Cambodia: Rapid Growth with Institutional Constraints

This paper examines Cambodia’s socioeconomic development since the early 1990s peace settlement. The country’s economic growth has arguably been the fastest among post-conflict societies, driven by the credible restoration of peace and security, large public and private capital inflows, economic openness, reasonably prudent macroeconomic management, and a dynamic, integrating neighborhood. A legacy of history and small size is that the government has limited policy space, although this has not necessarily retarded economic development. We also highlight some key challenges, including rising inequality, uneven spatial development, weak institutions, and high levels of corruption. Looking forward, we highlight the importance of strengthening supply side capabilities, broadening the benefits of growth, and developing stronger institutions and property rights.


EXECUTIVE SUMMARY
This paper examines Cambodia's socioeconomic development since the early 1990s peace settlement. The country's economic growth has been very rapid, especially over the period 1998-2007, when per capita GDP doubled. Its growth rate has arguably been faster than that of any other post-conflict society. It has also been the first period of sustained economic expansion in the country's recorded history. At this rate of progress, Cambodia will very shortly graduate from the 'least developed' group of countries. Within a decade or so, it can be expected to enter the middle-income developing group, an outcome that would have been unthinkable in the 1990s.
Several factors explain this rapid growth. The first has been the restoration of peace and security, enabling normal economic life to resume after almost 3 decades of conflict, including the terrible rule of the Khmer Rouge, 1975Rouge, -1979 it is estimated that about onequarter of the population either perished or left the country. Second, Cambodia has opened up to the regional and international economy, by choice, and also recognizing the reality of its porous international boundaries. Third, there have been very large capital inflows, both in the form of official development assistance, especially in the 1990s, and increasingly private capital. These flows have been motivated by both international generosity and commercial opportunity. A fourth factor has been largely prudent and stable macroeconomic management, resulting in reasonably low inflation rates. Finally, underpinning these developments has been a positive neighborhood effect, as Cambodia has been pulled along by, increasingly integrated with, and benefited from observing a dynamic region.
Reflecting the country's history and small size, and also as a result of deliberate government decisions, Cambodia has somewhat limited policy space. It is a heavily dollarized economy, with the United States (US) dollar and the national currency, the riel, freely circulating and interchangeable. The result is that the National Bank of Cambodia has little scope for monetary and exchange rate policy. The government in effect loses a 'policy lever.' For example, when the region experiences a negative external shock, as in 2008-2009, currencies would be expected to depreciate. Cambodia does not have this option, and its dollar-based cost structure therefore increases relative to countries with their own currency. The government also foregoes a small amount of revenue through the loss of seigniorage. Nevertheless, there are benefits of dollarization, particularly the fact that the potential for macroeconomic mistakes is limited. This is an important consideration, as illustrated by Viet Nam's recent macroeconomic difficulties. It is also the case that most small economies do not have monetary policy independence, and many deliberately decide to use the currency of a larger country. The government has adopted a pragmatic approach to dollarization, expressing a wish that the country will de-dollarize over time, as the community's confidence in the country's macroeconomic management increases.
In addition, Cambodia has limited space in its trade and commercial policies and this feature has worked to its advantage. It has signed on to commitments as part of its membership of the Association of Southeast Asian Nations (ASEAN) Free Trade Area (AFTA), the prospective ASEAN Economic Community, and also the World Trade Organization.
Moreover, its open borders with efficient, large economies to its east and west limit the scope for trade policy interventions. A dynamic political economy underpins the openness, in the sense that the country has few commercial vested interests to protect. For example, unlike the Lao People's Democratic Republic ( Lao PDR) and Viet Nam it has a very small state enterprise sector.
Not surprisingly, most institutions are weak. The country is more democratic than either the Lao PDR or Viet Nam, but it has had more or less continuous one-party rule for 2 decades. By most indicators, corruption levels are very high, as a result of the concentration of political power, the very low civil salaries, the very large capital inflows, and the weak civil society, the latter in spite of the fact that the country has a very large number of domestic and international nongovernment organizations(NGOs). Land ownership and titling is a particularly serious problem, reflecting the country's troubled past. There are frequent and much publicized allegations of land grabbing, while most of the legal cases center on disputes over land titles. The lack of secure title outside major urban centers has retarded rural development, including the growth of financial markets where land is the most form of collateral.
As a result of the rapid growth, the incidence of poverty has fallen quickly. There have been commensurate improvements in most other social indicators, including literacy, nutrition, and infant and maternal mortality. However, about three-quarters of the population remain poor or 'near poor,' below or marginally above the poverty line. Although the country in aggregate produces a food surplus, the incidence of child malnutrition remains quite high, the result of pockets of severe poverty, particularly in some of the more isolated rural areas. The evidence on inequality is less clear. From very low levels in the 1990s, inequality rose quite quickly, to be expected as the economy opened up but was centered on growth in Phnom Penh, and with the increasingly unequal distribution of land access.
