Africa Finance Forum Blog

Currently the posts are filtered by: May 1
Reset this filter to see all posts.

Early Warning Systems and Banking Crises in Low Income Countries in Sub-Saharan Africa

20.05.2013Pietro Calice

The global financial crisis has stimulated renewed interest among academics and policymakers in early warning systems (EWSs) or models able to provide risk alerts on the potential for systemic banking crises on an objective and systematic basis. Most of the attention has been devoted to advanced economies, which have been at the epicentre of the recent turmoil. With the notable exception of the IMF, which has recently initiated a research program aimed at enhancing its financial sector surveillance framework, low income countries (LICs), particularly in Sub-Saharan Africa (SSA), have not been explored as of late.  

SSA LICs’ banking systems have on average proved resilient to the recent episodes of global financial stress. This is primarily the result of the structural reforms implemented by many countries over the past decade within a context of sound macroeconomic policies. Most countries have improved the regulatory framework for supervision, bolstering prudential requirements and supervisory rules. However, as banking systems deepen and new sources of risks materialise SSA LICs need to move ahead with their plans to strengthen supervisory capacity and financials sector resilience. Further actions are warranted to explore the impact of macroeconomic and financial developments on systemic banking risk. In this respect, EWSs can represent a valuable tool for policymakers and regulators in the region.  

In a forthcoming paper co-authored with Giovanni Caggiano (University of Padua) and Leone Leonida (Queen Mary University of London) we develop an EWS of new generation to predict systemic banking crises in SSA LICs. In particular, we focus on a sample of 38 countries during 1980-2008. Focusing on a group of homogenous economies contributes to improve the predictive power of EWSs. Moreover, despite having experienced a number of systemic distress episodes, especially during the 80s and 90s, SSA LICs have received no attention in the relevant empirical literature. We test for a number of macroeconomic and financial variables that have been widely used in the literature as well as for indicators proxying the structural characteristics of SSA LICs’ banking systems.  

We find that a decline in real output, a depreciation of the nominal exchange rate, high domestic credit growth, and banking system illiquidity are all positively correlated with the probability of incurring a systemic banking crisis in SSA LICs. Our findings indicate that banking distress in SSA LICs is typically preceded by weakening macroeconomic fundamentals, in particular a slowdown in economic activity and a depreciation of the currency. Given the low economic diversification of SSA LICs, any decline in GDP growth is felt by the banking sector through a generalised deterioration in asset quality, which increases the probability of a systemic banking crisis, other things being equal.  

On the other hand, a depreciation of the exchange rate lends credence to the hypothesis that banking crises in SSA LICs may be driven by excessive foreign exchange (FX) risk exposure. Our results show that currency turmoil can indeed trigger a banking crisis when banks are highly exposed to FX risk. This can affect the banking system directly, through currency mismatches in the balance sheet, or indirectly, through exposures to unhedged borrowers. Moreover, the high dollarization of most SSA LICs exposes them to the risk of a generalized run on their foreign currency deposits. Given that our proxies for currency mismatch in the banking system’s balance sheet (net open FX position) and for liability dollarization (M2 to reserves) are not statistically significant, we can infer that the channel through which a currency depreciation works in SSA LICs is primarily indirect, i.e. via exposures to FX risk by bank borrowers.  

Our results also show that rapid domestic credit expansion relative to GDP precedes banking crises. This may signal the peak of a credit cycle, where relaxed credit standards and build-up of financial imbalances in the real economy lead to widespread defaults and therefore systemic banking problems. Related to this, we find that the banking system in SSA LICs is crisis-prone if it engages in excessive credit activity relative to the deposit base. In other words, the probability of a systemic banking crisis increases when banks tend to finance an increasing share of their loan portfolio with non-core liabilities, resulting in low liquidity and/or vulnerability to deposit withdrawals.  

