The answer to the title’s question is no. Let’s keep in mind that Bangladesh is the eighth most populous country in the world. It is also among the most densely populated, with 170 million people. It is unlikely that it can continue its high-income status development journey under a highly centralised government. Without decentralisation, it will not be possible to provide quality public services, to plan urban centres and boost local economies. Both economic theory and international experience suggest that decentralising governments provide better services, urban development, economic dynamism and accountability.

This writeup explores some radical reforms on Bangladesh’s decentralisation agenda. Bangladesh needs to make its 64 districts the main tier of subnational government. It must also transfer most governance and public service delivery functions to its elected subnational governments. The country’s districts can be declared provinces after consideration by a Decentralisation Commission. The structure would be like Vietnam’s, whose 58 provinces provide most public services, infrastructure and private investment support to the country’s 99 million people.

This is not a new idea for Bangladesh. Indeed, it is evocative of the vision of Sheikh Mujibur Rahman, the country’s founding father. Mujib, in 1975, appointed governors of 61 districts who would take responsibility for their areas and report to the prime minister. In effect, he set up 61 provinces.

Remarkable progress but just a start

Defying all pessimists and naysayers at the time of independence in 1971, Bangladesh has made remarkable progress. Its per capita income and food production have increased fourfold since that time. Poverty reduction and human development have been equally impressive. Poverty rates have declined by two-thirds, from more than 50% in 2000 to 19% in 2022. The country has essentially achieved universal enrolment and gender parity at primary school level. Most telling, life expectancy now stands at 72 years, higher than in neighbouring India and Pakistan.

This progress has been noticed. Reclassified by the World Bank in 2016 as a lower-middle-income country, Bangladesh has also started the process of graduation out of the least developed country group under the United Nations classification.

However, this progress needs to be contextualised. The truth is, Bangladesh is at an early stage of its development, with a per capita income one-eighth less than that of high-income countries measured in international purchasing power parity (PPP) dollars (and one-sixteenth in US dollars). While extreme poverty has declined sharply, about half of the population lives under the internationally recognised lower-middle-income poverty line of USD 3.20 PPP. Further, the poor are geographically concentrated: more than half live in one-quarter of the country’s districts in the north, the north-west and the eastern Chittagong Hill Tracts.

While extreme poverty has declined sharply in Bangladesh, about half of the population lives under the internationally recognised lower-middle-income poverty line…

Given Bangladesh’s population density and the vulnerability of its southern coastal areas, out-migration to address poverty in poor districts is not a practical solution. Instead, a spatially balanced local area development strategy could lead to the creation of good jobs in poorer districts. Local governments with more robust capacity can better counter the challenges.


Growth needs strong institutions

For Bangladesh, sustainable and quality growth will require more robust institutions and better governance. For instance, policy-makers will have to think beyond building schools. Children need also to attend school regularly, obtain a good education and graduate to the next level. Medical clinics have reached rural villages but more needs to be done to ensure health professionals and medical supplies are adequate. Building roads and infrastructure won’t be enough; good maintenance is also vital. Lastly, quality service provision will not suffice: local governments will also have to be competitive in attracting private investment. These are all areas where Bangladesh is lagging.

At present, Bangladesh’s excessive centralisation is a key obstacle to good service delivery. The country is one of the most centralised in the world. The central government spends more than 90% of all public expenditure. That means the 700 plus other city, municipal and subdistrict (known as upazila) governments spend less than 10% of public money. Their budgets are so small that employee salaries take up most of the allocation. In other words, local governments play virtually no role in development.

Decentralisation: economics and experience

Let’s delve into the economic case for decentralising governments. This is best illustrated with an example from south-western Bangladesh. It is the people of Khulna city, and the broader Khulna district, who have the greatest interest in improving the services they get. These may be public services, education, health, water, sanitation, drainage, roads, electricity, employment or business support. They are also better informed than outsiders about their own problems and needs.

Khulna’s elected local governments (like its city government) will have the incentives to provide these services effectively or else they will lose the next election. Compared with them, Bangladesh’s 39 ministries and the five key utilities, centralised in Dhaka, do not have the knowledge or incentives to effectively provide any public services for Khulna’s people. This is the fundamental rationale for decentralising governance and public services to the people and their government in Khulna.

