WPS A139 POLICY RESEARCH WORKING PAPER 213 9 Sm all States, Small states, no different from large states in income and Small Problems? growth, should receive the same policy advice large states do Because of their William Easterly greater openness, they may Aart Kraay be more vulnerable to volatility in terms-of-trade shocks - but their openness pays off in growth The World Bank Development Research Group Macroeconomics and Growth June 1999 POLICY RESEARCH WORKING PAPER 2139 Summary findings Small states have attractecl a good deal of research. Their annual growth rates are more volatile, partly Easterly and Kraay test wlhether microstates are any because of their greater volatility in responses to terms- different from other states in income, growth, and of-trade shocks - to which they are exposed because of volatility. their greater openness. But on balance their greater They find that, controlling for location, smaller states openness pays off positively in growth. are actually richer than other states in per capita GDP. Easterly and Kraay do recommend that srmall states This income advantage largely reflects a productivity diversify their risk by opening up more to interniational advantage - evidence against the idea that microstates capital markets, although the benefits of doing so are still are unable to exploit increasing returns to scale. unresolved in the literature. Small states do not have different per capita growth In general, they conclude, small states are no different rates, with or without controls. from large states and should receive the same policy advice large states do. This paper - a product of Macroeconomics and Growth, Development Research Group - is part of a larger effort in the group to study the needs of small states. Copies of the paper are available free from the World Bank, 1818 H Street NW, Washington, DC 20433. Please contact Kari Labrie, room MC3-456, telephone 202-473-1001, fax 202-522-1155, Internet address klabrie@o!worldbank.org. Policy Research Working Papers are also posted on the Web at http:,'/ www.worldbank.org/htmi/dec/Publications/Workpapers/home.html. The authors may be contacted at weasterly (aworldbank.org or akraay@worldbank.org. June 1999. (36 pages) The Policy Research WVorking Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers canry the names of the authors and should be cited accordingly. The finidings, interpretations, and conclusions exp^ essed in this paper are entirely those of the authors. They do not necessarily represent the vietc of the World Bank, its Execsitive Directors, or the countries they represent. Produced by the Policy Research Dissemination Center Small States, Small Problems? William Easterly and Aart Kraayl The World Bank X Views expressed here are not necessarily those of the World Bank or its member governments. 2 "Smallness is neither a necessary nor sufflcient condition for slow economic development" T.N. Srinivasan (1986) "Economic storm clouds are gathering over paradise and the outlook is undeniably gloomy. " A. Dolman (1985) Do small states suffer from their smallness? There are good theoretical reasons to believe that they do. The provision of public services may be subject to indivisibilities that lead to increasing returns to scale (Alesina and Spolaore 1997), especially fiscal institutions (Easterly and Rebelo (1993)) and defense (Kuznets (1960), Harden (1985)). Many theories of economic growth suggest increasing returns to scale in the private economy as well (Romer 1986, Barro and Sala-I-Martin 1995, Aghion and Howitt 1998), which may be difficult to realize in small states. Small economies may also be at a disadvantage because their size prevents them from diversifying into a wide range of activities, making them more vulnerable to terms of trade shocks than large states (Commonwealth Secretariat (1998)). Many small states suffer from poor location in that they are remote and/or landlocked, and many are located in regions prone to hurricanes and volcanic activity (Srinivasan (1986)). Public officials in small states may be much more likely to be subjected to conflicting pressures (Farrugia (1993)), and it may be difficult to recruit a high-quality civil service given the limited pool of candidates in small states (Streeten (1993)). These potential difficulties facing small states have not been lost on policymakers or academics. Numerous conferences and seminars on the special difficulties of small states have been convened over the past forty years (Robinson (1960), Benedict (1967), Dobozi et. al. (1982), Commonwealth Consultative Group (1985, 1997), Small States Financial Forum (1987,1988), Kaminarides et. al. (1989)). International organizations such as the United Nations have commissioned studies on the problems confronting small states for many years (United Nations (1971), Doumenge (1983)), and the United Nations has formally recognized the special difficulties of small states in a resolution to that effect (Briguglio (1995)). Titles of papers on small states (see our bibliography) frequently feature ominous terms and phrases such as "Problems", "Vulnerability", "Small is Dangerous", and even (twice) "Paradise Lost". In this paper, we look for empirical evidence of alleged disadvantages of size by examining microstates with population 1 million or less. (We will use the terms "small states" and "microstates" interchangeably). If smallness is a disadvantage, then microstates must suffer with a vengeance. In particular, we would expect that microstates must on average be less 3 developed and grow less rapidly than larger states. In this paper, we test this hypothesis using cross-country data ia a large sample that includes many microstates. In light of the grim predictions of theory, the picture of microstates which emerges from this analysis is somewhat surprising. After controlling for a range of factors, we find that microstates have on average higher income and productivity levels than small states, and grow no more slowly than large states. Per capita GDP growth rates are more volatile in microstates, due to their much greater exposure to international trade and fluctuations in their terms of trade. However, any growth disadvantages of this greater volatility are more than outweighed by the growth benefits of trade openness reaped by microstates by virtue of their necessarily large trade volumes. Finally, microstates are well-positioned to take advantage of opportunities international risk sharing, since the correlation of economic fluctuations in microstates with the world business cycles is surprisingly low. These resulls contribute evidence in support of the growing view in the literature that small size might not be a disadvantage after all. Kuznets (1960) notes that small states also have advantages: primarily that many are lucky to have good natural resources and have a small and more cohesive