COVID-19 and Increased Sub-Saharan African Debt: Why Aid Isn't the Answer

Introduction

“COVID-19, through economic and health shocks, has greatly awakened us to African debt fragility and unsustainability.”^[1]

There is no question that the COVID-19 pandemic has negatively impacted the global economy. However, countries in sub-Saharan Africa are uniquely affected. Their already unsustainable debt burden has been made worse over this period. This effect comes from two factors: increased borrowing to finance the purchasing of PPE and funding of stimulus programs, and decreased economic opportunities for tourism-reliant countries as the world saw a broad-based decrease in travel. Heitzig et al.^[2] estimate that African countries have taken, on average, 4.5% of increased debt year-over-year during the pandemic, over and above the already unsustainable levels that plagued the continent pre-COVID. With this increase in debt and decrease in GDP due to COVID, there have been increasing calls to forgive or relieve African debt.

The African continent now finds itself in the same context as the 1980s—heavily overloaded in debt, discussing new debt relief programs. In the past, when in this situation, the IMF and World Bank created two debt relief initiatives: the Heavily Indebted Poor Countries (HIPC) and Multilateral Debt Relief Initiative (MDRI). Though these programs were well-intentioned, they were unsuccessful in that they failed to meet their three main aims: reduce debt, increase development, and decrease corruption. Instead, they created incentives for increased borrowing, allowed politicians to line their pockets with money targeted for development, and overall left HIPC countries worse off.

Since neither the structure of debt relief nor the political conditions present on the continent have significantly changed, debt relief as it is being proposed now would be catastrophic for African countries’ futures—leading, as in the past, to insignificant developmental increases, corrupt governments, and more unstable debt policies.

The History of Debt Relief in Africa Pre-HIPC

The history of high levels of debt in Sub-Saharan Africa begins in Cold War politics. Many dictatorial African states throughout the 1950s and into the 1960s post-independence received sympathetic financing in the form of loans from Cold War powers. These loans were not doled out with the intention of making money. Instead, they created political ties between African states and their creditors in order to induce a sense of loyalty. While these loans were framed as beneficial to African states, they in fact contained payback structures that were infeasible given states’ average development and GDPs. Rather, the intention behind this was to ensure that African states remained loyal to Cold War powers for long periods of time, as loans would accrue interest that made them even harder to pay back, thus keeping African states in relations with these powers for extended periods of time. This can be quantified by the large debt-to-GDP ratios of the time, in some states reaching over 350%. In comparison, Krugman^[3] estimates that only a 70% debt-to-GDP ratio is sustainable.

To make matters worse, most African regimes in this period were corrupt and dictatorial. Many African dictators entered loan contracts with the aim of advancing their own economic standing, typically through corruptly siphoning off loans at the expense of the country and its future development.^[4] Furthermore, as some regimes were tenuous, African leaders had little incentive to engage with long-term planning and thus did not find issue with taking on a high debt burden.^[5] The result was poor development, absent planning, and high African debt.

As countries began to democratize in the 1960s and 1970s, there was a push to forgive debt taken on by these dictatorial regimes. The “We Are the World” movement, in particular, aimed to give newly democratized African countries a fresh start by forgiving past debts and moving forward with a clean fiscal slate. Several independent, country-specific debt relief programs were proposed by Bretton Woods institutions, aimed at reducing debt burden. On the whole, these programs—despite being tailored to a specific country’s needs—failed to understand the cultural and political context of many of the states that they operated in. For example, in several cases, the debt relief programs pushed for increased industrialization at the expense of agrarian lifestyles, failing to appreciate the cultural importance of farming for many tribes as well as generating mass displacement.^[6] These programs were unsuccessful and frequently left in their aftermath higher rates of borrowing, corruption, and poor economic forecasting.^[7] Overall, pre-HIPC debt relief measures were ineffective and led to increased corruption and debt burdens, later hindering the growth of African countries in the 1980s and 1990s.

HIPC And The MDRI

The HIPC program, and its subsequent follow up, the MDRI, were supposed to be different. As part of the millennium development goals, the World Bank and the IMF sought to use debt relief as an incentive for African countries to be able to manage their debt more effectively in the future. To do this, the new programs included two new components. The first was the pre-decision-point Poverty Reduction Strategy Paper (PRSP), as African countries had to lay out a plan—alongside the guidance of World Bank and IMF experts—indicating how the presence of debt relief would help to reduce poverty in their country. These PRSPs had to be reviewed and accepted for countries to be permitted to utilize the HIPC program. The second was the presence of strong conditionality measures both during and after the program, which implemented changes to how African countries would budget and manage their finances. This often included austerity measures aimed at ensuring that African countries would be able to save enough money to pay off loans in the future.