Across the world, as countries develop, they decentralise, and the share of local governments’ spending increases. Local government expenditures are 30% of all public spending for high-income countries. For the upper-middle-income group the share is 25%. Meanwhile, it is 20% for the lower-middle-income group, to which Bangladesh belongs. However, Bangladesh is an outlier, because its local governments’ share in public spending is a miniscule 7–8%.

Bangladesh can look east for lessons on decentralisation. Neighbouring East Asia’s experience with decentralisation is noteworthy. As it happens, these dynamic, fast-growing countries that created the ‘East Asian Miracle’ are highly decentralised. Their experience illuminates that decentralisation was pivotal to their development. They also provide lessons on the design for decentralisation.

Decentralisation to Bangladesh’s rescue?

There are five ‘decentralising’ channels that will stimulate Bangladesh’s development:

  • Decentralisation will energise When local people and governments have the incentives to develop the local economy, they will be energised. That will happen because people’s loved ones will clearly see the benefits from the economy; governments will be closer and more accountable to the people; the taxes the governments receive will depend on performance; and the future careers of their politicians will depend on the development record. All over the world, there are examples of effective local leaders who have become the next head of their country’s government. Key examples are mayors of cities (e.g. Joko Widodo of Indonesia, Lee Myung-bak of South Korea, Rodrigo Duterte of the Philippines) and governors of provinces (Xi Jinping of China, Narendra Modi of India).
  • Decentralisation will encourage ‘development competition’: If the tenures of different local governments depend on their performance, this will encourage healthy competition. This will boost the economy. In East Asia, decentralisation policy in China, Vietnam, Indonesia and the Philippines created an intense competition to develop among local governments. They competed to increase agricultural productivity, attract investment, build the manufacturing sector and create good jobs. Local governments created industrial parks, offered business licences, recruited workers and organised electricity. Tools such as the Provincial Competitiveness Index in Vietnam and the Local Economic Governance Survey in Indonesia regularly track the performance of provinces and cities to engender a spirit of competition. The same can happen in Bangladesh.
  • Decentralisation will facilitate reforms: Not widely known is that it was reforms by local governments that started the spectacular rise of China and Vietnam in the Famously, in an era of collective farming, local communist party officials of a village in the then-impoverished Anhui province of China decided to experiment by giving 18 farming houses the right to cultivate their own plots and sell their produce, after they had met their supply quotas. Production boomed. The news of this successful experiment spread. In 1979, Deng Xiaoping, China’s reforming leader, directed all provinces to emulate the system.

Local government leaders in Vietnam led reforms through ‘fence-breaking’ experiments, often without Hanoi’s approval. In the north, experiments were secretly carried out in Haiphong province when members of the Doan Xa community were given land on lease to cultivate on their own using a ‘sneak contract.’ Production increased by six times.

Later, the Ho Chi Minh City government became the ‘reform leader.’ It carried out bold reform experiments and drove growth. In Indonesia, too, policy reform experiments such as e-procurement by the city of Surabaya were first carried out by district governments before being adopted nationally. This made regulations more transparent, and introduced one-stop centres for investors. The same can be put in place in Bangladesh.

  • Decentralisation will find leaders: Finding and training able national leaders is a crucial challenge for all countries, including Bangladesh. Decentralisation helps meet this challenge. Decentralised governments give their leaders the space to prove their ability to the people. Conversely, the lack of options to identify competent leaders curtails the benefits of democracies. As Nobel prize-winning economist Roger Myerson has pointed out, if there are no means to recognise emerging leaders, people may perceive that there are no alternatives – even when the existing leaders perform below expectations.
  • Decentralisation promotes high-quality urban development: Development and growth typically start with a solid boost to agriculture. But soon cities become the engines of growth. The world’s 3% of urban land produces 80% of its gross domestic product (GDP), while Bangladesh’s 7% of urban land produces 66% of its GDP. Cities provide essential services – such as marketplaces, finance access, education, job facilities, health care, cultural endeavours, sports and recreation – that make city life attractive. These cities will need dense infrastructure such as internal transport, power, telecommunications, a water supply, sanitation services, housing, land zoning and open spaces to better people’s lives.

