• Large infrastructure projects have been a strong driver of real GDP growth in East Africa.
  • Due to the infrastructure investments the public finances of Djibouti and Kenya have deteriorated.
  • Import demands have surged given the need to import capital goods and construction material related to infrastructure projects.
  • Export growth has been subdued in most countries.
  • Current-account deficits have been funded increasingly through external borrowing.
  • External borrowing has outpaced export receipt growth, which is putting pressure on external debt sustainability.

Large infrastructure investments in East Africa

It is evident that countries all over Sub-Saharan Africa are investing heavily in large infrastructure projects. These investments have been one of the most important factors that have driven external and public debt to their highest levels since the region received debt relief. The IMF estimates that since 2013 the share of low-income developing countries at high risk of debt distress has doubled to around 40% in Sub-Saharan Africa. This publication focuses on the economic impact of infrastructure investments in six East African countries, namely Djibouti, Ethiopia, Kenya, Rwanda, Tanzania and Uganda. While most countries in East Africa have performed better than other countries in Sub-Saharan Africa, two of them, Ethiopia and Djibouti, are currently considered by the IMF to be at high risk of debt distress and in all countries safety margins have been reduced in recent years.

Large-scale infrastructure programmes have been launched in all East African countries as an essential part of long-term development plans outlined by the government and these have had a significant impact on the whole economy.

In Djibouti, where infrastructure development has led to a strong deterioration of public and external debt sustainability, infrastructure programmes have been part of government’s “Vision Djibouti 2035” plan launched in 2014. The plan aims to transform Djibouti into a logistics and commercial hub for the whole of East Africa by developing new deep-water harbours and expanding existing ports. Additionally, a railway and a water pipeline to connect Djibouti and Ethiopia have been constructed. In Ethiopia, infrastructure investments have been a key part of two successive growth and transformation plans (2010-2015, 2016-2020). Besides joint projects with Djibouti, the government has also expanded domestic railroads and launched the construction of Africa’s largest dam (the Grand Ethiopian Renaissance Dam) and a number of large industrial zones. For Tanzania, another country which aims to become the logistics hub of East Africa, investments are also part of two consecutive five-year development plans (the second one started in 2016) which aim to capitalise on Tanzania’s comparative advantages, in particular its agricultural and mining potential and its geographical location. Investments include, for example, the construction of a railway that connects Dar es Salaam with Mwanza, the Stiegler’s Gorge hydropower project, the development and expansion of a number of mines, the expansion of ports, and the planned construction of the oil pipeline that will connect the Ugandan oil fields with the Tanzanian port of Tanga.
In Rwanda, Kenya and Uganda infrastructure investments are also an essential part of their development strategy. In Kenya, for example, they have resulted in the construction of the USD 3.5 bn railway connecting the capital to the port of Mombasa, a controversial project given the significant cost compared to the railway developed in Ethiopia. In Uganda one of the priorities is to develop the necessary infrastructure to extract the country’s oil resources.

These infrastructure projects have contributed to the region’s robust real GDP growth. Graph 1 shows the average GDP growth rate for the six countries since 2010. All of them have increased by more than 4% on average since 2010 and are expected to maintain robust growth in 2018. However, while investments have supported economic growth, real GDP growth has not been sufficient to reverse the debt burden in East Africa and public and external debt levels have risen faster than GDP.

Impact on public finances

In some countries, infrastructure investment programmes have drained public finances. The most striking example is Djibouti, where investments implemented after 2014 led to fiscal deficits of 22% of GDP in 2015 and 16.3% in 2016 compared to an average deficit of just 2.1% in the 5 years before the start of the investment programme. The IMF estimated during their last visit in February 2017 that the deficit would be reduced to just 2% of GDP for that year, but this target is unlikely to have been achieved. The increased deficits have led to Djibouti’s public debt almost doubling. While the public debt-to-GDP ratio was 51% of GDP at the end of 2013, it is currently estimated to be around 90%.

Since 2005, infrastructure investments combined with fiscal slippages in Kenya have led to increasingly large fiscal deficits. In the last three years, the deficit has been above 8% of GDP, which pushed the government debt-to-GDP ratio towards 60% at the end of 2017, a high level for a lower middle income country.

In Uganda the impact on public finances has been more limited but public deficits have nevertheless been rising (averaging 4.0% of GDP for the last five years). Given that investments are needed to develop oil exports, government deficits are likely to increase further in the coming years and are expected to peak in 2020 at around 7% of GDP.
In Ethiopia, Tanzania and Rwanda, the deterioration of public finances has been more limited given that their governments have been able to limit their deficits. For example, in Ethiopia and Tanzania, government deficits have been limited to around 2.5% of GDP, which has led to only a limited rise in the public debt-to-GDP ratio.

