Harbingers of doom? Bank failures in Africa – how to interpret these

Yesterday, Zambia’s central bank announced it had taken over a commercial bank, Intermarket, after the latter failed to come up with the capital it needed to satisfy new minimum capital requirements. Three weeks ago, a Mozambican bank – Nosso Banco – had its licence cancelled, less than two months after another Mozambican bank, Moza Banco, was placed under emergency administration.

At the end of October, the Bank of Tanzania stepped in to replace the management at Twiga Bancorp, a government-owned financial institution which was reported to have negative capital of TSh21 billion.  A week before that, just over the border in Uganda, Crane Bank, with its estimated 500,000 customers, was taken over by the central bank, having become “seriously undercapitalised”. In DR Congo, the long-running saga of BIAC, the country’s third largest bank, continued in 2016, forced to limit cash withdrawals after the termination of a credit line from the central bank. And in Kenya, Chase Bank collapsed in April, barely six months after the failure of Imperial.

How are we to interpret this? It seems that 2016 is the year in which latent fragility in Africa’s banking sectors is being laid bare.  After years in which observers have favourably contrasted the relative stability of African banking with the financial sector chaos in Europe and the US, it seems that three critical perils – mismanagement, political interference and economic woes – are conspiring to transform the landscape of African banking into a decidedly treacherous place for depositors and investors.

We have had remarkably few bank failures in Africa in recent years and yet this sudden uptick in stories like Crane and Chase, against a backdrop of economic challenges in many places, raises the question as to whether there is worse to come.

Mismanagement and/or political interference have been at the root of most bank collapses over the past few decades. Martin Brownbridge’s grimly fascinating analysis[1] on this subject from 1998 concluded that “moral hazard, with the adoption of high-risk lending strategies, in some cases involving insider lending” was behind most of the bank collapses in the 1990s. This certainly resonates today. Catastrophic lapses in governance rather than economic malaise are alleged to be behind the recent Kenyan bank failures (although their shareholders and directors vigorously refute this) – but how else can you explain why a small number of banks fail when the sector as a whole has been returning well over 20% on its equity for the past several years?

There are some excellent programmes like Accions’s Africa Board Fellowship Program, which aims to strengthen capacity at financial institutions because their promoters understand that weak governance undermines trust in the financial system and is therefore very bad for financial inclusion. But it is one thing to know what you’re supposed to do as a bank board director – quite another to actually do it.

Each bank failure seems to have its own special story – and we derive comfort from this. It is somehow reassuring to think that that might be the case because the prospect of a system-wide failure is so awful.

And each country context has particular features that impinge on the stability of the financial system. There are deep concerns in Kenya, for example, that the recent imposition of interest rate caps is going to result in a very messy period of bank failures and/or consolidation.

But are there common patterns that we should be taking note of?  Is there a system-wide issue that we should be facing up to?

Well, one pattern might actually be positive – that central banks are intervening more, and more quickly, to weed out the miscreants, less cowed by the politicians than they might have been in the past and more concerned to protect their well-earned professional reputations. Another is that central banks are finally implementing the increases in minimum capital requirements which many have been talking about for years with the inevitable intended consequence that some banks will be forced to get out of the market.

These might be two good reasons why we are seeing more collapses. You could say that’s excellent news for the future of African banking. But perhaps only to a point. There is still the risk that the cumulative effect of bank failures as a result of zealous supervisory action causes a loss of faith in the entire system resulting in mass panic and the withdrawal of deposits and credit lines.

Also, the inevitable result of this would be fewer, bigger banks which may have negative consequences for competition and access – although it worth pointing out that Tanzania, which has 55 commercial banks, still only manages to bank around 12% of its adult population (FinScope).

The more concerning issue is the impact of underlying economic weakness. Leaving aside the paradox that some of these bank failures are taking place in economies that are growing quite fast (Kenya and Tanzania forecasting 6-7% GDP growth), lower commodity prices and their pervasive impact across African economies are going to make life much tougher for banks – especially if they are poorly managed and have political skeletons in their cupboards.

One problem we have, especially when economic conditions are changing fast and for the worse (as in Mozambique), is that data is often out of date and is not sufficiently disaggregated. So, when we look at Africa as a whole, or even the banking system of one country as a whole, the averages we tend to look at create a blithely benign picture which masks dramatic variations.

So, non-performing loans (NPLs) across Africa up to 2014 were a little over 5% but NPLs in Ghana were more like 11-12%. NPLs in Tanzania are currently a little over 8%, yet Twiga Bancorp’s NPL’s were – unbelievably – at 34% in early 2015, according to media reports.

