Beyond Macroeconomic Stability: The Quest for Industrialization in Uganda1

Abebe Aemro Selassie

doi:10.5089/9781451870893.001.A001

Jump to Content Jump to Main Navigation

Beyond Macroeconomic Stability: The Quest for Industrialization in Uganda1

Author: Mr. Abebe Aemro Selassie

Contributor Notes

Author’s E-Mail Address: aselassie@imf.org

Publication Date:: 01 Sep 2008

eISBN:: 9781451870893

Language:: English

Keywords:: WP; SG country; exchange rate; per capita income; growth acceleration; economic activity; take-off date

Download PDF (540.2 KB)

Uganda has registered one of the most impressive economic turnarounds of recent decades. The amelioration of conflict and wide ranging economic reforms kick-started rapid economic growth that has now been sustained for some 20 years. But there is a strong sense in policy making circles that despite macroeconomic stability and reasonably well functioning markets, economic growth has not translated into significant structural transformation. This paper considers (i) Uganda's record of economic transformation relative to the high growth Asian countries and (ii) the contending explanations as to why more transformation and higher growth has proved elusive.

Abstract

I. Introduction and Summary

Uganda has registered one of the most impressive economic turnarounds of recent decades. General Idi Amin’s power grab in 1971 had prompted a period of political instability and insecurity, causing the economy to go into a tailspin (Figure 1). Since the late-1980s, however, the economy has enjoyed a period of high uninterrupted growth, broadly attributable to peace, easy taxes, and a tolerable administration of justice.² The civil strife that characterized the period through the late 1980s diminished greatly in most parts of the country with the advent of the National Resistance Movement (NRM) administration in 1986; a comprehensive pro-market economic reform program was implemented to reduce the many onerous taxes and economic restrictions that were in effect; and economic and political institutions have also improved to some extent, with checks on the executive branch emerging (reasonably independent judiciary, an improving parliament, a vibrant media, and active civil society organizations). Beyond fairly rapid economic growth (an average of 7½ percent between 1990 and 2007), these developments have contributed to a sharp drop in poverty from 55 percent of the population in 1993 to 31 percent in 2006.

Despite this growth record, macroeconomic stability and reasonably well functioning markets, however, there is a strong sense among policy makers that economic growth has not translated into significant economic transformation and industrialization. This disappointment is in turn prompting more and more ad hoc and enterprise specific interventions. Increasingly, the stress is on how best Uganda can emulate successful government interventions, particularly in East Asia. On occasion, the more statist and expansionary model pursued by Ethiopia, has also been cited as a possible example for Uganda to follow. As one senior official put it “there is a quiet rebellion underway.”

It is useful to think about three distinct but overlapping challenges for economic take-off in countries like Uganda. The first is getting growth started, and the second sustaining growth. Uganda has done well on these two counts, with economic growth being sustained for some 20 years notwithstanding significant exogenous shocks along the way. The third challenge is economic transformation—the change from being a primary-commodity based economy to one where industry and services dominate economic activity. Such transformation is the kernel of economic development. And since this is what engenders high economic growth, the Government of Uganda, rightly, considers it as the overriding economic challenge facing the country.

Against this backdrop, this paper tries to address three questions.

To what extent has growth in Uganda been accompanied by economic transformation? To gauge progress on this front (Section III), we look at the evolution of a range indicators that track economic transformation by normalizing the starting points of Uganda’s growth acceleration and that of the handful of (mainly Asian) countries that have sustained rapid rates of economic and attained middle income status in the last few decades. We compare transformation in “take-off” time, as it were. And the paper’s main finding is that while there has some economic transformation in Uganda over the last 20 years it has been from a lower base and at a slower pace than in the “sustained growth” countries.
Second, why has economic transformation been limited? Section IV considers the extent to which the arguments that have been advanced as constraints on growth in low income countries account for Uganda’s current predicament. These arguments include: weak fundamentals (poor geography, low human capital, weak institutions, etc.); limited openness to trade (either due to weak initial endowments or foreign aid induced dutch disease type effects); failure to strike the right balance between states and markets (too much “Washington Consensus” and too little “Developmental State”); and, “globalization” or more specifically the adverse consequences of the “rise of China.”
Third, we consider what next for policies (Section V)?

The paper’s main conclusions are that the uninterrupted growth that the country has registered since 1987 is a remarkable achievement. Private investment is high by African standards. Export diversification has been impressive, and more recently the volume of exports has started to grow rapidly. All this suggests that the current economic framework has served Uganda well, and only in need of improvements in a few areas.

Perhaps the first priority is to improve the country’s infrastructure (particularly roads and electricity). Industrial type activities but also more modern forms of agriculture make intensive use of infrastructure services. In their absence, it is perhaps not so surprising that the Ugandan economy remains centered on small-holder agriculture. And it is not just more money that is needed to address the infrastructure deficit (indeed, this is already happening) but also a review of whether the current institutional structures are consistent with much higher levels of investment and improved service delivery.

Some attention also needs to be given to developing a focused growth strategy. Chastened by poor returns to government intervention in the past, there has been little effort in Uganda to settle on the 3 or 4 sectors that can serve to catalyze modernization. The result has not been fewer interventions but a broad range of uncoordinated and enterprise-specific interventions. A more focused strategy is needed instead.

The other major challenge will be to try and improve export competitiveness, including by striking the right balance between reliance on foreign aid and external competitiveness. The debate on this started some years ago in Uganda but remains very relevant today. The performance of the tradables sector has until recently been weak. Going forward, sustaining the recent pick-up in export performance will be key for greater economic transformation.

To be sure, there is no unanimity on whether land-locked resource scarce countries like Uganda should strive for rapid industrialization. Paul Collier (2007) for one stresses that these countries should look to specialize in regional trade and give priority to policies on rural development—where he expects the bulk of these countries’ population to remain for the foreseeable future. But the view in this paper is that beyond the reasons stated above, industrialization is also important in Uganda from a demographic perspective. The country has one of the highest population growth rates in the world, with its current population of about 30 million set to reach 100 million people by 2050. Reliance on agriculture, where current levels of productivity are quite low, certainly offers scope for growth. But industrial type activities offer much more scope for rapid capital accumulation that can stay well ahead of population growth.

As one of the first sub-Saharan African (SSA) countries to embark on the process of liberalization and pro-market reforms in the late-1980s, the ongoing debate on the appropriate policy direction for Uganda holds lessons for other countries in the region. The broader international context is also likely influencing the debate, including the swing in the pendulum in development economics literature (Kanpur, 2005) and indeed practice (Latin America) towards a greater role for the state in the economy. The debate in Uganda has been prompted by the desire to hasten economic transformation, but only partly. The quality of domestic institutions have gotten better, improving the prospects for better interventions this time round. In the wake of the debt write-offs by external creditors, there is also more fiscal (and policy) space to permit government interventions. Whatever it is due to—the changing fads of development economics or a case of history repeating itself—the post-stabilization challenges that African countries are facing are increasingly reminiscent of those challenges faced in the immediate aftermath of independence. This time (as then) it is more and more all about how best to foster rapid industrialization.

II. The Stylized Facts of Uganda’s Recovery

At some level, any despondency over Uganda’s recent economic performance is hard to fathom. Between 1990 and 2007, real GDP growth averaged close to 7½ percent, compared with 3 percent in the rest of Sub-Saharan Africa (SSA). In per capita terms the corresponding figures are 3¼ percent and 1 percent (Figure 2). Uganda’s growth since 1990, then, has been among the strongest in Africa or indeed elsewhere (Figure 3).

Importantly, growth has been sustained for a remarkably long period—20 years and counting. Historically, getting growth started has not been all that difficult. Work by Haussmann, Pritchett and Rodrik (2004) for instance shows that between 1957 and 1992 there were 83 episodes of “growth accelerations” around the world involving some 60 countries.³ Thus increases in the growth rate are relatively common, and can be triggered by factors such as better leadership and commodity price booms. The greater challenge has been sustaining growth beyond 8–10 years, which is the average duration of high growth episodes (Berg, Ostry and Zettlemeyer, 2007). With growth lasting for some 20 years, Uganda’s accomplishment on this front is notable. Growth in Uganda has also been quite effective at reducing poverty (Figure 4)—from 56 percent in 1993 to 31 percent in 2006.

The country’s achievement is all the more impressive when one considers the regional and internal conflicts that have raged over the period. For much of the last 20 years, northern Uganda has contributed very little to measured economic activity because of the low- intensity (but no less brutal) conflict that until recently was underway in the area. Conflicts in neighboring Democratic Republic of Congo and South Sudan have until recently also prevented trade with these large and important markets for Uganda. Globally, estimates show that 1 percentage point growth in a neighboring country raises home-country growth by 0.4 percentage points (Collier and O’Connell, 2007). If instead conflict prevails, negative spillovers abound.

As might be expected for a country emerging from conflict and rebuilding its economic base, much of the growth since the late 1980s was driven by increased factor accumulation. Mikkelesen (2005) calculates that the contribution of productivity to growth during 1986-2003 was negligible, with capital accumulation explaining 85 percent of the increase in output and the balance being explained by labor input. Such over reliance on capital accumulation is not particularly healthy, of course, but also not uncommon in the early years of sustained growth episodes—see for example Mauritius (Subramanian and Roy, 2001).

