Governance

Bangladesh Governance: Institutional Reform, Bureaucratic Capacity, and Anti-Corruption

Executive Summary

Bottom line: governance is the binding constraint on Bangladesh's next development decade. All six World Bank Governance Indicators (WGI) place Bangladesh in the bottom quartile, with a composite percentile of 22.5/100. The weakest dimension is Political Stability at 15.0. The Transparency International CPI score of 28.0/100 (rank 151/180) marks corruption as deeply entrenched. The CPI score has improved by +2.0 points over the past year. This institutional deficit persists alongside 4.2% GDP growth, demonstrating that growth and governance can decouple for a time. They cannot decouple indefinitely: LDC graduation on 24 November 2026 removes trade preferences that have subsidized the current growth model, and the FDI needed to offset that gap requires rule of law (27.0 percentile) and regulatory quality (22.0 percentile) that Bangladesh does not yet deliver. The July 2024 political transition to an interim government is a one-time structural window that will not recur under normal partisan rule.

Base case: Without structural reform in this window, WGI composite remains anchored below 25, FDI stays below 0.7% of GDP, and post-LDC export competitiveness erodes.

Reform case: A sequenced civil service, judiciary, and fiscal devolution program, if launched in 2025-2026, can push the WGI composite toward the South Asian median (Vietnam: 42, Sri Lanka: 37) within one decade. The returns are non-linear: each percentile point improvement in Rule of Law is associated with measurable FDI and contract enforcement gains across comparable economies.

Civil Service: Patronage Beats Merit at Every Stage

The verdict: A civil service of 1,400,000 cannot modernize Bangladesh if postings, promotions, and pay are determined by political alignment. Government effectiveness at the 26.0 percentile is the direct output of that system.

• Entry is competitive; advancement is not. The Bangladesh Civil Service (BCS) examination is rigorous, but merit stops mattering after placement. Officers perceived as opposition-sympathetic receive punitive transfers; loyalists receive high-value postings irrespective of performance. This inverts the incentive structure at every level.
• No lateral entry mechanism exists. India has piloted specialist lateral entry at joint-secretary level, acknowledging that generalist cadre officers cannot lead digital governance (UN e-gov rank 100), climate finance, or post-LDC trade negotiation. Bangladesh has no equivalent pipeline, so specialist knowledge remains chronically scarce precisely where it is most needed.
• Regulatory fragmentation compounds the gap. Bangladesh operates 28 major regulatory bodies with overlapping mandates, no sunset review, and limited technical capacity. Regulatory quality at the 22.0 percentile is the investor's daily experience of this system.

So what: Every percentage point drop in regulatory quality is associated with higher compliance costs, slower business startup (now 19.5 days; regional comparators average under 10), and a Doing Business rank of 168 that reflects structural friction rather than a recent shock.

Anti-Corruption: Accountability Without Independence Fails

The verdict: At 850 cases filed annually, the Anti-Corruption Commission (ACC) is active on paper but compromised in practice. CPI 28.0/100 and WGI corruption percentile 17.0 are consistent with an institution that prosecutes by political affiliation, not by severity.

• Procurement is the highest-risk vector. Electronic Government Procurement (e-GP) covers 65% of central government procurement, a genuine improvement. But local government and autonomous-body procurement, where price inflation on infrastructure projects runs 20 to 30 percentage points above regional benchmarks, remains outside the e-GP perimeter.
• Parliamentary oversight is nominal. 50 standing committees exist but lack investigative independence, public hearing authority, or follow-up enforcement. Committee chairs are allocated by party loyalty, not subject expertise. The contrast with South Korean or Filipino legislative committees, which have compelled executive document disclosure, is instructive.
• RTI is structurally under-utilized. The RTI Act 2009 generates 12,000 requests annually. India's equivalent law processes over 6 million. The gap reflects low public awareness and institutional non-responsiveness, not a lack of citizen demand for transparency.

So what: Corruption at this scale does not only raise costs. It selects against productive investment and selects for rent-seeking capital allocation. Press freedom at rank 165/180 (RSF 2024) means the investigative journalism that historically exposed ACC capture in peer countries is absent.

Local Government: Administrative Deconcentration Is Not Devolution

The verdict: 64 districts, 495 upazilas, and 4,571 union parishads constitute a three-tier structure that delivers central government programs, not local governance. Fiscal architecture explains why.