Investment projects have led to widening current-account deficits

Large infrastructure investments are associated with a requirement for significant quantities of capital goods and building materials. On top of that, in most countries, there has also been strong demand for imports of consumption goods, fuel and capital goods unrelated to the infrastructure investment programmes such as the import of transportation and telecommunication equipment. Indeed, these products are not often produced domestically.

At the same time, export growth has been subdued in the six countries that this report focuses on, and consistently lower than what was predicted in recent years. There are a number of reasons for this. First of all, while infrastructure investments are expected to lead to more exports in the future, most of the projects are not yet finished and have been delayed. Therefore, export receipts cannot yet be generated by these projects. Secondly, export growth has been limited, as most of these East African countries have a relatively small export base; they mainly rely on primary exports that are vulnerable to global price volatility and weather conditions. Thirdly, given the limited availability of technology, they have not all been able to use production technologies in order to diversify exports and create added value. Additional factors include subdued regional trade (due to increased competition from other countries in the regional market) and country-specific elements such as foreign-exchange shortages and the overvalued exchange rate in Ethiopia or, in Uganda’s case, conflicts in South Sudan and the Democratic Republic of Congo.

Rising imports and subdued export growth have resulted in increasing current-account deficits in East Africa (see graph 2). This is particularly evident in Djibouti, where the current-account deficit has been exceptionally large, peaking at 31% of GDP in 2015. However, Rwanda and Ethiopia have also struggled with large current-account deficits. In Rwanda, they reached almost 17% in 2016. The need to import capital goods and materials for the construction of the Kigali convention centre, the new airport and the expansion of the Rwandese national airline played a large role here.

Tanzania, on the other hand, has been able to structurally decrease its current-account deficit. Although it peaked at 11% of GDP in 2012, it was reduced to less than 4% by the end of 2017, mainly due to increased service receipts and reduced imports. Nevertheless, with the implementation of the infrastructure investment programme, it is predicted that the current-account deficit will increase to around 6% of GDP in the medium term.

For Uganda the current-account deficit is expected to worsen further given the great need for imports related to oil investments.

Given that all the countries are oil importers and oil imports are of great importance, it is remarkable that the oil price collapse after mid-2014 did not have a more positive effect on the current-account balance in the period 2015-2017 and even significant deterioration for Djibouti, Ethiopia and Rwanda.

Current account-deficits are mainly funded through external borrowing

There are various ways to fund a current-account deficit, such as external borrowing, foreign direct investments (FDI) and foreign-exchange reserves. Graph 3 shows the share of the current-account deficit that is funded through FDI for the six countries covered in this report. While Tanzania and Ethiopia have been able to attract significant and increasingly large (net) FDI in recent years, this has not been the case for the other countries; particularly in Kenya, FDI inflows have covered only around 10% of the current-account deficit, meaning that significant external borrowing has been necessary. Nevertheless, in all the countries the FDI inflows have not been sufficient to cover the current-account deficit, meaning they have had to rely heavily on external borrowing.

The result is that external debt levels have risen in all six countries, and while lending has remained mostly on concessional terms from official creditors and multilateral organisations, in the last five years the commercial share has risen in all of the countries (for example through the issuance of government bonds) except Djibouti, which has relied extensively on Chinese concessional government loans. Tanzania was the first country in East Africa to issue bonds in 2013. It was soon followed by Rwanda, Ethiopia and Kenya.

The rise in external debt is particularly worrying when compared to the subdued growth of total export receipts. Given that external debt needs to be repaid in foreign currency, an economy must generate sufficient foreign-exchange revenues in order to pay back the debt. Currently, however, we see that external debt-to-export receipts have risen strongly in all six countries studied (see graph 4).

The rise in external debt has also pushed up the debt service ratios – most drastically in Kenya – meaning that the countries need to spend more of their current-account receipts on the repayment of external debt.

Risks and factors worsening the debt build-up

The external debt build-up also brings a number of risks. First of all, increased external borrowing in foreign currency exposes the countries to currency fluctuation. A currency devaluation risks pushing up the nominal value of the external debt. This happened in October 2017, for example, when the Ethiopian birr was devalued by 15% against the US dollar. Uganda has also been vulnerable to this. In the last two years the Ugandan shilling has lost around 11% of its value compared to the dollar, which is pushing up the nominal value in local currency of the external debt.

Secondly, increased external debt levels expose the countries to risks associated with sharp decreases in risk appetite for emerging markets. This risk could materialise if we see a faster than expected normalisation of monetary policy in advanced economies or an increased risk perception of the emerging markets, as is currently the case with Turkey and Argentina.

