Corruption
How common is corruption? What impact does it have? And what can be done to reduce it?
This page was first published in October 2016. We made minor changes to the text in January 2024.
This topic page presents research and data on corruption — an important problem imposing political, economic, and environmental costs on societies worldwide.
Corruption involves many different aspects, and it is therefore hard to give a precise and comprehensive definition. However, at the core of most definitions of corruption is the idea that a corrupt act implies the abuse of entrusted power for private gain. Classic examples include bribery, clientelism, and embezzlement. Other, often more subtle and even legal examples of corruption include lobbying and patronage.
While long-run data on corruption is very limited, historical examples suggest that corruption has been a persistent feature of human societies over time and space. Two examples are the sale of parliamentary seats in ‘rotten boroughs’ in England before the Reform Act of 1832, and ‘machine politics’ in the US at the turn of the 19th century.1
The unethical and often illegal nature of corruption makes measurement particularly complicated. Corruption data usually comes from direct observation (e.g. law enforcement records and audit reports), or perception surveys (e.g. public opinion surveys or expert assessments). In this topic page, we discuss data from both sources and discuss their underlying limitations.
As we show, although precise corruption measurement is difficult, there is a clear correlation between perception and behavior; so available corruption data provides valuable information that, when interpreted carefully, can tell us something important about our world and contribute to developing effective policies.
For example, the data from perception surveys suggests that corruption correlates with human development, and several studies exploiting rich data from law enforcement records have shown that education is an important element explaining this relationship. Specifically, the data supports the idea that voters with more education tend to be more willing and able to monitor public employees and take action when these employees violate the law.
Related topics
Government Spending
What do governments spend their financial resources on?
State Capacity
Do governments worldwide have the ability to implement their policies? How has this changed over time?
See all interactive charts on corruption ↓
Empirical View
Evidence from surveys
Where is perceived corruption high?
The non-governmental organization Transparency International (TI) estimates a 'Corruption Perception Index', which is arguably the most widely used indicator of corruption worldwide and shown in the map here.
The index scores countries on a scale of 0-100, where 0 means that a country is perceived as highly corrupt and 100 means that a country is perceived as very clean. The indicator is representative of experts’ opinions, as it is constructed by taking the averages of various standardized expert surveys, including those from the Bertelsmann Foundation, the World Economic Forum, the World Bank, and many others.2
While TI's Corruption Perception Index has been estimated since 1995, the methodology has been changed recurrently up until 2012. As such, the data presented here starts in 2012. You can explore country-specific trends by clicking on the 'Chart' tab and then clicking 'Edit countries and regions' in the upper-right corner.
As we can see, countries that have received high scores (and thus are perceived as the ‘cleanest’) are Denmark, New Zealand, Finland, Singapore, and Sweden. At the other extreme, countries that have received low scores (and the highest perceived corruption) include Somalia, Syria, South Sudan, Yemen, and North Korea.
We can see that scores are fairly stable, and drastic changes in ranking are not common.
Where do people perceive corruption to be a major problem?
The data visualized previously relied on the perception of experts (e.g. business people, country analysts, etc.). Now we analyze data representing the perceptions of everyday people confronting corruption around the world.
The Global Corruption Barometer, also produced by Transparency International, surveys individuals worldwide, asking them about their opinions and experiences regarding corruption. The visualization shows the average national perception of corruption, as rated on a scale of 1 to 5 by respondents who were asked: "To what extent do you think that corruption is a problem in the public sector in this country?".
The map shows that everyday people's perception of the problems associated with corruption correlates with expert opinion (seen in the previous section) about how much corruption there is. However, the correlation is far from perfect, indicating that these two indicators present us with different perspectives.
The fact that the perceptions of experts and other citizens diverge is at the heart of perception measures: experts and non-experts have different reference points against which they assess whether corruption is a problem. Indeed, in countries where the public finds corruption extremely intolerable, the perception of its implications may be extreme, even if baseline corruption levels are lower than in other countries.
