A4liddle East Pefor n d cE1 Ct>sive~ andlc 6vtowtl inV the- J\ov±k Africa A\Aiddle-east Wkcd l AF1as ta ppeAed Workin0 Paper iVA tteA 1 990s? Serpies NoJ0. 25 3ujy 2002 Tke WoAdl Bk by P-OdACed by ke- D i pCIk Das Apta, Office of tke* >1ikef Ccovnoznv st r-kie-f~~ GC1miteVI VA f e-v e IIe i t a ncl4 Fa;ies c Arabic avwcl F=rencl Reform and Elusive Growth in the Middle-East --- What Has Happened in the 1990s? By Dipak Dasgupta, Jennifer Keller and T.G. Srinivasan Social and Economic Development Group Middle East and North Africa Region July 2002 Discussion papers are not ftiinial: i aons of the World Bank. They irepresent preliminary and often unpolished results of country analysis andl eOearch C-@rculation is intended to encourage discussion and comments; citation and the use of the paper should take-account of its probvisional character. The findings and conclusions of the paper are entirely those of the author and should not be attributed to the World Bank, its affiliated organizations, or to members of its Board of Executive Directors or the countries they represent. Table of Contents Reform and Elusive Growth in the Middle-East -What Has Happened in the 1990s? Summary A. Introduction ................................... B. Oil Prices and Growth in the MNA Region ..................................3 C. Measuring Reform Effort in MNA ...............................6 Framework ..............................6 * Progress of Structural Reform in MNA ..............................8 * Positive Signs ............................... 14 D. Sources of Growth in MNA .................... 16 * What characteristics define strong performance in productivity growth? .................. 23 E. Conclusion .................................................................. 24 Charts Chart 1: Real Oil Price (1990 $/bbl), 1976-2000 .4 Chart 2: Real Oil Price and MNA Growth, 1976-99 .5 Chart 3: Real Oil Price and GCC Growth, 1976-99 .5 Chart 4: The Correlation of Oil Prices and GDP growth in non-GCC Oil Producers, 1976-99 (Algeria, Iran) .6 Chart 5: Oil Prices and Non-GCC Diversified Economist Growth, 1976-99 .6 Chart 6: Composite Structural Reform Index in Developing Regions ..................................................9 Chart 7: Comparison Trade Policy Indicator Among Developing Regions ..........................................9 Chart 8: Highest Marginal Tax Rate on Individuals Among Developing Regions ............................ 10 Chart 9: Highest Marginal Corporate Rate on Individuals Among Developing Regions ................... 10 Chart 10: Privatization Proceeds Among Developing Regions ..11................................... 1 Chart 11: Real Exchange Rate Over-Valuation in Developing Regions . .11............................ 1 Chart 12: Change in Composite Reform Index .................................................................. 12 Chart 13: Pace of Reforms and Per-capita GDP Growth .................................................................. 14 Chart 14: Composite Index of Economic Stabilization by Regions . .................................................... 15 Chart 15: Credit Rating Improves with Structural Reforms for MNA . ............................................... 16 Tables Table 1: Pace of Structural Reforms in Developing Countries ............................................................ 13 Table 2: GDP per capita growth over the 1980s and 1990s, MNA Region ......................................... 17 Table 3: Regional TFP Growth, by decade ..................................................................1 8 Table 4: GDP per capita growth and growth of accumulation and productivity by region 1960-1990 .................................................................. 20 Table 5. Change in MENA's growth and TFP growth between 1980s and 1990s (%) ....................... 21 Bibliography .................................................................. 25 Annex 1. Data sources for Construction of Structural Reform Index ................................................. 27 Annex 2. Measuring Factor Accumulation and TFP growth contributions to Aggregate Growth ...... 28 Annex 3. Emerging economies in terms of improved productivity growth Performance, 1960-1999.30 dIjJ1I -AI .fl Lhul43t j.Cy1'L A ; _ , A1 ws UA s9S wJ*1 JtAJy JJ't zjy1 4Jn9 cj~ oRA .ma @ zAU > , ~Aj l aU;.L oljl ZY A PJ 411 ~ ayl zaI jA - *LS alJj iI S y :A.l,sa l tl.ill jsil1 sil ;! l t=L l Ysall 1,1J ) ~~~~~-4OAv Y- 241-1 j4 Li" o^ll lelil SI*Wj i?