Both of these factors would push up the public and external financing requirements and impede refinancing possibilities. A sharp decrease in risk appetite could lead to large capital outflow and thus increased pressure on foreign-exchange reserves or the exchange rate. Additionally, debt service ratios in East African countries could increase, given that most of these countries have external debt with a variable interest rate, which means that the debt service cost will rise along with global interest rates. This is especially the case for Uganda, where more than 60% of external public debt has a variable interest rate according to World Bank data. For the others this figure is lower, at around 30% (except for Rwanda, which has barely any external public debt with a variable interest rate).

Thirdly, most East African countries remain dependent on agricultural products such as coffee, tea and horticulture or other natural resources. Fluctuations in the price of these resources could reduce export receipts.

Fourthly, there remains a risk of further fiscal slippages and thus further debt build-up. This is especially the case in Ethiopia and Kenya, given the current political environment.

Lastly, as a consequence of the debt build-up, they have less room to implement a countercyclical policy in the event of an adverse economic shock.

The road ahead: mind the debt trap

In the coming years, it will be vital for all countries to carefully balance further infrastructure investments and improve debt management. Fiscal slippages should also be minimised and additional external lending in all countries should be limited. New infrastructure investments in East Africa which require additional external lending will need to be carefully selected based on a transparent process, selecting only projects with a high and realistic return on investment. Furthermore, these countries should have a clear strategy for promoting exports. This is particularly critical for Djibouti, Ethiopia and Kenya.

While the six countries score better than most Sub-Saharan African countries in terms of debt-management strategies, there is still room for improvement. Graph 5 shows how they perform on the World Bank Country Policy and Institutional Assessment (CPIA) debt policy rating. This rating assesses how the debt-management strategy of a country minimises risks and ensures long-term debt sustainability on a scale from 1 (low) to 6 (high). It shows that while they score better than average for Sub-Saharan Africa, only Rwanda has improved compared to 2009. When we look at the CPIA fiscal policy rating, which takes into account the short- and medium-term sustainability of fiscal policy, it becomes evident that all except Djibouti and Rwanda have performed worse compared to 2009.

Some governments have already taken measures to limit external borrowing. Since 2016, Ethiopia has been implementing tighter controls on external borrowing by public enterprises, and since it was classified at high risk of debt distress by the IMF in January 2018, the government has indicated that external borrowing will be further slowed down. Nevertheless, its high external debt level continues to put pressure on the country’s medium/long-term (MLT) political risk outlook, which is currently in category 6. Tanzania has made it an essential part of its five-year development plan to limit borrowing for infrastructure projects, and instead the government wants to rely more on public-private partnerships for large infrastructure projects. However, the difficult relationship between President Magufuli and foreign investors risks making this difficult. Given that Tanzania has balanced past lending more carefully, the country has remained stable, in category 5 for the MLT. On the other hand, Djibouti’s wide current-account deficit, the poor state of public finances and the high level of external debt have motivated Credendo to classify the country in category 7 for MLT political risk.


For Kenya, increased external debt levels and external debt service ratios and the deterioration of public finances prompted the downgrade of MLT political risk to category 6 in October 2017.

Uganda is an exception in East Africa given that it is facing the prospect of significant oil revenue, and external debt is expected to decrease once the oil starts to flow. That said, major obstacles need to be overcome and significant additional investment is needed. The government’s timeline indicating that oil will flow as of 2020 is likely to be too optimistic; for example, the pipeline connecting Uganda to the international markets is already delayed by more than a year, and significant road development is needed in order to access the oil fields in Uganda. The significant oil revenue prospects bring additional challenges, however, as Uganda will need to overcome a replication of the boom and bust cycles seen in a number of other oil-exporting countries on the continent. Policy measures such as an expansion of the tax base and saving of oil windfalls could mitigate such a risk.

However, besides limiting external borrowing, the countries will also need to increase their export revenues. Exports can be supported through carefully selected policy measures. For example, Rwanda noted a 50% rise in the export of goods in 2017, albeit from a low level. While this was partly due to favourable price movements of its traditional exports such as coffee, tea and minerals, Rwanda also succeeded in increasing the value added of existing exports and has become a regional distribution hub for a number of products. This recent rise in export receipts and the fact that the government has been limiting external borrowing explains why the outlook for the MLT political risk currently remains stable, in category 6, for Rwanda. In the future, exports in the region should be supported by the completion of a number of large infrastructure projects.

East Africa remains a region with great potential for further economic development. Nevertheless, keeping public and external debt at a sustainable level will be necessary in order to ensure long-term economic growth.

Analyst: Jan-Pieter Laleman – jp.laleman@credendo.com