• Union parishads collect less than 5% of the revenue they spend. Formula-based block grants from Dhaka determine what local chairs can and cannot do, making accountability run upward to the administration that controls funding rather than downward to voters. Bangladesh allocates under 10% of the national budget to local government; decentralized comparators (Philippines: 25 to 30%) allocate two to three times as much.
• Land administration is the citizen-facing manifestation. An estimated 60 to 70% of civil litigation traces to land disputes. Digital survey modernization is underway but proceeding below the pace needed to absorb demand. Until land titles are digitally verifiable and mutation processes are automated, the 28.0 voice and accountability percentile reflects an institutional reality: citizens have a vote but not a lever.

So what: Genuine devolution, with own-source revenue capacity and community-level accountability, is the most direct route to improving citizens' daily experience of the state. Administrative deconcentration without fiscal autonomy produces neither efficiency nor accountability.

Judiciary: 4.7 Million Cases, No Fast Exit

The verdict: The judiciary backlog of 4,742,000 pending cases is not a processing lag; it is a structural denial of justice. At current disposal rates, clearing the queue would take decades. Rule of law at the 27.0 percentile is the investor-facing consequence.

• Commercial case delay is the investment deterrent. Contract enforcement, a standard indicator in Doing Business assessments, is defined by average time to resolve a commercial dispute. Long delays transform contracts into aspirations and raise the risk premium on private capital deployment.
• Judicial independence is formal, not behavioral. The separation of judiciary from the executive (2007) was a constitutional milestone. Appointment processes for higher court positions remain opaque and subject to executive influence, as evidenced by post-transition judicial conduct inquiries.
• Alternative dispute resolution (ADR) is the fastest lever. Village courts and mediation panels exist but are underutilized. Diverting minor civil and family disputes through mandatory ADR before formal filing could reduce new case inflows by an order of magnitude, allowing judges to process the existing backlog rather than absorb new additions.

So what: Without a binding backlog-reduction target, case numbers compound. The leverage point is supply-side (court automation, ADR, specialized commercial tribunals) not simply judicial appointments.

Digital Governance: Infrastructure Ahead of Process

The verdict: UN E-Government rank 100 and e-GP coverage 65% represent real digital progress. The gap is that front-end digitization has outpaced back-end process reform; citizens can apply online for services that are still approved through paper-based bureaucratic chains.

• e-GP extension to local government is the next-highest-return investment. Central government procurement at 65% coverage represents the low-corruption tier. The high-corruption tier, local government and state-owned enterprise procurement, remains opaque.
• Political stability at 15.0 percentile is the reform continuity risk. Digital governance programs require multi-year institutional commitment. Bangladesh's political instability track record means program continuity across government transitions is not guaranteed without legislative anchoring.

Prioritized Recommendations

Three interventions, ranked by near-term tractability and impact:

1. Depoliticize civil service postings and introduce lateral entry (owner: Ministry of Public Administration; timeline: 2025-2026)

Establish an independent Civil Service Commission with statutory authority over postings and transfers, removing the ministry from individual placement decisions. Simultaneously, create a limited-scale lateral entry pathway for 50 to 100 specialist positions annually in digital governance, trade, climate finance, and economic regulation. India's lateral entry outcomes and Vietnam's technocratic track record both demonstrate that this change is implementable within a 12-month legislative cycle under an interim government with reform mandate.

2. Set a binding 5-year judiciary backlog target with ADR expansion as the primary instrument (owner: Supreme Court + Ministry of Law; timeline: 2026-2030)

Legislate a target to reduce the 4,742,000-case backlog by 50% within 5 years, using three instruments: mandatory pre-filing ADR for civil disputes below a value threshold, specialized commercial courts in Dhaka and Chittagong with dedicated case tracks, and full e-filing with automated scheduling. Philippine experience with commercial courts cut average contract enforcement time by 40%. Judicial appointment transparency should be anchored in this same legislative package.

3. Restructure local government fiscal architecture toward own-source revenue (owner: Local Government Division + Finance Ministry; timeline: 2026-2028)

Increase the local government share of national budget from the current sub-10% toward 15% within three years, financed by formula-based transfers tied to service delivery indicators rather than political patronage. Pair this with property tax administration reform at the union parishad level to build own-source revenue capacity. The ACC should extend e-GP coverage to all union parishad and upazila procurement above a minimum contract value, closing the transparency gap where corruption risk is highest.

Data sources: World Bank Worldwide Governance Indicators 2022, Transparency International CPI 2023, UN E-Government Survey 2024, RSF Press Freedom Index 2024, Bangladesh ACC Annual Report, LGRD Division, Supreme Court case-disposal statement (31 Dec 2025), CPTU e-GP data, WB Doing Business 2020.