>ip; JI c-Ul , > LL 6UL ej sj; *JUA1~ 4.12uJ AcL. A S, 49 WU _L= Lay Ly_*1 Ie ^ 1cS)sJ iil DJ 11D . ¾yaL 41~ WŽiL.4 ^ USJ - N . il 't^ Q fl~Iut i sI; .rl @ 9 'U4Jj v 1 L Us UZl ; 9 101 D1 Sjl I A~~~~~L -- . . W A ylUj 11 ,91aL 19 fLJ @ =ffi _ . ,,l -^~~~~~~~~~~~~~~~~~~~~_A Y.3L)+. SII .6L!tU l .R ,}. Y ll1W5jJ &1*SLL Risum6 Pour les decideurs, il est tres important de realiser une croissance economique plus rapide. L'un des postulats fondamentaux de la reforme est qu'elle devrait favoriser cette croissance plus rapide. En raison de la baisse importante des cours petroliers internationaux au milieu des annees 80, la majorite des pays de la Region MENA ont connu un ralentissement marque de la croissance et/ou des crises macro-economiques. Cette situation a impose une reforme relativement etendue et des efforts de restructuration (mais A des degres divers) pour ramener la stabilite macro-economique et diversifier les economies loin des produits petroliers et visant une croissance induite par le secteur prive. Et pourtant, pour de nombreux observateurs dans la region et en dehors de la region, les resultats en termes de croissance acceleree dans la region semblent evasifs. Le document examine la relation reelle, si elle existe, entre la reforme et la croissance dans la Region Moyen-Orient et Afrique du Nord (MENA), en particulier au cours des annees 90, en la comparant A la performance d'autres regions et pays en developpement. Les resultats de la croissance des pays MENA examines dans ce document donnent a entendre que les reformes dans cette Region ressemblent fort A une tache A moitie termin6e. II est clair que les reformes ont commence, mais la majorite des pays de la Region MENA sont des reformateurs lents et qui continuent A etre en retard par rapport aux resultats d'autres pays et d'autres regions. L'effet des reformes est difficile A distinguer mais il peut au moins avoir contribue a inverser les pertes importantes de productivite de facteur des annees 70 et 80. Mais la reponse de l'investissement prive se fait attendre et il est prioritaire d'en comprendre les raisons - et il faudra que les reformes requises aillent au-dela de la premiere etape des reformes structurelles dans ces pays. Enfin, des ameliorations soutenus des resultats de la productivite seront critiques et semblent etre lies, A premiere vue, a la reussite d'une diversification rapide de l'activite economique sur les annees, des echanges de biens et de services et A l'eloignement de la dependance excessive des ressources et des produits comme le petrole (qui donne lieu a une volatilite elevee et une faible productivite). Abstract Achieving faster economic growth is of pressing importance to policy-makers. One of the fundamental premises of reform is that it would be expected to foster such faster growth. Since the large fall in international oil prices in the mid-80s, most MNA countries experienced a marked growth slowdown and/or macroeconomic crises. This forced fairly widespread economic reform and restructuring efforts (albeit in varying degrees) to bring back macroeconomic stability and to diversify the economies away from oil and from a public sector to a private sector led growth. Yet, to many observers within and outside the region, the pay-offs in terms of accelerated growth in the region are seen to have been elusive. The paper explores the actual relationship, if any, between reform and growth in the Middle-East and North Africa (NNA) region, especially in the 1990s, benchmarking against the performance of other developing regions and countries. The growth performance of MNA countries reviewed in this paper suggests that reforms in MNA remain very much a half-finished business. Reforms have clearly begun but most NMNA countries are late reformers and continue to lag other countries and regions' performance. The effect of reforms is difficult to distinguish but may have at least contributed to reverse the large negative total factor productivity losses of the 1970s and 1980s. But the private investment response has been lagging and understanding the reasons are a priority---and the needed reforms may need to go well beyond the first stage of structural reforms in these countries. Finally, sustained improvement in productivity performance will be critical and appears to be tied, at least at first glance, to success in rapid diversification in economic activity over time, in traded goods and services and away from excessive dependence on resources and commodities such as oil (which induce high volatility and low productivity performance). I A. Introduction Achieving faster economic growth is of pressing importance to policy-makers. One of the fundamental premises of reform is that it would be expected to foster such faster growth. The "Washington consensus" of the early 1990s had advocated a certain core set of reforms as essential in a rapidly globalizing world---minimum standards of macroeconomic stability, and structural reforms in openness and the investment environment. But questions have started to be asked by the late 1990s whether these were having their intended effects. Despite apparent reforms---albeit not to the same extent in all countries and