populations which allows them to adapt better to change. Srinivasan (1986) and Streeten (1993) argue without systematic empirical evidence that small may also be beautiful. Using a sample of 48 countries Millner and Westaway (1993) fail to find evidence that the effect of a number of growth determinants varies with country size. Armstrong et. al. (1998) uses cross- sectional regressions covering a large number of small states and independent regions to argue that population size. does not significantly affect growth, controlling for initial income and regional effects. The remainder of this paper proceeds as follows. In the next section we document that small states are richer and have higher productivity levels than small states. In the following section we observe that small states suffer no obvious growth rate disadvantage, and attribute this to a number of offsetting advantages and disadvantages of small states. In Section 3 we document that although trade openness contributes significantly to the greater volatility of growth in microstates, this is not the whole story. In Section 4 we note that microstates are relatively well-positioned to take advantage of opportunities to diversify away their special risks since they currently are not particularly financially open and the shocks they receive are relatively uncorrelated with those experienced by the rest of the world. Section 5 offers concluding remarks. I 4 1. Microstates and per capita income levels In this paper, we consider a large cross section of 157 countries for which at least 10 years of annual data on per capita GDP adjusted for differences in purchasing power parity is available. Of these, 33 are microstates defined as having an average population over the period 1960-1995 of less than one million. These countries are listed in Table 1, and range in size from tiny St. Kitts and Nevis with population of 42,000 to Mauritius with 912,000. The income range is similarly wide, from very poor African countries such as Guineau-Bissau and Comoros with real PPP-adjusted per capita GDPs around $600 to wealthy oil-exporting countries such as Qatar with per capita GDP of over $18,000. If microstates suffer from the disadvantages of smallness, they should be poorer on average than larger states. What do we actually find? Not controlling for any other characteristic, microstates have the same range of per capita incomes as the rest of the samnple (Figure 1). However, if we control for the location by continent of all countries, whether they are oil producers, and whether they belong to the OECD, then microstates are actually significantly richer than other states (Regression 1). Microstates are 50 percent (=-exp(.4025)-1) richer than other states, controlling for location. We note that this result does not reflect the obvious outliers in the sample, since the oil exporting countries Qatar and Bahrain are picked up by the OIL dummy, and Luxembourg and Iceland are picked up by the OECD member dummy. Even if we exclude two other particularly wealthy microstates (Bermuda and Bahamas), we still find that microstates are nearly 40 percent richer than other states. Figure 2 shows the income residual by quintile of population, and we see the very strong income effect in the microstate bottom quintile of population. We also note that the favourable performance of small states carries over to other quality of life indicators. For example, if we in tum use under-five infant mortality and life expectancy at birth as the dependent variable in the above regression, we find that infant mortality is significantly lower in small states by 22 per thousand, while life expectancy is about four years higher. We are not sure why microstates are so much richer than their regional neighbours and have so much better human development indicators, but we see this as a decisive refutation of the macro arguments that microstates suffer from a development disadvantage. Are microstates richer than others, controlling for location, because they save more or because they have a higher productivity level? We use the Mankiw-Romer-Weil (1992) 5 regression to answer this question. In the steady-state of the Solow model, output per person is given by: (1) Y/L = A (s/(x+&+n))x('-O') where Y/L is output per person, A is the level of labor-augmenting productivity, s is the investment to GDP ratio, x is the rate of labor-augmenting productivity growth, 8 is depreciation, n is population growth, and a is the share of capital income in GDP. We assume productivity growth of 2 percent and a depreciation rate of 7 percent. Following MRW, we take logs of both sides and regress the log of output per person on the same dummies as above (capturing continental and other productivity differences) and the log of the second multiplicative term in (1): (2) ln(Y/L) = In A + a/(1-a) [In s - ln(x+5+n)] We call this second term MRW, and the results of estimating this specification are in Regression 2. We find that small states' productivity advantage accounts for two-thirds of their income per capita advantage. Again this decisively refutes the notion that small states have a productivity disadvantage due to increasing returns to scale. We also refute the original MRW idea that productivity levels are the same across countries, as Asia and especially sub-Saharan Africa have significantly lower productivity than other regions. Once we allow the productivity level to vary, the coefficient on MRW implies a capital share of .28 - which is in line with most estimates from national income accounting. When we decompose MRW into its numerator and denominator from equation (1), we find that small states have significantly higher (log) investment rates - see Regression 3 - but not lower population growth rates (not shown). Hence, the one third of the small-state income effect in Regression 1 that is not attributable to productivity differences is attributable to higher investment in small states. However, we should take with more than a grain of salt the result that investment accounts for even as much as one-third of the income advantage of small states. The significance of the MRW term in Regression 2 may reflect reverse causality - richer states can afford to invest more and are usually thought to choose lower population growth than poor states. Or it may 6 reflect an omitted third factor, like incentive policies that affect both investment and income. One way of dealing with omitted third factors is to estimate equation 2 in chages. The results are not encouraging to the MRW explanation of income and growth. In Regression 4 we estimate the change in the MRW steady-state level using population growth and investment rates for 1960-70 and then 1985-95. The change in the MRW steady state levels does not explain the cross-country differences in growth rates over 1960-95. Figure 3 shows the variation in growth rates and