Pakistan’s score on Transparency International’s Corruption Perceptions Index (CPI) has improved by one point, but a closer examination of the data and methodology raises questions about how the score is calculated, how much of the underlying evidence can be independently verified and how closely international perceptions reflect the everyday experience of ordinary Pakistanis.
The issue extends beyond Pakistan’s position in the global ranking. Corruption and governance indicators can form part of the broader country-risk picture considered by investors, businesses, lenders and international institutions. The CPI is not an investment or credit rating, nor is it an IMF lending condition, but such indicators can influence perceptions of a country’s institutional environment.
The IMF’s 2025 Governance and Corruption Diagnostic for Pakistan said corruption and governance weaknesses have significant effects on economic growth and investment. Citing research, the IMF noted that every one-point improvement in the CPI is associated with a 4% increase in a country’s investment rate, while stressing that the relationship does not establish direct causation.
Pakistan’s score rises, but rank falls
Transparency International’s 2025 CPI, released in February 2026, gave Pakistan a score of 28 out of 100 and placed it 136th among 182 countries.
The previous CPI gave Pakistan 27 points and ranked it 135th among 180 countries. This means Pakistan’s score improved by one point, while its numerical position slipped by one place.
Transparency International says CPI scores are more meaningful than rankings because changes in the number and performance of countries can affect a country’s position.
What is behind Pakistan’s score of 28?
The CPI is not based on a survey of Pakistani citizens. Transparency International combines information from external sources and requires each country to have data from at least three sources. Pakistan was assessed using eight of the 13 sources used globally.
The index measures perceived public-sector corruption among experts and businesspeople.
Five of Pakistan’s eight published source scores remained unchanged, two declined and one, V-Dem, recorded a substantial increase.
Bertelsmann remained at 21, the Economist Intelligence Unit at 18, Global Insights at 32, PRS ICRG at 33 and the World Bank’s CPIA at 39. The World Economic Forum fell from 33 to 32, while the World Justice Project declined from 26 to 25.
V-Dem, meanwhile, rose from 14 to 19.
The published whole-number source scores add up to 219, producing an average of 27.375, compared with Transparency International’s published CPI score of 28.
This does not establish a mathematical error because Transparency International uses unrounded underlying figures. However, the exact calculation cannot be reproduced from the publicly displayed whole-number scores.
That makes the precise unrounded average and its distance from the threshold for a score of 28 important questions, particularly because V-Dem was the only source showing a substantial improvement.
Why did the V-Dem score rise?
V-Dem’s five-point increase is the only major positive movement among Pakistan’s eight CPI sources.
V-Dem uses expert assessments and statistical modelling, and its datasets can be revised as new information and versions become available.
Questions therefore remain about what specifically changed in Pakistan’s assessment, including whether the expert pool changed, historical data were revised, underlying indicators were updated or methodological changes affected the result.
The exact unrounded V-Dem score supplied to Transparency International is also not publicly disclosed in the CPI data.
Transparency International referred questions about the V-Dem increase to V-Dem, which produces the underlying data.
The five-point increase is particularly notable because seven of Pakistan’s eight published source scores either remained unchanged or declined.
Eight sources use different methods
The eight sources used for Pakistan are not identical surveys.
The World Economic Forum uses an executive opinion survey, while the World Justice Project combines household and expert information. Bertelsmann relies on expert assessments, V-Dem uses expert coding and statistical modelling, and the EIU, Global Insights and PRS ICRG use country-risk assessments. The World Bank’s CPIA is an institutional assessment.
The sources therefore involve different respondents, experts, analysts, definitions, assessment periods and levels of public disclosure.
Transparency International standardises the different scores before combining them into the CPI.
Using the standardised scores equally is a methodological choice and does not, by itself, establish that the CPI is flawed. However, it raises a question about whether assessments based on fundamentally different forms of evidence should carry the same arithmetic weight.
Another issue concerns statistical independence. Although the eight organisations are institutionally separate, their assessments may draw on some of the same court cases, government documents, media reports, international developments or expert networks.
This does not establish duplication or bias, but it means that eight sources should not automatically be interpreted as eight completely independent measurements of corruption.
The 2025 CPI does not measure one calendar year
The assessments underlying the CPI do not necessarily cover the same period.
Some sources may use information from 2024, others specific periods in 2025, while some assessments reflect longer timeframes. As a result, the 2025 CPI should not be interpreted as eight simultaneous measurements of corruption throughout calendar year 2025.
A reform introduced late in 2025 may not immediately appear in every source, while earlier events can continue to influence assessments.