Corruption is sometimes hard to tackle precisely because it is common, so people perceive it to be a natural economic transaction: it is easier to act corruptly if many other individuals think it is acceptable to be corrupt.
Which institutions do people perceive as most corrupt?
The Global Corruption Barometer produced by Transparency International asks individuals across countries whether they perceive specific institutions as corrupt. The chart presents, by institution, the global aggregate figures. The numbers correspond to the percentage of survey respondents who think that "Most" or "All" of each institution is corrupt in their home country.
The estimates show that, globally, people perceive domestic police forces and the legislature to be particularly corrupt.
Perceived corruption of political parties
The next visualization digs deeper into corruption perceptions, specifically in the context of politics. It shows the percentage of survey respondents in each country who think political parties are "corrupt or extremely corrupt".
As we can see, there is substantial cross-country heterogeneity, and patterns again show differences concerning general corruption perceptions.
Where are people more likely to pay bribes to access public services?
Bribery is one of the most common forms of corruption. The following visualization shows the share of people who report having paid a bribe to access public services in the 12 months prior to the survey. The data comes from the Global Corruption Barometer, produced by Transparency International, and the public services in question are: education; judiciary; medical and health; police; registry and permit services; utilities; tax revenue and/or customs; and land services.
Once again, cross-country heterogeneity stands out. In some countries, a majority of survey respondents report having paid a bribe within the past year. In other countries, only a few percent of the respondents report having paid a bribe in the same time window.
While these results show that in some countries the paying of bribes is very rare, we have to consider that these estimates come from ordinary people interacting locally. As we show below, many countries where normal people do not frequently pay bribes, have far-from-perfect international records regarding international private-sector bribery.
How does petty corruption affect people with low incomes?
For those without money and connections, paying even small bribes to access basic public services such as public health or police can have significant consequences. Petty corruption in the form of bribes often acts as a regressive tax, since the burden typically falls disproportionately on people with low incomes.
This visualization, taken from the World Development Report (2011), uses data from Ecuador to estimate the cost of bribes paid relative to income. More specifically, the right panel shows the self-reported cost of bribes paid by households (as a share of household income), while the left panel shows the self-reported cost of bribes paid by firms (as a share of firm revenue). The presented estimates are in percent and come from a 1999 survey of 1,164 enterprises and another of 1,800 households.
As we can see from this example, petty corruption seems to act as a regressive tax: poor households tend to pay a larger share of their income on bribes to access public services.
Where are firms more likely to be asked for bribes?
Bribes are also requested from and paid by firms in the private sector. The following visualization shows, for a selection of countries, the percentage of firms experiencing at least one bribe payment request during six transactions dealing with utility access, permits, licenses, and taxes. The data comes from the World Bank's Enterprise Surveys.
According to this source, a large majority of firms report being asked for bribes in some countries, while in others, it is almost no firm.
Which countries are perceived by business executives as more likely to 'export corruption'?
The previous visualization provides evidence of the frequency with which firms are requested to pay bribes — the demand side of bribing. Here, we explore the supply side, using data from Transparency International's 'Bribe Payers Survey.'
The survey asks senior business executives around the world for their perceptions of the likelihood of companies from countries they have business dealings with engaging in bribery when doing business in the executive’s own country.
Based on the answers, Transparency International estimates the average of the scores given by all the respondents who rated each country (on a scale from 0 to 10, where 0 means firms always pay bribes in that country, and 10 means they never do). This gives the so-called 'Bribe Payers Index'.3
The map shows the results.
Evidence from law enforcement and regulation
Many firms from high-income countries engage in bribery across the world
The Foreign Corrupt Practices Act (FCPA) in the US was passed in 1977 to make the bribery of foreign officials illegal. The visualization from the Mintz Group shows the global distribution of all penalties in US Government FCPA cases since 1977. Darker colors in the map denote larger penalties for activities in the corresponding country. Total penalties are measured in US dollars, so this chart combines the number and magnitude of cases. In Argentina, for example, the dark color is mainly due to one very large case involving Siemens (450 Million USD penalty, in 2008). An interactive version of this map, including details regarding specific penalties, is available from the Mintz Group at http://www.fcpamap.com.