Looking forward, the challenges and the outstanding reform agenda remain formidable. The country has been treated very generously by the international donor community, but aid flows are now declining relatively, and within the next few years they will begin to decline in absolute terms. Perhaps inevitably, the economy is narrowly based, with limited policy levers to deal with the challenges of a volatile regional and global economy. The priority of growth at all costs will need to give way to greater attention to distributional considerations, between households, regions, and particularly the capital, Phnom Penh, and the rest of the country. Key supply side challenges remain, particularly education at all levels and rural infrastructure.

I. INTRODUCTION
Cambodia is a small, poor, transition economy. It was once the center of the thriving Khmer empire that spread well beyond the country's current borders, and for a period centered on the world famous Angkor Wat temples. However, its last century and a half have generally been a troubled period, with French colonization, an uneasy relationship with its much larger neighbors, a reluctant but ultimately deep and tragic enmeshment in the Indochina War over the decade to 1975, and then rule by one of the world's most murderous regimes in recent history, followed by the Vietnamese invasion (or 'liberation,' depending on one's historical perspective). The country has enjoyed periods of comparative peace and prosperity only in the first decade after independence in 1953, and more recently since the 1993 Paris Peace Accords.
Two features of Cambodia's development record are of particular interest, and they constitute the major themes of this paper. First, with the return of peace from 1991, and particularly from 1998, its economy has grown very quickly, arguably faster than any other postconflict country in recent times. There are numerous caveats to be attached to this achievement, but it is nonetheless a remarkable achievement. Second, the country's policy settings are unusual, owing to its history, size, location, and also deliberate policy choices. In particular, it has little scope for monetary and exchange rate policy, owing to the high level of dollarization. It is a very open economy, owing to explicit policy decisions, its porous borders, its location between two large, export-oriented economies, and its membership of AFTA and in the near future the ASEAN Economic Community (AEC). Its fiscal policy levers are also limited. A major challenge for the country is to manage constructively its location, and the attendant large official and private capital flows, with institutions and a political system still very much in their infancy.
Our organization is as follows. Section II outlines the development context, the factors that have shaped the country's economic development and its political economy. These include its complex, bloody history, together with its size, geography, and natural resource endowments. In turn, these contextual factors shape the policy parameters, including weak property rights, limited policy space, and the struggle to establish effective governance and institutions. In Section III, we examine development outcomes, including rapid but volatile economic growth since the 1990s, substantial improvements in living standards, and the uneven distributional outcomes. Section IV sums up. 1

II. THE DEVELOPMENT CONTEXT
Cambodian economic development and its policies have been shaped profoundly by its history. The major history of the country (Chandler 2008) highlights several key features of this history. The first has been the challenge of managing a neighborhood dominated by larger powers. The second was the country's immersion in the Indochina War, and its consequences, including extensive aerial bombardments. Third, there was the awful period of Khmer Rouge (KR) rule. When the Communist Party of Kampuchea (CPK) took control in April 1975, many of the country's institutions were destroyed or overturned. It is estimated that nearly two million Cambodians, or one in four, died of overwork, malnutrition, and misdiagnosed diseases or were executed. When the People's Republic of Kampuchea was established in early 1979, it is 1 All Cambodian statistics should be regarded as approximate. The government apparatus is, understandably, still in the process of being re-established on a nationwide basis. There are also large nonmonetized and informal sectors, some of the latter centered on illegal activities. The first national economic census was conducted in 2011 and its findings are contested.
difficult to imagine more difficult circumstances for a new administration (Naron 2011). The security situation was tenuous, the country was internationally isolated and subject to an embargo, and some of the social indicators were among the very lowest in the world.
History and geography have shaped the country's development trajectory and outcomes in numerous ways. We draw attention here to six. First, Cambodia is a small country, with a gross domestic product (GDP) of just $12 billion, surrounded by much larger economies. Thus, national sovereignty has been a prime consideration of all its leaders, from ancient times. Thailand's economy is about 27 times larger than that of Cambodia's, while Viet Nam's is almost 10 times larger. The country's most efficient port for international shipping is Ho Chi Minh (HCM) City. Its international civil aviation gateways are HCM City and Bangkok. As the backbone of the country's agriculture, the Mekong River is central to the livelihoods of millions of Cambodians, and the country is therefore vulnerable to arbitrary actions of upstream users. An inevitable consequence of expanding Mekong and ASEAN economic integration is ever more porous national boundaries, including transhipment of freight.
Second, notwithstanding the recent rapid growth it is a very poor country, still classified as one of the United Nations (UN) least developed economies. About three-quarters of its population live precariously, below or marginally above the poverty line. Reflecting its troubled history, also, most social indicators are low, especially rural education and health. Unexploded ordinance from past conflicts remains a problem in some of the border regions. However, the country has a food surplus, and therefore aggregate food supplies are adequate. The main nutritional challenges relate to distribution and the effects of climatic factors, particularly drought and the possibility of reduced water flows through the 500-kilometerMekong River corridor.