High rates of economic growth in Uganda have been delivered alongside macroeconomic stability and considerable structural reforms. Tighter monetary and fiscal policies reduced inflation from well over 100 percent in the 1980s to single digits in the early 1990s (text figure). The black market premium for foreign exchange was eliminated by the mid-1990s. Trade barriers were reduced sharply. Virtually all sectors of the economy were also liberalized. To give but two examples: in the banking sector, government involvement was reduced sharply with the privatization of the largest commercial bank and foreign bank entry was allowed; in the coffee sector, the state marketing board, which sought to stabilize coffee prices for farmers but ended up taxing them, was dismantled.

The result is an economy which has been subjected to a large number of shocks but has continued to expand at a fairly rapid impressive rate. One explanation for the economy’s seeming imperviousness to shocks must be the strongly pro-market policies that have been put in place. With many state enterprises in government hands and/or most prices administratively determined, the response to shock would likely have been sluggish. But with price decisions left to market forces they have tended to adjust quickly to reflect relative scarcities. The private sector has also acted nimbly in these circumstances. The result has thus been an economy that is more efficient and resilient.

But the growth record is not without its shortcomings. For one, much of this growth has merely served to reverse the sharp decline in per capita income between the early 1970s and the late 1980s. Consequently, it was only around 2000 that per capita income reverted to its previous peak (Figure 1). To some extent, one could argue that only the growth since then can really be said to have improved welfare. But in reality much economic capacity was destroyed in the 1970s and 1980s. So it would not be unreasonable to consider much of the growth since the start of recovery in the late-1980s as “new” growth. Second, as noted above, there is a strong sense among policy makers that this period of high uninterrupted growth has not delivered much industrialization and economic transformation. We turn to this issue next.

III. Economic Transformation

This section considers the extent to which economic growth in Uganda has been accompanied by economic transformation. Following, Johnson, Ostry and Subramanian (2007), the main approach we use is to compare Uganda’s performance relative to the benchmarks set by the small group of (mainly) Asian countries that have sustained high rates of economic growth and engendered significant economic transformation over the last 50 years—sustained growth (SG) countries hereafter.⁴ The reasoning is straight forward. The overriding objective in Uganda is to become a middle income country in the shortest possible timeframe, much as the SG countries have done. With 20 years of growth under its belt, how well is Uganda emulating the SG countries? To this end, we align the starting point of Uganda’s growth episode with that of the SG countries and look at the evolution of various indicators of economic transformation in what might be called take-off time.⁵

What is the appropriate take-off point for Uganda? The empirical literature on growth transitions uses either economic criteria (Haussmann, Pritchett and Rodrik, 2004) or statistical techniques (Berg, Ostry, and Zettlemeyer, 2007) to identify the inflexion points in countries’ growth trajectories. Uganda’s growth “acceleration” is captured by both these types of approaches, with either 1989 or 1990 identified as the start of the growth episode. This seems quite reasonable, coming about 3 years after the NRM assumed power and heightened civil conflict subsided.⁶ For the purposes of this study, we use 1990 as Uganda’s take-off year.

Starting with per capita income, at take-off Uganda’s per capita income was well below the average of the SG countries—and, in fact, only higher than one country, China. The average SG country’s per capita income was 2½ times more than Uganda’s (Figure 5). And it wasn’t just that the starting income levels were different but growth rates in the SG countries subsequently were also much higher—5½ percent versus 3 percent in Uganda. This is a statistically significant difference. Thus, 15 years after the start of growth acceleration the SG countries’ per capita income had increased by 100 percent compared to a 60 percent increase in Uganda. From the outset, then, it is clear that there is an order or magnitude difference between the growth accelerations in Uganda and those in the SG countries.

The composition of output in Uganda has altered considerably over the last 15 years, with services replacing agriculture as the largest sector. The share of the industrial sector has also increased significantly from 12 percent in 1990/91 to 24 percent in 2005/06 (Figure 6). Much of this increase had, however, taken place by 2001/02, and the share of industry in total output has since been flat. Within industry, the share of manufacturing was broadly unchanged over the entire period at some 7 percent of GDP. Overall, Uganda’s level of industrialization is well below that observed in the early stages of the SG countries’ take-off and subsequently (Figure 7, top panel). The SG country average nonetheless masks a fair amount of dispersion, with the share of industry in Korea and Indonesia being in the 12 – 17 percent range at the start of their growth acceleration and well over 35 percent in Chile and China. Also noteworthy is the relatively high share of services in economic activity in Uganda at the present juncture notwithstanding its significantly lower per capita income level (Figure 7, lower panel). The transition out of agriculture has to date thus been as much towards the services sector as the industrial sector. One explanation for this is the difficulties for a land-locked African country to nurture a significant industrial sector as in, say, the coastal Asian economies. But geography, as discussed more fully below, need not preclude further industrialization. Moreover, there is the precedent of at least one other land-locked African country—Zimbabwe, before is current predicament—that was able to develop a significant industrial sector and manufacturing exports to boot (Wood and Jordan, 2000).

The share of population in urban areas is another proxy for economic transformation, providing a sense of how much production is shifting away from agriculture to industrial type activities for which agglomerations are important. Urbanization in Uganda is still very low at around 12 percent in 2005, and does not appear to have increased much over the last 20 odd years (Figure 8). In contrast, at take-off, the share of people living in Urban areas in the SG countries was close to 40 percent and steadily increased thereafter. Indeed, the share of population in urban areas in Uganda is low even by sub-Saharan African standards where it averaged some 35 percent in 2005.

The development of the financial sector also closely tracks economic transformation. A well developed financial system mobilizes and pools savings, facilitates the exchange of goods and services, and allows the diversification and management of risk. These functions influence savings and investment decisions as well as technological innovations, and hence economic growth. Relative to the SG cases, the picture is of fairly slow progress in financial deepening in Uganda (Figure 9, top panel). Some 15 years into Uganda’s takeoff, the ratio of financial liabilities to GDP stands at around a third of the level observed in the SG countries at takeoff (Figure 9, lower panel). The picture with respect to private sector credit is worse still, with the ratio of private sector credit to GDP in Uganda at the current juncture standing at only one-eighth of the level in the SG countries.

The behavior of savings and investment also has an important bearing on economic growth (Figure 10). Both aggregates have increased markedly in Uganda since 1990, but remain lower than the average levels observed in the SG countries around their take-offs and subsequently. But the averages for the SGs again mask considerable variation. Current levels of savings in Uganda (about 9 percent of GDP) are not far off the levels observed in the low-savings SG countries (Dominican Republic and Egypt). In a similar vein, there are one or two countries among the SGs cases with whom Uganda’s current level of investment (21 percent of GDP) compares favorably.

Another indicator that has been found to be closely associated with growth transitions is openness to trade (Jones and Olken, 2007). We start by considering one dimension of openness, export. Uganda’s export to GDP ratio at the start of its growth episode was less than 5 percent compared to the SG average of around 15 percent. Fifteen years on, the SG average had doubled and was on an upward trajectory. In contrast, the level of goods exports in Uganda, after increasing in the early years had stabilized at around the 10 percent mark (Figure 11, top panel). Interestingly enough, merchandise exports were lower still in China and Korea before and in the early years of their growth take-offs. Subsequently, of course, they have grown at a remarkable pace. Also noteworthy is the fact that some 10 years before take-off (i.e. around 1980), Uganda’s export to GDP ratio was well above the average level in the SG countries at take-off. The subsequent collapse of exports likely reflects the sharply overvalued exchange rate that prevailed at that time (see below).

A number of authors also stress the importance of manufacturing exports for sustained economic growth. Rodrik (2006) notes that there is a premium on the exports of manufactured goods.⁷ Hausmann, Klinger, and Lawrence (2008) note that exports rather than domestic production is a better indicator of structural transformation given that domestic production can always take place behind protective barriers while “exporting more likely to reveal actual productive and efficient production.” How well has Uganda done on this front? Again, the share of such manufacturing exports in GDP is well below the levels observed in the SG countries (Figure 11, lower panel). Only Indonesia (an oil exporter) stands out as having had lower manufacturing exports than Uganda in the immediate aftermath of growth acceleration.

Using a slightly different approach, we also looked at Uganda’s openness to trade (proxied by the exports plus imports divided by GDP) in 1990 and 2005 relative to a much larger sample of countries. In doing so, we controlled for the factors that have a bearing on openness, including per capita income, country size and natural endowments (whether a country is land locked, an oil exporter etc). The results show that Uganda’s openness both in 1990 and at the current juncture are below what it ought to be (scatter plots). But the Uganda indicators in these regressions though negative are not significant. In contrast, the SG countries tend to trade well above their “potential” consistent with the oft-repeated characterization that their economic growth was highly export oriented. Similar cross country regressions were estimated for the export to GDP ratio and manufactured exports (not reported). Again, the results suggest that the values for Uganda are lower than they should be according to key determinants of export performance but not at statistically significant levels. So there are signs of openness and exports being below potential, but not definitively so.