areas---actual growth performance in many developing countries and regions has often been disappointing. MNA countries are no different in their experience with reform and growth in the 1 990s. Reform and its link to economic growth is also politically and socially central issue. It is especially critical in the MNA region because of the unemployment problem. At least some part of the critical pressure on governments (other than the decline in oil prices and macroeconomic crises) arises from the unemployment situation as the labor force grew by about 2.8 percent a year, about the highest rate in the world while job opportunities have been lagging. While rapid labor force growth can be a "demographic gift" in the right circumstances, it can also create enormous social pressures if actual job opportunities for this are not generated. Formal unemployment rates in the region (outside of the Gulf Cooperation Council or GCC countries), now average close to 25% - - the second highest in the world2. In some countries - Algeria, Iran, Syria, Libya, Yemen -- as much as one-third of the labor force is unemployed. Even the GCC economies, with comparatively low unemployment rates averaging about 5%, have begun to experience growing unemployment among their national populations. The absence of growth and rapidly rising unemployment also coincided with falling real wages in the region---as is to be expected. Consequently, the crisis of growth in the MNA region since the 1980s has translated into an "unemployment crisis", which could spill over into a larger poverty crisis (which the region has so far managed to avoid in most places because of the presence of formal and informal and implicit or explicit social safety nets). The biggest impact of unemployment is also often on young first-time job-seekers, and a potentially socially explosive situation. Unemployment rates for those under 25 are about twice as high as national averages, as in Algeria and Tunisia, and_as high as two and a half times to three higher in Lebanon and Iran. This same age group is also better educated. Rising education and unemployment are potentially powerful pressures for change and reforms. In the above context, the main question that is explored in this paper is to understand the actual relationship, if any, between reform and growth in the Middle-East and North Africa (MNA) region, especially in the 1990s, benchmarked against the performance of other developing regions and countries. Since the large fall in international oil prices in the mid-80s, most MNA countries experienced a marked growth slowdown and/or macroeconomic crises. This forced fairly widespread economic reform and restructuring efforts (albeit in varying degrees) to bring back macroeconomic 2Unemployment in Sub-Saharan Africa is estimated to be about 25%, but definitional differences make comparisons of unemployment rates across countries difficult. stability and to diversify the economies away from oil and from a public sector to a private sector led growth. Yet, to many observers within and outside the region, the pay-offs in terms of accelerated growth in the region are seen to have been elusive. Regional GDP growth averaged 3.2% a year over the l990s3, higher than in the 1980s (2.7% a year), but not the rebound one might have wanted or expected following a decade of stagnation. On a per capita basis, economic growth averaged less than 1% a year. Outside the Gulf economies, growth has been somewhat healthier, averaging 3.5% per year, but remains weak on a per capita basis. Were the reforms inadequate? Were they directed at the wrong areas? Or did the reforms miss out some other vital ingredients---such as governance, market institutions, end of conflict and favorable geography---for faster growth to happen? This paper attempts to provide a preliminary first evaluation of these broad questions. The structure and principal conclusions of this paper are the following: As a first step in understanding factors affecting growth outcomes, section B examines the relationship between oil prices and aggregate growth. A trend decline (and high volatility) in international oil prices, which persisted throughout the 1990s, was a major exogenous factor dampening growth outcomes in all MNA economies. Countries were forced to adjust to a sizeable real income decline (mainly by reduced public spending). While the negative income effects dominate, some positive effects of substitution to non-oil activities might also however be expected with lower oil prices (and reduced Dutch-disease). In the major oil-producing countries (GCC, hran, Algeria, Syria), the negative income effects clearly dominated aggregate growth outcomes, as might be expected both because of the very large weight of oil and slower reform. In the more diversified and earlier reform economies (Egypt, Jordan, Morocco), however, there was evidence to suggest a distinct break in the relationship of aggregate growth with oil prices and therefore relative success in diversification (the aim of the reform efforts). Section C then measures actual reform effort and its intensity in the diversified MNA economies. A variety of indicators are used that attempt to be closer to the choice of policies and instruments (rather than outcomes that are often wrongly used to measure policy effort). The reform effort is also benchmarked against a large number of countries in all other regions. The conclusions are nuanced. MNA countries have indeed undertaken sizeable reform effort, but the main area of progress is in macroeconomic stability (expenditure reduction, lower inflation) rather than in core structural reform. In structural reform (trade policy, private investment environment), MNA countries are a late-comer, with significant reform only happening from the mid-1990s onwards. At the same time, other competitors and comparators globally have been moving faster and/or catching-up, especially in middle-income Latin America, Europe and East Asia. All of the countries we would classify as diversified economies (Morocco, Egypt, Tunisia and Jordan) are rated in the paper as slow or gradual reformers. Nevertheless, institutional investor surveys are responding and beginning to positively re-evaluate reform intensity and effort in MNA countries. 3 Weighted (by population) growth and per capita growth rates for Algeria, Egypt, Iran, Jordan, Morocco, Syria, Tunisia, Yemen, Bahrain, Oman, and Saudi Arabia. - 2 - Section D then turns to a detailed discussion of the sources of growth in the MNA region, from an accounting perspective, and focuses on both trends in factor accumulation and total factor productivity, and their qualitative relationship to the extent of reform undertaken. Four main conclusions emerge. First, while reform has had a growth pay-off for some countries, for the region as a whole, per capita GDP growth remains anemic, increasing by an average of only 0.3 percentage points per year. Second, this growth stagnation occurs despite a significant increase in virtually all of the MENA region in total factor productivity growth. For the region as a whole, total factor productivity growth increased by an average of 1.3 percentage points per year, from an average of-1.3 % per year in the 1980s (linked to massive and inefficient public and private investments of the oil-boom years) to 0 percent per year in the 1990s. Structural reform has probably contributed to reversing this negative factor productivity in the region. In 8 out of 10 countries in the region, TFP performance improved. Third, the principal reason that growth has remained stagnant is the collapse in factor accumulation, especially physical capital, within most of MNA. In only two countries in the region has factor accumulation increased from the 1980s to 1990s. Public investment has certainly fallen, much as to be expected with the fiscal adjustment. But equally, private investment has not taken up the slack nor responded aggressively to the reform. In the 1990s, investment has declined dramatically, and without exception across the region. Fourth, preliminary evidence using cross-country regression panel data including the MNA countries seem to suggest that the macroeconomic reform effort has had the greater positive (and significant) association with the improvement in aggregate growth and in total factor productivity. The structural reform index shows no significant or discernible impact so far. However, this work needs to be extended and examined much more carefully, before any definite conclusions can be drawn. Finally, countries which rank high on an index of sustained total factor productivity growth appears to be those that are successful in rapid diversification, openness, and macroeconomic stability, while countries at the bottom tend to be those which are relatively heavily reliant on resources and relatively undiversified and subject to larger macroeconomic volatility For structural reform to have an effect, the private investment response and its redirection to the traded goods sector is essential. For that to happen, questions remain as to some other critical factors that we do not understand well-factors affecting the critical private investment response, the export diversification response, and geography (regional conflict and peace) factor---as to why faster growth continues to elude some most countries in MNA. This should now form the main agenda of research questions on the growth question in the MNA region, as indeed in most other middle-income countries around the world facing slow growth despite reform. B. Oil Prices and Growth in the MNA Region Oil has an usual degree of importance in the MNA countries. Three