Third, Cambodia's history is dominated by foreign intervention, colonial domination, and conflict. For several years beginning in 1963, it was a minor link in the so called Pyongyang-Peking-Ha Noi-Phnom Penh-Jakarta axis. Then, in addition to the KR rule and its associated barbarities and deprivation from April 1975 to the country's liberation by occupying Vietnamese troops in early 1979, the decades both before and after the KR rule were also dominated by war and conflict. From the mid-1960s, the Viet Nam War increasingly spilled over into Cambodia, especially in the period 1970-1975 following the military coup and the US border bombings. From 1979 until the 1991 Paris Peace Accord, the country was internationally isolated, with Viet Nam and the former Soviet Union its principal international supporters. Most of the countryside remained insecure, with frequent KR attacks. Following the collapse of the power-sharing arrangements, a brief but bitter civil war among the governing coalition partners also erupted in July 1997, which delayed the country's entry into ASEAN, and its re-entry into the UN (in December 1998). Thus, since around 1860 it is no exaggeration that the country has been at peace and free of conflict and outside domination only for brief periods-for a few years after achieving full economic sovereignty-from 1 January 1955 and since 1998.
A fourth contextual factor is that Cambodia has been highly dependent on foreign aid since the October 1991 Paris Peace Agreement. It was briefly administered by a special UN body, UNTAC (United Nations Transitional Authority in Cambodia) in the transitional period, which in turn led to very large international transfers, a factor also motivated by the international distress at the events of the 1970s. For over a decade after the accords, official development assistance (ODA) was equivalent to more than 10% of GDP, in some years as high as 15%. Foreign direct investment (FDI) has also been very large, overtaking ODA for the first time in 2007 ( Figure 1). Although declining, Cambodia's ODA remains one of the highest percentages in the developing world apart from a number of microstates. The country also has perhaps the dubious distinction of possessing arguably the world's largest number of nongovernment organizations (NGOs)-mostly international-per capita. ODA is likely to continue for years to come, owing to strong support constituencies in the major donor countries, and to the growing big power rivalry in the region, particularly with growing Chinese influence in Indochina. Almost all this aid has been provided on highly concessional terms, either as grants or low-interest, long-term loans. These large ODA flows have had several major implications. One is that, even though the effective value of the ODA is overstated, to the extent that it embodies service provision at expensive international prices, it has meant that the government's domestic tax effort has until recently been weak. There has been a tendency in some government circles to outsource key public sector responsibilities to donors and NGOs. Another implication is that economic policy making, in the context of a very weak and poorly paid government administration, has been complicated by the large foreign presence. Donor activities are often poorly coordinated and make heavy demands on the very few senior, well-educated civil servants fluent in English. Competing priorities and short time horizons among donors not infrequently result in erratic policy priorities, incomplete projects, and a bias towards capital expenditures over maintenance, especially in infrastructure and rural development projects. Moreover, the donor presence has substantial distributional consequences. It has led to an over-concentration of development in the capital, Phnom Penh, owing to the presence of the donors and their expenditures in that city, and also in some cases to their own spending priorities. The donor presence has also shaped the aspirational horizons of young, well-educated Cambodians, attracted to donor salaries that are typically five times or more than those in the public sector.

$ million
Fifth, Cambodia is a highly open economy, owing to its geography, its regional and international commitments, and some deliberate policy choices. Once the country rejoined the regional and international mainstream in 1992, it had little choice but to be open. Moreover, there are few vested interests to protect, and that demand special protection, and unlike neighboring Lao PDR and Viet Nam it has a very small state enterprise sector. The value of the country's merchandise trade relative to GDP is currently about 114%, not far short of Thailand and Viet Nam. The stock of inward FDI is estimated to be equivalent to about 53% of GDP, about midway between the ratios for Thailand and Viet Nam, and a very large figure considering that this FDI is of very recent origin. Its average tariff rate is a relatively high 12.4% but this reflects the fact that the tax structure is underdeveloped and that the country is still substantially reliant on trade taxes. In any case, there is little resort to non-tariff barriers while the informal border trade and various tariff exemptions (e.g., for the export-oriented garments sector) result in a much lower effective tariff rate. Cambodia also has a very open labor market, probably at least as open as that of Thailand, and there are foreign workers across most skill levels.
Cambodia's openness is underpinned by its membership of the World Trade Organization (WTO) and ASEAN. Cambodia joined the latter in 1999 and the former in 2004 (WTO 2011). Another factor facilitating continuing trade openness is the gradually reduced reliance on trade and related taxes. In 2004, customs duties, value-added taxes, excise taxes, and export taxes generated about 70% of government revenue. By 2010, the share had fallen to 56%. 2 The country now has just four tariff lines, at 0%, 7%, 15%, and 35%. There is some tariff escalation with the highest band applying to semi-processed and consumer goods. Import and export procedures have been streamlined. The foreign investment regime is an open one, with few restrictions on sectors and foreign equity shares, relatively simple registration procedures, and generous (perhaps too generous) fiscal incentives.
Sixth, the country's political and institutional features are also heavily shaped by its history. It is a quasi-authoritarian state with weak formal institutions (Hughes and Un 2011). As part of the UN accords, it has the trappings of democracy, with elections at national and local levels held every 5 years, and a reasonably open media. According to the World Bank's world governance indicators (WGI), it ranks in the 25th bottom percentile in terms of voice and accountability. This is well above the other Indochina latecomers, the Lao PDR, Myanmar, and Viet Nam (5.7, 0.9, and 8.5, respectively), and also the People's Republic of China (PRC), and not far below Malaysia and Thailand. The political structure established under the UN accords stipulated a power-sharing arrangement between the two major parties, the Cambodian People's Party (CPP), headed by the current prime minister, Hun Sen, and the royalist Funcinpec, together with a smaller third party named after its founder and leader, the Sam Rainsy Party. Since 1998, Hun Sen and his CPP have out-maneuvered the former coalition partners, and he now serves as the country's sole and increasingly all-powerful prime minister. The country remains a constitutional monarchy, with the King mainly undertaking ceremonial duties.