in MRW steady state changes across small states. A view that aU countries have the same productivity growth but have different long-run growth rates because of changes in steady state levels doesn't work in the data, either for small states or aU states. Moreover, the Solow/MRW sources of growth accounting implied in (2) sometimes gives unreasonable predictions. For example, figure 4 shows actual income in Guyana compared to Guyanese income assuming a constant productivity growth rate (x=.02) and allowing capital per person growth to evolve using actual investment rates and population growth.2 The sources of growth exercise based on the Solow/MRW model cannot account for the boom in the 70s or for the collapse in the 80s. Nor can the sources of growth exercise based on the Solow/MRW model account for the negative growth in Guyana over the 40 years 1950-90. Capital growth per person was so rapid that Guyana should have had six times the income in 1990 that it actually had. Even if we assumed that productivity growth was at a lower bound estimate of 1 percent over 40 years (grey-shaded area), we still arrive at 4 times the actual income in 1990. Nor is there is a negative level shift of steady state income in Guyana, because the MRW change is barely different from zero (see the Guyana -GUY-- point in Figure 3). Clearly the assumption of a constant (and positive!) productivity growth rate is untenable for Guyana. But negative productivity growth does not make sense in the Solow/MRW framework if x is interpreted as technological progress - - it's hard to believe that Guyana had technological regress. Nor does the Solow/MRW framework give us any explanation as to why productivity growth rates might differ across countries. We have to move outside of the model to recognize that capital growth sometimes does not pass into output growth as the SolowIMRW predicts. This only strengthens the presumption that income differences like the small state positive income differential have primarily to do with differing levels of A - however that is interpreted - and little to do with capital growth per person. 7 2. Microstates and Macro Growth Even if the microstates do not have a disadvantage in levels, they may nevertheless grow more slowly over time. Several endogenous growth theories have a scale effect on per capita growth. Moreover, rmicrostates exhibit greater output volatility which has negative effects on growth (Ramey and Ramey (1996)). Small states have the same range of growth experiences as other states (Figure 5), suggesting that there is no obvious scale effect for growth rates. There is also no growth difference for small states after controlling for continental location, oil, and OECD dummies, as shown in Regression 5. As in other work, sub-Saharan Africa has lower growth than the rest of the world (Easterly and Levine 1997), and Asia has higher growth. However, there is no evidence that microstates either grow faster or slower than non-microstates Why do small states not suffer any apparent growth disadvantages due to their small size? To answer this question, we consider a parsimonious cross-country growth regression which captures two of the factors prominent in the small states debate: openness to international trade measured as the shaLre of imports and exports in GDP, and volatility measured as the annual standard deviation of growth rates within each country (Regression 6). We also include initial income to capture convergence effects, and secondary school enrollment rates as a measure of human capital. All of the non-dummy variables are significant of the expected sign: there is conditional converg,ence (negative coefficient on initial income), a positive effect of secondary enrollment and trade openness, and a negative effect of the standard deviation of annual growth. This regression framework provides some useful clues as to why the microstate dummy is not significant in the basic regression 5. In particular, we can see from this regression that small states will have several offsetting advantages and disadvantages. They are richer than other countries controlling for continent dummies (see previous section) and hence will have slower growth than average by the conditional convergence effect. They have slightly higher secondary enrollment, which would give therm higher growth, but the difference is not statistically significant (results not shown). Most important, microstates tend to have much higher trade shares (which is good for growth), offset by much higher volatility of growth rates (which is bad 2 We follow the usuaL conventions, using the perpetual inventory method to calculate the capital stock and calculating the initial y/k as (x+5+nr/s where n and s are average population growth and investment rates, 8 for growth).3 The insignificance of the microstate dummy therefore suggests that the negative effects of high initial income and high volatility are roughly offset by the positive effects of trade openness and better educational attainment. In order to document the magnitude of these offsetting effects, we first need to know how different microstates are from non-microstates in terms of their trade volumes and volatility. We first document the well-known fact that microstates typically have much higher trade ratios than larger states, as illustrated by Regression 7. The consequences for openness of being a small state are truly remarkable. Small states have a ratio of trade to GDP that is 54 percentage points (1.2 standard deviations) higher than the average economy controlling for continent dummies! Second, real per capita GDP growth rates tend to be much more volatile in small states, as illustrated by Regression 8.4 In particular, the standard deviation of annual real per capita GDP growth is 1.4 percentage points higher in microstates than in non-microstates. We have already shown that small states have higher income, which is a growth disadvantage because of the convergence effect. They also have higher secondary enrollment controlling for the usual dummies, which is a growth advantage, although the effect is not statistically significant. If the small state dummy is not significant in the overall growth regression (Regression 5 earlier), then the advantages and disadvantages of smallness must be roughly offsetting. Interestingly, the positive growth effect of openness (0.012x0.54=0.65 percent) is 2.5 times larger than the negative growth effect (-1.79x0.014-0.25 percent) of small states' greater output volatility. If output volatility is one of the consequences of openness (on which more below), this suggests that small states' greater openness is still on balance a positive factor for small states' growth. This finding is of particular interest, given the widely held view that small states suffer from their openness.5 Any source of growth volatility that is not related to openness, on the other hand, is detrimental to small states' growth. In summary, there is no evidence that small states suffer any growth disadvantage from their small size. This finding can be explained by several offsetting advantages and respectively, over the entire 40 year period. 