Nadeem-ul-Haque
Every year, Transparency International releases its Corruption Perceptions Index, and the familiar ritual begins. Poor countries are named and shamed, rich countries are celebrated as clean, international agencies demand governance reforms, and governments such as Pakistan’s promise to do better. Newspapers reproduce the rankings as if they were scientific measurements of national integrity. Almost nobody asks who finances these indicators, who evaluates their accuracy, and why the institutions that judge developing countries are rarely subjected to equivalent public scrutiny.
The problem extends well beyond Transparency International. Over several decades, an elaborate international development industry has emerged that measures the governance, corruption, regulatory quality, institutional capacity and policy performance of poorer countries. There are indicators for almost everything. Meanwhile, developing-country governments must satisfy lending conditionality, undertake donor-prescribed reforms, establish monitoring units, hire consultants and produce an endless stream of reports. Yet the advisers/consultants are absolved of all blame as the indicators tend to put the blame on the morality of the people so that the advice given by consultants which probably creates the incentives to beat the system are not questioned.
But who measures the development industry itself?
Transparency International provides an instructive starting point. It was founded in Berlin in 1993 by Peter Eigen, a former senior World Bank official, and his associates. Initially, the organization challenged the reluctance of international institutions to confront corruption. At that time, some advanced countries even allowed foreign bribes to be treated as tax-deductible business expenses.
The organization’s first Corruption Perceptions Index appeared in 1995. The World Bank initially resisted joining the campaign. But under James Wolfensohn, who described corruption as a cancer in the mid-1990s, the Bank embraced anti-corruption policies. Transparency International and the Bank subsequently developed a close professional relationship. The Bank incorporated anti-corruption concerns into its operations and developed its own Worldwide Governance Indicators.
Who finances Transparency International? Its published accounts identify substantial support from Western governments and development agencies, including Germany, Britain, Canada, Denmark, Sweden, Australia and the United States, alongside the European Commission, foundations and some corporations. Its accounts are audited, and its governance arrangements are publicly documented. Funding does not by itself establish donor interference. Nevertheless, dependence on institutions embedded in the international development system makes independence an entirely legitimate subject for investigation.
The empirical foundations of Transparency International’s Corruption Perceptions Index (CPI) deserve much greater scrutiny. Our examination of 90 countries using 2024 data finds that income per capita alone explains approximately 67% of the variation in CPI scores. Rich countries consistently score better, but does this reflect actual differences in corruption, or do perceptions partly reflect prosperity, economic cycles, political stability and established country reputations?
There is no common international sampling frame. Even the European Commission’s 2017 methodological review primarily established statistical consistency, not that the index accurately measures actual corruption or that equivalent scores mean the same thing across countries. Agreement among experts does not establish accuracy, particularly when perceptions may be shaped by shared assumptions and information sources.
Equally troubling is the limited role of developing countries in the measurement process. International institutions and assessment organizations determine the indicators, select respondents, establish methodologies and produce the scores by which poorer countries are judged. Local experts and businesses sometimes participate, and Transparency International has national chapters, but independent universities and researchers in the countries being assessed have no guaranteed role in designing, validating or independently replicating the CPI.
This reflects a broader asymmetry in the development industry: donors finance an expanding universe of indicators judging recipient countries, while their own governance, consultancy networks, institutional inefficiencies and policy failures face far less comparable external scrutiny. Poor countries are continuously monitored, ranked and subjected to conditionality, often with considerable economic consequences, while those prescribing reforms bear much less direct responsibility when their advice fails. The essential question is not merely whether the CPI measures corruption correctly, but who has the authority to define, measure and judge governance—and why those exercising that authority are not subjected to equally rigorous independent evaluation.
The CPI measures perceptions of public-sector corruption, based on expert and business assessments. It does not measure all corrupt transactions or the full range of ways private wealth influences political decisions.
Research has long exposed this limitation. Donchev and Ujhelyi found that actual experiences of corruption are relatively weak predictors of corruption perceptions. National income and institutional characteristics also influence what experts believe about corruption.
Our examination of a 90-country sample using 2024 data confirms a striking association. Income per capita alone explains approximately two-thirds of the cross-country variation in CPI scores. Western countries have substantially higher average scores than Sub-Saharan African and South Asian countries
The more serious issue is conceptual: what is excluded from the definition of corruption? In poor countries, influence is often exercised through bribery, family connections, administrative discretion and political patronage. In wealthy countries, powerful interests can employ lobbyists, former regulators, political consultants, law firms, industry associations and campaign-financing networks to influence policy. Much of this is legal, and legitimate lobbying must be distinguished from improper influence. In both cases, wealth can purchase disproportionate political access without violating the law. Yet only in poor countries is this called corruption!
A businessman bribing an official to circumvent a regulation is readily identified as corrupt. A corporation employing professional intermediaries to shape the regulation in its favor may escape that description entirely. The first transaction violates the rules; the second can influence who writes the rules.