As would be expected, all comparative surveys highlight Cambodia's weak formal institutional structures, and the low levels of trust and social capital, more generally. In 2011, it was ranked 164th out of 182 jurisdictions in Transparency International's Corruption Perceptions Index within ASEAN, second only to Myanmar (ranked 180th). The WGI placed Cambodia at the 8th percentile in terms of control of corruption, also the lowest in ASEAN apart from the special case of Myanmar. It performs somewhat better on other rankings: 138th out of The country's weak property rights are a major challenge. The quality of both civil and commercial law is poor, and most disputes are settled informally, through direct negotiation or through informal payments to secure the desired outcome. Access to secure land title is a particularly serious issue. As a result of the Khmer Rouge's decision to abolish the private ownership of property and the general upheavals of the 1970s, much of the country's land has insecure or uncertain title. Some claims are based on alleged ownership or customary access from the pre-1975 period. The government has also allocated extensive land access rights, often on a 99-year lease basis, to foreign and domestic investors for plantations, forestry, and mining. Many of these allocations are non-transparent and highly contentious. There is a widespread suspicion that the allocation process is deeply corrupt, and the Cambodian press is replete with allegations of land grabbing. The rights of the local people are supposed to be taken into account, but in reality they are largely ignored according to Naron (2011, p. 29). Land disputes constitute about 60% of the cases brought before the Supreme Court, and reportedly much of the violence in rural areas.
Land is typically the most important collateral on poor agrarian economies and the insecure titles therefore inhibit financial sector development. Banks will generally lend only on the basis of 'hard title'. In practice, this extends only to the greater Phnom Penh region and to other major urban areas such as Battambang and Sihanoukville. The retarded rural financial development has the effect that it is difficult to finance projects such as rice milling, that might otherwise be profitable. The investment climate is also less attractive, again mainly in rural areas, where foreign investors who have been granted the necessary permits in Phnom Penh find that they are unable to undertake their investments. In addition, there are adverse distributional effects, to the extent that land is granted to the politically powerful on highly concessional terms. Admittedly there has been some progress. As noted, titling in the major urban areas has progressed. But the rural problems appear to be intractable. In July 2012, the prime minister attempted to resolve the problem by dispatching student volunteers to rural areas to verify land titling agreements, an initiative thought to be triggered by the ruling party's poor showing in the June 2012 commune elections.

III. DEVELOPMENT OUTCOMES
In this section, we examine Cambodia's development performance, focusing on the period since 1991. We address three major topics. First, there has been very rapid economic growth and structural change. We also highlight the sources of this growth, its narrow base, and its potential vulnerabilities. The second is the significant improvements in living standards, even with the substantial increase in inequality that has occurred since the mid-1990s. The third is the unusual macroeconomic policy framework, including extensive dollarization and very large ODA and private capital inflows.

A. Rapid Growth and Structural Change
Economic growth has been rapid since the early 1990s when the system of national accounts was re-established. Figure 2 shows that growth in most years has exceeded 5% and on some occasions has reached double digits. Although the process of economic normalization is generally dated from 1992 (Naron 2011), the real rapid growth occurred during the period 1998-2007, when per capita GDP doubled. This was the period when the political situation stabilized, security was restored to the countryside, investors became convinced that the country had adopted a business-friendly environment, and capital inflows were very large. The three major sectors, agriculture, industry, and services, all contributed to this growth (Figure 3).  Cambodia's economic growth has been quite volatile, for three interrelated reasons. The first has been the occasional political uncertainty, most significantly in July 1997 when the power sharing agreement between the two major political parties collapsed and a brief but bitter civil war erupted. The second factor has been the country's relatively narrow economic base, with rice dominating the rural economy, garments-the major manufacturing activity, and tourism, construction, and related services activities driving much of the urban economy.
The third explanation for the volatility has been major external shocks, most importantly the 1997-1998 Asian financial crisis and the 2008-2009 global economic recession. The former event impacted less severely on Cambodia than its neighbors, and was in fact overshadowed by the domestic political crisis. Cambodia was not then connected to regional financial markets, the principal contagion channel during that crisis, and it had experienced little of the asset price boom fueled by short-term capital inflows than its more developed ASEAN neighbors had experienced. Moreover, the rural economy was largely insulated from the crisis, while new drivers of economic growth, such as garment exports and tourism, were mainly connected to the still-buoyant Organisation for Economic Co-operation and Development economies.
