Bangladesh's government has started a serious conversation about deregulation. On August 9, the Cabinet Division formed a 22-member National Taskforce on Deregulation and Business Facilitation. Finance and Planning Minister Amir Khosru Mahmud Chowdhury is heading the taskforce. Other taskforce members are senior officials from Bangladesh Bank, NBR, BSEC and RJSC, along with representatives of major business organizations.
The taskforce will review existing rules, regulations, and procedures around business and investment, identify unnecessary hurdles, and develop a roadmap for removing or simplifying them.
This follows the deregulation package announced in the FY2026–27 budget. The package includes a single window for approvals, service-level agreements for government services, deemed approval when agencies miss deadlines, and an online grievance system for businesses.
There have already been some movements. BIDA's One Stop Service has brought services from dozens of government agencies into one platform. BanglaBiz has taken that idea further by connecting the investment-service portals of BIDA, BEZA, BEPZA, BHTPA and BSCIC. The government has also introduced a common business identifier through BanglaBiz.
Regulatory overreach and complex regulatory requirements have been a major challenge to entrepreneurship and private sector growth in this country. Successive governments talked about ease of doing business for years. But after the talks, things hardly change.
A founder wants to start a company, then begins running around for a myriad of licenses. Trade licence. Tax registration. VAT registration. Sector-specific licences. Then annual renewals. Inspections from several departments if you operate in certain sectors.
Of course, licenses are necessary. In fact, none of these requirements looks outrageous or even unnecessary on its own. But together, they create a system that consumes time, money and attention. In many instances, the design and structure are unhelpful and can be significantly improved without compromising the regulatory objectives.
Future Startup has written about these challenges from different angles for years. Our recent work on Bangladesh's startup policy argued that bureaucratic complexity, opaque licensing and policy uncertainty remain among the structural constraints on entrepreneurship and investment.
In another instance, our analysis of the new Travel Agency Ordinance tries to point out that regulation can address legitimate problems and still create new problems when its requirements are designed for sophisticated, well-capitalized companies while the actual market contains thousands of small businesses.
These nuances are important to consider in policymaking. At the end of the day, every policy decision is a trade-off and comes with real consequences for people and the economy.
We don’t suggest good deregulation means removing every rule. Regulations are necessary. But it is also worth asking whether every rule, form, licence, inspection and approval is doing enough useful work to justify the cost it imposes, both the implementation costs on the state and costs on the businesses and people trying to do something.
To that end, the current taskforce has an opportunity to start with some remarkably ordinary problems.
Here are nine.
Inspections serve important purposes. Factories need safety checks. Food businesses need health and safety oversight. Financial institutions need serious supervision. Environmental risks need monitoring.
However, one common problem many early-stage entrepreneurs and small businesses face is coordination. A business can have several agencies looking at different parts of the same operation. Each agency may have a legitimate mandate with its own timeline. If approval depends on clearance from all different agencies, the entrepreneur gets the combined burden.
For a large company, this is a compliance function. For a small business or someone getting started, it can eat up valuable time, energy, and resources of the company. In many instances, people get drained of both energy and resources before laying a single thing in operation.
Moreover, every physical interaction between an entrepreneur and a government official creates room for discretion. It doesn’t mean improper conduct on the part of the officials, but the system indeed creates more opportunities for uncertainty than necessary.
We’ve made a similar argument in our old coverage of Bangladesh's urban mobility regulation. Rules can become especially problematic when implementation is vague, and enforcement becomes discretionary.
The challenges of inspection can be fixed by a mix of establishing a simpler coordination mechanism and the right intention from the regulators. If several agencies need to inspect the same business, they should coordinate the inspection.
One visit where possible, a digital inspection record, one list of findings, one deadline for correction, and one approval so that the business can go into operation without unnecessary delays.
The current deregulation effort already talks about coordinated processes and service-level agreements. The task force should push this idea into everyday business regulation. The government should coordinate with itself. The entrepreneur has a business to run.
This is a common problem across citizen services in Bangladesh. Everywhere you go, you are asked for the same set of documents again and again. A company submits its incorporation information. Then another agency asks for it. A founder submits a tax document. Another office asks for a copy. A business provides its address. Another portal asks the business to enter the same address again. It is like that hedgehog day movie where you keep living the same day again and again.
The government has already started solving this problem in a way. BanglaBiz is built around integration between investment-related government services. The platform's business identifier is designed to help agencies work with a common reference for businesses. The NBR's own medium- and long-term strategy also recognizes the value of digital data sharing and reducing repeated requests for taxpayer information. The next step can be a government-wide rule where, if the government already has the information, the government should retrieve it.