These official records show that US firms have paid bribes in 80 countries since 1977—including in many OECD countries.
Diplomats from countries with high corruption perception tend to break traffic rules abroad more often
Corruption is commonly defined as "the abuse of entrusted power for private gain." Here, we provide evidence of how diplomats in New York City abused their diplomatic status to break traffic rules by parking illegally.
UN mission personnel and their families benefit from diplomatic immunity. In New York City, until November 2002, diplomatic immunity implied that UN mission personnel could park illegally and avoid paying the corresponding fines. The visualization maps the average unpaid annual New York City parking violations per diplomat by the diplomat's country of origin from November 1997 to November 2002. The data comes from Fisman and Miguel (2007)5, who in turn obtained information compiled by the New York City Department of Finance.
As we can see from the map, this ‘revealed-preference’ measure of corruption among diplomats correlates positively with the survey-based measures of corruption we have already discussed. Diplomats from countries where corruption perception is low (e.g. Denmark) seem to be generally less likely to break parking rules abroad, even in situations in which there are no legal consequences.
While the correlation is not perfect (e.g. Colombia has a high corruption perception but zero unpaid parking violations in the data) Fisman and Miguel (2007) show that statistically speaking, the positive correlation between corrupt behavior by diplomats 'abroad,' and corruption perception 'at home,' remains after controlling for factors such as national income in the diplomats’ home country, or the diplomats' salaries. This evidence suggests that cultural norms are one of the factors that affect corrupt behavior.
OECD countries are increasingly providing procedures for public officials to report corruption
Corruption is not something that only affects low-income countries—and in fact, many high-income countries have become increasingly aware of this in recent years.
The table details a list of OECD countries and whether they have specific procedures for public officials to report misconduct or suspected corruption.
As we can see, since 2000 many OECD countries have developed formal mechanisms to allow public officials to expose corruption ('whistle-blowing') more easily. Specifically, the table shows that the share of OECD countries that reported having legal whistle-blowing procedures went up from 44.8% in the year 2000 (13 out of 29 countries in the sample) to 79.3% in 2009 (23 out of 29 countries in the sample). Today, many of these countries supplement legal provisions with internal rules (e.g. minimum requirements for whistle-blowing programs).
These recent changes show that legal provisions to fight corruption are relatively new, and high-income countries are still trying to find new policy instruments to deal with corruption.
What is the relationship between corruption and development?
This visualization shows the cross-country relationship between development (as measured by the United Nations Human Development Index) and corruption (as measured by Transparency International's Corruption Perception Index).
As we can see, countries that score higher in the Corruption Perception Index (i.e. countries seen as less corrupt) tend also to have better scores in the Human Development Index.
Education and corruption
The relationship in the visualization above is just a correlation: many factors simultaneously drive corruption and development. Education is an important case in point.
The next scatter plot provides evidence of the cross-country relationship between educational attainment and corruption. The horizontal axis measures corruption using Transparency International's Corruption Perception Index, and the vertical axis measures average years of schooling.
As we can see, there is again a strong positive relationship: countries where people are more educated tend to have better scores in the Corruption Perception Index.
Several academic studies have tried to establish the extent to which this relationship is causal. Glaeser and Saks (2006)7, for example, show that within the US, states that are better educated tend to be less corrupt—notably, they show that this relationship holds even when using historical factors like Congregationalism in 1890 as a proxy for the current levels of schooling. In other words, they find that historical levels of education predict differences in levels of corruption across States several generations later. This is consistent with other studies that support the theory that voters with more education tend to be more willing and able to monitor public employees and to take action when these employees violate the law.
What is the relationship between corruption and accountability?