3 The model of the aforementioned Alesi;,m and Spolaore 1997 has the prediction that openness will make small states more viable. 4In interesting historical footnote is that the greater volatility of small states has not always been accepted. Tarshis (1960) finds little evidence of a relationship between the coefficient of variation of per capita income and size across US states, and poses this as a puzzle. 9 disadvantages of small states. Although they are richer and experience more volatile shocks tha non-microstates, they reap substantial growth advantages from their exposure to international trade. Finally, it is interesting to note that one often-heard benefit of microstates does not appear to be empirically very important. It is often argued that one of the advantages of microstates is that they tend to be ethnically very homogeneous, which may make it easier for such states to forge the political consensus required to adjust to a changing environment (for example, Kuznets (1]960)). Easterly and Levine (1997) and Alesina, Baqir, and Easterly (1999) find that measures of ethnic fractionalization are associated with a lower level of public goods provision and lower growth. However, the mean value of the ethnolinguistic indicator of ethnic diversity among those microstates for which data is available is insignificantly different from that among non-microstates, suggesting that the benefits of homogeneity may not be especially pronounced for microstates. 3. Openness and V'olatility In the previous Section we saw that microstates reap growth benefits from their openness to international trade, but suffer growth costs due to their greater volatility of growth rates. In this section we consider in more detail the relationship between trade openness and volatility in microstates. A significant portion of the growth rate volatility experienced by small states can be attributed to volatility in their terms of trade, but this is not the entire story. Even after controlling for terms of trade volatility, growth rates in microstates are significantly more volatile than in non-microstates. We note first that the volatility of terms of trade shocks experienced by microstates is much greater than for larger states. We define terms of trade shocks as the growth in the local currency price of exports times the share of exports in GDP less the growth in the local currency price of imports less; the share of iimports in GDP, which captures both the magnitude of price fluctuations (changes in export and import prices) and their importance for the domestic economy (weighted by the shares of exports and imports in GDP). We then regress the standard deviation of this measure of terms of trade shocks on the same set of regional dummies as before, dunmny 5 This view of small states dates back at least to Scitovsky (1960). Dolman (1985) goes so far as to suggest that many small island states would be better off reverting to auaic subsistence economies. 10 variables to capture oil exporters and commodity exporters who are more likely to suffer extreme fluctuations in their terms of trade, and the microstate dummy (Regression 9). We find that there is a highly significant microstate effect, with the standard deviation of terms of trade shocks larger by 0.0 13 (or about one-third of one standard deviation of the dependent variable) in microstates. This terms of trade volatility might be due to two factors. First, we have already seen that the share of trade in GDP is especially large in microstates, and this may contribute to the magnitude of our measure of terms of trade shocks (since it weights changes in import and export prices by the shares of imports and exports in GDP). Second, microstates' exports are likely to be more specialized than those of large states, both in terms of products exported and in terms of export markets (Kuznets (1960), Knox (1967), Annstrong and Read (1998)). The distinction between these factors is important because there is little that microstates can do about their overall trade volumes - autarky is simply not an option for small states that produce a much narrower range of goods and services than they consume, and moreover we have already documented the substantial growth benefits accruing to small states due to their openness. If in contrast the greater volatility of growth is due to excessive reliance on a few export products and a few export markets, then policies designed to help diversify exports may help to dampen economic fluctuations. We can get a rough idea of the relative importance of these two factors by redefining the terms of trade shock as the unweighted difference between the growth in export prices and the growth in import prices. When we use this alternative measure of terms of trade shocks as the dependent variable in Regression 9, we find that the microstates dummy is negative and insignificant (Regression 10). That is, the volatility of changes in the price of exports relative to imports are if anything lower in microstates relative to larger states. Although this is not conclusive evidence, it does cast doubt on the notion that microstates are especially vulnerable to external shocks simply because their international trade is more specialized. Rather, the greater volatility of terms of trade shocks in microstates is primarily due to their unavoidably large trade shares. Finally, it is worth noting that greater volatility of growth in microstates is not solely due to their greater susceptibility to terms of trade shocks. To illustrate this point, we re-estimate Regression 8, but include the volatility of the terms of trade as an explanatory variable 11 (Regression 11). We find that the microstate dummy remains significant even after controlling for the effect of greater terms of trade volatility on the volatility of overall growth. This additional volatility may be due tb several factors. Many of the microstates in our sample are located in areas prone to natural disasters such as hurricanes, and the higher growth volatility in small states may simply reflect the devastating effect of these natural forces. However, it is also possible that some of this observed volatility reflects difficulties in measuring per capita incomes, which may be particularly acute in small states where statistical institutions may be weaker than average. 