By contrast, there was a severe economic slowdown in 2009. Elsewhere (Hill and Menon 2011) we assess the proximate causes of this event, noting also the data caveats, especially the absence of reliable quick-release indicators of the real economy and financial markets. The real economy was affected by a sharp slowdown in the garments and tourism industries, that are both heavily dependent on western markets; tourism was also affected by the slower growth in some neighboring economies. Short-term capital flows also fell significantly, resulting in the collapse of some major modern-sector construction projects, particularly in Phnom Penh. However, other major sources of capital inflow remained buoyant, especially ODA but also FDI and remittances to a lesser extent. In addition, the banking sector remained largely intact, with no major bank runs, in spite of the sudden slowdown in construction activity to which the banks are heavily exposed. This benign outcome may be attributed to the banks' relatively cautious prudential standards reinforced at the margin by the National Bank of Cambodia's attempts to monitor and enforce international banking management practices. The major source of bank funding is domestic deposits, and these institutions have very little exposure to risky financial instruments at home or abroad. The crisis and the associated drop in asset prices, particularly for urban real estate, also had the beneficial effect of redirecting funds towards more productive activities such as agribusiness, coinciding as it did with buoyant food and commodity prices.
Returning to the growth story, there is no readily available comparator to benchmark Cambodia's performance. The rapid growth is partly a reflection of the extremely low starting point in the early 1990s. But as noted it also reflects some important policy choices, especially related to macroeconomic management and openness. Guimbert (2010) argues that Cambodia's growth over the period 1998-2007 was one of the fastest in the developing world. To further illustrate this proposition, we computed growth rates in GDP per capita by decade for all the developing East Asian economies and India since 1950. For the decade 2000-2010, Cambodia's growth rate of almost 7% was below only that of the PRC (1980s, 1990s, and 2000s), Singapore (1960s and 1970s), and Hong Kong, China (1960s and 1970s). Its growth rate also compares very favorably with that of its neighboring Mekong latecomers (Pomfret, forthcoming), as well as other post-conflict cases, such as Uganda, Rwanda, and El Salvador.
Rapid economic growth has driven equally rapid structural change (Table 1). Agriculture, dominated by rice, has grown strongly at over 4% per annum, but its share of GDP has fallen from over 50% to about 30% in less than two decades, with a commensurate, though imperfectly measured, proportionate decline in the share of agricultural employment from about 80% to 60%. There is considerable scope for broad-based agricultural growth, and associated processing activities. Rice yields are well below those of neighbouring Thailand and Viet Nam. Only about 10% of agricultural land is irrigated, fertilizer use is low, and only about 5% of rural roads are paved, even though these roads constitute about three-quarters of the country's road network. In the case of rice, much of the unmilled paddy is processed across the border, even for domestic consumption, while Cambodian exports of milled rice are very small. Manufacturing has grown rapidly, by about 15% per year but is dominated to an extraordinary extent by just one sector, garments, which accounts for over half of output. The large agricultural processing and home goods manufacturing that one typically observes in lowincome economies is largely absent owing to the proximity to large and efficient industrialized neighbors. Construction has grown rapidly, fueled by the country's major infrastructure backlog, the modern sector urban boom, and some speculative activity. Mining is small but likely to grow rapidly as the environment for exploration becomes more favorable, if commodity prices remain high.
Cambodia also has a large service sector for a low-income economy. The explanation is threefold. The first is the very large tourism sector and associated activities such as hotels, restaurants, transport, and personal services. Second, there are the large ODA and private inflows which generally have substantial modern-sector spinoffs in urban services. The third factor is the relatively small tradable goods sectors, particularly manufacturing, but also agriculture to the extent that rural infrastructure investments have been holding back growth in this sector.

B. Falling Poverty, Rising Inequality
Cambodia's social indicators reflect the country's history, its resource endowments, recent development trajectory, and current political economy. The incidence of poverty has fallen quite rapidly since the early 1990s, while health, education, and other social indicators have improved commensurately. Some, but not all, of these indicators have lagged the rapid economic growth as a result of the skewed development patterns, especially the rising inequality between households and regions. There is of course a 'missing generation' in the country's human capital owing to the fact that most of the country's educated elite perished in the late 1970s, a phenomenon that will take more than one generation to overcome. A legacy of past conflicts is the relatively high proportion of households headed by females, disabled persons and former border camp refugees, all of which feature significantly among the poorest households. All these judgments should be regarded as approximate, owing to the weak social database, especially for inter-temporal comparisons. Table 2 summarizes the key social outcomes. By comparative Southeast Asian standards, poverty (as measured by the World Bank's $2 per day, purchasing power parity benchmark) remains high, with just over half the population living below this line in 2008. This incidence is somewhat less than that for the Lao PDR, but higher than that for the three larger Southeast Asian economies, Indonesia, the Philippines, and Viet Nam, that are just in the lower middle-income group of countries. The rankings mainly reflect differences in per capita income. Cambodia's poverty also fell most rapidly during the period of accelerated economic growth, 2004-2007. There are significant differences in poverty incidence between urban and rural areas, illustrating the lopsided nature of Cambodian economic growth since the mid 1990s, with the improvement in urban living standards far outstripping those in rural areas.