One of the most widely accepted mechanisms of controlling corruption is to ensure that those entrusted with power are held responsible for reporting their activities. This is the idea behind so-called 'accountability' measures against corruption.8
The visualization shows the cross-country relationship between corruption and accountability. Here, corruption is measured as the share of people who admit having paid bribes in the past 12 months (as per the estimates from the Global Corruption Barometer), and accountability is measured by the Accountability Transparency Index developed by Williams (2015).9 This index is constructed from a number of underlying indicators that provide information about the extent of free media, fiscal transparency, and political constraints.
As we can see, people are less likely to pay bribes in countries with stronger institutions supporting accountability.
Ferraz and Finan (2011)10 show there is evidence that this relationship is causal. Specifically, they show that electoral accountability causally affects the corruption practices of incumbent politicians in Brazil. There is significantly less corruption in municipalities where mayors can run for reelection, and the positive effect of accountability via reelection is more pronounced among municipalities with less access to information and where the likelihood of judicial punishment is lower.
How effective are top-down audits to reduce corruption?
A common policy prescription to fight corruption is to increase monitoring and punishments. The logic supporting such policies is straightforward: better monitoring and harsher punishments increase the expected cost of acting corruptly, so people rationally choose not to break the rules.
To test the extent to which monitoring and punishments effectively reduce corruption, economists often rely on 'policy experiments,' where they administer these policies to 'treatment groups.' Olken (2007)11 follows this approach, increasing the probability of central government audits from 4 percent to 100 percent (the 'policy treatment') in the context of Indonesian village road projects.
The study compares the outcomes for villages that received this intervention with those that did not and finds that audits significantly reduced missing expenditures, as measured by discrepancies between official project costs and independent engineers’ estimates. The following visualization summarizes these results. The height of the bars shows the percentage of expenditures that engineers found missing.
As we can see, missing expenditures were much lower in villages where audits were certain.
The study provides further evidence of the extent to which officials in charge of road projects responded to private incentives: it finds that (i) audits were most effective when officials faced elections soon, and (ii) village elites shifted to nepotism (the practice of hiring family members), which is a form of corruption that was harder for audits to detect.
How important is the link between cultural norms and corruption?
A study by Fisman and Miguel (2007)13 shows that diplomats from countries where corruption perception is low (e.g. Denmark) seem to be generally less likely to break parking rules abroad, even in situations in which there are no legal consequences.
A world map showing the number of parking violations per diplomat across different UN diplomatic delegations in the New York, US, is available here. You can read more about the data in our section on Evidence from law enforcement and regulation above.
Fisman and Miguel (2007) show that the positive correlation between corrupt behavior by diplomats ‘abroad’ and corruption perception ‘at home’ remains after controlling for factors such as national income in the diplomats’ home country or the diplomats’ salaries. This evidence suggests that cultural norms are one of the factors that affect corrupt behavior.
Definitions, Measurement and Data Quality
Definition
Corruption involves many different aspects, and it is therefore hard to give a precise and comprehensive definition. However, at the core of most definitions of corruption is the idea that a corrupt act implies the abuse of entrusted power for private gain. Classic examples include bribery, clientelism, and embezzlement. Other, often more subtle – and sometimes even legal – examples of corruption include lobbying and patronage.
Measurement limitations
The unethical and often illegal nature of corruption makes measurement particularly complicated. Corruption data usually comes from direct observation (typically law-enforcement records) or perception surveys (e.g. general-population attitudinal surveys or expert assessments).
The main disadvantage of direct observation is that corruption is difficult to observe, as should be clear from its definition. This means that estimates are likely biased and are not generally suitable for cross-country comparisons since differences in levels may stem from differences in law enforcement rather than actual differences in corruption.
On the other hand, the main disadvantage of measuring the perception of corruption, rather than corrupt behavior directly, is that estimates are susceptible to how survey respondents — both experts and everyday people — form perceptions in the first place. For instance, differences in perceived corruption between countries may stem from differences in how corruption is defined or morally viewed rather than differences in actual behavior.
In any case, despite these limitations, both sources provide valuable information regarding underlying patterns. For example, studies relying on direct observation have been widely used for within-country studies, specifically in contexts where law enforcement is considered to be broadly constant across states or municipalities (see, for example, Glaeser and Saks 2006 14, and Ferraz and Finan 201115).