4. Opportunities fDr Diversification In the previous section we have seen that microstates experience much more volatile growth rates than non-microstates. This in part reflects their greater vulnerability to terms of trade shocks, and also the tendency of many microstates to suffer heavily from natural disasters. In this section we briefly consider the potential of small states to mitigate the adverse effects of this largely-unavoidable volatility by sharing risks with the rest of the world. One of the potential benefits of financial openness is that it allows countries to share risks with the rest of the world, by holding claims on assets located outside their borders whose returns are not perfectly cor.related with returns to domestic assets. The magnitude of these benefits depends on how volatile are shocks to the domestic economy, and the extent to which they are uncorrelated with shocks abroad. Small states are particularly well-situated to benefit from such risk sharing arrangements, for two reasons. First, small states suffer large shocks, as documented in Section 3. Second, in contrast to the often-heard view that small states are particularly susceptible to cyclical fluctuations in large states, we find that the shocks experienced by small states are not unusually correlated with the world business cycle. We illustrate this point with Regression 12, which regresses the correlation of per capita GDP growth in a country with OECD average real per capita GDP growth on the same set of dummies as before, as well as the logarithm of average per capita CiDP (to capture the stylized fact document by Kraay and Ventura (1998) that business cycles in poorer countries tend to be less correlated with the world average cycle), and a microstate dummy. The microstate dummy is insignificant, suggesting that microstates are not unusually correlated with the OECD cycle. However, it is important to note that growth rates in neighbouring microstates may be highly correlated, especially to the extent that growth rate volatility reflects natural disasters such as hurricanes. This suggests that regional 12 arrangements to share risk among microstates will be much less valuable than pooling risks with a wider range of countries. Despite the potential benefits of risk sharing through participation in intemational financial markets, microstates do not appear to be especially open financially. We illustrate this with Regressions 13-14, which regress two alternative measures of financial openess on a set of regional dummies as well as the logarithm of average per capita income. In Regression 13 the dependent variable is the fraction of years for which data is available in which the IMfF reports restrictions on capital account transactions in that country.6 The coefficient on the microstate dummy is positive, although insignificantly so. This suggests that microstates are not particularly open to financial flows, as measured by legal impediments to such flows. Combining this observation with the empirical results of Lewis (1995), who finds that consumption risks are less diversified in countries with this measure of capital controls, this suggests that microstates are not taking full advantage of the opporunities for risk diversification afforded by international capital markets. The outcome measure of financial openness paints a somewhat more favourable picture, as the microstate dummy is positive and statistically significant at conventional levels. This suggests that the volume of capital flows is slightly larger for microstates than for non- microstates, although the magnitude of this effect is small - only about 2 percentage points of GDP. Overall, this evidence suggests that microstates are not as financially open as they might be given the high volatility they face, and hence are not fully exploiting opportmities for international risk diversification. We conclude this section with the observation that although greater financial openness may help microstates insure against the large shocks they receive, financial openness is itself no panacea. Grilli and Milesi-Ferretti (1995) and Rodrik (1998) both note that there is no evidence that financially-open economies grow faster or enjoy higher investment rates. On the other hand, there is also no systematic evidence in favor of the popular view that by opening up financially, countries expose themselves to greater volatility due to the vagaries of international financial markets (Kraay (1998)). In summary, although financial openness may provide a valuable means for small states to diversify some of the large risks they face, existing evidence does not support the view that there will be a large growth payoff from such policies. 6 As reported in the Annual Report on Exchange Arrangements and Exchange Restrictions. The disadvantages of this measure are well-known. First, it captures only the presence, and not the intensity of controls. Second, it captures only controls on residents, and not on non-residents, although there is some presumption that these two types of controls are correlated across countries. 13 5. Conclusions Our analysis suggests that small states have perhaps received excessive attention from the literature - notwithstanding our own addition to the literature! -as special cases calling for special policy measures. We find that microstates have, if anything, significantly higher per capita income than others in their region. There is no significant difference in growth performance between large and small states. It is true that growth volatility and volatility of terms of trade shocks as percent of GDP is higher in small states, but this is due entirely to their greater trade openness - and the net benefits of openness on growth are positive. The one missing piece in the current situation of rnicrostates is that they are not fully exploiting the potential to diversify their risks by opening up to international capital movements. But even the payoff to filling in this last missing piece is unclear from evidence in the literature. This is not to say that microstates are free of economic problems! Many microstates are still poor, and promoting growth out of that poverty is as important as it is in other poor countries. The good news is that the lessons of experience from all countries are applicable to small states, so they can benefit from the large amount of cross-country evidence on the determinants of long- run growth. 