Cambodia's demographic and health indicators also provide a clear picture of improvement, aided by two reasonably comprehensive population censuses in 1998 and 2008. 3 In summary, these data (not presented here owing to space limitations) show the following. First, Cambodia was the only Southeast Asian country to record negative population growth in the 1970s (as with Timor Leste, for similar reasons), followed by the 1980s decade of demographic catch up with annual growth of more than 3.5%. Since 1990, its population growth has approximated the region's low-income norms. Second, fertility rates reflect this pattern, albeit with continuing high levels into the 21st century. Cambodia therefore has a very young population, with just 3.8% of the population estimated to 65 years or above in 2010, about twothirds of the average ASEAN figure, and well under half the proportion in Thailand. Third, life expectancy has risen dramatically. In the first half of the 1970s, the average Cambodian lived about 40 years, a figure that fell to just 31 years later in that decade. By 2005-2010, the figure had doubled to 61.5 years, still 8 years below the region's average, but catching up to its lowincome norms. Fourth, similarly, infant mortality has fallen sharply from the exceptionally high levels of the 1970s, when it peaked at about 264 (per 1,000 live births) to the current figure of about 62. This is, however, the highest in ASEAN and almost three times the region's average. Although they suffer from well-known limitations, the country's MDGs at least enable major social policy challenges to be identified. Table 2 presents the most recent set of estimates as well as the 2015 targets. There have been some notable public health successes. Child immunization is now widespread and expected to be universal in the near future. Poliomyelitis has been virtually eradicated while malaria infection rates have fallen significantly. A decade ago the country appeared to be heading for a serious HIV-AIDS epidemic, exacerbated by the large crossborder movement of people. However, like its two larger neighbors, Cambodia has successfully controlled the problem, and the incidence is now estimated to be very low, about 0.9% in 2011. The major problem areas include the very high maternal mortality rates, thought to be a problem owing to the limited health services in poor rural areas, and child stunting, owing to deficient diets. The incidence of stunting and malnutrition in the under 5 years age group is very high, about 30%, indicating that there are serious food distribution problems in spite of the country's overall food surplus.
Trends in inequality are less clear. By the early 1990s, Cambodia probably had one of the lowest levels of inequality in income and wealth distribution as all large private wealth holdings had been wiped out. As best as can be estimated, expenditure inequality (the most reliable series) began to rise rapidly through the 1990s and into the next decade. By some accounts, it was one of the most rapid increases in inequality in developing Asia. However, the recent household surveys suggest that inequality has fallen significantly. It is possible that such a trend has occurred, as employment opportunities had spread rapidly, in tourism, rice, tree crops, garments, and construction. The major economic slowdown in 2009-2010 may have also been equalizing, to the extent that modern sector activities were most severely impacted. But the reported variability is very large, and we will need to suspend judgment pending the release of more accurate data.

C. Macroeconomic Policy and Outcomes
As noted, Cambodia's macroeconomic policy framework and flows are unusual. It has very large current account and fiscal deficits, both currently manageable. Its large fiscal deficits are financed mainly by very large ODA flows, much of it in the form of grant or highly concessional finance. The very large current account deficits are also financed by ODA, together with FDI and other capital inflows. Although it has a central bank, the National Bank of Cambodia, it has limited monetary and exchange rate policy authority, owing to the very high level of dollarization. Inflation is generally modest with occasional spikes, the result of either cost-push pressuresmainly food prices-or very large capital inflows that push up asset prices. We briefly examine each of these propositions. Figure 4 presents summary macroeconomic balances relative to GDP since the early 1990s. It shows clearly the magnitude of both the current account deficits (X-M) and the fiscal deficits (G-T), with the corresponding large investment gap financing (I-S). The government has been able to run large noninflationary budget deficits, in the 4%-7% range of GDP in the early post-conflict period, declining to 2%-3% after 2000, except for the very large deficits during 2008-2009 to cope with the sudden economic downturn. The current account deficits have been funded by both the ODA and FDI inflows, and have therefore posed very little macroeconomic dangers. The trade deficit was exceptionally large during the early rehabilitation period. Both savings and investment rates have increased impressively ( Figure 5). Savings rose from 5% of GDP in the early 1990s to a peak of over 20% during the boom period. Around this increasing trend, the series has been quite volatile, falling sharply in response to the political instability of the late 1990s and the 2008-2009 recession. Investment rates have been higher still, around a broadly similar trend, with a noticeable decline and then recovery after 2007.
The government's budgetary position reflects the weak tax effort, the reliance on trade taxes, and the importance of external financing. Over time, these indicators have been gradually improving, in the sense that the tax effort has been rising, the share of trade taxes has been falling, and the government has been able to finance more of its own expenditure requirements. Domestic revenue, for example, has risen from 4.3% of GDP in 1994 to over 10% currently. There has also been a discernible 'peace dividend' from the greater domestic and regional security. That is, the share of defense and security expenditure has been declining, from about 4.4% of GDP in 1994 to about 1.9% currently, while that devoted to the social sectors has risen from 1.3% to 2.1% over the same period (Naron 2011).  Cambodia's balance of payments similarly reflect the large excess of imports over exports, the narrow concentration of the merchandise trade (particularly in the case of exports), and the very large private and official capital inflows to finance them. Some numbers for recent years illustrate approximate orders of magnitude. For example, the current account deficits excluding official transfers have been in the range 8%-12% of GDP, declining to about 3%-6% including these transfers. The deficits on current account are generally more than matched by capital account surpluses, resulting in rising gross international reserves. The latter were of course negligible in the immediate post-conflict period, and now total over $3 billion, equivalent to about 4.5 months of imports. Garments constitute about 80% of merchandise exports, which in aggregate are generally equivalent to about two-thirds of merchandise imports. Tourism dominates the services flows, in most years exceeding both FDI and official transfers.
