase and the central bank could start losing foreign exchange reserves, and in the medium term, additional problems would arise. Slow growth of output and exports, in particular in the case of a substantial appreciation of the real effective exchange rate (REER), usually signals problems with the exchange regime. These problems are often related to the inconsistency between the exchange rate policy and other supporting policies and/or the inability of the economy to adjust to exogenous shocks of a given magnitude. As macroeconomic imbalances worsen, the authorities would need to adjust their policies, or change the exchange rate regime often in the midst of a financial crisis.
Based on the classification presented in Box 1, all countries under review have intermediate exchange regimes. In particular, Jordan, Lebanon, and Morocco have conventional fixed pegs; Tunisia has a crawling peg; and Egypt and Iran have managed floats.
Egypt pegged its currency to the U.S. dollar in 1991, but abandoned its fixed peg in mid-2000, following a deterioration in the external accounts. The latter reflected a combination of capital outflows after the 1997–98 Asian crisis, a decline in tourism receipts in 1998 on account of terrorism concerns, and currency overvaluation due to the strengthening of the U.S. dollar against other currencies in the late 1990s. The authorities initially responded to exchange market pressures through intervention and tighter credit policies, but then slowing economic growth strengthened the case for exchange rate adjustment and the currency was allowed to depreciate from mid-2000. After a period of managed floating, an adjustable currency band was adopted in January 2001, in which currency trades were required to be within a specified band. Over the following two-year period, the band was depreciated on five occasions, and widened twice, resulting in a cumulative depreciation of about 27 percent (in foreign currency terms) from the pegged rate against the U.S. dollar of the 1990s. This was not sufficient, however, to restore external balance, and shortages of foreign exchange in the official market gave rise to a more depreciated parallel market rate. In response to concerns about the inflexibility of the band regime, the exchange rate was floated from end-January 2003. This move was associated with further currency depreciation, increasing the cumulative nominal depreciation to more than 40 percent (in foreign currency terms). As of mid-2003, the exchange rate quoted by banks had not moved to a market-clearing level. With foreign currency still in short supply, the authorities reintroduced surrender requirements of export proceeds and tightened monetary policy, which led to a significant decline in the differential between the rate quoted by exchange dealers and that quoted by banks.
Iran adopted a managed float with a monetary aggregate as the
de facto nominal anchor following the unification in March 2002 of officially
recognized multiple exchange rates. The central bank operates the managed
float primarily by intervening in the foreign exchange market, because
the country has virtually no money markets and the financial rates of
return (equivalent to interest rates) are administratively controlled.
Fluctuations in the rial’s real effective exchange rate (REER) appear
to be driven by oil revenues (calculations of the REER should be interpreted
with caution in the presence of extensive price controls). An Oil Stabilization
Fund was recently established to help the government smooth out the impact
of oil price fluctuations. The choice of a managed float seems appropriate,
given Iran’s vulnerability to terms of trade shocks, labor market
rigidities, ongoing trade reforms, plans for gradual price liberalization,
and prospects of large capital inflows.
Lebanon also has a small open economy with a conventional fixed peg to the U.S. dollar, and its economy is highly dollarized. Although Lebanon achieved rapid disinflation during the 1990s, the economy, buffeted with structural problems including labor market rigidities, has suffered from a loss of competitiveness and has become increasingly vulnerable to the volatility of capital flows and transfers. The recent depreciation of the U.S. dollar has reduced competitiveness pressures somewhat, but the pound is still 38 percent more appreciated in real effective terms than in mid-1990s when the peg was introduced. Moreover, Lebanon’s large structural fiscal deficits, driven by the reconstruction effort following the civil war, have led to a massive buildup of public debt (178 percent of GDP by 2002), which has made the financial system more vulnerable to any significant adjustment in the nominal exchange rate. In the late 1990s to the early 2000s, Lebanon repeatedly had to raise its already high interest rates to defend the peg during periods of sluggish growth. During 2002, intervention in the foreign exchange market was the preferred instrument for defending the pound. More recently, market confidence was bolstered by the second meeting held in Paris in November 2002 to mobilize international financial assistance for Lebanon (Paris II) and ongoing fiscal adjustment measures. As a result, since late 2002, broad money (M3) growth has resumed, dollarization has declined, and interest rates have dropped markedly.