different typologies can be constructed. The first is the core set of OPEC oil producers, mostly the GCC countries4, whose economies are dominated by oil and whose objective function is arguably to maximize long-run oil prices and rents in global energy markets (including 4 Bahrain, Kuwait, Oman, Qatar, Saudi Arabia, and UAE. -3 - its relative stability). The second are larger and more populous OPEC oil exporters (Iran, Algeria, Syria, Iraq), whose economies are also dominated by oil, but whose objective functions are more diverse (in raising and diversifying their economies away from oil). The third is the set of diversified economies, whose oil exporting sectors are either relatively small or non-existent, but which are still relatively reliant on oil markets either directly as exporters or indirectly for regional export markets for traded goods or labor (workers remittances) and capital inflows from richer regional oil exporting countries-- Egypt, Morocco, Jordan, Yemen, and Lebanon. Real oil prices (1990 dollars per barrel), as shown below in Chart 1, collapsed after 1985 and reached a low in the late 1980s, before recovering modestly in the aftermath of the 1990 Gulf War. Following that, oil prices again fell steadily during much of the 1990s, reaching a low in 1999 (of less than $10 a barrel). Only since the closing months of 1999 have oil prices recovered sharply, to close to $30 per barrel. Much of the period under examination has therefore been one of a dramatic fall in oil prices in real terms. Chart 1. Real Oil Price (1990 $Ibbi), 1976-2000 50.0 45 0 40.0 35.0 30.0 25.0 20.0 15.0. 10.0- 50 co £0 0 N t SD 0 0 N 1- 1- £0 0 £0 £0 0 0 e £0 0 0 am c0 0 0 0 0 0 0 0 0 £ |Real COil Price| Source: World Bank. What has been the relationship of such trends in oil prices with regional GDP growth, and as differentiated by the three different typologies of countries described earlier? For the region as a whole, the relationship is mixed, as shown in below Chart 2. More precisely, during the 1970s till 1988 or a period of heavily oil-cartelized world, there is evidently a counter-cyclical relationship---rising oil prices matched by falling GDP growth and vice-versa. But since late 1980s, that relationship broke down and we see a shift to a close positive relationship. But how far is this aggregate picture different between the three types of MNA countries? Three conclusions are possible. First, GCC oil producers continue to show a counter-cyclical GDP growth relationship with oil prices between 1976-88---as would be expected as longer-term cartel producers restrict oil supplies (lowering GDP growth) to raise prices and vice-versa---but when the cartel is pressured beyond a point, they reverse, as evident since 1988 (Chart 3). Second, for non- GCC oil exporters, there is a clear close positive relationship between oil prices and GDP growth throughout, as higher oil prices permit faster growth and vice-versa (and some free-riding on cartel behavior, Chart 4). Third, for the diversified economies, falling oil prices caused collapsing GDP growth till the late 1980s, but since then, the trend is a -4 - promising disassociation between oil prices with GDP growth trends picking up mildly despite continuing trend decline in oil prices (Chart 5). Chart 2. Real Oil Price and MNA Growth, 1976-99 50 -5 t40 10 a30 20~~~~~~~~~~~~~~~ *& 10 a> o0 0 0 0 o- 0 0 a5 0 0 0 0 0 0 0 0 0) 0 0 0) 0) - Real Oil rice - hiA GDPGrowth Source: Staff estimates. Chart 3. Real Oil Price and GCC Growth, 1976-99 50.0 20.0 <°n>830°0 - 15.0 D 40.0 ._ 10.0 I l l l l l l l l l l l l l l l l l l l l l l -10.0 30.0 5.0 2 o t 20.0 0.0 -5.0 - 0.0 -10.0 *6_________________________ N-r0)0)0W W X 0)0 0coo)0))00CD 0 | Real 0 Price - GDPGrowth Source: Staff estimates. -5- Chart 4. The Correlation of Oil Prices and GDP growth in non-GCC Oil Producers, 1976-99 (Algeria, Iran) 50.0 - 20.0 ; 40.0 15.0 10.0 a 0130.0 5.0 0.0 -10.0 10. -5.0 NXtu o wo ) Ds00 LD CX -a)a)O~~NCLO C (D- Co C) a)00 0 0 0ax amaxala) a) a) a) a a) C) a) cn o> a) a) am a)a) a) a a)a)o a) ox a) a) o a) a) 0) a al a) a)a)O - Real Oil Price + GDP Growth| Source: Staff estimates. Chart 5. Oil Prices and Non-GCC Diversified Economist Growth, 1976-99 50.0 14.0 .0~ ~ ~~~~~~~~0 40.0 1. 30.0 0.0 (20.0 6.0 4.0 0 0.