14 Tables and Figures Table 1 - Small States Population Average Per Capita (Thousands) GDP, 1985 PPP-Adjusted Dollars ATG Antigua and Barbuda 63 5329 BHR Bahrain 419 10342 BHS Bahamas, The 237 11136 BLZ Belize 178 3548 BMU Bermuda 58 15356 BRB Barbados 247 5341 BWA Botswana 880 1516 COM Comoros 340 632 CPV Cape Verde 295 746 CYP Cyprus 638 5084 DJI Djibouti 344 1479 FJI Fiji 602 3149 GAB Gabon 777 3853 GMB Gambia, The 628 803 GNB Guinea-Bissau 739 644 GRD Grenada 92 2632 GUY Guyana 719 1630 ISL Iceland 223 9689 KNA St. Kitts and Nevis 42 4399 LCA St. Lucia 148 3264 LUX Luxembourg 358 11934 MDV Maldives 201 1908 MLT Malta 341 4049 MUS Maurtius 916 4092 QAT Qatar 384 18278 REU Reunion 496 2253 SLB Solomon Islands 299 1845 SUR Suriname 378 2877 SWZ Swaziland 556 2358 SYC Seychelles 59 2214 VCT St. Vincent and the Grenad 107 3312 VUT Vanuatu 145 1633 WSM Samoa 160 1844 15 Regression 1 Dependent Variable: log of average income 1960-95 Method: Least Squares Included observations: 157 White Heteroskedasticity-Consistent Standard Errors & Covanance Variable Coefficient Std. Error t-Statistic Prob. SMALLSTATE 0.402504 0.108228 3.588670 0.0005 ASIA 7.517973 0.111549 52.12283 0.0000 AFRICA 6.691796 0.085404 81.37846 0.0000 WESTERN 7.932229 0.098881 85.54604 0.0000 HEMISPHERE MIDDLE EAST & N. 7.863013 0.169759 45.58837 0.0000 AFRICA EUROPE & 8.100983 0.116053 76.49556 0.0000 CENTRAL ASIA OIL 0.814728 0.178268 4.900633 0.0000 OECD 1.168653 0.148174 9.309253 0.0000 R-squared 0.708909 Mean dependent var 7.855922 Adjusted R-squared 0.695233 S.D. dependent var 0.982946 16 Regression 2 Dependent Variable: log of average income 1960-95 Method: Least Squares Included observations: 139 White Heteroskedasticity-Consistent Standard Errors & Covariance Variable Coefficient Std. Error t-Statistic Prob. ASIA 7.438718 0.142546 52.18481 0.0000 EUROPE & 7.714671 0.133412 57.82584 0.0000 CENTRAL ASIA WESTERN 7.800517 0.094056 82.93488 0.0000 HEMISPHERE MIDDLE EAST & N. 7.763342 0.143795 53.98906 0.0000 AFRICA OECD 1.122059 0.113640 9.873815 0.0000 OIL 0.691713 0.150642 4.591781 0.0000 AFRICA 6.865222 0.093269 73.60652 0.0000 SMALLSTATE 0.267340 0.132294 2.020799 0.0454 MRW 0.389346 0.101618 3.831453 0.0002 R-squared 0.761487 Mean dependent var 7.795722 Adjusted R-squared 0.746809 S.D. dependent var 0.994449 17 Regression 3 Dependent Variable: Log of average investment rate/GDP (PPP) 60-95 Method: Least Squares Included observations: 139 White Heteroskedasticity-Consistent Standard Errors & Covanance Variable Coefficient Std. Error t-Statistic Prob. ASIA -1.867634 0.117983 -15.82967 0.0000 AFRICA -2.581324 0.117818 -21.90938 0.0000 WESTERN -1.968057 0.072648 -27.09038 0.0000 HEMISPHERE MIDDLE EAST & N. -2.064563 0.158454 -13.02945 0.0000 AFRICA EUROPE & -1.493982 0.084612 -17.65692 0.0000 CENTRAL ASIA OIL 0.137961 0.160770 0.858126 0.3924 OECD 0.198926 0.098043 2.028980 0.0445 SMALLSTATE 0.381396 0.107068 3.562201 0.0005 R-squared 0.440238 Mean dependent var -1.956707 Adjusted R-squared 0.410327 S.D. dependent var 0.680181 18 Regression 4 Dependent Variable: Average per capita growth 1960-95 Method: Least Squares Included observations: 133 Variable Coefficient Std. Error t-Statistic Prob. C 0.018962 0.001613 11.75912 0.0000 DMRW6095 0.003758 0.002847 1.319799 0.1892 R-squared 0.013122 Mean dependent var 0.019050 Adjusted R-squared 0.005589 S.D. dependent var 0.018633 19 Regression 5 Dependent Variaible: Average per capita growth 1960-95 Method: Least Siquares Included observations: 154 White Heteroskedasticity-Consistent Standard Errors & Covariance Variable Coefficient Std. Error t-Statistic Prob. ASIA 0.026679 0.004888 5.457927 0.0000 AFRICA 0.006639 0.002874 2.310184 0.0223 WESTERN 0.016090 0.002569 6.263201 0.0000 HEMISPHERE MIDDLE EAST & N. 0.018039 0.006376 2.829106 0.0053 AFRICA EUROPE & 0.020941 0.004848 4.319413 0.0000 CENTRAL ASIA OIL -0.014381 0.007439 -1.933011 0.0552 OECD 0.006522 0.004430 1.472347 0.1431 SMALLSTATE 0.002222 0.004983 0.445971 0.6563 R-squared 0.195323 Mean dependent var 0.016336 Adjusted R-squared 0.156743 S.D. dependent var 0.021323 20 Regression 6 Dependent Variable: Average per capita growth 1960-95 Method: Least Squares Included observations: 130 White Heteroskedasticity-Consistent Standard Errors & Covarance Variable Coefficient Std. Error t-Statistic Prob. ASIA 0.137565 0.024502 5.614343 0.0000 AFRICA 0.117585 0.023412 5.022435 0.0000 WESTERN 0.134214 0.026701 5.026535 0.0000 HEMISPHERE MIDDLE EAST & N. 0.137746 0.024181 5.696354 0.0000 AFRICA EUROPE & 0.136707 0.025164 5.432535 0.0000 CENTRAL ASIA OIL -0.000407 0.004736 -0.085912 0.9317 OECD 0.009192 0.005901 1.557607 0.1220 LQIN60 -0.017360 0.003702 -4.689944 0.0000 SECONDARY 0.000342 0.000111 3.069972 0.0027 ENROLLMENT 60-95 Share of Trade in 0.012076 0.002746 4.397740 0.0000 GDP 60-95 Standard Deviation of -0.179370 0.082047 -2.186187 0.0308 Growth 60-95 R-squared 0.535076 Mean dependent var 0.016972 Adjusted R-squared 0.496006 S.D. dependent var 0.019612 21 Regression 7 Dependent Variable: Share of Trade in GDP 60-95 Method: Least Squares Included observations: 158 White Heteroskedasticity-Consistent Standard Errors & Covanance Variable Coefficient Std. Error t-Statistic Prob. ASIA 0.657933 0.128117 5.135396 0.0000 AFRICA 0.522459 0.042598 12.26481 0.0000 WESTERN 0.613623 0.063168 9.714153 0.0000 HEMISPHERE MIDDLE EAST & N. 0.698416 0.081605 8.558546 0.0000 AFRICA EUROPE & 0.788507 0.059614 13.22688 0.0000 CENTRAL ASIA OIL 0.145042 0.094684 1.531846 0.1277 OECD -0.178562 0.077809 -2.294883 0.0231 SMALLSTATE 0.538525 0.070860 7.599897 0.0000 R-squared 0.285557 Mean dependent var 0.739223 Adjusted R-squared 0.252217 S.D. dependent var 0.436784 22 Regression 8 Dependent Variable: Standard Deviation of Growth 60-95 Method: Least Squares Included observations: 154 White Heteroskedasticity-Consistent Standard Errors & Covarance Variable Coefficient Std. Error t-Statistic Prob. ASIA 0.047770 0.002558 18.67561 0.0000 AFRICA 0.061945 0.002457 25.21095 0.0000 WESTERN 0.049090 0.002689 18.25762 0.0000 HEMISPHERE MIDDLE EAST & N. 0.060996 0.006357 9.595790 0.0000 AFRICA EUROPE & 0.055067 0.004943 11.14022 0.0000 CENTRAL ASIA OIL 0.015740 0.004851 3.244719 0.0015 OECD -0.025540 0.004311 -5.924337 0.0000 SMALLSTATE 0.014339 0.003346 4.285854 0.0000 R-squared 0.467655 Mean dependent var 0.055809 Adjusted R-squared 0.442131 S.D. dependent var 0.021426 23 Regression 9 Dependent variable: Standard Deviation of Terms of Trade Shocks 1960-95 Number of observations: 114 Mean of dep. var. .038019 LM het. test = 1.87192 [.171] Std. dev. of dep. var. .021479 Durbin-Watson = 2.07124 [<.8691 Sum of squared residuals = .026595 Jarque-Bera test = 2.17388 (.337] Variance of residuals = .250897E-03 Ramsey's RESET2 = .137189 (.712] Std. error of regression .015840 F (zero slopes) = 14.5415 [.000] 1R-squared = .489871 Schwarz B.I.C. = -8.03086 Adjusted R-squared = .456183 Log likelihood = 314.945 Estimated Standard Variable Coefficient Error t-statistic P-value SSA .04:2971 .419929E-02 10.2330 [.000] ASIA .022997 .323085E-02 7.11797 [.000] ECA .016725 .219593E-02 7.61616 (.000] MENA .039738 .690534E-02 5.75471 [.000) LAC .037033 .317373E-02 11.6685 [.000] OIL .02:3363 .503847E-02 4.63685 [.000] COMMOD .740058E-02 .452971E-02 1.63379 [.105] MICROSTATE .01:3328 .473321E-02 2.81585 [.006] Standard Errors are heteroskedastic-consistent (HCTYPE=2). 