A one-point change in the CPI therefore cannot automatically be attributed to a specific government, policy or reform.
How much of the data can be independently verified?
Transparency International publishes its methodology and explains how sources are selected, standardised and combined. However, it also acknowledges that some underlying information is private or proprietary.
This creates a distinction between methodological transparency and data transparency.
Researchers can understand the design of the CPI, but they cannot reproduce every Pakistan source score from the original observations.
More detailed scrutiny would require country-level data, assessment dates, respondent or expert numbers and types, response rates, sampling and weighting methods, treatment of missing data, unrounded scores and information about changes in methodology or expert pools.
Domestic surveys paint a different picture
The distinction between international perceptions and citizens’ experiences is also visible in Pakistan’s own surveys.
Transparency International Pakistan conducts a separate National Corruption Perception Survey and has clarified that it does not collect the data or calculate Pakistan’s global CPI score.
Its 2025 survey covered 20 districts across all four provinces and received 3,989 completed responses. Police were identified as the most corrupt sector by 24% of respondents, followed by tendering and procurement at 16% and the judiciary at 14%.
The survey found that 66% of respondents had not paid a bribe for a public service during the previous 12 months, while 77% were dissatisfied with government efforts to combat corruption.
Transparency International Pakistan said it changed its sampling approach in 2025 from non-probability convenience sampling to a multistage stratified cluster design.
Around 1,000 respondents were allocated to each province, with participants selected at public venues such as markets, parks, government buildings and community gathering points.
However, the published methodology does not provide full details on individual selection within venues, refusals, replacements or selection probabilities. Equal allocation among provinces also requires appropriate weighting because the provinces differ significantly in population size.
The published information is therefore insufficient to independently reproduce the national weighting and selection probabilities.
Court ruling highlights importance of survey methodology
A 2024 judicial ruling also illustrates why transparency over sampling and data treatment matters in domestic corruption surveys.
In April 2024, the Peshawar High Court ordered Transparency International Pakistan to recall and republish its 2023 National Corruption Perception Survey after examining findings concerning the Khyber Pakhtunkhwa judiciary.
The court found deficiencies in the survey’s “ethical and methodological rigour” and held that the findings concerning the KP judiciary were not credible.
The judgment questioned the treatment of responses concerning whether respondents had paid bribes to access public services. Transparency International Pakistan reported 760 respondents saying they had paid a bribe, while the court noted that responses from another 788 respondents had not been adequately explained.
The court described the issue as a major data-analysis flaw affecting the reliability and validity of the findings. It also criticised the survey for failing to make sufficiently clear that perceptions do not necessarily represent reality.
The court ordered the report to be recalled and republished with the relevant data and public-opinion ratios.
The judgment does not establish that the 2025 NCPS suffered from the same weaknesses, particularly because its methodology was substantially changed. It does, however, provide a documented example of why sampling, response handling and data treatment are important when corruption surveys are used to draw conclusions.
A larger sample does not guarantee accuracy
Transparency International Pakistan increased its 2025 sample to almost 4,000 respondents.
A larger sample can reduce sampling error, but it does not automatically eliminate selection bias. The methodology cites an approximately three-percentage-point margin of error and a cluster design, but design effects, weighting and respondent-selection probabilities are needed to independently evaluate the national precision of the estimates.
Some questions also received fewer responses than the headline sample of 3,989. The published survey does not prominently provide separate missing-response categories for every question or fully explain how missing responses were treated.
These limitations do not establish that the findings were manipulated. They mean that the published information is insufficient for an independent survey expert to fully reproduce the national estimates.
Khyber Pakhtunkhwa Chief Minister Accuses Federal Government of Massive Corruption
Perception and personal experience can differ
A separate survey by Ipsos and the Federation of Pakistan Chambers of Commerce and Industry (FPCCI) provides another perspective.
The Index of Transparency and Accountability in Pakistan, based on fieldwork conducted in December 2025 and January 2026, interviewed more than 6,000 respondents across 82 urban and rural districts and more than 195 tehsils. A separate sample of around 300 respondents was drawn from government institutions.
The survey found that 68% of respondents believed bribery was common in public institutions, while 27% said they had personally faced a demand for a bribe.
Similarly, 56% believed nepotism was widespread, compared with 24% who reported personally experiencing favouritism that undermined merit.
While 59% believed public servants became wealthy through illegal means, only 5% said they had personally observed such illicit enrichment.
These figures do not disprove the CPI or establish that corruption is low. Rather, they illustrate how perception and personal experience can measure different aspects of corruption.