Morocco’s currency, the dirham, which is pegged to a basket of currencies, appreciated in both nominal and real terms over a long period. Moreover, the Moroccan economy is vulnerable to weather conditions and to some extent to terms-of-trade shocks. With capital account restrictions (mainly on outflows by residents) in place, Morocco’s central bank has preserved some monetary policy autonomy. Owing to generally prudent monetary policy, consumer price index inflation rates have converged with those of developed countries. However, the dirham appreciated by about 21 percent in real effective terms during January 1991–March 2001, mainly because of the U.S. dollar’s large weight in the basket and downward rigidity of nominal wages and prices. This, together with Morocco’s growing integration with the European Union, prompted the central bank to adjust the composition of the basket in April 2001 in favor of the euro, which resulted in a relatively small depreciation of the nominal effective rate. Despite the macroeconomic stability achieved by Morocco in recent years, growth has been relatively weak, reflecting the country’s dependence on agriculture and the slow pace of structural reforms. In the absence of productivity gains, the real appreciation of the dirham may have also slowed export growth.
Tunisia targets the real exchange rate, which is akin to a crawling peg regime (see Box 1). Such a policy had mixed results outside the MENA region, but Tunisia has achieved low inflation and rapid growth since the mid-1990s. This is mainly attributable to prudent fiscal and monetary policies, the latter increasingly relying on indirect instruments of monetary control. Tunisia’s exchange rate policy has been facilitated by the absence of major terms of trade shocks and by capital controls for nonresidents. Tourism has become increasingly important, while dependence on agriculture has declined.
Clearly, the exchange rate regimes in the six countries have had varying degrees of success. Exchange regimes in Jordan, Morocco, and Tunisia have not recently come under pressure, as reflected in low real interest rates and stable gross official reserve positions. In these countries, real shocks varied in magnitude but were relatively manageable, and macroeconomic policies were generally consistent with the choice of the exchange rate regime.
In contrast, the recurrent pressures in the foreign exchange market in Lebanon are reflected in high, albeit declining, real interest rates and a reduction in gross official reserves from their peak levels during the past 18 months. This demonstrates that vulnerability to real exogenous shocks (including terms of trade, foreign demand, and volatile capital inflows) and large structural fiscal deficits financed by heavy domestic and foreign borrowing are incompatible with a pegged exchange rate regime combined with downward rigidity of nominal wages and prices.
Egypt and Iran offer two different experiences of a move to a flexible exchange rate regime. Iran initiated the exchange regime change from a position of strength, benefiting from high oil prices, renewed reform efforts, and a few years of experience with a market-determined exchange rate—under the so-called Tehran Stock Exchange (TSE) market operating within a dual exchange rate system. In addition, Iran opted for a managed floating system with most transactions being carried out through the recently established interbank foreign exchange market. Moreover, the Iranian authorities have been able to minimize the pass-through to prices of the depreciation of the official exchange rate, relying on resources previously accumulated in the Oil Stabilization Fund and high oil revenue. Following the exchange rate unification and the move to a managed float, the exchange rate quoted in the interbank market remained relatively stable, which helped bolster business confidence and ensured a smooth transition to the new system.
In contrast, Egypt moved to greater exchange rate flexibility since 2000 under market pressures and against a background of a series of negative exogenous shocks. As a result, the depreciation of the currency has been substantial, even though the official exchange rate has adjusted more slowly than might have been expected given market conditions. While this has contained the inflation pass-through, it has also prevented the exchange rate from reaching a market-clearing level. As a result, credibility of the flexible exchange rate regime has not yet been fully established, and a parallel exchange market continues to operate.
Notwithstanding the exchange regime choice, five of the six countries (Iran is the exception) registered low inflation rates, while real growth averaged about 3–5 percent during 1991–2001 (see Figure 2). The good inflation performance mirrors the worldwide downward trend in inflation during most of the 1990s and the impact of prudent demand-management policies, as well as, to varying degrees, the persistence of price controls.
Figure 2. Selected MENA Countries: Economic Indicators, 1991–2002
Sources: IMF, International Financial Statistics;
World Economic Outlook; and IMF staff estimates.
