Chapter 5. Foreign Aid Flows and Foreign Direct Investment in Cambodia

Sumio Ishikawa, Sibel Beadle, Damien Eastman, Srobona Mitra, Alejandro Lopez Mejia, Wafa Abdelati, Koji Nakamura, Il Lee, Sònia Muñoz, Robert Hagemann, David Coe, and Nadia Rendak
Published Date:
February 2006
  • ShareShare
Show Summary Details
Koji Nakamura

When Cambodia embarked on its economic reconstruction in 1993, its physical infrastructure as well as human capital were nearly decimated by the decades-long civil war. With few domestic resources at its disposal, Cambodia relied heavily on external financing and technical assistance. Foreign direct investment (FDI) played a more limited role. Section A reviews the amount and sectoral distribution of foreign aid, and Section B examines its contribution as well as any side effects on the economy. Section C reviews factors that contributed to foreign direct investment inflows, and in Section D, future FDI prospects are assessed in the context of the current investment climate.

A. Recent Developments in Foreign Aid Flows

Foreign aid flows in Cambodia have averaged 12 percent of GDP a year in the past decade, reaching $500 million in 2003 (Table 5.1). About 70 percent of aid flows were in the form of official grants, largely provided by bilateral donors, while the rest were concessional loans mainly from the World Bank and the Asian Development Bank.

Table 5.1.Aid Flows(In percent of GDP, unless otherwise indicated)
Official grant10.09.911.511.
(in millions of U.S. dollars)257355502443391354342418423491504
Sources: Council for the Development of Cambodia, Development Cooperation Report; and Ministry of Economy and Finance.
Sources: Council for the Development of Cambodia, Development Cooperation Report; and Ministry of Economy and Finance.

Cambodia is one of the largest recipients of foreign aid among the Asian countries in percent of respective GDP, along with Lao P.D.R., Bhutan, and Mongolia (Figure 5.1). Moreover, it has the lowest average interest rates on its borrowing (Table 5.2) and a very large share of official grants, in part reflecting the extensive technical cooperation provided by donors.

Figure 5.1.Comparison of Aid Flows1

(In U.S. dollars per capita)

Source: World Bank, Global Development Finance.

1 Average of 1995–2001.

Table 5.2.Aid Flows in Low-Income Countries(In percent of GDP; average 1995 – 2001)
GrantLoanTotalAverage Interest Rate (In percent)
Lao P.D.R.12.16.618.71.8
Sri Lanka1.
Source: World Bank, Global Development Finance.
Source: World Bank, Global Development Finance.

From a functional perspective, most of the aid flows were provided for specific purposes, initially as food aid and emergency relief assistance, and then investment projects and technical assistance (Table 5.3). By 2003, technical assistance accounted for about 40–50 percent of the total, most of which was spent on compensation of technical assistance advisors. Investment projects accounted for about 35–45 percent of the total aid flows.

Table 5.3.Share of Aid Flows by Type(In percent of total)
Technical assistance24.229.733.736.148.054.849.844.937.342.946.1
Investment project23.238.740.740.540.844.033.435.445.845.043.4
Budget support22.819.315.
Food aid29.712.310.510.610.61.37.711.
Source: Council for the Development of Cambodia, Development Cooperation Report.
Source: Council for the Development of Cambodia, Development Cooperation Report.

Sectoral distribution of aid was highly skewed toward education and health, and infrastructure, which together accounted for about 50 percent of total aid flows. The share of education and health has increased with greater involvement by the World Bank and the Asian Development Bank, while the share of institutional building has gradually declined as the country emerged from a post-conflict situation. By contrast, less than 10 percent of the aid was spent for agricultural development (Table 5.4), which is the main source of income for the poor.

Table 5.4.Share of Aid Flows by Sectors(In percent of total)
Education and health17.813.913.
Agriculture and forestry8.97.27.313.
Institutional building21.328.328.731.125.422.015.215.517.919.010.8
Source: Council for the Development of Cambodia, Development Cooperation Report.
Source: Council for the Development of Cambodia, Development Cooperation Report.