~~~~~~~~~~~~~~~~~~~~~~~~~~~0 =10.0 0 ~~~~~~~~~~~~~~~~2.0 ~ t-t'Dto D 0 D 0 00 CD C0 0 0 am )a)a) a) as a) a) aW 0 ) a) a) a) ) a) a) a a oa a) a) a) a) a) a) ) a) a) a) a) a) - Real Oil Price - GDP Growth - Poly. (GDP Growth) Source: Staff estimates. C. Measuring Reform Effort in MNA Framework Nearly after two decades after the start of structural reform programs supported by international financial institutions and donor community, consensus has eluded reaching firm conclusions on the effect of these programs on growth. A critical step in such evaluation exercises is the measurement of structural reforms efforts. There have been broadly three approaches to measuring reforms. First, the early studies distinguished only between presence or absence of structural reform programs and proceeded to evaluate their effect on growth in a "with" or "without" comparisons. Often, - 6 - the researchers used the number of loans or scale of adjustment lending as indicators of the intensity of structural reform efforts (see McGillivary (1999) for a short-survey of empirical methodologies for this approach). Even when used in an econometric regression to separate out the influences of external factors, these measures fail to adequately reflect the varying reform intensities in different countries resulting in possibly differentiated payoffs. Second, outcome measures under different components of reform programs such as outward-orientation came to be used because these datasets are easier to assemble (Easterly and others (1997) for example). However, the outcome measures fails to distinguish the between reform efforts by the government and the response of economic agents to these reforms. The response of the agents to reform inputs in itself is a legitimate object of study and therefore a failure to distinguish between reform input and outcomes is a serious shortcoming. The third and last approach seeks to directly measure structural reform policy inputs and this is the one the current study also adapts. The earliest example of direct measures of policy reform efforts was in Agarwala (1983), which focused on price distortions in 31 developing countries by means of 7 indicators in three areas of foreign exchange pricing, factor pricing and product pricing. More recent research studies following this approach are listed with indicators used in Srinivasan (2001). Identifying what constitutes structural reform is important for putting together a set of policy indicators that are monitorable. In broad terms, structural reforms can be defined as the set of policies that increase an economy's market orientation. Structural reforms are a large part of policies that came to be termed as the "Washington Consensus" by Willamson (1990). Williamson provided a convenient overview of 10 policy areas for government action that formed this consensus which can be grouped in his own words under "macroeconomic prudence, outward orientation and domestic liberalization": Fiscal discipline Public expenditure quality Tax Reform Interest Rate liberalization. Competitive real exchange rate Freer Trade Fostering FDI Privatization Deregulation Property rights Leaving aside macroeconomic prudence as the subject matter identical with macroeconomic stabilization, we are left with outward orientation and domestic liberalization as the two key components of structural reform. Research on construction of structural reform indices have implicitly followed the Washington Consensus - 7 - typology, but emphasized in greater detail particular sub-components and different weights to derive a composite indicator based on the developing country region under investigation. For example, Lora (1977) and Morley et al (1999) in their work on Latin America emphasize financial indicators, Dicks-Mireaux and others (1997) and Bonaglia and others (2000) in their work on SSA, or, Martha de Melo and others (1996) on transition countries, stress on public enterprise reform, price controls and state intervention in marketing, and on European transition countries). For the current study in its initial phase, we have identified 4 components (Trade Policy, Tax Policy, Real Exchange Rate Over-valuation, and, Privatization)---see Srinivasan (2001) for more details. We opted for a parsimonious set of components because we wanted to quickly assemble indicators for as many countries as possible with a view to compare NINA countries with all other developing countries. Progress of Structural Reform in MNA In the current phase of structural reform initiated since the mid-1980s, MNA reforms were the last to start among the four regions of MNA, East Asia and Pacific (EAP), Latin America and Caribbean (LAC) and South Asia (SAR)5. MNA reform index did not show a strong pick up until 1994 compared to LAC which had a steady pace from 1985, EAP reforms that accelerated since 1990, and South Asia that dramatically accelerated reforms since 1990. Also, the pace of reforms have been slower in MNA than other regions. As a result, the regions ahead of MNA in 1985 (East Asia and Latin America) have raced further ahead as shown in Chart 6. Those who were behind MNA in 1985 (South Asia and possibly Europe and Central Asia) have closed in. At a time when global competition is becoming intense, without accelerated reforms to improve efficiency of domestic production, MNA producers are likely to feel disadvantaged in global markets. Considerable