Citizens, business executives and international analysts may therefore view the same institutional problem through different forms of evidence.
Why the distinction matters
The CPI is sometimes reduced to a statement that Pakistan is the “136th most corrupt country”. Such a description can be misleading.
A score of 28 does not mean that 28% of Pakistanis are corrupt, that a particular percentage of public money is stolen or that a specific proportion of citizens have paid bribes.
The CPI measures perceived public-sector corruption and is an indicator rather than an audit of actual corruption.
At the same time, dismissing the index simply because it is perception-based would overlook the broader governance concerns identified by international institutions.
The IMF’s 2025 diagnostic identified governance and corruption vulnerabilities involving areas such as public financial management, procurement, regulation, accountability and the rule of law, and linked these constraints to private-sector development and economic performance.
Why the score matters economically
Foreign investors and financial institutions do not make decisions based on the CPI alone. They consider a much wider range of economic, political, financial and institutional risks.
Corruption and governance form part of that broader country-risk assessment.
The IMF has linked governance weaknesses and corruption with investment and economic growth and conducted its Pakistan diagnostic because these issues can have macroeconomic consequences.
The CPI therefore carries significance beyond a simple media ranking. The key issue is whether the underlying evidence provides enough information for researchers, policymakers and the public to understand how the final score was produced.
Transparency International responds
Transparency International was asked a series of detailed questions about Pakistan’s 2025 CPI.
Regarding the increase from 27 to 28, TI referred to its technical methodology but did not provide the unrounded scores or exact arithmetic used to arrive at the final figure.
The organisation said V-Dem was the only source showing a positive change, while the World Justice Project and World Economic Forum declined slightly.
TI also said Pakistan’s measurement error became smaller because V-Dem had previously provided its lowest source score.
When asked why V-Dem’s score increased from 14 to 19, TI referred the question to V-Dem.
TI said none of Pakistan’s historical CPI changes has been statistically significant and described the country’s five-year trend as “stagnant”, with Pakistan consistently remaining in the bottom quartile.
The organisation also confirmed that it did not conduct a sensitivity test excluding V-Dem, saying the CPI uses all available source data rather than simulating results by removing individual country observations.
On the different methodologies used by the eight sources, TI said they all measure aspects of public-sector corruption and are highly correlated, citing an independent statistical audit by the European Commission’s Joint Research Centre.
TI said it cannot formally test statistical independence because of confidentiality constraints and instead uses methodological diversity as a proxy.
Regarding differences in reference periods, TI referred to its published source descriptions. It also said EIU, Global Insights and PRS ICRG provide the data points required for the CPI but did not provide details of their proprietary assessments or separate verification processes.
TI said an individual country score cannot be reproduced in isolation because rescaling depends on global means and standard deviations. However, it said the complete CPI has been independently reproduced and statistically audited by the European Commission.
The organisation also said some sources, including V-Dem, the World Justice Project and the World Economic Forum, may have different sample sizes, while their methodologies and questionnaires have remained stable.
TI acknowledged that the CPI is a perception-based proxy for public-sector corruption and said there is no independently verified measure of actual corruption.
It also noted that the index does not capture illicit financial flows, private-sector corruption, money laundering or citizens’ direct experiences and perceptions.
Transparency International further said the CPI does not measure individual government institutions and therefore could not identify a Pakistani institution responsible for the 2025 improvement, referring that question to Transparency International Pakistan.
The bigger picture
Transparency International stands by the CPI methodology and says the index has been independently reproduced and statistically audited.
Its responses also confirm several limitations: Pakistan’s one-point improvement was not statistically significant, no sensitivity test was conducted to determine the effect of excluding V-Dem, statistical independence among the sources could not be formally tested, and the CPI remains a perception-based proxy rather than a direct measure of actual corruption.
Pakistan’s CPI score has remained within a relatively narrow range in recent years: 28 in 2021, 27 in 2022, 29 in 2023, 27 in 2024 and 28 in 2025.
The available evidence does not establish that Transparency International manipulated Pakistan’s CPI score or that Transparency International Pakistan manipulated its domestic survey.
Instead, the different measurement systems offer different perspectives. The global CPI measures perceived public-sector corruption using eight external assessments, the domestic NCPS measures perceptions through a revised survey methodology, and the Ipsos-FPCCI index combines perceptions with reported personal experiences.
The central questions are therefore not simply whether Pakistan moved up or down the ranking, but what evidence produced the score, how much of that evidence can be independently examined and how closely international perceptions correspond with the experiences of people dealing with public institutions in Pakistan.
Originally published in The News.