We also looked for the presence of real exchange rate overvaluation, something that might account for the trade outcomes just considered. To this end, we first compared Uganda’s price level with its income level and those of other countries via annual cross country regressions for the 1960 – 2003 period (Frankel, 2004 and JOS, 2007). According to this indicator, Uganda’s exchange rate was sharply overvalued through the late-1980s (Figure 14). More limited overvaluations is also evident in mid-1990s. For the SG countries this indicator shows consistent undervaluation in their take-off period. Second, Uganda’s real effective exchange rate was highly volatile in the 1980s but has been more stable and more depreciated since the 1990s (Figure 15).

In sum, the indicators of outward orientation can, at best, be interpreted as showing limited progress with integration into the rest of the global economy. Growth in Uganda, at least until very recently, has not been of the export-led variety.

Lastly, a brief look at demographics. Both the rate of population growth and structure have a bearing on economic growth (Bloom et al, 2003). Perhaps the most significant of these is the effect of demographic transition—the change in the age structure of the population. Some authors attribute a significant part of East Asia’s stellar economic performance in recent decades to such a “demographic dividend” (Bloom, Canning et al, 2007). As life expectancy increases, fertility rates tend to drop and population numbers start to stabilize. For a while at least, this leads to declining dependency ratios, with a large working age cohort relative to the young and elderly. This generally implies a higher national savings rate as people in the working age cohort (15 to 64) tend to save more including to tide them over in old age. Such a demographic transition has yet to take place in Uganda, and the dependency ratio remains much higher than in the SG countries (Figure 16). Improvements in life expectancy have not been matched by declines in the fertility rate, and the effect has been one of the highest population growth rates in the world at some 3½percent. In the very long-term and provided fertility rates decline, the demographic transition may have a positive bearing on economic growth. But in the near to medium-term, the effect of rapid population growth and attendant high dependency ratio in Uganda is likely to be somewhat negative.

Is growth without structural transformation, then, the Uganda record?

No, this would too harsh a characterization:

For one, there has been some economic transformation over the last 20 years. Uganda’s levels of industrialization, savings, and investment have all improved markedly. Encouragingly also, these indicators are now within the range observed in the early years take-off in the SG countries (see also JOS, 2007). The improvement in the level of industrialization is all the more noteworthy given the considerable restructuring of the sector as trade barriers were lowered including more recently with Kenya—the regional manufacturing powerhouse—in the context of the East African Customs Union. It is reasonable to expect more industrialization from this stronger, more competitive base. And the pick-up in export growth in the last few years, including of manufactured goods, is suggestive of this.

Second, most of the different indicators of economic transformation generally show Uganda to be following the same path as the SG countries, but from a low base and at a slower pace. One possibility is that there is a threshold effect at work. Perhaps countries have to have a certain level of per capita income and/or human capital to enjoy rapid economic transformation. Evidence of an income threshold is scant in the recent literature on growth transitions and what it takes to make growth sustained.⁸ This would be clearer still if population rather than countries is used as the unit of analysis. China and India (which is not in the SG sample) started with an income level well below Uganda and have engendered tremendous economic transformation for many times the number of people in the SG countries. But perhaps there is indeed a threshold effect with respect to human capital. We consider this in section IV.

Third, the performance of the tradables sector in Uganda gives some cause for concern. There has certainly been considerable export diversification. A look at direct measures of competitiveness also does not suggest that there is anything amiss. The current level of the real effective exchange rate (REER) is considerably more depreciated than at anytime in Uganda’s recent history. Our other measure of overvaluation also does not suggest any overvaluation at the current juncture. However, outcome variables—the ultimate test— suggest otherwise. Both the trade and current account balance are much wider now than at the start of the reform period (Figure 17). The trade balance was mildly negative in the late 1980s reflecting the external financing constraint. But once this constraint was removed through the availability of concessional foreign loans and grants in the early 1990s, the trade deficit widened sharply (to around 15 percent of GDP) and has remained at that level since (Figure 17, lower panel). This contrasts sharply with experience of the SG cases. In those cases, the typical pattern was for the trade balance to improve over time.

One final issue to consider here: Could we be concluding that economic transformation has been limited because we picked the wrong take-off point? Perhaps the appropriate take-off point for Uganda should be circa 2000, when per capita income had reverted its previous peak. After all, in places like Korea and Vietnam, their growth transition did not happen until some 10 years after military conflict ended—1962 and 1985, respectively. Shifting the start of Uganda’s take-off date would, however, not change much. First, Uganda’s indicators for industrialization, savings and investment would still be at the low-end of the levels observed in the SG cases and systematically lower with respect to indicators of openness. Second, shifting the take-off period for Uganda would belie the considerable improvement there has been over the years.

IV. Explaining the record

The preceding discussion suggests that while Uganda’s growth performance has been impressive by historic standards, it has been lower than in the sustained growth countries. Economic transformation has also been much more limited than in these countries whom Uganda is looking to emulate. This section considers the possible explanations for this outcome. The arguments presented here are not exhaustive of all the possible explanations nor mutually exclusive. Indeed, their proponents would likely be quick to acknowledge the role that the other explanations play. But there are important differences in emphasis.

The first set of explanations essentially suggest that what matters for growth are a country’s initial endowments: geography, quality of institutions, etc. In this view, spurts of growth are always possible, but with weak fundamentals growth is unlikely to be sustained.
The second explanation would essentially attribute Uganda’s failure to industrialize to aid-induced dutch disease. The performance of the tradable sector in Uganda has until very recently been weak, and this likely reflects the more appreciated real exchange rate that aid has engendered.
The third argument faults the strongly pro-market policy framework that Uganda has followed. The argument is along the lines that in the absence of strong institutions and infrastructure, market allocation of resources has proved as inefficient as state allocation of resources.
The last explanation we consider is the impact that “globalization” (for want of a better phrase) has had on Uganda’s efforts to industrialize. In this view, by industrializing before Africa, Asia has gained a first-mover advantage that will endure because of the importance of agglomeration for production. Africa may simply have “missed the boat.”

A. Fundamentals

The economic growth literature identifies initial endowments such as geography and the quality of institutions among the fundamental determinants of a country’s long-term economic performance. The proximate cause of economic growth are physical and human capital accumulation and technological change. But, the argument goes, it is these deeper fundamental attributes that “determine which societies will innovate and accumulate, and therefore develop and which will not” (Rodrik et al 2002). Geography or natural endowments are seen to affect economic growth directly (through its effect on land productivity, disease burden and thus human capital, etc.) and indirectly (through its effects on the form of institutions a country adopts and distance to markets). And in recent years much has been written about institutions—the formal and informal rules that shape economic interaction within society—as being of fundamental importance for economic growth. (Another fundamental determinant of long-term economic performance identified in the literature is openness to trade. But unlike institutions and geography, openness seems more amenable to near-term change as a result of policy action. Accordingly, we consider this separately below.)

Geography and human capital

Uganda’s geographic attributes mark it as having a particularly difficult environment for economic growth. The fact that it is land locked permanently constrains access to large markets, reducing the country’s ability to exploit economies of scale. Uganda also has a high disease burden even by African standards (Table 1). And the diversity of its population as reflected by the high value of its ethnic fractionalization index also points to high potential for social conflict and makes collective action difficult (Collier, 2007).

Table 1.

Indicators of Fundamental Country Attributes

^{^1/}Index showing probablity that any two randomly selected individuals in a country belong to different ethnolinguistic groups.

Source: World Bank, World Development Indicators.

Table 1.

Indicators of Fundamental Country Attributes

			Uganda	Sub-Saharan Africa	Sustained Growth Cases	Other Developing Countries
Geography
	Distance from coast line (km)		969	526	202	274
	Disease burden (% of pop. in Malaria prone a		100	85	22	34
	Ethnolinguisitc fractionalization ^1/		0.90	0.66	0.34	0.32
Integration
	Export plus import to GDP (percent)
		1990	26	69	100	84
		2003	40	73	118	92
Institutions
	Heritage Foundation (low score better)
		1995	3.15	3.6	3.01	3.24
		2005	2.95	3.4	2.97	3.07
	Economic Freedom of the World (aggregate index, high score better)
		1990	2.68	4.67	6.18	5.29
		2004	6.35	5.59	6.66	6.44
	World Bank Government Effectiveness Index (high score better)
		1996	-0.39	-0.65	0.54	-0.15
		2005	-0.48	-0.79	0.48	-0.08

^{^1/}Index showing probablity that any two randomly selected individuals in a country belong to different ethnolinguistic groups.

Source: World Bank, World Development Indicators.

Table 1.

Indicators of Fundamental Country Attributes

			Uganda	Sub-Saharan Africa	Sustained Growth Cases	Other Developing Countries
Geography
	Distance from coast line (km)		969	526	202	274
	Disease burden (% of pop. in Malaria prone a		100	85	22	34
	Ethnolinguisitc fractionalization ^1/		0.90	0.66	0.34	0.32
Integration
	Export plus import to GDP (percent)
		1990	26	69	100	84
		2003	40	73	118	92
Institutions
	Heritage Foundation (low score better)
		1995	3.15	3.6	3.01	3.24
		2005	2.95	3.4	2.97	3.07
	Economic Freedom of the World (aggregate index, high score better)
		1990	2.68	4.67	6.18	5.29
		2004	6.35	5.59	6.66	6.44
	World Bank Government Effectiveness Index (high score better)
		1996	-0.39	-0.65	0.54	-0.15
		2005	-0.48	-0.79	0.48	-0.08

^{^1/}Index showing probablity that any two randomly selected individuals in a country belong to different ethnolinguistic groups.