A well-run city requires dynamic planning and, importantly, effective governance to provide these essential services. For this, city governments will need resources and authority. It is readily apparent that centralised governments cannot perform this role from a distance. Central governments will lack the last-mile knowledge to adequately respond. Strong local governments thus become essential for the proper development of cities.

Nobel Laureate economist Roger Myerson, who is an expert in decentralisation, gives a lecture at the University of Chicago, Illinois, United States, 16 January 2014 | Photo by Becker Friedman Institute.

Thoughts on decentralising Bangladesh

The local governance agenda has a head-start in Bangladesh’s unitary government system. Local governments have a firm constitutional basis in the country. Bangladesh’s constitution states, ‘Local government in every administrative unit of the republic shall be entrusted to bodies, composed of persons elected in accordance with law’ (in Article 59). It goes on to say that, ‘Parliament shall, by law, confer powers on the local government bodies to impose taxes for local purposes, to prepare their budgets and to maintain funds’ (in Article 60). There are also different laws upholding the constitutional articles.[1]

The local governance agenda has a head-start in Bangladesh’s unitary government system.

Based on these laws, a local government structure of 64 district councils, 11 city corporations, 329 municipalities, 495 subdistrict (locally known as upazila) councils and 4,573 union councils (the grassroot local government bodies) operate in Bangladesh. De jure, these governments have some power to levy taxes and service charges. They also oversee some public health services, education and social/development undertakings (like public libraries and local roads).

However, de facto the picture is vastly different. Bangladesh is one of the least decentralised countries in the world, especially given its size. Implementation of laws lags well behind. Further, according to the laws, Bangladesh’s decentralisation focuses chiefly on subdistrict (locally known as upazila) governments overlying a network of union council governments. Nominally, district councils are the upper-tier government for the subdistricts. In reality, the districts are the weakest part of the local government structure.

Implementing decentralisation needs detailed rules, by identifying key agencies, guidelines, enforcement mechanisms and scope for public input. This includes: 1) defining different tiers of local government including the leading tier; 2) realising fiscal decentralisation by specifying revenue raising, revenue sharing, and expenditure authority; 3) strengthening civil service capacity for local government; and 4) ensuring accountability through monitoring and evaluation by audit bodies, civil society groups and central government.

For Bangladesh, a key weakness is that the functions of the district councils are unclear and residual. These are often those that are not assigned to the subdistrict or the central government. Consequently, resource flows are most limited in this area. District governments are weak because district councils are not directly elected by the people. Instead, they are elected indirectly by other local government members.

Getting the main local government tier correct

There are essentially three tiers of elected local government in Bangladesh: district – the top tier; sub-district (locally known as upazila) – the second tier; and union (the grassroot unit) – the bottom tier. Their elected bodies are their respective ‘councils.’ Although subdistricts are the principal local government hub, fiscal resources allocated to subdistricts and their sub-tier, union governments, are highly inadequate. They are spent almost entirely on staff salaries. Further, elected local government leaders have no authority over central government’s bureaucrats working as local government officials. These bureaucrats are accountable only to their parent ministries in the central government.

Bird’s-eye view of district 4, Ho Chi Minh City, Saigon, Vietnam, 15 July 2020 | Photo by S. Ken.

There are critical reasons for the ineffectiveness of subdistrict governments. The subdistrict government’s sphere of interest overlaps with the interests of the member of parliament (MP) of that area. MPs are far more influential than are the chairs of subdistrict governments. MPs effectively run the local government by colluding with the locally posted bureaucrat or the chair of the subdistrict council. Beyond politics, there are capacity constraints for subdistrict governments. They are too small in scale to provide an adequate tax base for a functional government. Bangladesh’s 495 subdistricts are simply too numerous to coordinate properly.