B. Contribution of Aid Flows

Aid flows have not only played a critical role in helping Cambodia rebuild the basic economic system, but have also helped the government run the country. While Cambodians returning from abroad have partially filled the hole in human capital left by the Khmer Rouge rule, a large number of foreign experts were needed to fill in the remaining gap. Moreover, the near absence of institutions, including a legal structure, required a large amount of foreign technical assistance to draft laws, especially for WTO accession, and establish basic operating procedures. Such needs were reflected in the large share of technical cooperation in total aid flows.

The remaining aid flows were used largely to enhance the country’s long-term growth prospects. Aid flows were critical in supplementing investment in education and health, and basic infrastructure, which lagged substantially behind, even relative to other low-income countries (Table 5.5). With much of the attention paid to these urgent issues, the amount of resources allocated to a more direct means of alleviating poverty has been small. This is partly reflected in the lesser aid flows to agriculture and rural development, which averaged 19 percent a year in the recent past.

Table 5.5.Education Indicators, 2002
Literacy ratio6974
Secondary school enrollment ratio (gross)2257
Source: World Bank, World Development Indicators.

Average of low-income Asian countries excluding Cambodia, Lao P.D.R., Nepal, Sri Lanka, and Vietnam.

Source: World Bank, World Development Indicators.

Average of low-income Asian countries excluding Cambodia, Lao P.D.R., Nepal, Sri Lanka, and Vietnam.

Aid flows appear not to have led to a “Dutch disease” situation in Cambodia. Much of the country’s labor force (largely unskilled) is still underutilized and aid flows were spent on payments of imports of goods and services. Little was spent on locally produced goods and services, reflecting the limited scale of domestic manufacturing output and the capacity of the services sector. Even the construction sector largely uses imported materials.

Providing a quantitative assessment of the contribution of aid inflows is marred by weak data and other parallel developments. The bilateral trade agreement with the United States, which led to a sharp increase in garment exports, and the pickup in tourism reflecting pent-up demand for visits to historic sites following political stability in the late 1990s led to rapid growth in these sectors. The contribution to growth from these two sectors was large, such that it is difficult to separate the net impact of aid flows on the relative growth of the tradables and nontradables sectors, which is one measure of assessing the Dutch disease22 (Figure 5.2).

Figure 5.2.Sector Contribution to GDP

(In percent)

Source: Cambodia, National Institute of Statistics.

The real exchange rate has appreciated, even though data weakness bars drawing a definite conclusion23 (Figure 5.3). Prices of tradable goods as measured by prices of clothing and footwear have risen by less than the prices of nontradable goods such as housing and utility prices. However, utility prices are affected by world oil prices, and housing prices might have been driven up not only by aid inflows but also by the recent increase in wealth of the urban areas that led to a sharp increase in demand for housing, faster than housing supply.

Figure 5.3.Consumer Price Index (CPI)

(December 2000 = 100)

Source: Cambodia, National Institute of Statistics.

There is an ongoing debate as to whether or not aid flows induce corruption in the recipient countries. There is little evidence to suggest that aid flows in Cambodia have, or have not, induced corruption. About half of the total aid flows are executed outside the government budget and are thus subject to close scrutiny and control of donors, and even those that are channeled through the budget are closely monitored by the respective donors. A case could be made, however, that donors’ budgetary financing weakens the authorities’ resolve to raise fiscal revenue.24

C. Foreign Direct Investment

FDI approvals increased dramatically following the peace accord and the reconstruction efforts led by the United Nations Transitional Authority in Cambodia (UNTAC) that began in 1993, but declined thereafter (Table 5.6). Actual FDI disbursement, while more phased, broadly exhibits a similar trend.

Table 5.6.Foreign Direct Investment in Cambodia(In millions of U.S. dollars)
FDI approval12821,91061757855619616014014565
In percent of GDP105718171864442
FDI disbursement28015129416824323014914914587
In percent of GDP3495874442
Sources: Council for the Development of Cambodia; National Bank of Cambodia; and National Institute of Statistics.

Approvals issued by the CDC.

National Bank of Cambodia’s estimates.