scope for further reforms exist for MNA countries. How representative is our structural reform index for MNA? The MNA index used in the discussion that follows is the average of four reformers among developing MNA - Egypt, Jordan, Morocco and Tunisia. Building a comprehensive picture of reform indicators for the rest of developing MNA (that is, excluding high income countries of GCC) is hard. But the available reform indicator evidence on these countries suggests that the MNA indicators would not alter the force of arguments made in the current study for MNA. On trade policy reforms, for example, most recent available estimates6 of average import tariff rates for six developing MNA countries not included in this study (Algeria, Iran, Lebanon, Libya, Syria and Yemen) amounts to 24 percent. If measures of non-tariff barriers were also included then the excluded countries would significantly worsen the reform index as they continue to have very high NTBs, compared to the MNA economies included. On other dimension of reforms such as privatization (outside oil sector), or removal of price controls in the economy or taxation, the late reformers in developing MNA such as Syria, Libya or Iran have a long way to go to catch up with the rest of the world. Therefore, we conjecture that if we include reform measures for all countries of NIA, the aggregate measure could look considerably worse. 5 There are insufficient observations from 1985 to derive regional average for ECA and SSA. 6Oliva (2000) -8- Chart 6. Composite Structural Reform Index in Developing Regions Structural Reform Index 75 7 0 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 6 5 - - - - - - - - - - - - - - - - -- - - - - - - - - - - - - - 55 _------ --_---- MNA 50 _7 -- - - -- - - - -- - - -- - - - _ ___ ___ ____ __ EAP 45 -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -_ LA C 40 _-- - ----- ------- --------- ------ -- - - - SA R 3 5 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 30 0 8 X 8 X 0 0 0 0 0 0 0 0)0 0) 0) 0 0 0) 0 0 0 0 0 Source: Staff estimates. This weak pace of reforms in MNA is because of slow progress in all components of structural reforms excluding real exchange rate over-valuation. In trade policy reforms, measured by the un-weighted average tariff rate, MNA region is the second worst among the four developing country regions averaging around 24 percent. LAC region, which had a tariff rate higher than MNA in 1985, has cut them rapidly to reach around 10 percent in 1998, marginally lower than EAP (Chart 7). Chart 7. Comparison Trade Policy Indicator Among Developing Regions Unweighted Tariff Rates (%) 80 7 0 - - - - - - - - - - - - - - - - 60 -- -- -- -- -- -- - - - - -------- ---------- - -- -- -- -- -- -- -- --- 40 ------------- ---- _--- _-_-_-_-_-_ _ =M N A 40 --- ----- ----- ------ ----- -- --- ---- --U '- E A P 30, . -, -KLAC 30 20 --- - - - - -- - - - _-* SAR _______ ___ O------------ -- n CD 1- CO 0c 0 _ tN to I- co o co 0c co co 0) 0 0 0 ) 0 ) 0) 0) 0) 0) a) a) 0 ) 0) a) 0) 0) a) a 0) Source: See Annex 1. Tax policy is measured by two sub-components - highest marginal tax rates on individuals and corporations. In terms of highest marginal individual income tax rate, there is not as much divergence among countries in 1998 as there was in 1985. MNA region has closed the gap considerably by bringing it down to 35 percent from being the highest taxed (57 percent) region among the four regions compared (Chart 8). -9- Chart 8. Highest Marginal Tax Rate on Individuals Among Developing Regions Individual Tax Rates (%) 65 55 0 45-- - - - -- - -- - - --o MNA 40 - -- - - ---- --w-EAP 35 -- - - - - - - - - _" * LAC 25 __---- 20 . Source: See Annex 1. Chart 9. Highest Marginal Corporate Rate on Individuals Among Developing Regions. Hihgest Marginal Corporate Tax Rates (percent) 65 55--- - - - - 50 --- - - - 45 ---- MNA 40 ---------- LAC 35 -- - - - - -- - - - - - _-- -LASC 30 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -S A R 20 co r- s 2 0 0 04 co 0 co 0 0_ 20 co 0 0 CD CD 0 0CD 0 0 Source: See Annex 1. In terms of highest marginal corporate tax, MNA region continues to be the second worst in 1998 as it was in 1985 (Chart 9). Though for the region as a whole the index was on a declining trend, some countries like Egypt actually moved in the opposite direction until 1990. Only the LAC region has aggressively moved to match the 30 percent corporate tax average in EAP. In terms of privatization, measured by the cumulative privatization proceeds expressed as a ratio of GDP, MNA region bears the rear guard averaging around 3 percent in 1998, some 50 percent below LAC (Chart 10). - 10- Chart 10. Privatization Proceeds Among Developing Regions Cummulative Proceeds from Privatization to GDP Ratios (%) 7 4 iA 3 ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~~~~~~~i-EAP LAC 2 -~~~~~~~~~~~~~~~~-*4-SAR 0 v - I I I I .