The Banking Crisis and Private Power
This second article in a three-part series argues that bad loans, political patronage, and cosmetic accounting turned Bangladesh’s banks into a public crisis.
I will admit that when I first started looking at Bangladesh's banking numbers, my instinct was to move on quickly. Non-performing loan ratios, capital adequacy requirements, Basel III classification standards -- it all sounds like the kind of thing you read in an IMF staff report and forget by lunchtime.
But banking crises stop being technical the moment they enter ordinary life. They show up in the depositor who wonders if her savings are safe. In the small manufacturer in Narayanganj who cannot get a working capital loan. In the entrepreneur who is quoted 16/17% interest and decides the risk is not worth it.
And eventually, in the taxpayer who is quietly asked to absorb losses that someone else created.
That is why this part of the story matters. It is not a side chapter to the wider economic crisis. It sits right at the centre of it.
For a long time, if you asked anyone in an official position about the banking sector, you got more or less the same answer. Yes, there were problems. Yes, some loans were not performing. But the sector was broadly stable, the issues were manageable, and a few policy adjustments here and there would sort things out.
I have heard versions of this line for years. I suspect most people who follow Bangladesh's economy did too.
However, the newer numbers make that reassurance much harder to sustain. By September 2025, non-performing loans had risen to Tk 644,000 crore, close to 36% of all outstanding credit, while the system-wide capital adequacy ratio had fallen to 4.5%, less than half the regulatory minimum.
That is not a normal level of stress. It points to a deeper failure in the way public institutions were governed. In Bangladesh, the banking problem cannot be understood simply as a neutral market accident. It has to be seen alongside political power, regulatory weakness, and the longstanding protection enjoyed by large and well-connected business groups.
For years, loans were rolled over, rescheduled, or dressed up in ways that kept the official picture from looking as bad as the underlying reality.
This matters because when loan classification rules tightened after the political transition of 2024, the official numbers changed sharply. The jump was dramatic, but the rot did not begin at that moment. The stricter standards mainly exposed how much distress had already been building underneath. In other words, the crisis was not created overnight by a new reporting method. It was revealed more clearly by it.
To understand why, you need to know one technical detail. Under the old Bangladeshi rules, a loan was not classified as non-performing until payments were overdue by 270 days -- nine months. Under the international Basel III standard, that threshold is 90 days. The difference is enormous.
Under the old system, a borrower could stop paying for the better part of a year and the loan would still appear healthy on the bank's books. When the interim government adopted the 90-day standard after August 2024, it was like turning on a light in a room that had been kept deliberately dark. The loans had been bad for years. The classification simply caught up with reality.
I think Figure 3 tells this story more clearly than I can in words. Look at the red line -- the NPL ratio -- between 2015 and 2023. It barely moves. It hovers in a narrow band around 8 to 10%, as if the banking system were in a mild but manageable state of discomfort. But many people who followed the sector closely -- economists, journalists, a few brave Bangladesh Bank officials -- suspected for years that the real picture was much worse.
Then, once tougher recognition standards and political change made concealment harder, the line shot upward. At the same time, capital adequacy moved in the opposite direction and fell below the regulatory floor. The divergence between those two lines tells us something simple but important: The banking crisis is not just about more defaults, but about a system that no longer has enough real cushion to absorb them.
Figure 3: The scissors of crisis -- NPLs surge as capital adequacy collapses (2015–Sep 2025)
The picture becomes even clearer when we ask which institutions are carrying the heaviest burden. Figure 4 breaks the problem down by bank category. State-owned commercial banks sit in the worst position, with an NPL ratio of 44.4%. Foreign banks, by contrast, stand at 4.5%. That gap is too large to dismiss as bad luck or a few isolated errors. It points to governance, oversight, and political interference as central parts of the story.
I keep coming back to the word "capture," because I think it describes what happened more accurately than words like "mismanagement" or "inefficiency." This was not a system that failed accidentally. Public institutions that were supposed to mobilize savings and direct credit toward productive activity were captured -- bent deliberately toward patronage, accommodation, and protection for powerful borrowers. The names are known.
The S Alam Group, Beximco, Bashundhara -- these are not obscure entities. They are among the largest and most politically connected conglomerates in the country. When those loans sour, the losses do not remain private. They are pushed outward and downward. The public balance sheet absorbs them, and ordinary people pay through inflation, weaker credit conditions, poorer services, or future recapitalization.
Figure 4: NPL ratio by bank category (December 2025)
The merger of five distressed Islamic banks into Sammilito Islami Bank PLC in September 2025 is worth pausing on, because it shows exactly how this kind of damage gets socialized. The estimated capital injection was around Tk 35,200 crore -- an enormous sum. You can describe that as stabilization, and in a narrow operational sense it is.
But step back and look at what actually happened. Of that sum, Tk 20,000 crore came from the government -- meaning public funds: Private individuals extracted reckless loans from institutions they effectively controlled, those loans went bad, and the public was then asked to recapitalize the wreckage. The gains were private. The losses are collective.
The key question is always the same: who took the gains when lending was reckless, and who carries the burden when the bill arrives?
I think it is important to say plainly who the human face of a banking crisis actually is. It is not the conglomerate owner -- he has lawyers, offshore accounts, and political connections to cushion the fall. It is not the politician who arranged the loan -- she has moved on to other concerns.