Inflation: Cambodia experienced highly turbulent macroeconomic conditions in the early 1990s. It suffered hyperinflation in the very early part of the 1990s as a result of inflationary financing or monetization of the budget deficit. Immediately prior to the peace settlement, Soviet aid constituted about 15% of budget expenditure. As it was quickly scaled back, the government resorted to large-scale deficit financing, as much as half the budget gap in 1991, hence triggering inflation. Between 1990and 1993, inflation averaged 130%, peaking at 191% in 1991). By 1994, however, inflation had eased significantly to 10.4%. Tight monetary and fiscal policy and increasing dollarization have helped to keep inflation relatively low, except for a couple of occasions: in 1998, inflation rose to 14.7% as a result of the Asian Financial Crisis (AFC), and in 2008, inflation spiked again to 25% due to rising food and oil prices, as well as rapid growth in government spending and broad money. The latter was a result of government efforts to mitigate the impact of the global financial crisis (GFC) by boosting domestic demand through fiscal and monetary expansion.  1990: 142% 1991: 191% 1992: 75% 1993: 114% 1990: 241% 1991: 29% 1992: 214% Inflation turned slightly negative in 2009 at -0.7%, but with the pick-up in growth, inflation reached 4% in 2010. It rose further to 5.5% in 2011, again due to higher global food and commodities prices. It is expected to moderate to about 4% in 2012, in line with lower global commodity prices. Putting aside the turbulence in the early 1990s, and a couple of episodes characterized by external shocks, Cambodia has managed to keep inflation under control.
Exchange rate: Since 1993, Cambodia's exchange rate system has consisted of an official rate and a parallel (market) rate. The National Bank of Cambodia (NBC) adjusts the official rate daily, keeping the spread between official and market rates at less than 1% since 1995. The shift to a managed float led to a sharp depreciation in the riel in 1993. Another sharp depreciation occurred in 1998, when the riel depreciated by 27% due to the AFC and rising inflation. The nominal exchange rate has remained broadly stable since 1999 (Figure 7). There has been a very mild trend of depreciation since then, with the riel\dollar rate rising above the 4,000 mark, but it appreciated to bring it back toward 4,000 in 2010-2011. Since the riel continues to be used quite widely in rural areas, the government has been keen to keep the rate stable in order to maintain its purchasing power in line with social and distribution objectives.

Figure 7: Nominal Exchange Rates and Dollarization, 1990-2011
Sources: IMF International Financial Statistics (ER), Beresford et al (2004), Menon (2008), and IMF Staff Reports for Article IV Consultations (dollarization) As a heavily dollarized economy however, the riel\dollar exchange rate is not the only measure that matters, especially when it comes to determining international competitiveness of traded goods. Dollarization compromises the ability of exchange rate policy to initiate adjustments to macroeconomic shocks, as will be shown below. If the US dollar is Cambodia's de facto currency, then movements in the value of the dollar against the currencies of its major trading partners may matter more in determining external competitiveness. For instance, export contracts of Cambodia and its competitors in the garments industry are invoiced in dollars, and therefore a depreciation of the dollar against the currencies of these countries boosts Cambodia's competitiveness. This is because, while costs such as money wages are unaffected in Cambodia, as they are set in dollars, these costs would now be lower than those in nondollarized economies.
Dollarization: Cambodia is one of the most heavily dollarized countries in the world. It is also the most dollarized economy in Asia among those that are not officially dollarized, such as Timor-Leste. Cambodia was not yet heavily dollarized in 1990, although foreign currency began circulating in the economy in 1989 following a gradual liberalization of foreign investment. Dollarization occurred mainly during the UNTAC period of 1991-1993, when dollars poured into the economy. Therefore, dollarization in Cambodia can be viewed as a direct legacy of the destruction of economic and financial institutions after the 1970s, economic mismanagement in the 1980s, and the large inflows of US dollars that occurred during the UNTAC period in the early 1990s. Thus, unlike other countries where bouts of macroeconomic instability and hyperinflation largely induced or increased dollarization, the last and determining source of dollarization in Cambodia was a result of an administrative and political event.
Cambodia today is still as dollarized, if not more so, than it was 10 years ago. The US dollar still serves all three functions of money: it is widely used as a medium of exchange, store of wealth, and unit of account. The riel is used in the payment of taxes and public sector wages, and for small transactions, usually involving less than $1. The riel is widely used in the rural economy however (Menon 2008, World Bank 2009. The IMF estimates the share of dollars in currency in circulation at about 90%, little changed from what it was a decade ago. Recent estimates put the share of foreign currency deposits (FCDs) in broad money (M2) to have risen to its highest level ever, at 82% in December 2011 (Figure 7). Dollarization has costs and benefits. Arguably the main costs are the loss in seigniorage and the reduced ability of the monetary authority to implement discretionary monetary policy. The monetary authority also loses its capacity to act as lender of last resort in order to guarantee the payments system in the event of a banking crisis. The possible loss in seigniorage is estimated by Duma (2011) to range between 0.1%-0.5% of GDP. Although this in non-negligible, it is also not very significant.