Source: World Bank, World Development Indicators.

But, as they say, geography is not destiny. Overcoming these weak initial conditions Uganda’s economy has been expanding at a healthy clip. So to the extent initial endowments matter the argument has to be more that even if they do not preclude growth accelerations they nonetheless prevent higher growth rates (and economic transformation) from being realized (JOS). But the exact mechanism for this is unclear. To be sure the SG countries are all costal economies, but Botswana, another land-locked African country, has come close to matching their higher growth record. Long distance from markets has also not hindered Mauritius from growing rapidly nor industrializing. Another, less celebrated success of industrial development in Africa is Zimbabwe. Through the late 1990s, manufactured exports accounted for some 30 percent of the country’s total exports (Wood and Jordan, 2003)—a remarkably high share for a land-locked country. Nor was this high share explained by trade with South Africa since through the mid-1990s the two countries barely traded with each other.

Beyond this, it could be argued that some of Uganda’s geographic attributes, such as land-lockedness, could actually have a positive bearing on economic growth. First, in so far as remoteness provides protection from imports, being landlocked should makes it possible for some industries to flourish. Second, according to Nunn (2007), in the African context being in the hinterland likely minimized the long-lasting damage that coastal areas suffered from the slave trade, the negative effects of which, he argues, are still being felt.

A more plausible channel through which initial endowments may constrain faster economic growth and transformation is the human capital channel. The idea here is that economies likely need a certain level of economy-wide skill level to industrialize rapidly. If a country is below this threshold, economic take-off will not happen. And if on top of this health outcomes are poor, a vicious cycle develops: with low life expectancy the incentives to invest in education diminishes and with it the prospect of higher incomes. An even more direct link between income and human capital is the fact that the poorer one is the less affordable education also is—children from a relatively young age are needed to work to raise family incomes.

Life expectancy in Uganda at 50 years is very low (Figure 18). Indeed, it has only recently recovered to this level, having dropped to the low 40s in the mid-1990s partly on account of the AIDS pandemic. In contrast, life expectancy in the SG countries averaged 57 years at take-off. Only one country (Indonesia) in the group had life expectancy close to Uganda’s at take-off. In the case of this particular dimension of human capital, Uganda is clearly well behind the SG countries at the point of “take-off”. But overall, as Johnson et al (2007) note, there is no clear theory why poor health conditions would allow growth accelerations but not sustained higher growth. Second, the available formal evidence is that while good health increases income levels, the predicted effect is small (Acemoglu and Johnson, 2007).

Turning to education, according to the imperfect proxy that we have for this, years of schooling, Uganda is now within the ball park of the education attainment observed in the SGs in their early years of take-off (JOS, 2007). Average years of schooling in Uganda as of 2000 were 3.5 years. In 1985, the earliest year for which comparable cross country data is available, the average year of schooling in the SG countries stood at 5.2 years (and was presumably lower still in earlier years).

If indeed the stock of human capital is the key constraint on higher economic growth then Uganda is in a quandary. The importance of improvements in human capital has been accepted and acted on by the government. Spending on health and education now accounts for around 30 percent of total government spending. With the recent launch of universal secondary education (to supplement the universal primary education first initiated in 1997), even more outlays on education are likely in the coming years. But despite the additional resources, education and health outcomes are only likely to improve gradually: teacher training colleges have to be expanded before a new cadre of teachers can be trained and deployed; medical colleges have to train more health sector workers; and so on. More broadly, the progress that Uganda and other African countries have made towards universal primary schooling in recent decades is by historic standards quite impressive—possibly even unprecedented (Clemens, 2004).

Institutions

The extensive literature on this in recent years points to the importance of property rights protection, apportionment of economic and political powers, reigning in of excessive corruption and rent-seeking etc. for the level of per capita GDP. Weak institutions reduce the return to investment, thus affecting the incentive for accumulation and adoption of new technologies. Those that champion the importance of institutions stress that even where geography and openness matter, it is through institutions that they exert an influence on economic development (Rodrik, Subramanian, and Trebbi, 2002, and Easterly and Levine, 2002). Thus for example in places where the disease burden was high, colonial governments introduced extractive institutions which have persisted to this day. In contrast where the disease burden was low, they established settler colonies with broadly the same quality of institutions as in Colonial capitals (Acemoglu, Johnson and Robinson, 2000). In a similar vein, it is argued, even if openness matters, it is because trade creates constituencies that push for better domestic institutions (JOS, 2007). Institutional development has also been identified as playing a role in structural transformation. Amin and Mattoo (2006) argue that better quality institutions are particularly important for the services sector, as they will reduce the transaction costs for the more complex web of transactions that typically pertain in such activities.

Are weak institutions the binding constraint on faster economic growth and structural transformation in Uganda? This is not clear from the data. For one, the quality of Uganda’s economic and political institutions have improved markedly from being well below what its income level would predict in 1990 to above this level by 2005 (Figures 19 and 20). The quality of institutions are now well within the range observed in other developing countries.⁹ Second, and somewhat more formally, we also looked for evidence on the extent to which Uganda might be below its steady state income level on account of its institutions. To this end, we regressed current income on the fundamental determinants of income (geography, openness, and institutions) together with an indicator for Uganda.¹⁰ The estimated coefficient on the Uganda indicator is negative but insignificant (Table 2), suggesting that neither institutions (nor the other deep determinants of income) are playing a significant role to keep income below its potential.

Table 2.

Economic potential and initial endowments

Institutions instrumented by settler mortality data from AJR 2001; *, **, *** significance respectively at the 90, 95 and 99 % levels

Nuanced and extensive as the literature on institutions is, one dimension that has not been given enough attention is state capacity. Note that state capacity—which here is used to refer to the ability of government’s to deliver effective health, education and infrastructure services—is fairly different from what are typically taken as the attributes of a country with strong institutions in this literature, which typically refer to rule of law and constraints on the executive. And as Fukuyama (2007) argues, state capacity may be particularly important factor at early stages of development. How strong is state capacity in Uganda? Crude cross-country comparisons paint a mixed picture. Starting with infant mortality as a proxy—on account of its usefulness as a ready measure of development in which government services play a significant role—Uganda’s administrative capacity is found to about where its per capita income would predict.¹¹ More direct measures of state capacity, such as indicators of “government effectiveness” and “control of corruption” World Bank’s governance indicators database similarly show Uganda to be about where its per capita income would predict.¹² Again, a case of Uganda being about where it needs to be.

The fundamentals-matter-most argument, then, has some appeal, but does not provide a satisfactory explanation for Uganda’s current predicament. Perhaps its biggest weakness is its limited policy implications. If being in the tropics is detrimental to growth, what is a country like Uganda, straddling the equator, to do? If life-expectancy of close to 60 years is a sine qua non for rapid economic transformation and it takes decades to get there, what should Uganda do in the interim? Similarly, other fundamentals, such as institutions, exhibit strong persistence. As Avinash Dixit (2005) suggests, it is not clear that such initial endowments type explanations pass the “prospective test of usefulness.” More memorably, he likens them to “‘Irish recipes for success’, after a story about an Irishman who was asked for directions and replied ‘if I wanted to go there, I wouldn’t start here.’”

B. Openness

Work by a number of authors (Sachs and Warner 1995, Olken and Jones 2007 and many others in between) has shown the importance of outward orientation in sustaining economic growth. This is quite intuitive. If industrialization is the kernel of development, it follows that the more competitive the environment that firms face the more efficient they are likely to be. Trade also permits economies of scale in production. Production of non-tradable goods tends to face fairly inelastic demand so that an increase in production tends to cause a decline in prices. Expansion through the production of tradables faces no such constraint, with world markets providing virtually unlimited demand for such goods (Rodrik, 2006). Last but not least, openness facilitates diffusion and absorption of technology. For these reasons, the tradables sector is seen as being “special,” particularly at low-to middle-income levels (Rodrik, 2007).

The performance of Uganda’s tradables sector, as noted above, has been relatively weak. The export to GDP ratio has not improved much over the years. Manufacturing exports has done worse still, only increasingly significantly very recently. Further, the trade and current account balances, the litmus test for the performance of the tradables sector, remain strongly negative. What accounts for this weak performance of the tradable sector? Could this be the reason why economic transformation has been limited in Uganda?

One reason for the large trade and current account deficits is the high level of foreign aid which the country receives.¹³ Between 1996 and 2005 donor support channeled through the government averaged around 10 percent of GDP. The aid allows more outlays on both tradable and nontradable goods, and bids up the price of nontradable whose supply is more limited relative to tradable goods. The result is a more appreciated (less competitive) real exchange rate than would otherwise be the case. This in turn reduces the incentive to engage in the production of tradable goods. This is the infamous Dutch Disease. It imparts a “deindustrializing” influence by causing the relative price of imported manufactured goods to fall.

This contrasts with the competitive real exchange rate which has been found to be necessary for higher economic growth and economic transformation (Rodrik, 2007 and Eichengreen, 2007). Rodrik advances two reasons a competitive exchange rate matters. First, the tradables sector is affected more by weak institutions (typically more prevalent in poor countries) than the nontradables sector. A manufacturing firm has to have many more implicit and explicit contracting arrangements with other economic agents than, say, a building contractor who likely deals with a more limited number of suppliers and clients. Second, the tradable sector may face more market failures than the nontradables sector, including for example coordination externalities to get new activities going. For these reasons (Rodrik, 2007): “developing countries devote too few resources to tradables and grow less rapidly than they should… [a more depreciated real exchange rate] can act as a second best mechanism for spurring tradables and for generating more rapid growth.” A competitive real exchange rate in effect helps subsidize industrialization.