Our results have important policy implications at a time when financial regulators and central banks in SSA LICs are reassessing their financial regulatory agenda in the context of recent reforms spurred by the global financial crisis, in particular Basel III. Our findings underline the need to implement an effective macroprudential framework in addition to standard microprudential regulation, monetary policy and administrative measures to address the risks to financial stability emanating from undiversified and dollarized environments. In this respect, SSA LICs’ financial systems might benefit from credit-related measures such as sectoral ceilings on credit exposures and/or credit growth, and higher risk weights and/or provisioning rates on foreign currency lending to ensure adequate protection from concentration risk and currency-induced credit risk.  

In terms of microprudential regulation, implementing the liquidity requirements set out in Basel III might also be a useful complement to existing liquidity rules. On the other hand, our results do not warrant tightened capital requirements, which represent the cornerstone of recent international regulatory reforms. Finally, it is important that financial regulators in the region, with the assistance of international financial institutions, continue efforts to strengthen supervisory capacity, especially in contexts of sustained credit growth, as was the case in Nigeria before the 2009 banking crisis . Ultimately, EWSs are useful tools for policymaking but should never substitute the judgment of financial regulators.  

Pietro Calice is Principal Investment Officer with the African Development Bank

Developing a Sound Banking Industry in Africa: How could Development Financial Institutions Help?

07.05.2013Julien Lefilleur

Sub-Saharan Africa has the least developed financial sector in the world. With the exception of South Africa, the total African bank assets amount to less than USD 300 billion, which is nearly ten times less than the largest Chinese bank and about the size of the third largest Swedish bank. Even after taking into account the differences in GDP, the African financial sector remains very much underdeveloped, with a penetration rate around 30 %, twice inferior to the average figure in developing countries. In addition to its very small size, the African banking sector remains very much fragmented: the largest banking group in SSA totalled USD 17 billion – three times less than the first Cypriote bank – and only a dozen banking groups have total assets in excess of USD 5 billion.

Africa hosts more than 500 banks, including plenty of very small sized banks that are inefficient since they are unable to generate returns of scale, are not very innovative, and often display poor performances. These banks therefore cannot generate a healthy and competitive environment, and occupy low risk in very profitable niche markets, such as public debt, change or money transfer markets (Western Union, etc.), with hardly any impact on private sector financing. As a result of this underdevelopment in Africa’s banking sector, the amount of private sector credits does not exceed 20% of the African GDP, the lowest rate in the world. Furthermore, a significant portion of the financing needs of African economies – particularly in the areas of agribusiness, building and civil engineering, petrol and mining industries – are covered by external financial systems (donors, international banks, suppliers’ credits, etc.).  

Fostering New Pan African “Leaders”  

In order to fully benefit of the opportunities in the continent, the African banking sector should identify pan African “leaders” capable of carrying the pan-African capital-intensive operations. Development Financial Institutions (DFIs) can contribute, through targeted actions in order to support some banking groups that are well positioned to play this role. Apart from existing groups (Ecobank, BOA, UBA) – which already receive help from DFIs – two kinds of players are likely to prove good candidates: regional groups with an already critical size, and major banks in the African financial centres. Players in the first group include Orabank or BGFI Bank in Central and West Africa, I&M, Equity Bank or Kenya Commercial Bank in East Africa, and BancABC in South Africa. These banks have good knowledge of the markets but are limited in their means. They could benefit from capital injections by DFIs, such as FMO in Afriland First Bank, Proparco in I&M, and Orabank or IFC in Equity Bank. As for the main African financial centres, three countries could lead the way at a regional scale: Nigeria, Morocco, and South Africa. If the last two do not belong to sub-Saharan Africa, their geographical area represents their natural growth territory. Banks in these countries traditionally stayed within their geographical boundaries, and their pan African expansion is a recent development: Moroccan banks in Francophone West Africa, and Nigerian or South African banks in Anglophone Africa. However, their expansion over not well-known and not very accessible markets remains limited. The largest banks in these countries remain still very much interested in their domestic markets. This is where DFIs have a role to play in encouraging and supporting these banks outside their borders. Proparco and FMO supported BMCE's regional growth in this very approach, by contributing to the establishment of closer ties with the BOA Group. In the same way, Société Financière Internationale (SFI) invested in BCP Maroc to finance Group Banque Atlantique's buy-back in West Africa.  