The picture for growth is somewhat mixed, given the vulnerability of agriculture to weather fluctuations (Morocco), the volatility of oil prices (Iran), and the weak improvements in total factor productivity in all six countries. In addition, international or regional exogenous shocks, including repeated breakdowns of the Middle East peace process and the events of September 11, 2001, have affected growth. The cost of adjusting to these shocks would arguably have been smaller under a flexible exchange rate regime than under a peg, other factors being equal. This conclusion is corroborated by the results of an analysis of pooled data for the six countries during 1996–2001, when all countries had some form of pegged exchange rate. This analysis suggests that periods of steep REER appreciation may have adversely affected growth. This is not surprising, as the fixity of exchange rates or their tight management combined with rigid goods and labor markets in most countries added to the real appreciation, thereby hampering growth. Empirical research also indicates that emerging markets, in general, grow faster with more flexible exchange rates.
While the move to a more flexible exchange rate arrangement by Iran and Egypt seems appropriate, the challenge is to ensure adequate exchange rate flexibility. Both countries need to develop a supporting institutional framework for monetary and fiscal policies. Iran needs to reduce fiscal dominance over monetary policy and develop market-based instruments of liquidity management if it is to keep its foreign exchange market functioning successfully, while Egypt must enhance its fiscal performance and establish a market-clearing exchange rate. Both countries need to strengthen their financial systems through structural reforms.
For the other countries, however, the need for greater exchange rate flexibility may arise in the medium or long term, as the economies become more diversified and more integrated into the world economy. Lebanon could maintain its conventional fixed peg in the near term while implementing a strong fiscal adjustment complemented with structural reforms aimed at reducing the public debt and enhancing competitiveness. The authorities’ economic strategy for 2003 and the medium term aims at reducing the government debt and its servicing cost, and accelerating growth. The strategy includes further fiscal adjustment, large-scale privatization, burden sharing with the banking system, and external financial assistance on concessional terms. The government intends to use the counterpart of the privatization and Paris II–related inflows to retire expensive short-term debt. Moreover, the authorities have agreed with commercial banks on a scheme to reduce interest payments during the next two years, which has already started to be implemented. In the medium term, however, it is not clear whether fiscal adjustment can be sustained for long without affecting growth, or whether structural reforms will be deep enough to restore competitiveness under the conventional fixed peg. While the high dollarization of the economy precludes exchange rate adjustments in the near term, some exchange rate flexibility in connection with efforts to strengthen the financial system and public finances over the medium term should not be ruled out.
For Jordan and Morocco, the choice of an exchange regime must take into account the increase in potentially volatile capital flows that could follow further trade and capital account liberalization and other reforms, and the difficulties their fledgling financial systems may have in intermediating large capital inflows. Indeed, it would become increasingly difficult for the central banks of these countries to maintain conventional fixed pegs under more liberalized capital accounts. Both countries also need to consider other real exogenous shocks to which their economies are exposed, such as terms of trade and security issues. The experience of other emerging market economies illustrates that more flexible exchange rate arrangements may be better suited to help these countries adjust to increased capital inflows and exogenous shocks. Conventional fixed pegs or narrow bands could have heavy economic costs, in particular if the rigidity of labor markets and nominal wages (Morocco) and the relatively weak fiscal positions are not addressed in a timely manner.
Tunisia would benefit from moving away from real exchange rate targeting toward more flexible exchange rate management. Although Tunisia is not subject to large exogenous shocks for now, trade and capital account liberalization could amplify the size of shocks and warrant greater exchange rate flexibility and more independent monetary policy.
With regard to how much flexibility is desirable and the appropriate timing of transition, there is no one-size-fits-all approach. The speed and sequencing of improvements in financial systems and progress in fiscal adjustment, as well as the importance of real shocks, would be major factors. The scope for exchange rate flexibility may range from loosely managed floats to less flexible intermediate regimes. More important, perhaps, is the need for these countries to initiate the transition from a position of relative stability without waiting for a crisis to occur.
Countries opting for greater exchange rate flexibility must choose a nominal anchor for their monetary policy to stabilize inflation expectations. Possible options include targeting the growth rate of a monetary aggregate or targeting inflation. Targeting a monetary aggregate such as reserve money or broad money is perhaps more familiar to policymakers in developing countries and has less stringent institutional and policy requirements. However, it is based on the assumption that the relationship between money supply and inflation is stable, which may not always be the case.
