Dhaka Metropolitan Police’s own records, reported in May 2026, list 1,280 people actively running extortion rackets across the capital, 148 of them armed. They operate in named groups, each controlling a defined territory, with rates that range from BDT 200 a day from a pavement vendor to BDT 20,000 a month from a business in areas like Gulshan and Banani. This is not an outside estimate or a reconstruction after the fact — it is the police’s own accounting of an economy that runs in daylight, street by street, inside a system whose stated job is to shut it down. The question this raises is not whether extortion exists in Dhaka; the police’s own numbers already answer that.
It is why an enforcement apparatus that can name 1,280 individuals and knows exactly which streets they control has not managed to dismantle the extortion network those individuals run. But police have not been idle. Operation Devil Hunt, launched in February 2025, produced more than 9,000 arrests within a matter of weeks. A nationwide drive launched in May 2026 added more than 18,000 arrests on top of that. In June 2026, alone, Dhaka police reported 2,314 arrests, of which 1,446 were for drug offences. In May 2025, the army arrested Subrata Bain, one of the police’s own listed “top criminals,” along with an arms cache. Read on their own, these are the kind of numbers that read as proof of resolve, and by the standard of activity, they are real.
Read against the extortion figures, they look different. None of the reporting from this period — not the coverage of Devil Hunt, not the nationwide drive, not the June arrest tally — documents a sustained fall in the crimes these operations targeted. An arrest count is not a measurement of reduced extortion, and nothing in the public record connects the two. Part of the explanation is supply. More than 1,000 weapons looted from police stations during the August 2024 unrest remain unrecovered, and reporting has linked the resurgence of armed groups to convicted criminals who were released or returned to the country after the political transition.
Those two facts sit next to each other for a reason: a network can absorb a few thousand arrests a year and still function if its weapons stay in circulation and its ranks keep refilling. A structure of 1,280 people, 148 of them armed, does not shrink because individuals inside it are cycled through arrest. The people are replaceable; the territories and the rates charged in them are what would actually need to be disrupted, and neither shows up anywhere in this record. Mass arrests and lasting disruption have turned out, in Dhaka’s extortion economy, to be two different things, and so far only one of them is happening at scale.
None of this happens in isolation from how Bangladesh’s law-enforcement institutions actually function day to day. Transparency International Bangladesh’s 2025 household survey, covering November 2024 to October 2025, found that 69.4 percent of households who sought law-enforcement services, and 71.3 percent who sought judicial services, reported paying a bribe to get them. Bribery for public services nationwide totalled an estimated BDT 12,633 crore over that year, and for households below the poverty line, bribes to law enforcement alone consumed an average of 34 percent of monthly income.
Put that beside the extortion figures. Roughly seven in ten households who go to the police to get something done — a report filed, a complaint acted on, or protection sought — pay informally for it. That is not a description of a police force with a corruption problem at the margins; it is a description of an institution where informal payment is close to the default mode of transaction. An organization that runs, in most of its everyday interactions with the public, on the same currency the extortion economy runs on and money paid outside any formal channel to secure an outcome, is not well positioned to dismantle that economy from within.
The distinction between a resident paying an officer to get a report filed and a shopkeeper paying a racket to be left alone is one of framing, not of mechanism. Both are payments made because going through the system as designed does not reliably work. It is not required to assume individual officers are complicit in extortion rackets specifically; the survey makes no such claim, and neither does this piece.
What it shows is an institutional culture in which informal payment already substitutes for formal process across a majority of interactions. Dismantling 1,280 people’s worth of organized extortion asks that same institution to behave, in this one domain, entirely unlike how it behaves everywhere else households encounter it. Arrests can happen inside a system like that — and they have. Sustained disruption of an income stream this large is a different kind of task, and a bribery rate above two-thirds is not evidence that this institution is currently built to do it.
Abstract numbers change shape once you leave Dhaka. In the industrial city of Gazipur, an investigation linked a rise in street crime directly to factory closures and job losses: 72 factories closed, and with them, roughly 200,000 jobs disappeared. People who used to walk to a shift now walk the same streets with nothing to walk toward. The same investigation documented the killing of a journalist who had filmed muggers at work. He did not survive it. Gazipur is where an enforcement gap that shows up as a percentage point in a report shows up, on the ground, as a factory floor gone dark and a life lost for doing, with a camera, what the state itself has struggled to do at any scale: point directly at who is responsible.
None of this points to a shortage of state capacity. Bangladesh has shown it can mobilize thousands of arrests within weeks when it decides to. It points, instead, to a state measuring the wrong thing. An enforcement effort that reports arrests, rather than whether a specific racket in a specific territory still collects money six months later, will keep producing headline numbers that never touch the underlying economy.
To end extortion successfully in Dhaka, the very extortion rackets need to be dismantled. But two changes are needed from what is documented here. First, both things already identified here as keeping this network replenished, more than 1,000 looted weapons still in circulation, and the return of released or returned convicts into the same rackets, need to be tracked and reported on their own, separate from arrest tallies. Neither shows up in an arrest count.
Second, extortion complaints need a reporting channel independent of the same police structure that a majority of households report paying informally to deal with; a channel that publishes recovery and prosecution rates by territory, not arrest counts, would let the public judge whether a given racket is actually shrinking rather than merely whether officers have been busy around it.
DISCLAIMER: The views expressed are those of the author(s) and do not necessarily reflect the official policy or position of the Magazine and its editorial team. The views published are the sole responsibility of the author(s).
