The problem of the high and growing level of non-performing loans (NPLs) is the most difficult issue confronting the Bangladesh financial system.
Of course, it is not a new problem.
Discussions and writings about the financial sector have emphasized this problem for decades.
One might first ask why there is such a high level of NPLs in the Bangladesh banking sector.
There are a lot of views on this issue: Some believe the borrowers are dishonest and simply avoid paying the loan back by one device or another; the most common reason advanced is the access to persons of power who will for a price protect the borrower from harassment.
Others suggest that the slow actions in carrying out their normal function by the government cause delays that undermine the profitability of the project.
The high cost of doing business reported in the World Bank surveys reflects costs that cannot easily be incorporated in the financial analysis of a project.
Some suggest that the procedures followed by banks slow down implementation and make implementation of complex projects slower.
Entrepreneurs starting projects in Bangladesh find that the time to get things done is far longer than they planned.
All of these factors delay the start of earnings for the project and throw off the repayment schedule; often firms never recover from the slow start.
The government itself is notorious for completing projects late and usually at much higher cost than the original plan.
Delays and cost increases can significantly reduce the return on a project.
As a simple example consider a project that costs Tk1 crore and one year to construct and earns Tk25 lakh net of costs.
The return on this project is 25%.
Delay the project by one year and the return drops to 22%; delay one year and increase the cost by 20% and the return declines to 19%.
Rather than blame the businessmen, those working on these matters should first try to understand the real reasons why loans went bad.
I believe that the result will be recognition that the inefficiency and corruption involved in processes of getting started are an important element.
The bureaucracy, taking a haughty superior stance over businessmen who they perceive as corrupt, are themselves partly responsible for the difficulties of getting projects going on time.
One indicator of this problem is the inability of the senior authorities to require government organizations to manage in a speedy way application for services of investors.
This matter has been argued about for 40 years and while some progress has been made it is far from enough.
The bureaucrats have not been willing to do their part in enabling investments to get started promptly.
The arithmetic of non-performing loans is clear: During the pandemic many companies found revenues declined and it was impossible to repay loans.
Bangladesh Bank allowed banks to avoid classifying loans during this period.
However, once this forbearance, allowing banks to not classify loans as NPLs, is lifted the volume of NPLs will increase sharply.
But the key point is that to improve the recovery of loans one must know why the loans become non-performing.
This seems simple enough but in fact there is little known,
Inability to repay
It should have been clear as the pandemic got underway that many small companies would never be able to repay the loans without a significant reduction (forgiveness) in the size of the loan.
A good part of the loan portfolio would become non-performing and delays in repayment would never solve this problem.
I wrote at the time of the absolute necessity of a loan forgiveness program for small loans, pointing out that the potential for recovery of these small borrowers was unlikely.
That is what is happening!
If the IMF and the central bank do not realize this then a large number of small businesses will cease operation, be branded as loan defaulters and have difficulty obtaining credit.
The statistical framework of the Bangladesh Bureau of Statistics (BBS) is insensitive to the condition of small businesses so estimates of value added in manufacturing can sail on higher and higher when the reality is a crash landing.
The banks cannot solve this NPL problem without assistance from the courts.
The money loan courts are supposed to dispose of loans in six months, but it takes five years.
First, the NPLs in the state banks are very high.
There has been only one international standard review of the lending from a state-owned bank and that was done for Rupali Bank at the time of attempted privatization.
The NPLs came to 60%!
One should forget for the moment trying to improve the state-owned banks and concentrate on the private banks.
There are six things to do:
| 1. Do a careful study of 2,000 loans to find out what has happened in the borrower's attempts to carry out a project and repay the loans. Rather than guess one should make a serious attempt to find out. |
| 2. The money loan courts should have many more judges; a detailed computer record of all cases should be maintained and regular reports prepared of the status of the cases. There should be regular training for judges; a research unit should study the cases and determine the source of difficulty in resolving such cases. These cases should be simple (there is clear evidence that the money was borrowed and that it has not been repaid; what is there to argue about?) take a few hours to conclude, not years. Work on improving the bankruptcy court can wait until the money loan court's function better. |
| 3. The rules for rescheduling should be redone. Currently the rules set down conditions for how much of the outstanding amount must be repaid, how long the loan can be extended, etc. technical issues. Arguments about these conditions are a waste of time as none of this touches the essential issue. A loan that is non-performing must be seen within the overall economics of the project or the company, not conditions dealing with the loan. It is naive to believe that the rescheduling will improve the performance of the project or the company. Hence a rescheduling must include a series of steps that will improve the probability that improved conditions in the operation of the company will generate the returns to pay the loan. This might include cost cuts reducing staff or staff compensation, reducing the scale of the project, increased investment by the owners, tighter quality controls, etc. |
| 4. Refrain from changing the rules for determining if a loan is non-performing. It would seem that first one should try to improve the processes for collecting NPLs before taking off to increase the volume of these. Inspections to ensure the current rules are properly implemented, however, should certainly be managed. The IMF and World Bank always try to make conditions tougher. A moment of thought indicates how inappropriate such an action is. Try to improve recovery and then think about tightening conditions. |
| 5. Central to improved loan recovery is more accurate and better estimates of the value of the collateral. At present valuations are irregular, nor readily reproducible. Often there remain claims on the collateral that have not been adjudicated. It may be possible to pass legislation that would establish a process where collateral is cleaned prior to the loan being granted with a period of time allowed for challenges. Once the collateral is cleaned of competing claims, then the bank would have permission to seize it when the borrower defaults. All this requires clear definitions and conditions. The point is to make the seizure of the collateral simple enabling the money loan courts to dispose of cases rapidly. |
| 6. Finally, is the issue of writing off loans and how continued effort at recovery can be established. At present rules for writing off loans require a certain level of provisions and some percent of the estimated value of the collateral. Experience indicates that the real value of the collectable collateral is a bit less so the impact of the write-off on the balance sheet of the bank is not too large. The loan is still to be collected, the borrower is not off the hook after the write-off. But in reality, little progress is made in such loan collection efforts. One approach that has been discussed is the establishment of a company that would collect these written off loans, keep a substantial portion and pay something to the lending bank. The IMF is reported to have recommended the establishment of such a company and there have been various proposals along similar lines. |
A few points on such loan collection companies:
- There is no reason to believe that a government owned company carrying out this task would be successful.
Two or three private companies are likely to be more successful.
A government company would take years to establish and staff; based on experience 7-10 years is needed. - It is not clear what the courts would think of a government-owned company collecting the loans.
- Better for the bank to sell the NPL to the private collection company allowing the collectors to keep whatever they can obtain.
Conclusion
There is no doubt that the high level of NPLs is the most important problem facing the financial sector.
The high level of non-performing loans reduces the growth rate of the economy.
At the level reached in the Bangladesh banking system with the savings rate near 30% the loss of the growth rate is 1.5-2%.
The second point is that there is a cost to the banking system of 6.5-7% of a loan.
For the best private banks, the loss is 5-5.5%.
An interest rate that covered these losses would result in an average lending rate of 12-13%.
At present the banks are not earning enough from their loans to cover their cost.
I do not understand why any private bank is lending except to high quality corporations.
The banks cannot improve the loan recovery without assistance; blaming the banks for the level of bad loans is a complete misunderstanding of what is happening.
What is needed is a thorough improvement of the judicial management of default cases.
The six recommendations are suggested for NPLs held by private banks.
For state-owned banks it is essentially a matter for the Ministry of Finance to resolve.
Forrest Cookson is an economist who has served as the first president of AmCham and has been a consultant for the Bangladesh Bureau of Statistics.