We think the African banking sector is in for a rocky ride in 2017 and 2018 and, in the short term, this is not good news for the real economy. However, one industry that is set to grow, surely, is central banking supervision.

[1] Brownbridge, M (1998): “Financial distress in local banks in Kenya, Nigeria, Uganda and Zambia: Causes and implications for regulatory policy” Development Policy Review, vol. 16, no.2


  • Swifter actions to identify and resolve non-performing loans as they arise is critical.

  • Mark, this is interesting and insightful! It does look a bit rocky for the next few years. I hope that the growth of bank supervision is accompanied by a growth in central banks’ ability to use technologies available for better oversight (such as regtech vendors increasingly provide), rather than simply in the cost of compliance for the surviving banks!

  • Good article Mark. I see a few significant challenges:
    — trust is crucial for financial inclusion, so every failure where depositors lose money is a real disaster for consumer confidence. This is an area where central banks must step up and ensure adequate and fast deposit insurance payouts and resolution action.
    — have a look at the returns. They are markedly high, showing that there is much more noise about banks pricing risk and services than reality.
    — consider how the banks do product innovation, or not. We see a bit too much herd mentality and not enough consumer-centricity and piloting with banks, and central banks not using a sandbox approach to new services.
    — reporting timelines are always a problem; banks vary markedly in their automation spend, and central banks in their insistence on automation links to their own reporting databases. This needs to change.

  • Excellent analysis, Mark. I would emphasise the corporate governance failures here, in the broadest sense, and focus on the four pillars of Responsibility, Accountability, Fairness and Transparency. Boards, Management and Shareholders must invest more in this. Ethics and sustainability need to be top agenda items.

  • While the current commodity price crisis has brought banks into more severe difficulties than for a number of years, it’s difficult to see the collapse of these banks as “harbingers of doom”. As demonstrated by the cited cases in Mozambique, Tanzania and DRC, banks in Africa are usually only subject to sanction once their capital is negative. Indeed in most cases sanctions are only imposed once banks become illiquid. Are central banks really “intervening more, and more quickly”? Is the fact that there have been “remarkably few bank failures in Africa in recent years” a sign of strength or rather pervasive supervisory forbearance? Unfortunately, it is rather the latter – a sign of silent distress.

    It’s true – as pointed out by Martin Brownbridge so many years ago – that in many cases it is “catastrophic lapses in governance”, such as insider lending, that brings banks to their knees, but should bank supervisors only be addressing bank weakness once it becomes ”catastrophic”? Surely the purpose of having a capital buffer is to allow banking supervisors and bank management to address weaknesses while they are nascent rather than catastrophic, and while bank owners still have ‘skin in the game’? Instead shareholders of banks in many sub-Saharan Africa countries continue to extract sizeable rents long after the level of capital has fallen below supervisory minimum thresholds. So while bank collapses may have become more frequent in recent years, banking supervisors have not even embarked on the task of identifying and addressing bank weakness. Among the reasons are:

    • Weakness in the ability to assemble reliable data about the situation of banks and to analyze this data. Given the potential importance of banks in allocating financial resources, the attention paid by banks and their supervisors to the quality of data that banks report and publish about their financial situation is remarkably weak. Banks are potential conduits of capital to the local economy, and yet data about them is often difficult to find, and where analysis of the data is available, it is fragmented and partial.
    • The authority of supervisors to sanction banks that demonstrate weaknesses needs to become more rule-bound and less discretionary. Banking supervisors rarely exercise their authority when banking difficulties arise. There is a tendency to delay and little appreciation of the potentially high costs of delayed intervention – the less capital they stand to lose, the greater the incentive of bank owners and managers to ‘bet the bank’ by taking potentially hazardous risks.
    • The process of bank resolution is uncertain and lengthy. If banking supervisors are to take early and effective action in sanctioning banks, they also need to be sure that, should the need arise, banks can be efficiently resolved. Exit needs to be a viable option. In almost all countries the process of exit embroils banking supervisors in court challenges and resolution authorities in tortuously long and uncertain processes. They have every reason to want to avoid this sanction and this is one of the reasons it is so rarely applied. So measuring the health of banks by the rarity of bank closures is quite misleading.

    Altogether the current wave of bank closures only scratches the surface. The focus needs to shift towards strengthening the quality of the supervisory process and the value-addition provided by the banking sector. Supervisors need to have the authority to sanction banks on a level playing field: sanctioning well-connected owners and borrowers as they would those that do not have such connections.

  • Great article that gives important insights into the fast changing banking scenario in sub Saharan Africa.