Long Term Financing Needs  

Given the fact that DFIs contribute to the development of a local, pan African banking sector, their main actions up until now were essentially based on very traditional funding based on long-term debt and equity, in order to provide banks with the missing long-term resources. Today however, other forms of support solutions that are more innovative should be considered. First of all, giving the banks access to international markets could permit to lower their long term refinancing dependency of DFIs. Groups such as Ecobank offer fundamentals, that would enable them to obtain good financing solutions in these markets, but that are too little known by investors outside Africa. DFIs could help by integrating African banks in the international financing systems, for example by guaranteeing their first bond offerings. DFIs could also help local banks to better exploit home market available resources. For example, insurance companies and other social security funds dispose of important long term capital but – due to their missing knowledge of the banking sector –may be reluctant to invest them long term with banks. A significant portion of these resources, when not invested in developed countries banking markets, are placed locally in government debt securities. By guaranteeing loans lent by these institutions to local banks, DFIs would contribute to feed these long-term resources to the banking sector and finance a more productive sector. Similarly, most African banking sectors have excess short-term resources, which although very stable, cannot be used to finance long-term investment due to cash banking regulations. In order to make better use of these resources, DFIs could explore the possibility of offering liquidity guarantees: providing banks with refinancing in case of a liquidity crisis, which would enable them to transform a greater share of short-term resources into long-term ones. Finally, in order to help banks increase their resources' maturity, DFIs should also consider stimulating long-term interbank market refinancing, or encourage banks to develop local bond markets.   

Developing synergies between the various markets  

Developing the bank's resources is not the only way to increase their response capabilities. Better use of resources would also overcome the constraints imposed by their limited size. Developing the syndication market could prove especially useful in this regard. As syndication develops on a per country base, “transnational” syndication remains very rare: only bank groups with subsidiaries in several countries (like Ecobank, BOA Group, Standard Chartered, Standard Bank, etc.) are able to initiate them and these syndications generally remain within the group. DFIs certainly have a role to play to organise this market and encourage increased bank cooperation. They could, for example, develop their coordination capabilities by offering local banks intermediation services. This path has been little explored by DFIs, whose syndication attempts are usually limited to the projects they are involved in, bankable in hard currency. Yet, it would stimulate local currency finance markets, over which DFIs have very little control. From a more general point of view, and always with a view to develop the capacity of local banks, it seems possible to increase cross-border cooperation by encouraging banks operating in different markets to develop synergies between them. There are very few partnership examples, such as the one developed between Nedbank (based in South Africa) and Ecobank (present in the rest of sub-Saharan Africa). Most Ghanaian banks, for example, have no partner banks in the WAEMU region, even though Ghana is located in the middle of the area, which greatly limits the scope of the operations they can perform with their active clients in this region. DFIs could, for example facilitate regional trade by guaranteeing African vis-à-vis their banking counterparts in neighbouring countries (as they do for international trade between Europe and Africa, for example). Generally speaking, a transverse DFI approach would enable them to gather banks likely to develop synergies, and thereby stimulating regional banking integration and dissemination of knowledge between different markets. Such support would allow banks to know their neighbouring markets, and thus contribute to strengthening the capacity of African banking systems.

Julien Lefilleur joined Proparco in 2004, a French development financial institution that is a member of the French Development Agency Group. Having occupied a series of positions in Proparco - mainly in the Banks and Financial Markets department - he opened in 2010 Proparco's regional office for West Africa in Abidjan. Julien Lefilleur is also the Chief Editor of the Proparco magazine Private Sector & Development, which he founded. He graduated at École Centrale de Paris with a PhD in economics from the Sorbonne University.


What do renowned economists, financial sector practitioners, academics, and activists think about current issues of financial sector development in Africa? Find out on the blog - and share your point of view with us!


Harnessing the Buy Side’s Potential to Develop East...Jim Woodsome, Senior Research Analyst, Milken Institute
Getting ahead of the Curve: How the Regulatory Discourse on...Catherine Denoon-Stevens, Senior Research Associate, Cenfri