The deeper question concerns the enormous governance industry surrounding developing countries. For decades, the World Bank, IMF, bilateral donors and international foundations have financed studies documenting the institutional weaknesses, corruption and policy failures of poorer countries, accompanied by an expanding apparatus of consultants, rankings, reform prescriptions and lending conditionality. Yet the performance of these interventions has attracted serious criticism, including from the institutions’ own evaluators.
Where, then, is the comparable international index evaluating donor governance, consultancy costs, procurement networks, conflicts of interest, duplication, failed reforms and long-term development outcomes? Independent evaluation offices exist, but they do not constitute a system of accountability controlled by the countries bearing the consequences. Borrowing governments face conditionality, debt obligations and relentless monitoring, while consultants and international advisers rarely face comparable consequences when their recommendations fail. They have far less skin in the game. Why should developing countries be continuously rated for governance failures while the institutions prescribing their reforms escape equally systematic, externally directed scrutiny?
There is no comparably influential, sustained system through which independent universities and research institutions in developing countries evaluate and rank donor performance, with the same visibility and consequences that donor-backed indicators can have for recipient governments.
Indeed, the incentives run in the opposite direction. Research institutions in poorer countries frequently depend on international grants, contracts and consultancy opportunities. Studying corruption in their own governments attracts funding. Investigating the governance of donors, their costly consulting practices, their policy failures and their institutional incentives is a much more awkward proposition for organizations seeking continued donor financing.
That is an imbalance in the production of knowledge itself. The agencies that finance research help determine which questions are investigated. If the quality of developing-country governance is continuously examined while the quality of donor governance receives less independent, locally directed scrutiny, the research agenda becomes institutionally asymmetric.
The World Bank’s own evaluation findings illustrate why the problem matters. Its Independent Evaluation Group reported in 2025 that about 40% of completed advisory assignments lacked evidence of results in their intended outcome areas. Yet 97% of these assignments had been self-rated moderately satisfactory or better during 2017–24.
Imagine a developing-country ministry reporting that it could not demonstrate results for a substantial share of its activities while declaring almost all of them satisfactory. The governance industry would rightly demand accountability. Should the same standard not apply to the institutions providing advice?
Development agencies also contribute to the institutional proliferation they subsequently criticize. Rather than strengthening existing government systems, projects can establish separate implementation units, special-purpose arrangements, parallel monitoring mechanisms and externally managed consulting structures. These arrangements may deliver some immediate results while weakening incentives to build sustainable domestic capacity.
Who evaluates the resulting duplication? What do these institutional arrangements cost? How much expenditure reaches actual services rather than administrative structures and consultants? What happens to the organizations once external financing ends?
And what about conditionality? Pakistan and other frequent borrowers must repeatedly adjust policies to satisfy international institutions. Governments negotiate targets, draft legislation, restructure agencies, increase taxes and revise public expenditures, often under considerable financing pressure. The development profession then judges their compliance and subsequent performance.
But when a donor-supported reform fails, accountability becomes diffuse. Officials and citizens in the borrowing country live with the consequences. Public debt remains a national obligation, while the international experts who recommended the policy generally do not bear a comparable personal cost.
This is the problem of having insufficient skin in the game. Development agencies do face financial, reputational and institutional risks, and some undertake serious evaluations. But the costs of misguided reforms are not distributed symmetrically. Borrowing countries remain responsible for debt and implementation, while international advisers can move on to the next program, country or assignment.
The issue is not that development institutions never contribute useful knowledge or that domestic corruption is imaginary. Pakistan’s system of privilege, discretionary regulation, opaque contracting, weak accountability and restricted competition demands serious reform. Foreign assistance has also produced worthwhile results.
But acknowledging these realities cannot justify a system in which poorer countries are continuously examined while the international institutions directing substantial portions of their policies enjoy considerably greater freedom from comparable scrutiny.
We need to reverse the direction of some of this research. Developing-country universities and independent research institutions should be financed through arrangements that give donors no editorial or methodological control to examine international agencies, assess consultant contracting, measure reform outcomes, estimate administrative costs and evaluate the economic consequences of conditionality. Their findings should be public, independently reviewed and debated in recipient-country parliaments.
Equally, corruption indices should be complemented by measures that capture lobbying transparency, political finance, regulatory capture, financial secrecy, procurement networks and professional enablers across rich and poor countries.
Transparency International made an important contribution by bringing corruption into international debate. But the development industry must now confront the logical implication of its own message: accountability cannot be reserved for the recipients of advice and assistance. It must extend to those who finance, design and prescribe the reforms.
For too long, poor countries have been measured, monitored, rated and instructed. It is time to measure those who do the measuring.
The ultimate test of development institutions is not how many reforms they prescribe or how many indicators they produce, but whether the countries that follow their advice achieve sustained economic improvement. On that test, advisers and borrowers alike must face scrutiny.
Get our Books: 0341-4222501