Of course, there will be exceptions. Some information may require fresh certification. Some documents may need updating. Some agencies may lack legal authority to access particular data. Those exceptions can be defined. The default should still be information sharing inside the government. This would also reduce the importance of physical copies, notarization, and repeated certification where digital verification is possible.
We routinely have conversations with small businesses and startups who complain about the pain of routine license renewals and update submission requirements where an alternative approach might have served both the government and the entrepreneurs better.
Consider the ordinary trade licence. For a business that operates in the same place, under the same ownership, doing the same thing, the entrepreneur still has to deal with renewal every year.
This is not a challenge for large businesses that have administration departments to manage these procedures. But for a large majority of small and medium enterprises and startups, this is a costly challenge.
There could be straightforward solutions. For low-risk businesses, a licence should remain valid as long as the business remains compliant. The business pays the required annual fee. The licence stays active. If the business changes ownership, address or activity, it updates the information. If it stops operating, it closes the licence. Call it an evergreen licence. Or call it continuous licensing. The name does not matter.
The principle is do not require the entrepreneur to repeatedly renew permission to exist.
We understand there are complexities and categories of business where yearly renewal is necessary. In those cases, this should apply only to appropriate categories. A small software business does not create the same public risk as a chemical manufacturer. A design studio does not need the same licensing logic as a factory handling hazardous materials. The whole thing can be risk-based or based on some other criteria.
The government's current deregulation agenda already recognizes the need for proportionate, time-bound regulation. Evergreen licensing would turn that principle into something an entrepreneur can actually feel.
A regulation can be sensible in principle, but its requirements can still be poorly calibrated.
Our analysis of Bangladesh's Travel Agency Ordinance illustrated this problem. The ordinance responds to genuine concerns around consumer protection, ownership transparency and financial risk. But some requirements appear calibrated toward sophisticated, well-capitalized operators in a market where thousands of smaller businesses operate on thin margins. And some other provisions simply had no useful function there. It is good that the government later scrapped the ordinance.
But this pattern of policy thinking can appear elsewhere. The government writes a rule for the market it wants to have. Businesses have to operate under the rule in the market that actually exists. It can lead to higher barriers to entry, challenges for smaller companies, and make entering the market difficult for new businesses. Consequently, larger companies gain an advantage, competition is stifled, and eventually, it doesn’t serve the consumers well. Those are not the outcomes policymakers always intend. But certain policy thinking can lead to those outcomes.
And the solution is proportional regulation. A three-person software company should not face the same reporting burden as a bank. A small travel agency should not automatically face the same financial requirements as a large operator handling enormous volumes of customer money. A low-risk service business should not carry the same compliance burden as a hazardous industrial operation.
Regulatory obligations should match actual risk and business scale. This does not mean giving small businesses a free pass. It means asking a better question. What is the minimum regulatory requirement needed to manage the actual risk? It is a different question from asking what would be the most comprehensive set of requirements the government could impose. It also encourages better regulation. When a regulator knows that its requirements will have to be justified by risk, it has an incentive to distinguish between important and routine matters.
A large share of Bangladesh's regulatory code still carries the reflex of an earlier era: criminal liability, sometimes imprisonment, for offenses that are really paperwork failures, not fraud. The cost of the reflex is often more expensive. An entrepreneur who fears a magistrate over a missed filing date has a strong incentive to stay small, stay informal, and stay off the books entirely, which is exactly the opposite of what a deregulation task force is trying to encourage.
India has spent three years methodically working through this exact problem under its Jan Vishwas Act. The 2023 version converted 183 minor provisions across 42 laws from criminal offenses into civil penalties. A 2026 expansion pushed that to more than 700 provisions across 79 laws.
The logic is that a first-time, minor, technical lapse gets a warning or a fine, and prison stays reserved for fraud and genuinely harmful conduct.
One analysis of the reform put its purpose plainly: ending the compliance anxiety that keeps small businesses from formalizing in the first place.
Bangladesh's taskforce could run the same audit: go through the statutes that touch ordinary business filings, and convert the ones written for fraud into penalties fit for a missed deadline.
Bangladesh's regulatory history has made it hard for global companies, and by extension globally connected local startups, to move money in and out with any predictability. We've written about the downstream cost of this: a country that multinationals treat at arm's length also trains fewer local engineers, produces fewer managers with global operating experience, and builds fewer of the professional networks that early-stage founders eventually draw on. We laid this out at length after watching how thin Bangladesh's talent pipeline stays without deeper multinational roots here.