Sources: Council for the Development of Cambodia; National Bank of Cambodia; and National Institute of Statistics.

Approvals issued by the CDC.

National Bank of Cambodia’s estimates.

1994–95 period: The initial FDI approvals concentrated on tourism and construction sectors (Figure 5.4). Cambodia’s cultural heritage, notably the country’s ancient monuments, were perceived by investors as having a strong potential to attract a large number of foreign tourists, especially with the advent of political stability. The ensuing construction of hotels and basic infrastructure attracted large investment into the construction sector.

Figure 5.4.FDI Approval by Sector1

(In millions of U.S. dollars)

Source: Council for the Development of Cambodia.

1 Estimated by IMF staff based on CDC investment approval data.

1996–98 period: The second wave of investment concentrated on logging. The high-quality hardwood in Cambodia, which required little formalities to log, attracted investment into the logging industry. However, due to strong complaints from the donor community regarding the rapid and chaotic depletion of the forests, the government finally embarked on a major reform of its forest policy in January 1999.

The next wave of foreign investment was in the garment industry, in response to the 1996 signing of the bilateral trade agreement with the United States that substantially reduced the effective tariff rate imposed on Cambodian garment exports. However, since the relatively labor-intensive garment manufacturing industry did not require a large amount of investment to set up factories, the associated amount of FDI inflows was modest.

1999–Present: FDI inflows to Cambodia have been weak since 1999. Annual FDI approvals fell below $200 million, a decline that occurred against a backdrop of restored macroeconomic and political stability. The decline was partly related to changes in the outlook for garment exports with the expected phaseout of the quota system that would put Cambodia on an equal footing with all WTO members. Only moderate investment in tourism-related activities continued to attract foreign interest.

D. Future Prospects of FDI

Except for the three sectors identified above that provided surges at different periods in the past, FDI inflows to other sectors have been weak. With the elimination of quotas on garments, and the pent-up demand on tourism gradually reaching a satiation point, there are no obvious factors left that would attract foreign investment in the future. Accordingly, Cambodia will be subject to competition on par with its neighboring countries for general FDI inflows to the region for which competitiveness and investment climate will become more important. The following factors, however, argue for poor prospects.

  • Widespread governance problems. As confirmed by a recent study by the World Bank (2004), the informal costs in Cambodia are high: the so-called bribe tax is roughly 5 percent of total sales in the manufacturing sector, the highest among five countries for which similar data are available.

  • Weak rule of law. Although the legal framework could be strengthened with the expected enactment of numerous laws related to WTO accession, the reform of the judiciary, critical for implementation of these laws, remains uncertain.

  • High cost of production. Energy costs in Cambodia are high, reflecting the dilapidated state of the country’s diesel generators and the lack of domestic sources of fossil fuels.

  • Small local market. The economy is too small to attract foreign investments targeted for the local market.

Private sector–led growth in Cambodia clearly depends on mobilizing FDI given the scarcity of domestic savings. Although it is difficult to address all the problems mentioned above in the short term, the government needs to take bold steps to provide the basis for a more competitive economy. In the short run, at least providing a more predictable business environment through streamlining red tape could reduce the cost associated with uncertainties. The agenda for the medium to long term remains large: given governance is a cross-cutting issue, strengthening the judiciary will need to receive priority while at the same time the government should foster human capital formation through expanding education to everyone (Borensztein, De Gregorio, and Lee, 1998).

Nkusu (2004) suggests that Dutch disease in the context of aid flows is suspected if (1) the real exchange rate appreciates, and (2) the tradables sector shrinks relative to the nontraded sector.

The consumer price index (CPI) is calculated only in the main urban areas.

Aid flows are said to induce corruption when resources are transferred without the accompanying accountability of the decision maker. An alternative view is that aid flows are associated with improved rules and conditions that limit the discretion of the recipient country’s officials, thus decreasing corruption. The results of recent empirical studies are mixed. While Alesina and Weder (2002) suggest that aid flows are positively correlated with corruption, results by Tavares (2003) suggest the opposite.

    Other Resources Citing This Publication