Restoring trust in official numbers: Why the BBS needs urgent reform

In recent months, the Bangladesh Bureau of Statistics (BBS) — the country's premier data-generating agency — has come under increasing scrutiny. 

Accusations of data manipulation, methodological flaws, and delayed publication have shaken public trust. A widely circulated investigative report by The Business Standard, titled "The agency that helped AL weave its development fable", has reignited the debate over the objectivity and transparency of economic data in Bangladesh.

Critics argue that official statistics, particularly those related to GDP, inflation, and poverty, no longer reflect reality and are often used to serve political narratives rather than to inform policy.

At a time when evidence-based policymaking is essential for navigating post-pandemic recovery, climate challenges, and demographic shifts, data integrity is not a luxury — it is a necessity. The crisis surrounding BBS should prompt urgent reform. Here are three critical areas that require attention if we are to restore confidence in our national data systems.

Institutional independence and capacity: Let BBS stand on its own

No statistical agency can maintain objectivity if it operates under the shadow of political influence. 

Currently, BBS functions as a department under the Ministry of Planning, with a bureaucrat often placed at its helm. This structural arrangement inherently creates a conflict of interest. A government that is both the subject and the manager of its own statistical reporting risks undermining the credibility of its data.

Global best practices suggest a different approach. 

For example, Statistics Canada operates independently under the direction of the Chief Statistician, who reports to Parliament, not a political ministry. Similarly, the UK's Office for National Statistics (ONS) is an executive agency of the UK Statistics Authority — an autonomous body that ensures impartiality and transparency in official data.

Bangladesh must consider a similar model. An autonomous BBS with its own governing board — comprising statisticians, economists, and civil society representatives — could ensure that data collection and dissemination are conducted professionally, without political interference. 

Furthermore, BBS must be adequately funded. Without sufficient human and financial resources, even the best institutional framework will fail. Regular investments in IT infrastructure, field operations, and statistical modelling tools are essential to meet global standards.

Data quality, regularity and relevance: Get the basics right

One of the most persistent criticisms of BBS is its failure to deliver high-quality data consistently. Key national surveys such as the Household Income and Expenditure Survey (HIES) or the Labour Force Survey are often delayed or conducted at irregular intervals. This undermines the ability of policymakers to track poverty, inequality and employment trends in real time.

Equally concerning is the quality of data. Surveys frequently contain missing values, measurement errors, and outdated sampling frames. These issues not only reduce the reliability of headline statistics but also affect downstream analysis by researchers and development partners.

Moreover, several data domains critical for inclusive and employment-centric policymaking are completely ignored. For example, BBS does not currently maintain any comprehensive database on small and medium enterprises (SMEs), startup activity, or the informal sector, even though these represent the largest sources of job creation in Bangladesh. Without such data, planning for youth employment, skills development, or social protection becomes speculative at best.

To address these shortcomings, BBS should adopt modern survey methodologies and statistical software that allow for real-time monitoring and automated validation. Collaboration with academic institutions and think tanks can enhance methodological rigour, while capacity-building programmes — such as those supported by UNDP and the World Bank — should be institutionalised rather than ad hoc.

Additionally, the creation of a central coordination unit that brings together BBS and other data-generating bodies, such as Bangladesh Bank, the Export Promotion Bureau, the Ministry of Food, and the Department of Agricultural Marketing, would help avoid duplication, ensure consistency, and streamline data flows.

Transparency and accessibility: Free the data

Even when BBS produces useful data, access to it remains a major barrier. Datasets are often released months — sometimes years — after surveys are completed. For instance, HIES 2022 was conducted in that year, but the microdata was released to researchers only well into 2023. In many cases, researchers must navigate a labyrinth of bureaucratic approvals just to gain access to anonymised datasets.

This secrecy runs counter to the very ethos of public data. Internationally, agencies like Statistics South Africa and the US Census Bureau provide open access to anonymised microdata, along with detailed methodology notes and codebooks. The OECD and IMF also advocate for open data portals as part of fiscal transparency initiatives.

Furthermore, BBS does not release the statistical codes used to generate published summary tables. As a result, researchers are often unable to replicate findings or reconcile discrepancies. This lack of reproducibility undermines both academic credibility and public trust.

To reverse this trend, BBS should adopt an open data policy. All major datasets should be published in machine-readable formats with accompanying metadata and R/Stata codes for replication. A public release calendar, updated annually, would allow users to plan their research and analysis more effectively.

The way forward

BBS is not merely a technical institution — it is a public good. Its ability to collect, analyse, and disseminate reliable data is foundational to good governance, effective service delivery, and democratic accountability. 

The current controversy must be seen as an opportunity to reform, not retreat. By ensuring institutional independence, investing in quality and coverage, and embracing transparency, we can build a statistical system that earns public trust and serves the national interest.

The government, development partners, and civil society must come together to ensure that BBS is not just an agency that counts the numbers — but one that makes the numbers count.

Md Deen Islam is an Associate Professor of Economics at the University of Dhaka and Research Director at RAPID.

Disclaimer: The views and opinions expressed in this article are those of the author and do not necessarily reflect the opinions and views of The Business Standard.