24 Regression 10 Dependent variable: Standard Deviation of Unweighted Terms of Trade Shocks 1960-95 Number of observations: 114 Mean of dep. var. = .126255 LM het. test = 3.77342 [.052] Std. dev. of dep. var. = .072712 Durbin-Watson = 2.36931 [<.9971 Sum of squared residuals = .314207 Jarque-Bera test = 3.85662 (.145] Variance of residuals = .296422E-02 Ramsey's RESET2 = .084551 [.772] Std. error of regression = .054445 F (zero slopes) = 13.6498 (.000] R-squared = .474073 Schwarz B.I.C. = -5.56154 Adjusted R-squared = .439342 Log likelihood = 174.193 Estimated Standard Variable Coefficient Error t-statistic P-value SSA .146951 .011390 12.9012 t.0003 ASIA .094062 .011595 8.11254 [.000] ECA .046957 .553333E-02 8.48622 [.000) MENA .138424 .029647 4.66910 (.000] LAC .140449 .014181 9.90428 [.000] OIL .074365 .023821 3.12183 [.002] COMMOD .023068 .012845 1.79589 [.075] MICROSTATE -.011439 .011375 -1.00558 [.317] Standard Errors are heteroskedastic-consistent (HCTYPE=2). 25 Regression 11 Dependent variable: Standard Deviation of Real Per Capita GDP Growth 1960-1995 Number of observations: 114 Mean of dep. var. = .052993 LM het. test = .058422 [.809] Std. dev. of dep. var. = .021039 Durbin-Watson = 1.82486 [<.466] Sum of squared residuals = .023269 Jarque-Bera test = 7.39687 (.025] Variance of residuals = .221607E-03 Ramsey's RESET2 = .32S584E-02 (.955] Std. error of regression = .014886 F (zero slopes) = 15.0879 [.000] R-squared = .534787 Schwarz B.I.C. = -8.12293 Adjusted R-squared = .499342 Log likelihood = 322.561 Estimated Standard Variable Coefficient Error t-statistic P-value SSA .045460 .524191E-02 8.67245 [.000] ASIA .0376574 .424625E-02 8.87235 [.000] ECA .028!505 .330153E-02 8.63375 [.000] MENA .046493 .842420E-02 5.51904 [.000] LAC .034650 .462190E-02 7.49686 (.000] OIL .012151 .580664E-02 2.09255 [.039] COMMOD .709771E-02 .396159E-02 1.79163 [.076] MICROSTATE .020540 .518460E-02 3.96181 (.000] Standard .249983 .105027 2.38017 (.019] Deviation Of Terms of Trade Standard Errors are heteroskedastic-consistent (HCTYPE=2). 26 Regrssion 12 Dependent variable: Correlation of Real Per Capita GDP Growth with OECD Average Real Per Capita GDP Growth, 1960-95 Number of observations: 155 Mean of dep. var. = .246648 LM het. test = .866619 [.352] Std. dev. of dep. var. = .267756 Durbin-Watson = 1.95295 (<.711] Sum of squared residuals = 7.23486 Jarque-Bera test = .769712 (.681] Variance of residuals = .049896 Ramsey's RESET2 = .034630 [.853] Std. error of regression = .223373 F (zero slopes) = 8.47520 (.000] R-squared = .344712 Schwarz B.I.C. = -2.73913 Adjusted R-squared = .304039 Log likelihood = 17.5644 Estimated Standard Variable Coefficient Error t-statistic P-value SSA -.717189 .199481 -3.59529 [.000] ASIA -.712958 .231572 -3.07877 (.002] ECA -.638812 .239380 -2.66861 (.0081 MENA -.822765 .233648 -3.52138 (.001] LAC -.692913 .232984 -2.97408 [.003] OECD .082437 .068673 1.20043 [.232] LOGQAV6095 .119230 .029067 4.10185 (.000] OIL -.077165 .065326 -1.18123 (.239] COMMOD .026310 .057571 .456996 [.648] MICROSTATE .013354 .053808 .248181 (.804] Standard Errors are heteroskedastic-consistent (HCTYPE=2). 27 Regrssion 13 Dependent variable: Fraction of Years in Which Capital Controls in Place, 1960-1995 Number of obse:rvations: 139 Mean of dep. var. = .787387 LK het. test = 12.2811 J.0001 Std. dev. of dep. var. = .342953 Durbin-Watson = 2.03770 [<.867] Sum of squared residuals = 11.7237 Jarque-Bera test = 35.1819 [.000] Variance of residuals = .090881 Ramsey's RESET2 = 2.65476 (.106] std. error of regression = .301465 F (zero slopes) = 5.51078 (.000) R-squared = .277703 Schwarz B.I.C. = -2.11787 Adjusted R-squared = .227311 Log likelihood = -25.3683 Estimated Standard Variable Coefficient Error t-statistic P-value SSA 2.09580 .328127 6.38717 [.000] ASIA 2.0)6680 .375284 5.50729 (.000] ECA 2.39422 .403090 5.93967 [.000] MENA 2.13384 .384726 5.54640 (.000] LAC 2.0)8108 .393251 5.29200 (.000] OECD -.149049 .100129 -1.48857 (.139] COMMOD .541426E-02 .062675 .086386 [.931] OIL -.097368 .115647 -.841936 [.401] LOGQAV6095 -.170939 .049805 -3.43213 t.001] MICROSTATE .025574 .080311 .318431 (.751] Standard Errors are heteroskedastic-consistent (HCTYPE=2). 28 Regression 14 Dependent variable: Average Capital Inflows Plus Capital Outflows as a Fraction of GDP, 1960-95 Number of observations: 132 Mean of dep. var. = .062080 LM het. test = 12.0349 [.001] Std. dev. of dep. var. = .058247 Durbin-Watson = 2.03507 [<.866] Sum of squared residuals = .329045 Jarque-Bera test = 111.565 [.000] Variance of residuals = .269709E-02 Ramsey's RESET2 = .463370 (.497] Std. error of regression = .051933 F (zero slopes) = 4.75418 [.000] R-squared = .259653 Schwarz B.I.C. = -5.62445 Adjusted R-squared = .205037 Log likelihood = 208.328 Estimated Standard Variable Coefficient Error t-statiozic P-value SSA -.109303 .089036 -1.22763 [.222] ASIA -.123786 .091137 -1.35824 [.177] ECA -.126912 .102485 -1.23834 [.218] MENA -.116162 .103124 -1.12643 (.262] LAC -.153697 .104193 -1.47512 [.143] OECD .016410 .028921 .567432 [.571] COMMOD -.013906 .011405 -1.21930 [.225] OIL -.027007 .015827 -1.70637 [.090] LOGQAV6095 .023934 .013285 1.80152 (.074] MICROSTATE .027312 .013028 2.09646 [.038] Standard Errors are heteroskedaBtic-consistent (HCTYPE=2). Per capita income r'~~~~~~~~~~~~~~~~E~~~r la z P8 | 8 f~~ | I ' . , ,~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~% 30 Figure 2: Unexplained income level and population size 0.3 - Small states C 0.25 - O 0.2- 0 0.15 - 0 0.1 A) M -0.05 -0.1 -0.1505 1 4 9 22 935 - ~~~~Population quintiles less thn or equal to (in millions): GDP Per Capita Growth, 1960-9' 42 I al 1,r a aa I o o 0 0 0 0~~~~~~~~~~~~~~~~~~~~~~~~~~~~~r Cl rA~~~~~~~~ 9A 0 0 0 0 0 0~~~~~~~~~~~~~~~~r .. . .......... ........ 0...... .. Figure 4: Predicted Income in Guyana from Solow/MRW model and actual income 6000 Predicted income from sources of growth 5500 accounting with productivity growth 2% 5000 - 4500 - 34000 - 3500 3000 0 2500 2000 1500 S / _ ~~~~~~~~~~~ACUl 11 wcoESMl#c 1000 -1 i i I I I I I I I_ _ . 2500 4/ _p 0b - c . -- -- c NC)-.,'', -l .C -.' -.--' .q .''- "'q "') NC NC IC C C Per capita growt} o In I PO 0 t-) 0 C 0 I I I IX ;I I H o H 0~~~~~~~~~~~~~~~~ 34 References Aghion, Philippe and Peter Howitt (1998). Endogenous Growth Theory. Cambridge: MIT Press Alesina, Alberto, Reza Baqir, and William Easterly (1999). "Public goods and ethnic divisions," forthcoming, Quarterly Journal of Economics. Alesina, Alberto, and Enrico Spoalare (1997). "On The Number and Size of Nations". Quarterly Journal ofEconomics. November. Armstrong, H R-J. de Kervenoael, X. Li and R. Read (1998). "A Comparison of the Economic Performance of Different Microstates, and Between Microstates and Larger Countries". World Development. 26(4):639-656. Barro, Robert and Xavier Sala-I-Martin (1995). Economic Growth. New York, NY: McGraw- Hill. Benedict, Burton (1967). Problems of Smaller Territories. London: Athlone Press. Briguglio, Lino (1995). "Small Island Developing States and Their Economic Vulnerabilities". World Development. 