Bank of Uganda’s decision to take over the management of Crane Bank — the country’s fourth largest bank by total assets — was all too familiar.
Central banks in the region are having a torrid time maintaining financial sector stability. Over the past 20 months, there have been 15 regulatory seizures in eight countries — Kenya, Zimbabwe, DRC, Nigeria, Zambia, Mozambique, Tanzania and Uganda.
Some of these banks being taken over appear to be systemically important institutions — like in the case of Uganda, DRC and Mozambique. In all these cases, the central bankers have been at pains to reassure markets that everything is normal.
Banking is all about confidence anyway. For Uganda, the fall of Crane Bank now calls for five key adjustments in the regulatory environment, Bank of Uganda now needs to recalibrate commercial banks’ management of past-due loans.
In Uganda, commercial bank’s loan book performance is classified into five categories: Normal, watch, substandard, doubtful and loss. Normal refers to the good book; while doubtful and loss categories constitute non-performing loans. Past-due loans or watch are credit facilities with fixed repayment dates but, which have not been serviced by between a month and 90 days.
On the 91st day, it becomes a non-performing loan. Usually, a non-performing loan qualifies for a restructuring and is not subject to any specific provisioning. But there are some restrictions to it: only credit facilities under watch and substandard (in arrears for up to 180 days categories can be restructured. Substandard loans are subject to specific provisions of up to 20 per cent of total value.
However, the regulation limit restructuring to only two times during the life of a loan. In my view there is a loophole in allowing substandard loans to be restructured: It allows banks to under-provide, hence understate underlying credit risks as was evident with Crane Bank.
The bank’s credit risks disclosures showed that it continued to amass a huge portfolio of past-due loans, compared with reported non-performing loans. This should have been a red flag because the past-due loans should not tower actual non-performing loans. It’s a sign of risk understatement.
To correct this, Bank of Uganda needs to exclude loans classified as substandard from the list of restructureable facilities.
Second, commercial banks’ liquidity measurement regime needs a review. At the moment, banks in Uganda measure their liquidity positions by discounting current liquid assets against current liabilities.
Bank of Uganda has set the minimum ratio at 20 per cent. This is a flawed measurement framework because liquid assets also include encumbered assets, which are never realisable on demand. And, if you do liquidate, you’ll have to take a haircut. This often defeats the spirit of asset liquidity.
Consequently, liquidity measurement in post-Crane Bank, should now be based on a bank’s ability to cover for its liabilities as they fall due: On a three-month and six-month basis. And a bank needs to demonstrate that it has adequate unencumbered liquidity assets for coverage purposes. The coverage ratio should be 100 per cent at all times.
Third, to a large extent, at some point there’ll be a need for the Bank of Uganda to separate its financial institutions licensing and supervisory functions, by way of either establishing a registrar of banks or gradually ceasing to warehouse its supervision functions (or both). Separation of these two functions will be key in minimising any possibility of conflict between Monetary policy and supervision of the banking sector.
Fourth, credit committee should now be mandatory. Bank of Uganda’s corporate governance regulation, 2005, only prescribes four specialised committees — audit, asset liability management committee, risk and compensation.
The credit committee is crucial in approving credit requests of certain limits-mostly large corporate files; taking into consideration a full understanding of the applicant’s risk profile. At the moment, this function rests within the risk committee.
The risk committee, to me, is always inundated. Credit risk oversight needs to be standalone. Finally, as part of corporate governance practices, beneficially majority owners, whether directly or indirectly, and who own more than five per cent of a bank and have a board representation, in my view, should not be allowed to sit in either risk or credit committees.
In the Crane Bank case, if indeed Dr Sudhir Ruparelia, who controlled 48.67 per cent of the bank, sat in any credit-approving committee, then I would imagine how board credit decisions may not have been fully independent.
This is also important where beneficial majority owners of a bank have strong peripheral businesses, as was in Dr Sudhir Ruparelia’s case. You just don’t know how those businesses are funded.
George Bodo is a banking sector analyst.