The HIPC process consisted of several stages that implemented these conditionality measures and changes, as illustrated in Figure 1. After countries created their PRSPs and had them approved, they had to prove their willingness to adopt new governance measures outlined by the World Bank that would ensure that they are moving towards fiscal responsibility.

Figure 1: The HIPC process

Once these initial steps were completed, countries reached the decision point, termed this because it was at this moment when the country would either be accepted into the program or not. After the decision point had been passed, countries in the program would receive interim debt relief that was meant to be put towards structural reform. This relief would be supplied throughout the interim period and would be deducted from the final lump-sum debt relief received after the completion point. Once countries completed the reforms necessitated by the World Bank, they would reach the completion point, in which countries received the full amount of debt relief aid and were released from the program. The aim was that once countries are released from the program, they had developed the necessary governance structures to prevent subsequent mass accumulation of debt in the future.

Empirical evidence suggests that both the HIPC and MDRI programs failed at their three main aims of increasing development, reducing corruption, and keeping debt-to-GDP ratios (and therefore debt burden) to a minimum. A difference-in-difference analysis, following the framework of Ferry,^[8] makes these empirical findings become clear.

Methodology

This empirical study is done using a Difference-In-Differences (DID) approach in order to measure the cross-interaction effect between treatment and time for the corruption indicator in both HIPC and the control group, non-HIPC Sub-Saharan African countries. The DID coefficient measures the effect of treatments (i.e., HIPC versus non-HIPC) and time (i.e., pre- or post-debt relief completion points). To fit this model most accurately, time will be measured on a scale ranging from [-5, 15], representing time relative to HIPC completion point dates, with time t=0 referring to the year of the completion point. For the control, non-HIPC countries, the average HIPC completion date, 2005, was used as t=0 to best represent the trends in Sub Saharan African countries at the time. This method was employed to accommodate the different years of completion across the HIPC countries as well as the lack of completion dates for the control countries, as done in Ferry’s^[9] study on the effects of debt relief.

The interpretation of the DID coefficient will depend first on its level of significance, as an insignificant DID estimate indicates that there is no treatment effect over time and thus leading to the conclusion that the rate of change in the relevant index does not depend on the treatment group after t=0. Afterwards, if the DID coefficient is proven to be statistically significant, it can provide the direction of the difference.

Index = β₀ + β₁u₁ + β₂u₂ + β₃u₁u₂

Where u₁ and u₂ are defined as follows:

The β₃ parameter thus determines the cross-effect of time and HIPC status.

This analysis will look at three different data sets to analyze the impact of debt relief on debt burdens, development indexes, and finally corruption. The first data set is World Bank-recorded debt-to-GDP ratios,^[10] selected because these figures used to determine whether countries had significant overwhelming debt burdens. The second data set is the Human Development Index,^[11] a measure of a state’s social and economic development. Finally, the third data set is the World Bank Control of Corruption index,^[12] a measure of how effectively a state can curb political corruption.

Results

Considering first the main aim, reducing future debt burden, Figure 2 plots World Bank-collected debt-to-GDP ratios for HIPC and non-HIPC participating countries to analyze the impact of HIPC on a measure that determines a country’s debt sustainability. Figure 2 thus demonstrates that not only was HIPC ineffectual at creating reduced debt-to-GDP ratios in the long run but that countries participating in the program averaged higher debt-to-GDP ratios than countries not participating in the program after only eight years post completion date. In fact, the difference-in-difference analysis suggests there is a negative impact of debt relief programs on these countries’ ability to maintain a manageable debt-to-GDP ratio after the program. This increase starts, on average, only five years after a country reaches its completion point.

Figure 2: Difference in Difference Comparison of HIPC Participating and Non-HIPC Participating Sub-Saharan African Countries’ Debt-to-GDP Ratios

Furthermore, when comparing the Human Development Index of HIPC and non-HIPC countries in Figure 3, there is no significant difference in development growth after HIPC completion points. While countries in SubSaharan Africa which did not participate in the HIPC program started out on average more developed according to the index, the rate of development growth in both groups was the same even 15 years after the completion point, with a slight dip in development growth in HIPC countries nearing 15 years post-completion. The difference-in-difference analysis indicates that there was no impact of debt relief on development throughout the period.