23(9): 1615-1632. Cashin, Paul and Norman Loayza (1995). "Paradise lost? Growth, convergence, and migration in the South Pacific". AMFStaffPapers. 42:608-41. Commonwealth Consultative Group (1985). Vulnerability: Small States in the Global Society. London: Commonwealth Secretariat. Commonwealth Consultative Group (1997). A Future for Small States: Overcoming Vulnerability. London: Cormmonwealth Secretariat. Dobozi, Istvan, Clare Keller and Harriet Matejka (1981). Small Countries and International Structural Adjustment: A Collection of Hungarian and Swiss Views. Dolman, Antony (1985). "Paradise Lost? The Past Performance and Future Prospects of Small Island States", in Edward Dommen and Philippe Hein, eds. States, Microstates and Islands. London: Croom Helm. Doumenge, Francois (1983). Viability of Small Island States. United Nations Conference on Trade and Development. Easterly, William and Ross Levine (1997). "Africa's Growth Tragedy: Policies and Ethnic Divisions". Quarterly Journal of Economics. November. Easterly, William and Sergio Rebelo (1993). "Fiscal Policy and Economic Growth: an Empirical Investigation". Journal ofMonetary Economics32:417-57. Farrugia, Charles (1993). "The Special Working Environment of Senior Administrators in Small States". World Development. 21(2):221-226. 35 Grilli, Vittorio and Gian Maria Milesi-Ferretti (1995). "Economic Effects and Structural Determinants of Capitail Controls". International Monetazy Fund Staff Papers. 42(3):517- 551. Harden, Sheila (1985). Small is Dangerous: Micro States in a Macro World. London: Frances Pinter. Kaminarides, John, Lino Briguglio and HenkN. Hoogendonk (1989). The Economic Development of Small Countries: Problems, Strategies and Policies. Delft: Eburon. Kraay, Aart and Jaume Ventura (1998). "Comparative Advantage and the Cross-Section of Business Cycles". World Bank Policy Research Department Working Paper No. 1948. Kraay, Aart (1998). "In Search of the Macroeconomic Effects of Capital Account Liberalizalion". Manuscript, World Bank. Kuznets, S. (1960). "Economic Growth of Small Nations", in E.A.G. Robinson, ed. (1960). The Economic Consequences of the Size ofNations: Proceedings of ofA Conference Held by the Interrational Economic Associations. Toronto: MacMillan. Lewis, Karen (199,6). "What Can Explain the Apparent Lack of International Consumption Risk Sharing?". Journal of 'olitical Economy. 104:267-297. Mankiw, N. Gregory, David Romer and David Weil (1992). "A Contribution to the Empirics of Economic Growth". Quarterly Journal ofEconomics. Milner, Chris and Tony Westaway (1993). "Country Size and the Medium-Term Growth Process: Some Cross-Country Evidence". World Development. 21(2):203-211. Rainey, Garey and Valerie A. Ramney. (1995). "Cross-country Evidence on the Link Between Volatility and Growth". American Economic Review. 85:1138-5 1. Robinson, E.A.G., ed. (1960). 7he Economic Consequences of the Size ofNations: Proceedings of ofA Conference Held'by the International Economic Associations. Toronto: MacMillan.. Rodrik, Dani (1998). "Who Needs Capital Account Convertibility?", in Peter Kenen, ed. "Should the IMF Pursue Capital-Account Convertibility?". Princeton Essays in International Finance, No. 207. Romer, Paul M. (1986). "Increasing Returns and Long-Run Growth" Journal ofPolitical Economy. 94:1002-37. Scitovsky, Tibor (1960). "International Trade and Economic Integration as a Means of Overcoming the Disadvantages of a Small Nation", in E.A.G. Robinson, ed. (1960). The Economic Consequences of the Size ofNations: Proceedings of ofA Conference Held by the International Economic Associations. Toronto: MacMillan. Small States Financial Forum (1987). Round Table on Foreign Investment and Commercial Finance for Small States. London: Crown Agents Financial Advisory Service. 36 Small States Financial Forum (1987). Round Table on Some Practical Possibilities of Financial Co-operation Between Small States. London: Crown Agents Financial Advisory Service. Streeten, Paul (1993). "The Special Problems of Small Countries". World Development 21(2):197-202. Srinivasan, T.N. (1986). "The Costs and Benefits of Being A Small, Remote, Island, Landlocked or Ministate Economy". World Bank Research Observer. 1(2):205-218. Tarshis, L. (1960). 'The Size of the Economy and Its Relation to Stability and Steady Progress", in E.A.G. Robinson, ed. (1960). The Economic Consequences of the Size of Nations: Proceedings of of A Conference Held by the International Economic Associations. Toronto: MacMillan. United Nations Institute for Training and Research (1971). Small States and Territories: Status and Problems. New York: Amo Press. Policy Research Working Paper Series Contact Title Author Date for paper WPS2124 Social Exclusion and Land Robin Mearns May 1999 G. Burnett Administration in Orissa. India Saurabh Sinha 82111 WPS2125 Developing Country Agriculture and Bernard Hoekman May 1999 L. Tabada The New Trade Agenda Kym Anderson 36896 WPS2126 Libert6, Egaiite, Fraternite: Monica Das Gupta May 1999 M. Das Gupta Exploring the Role of Governance 31983 In Fertility Deciine WPS2127 Lifeboat Ethic versus Corporate Monica Das Gupta May 1999 M. Das Gupta Ethic: Social and Demographic 31983 Implications of Stem and Joint Families WPS2128 Learning Outcomes and School Gladys Lopez Acevedo May 1999 M. Geller Cost-Effectiveness in Mexico: 85155 The PARE Program WPS2129 Agricultural Extension: Generic Gershon Feder May 1999 P. Kokila Challenges and Some Ingredients Anthony Willett 33716 for Solutions Willem Zijp WPS2130 Deep Integration, Nondiscrimination, Bernard Hoekman May 1999 L. Tabada and Euro-Mediterranean Free Trade Denise Eby Konan 36896 WPS2131 Efficiency Wage and Union Effects William F. Maloney May 1999 T. Gomez In Labor Demand and Wage Eduardo Pontual Ribeiro 32127 Structure in Mexico: An Application Of Quantile Analysis WPS2132 A Regime-Switching Approach to Maria Soledad Martinez June 1999 A. Yaptenco Studying Speculative Attacks: Peria 38526 A Focus on European Monetary System Crises WPS2133 Resolution of Corporate Distress: Stijn Claessens June 1999 R. Vo Evidence from East China's Simeon Djankov 33722 Financial Crisis Leora Klapper WPS2134 interlinkage, Limited Liability, and Kaushik Basu June 1999 M. Mason Strategic Interaction Clive Bell 30809 Pinaki Bose WPS2135 Hungary's integration into Bartlomiej Kaminski June 1999 L. Tabada European Union Markets: 36896 Production and Trade Restructuring Policy Research Working Paper Series Contact Title Author Date for paper WPS2136 An Empirical Analysis of Competition, Scott J. Wallsten June 1999 P. Sintim-Aboagye Privatization, and Regulation in 38526 Telecommunications Markets in Africa and Latin America WPS2137 Globalization and National Andr6s Solimano June 1999 D. Cortijo Development at the End of the 84005 20th Century: Tensions and Challenges WPS2138 Multilateral Disciplines for Bernard Hoekman June 1999 L. Tabada Investment-Related Policies Kamal Saggi 36896 , /