The weak performance of the tradables sector in Uganda is, however, not just (or even mainly) due to Dutch Disease. Higher exports are also likely being constrained by infrastructure deficiencies and limited human capital. With, say, better roads to regional and international markets, transaction costs would be lower and more industries would likely become more competitive. Thus, a more depreciated real exchange rate is not the only way that a government can bolster the tradable sector. Other factors constraining a stronger supply response also have to be addressed. And, of course, the aid that Uganda receives is intended to help alleviate these constraints.

The government thus faces a delicate balancing act. On the one hand, aid inflows engender a more appreciated real exchange rate. On the other, these resources are needed to build the schools, clinics, roads etc. that are necessary for a stronger supply response. Given that fairly strong economic growth that has been recorded over the last 20 years, it is not clear that the adverse effects of aid have (economically at least) outweighed its beneficial effects.¹⁴ As noted above, there is little evidence of overvaluation at the moment—though equally no undervaluation so as to “subsidize” the tradable sector because of the particular obstacles that it faces. Export growth in the last few years has also been strong.

Against this backdrop, the government’s stated policy seems appropriate. It calls for a reorientation of spending towards infrastructure projects coupled with a gradual reduction of the fiscal deficit over the medium-term in order to ameliorate the adverse effects of aid inflows on the tradables sector (Brownbridge and Tumusiime-Mutebile, 2007). By increasing public savings, this should help ease real exchange rate appreciation pressures.

All in all, evidence of an economically adverse effects of aid on economic activity are difficult to observe in Uganda (see also Nkusu, 2004).¹⁵ But it will of course never be known if in the absence of aid higher growth and more industrialization would have been possible. Given the historic weakness of the tradable sector it will be important to keep a close eye on the measures of competitive, and a closer eye still on export growth.¹⁶

C. Policies

Another argument as to why Uganda has not enjoyed higher still growth and economic transformation is the view that the country got the balance between the state and markets wrong. In this view, the premise for pro-market reforms—that markets are a more efficient allocator of scarce resources than states—is mistaken. Market allocation does not dominate state allocation in the absence of the institutions and infrastructure that are needed to make markets function smoothly. And these features are largely missing in Africa. Consequently, it is not surprising that the strongly pro-market policy frameworks pursued by countries like Uganda have not engendered enough economic transformation and growth.¹⁷ The approach is instead seen as having caused two problems.

First, it has led to the weakening of the African state when the state should be the cornerstone of development. As Mukandawire (2000) puts it “the African state today is the most demonized social institution in Africa.” And this is unfortunate because the state in Africa was never very large to begin with. While many African states distorted prices and these policies needed to cease, the extensive liberalization and privatization that then followed was neither necessary nor helpful. The main legacy of the pro-market reforms has been to engender extremely fragile economies.

Second, in the absence of strong institutions and adequate infrastructure, liberalization has simply allowed inefficient markets to replace inefficient states. Take the case of the financial sector. Ethiopia’s prime minister, Meles Zenawi, argues that the limited financial deepening post liberalization in Africa partly reflects missing institutions such as credit rating bureaus, but more so high real lending rates due to wide intermediation spreads as well as high policy rates. “One of the objectives of the reform program in Africa was to undo financial repression and it has succeeded admirably. The consequence [has been] to depress investment, reduce demand for credit and generate excess liquidity” (Zenawi, forthcoming).¹⁸

In this view, the government needs to play a more central role in the economy—much along the lines of the “developmental states” in places like Korea and Taiwan. Government interventions in these countries were after all highly effective. In their early stages of take-off, most of these countries maintained interest and foreign exchange controls, provided subsidies and cheap credit to preferred sectors, and offered trade protection. Why shouldn’t African countries rely on some of these proven approaches now? Earlier interventions in Africa failed because violence and economic decline had reduced state capacity. But with peace and economic expansion, state capacity has recovered.

There are however a number of reasons why this explanation is not satisfactory:

First, growth in Uganda (and indeed much the rest of Africa) in the pro-market era has been much better than in the earlier state-led period. Of course, this can not be entirely attributed to economic reforms. The lessening of armed conflict, move towards democracy, better institutions have all also played a role. Still, it is difficult to fault a framework which has delivered 20 years of uninterrupted growth for Uganda. No doubt there are some areas where the market outcomes can be improved upon, but by taking actions to address the relevant distortions it should be possible to engender higher still growth.

Second, maligned it might have been, but the size of the state in Uganda and Sub-Saharan Africa more generally has not declined over the years (Figure 22). In fact, in the case of Uganda, the ratio of government spending to GDP has been trending upwards since the late-1980s when the government adopted IMF and World Bank supported reform programs in earnest, showing signs of plateauing only since around 2000. If a “night watchman state” is what these reforms were after they have not realized it. A more nuanced criticism of the pro-market reforms, is that while they did not lead to a reduction in the size of the state they have made it grow in all the wrong directions—ministries of health and education have expanded instead of more direct growth promoting agencies such as ministries of trade and industry—and regulatory structures have not been strengthened to keep an eye on markets. There could be something to the point on needing to strengthen regulatory agencies, but the increased investment in human capital which account for a large part of the spending increase can certainly not be faulted for the reasons discussed above.

Third, in terms of outcomes, it is not clear that a more state-led development model as practiced in, say, Ethiopia has been all that superior to Uganda’s. Average per capita income growth between 1995 and 2005 was 3 percent in Uganda versus 2 percent in Ethiopia (Figure 12). Growth in Uganda has also been much less volatile. Private investment and domestic savings are also doing better in Uganda. One area where Ethiopia outstrips Uganda though is export performance, which from a low base has improved sharply in Ethiopia in recent years. So Uganda’s macroeconomic outcomes are overall better, but Ethiopia’s growth transition happened later it could yet catch-up and it has moved ahead in export growth. On balance, the jury is still out.

A related point is that development strategies have to be time and context specific. A “developmental-state” approach may work in Ethiopia, where state capacity has historically been strong, but it is less clear if this would work in Uganda. A look at tax collections—a reasonable proxy for state capacity (Besley, 2007)—is instructive. Tax revenue collection accounts for some 16½ percent of GDP in Ethiopia at present versus 13½ percent in Uganda, despite the former’s lower level of per capita income and larger (and harder to tax) agricultural sector. At the height of civil unrest in Ethiopia (1990), the government was collecting some 10 percent of GDP, compared to about 5 percent at the same proximate point (1986) in Uganda. With state capacity having atrophied so much, it would have been foolhardy for Uganda to have opted for a state-led development strategy.

Lastly, it is unclear if the Asian development model, which may have been appropriate for the technology of the 1960s and 1970s, is still relevant for African countries in the early 21^st Century. The incredible speed with which information flows makes it difficult to continue segmenting markets. Foreign exchange and interest rate controls can be circumvented via swaps, derivatives, and related instruments. In this brave new world, one can’t help but think that state control is not as easy or at least only achievable at high cost.

In sum, it is difficult to conclude that Uganda got the balance between markets and the state wrong. The right balance between the two is also a subjective matter. And given the strong outcomes that the current policy framework has yielded, the country seems to have struck the right balance. But is this as far as the current policy framework can take Uganda? Looking ahead, is more intervention the answer? We return to this issue below.

D. Globalization

Another factor that is increasingly mentioned as having a bearing on Africa’s economic prospects is the rise of China. The typical refrain is that China’s rapid growth and immense appetite for commodities is going to further push-up their prices and will be a boon for most African countries who are heavily reliant on such exports. This will serve to boost domestic investment and consumption, and is being supplemented by increased foreign direct investment (particularly in the extractive industries) as well as aid from China. And indeed the recent pick-up in growth across much of Africa, including Uganda, probably has something to do with the surge in commodity prices.

But less closely discussed are potentially adverse implications of China’s rise for Africa. To the extent China is using a depreciated real exchange rate as a development strategy to facilitate the transfer of employment from agriculture to manufacturing, one side effect of this would be to delay its progress along the technology ladder (from textiles, to toys, to tractors and so on to increasingly more sophisticated manufactured goods).¹⁹ This in turn could be closing off avenues for advancement for African countries which would be the next natural place for activities at the low end of the technology ladder to migrate.

Relatedly, and more bleakly still, Paul Collier (2006) has argued that Africa may have “missed the boat” in terms of industrialization and export orientation:

“policy mistakes happened to occur at precisely the critical time when Africa could have otherwise broken in on level terms with Asia. Now Asia has huge agglomeration advantages and so [having the right policies] is not enough. When will Africa be able to repeat Asia’s success? … the logic of the new economic geography is that Africa will have to wait until the wage gap between Africa and Asia is approximately as wide as that between OECD and Asia at the time when Asia broke into OECD markets. If this is approximately right then Africa will have to wait for several decades.”