In Cambodia, a reverse relationship between dollarization and inflation can be observed. Hyperinflation in the early 1990s soon dissipated as the level of dollarization increased exogenously. With dollarization, Cambodia's nominal exchange rate acts as a nominal anchor to manage inflation. Indeed, the objective of Cambodia's exchange rate policy is confined to 'price stability' rather than supporting export industries. In this respect, the inability of the central bank to conduct an independent monetary policy, and therefore similarly exchange rate policy, may not have been as severe a shortcoming as might appear to be the case. In an environment of price and exchange rate stability, both trade and growth have increased sharply in Cambodia. This contrasts with the recent experience in neighboring Viet Nam, for instance, where declining dollarization has been accompanied by significant increases in the volatility of prices and exchange rates.
In this sense, it could be argued that dollarization does not impose constraints on policy independence that are significantly different from those faced by small open economies, without capital controls, operating in a world with significant capital movements. But there are differences that are attributable to the presence of high dollarization that must be recognized, even if only by degree.
With regard to monetary policy, for example, although it is unlikely that Cambodia could pursue a totally independent monetary stance without capital controls in place, the issue is about some degree of discretionary power to implement countercyclical monetary policy that stabilizes the business cycle and dampens the effects of external shocks, when they occur. Being heavily dollarized is also quite similar to having a highly managed or fixed exchange rate. But it is different in important ways. Fixed exchange rates can be floated through policy decision, or if countries are forced to do so when they run out of reserves, but this is not the case with the exchange rate effect of dollarization. Furthermore, in dollarized economies, prices of a significant portion of non-traded goods may also be set in dollars, unlike in non-dollarized economies. The higher the share of non-traded goods that are priced in dollars, the greater will be the rigidity of the real exchange rate to changes in the nominal exchange rate (Menon 2008).
If the adjustment to economic shocks must occur through changes in money wages, rather than through changes in the nominal exchange rate, there are policy implications. For one thing, there are likely to be nominal rigidities in labor markets that limit, or at least slow down, the adjustment to external shocks. Even if there is significant unemployment of labor, wages rarely adjust quickly, or indeed fully, to demand-supply imbalances. There might also be an asymmetry in the response of wages. Depending on labor market conditions and factors such as the level and strength of unionization, nominal wages may be rigid downward. If nominal wages are initially close to subsistence levels, then any downward adjustment in response to an external shock will be difficult. Thus, the adjustment mechanism that requires changes in wages instead of changes in nominal exchange rates is a relatively inefficient instrument, and could result in social costs such as rising unemployment.
To sum up, while the constraints on macro policy management imposed by dollarization do not appear to have had a significant negative impact on the economy so far. Looking forward, therefore, the question is not whether Cambodia should de-dollarize, but rather when and how. Any draconian policy action, such as attempts to enforce de-dollarization, is not only likely to fail, but could prove counter-productive. Even without enforcement, the economy needs to be prepared and ready for de-dollarization for it to be both beneficial and sustainable. Moreover, the experience of Viet Nam and its current macroeconomic difficulties needs to be heeded. It is important to recognize that dollarization is more of a symptom that it is a problem. The problem (or the cause) is a lack of confidence in the riel, whilst the symptom (or the effect) is the use of another currency such as the US dollar. The causes of the problem emanate from the various constraints that Cambodia continues to face as an economy in transition, ranging from an underdeveloped financial and monetary system to weak legal and institutional structures. These are the problems that need to be addressed directly. When these problems are addressed, then the symptom, which is dollarization, will also cease to be a constraint.

IV. CONCLUSIONS
Looking backwards, Cambodia has achieved much more rapid economic development over the past 2 decades than even the most optimistic forecasts could have projected at the time of the 1991 Paris peace settlement. This has been a period of prosperity and peace almost without precedence in the country's history. The Cambodian people are now better educated and fed, they live longer, they have greater social and occupational choice than ever before, and they have a limited measure of democratic space. At this rate of progress, the country will very shortly graduate from the 'least developed' group of countries, and within a decade or so, it would enter the middle-income developing group.
But the challenges and the outstanding reform agenda are equally formidable. The country has been treated very generously by the international donor community, but aid flows are now declining relatively, and within the next few years they will begin to decline in absolute terms. Perhaps inevitably, the economy is narrowly based, with limited policy levers to deal with the challenges of a volatile regional and global economy. The priority of growth at all costs will need to give way to greater attention to distributional considerations, between households, regions, and particularly the capital, Phnom Penh, and the rest of the country. Key supply side challenges remain, particularly education at all levels and rural infrastructure. Levels of corruption remain high, while community tolerance for abuse of office will fade as education levels improve and memories of the awful past recede. Although democratic processes are in place, in the not-too-distant future the country will also have to manage its first major political succession in the post-conflict era.