The human face is the small business owner who walks into a bank branch in Narayanganj and is told there is no credit available because the bank is nursing its wounds by parking money in government treasury bonds.
It is the depositor who reads the newspaper and wonders whether her savings are safe. And it is the garment worker whose employer, cut off from fresh working capital, delays wages for weeks. As Kalpona Akter, executive director of the Bangladesh Center for Workers Solidarity, has said: "We don't want relief programs, we don't need any training, we just want dignified wages. We just want the laws to be implemented and protect us."
When the banking system fails, it is not an abstraction. It filters down through delayed salaries, shuttered factories, and families who have to choose between rent and food.
Bangladesh also looks troubling in comparative perspective. Figure 5 places the country beside other benchmarks, and the contrast is stark. At 35.7%, Bangladesh’s NPL ratio is far above the Asian average of 1.6%. India, after extensive clean-up efforts, is down to 2.5%.
Even countries dealing with severe political or economic strain do not sit near Bangladesh’s level. That does not mean international comparison explains everything, but it does make one point unmistakable: This is not a routine banking problem.
Figure 5: Bangladesh NPL ratio vs. global comparators (latest available)
What worries me most about these numbers is not the embarrassment -- though it should embarrass anyone in a position of responsibility. It is the downstream consequences. A banking sector this damaged does not just sit there as a problem on a balance sheet. It actively chokes the real economy.
Credit dries up. Banks stop taking risks on new enterprises. Lending rates climb to levels where only the already-wealthy can afford to borrow. The IMF has already spoken of vulnerabilities reaching macro-relevant proportions, including fraud and embezzlement. Once that happens, finance stops acting as a support for development and begins acting as a drag on it.
The scale of the dysfunction becomes clearer when you consider where bank lending actually goes. Rather than extending fresh credit to manufacturers, small businesses, or entrepreneurs -- the kind of productive investment that generates jobs and output -- large banks have retreated into high-yield government treasury bonds. It is safer, more predictable, and requires no due diligence on a borrower.
But it means that the banking system, instead of channeling savings into the real economy, is recycling public debt while the private sector is starved of capital. Private investment fell to 22.48% of GDP in FY2025, its lowest level in five years. That is not just a banking statistic. It is a direct constraint on employment, wages, and future growth.
I should be fair here. Not every bank is equally rotten, and not every large borrower is a crony. There are well-run private banks in Bangladesh, and there are businesspeople who repay their loans and run their operations honestly.
But those individual exceptions do not alter the structural point. The pattern is too broad, too persistent, and too costly to treat as a series of individual failures. Bangladesh’s banking crisis reflects a development model in which political access often counted for more than productive discipline.
That is why simply throwing public money at the problem -- recapitalizing banks without asking hard questions about where the money went -- will not solve anything. I have watched this pattern before, not just in Bangladesh but in other countries where I have followed financial crises. The balance sheets get patched. The worst offenders get quietly rescheduled.
The regulators declare the system stable again. And five or ten years later, the same thing happens, because nobody dismantled the political machinery that produced the crisis in the first place. Without forensic audits, serious asset recovery, and genuine insulation of the central bank from political pressure, any cleanup will be cosmetic. And the public will be asked to pay again.
A serious response would begin with transparency. It would identify where losses were hidden, who benefited from repeated rescheduling and regulatory leniency, and which institutions were used as channels of patronage. It would also reorient finance toward productive sectors rather than leaving banks as holding spaces for bad assets and protected relationships. Without that, any promise of reform will remain shallow.
In the first part of this series, I looked at how Bangladesh's growth model squeezed labour while delivering diminishing returns on investment. The banking crisis is not a separate problem. It is another face of the same structure -- a system in which capital was not allocated on the basis of productive merit but on the basis of who you knew and how close you sat to power.
Instead of allocating capital efficiently, large parts of it were captured, distorted, and used to protect private power. That is one reason why the economy now feels brittle from so many angles at once.
Jamil Iqbal is a UK-based researcher and writer from Bangladesh whose work spans qualitative research, public policy, and social analysis.
Sources for Figures:
Figure 3: NPL ratio data from Bangladesh Bank quarterly reports and CEIC Data. Capital adequacy ratio from CEIC Data. NPL ratio of 35.73% as of September 2025 reported by Bangladesh Bank on 26 November 2025 (Bonik Barta, 27 November 2025). Capital adequacy of 4.5% from CEIC and BBF Digital, "Why 2025 Marked a Turning Point for Bangladesh's Banks," March 2026.
Figure 4: Bangladesh Bank quarterly data via Pressenza, "Bangladesh Sees Sharp Fall in Defaulted Loans," March 2026. Breakdown: State-owned commercial banks 44.44%, private commercial banks 28.25%, foreign banks 4.51%, specialized banks 30.60%. All figures as of December 2025.
Figure 5: Bangladesh (35.7%) from Bangladesh Bank via Bonik Barta, 27 November 2025. Regional and global comparators from Asian Development Bank, "Nonperforming Loans Watch in Asia 2025," as reported in The Daily Star, "Bangladesh Tops Asia's Bad Loan Charts." Asia average of 1.6% from ADB.
What's Your Reaction?