Chamon and Kremer’s (2006) work suggests whether this scenario plays out may hinge on a race between African population growth and China and India’s economic growth. If the latter dominates, global growth would accelerate; otherwise, the outlook for the global economy, and poor countries could be bleak. But perhaps one of the more insightful parts of their paper is their observation “country characteristics that lead to poor performance today may well allow for rapid growth in the future if and when the world economy reaches a sufficiently advanced stage.”

Context thus matters. The timing and pace of Uganda’s industrialization may well be driven as much by circumstances beyond its control as the appropriateness of policies, institutions, and better infrastructure. This, though, does not imply that there is nothing that Uganda can do. Rather, in the global market place that Uganda finds itself in, it has no option but to compete. And this competition will, as Dollar (2007) has observed, likely be at the micro level: “which locations can create good environments for firms to start-up, access finance, find skilled labor, and connect to the global market.”

Overall, it is difficult to discern what, if any, effect China’s super competitiveness has had on welfare in Uganda. In all likelihood, the effect has been mixed. The adverse implications have been considered above. But on the plus side, Chinese companies have lowered the price of many consumer and investment goods. The current improvement in the terms of trade for commodity producers, largely on account of China’s blistering pace of economic expansion, has also been a boon to Uganda.

V. What next?

Ultimately, economic policies have to be judged by their results. And Uganda’s achievements over the last 20 years, against the backdrop of weak fundamentals and significant exogenous shocks, are remarkable. Moreover, the outlook remains good. This is evident from the high levels of private investment that the current policy framework is continuing to engender. Private investment has moved from being one of the lowest in sub- Saharan Africa in 1990 to well above the average for the region in 2005 (Figure 24). Export diversification has also been impressive—the share of non traditional items in total exports now stands at some 60 percent, compared to 20 percent of total exports in 1995. The current anxiety about the country’s economic prospects is thus a bit over done. Accordingly, any changes to the current economic framework have to be considered very carefully so as not to undermine the hard won gains. Reforms should be complimentary to the current robust and generally successful framework.

There is nonetheless scope for improvement. The weak point of Uganda’s recent economic performance is the fact that economic transformation has been limited—relative to the very high benchmarks set by the sustained growth countries. And it is right that Uganda should be measuring itself relative to the high standards set by these countries.

But as the discussion above has sought to highlight, there is no single cause for why there has not been more industrialization in Uganda and so it is unlikely that there is a single solution. There are “hardware” type problems constraining economic transformation, which include limited human capital and inadequate infrastructure. But “software” problems are also at work: the quality of institutions, governance, regulatory framework etc. could all be better. Last but not least, developments beyond Uganda’s borders—the global economic environment and, closer to home, peace and economic growth in Sudan and DRC, as well as better access to the sea via Kenya and Tanzania—also have a bearing on Uganda’s economic performance. Given the range of problems that need to be fixed and limited government capacity and resources to address these problems, the interventions have to be focused and limited. Three first-order interventions are identified here.

Infrastructure

The government is right to stress the importance of industrialization and economic transformation. But the main intervention needed to facilitate this is improvements in the country’s infrastructure. Cheap and effective infrastructure is a pre-requisite for industrialization.

And there are three important reasons for focusing on infrastructure services. First, industry and services in general use infrastructure services more intensively than agriculture. Hence, poor infrastructure tilts the comparative advantage away from such activities (Eifert et al, 2007). Second, infrastructure services are one of the inputs in the production process which enterprises find it near impossible to substitute away from. If labor is expensive, companies can opt for capital intensive technologies. And where capital is scarce, labor intensive methods offer an alternative. But it is difficult to, say, do without electricity. So when there is an interruption in its supply, firms often have little alternative to but to rely on back-up generators and the like, with the attendant rise in indirect costs. Third, such infrastructure services help expand markets and facilitate trade and production—which is perhaps the ultimate development objective.²⁰

The government has increased spending to address the infrastructure constraint in the last couple of budgets, but there remain questions about whether the current policy framework for railway and energy infrastructure, in particular, are conducive to significant additional public investment. The expectation in the energy sector, for example, is that the private generation companies will provide the required additional capacity. It has, however, taken close to 7 years since the inception of the Bujagali dam project for construction to begin. And even then at significant cost and with considerable government inducement. It also remains to be seen if there is enough private sector capacity to do the many other similar projects that are needed to improve electricity coverage. In a similar vein, the single railway line that links Uganda to the sea is now more than 100 years old. How or when a more modern line capable of accommodating higher capacity freight traffic will be built is unclear.

Growth strategy

Chastened by the ruinous results of extensive state involvement in the economy in the pre-reform area as well as the limited pay-off to more recent interventions, there has not been much effort in Uganda to identify and provide coordinated government support to the 3 or 4 sectors through which the country can modernize. And in the absence of such a strategy around which government policies could coalesce, interventions have been diffuse and uncoordinated. In recent years, tax breaks and/or government financial support have been provided to activities as varied as the hotel and tourism sector, hides and skins, textiles, palm oil production, micro-finance etc. These have not been very successful.

A more focused growth strategy is needed instead. Uganda of course has a Poverty Eradication Action Plan (PEAP), which is meant to guide the country’s economic policies. But this document, as Lant Pritchett (2005) suggests, is akin to a menu rather than a meal. It does not make policy trade-offs, but rather emphasizes all of the issues that the country needs to address. What is needed is a more focused growth-cum-industrialization strategy, with a focus on sectors rather than specific enterprises. All parts of the government could then swing their full support behind this strategy.

Whatever support is provided to the identified sectors should nonetheless be limited, both in terms of resources and time horizon. We bureaucrats continue to have very limited ability to distinguish between winners and losers. The recent pick-up of manufactured exports to Uganda’s neighbors illustrates this point nicely. In recent years, the government has devoted considerable attention and resource to promote textile exports to the United States (under the African Growth and Opportunity Act) and Europe (Everything But Arms initiative). The objective of these interventions has been to promote exports of valued added products and as a result of these initiatives, textile exports now amount to around US$3 million per annum. But much more impressive has been the increase in manufactured exports to regional markets in Rwanda, Sudan and DRC—without much active government intervention. These have risen from close to nothing in the late-1990s when the region was racked by civil conflict to well over US$20 million per month in 2006. With better roads and other transportation links, even more exports would likely be these countries and further afield.

To be sure, there is empirical that shows that it is not just the level of exports that matter but what you export and to whom you export are also important (Hwang, 2007). Exporting higher value added goods to more developed markets requires local producers to raise the quality of their products. And this move up the technology and quality ladder is among the intangible benefits of trade that contributes to economic development. But with the volume of manufactured exports in Uganda still limited, the emphasis should probably be in raising the overall volumes rather than the type of exports. This again implies that the emphasis should be improved infrastructure services (adequate electricity, road, railway, etc networks) followed by time-bound and financially-limited government support at the sectoral rather enterprise level.

Export competitiveness

The other important challenge for Uganda will be to sustain the rapid export growth of the last few years. The recent pick-up in export performance is certainly good news. But this is very slender evidence to draw much comfort from. The problem remains that the improvement is starting from a very low base. Better infrastructure will certainly be important to sustain export growth. But while this is being addressed is there a role for policies? Perhaps a tax break for exporters? Alternatively, a more competitive real exchange rate?

Any such intervention to help the tradable sector, however, would be subject to trade-offs:

If we take tax breaks for exporters, this will entail some revenue loss. Unless compensated for through other measures, this will result in a wider fiscal deficit. And the lower domestic savings that this implies would be contrary to what is needed to sustain a more depreciated real exchange rate. Note also that the tax breaks would benefit exporters, but not import competing industries.
As for having the Bank of Uganda aim for a more depreciated nominal exchange rate, this could only work in the short run. Unless it is supported by a lower fiscal deficit—i.e. higher domestic savings—the nominal depreciation will not stick. It will instead be dissipated by higher inflation. And of course the need for a tighter fiscal stance has to be weighed against the large spending needs that the government faces from education to health to roads.

The “structural adjustment” era in Africa started with the need for tighter fiscal policies being emphasized to help reduce macroeconomic imbalances. With macro stability at hand, the more recent avatar to address the development challenge has been “fiscal space,” and mainly with reference to higher aid financed fiscal deficits. But to some degree this runs counter to what is needed to sustain a competitive tradables sector. The latter requires a competitive real exchange rate, which in turn needs a tighter fiscal stance. Prudent fiscal positions thus remain a very relevant variable for economic development. The presence and importance of this policy tension varies from country to country, but is likely present in most African countries. The extent to which countries strike the right balance between the “development” and “competitiveness” needs of their economies, as it were, will have an important bearing on the pace of industrialization. The debate on this in Uganda started early-on, and has been resolved in favor a gradual fiscal deficit reduction. This seems like the right choice for the country.

VI. Conclusions

All told, Uganda’s growth performance over the last 20 years has been very impressive. But policy changes are needed to improve the quality of growth. Perhaps the most important of these adjustments is an aggressive focus strengthen the country’s infrastructure—starting with roads and electricity. This has not been lost on policy makers. The increased allocation towards increased infrastructure spending observed in the last few is welcome, but much more needs to be done. Second, a focused economic growth strategy needs to be developed. Lastly, export competitiveness needs to be enhanced, including by affecting the targeted gradual medium-term fiscal consolidation.