Figure 3: Comparison of Development Indices of HIPC and NonHIPC Sub-Saharan African Countries

Finally, analysis of the project’s last aim, curbing corruption, shows that HIPC was similarly unsuccessful. Figure 4 illustrates the comparison between the World Bank Control of Corruption index figures for HIPC and Non-HIPC African countries. The higher the Control of Corruption index is, the less corrupt the state. As can be seen, while non-HIPC Control of Corruption scores remain static over time, HIPC countries’ Control of Corruption scores decrease post-completion point—indicating a higher level of corruption after the HIPC program. This indicates that funds that became available through HIPC debt relief were quickly swindled by politicians once the IMF or World Bank stopped watching. Furthermore, with increased access to new streams of debt, there were increased opportunities for politicians to access more money on the countries’ debt roll, allowing for increased opportunities for corruption. For example, in Nigeria, 1 million USD of debt relief funds were used on a Beyonce and Jay-Z concert for high-ranking political officials, instead of being used for development.^[13]

Figure 4: Comparison of HIPC and Non-HIPC Corruption Scores

Why, then, did HIPC, this new and improved means of conducting debt relief, fail at all three of its aims? It fell prey to the same traps of earlier debt relief initiatives: a fundamental misunderstanding of the African context and poor conditionality measures that did not address the core issues leading up to the African debt problem. As Franceschi et al.^[14] argue, conditionality measures associated with HIPC PRSPs failed to correct some of the key issues leading to poor debt management, including overall state weakness, corruption, and bureaucratic inefficiencies.

Many scholars have attributed African state weakness to international institutions and the colonial legacy. Jackson and Rosberg^[15] observe that the borders of African states were artificially drawn by European colonizers and that these borders were preserved by the international community post-independence. As a result, they argue that African states never had to undergo the process of ensuring that they had the necessary capacity and authority to maintain rule over their entire territory. This lack of capacity and authority, coupled with the continued maintenance of boundaries and “statehood” by the international community, is what they refer to as juridical statehood.

This weakened capacity and inability to maintain authority, especially along international borders, creates difficulties for the implementation of the full conditionality measures as prescribed by the World Bank in the HIPC program. For example, many of the countries in the HIPC program were required to collect and report data on incomes, rates of education, and rates of unemployment. However, according to former World Bank consultant Bernhard Gunter,^[16] these same countries had underfunded and ineffectual Bureaus of Statistics, making this task nearly impossible—especially for regions along the border. It is unclear how states eventually devised these figures, and yet these same markers were used to determine whether debt was sustainable: an obvious shortcoming of the HIPC program.

Moreover, in monitoring the progress of imposing conditionality measures, the IMF left tracking completion largely to corrupt governments, which had no incentive to indicate if they were lagging behind in fixing key issues, especially when countries that completed all the necessary steps to reach the completion point received the lump-sum total of their debt relief. This money, without continued oversight from the IMF, could go straight to lining those same leaders’ pockets.

State weakness and poor economic performance also meant that African states were no longer servicing their debts.^[17] This means that African countries participating in the HIPC program weren’t paying loans off, even with balances outstanding. Thus, forgiving debt payments did not create new capital that could fund investment, since there was no capital being devoted to servicing debt in the first place.^[18] In this light, it is no wonder that HIPC did not cause significant increases in development indices.

Why This Time Around Is No Different

The factors that caused the poor results of the HIPC program have not been resolved as new calls have been made for additional debt relief schemes. In fact, if anything, they have become worse.

The Fragile States Index,^[19] a measure of state weakness, indicates that in the period from 2007, post-HIPC completion, to 2020, before the COVID-19 pandemic, countries in Africa have, on average, become weaker, specifically in their ability to effectively govern at the border. This indicates that collecting data and imposing taxes associated with conditionality will be even harder this time around. Furthermore, at the same time, Control of Corruption scores on average for African governments have decreased, meaning countries have less control of the corruption that does occur with government funds.^[20] Countries in Sub-Saharan Africa have largely not begun servicing their pandemic-era debts yet, either.

Moreover, due to increased access to creditors that HIPC debt relief provided, countries this time around have increasingly taken out loans of private origin in global financial markets.^[21] This suggests that IMF-related debt relief schemes, unless private creditors are co-opted, will be largely ineffectual. As a significant share of loans currently outstanding come from institutions outside of the IMF’s control, it is unclear to what extent those institutions will be, or be able to be, incentivized to relieve African debt.

It is unclear how to best proceed with the problem of increasing debt burdens of African states. As Moyo suggests, perhaps it is best to no longer allow Africa to rely exclusively on outside aid and instead to build up institutions to confront the debt problem. While debt relief initially aimed to do just this, relief instead provided little incentive for reform, as the opportunity for another bailout was just around the corner. If instead debt relief becomes unavailable to those struggling Sub-Saharan African countries and they are then unable to access global debt until they could do so sustainably, there would be aligned incentives to build up the institutions necessary for tracking, maintaining, and paying off existing debt.