Bangladesh’s experience with subdistrict governments, and similar lessons from other countries, recommends against decentralising to overly small governments. Decentralising to a higher level, the district level in Bangladesh’s case, is a better choice. This larger scale would help address both the political and the capacity constraints. The district, which covers an area greater than an MP’s constituency, would exceed the authority of any single MP. And if the district is refashioned into a province, the prestige of the elected provincial chairperson would overshadow that of the MP.

A district government level would also have an optimal revenue base from which to derive revenues. A governance structure of 64 such district (or province) governments, with autonomous authority over the administration and the budget, would work like the provinces in Indonesia, the Philippines and Vietnam. Interestingly, for Bangladesh, these powerful district governments are what the country’s founder also envisioned.

Implementation steps

Bangladesh’s decentralisation agenda needs a roadmap. Decentralisation will be a gradual journey, one that needs preparation. However, the process can start quickly with what is termed ‘deconcentration’ – that is, relocating relevant government agencies outside Dhaka, the centralising hub. Take the case of the Bangladesh Shipping Corporation. Its headquarter is in the south-eastern port city of Chattogram. But due to limited decision-making powers out of Chattogram, the chairperson has to spend most of his/her time in Dhaka. Hence, Dhaka is the de facto headquarters. This reality needs to change.

In some cases, when the headquarters cannot be moved, most departments and personnel should be moved closer to the field. For example, most of the personnel of the shipping ministry and the Navy headquarters should be located in the coastal cities of Chattogram and Khulna. Policies could encourage some commercial banks to move their headquarters to Chattogram (or even to Khulna) because economic activity is growing rapidly there. Most of the staff of the education and health ministries can be relocated to regional (locally known as divisional) headquarters. Moving departments and staff means shifting resources, which are excessively concentrated in Dhaka.

Decentralisation Commission

Decentralisation, or the devolution of government, is a deep process needing careful preparation. For Bangladesh, first a Decentralisation Commission, consisting of decentralisation experts, civil servants, politicians and civil society members, should be formed.

For Bangladesh, first a Decentralisation Commission, consisting of decentralisation experts, civil servants, politicians and civil society members, should be formed.

The Decentralisation Commission would have the following tasks:

  • Determine the main hub of the local government tier.[2] For instance, if the Decentralisation Commission decides the district (which can be rebranded as the province) should be the central hub, then it should revise the functions between the central and other tiers.[3]
  • Determine the constituencies for the district council and arrange for direct voting by the electorate.
  • Ensure the government’s rules of business give coordination authority to district governments to provide public services.
  • Recommend integrating the amendments to the existing legal framework, and formulate rules to implement the new amendments.

Finance Commission

Second, a decentralisation process in Bangladesh will need to set up a Finance Commission. Its job will be to recommend the reallocation of fiscal resources. The resource allocations have to factor in the verticals (among the different tiers of government) and the horizontals (across districts and subdistricts).

A Finance Commission will devise a formula for allocating central tax revenues among the subnational governments. It should also distribute tax bases among the other tiers. A vital task here will be formulating policies to strengthen local government property and land tax bases. Here, there is a huge potential for revenues, which will help finance Bangladesh’s overall development.

Joint committee for local governance

The third step of the decentralisation process is the formulation of a joint committee with representatives from both the Decentralisation and the Finance Commissions. Its job will be to enhance the administrative capacity in local governments. A local government civil service, accountable to local governments, should be set up.

To incentivise them, the upper echelons of the local civil service should have a performance-based pipeline to the national civil service. Local governments will need to have authority (jointly with the central government) over the top-down national civil servants assigned to them. They should have a say in appointments and tenures.

Accountability and expertise

Finally, for effective decentralisation in Bangladesh, transparent arrangements for monitoring, accounting for and evaluating local government performance will have to be set up. The key is to strengthen financial management (accounts and auditing system). The arrangement should also identify performance indicators for public services and economic activity. The monitoring and evaluation of district governments and below should be undertaken by central government agencies (such as the Office of the Comptroller and Auditor General) and specialised civil society groups.

…for effective decentralisation in Bangladesh, transparent arrangements for monitoring, accounting for and evaluating local government performance will have to be set up.