REFERENCES

Acemoglu, Daron, and Simon Johnson, 2006. “Disease and Development: The Effect of Life Expectancy on Economic Growth,” NBER Working Papers 12269.
- Search Google Scholar
- Export Citation
Acemoglu, Daron, Simon Johnson and James A. Robinson, 2001, “The Colonial Origins of Comparative Development: An Empirical Investigation,” American Economic Review, Vol. 91, No. 5 (December), pp. 1369–1401.
- Search Google Scholar
- Export Citation
Aizenman, Joshua and Mark M. Spiegel, 2007. “Takeoffs,” NBER Working Papers 13084.
- Search Google Scholar
- Export Citation
Berg, Andrew, Jonathan D. Ostry, and Jeromin Zettlemeyer, 2006, “What Makes Growth Sustained?” IMF Working Paper 08/59.
- Search Google Scholar
- Export Citation
Besley, Timothy and Torsten Persson, 2007. “The Origins of State Capacity: Property Rights, Taxation, and Politics,” NBER Working Papers 13028.
- Search Google Scholar
- Export Citation
Blattman, Christopher. Jason Hwang, and Jeffery G. Williamson, 2006, “Winners and Losers in the Commodity Lottery: The Impact of Terms of Trade Growth and Volatility in the Periphery, 1870-1939,” forthcoming, Journal of Development Economics
- Search Google Scholar
- Export Citation
Bloom, David, David Canning and Jaypee Sevilla, 2003, “The Demographic Dividend: A New Perspective on the Economic Consequences of Population Change,” Rand Publications.
- Search Google Scholar
- Export Citation
Bloom, David, David Canning and Günther Fink & Jocelyn Finlay, 2007. “Realizing the Demographic Dividend: Is Africa any different?” PGDA Working Papers 2307, Program on the Global Demography of Aging.
- Search Google Scholar
- Export Citation
Brownbridge, Martin and Emmanuel Tumusiime-Mutebile, 2007, “Aid and Fiscal Deficits: Lessons from Uganda on the Implications for Macroeconomic Management and Fiscal Sustainability” Development Policy Review, Vol. 25 No. 2 (March), pp. 193-213.
- Search Google Scholar
- Export Citation
Chamon, Marcos, and Michael Kremer, 2006, “Asian Growth and African Development” American Economic Association Papers and Proceeding, May 2006, Vol 96(2), pp. 400-404.
- Search Google Scholar
- Export Citation
Clemens, Michael, Charles Kenny, and Todd J. Moss (2007), “The trouble with the MDGs: Confronting expectations of aid and development success”, World Development, 35(5): 735-751.
- Search Google Scholar
- Export Citation
Collier, Paul, 2007, “Growth Strategies for Africa”, paper prepared for the Spence Commission on Economic Growth, Centre for the Study of African Economies, Oxford University.
- Search Google Scholar
- Export Citation
Collier, Paul and Stephen O’Connell, 2006, “Opportunities and Choice”, Chapter 2, in Political Economy of Economic Growth in Africa, 1960 – 2000, Volume 1, Cambridge University Press.
- Search Google Scholar
- Export Citation
Dixit, Avinash, 2006. “Evaluating recipes for development success,” Policy Research Working Paper Series 3859, The World Bank
- Search Google Scholar
- Export Citation
Dollar, David, 2007. “Asian century or multi-polar century?” Policy Research Working Paper Series 4174, The World Bank.
- Search Google Scholar
- Export Citation
Easterly, William and Ross Levine, 2002. “Tropics, Germs, and Crops: How Endowments Influence Economic Development,” NBER Working Papers 9106.
- Search Google Scholar
- Export Citation
Easterly, W., 2006, “Reliving the 50s: the Big Push, Poverty Traps, and Takeoffs in Economic Development” Journal of Economic Growth, Vol. 11(4), pages 289-318.
- Search Google Scholar
- Export Citation
Easterly, William, 2007, “How the Millennium Development Goals are Unfair to Africa” Professor of Economics, NYU Visiting Scholar, Brookings Institute.
- Search Google Scholar
- Export Citation
Eichengreen, B., 2007, “The real Exchange Rate and Economic Growth,” mimeo, University of Califonia, Berkeley.
- Search Google Scholar
- Export Citation
Eifert, P.B., 2007, “Infrastructure and Market Structure In Least-Developed Countries,” mimeo, University of California, Berkeley.
- Search Google Scholar
- Export Citation
Frankel, Jeffery, 2005. “On the Renminbi: The Choice between Adjustment under a Fixed Exchange Rate and Adjustment under a Flexible Rate,” NBER Working Papers 11274.
- Search Google Scholar
- Export Citation
Glaeser, E.L., R. L. Porta, F, Kioez-de-Silenes and A. Shleifer, 2004, “Do Institutions Cause Growth?
- Search Google Scholar
- Export Citation
Hausmann, Ricardo, Lant Prichett and Dani Rodrik, 2005 “Growth Accelerations,” Journal of Economic Growth, Vol. 10, pp. 303–29.
- Search Google Scholar
- Export Citation
Hausmann, Ricardo, Jason Hwang, and Dani Rodrik, 2006, “What You Export Matters” (unpublished, Kennedy School of Government).
- Search Google Scholar
- Export Citation
Hausmann, Ricardo, Bailey Klinger, and Robert Lawrence, 2008, “Examining Beneficiation,” Center for International Development Working Paper No. 162, Harvard University.
- Search Google Scholar
- Export Citation
Johnson, Simon, Jonathan D. Ostry, and Arvind Subramanian, 2007, “The Prospects for Sustained Growth in Africa: Benchmarking the Constraints”, IMF Working Paper 07/52.
- Search Google Scholar
- Export Citation
Jones, Benjamin F. and Benjamin A., Olken, 2005, “The Anatomy of Start-Stop Growth,” Working Paper 1152 (Cambridge, Massachusetts: NBER).
- Search Google Scholar
- Export Citation
Kanbur, Ravi, 2005, “The Development of Development Thinking,” mimeo, Cornell University.
- Search Google Scholar
- Export Citation
Lindauer, David and Lant Pritchett. “What’s the Big Idea? Three Generations of Development Advice” Economia (2002).
- Search Google Scholar
- Export Citation
Lucas, Robert E., 2007, “Trade and the Diffusion of the Industrial Revolution”, National Bureau of Economic Research.
- Search Google Scholar
- Export Citation
Mkandawire, Thandika, 2005, “Maladjusted African Economies and Globalisation,” Africa Development, Vol. XXX, Nos. 1 & 2, 2005, pp. 1–33
- Search Google Scholar
- Export Citation
Murphy, Kevin, Andrei Shleifer and Robert Vishny, 1989. “Industrialization and the Big Push,” Journal of Political Economy, University of Chicago Press, vol. 97(5), pages 1003-26, October.
- Search Google Scholar
- Export Citation
Mwenda, Andrew, 2006, “Foreign Aid and the Weakening of Democratic Accountability in Uganda,” Cato Institute Foreign Policy Briefing No. 88.
- Search Google Scholar
- Export Citation
Nkusu, Mwanza, 2004. “Financing Uganda’s Poverty Reduction Strategy: Is Aid Causing More Pain Than Gain?” IMF Working Paper 04/170
- Search Google Scholar
- Export Citation
Nunn, Nathan & Puga, Diego, 2007. “Ruggedness: The Blessing of Bad Geography in Africa,” CEPR Discussion Papers 6253.
- Search Google Scholar
- Export Citation
Prasad, Eswar, Raghuram Rajan, and Arvind Subramanian, 2007, “Foreign Capital and Economic Growth,” Brookings Papers on Economic Activity 1, pp. 153–209.
- Search Google Scholar
- Export Citation
Pritchett, Lant, 2000, “Understanding Patterns of Economic Growth: Searching for Hills Among Plateaus, Mountains, and Plains,” World Bank Economic Review, pp. 221–50.
- Search Google Scholar
- Export Citation
Pritchett, Lant, 2007, “Does Learning to Add-up Add-up? The Returns to Schooling in Aggregate Data,” mimeo, Kennedy School of Government, Harvard University.
- Search Google Scholar
- Export Citation
Rajan, Raghuram, and Arvind Subramanian, 2005, “What Undermines Aid’s Impact on Growth? IMF Working Paper No. No. 126 (Washington: International Monetary Fund).
- Search Google Scholar
- Export Citation
Rodrik, Dani, Arvind Subramanian and Francesco Trebbi, 2004, “Institutions Rule: The Primacy of Institutions over Geography and Integration in Economic Development.” Journal of Economic Growth 9.2 (June), pp. 131–65.
- Search Google Scholar
- Export Citation
Rodrik, Dani, 2007, The Real Exchange Rate and Economic Growth: Theory and Evidence, Kennedy School of Government, Harvard University.
- Search Google Scholar
- Export Citation
Rodrik, Dani, 2007, Industrial Development: Stylized Facts and Policies, Kennedy School of Government, Harvard University.
- Search Google Scholar
- Export Citation
Rose, Andrew K & Supaat, Saktiandi, 2007, “Fertility and the Real Exchange Rate,” CEPR Discussion Papers 6312.
- Search Google Scholar
- Export Citation
Roy, Devesh and Arvind Subramanian, 2001, “Who Can Explain the Mauritian Miracle: Meade, Romer, Sachs, or Rodrik?” IMF Working Papers 01/116.
- Search Google Scholar
- Export Citation
Sachs, Jeffery, and Andrew Warner, 1995, Economic Reform and the Process of Global Integration, in Brooking Papers on Economic Activity, 1995, Volume 26, Issue 1995-1, pp. 1-118.
- Search Google Scholar
- Export Citation
Subramanian, Arvind, 2007, The Evolution of Institutions in India and its Relationship with Economic Growth, mimeo, Peterson Institute for International Economics and Center for Global.
- Search Google Scholar
- Export Citation
Wacziarg, Romain, and Karen Horn Welch, 2003, “Trade Liberalization and Growth: New Evidence,” NBER Working Papers 10152.
- Search Google Scholar
- Export Citation
Wood, Adrian and Kate Jordan, 2000. “Why Does Zimbabwe Export Manufactures and Uganda Not? Econometrics Meets History,” The Journal of Development Studies, Taylor and Francis Journals, vol. 37(2), pages 91-116, December.
- Search Google Scholar
- Export Citation
Zettlemeyer, J., 2006, “Growth and Reforms in Latin America: A survey of Facts and Arguments,” IMF Working Paper 06/210.
- Search Google Scholar
- Export Citation