The problem of debt in Sub-Saharan African states is looming, yet increased debt relief along historical lines is not the answer. If the HIPC program has demonstrated anything, it is that forgiving debt without proper oversight and aligned incentives only leaves states even more poor and indebted than before.

[1] Chris Heitzig, Aloysius Uche Ordu, and Lemma Senbet, “Report: Sub-Saharan Africa’s debt problem: Mapping the pandemic’s effect and the way forward,” Brookings, October 21, 2021, https://www.brookings.edu/research/sub-saharan-africas-debt-problem-mapping-the-pandemics-effect-and-the-wayforward/.

[2] Chris Heitzig, Aloysius Uche Ordu, and Lemma Senbet, “Report: Sub-Saharan Africa’s debt problem.”

[3] Paul Krugman, “Financing vs. Forgiving a Debt Overhang,” NBER Working Paper Series, January 1988, https://www.nber.org/system/files/working_papers/w2486/w2486.pdf.

[4] Darlington Richards, Gladson Nwanna, and Sonny Nwankwo, “Debt Burden and Corruption Impacts: African Market Dynamism,” Management Decision 41, no. 3 (April 2003): https://doi. org/10.1108/00251740310469486.

[5] William Easterly, “How Did Highly Indebted Poor Countries Become Highly Indebted? Reviewing Two Decades of Debt Relief,” Policy Research Working Papers, The World Bank, November 1999, https:// doi.org/10.1596/1813-9450-2225.

[6] William Easterly, The Elusive Quest for Growth (Cambridge, MA: The MIT Press, 2001).

[7] Easterly, “How Did Highly Indebted Poor Countries Become Highly Indebted?”

[8] Marin Ferry, “The Carrot and Stick Approach to Debt Relief: Overcoming Moral Hazard,” Working Paper DT2015-14, September 2015, https://dial.ird.fr/wp-content/uploads/2021/10/2015-14-TheCarrot-and-Stick-Approach-to-Debt-Relief-Overcoming-Moral-Hazard.pdf.

[9] Ferry, “The Carrot and Stick Approach to Debt Relief.”

[10] “Central Government Debt, Total (% of GDP) | Data,” FRED, Federal Reserve Bank of St. Louis, https://fred.stlouisfed.org/categories/32264.

[11] United Nations Development Programme, “Human Development Index,” https://hdr.undp.org/datacenter/human-development-index#/indicies/HDI.

[12] “Worldwide Governance Indicators | DataBank,” World Bank, https://databank.worldbank.org/ source/worldwide-governance-indicators.

[13] Alex Ward, “Nigerian President ‘spent $1million of Aid Money Meant for Poverty-Stricken Country on Star-Studded Festival Featuring Beyoncé and Jay-Z,’” The Daily Mail, February 23, 2013, https://www. dailymail.co.uk/news/article-2283453/Nigerian-president-spent-1million-aid-money-meant-povertystricken-country-star-studded-festival-featuring-Beyonc-Jay-Z.html.

[14] Luis Franceschi, “How Debt Relief Can Turn into Punctured Life Boat,” Nation, October 2, 2020, https://nation.africa/kenya/blogs-opinion/opinion/how-debt-relief-can-turn-into-punctured-lifeboat-2454904.

[15] Robert H. Jackson and Carl G. Rosberg, “Why Africa’s Weak States Persist: The Empirical and the Juridical in Statehood,” World Politics 35, no. 1 (Oct., 1982): 1-24.

[16] Gunter Bernhard, “What’s Wrong with the HIPC Initiative and What’s Next?” Development Policy Review 20, no. 1 (2002): 5-24.

[17] Sumit Roy, “Globalization, Debt Relief and Poverty Reduction: Concepts and Policies,” Debt Relief Now, n.d., http://www.debtweek.org/content/globalization-debt-relief-and-poverty-reduction-conceptsand-policies/index.html.

[18] Roy, “Globalization, Debt Relief and Poverty Reduction.”

[19] “Fragile States Index,” The Fund for Peace, https://fragilestatesindex.org/.

[20] “Worldwide Governance Indicators | DataBank,” World Bank, https://databank.worldbank.org/ source/worldwide-governance-indicators.

[21] Chris Heitzig, Aloysius Uche Ordu, and Lemma Senbet, “Report: Sub-Saharan Africa’s debt problem.”