The FY27 package already has a good example in it: valuation reports are no longer required for unlisted-company transactions up to Tk 1 crore, and capital repatriation deals up to Tk 100 crore no longer need prior Bangladesh Bank approval. The low-hanging fix is to stop treating it as a carve-out. Raise the thresholds on a schedule, publish the schedule, and let founders and investors plan around it years in advance instead of hearing about the current ceiling only when they hit it.
BanglaBiz already proves the taskforce's own toolkit works. Five approvals that used to mean five separate offices, name clearance, a temporary bank account, incorporation, e-TIN, and trade license, now clear in three working days through one application. That's a genuine result, not a pilot promise. The gap is what happens next: the post-incorporation licenses that follow incorporation, sector clearances, environmental sign-off, and fire safety still run through their old, separate, undated processes. We noted this plainly in an earlier piece: starting a company here remains expensive and slow, largely because of what happens after the certificate, not the certificate itself.
The fix doesn’t need new infrastructure. It's an extension. Take the same starter-package model, service-level agreements plus deemed approval on missed deadlines, and apply it to the next tier of licenses a new business actually needs before it can legally operate: the sector-specific ones. Publish the completion times monthly, by agency, so the three-day standard is something every founder can expect, not something one platform happens to deliver.
Bangladesh has nearly 30 SEC-licensed Alternative Investment Fund Managers. Most haven't raised a meaningful fund. We've argued that governments here talk about startups while doing little to build the regulatory and tax scaffolding that angel investing and domestic VC formation actually need to exist at scale, and the AIFM numbers back that up. Bangladesh Bank's July 2025 revamp of its startup financing circular is proof that targeted deregulation on a narrow rule can move fast: higher loan ceilings, a 4 percent interest rate, an exemption from internal credit-risk rating rules through 2030. That happened inside a single central bank circular.
Private, early-stage risk capital needs the same kind of narrow, specific attention on the securities side: a clear fund vehicle for angel pools, defined tax treatment for early-stage losses and gains, and a workable path for corporate venture arms to invest without triggering the compliance overhead of a full financial institution. This is the deregulation idea with the longest payoff on this list, because it doesn't just cut a cost. It builds a layer of the ecosystem that barely exists yet.
Ask any SME owner in Bangladesh what actually keeps them up at night, and it's rarely the big regulatory questions. It's the thirty or forty trade payables sitting with retailers and buyers who pay whenever they feel like it, while wages and supplier bills don't wait. We described this pattern closely in a recent letter: an SME owner managing a permanent working capital gap not by borrowing, but by juggling who gets paid this week. The IFC puts Bangladesh's SME financing gap at $2.8 billion, and a meaningful share of that gap is really a late-payment problem wearing a financing costume.
India built a fix that doesn't touch the banking system at all. Under the MSMED Act, a buyer must pay a registered micro or small enterprise within 45 days, or 15 days if there's no written agreement. Since 2024, Section 43B(h) of India's Income Tax Act backs that timeline with a real cost: if the buyer pays late, it loses the tax deduction on that expense for the year, and can only claim it once payment actually happens. Suppliers can also file a complaint directly on the government's Samadhaan portal, which routes disputes to a facilitation council instead of a civil court. Bangladesh doesn't need a new lending program to close its SME financing gap. It needs large buyers to feel a cost, not just an inconvenience, when they sit on a small supplier's money.
The government has put together a serious institutional mechanism for deregulation. The political will is also apparent. Now comes the harder part: execution. We don’t think the success will be measured by the number of rules the taskforce reviews or portals it launches. The success will be measured by the experience of the entrepreneurs and businesses.
Can a founder register and operate a company without learning the internal geography of government? Can several government agencies inspect a company without making the entrepreneur manage their coordination? Can a business submit information once and have the government use it across agencies? Can a small company comply with the law without needing a full-time compliance professional? These may appear small questions, but collectively, they add up to the ultimate question: How much easier is it to build a business in Bangladesh?
That should be the measure. Deregulation does not have to mean deregulating everything. It can mean removing repetition, duplication, and unnecessary discretion.
As we see in this list, the best deregulation reforms may also be the least exciting ones. Many of them don’t require much to change. But they could save thousands of businesses thousands of hours. We believe that is a good place to start.
The government should make it easier for people to comply with the law than to navigate the bureaucracy around the law. If Bangladesh can achieve that, deregulation will have moved beyond a policy announcement and become part of how the economy works. Then the government can build on top of it.