For Bangladesh, the tasks for a decentralisation journey are complex. They are also likely to be contentious. But the country’s policy reform history and experience from overseas can act as guides. It will be essential that the mentioned institutional setup – with the Decentralisation Commission, the Finance Commission and the joint committee – draw expertise from across the government and society. The structure will also need to have strong technical support. The instruments should promote piloting and experimenting with a few decentralisation designs in different parts of the country.

Decentralisation and politics

In Bangladesh, the common objection to decentralisation is that it is politically unrealistic because members of parliament (MPs) will obstruct it. MPs will be unwilling to share power in their constituencies. Another objection is that local-level technical capacity is very limited. A third objection, coming from policy analysts, is that Bangladesh is a small, homogenous country and decentralisation is unnecessary.

Campaigners for an independent candidate organised a motorcycle rally during a local government election campaign in Sreenagar sub-district of Munshiganj, central Bangladesh, 5 November 2021 | Photo by Mahmud Hossain Opu.

This writeup addresses all these objections. Bangladesh can form 64 district governments (declared as provinces) in an effective subnational government tier. These districts are large enough to create political space for both MPs and local governments. Their optimal size can also address capacity constraints by drawing on much-needed resources (administrative or revenue-related).

Creating powerful district governments will also respond to the third objection. Bangladesh, with a population of 170 million people, is not a small country: it is larger by population than the Philippines, South Korea, Thailand and Vietnam. Even geographically much smaller countries such as Belgium and the Netherlands are highly decentralised.

An interesting point is that, de jure, local governments in Bangladesh already have a firm constitutional and legal basis. But de facto their implementation lags well behind. The reality is, Bangladesh just has to do it! As discussed, a well-designed decentralisation will address both political opposition and capacity constraints.

The final response to all the objections is that Bangladesh does not have a choice but to decentralise. The country’s future development will have to be more regionally focused, and accountable to the grassroots, if it is to meet its economic and climate-related challenges. Today, when an average Bangladeshi in towns and villages faces the commonly experienced problem of uncoordinated public works (like cut-up roads or encroachment into water bodies), he or she is clueless as to where to turn for help.

In Bangladesh, dynamic urban centres will be possible only if urban governments have the authority to directly provide public services. But regional, and especially urban, development needs the knowledge of ‘local needs and conditions.’

Providing this coordination and addressing the problems that plague local-level citizens will not be possible for 39 ministries headquartered in Dhaka, the capital city of Bangladesh. Empowering local governments will help in this vital coordination task. More importantly, it will energise the locals, both the government and the people, to take control of their own fate.


[1] The laws are the Hill District Local Government Parishad Act of 1989, the Zila Parishad Act of 2003 (for district councils), the Local Government (Municipality) Act of 2009, the Local Government (Union Parishad) Act of 2009 (for grassroots local government councils), the Local Government (Upazila Parishad) Act of 1998 amended in 2009 (for subdistrict councils) and the Local Government (City Corporation) Act of 2009.

[2] As of 2023, the main hubs are the divisional headquarters (there are eight divisions in Bangladesh, which are essentially a bureaucratic tier without any elected bodies to match them). The different local government hubs are divisional headquarters, districts and subdistricts (locally known as upazilas). Divisional headquarters play a minimal role and they are too large to provide effective governance for, say, school systems. The advantage with districts is that there are sufficient laws to support their consolidation.

[3] At present, district functions are mostly residual, with most powers assigned to the subdistrict (locally known as upazila) and the central government.


Photo © Mahmud Hossain Opu

Ahmad Ahsan is Director of the Policy Research Institute (PRI), Dhaka. He is an economist. He was a Lead Economist at the World Bank and a Professor at the University of Dhaka and Columbia University, New York. He specialises in policy dialogue, technical assistance operations, regional economic integration programmes, international migration, climate change and economic policy in Africa, East Asia and South Asia. He has worked with the International Monetary fund, the Asian Development Bank, the African Development Bank, Asia-Pacific Economic Cooperation, the Association of Southeast Asian Nations, the EU, the Government of Japan and the United States Government. He pursued his doctoral studies in economics at Columbia University.