-.-I 1990 1991 1992 1993 1994 1995 1996 1997 1998 Source: See Annex 1. In terms of real exchange rate overvaluation, as measured by David Dollar's overvaluation index 7, MNA region compares well with little overvaluation compared to strong under-valuation noted in the case of EAP or South Asia. However, good indicators of real exchange rate overvaluation are hard to come by, and the strength of MNA reformers (excluding Tunisia) in this regard does not accord well with regimes dominated by fixed pegs to US dollar for most of the period under investigation with occasional changes. However, an important conclusion to emerge from the comparison in movements of this index in Chart 11 is that MvNA region corrected largely the overvaluation episode that occurred in late 1980s but the Asian region have remained relatively undervalued to LAC and MNA, maintaining the edge in competition. Chart 11. Real Exchange Rate Over-Valuation in Developing Regions Deviation from PPP 160 1 20 - - - - - - - - - - - - - - - - - - - - 100 --- -- -- --- -- -- _ M NA 80 LAC 60 --=- SAR 20 0a0)0_ 0 0) 0 0 a) 0 _ 0) 0 0 Source: See Annex 1. Dollar (1992) -11- How does the pace of reform in MNA countries, individually, as against the regional average discussed before, compare with others ? The countries that are relative advanced in MNA (Egypt, Jordan, Morocco and Tunisia) are in the bottom two-thirds of countries in terms of cumulative liberalization since 1985. Comparing countries keeping their starting positions on reform index in 1985, MNA reformers can be classified as either gradual or slow reformers. Chart 12 shows the ranking of change in composite reform index between 1985-98 for 27 countries across the developing world. Of the four reformers in MNA, two (Jordan and Tunisia) fall in the bottom third, and two (Morocco and Egypt) fall in the middle third. Chart 12. Change in Composite Reform Index Change in Composite Reform Index (1985-98) Hukngary i Peru Bangladesh Malaysia Costa Rica Colonbia India Venezuela, Sri Lanka Mexico Argentina Nbrocco Chile Indonesia Egypt Arab Korea, Rep. Kenya Thailand Jordan Nblaw X Tunisia Botsw ana China South Africa Paklstan Pbland 0.0 10.0 20.0 30.0 40.0 Source: Staff estimates. Another way to compare reform performance of countries is to track the pace of reforms against their starting levels and the average for all reformers. Table 1 makes such a comparison. Morocco started off as below average in 1985 and continues to be below average in 1998, falling under the group of "slow reformers". The other three from MNA (Egypt, Jordan and Tunisia) had above average reform index in 1985 but slipped below average in 1998 as they turned out to be "gradual reformers". While it is instructive, as in Chart 13, to relate the pace of reforms in Table 1 to growth outcomes, it is important to remember the additional factors that facilitate or weigh down on a country's growth performance. For instance, in some countries in "early reformers" group, including Indonesia or Argentina, the weak financial regulatory environment has precipitated - 12- financial crises stripping away years of growth. Though Colombia, Peru and Venezuela are counted among intense reformers, their weak performance has possibly has origins in weak governance. The varied growth performance of countries under the general group of "gradual reformers" alludes to the strengths of spillover effects of emanating from location in a region of dynamic growth or not. While the East Asian countries in this group - China, Korea, Thailand - or, Poland and Tunisia reaped the benefits of location to major growing markets, Jordan and South Africa suffered because of slow growing regional economies. Among slow reformers, a common characteristic appears to be their slow speed of integration with the rest of the world. India is an exception in the group which has achieved high growth despite the slow pace of reforms benefiting from the large size of domestic market that has undergone deregulation. Table 1. Pace of Structural Reforms in Developing Countries Early reformers: Gradual reformers: Argentina China Botswana Egypt, Arab Rep. Above razil ordan he average Chile orea, Rep. Hungary Malawi 1985 Structural Indonesia oland Reform Index level Malaysia South Africa Mexico I'hailand Philippines Tunisia Intense reformers: Slow reformers: Colombia angladesh Below Costa Rica dia he average Peru enya Venezuela, RB orocco Pakistan Sri Lanka Source: Staff estimates. - 13 - Contrasting the pace of reforms with the realization of real per-capita GDP growth rates (Chart 13), we see a pattern that higher growth perforrnance has largely concentrated in gradual and early reformers. Chart 13. Pace of Reforms and Per-capita GDP Growth Reforms and Per-capita GDP Growth 90 Intensive Hungary Reformers Ea rly 85 - F