My thanks, without implication, to Michael Atingi-Ego, Robert Corker, Peter Doyle, Dmitry Gershensen, John Green, Louis Kaskende, Steve Kayizzi-Mugerwa, Damoni Kitabire, Mark Plant, and Manrique Saenz for helpful comments on earlier drafts. All remaining errors are mine.

Adam Smith’s turn of phrase.

Accelerations are defined as episodes which satisfy the following conditions over an eight year horizon: per capita growth of greater than 3½ percent; an acceleration in the per-capita growth rate of 2 percentage points or more: and post acceleration output level has to exceed pre-acceleration output level.

The sustained growth (SG) sample and starting dates of growth episodes used in this note are those identified by Johnson, Ostry and Subramanian (2007) and comprise: Chile (1986), China (1978), Dominican Republic (1969), Egypt (1976), Indonesia (1967), Korea (1962), Malaysia (1970), Singapore (1969), Taiwan Province of China (1961), Thailand (1960), Tunisia (1968), and Vietnam (1985).

One point of departure from Johnson et al. is to assign the start of Uganda’s growth episode to its appropriate starting point of circa 1990; they treat growth episodes of the African countries in their sample as having started in the late 1990s.

The first year when positive growth was recorded is actually 1987. But both the economic and statistical filters used by the various authors ensure that the trend is sustained so do not always pick the first year of positive growth as a take-off point

Subramanian (2007) goes as far as to argue “…manufacturing growth is a concomitant, perhaps even a sine qua non of, overall, growth.”

Recent papers in this vein include: Wacziarg and Welch (2003), Hausmann, Pritchett, and Rodrik (2004), Olken and Jones (2005), Easterly (2006), Johnson, Ostry and Subramanian (2007), Berg, Ostry and Zettlemeyer (2007) and Aizenman and Spiegel, 2007.

As with much of the empirics in this paper, the aim is not so much to establish causality but to see if the various bivariate (and occasional multivariate) associations are consistent with the predictions of theory.

Various measures of institutions (instrumented by settler mortality data from Acemoglu, Johnson and Robinson, 2001) were used in these regressions but the results are broadly similar.

Beyond per capita income, infant mortality rates are also influenced by factors such as the distribution of income, maternal education etc. For this exercise, we control for disease burden using data on share of territory affected by malaria, a major cause of mortality in sub-Saharan Africa, as a proxy.

The World Bank governance indicators database has six indicators—voice and accountability, political stability and absence of violence, government effectiveness, regulatory quality, rule of law, and control of corruption. Uganda generally ranks somewhat better than its per capita income level would predict on all of these dimensions, except for political the measure on political stability.

We identify aid here because the volumes involved are so large and the influence on both quantities (savings and investment) and prices (the real exchange rate) seems fairly direct. But other, more subtle and to date only weakly explored factors such as fertility may also be influencing the performance of the tradables sector (Rose, Suppat, and Braude, 2008). In particular, high fertility rates (and thus a higher dependency ratio) generally imply higher lower savings rates as well as greater outlays on nontradables (health and education).

There is a growing body of literature on the deleterious effect of aid on governance, by weakening the social contract between rulers and the electorate (see Rajan and Subramanian, 2005 and Mwenda, 2006). But this is beyond the scope of this paper.

As Nkusu argues, large private capital inflows can also have the same deleterious effect as aid. And exchange rate appreciation pressures in recent months have mainly been due to such inflows.

The foregoing discussion raises an interesting question: has the option of an export-led development strategy (the Asia model, as it were) been closed off for African countries because of their reliance on foreign aid inflows?

Note that these arguments are not an appeal to African exceptionalism. Very similar concerns have also been expressed with respect to Latin America’s experience with reform (Stiglitz, 2003). See also Zettlemeyer, 2007.

Extracts from the monograph can be found at http://www0.gsb.columbia.edu/ipd/pub/Meles-Extracts2-AfTF2.pdf.

And much economic transformation remains in China for it to want to keep its exchange rate competitive for some time to come. The share of the population in rural areas is still very high at around 60 percent.

Another argument for better infrastructure services is that it may be constraining growth (World Bank, 2006). But with economic growth having averaged some 9 percent during 2005-07, it is difficult to think of growth having been constrained.

Other Publishers

Save
Cite
Email this content

Share Link

Copy this link, or click below to email it to a friend
Email this content
or copy the link directly:

https://www.elibrary.imf.org/view/journals/001/2008/231/article-A001-en.xml
The link was not copied. Your current browser may not support copying via this button.

Link copied successfully

Beyond Macroeconomic Stability: The Quest for Industrialization in Uganda

Author: Mr. Abebe Aemro Selassie

Volume/Issue:: Volume 2008: Issue 231

Publisher:: International Monetary Fund

ISBN:: 9781451870893

ISSN:: 1018-5941

Pages:: 43

DOI:: https://doi.org/10.5089/9781451870893.001

Keywords
I. Introduction and Summary
II. The Stylized Facts of Uganda’s Recovery
III. Economic Transformation
IV. Explaining the record
V. What next?
VI. Conclusions
REFERENCES

View raw image

Uganda: Per Capita Real GDP Index, 1990 = 100

View raw image

Per Capita Real GDP Growth

(percent)

View raw image

Average Real Per Capita GDP Growth in Developing Countries, 1990 - 2007

(percent)

View raw image

Economic growth and poverty reduction

View raw image

Per Capita Income and Growth

View raw image

Sectoral Composition of GDP, 1990/91 and 2005/06

View raw image

Share of Industry and Services in Total Output

View raw image

Share of Urban Population

(percent)

View raw image

Indicators of Financial Deepening

View raw image

Savings and Investment

View raw image

Export Performance

View raw image

Trade Intensity in 1990

View raw image

Trade Intensity in 2005

View raw image

Overvaluation

View raw image

Real Exchange Rate and Per Capita Income 1982 - 2005

View raw image

Dependency Ratio

View raw image

Trade and Current Account Balances

View raw image

Life Expectency

(years)

View raw image

Institutions and Income in 1990

View raw image

Institutions and Income in 2005

View raw image

Infant Mortality and Per Capita Income

View raw image

Government Spending

View raw image

Selected Indicators for Uganda and Ethiopia

View raw image

Private Investment in 1990 and 2005

dependant variable: log of GDP
Explanatory Variables:		1990	2005
	Institutions	0.05 (8.01)***	0.10 (9.09)***
	Openness	-0.00 (-1.35)	-0.00 (-1.57)
	Geography	-0.16 (2.19)**	-1.03 (-1.30)
	Uganda dummy	-0.00 (-0.00)	-0.62 (-0.96)
	R-Square	0.57	0.67
	No of Observations	59	57

dependant variable: log of GDP
Explanatory Variables:		1990	2005
	Institutions	0.05 (8.01)***	0.10 (9.09)***
	Openness	-0.00 (-1.35)	-0.00 (-1.57)
	Geography	-0.16 (2.19)**	-1.03 (-1.30)
	Uganda dummy	-0.00 (-0.00)	-0.62 (-0.96)
	R-Square	0.57	0.67
	No of Observations	59	57

dependant variable: log of GDP
Explanatory Variables:		1990	2005
	Institutions	0.05 (8.01)***	0.10 (9.09)***
	Openness	-0.00 (-1.35)	-0.00 (-1.57)
	Geography	-0.16 (2.19)**	-1.03 (-1.30)
	Uganda dummy	-0.00 (-0.00)	-0.62 (-0.96)
	R-Square	0.57	0.67
	No of Observations	59	57

Topics

Countries

Series

About

Resources

Contributor Notes

Abstract

I. Introduction and Summary

II. The Stylized Facts of Uganda’s Recovery

III. Economic Transformation

IV. Explaining the record

A. Fundamentals

Geography and human capital

Institutions

B. Openness

C. Policies

D. Globalization

V. What next?

Infrastructure

Growth strategy

Export competitiveness

VI. Conclusions

REFERENCES

Same Series

Other IMF Content

Other Publishers

Asian Development Bank

Inter-American Development Bank

The World Bank

Share Link

Table of Contents

Headings

Figures

Metrics