As stated above, the income velocities of money in the six countries show a stable declining trend since 1991, except in Iran and Jordan. But money demand could become less stable in the other countries following reform of the financial sector—including the introduction of new financial instruments—and further opening up of the capital account. Moreover, unsustainable public debt, in particular in countries with a high level of dollarization, could undermine confidence, making money demand more unstable. These considerations need to be taken into account in the design of a monetary-aggregate-targeting framework. Egypt, Iran, and Tunisia have already started to target monetary aggregates.
The second option for the conduct of monetary policy under flexible exchange rate arrangements is inflation targeting. Under this framework, the central bank commits to meeting a certain inflation target in a given period and uses appropriate instruments to achieve it. It also commits to subordinating other policy objectives to the inflation target and to operating with a high degree of accountability. The central bank adjusts short-term interest rates that it charges on its facilities in response to developments in expected inflation, the output gap (usually defined as the percentage deviation of actual GDP from “potential” GDP), and some measure of the exchange rate.
Although the shift away from a pegged exchange regime is one of the rationales for introducing inflation targeting, initially the exchange rate continues to be an important element for the interest rate–setting rule of the central bank under inflation targeting in open economies. This is because the exchange rate directly affects the prices of tradable goods (exports and imports) and output through changes in the relative prices of tradable and nontradable goods (usually services). Also, the exchange rate can continue to play a role in forming inflation expectations. With improvements in credibility and a concomitant reduction in the pass-through effects of exchange rate fluctuations on internal prices, the role of the exchange rate in the interest rate–setting rule of the central bank would decline.
The prerequisites for the successful operation of an inflation-targeting framework are quite challenging for emerging markets. These include:
A number of emerging market countries, however, have moved to inflation targeting gradually, without meeting all the preconditions first. In South Africa there was a transition to inflation targeting from a monetary-aggregate-targeting framework in which inflation was an informal objective. In Chile and Poland, the transition took place through a gradual shift from targeting exchange rate bands and inflation toward focusing on inflation. Moreover, as many as 19 developing countries floated their exchange rates in 2002 and announced their inflation targets without a formal inflation-targeting framework. Economic research on transition to inflation targeting does not provide conclusive arguments on whether transitional regimes are beneficial, but it states clearly that transparency in monetary policy operations and a commitment to adoption of an inflation-targeting framework in the medium term would strengthen credibility and facilitate the transition to a single numerical objective for inflation.
None of the six countries under review meets all the requirements for inflation targeting. Low inflation was achieved in all countries, except Iran, but there are more problems with the other preconditions (see Table 3). This means that the transition period to inflation targeting could be relatively long. The floaters—Egypt and Iran—could gradually change the emphasis of their policies from monetary aggregates to inflation targeting, within a clearly defined timetable for supporting reforms.
Most of the countries still have to gain higher policy credibility. Sovereign
ratings by major credit rating agencies represent a good indicator of
market perceptions of the credibility of countries’ economic policies.
Only Tunisia has received a rating by Standard and Poors or Fitch equivalent
to the average sovereign rating of the emerging market countries that
The independence of the central bank in Jordan is enshrined in the law and asserted in practice, while the central banks of Lebanon, Morocco, and Tunisia seem to have a high degree of independence in the use of monetary policy instruments. Egypt and Iran seem to have little central bank independence, though steps toward greater independence are being considered in Egypt under a new banking law currently before parliament. Development of market-based indirect instruments of monetary policy and financial sector reforms are also of paramount importance for the proper operation of transmission mechanisms of monetary policy. Indirect instruments of monetary policy are well developed in Jordan, Lebanon, Morocco, and Tunisia, and are also in regular use in Egypt. Iran has limited and rudimentary indirect instruments with administrative controls on rates of return (interest rates) and centralized allocation of credit.
Banks are predominantly privately owned in Jordan, Lebanon, Morocco, and Tunisia, while public sector control is relatively high in Egypt and Iran. Moreover, in most countries, money markets are shallow and would need to gain depth and sophistication along with the development of indirect instruments of monetary policy. Finally, all six central banks need to enhance their analytical capabilities, particularly developing adequate tools to monitor price movements and establishing clear transmission mechanisms of monetary policy.