Relationship between sources
This visualization shows the relationship between two different survey-based measures of corruption. In the vertical axis, perception is measured using Transparency International's Corruption Perception Index (lower scores reflect higher perceived corruption). In the horizontal axis, corruption is measured using self-reported bribe paying behavior (estimated as the share of survey respondents who reported having paid a bribe to access public services in the last year).
As we can see from this plot, there is a clear association between both measures: countries where people often report having to pay bribes also tend to score low in the Corruption Perception Index.
The fact that corruption perception does contain information about corrupt behavior has been corroborated by detailed case studies. Olken (2009)16, for example, shows that citizens' perception of corruption in infrastructure projects correlates with objective measures of 'missing expenditures', as measured by independent engineers’ estimates in Indonesia.
Key Charts on Corruption
See all charts on this topic
Endnotes
'Rotten boroughs' were constituencies that had a tiny electorate, and could thus be used by a wealthy patron to secure influence; at the turn of the 19th century in immigrant US cities, 'machine politics' involved hierarchical political organizations that provided social services and jobs in exchange for votes. For more information, see: Aidt, T. S. (2003). Economic analysis of corruption: a survey. The Economic Journal, 113(491), F632-F652.
The list of sources that are considered (subject to availability for each country) are:
African Development Bank - Governance Ratings
Bertelsmann Foundation - Sustainable Governance Indicators
Bertelsmann Foundation - Transformation Index
IMD World Competitiveness Yearbook
Political Risk Services - Country Risk Guide
World Bank - Country Performance and Institutional Assessment
World Economic Forum - Executive Opinion Survey
World Justice Project - Rule of Law Index
Economist Intelligence Unit - Country Risk Assessment
Global Insight - Country Risk Ratings
Political and Economic Risk Consultancy - Asian Intelligence
Freedom House - Nations in Transit
Methodological details are available from Transparency International
Copyright© 2016 Mintz Group LLC. All Rights Reserved. The information and graphics contained in this map are copyrighted and may not be distributed, modified, reproduced in whole or in part without the prior written permission of Mintz Group LLC.
Fisman, R., & Miguel, E. (2007). Corruption, norms, and legal enforcement: Evidence from diplomatic parking tickets. Journal of Political economy, 115(6), 1020-1048.
This is Table D3 in OECD Government at a Glance (2009)
Glaeser, E. L., & Saks, R. E. (2006). Corruption in America. Journal of public Economics, 90(6), 1053-1072.
Transparency International defines accountability as "the concept that individuals, agencies and organisations (public, private and civil society) are held responsible for reporting their activities and executing their powers properly."
Williams, A. (2015). A global index of information transparency and accountability. Journal of Comparative Economics, 43(3), 804-824.
Ferraz, C., & Finan, F. (2011). Electoral accountability and corruption: Evidence from the audits of local governments. The American Economic Review, 101(4), 1274-1311.
Olken, Benjamin A. (2007). "Monitoring Corruption: Evidence from a Field Experiment in Indonesia." Journal of Political Economy 115(2): 200-249.
This Policy Briefcase was issued in March 2008, and revised in May 2012. The data comes from Olken, Benjamin A. 2007. "Monitoring Corruption: Evidence from a Field Experiment in Indonesia." Journal of Political Economy 115(2): 200-249.
Fisman, R., & Miguel, E. (2007). Corruption, norms, and legal enforcement: Evidence from diplomatic parking tickets. Journal of Political Economy, 115(6), 1020-1048.
Glaeser, E. L., & Saks, R. E. (2006). Corruption in america. Journal of public Economics, 90(6), 1053-1072.
Ferraz, C., & Finan, F. (2011). Electoral accountability and corruption: Evidence from the audits of local governments. The American Economic Review, 101(4), 1274-1311.
Olken, B. A. (2009). Corruption perceptions vs. corruption reality. Journal of Public Economics, 93(7), 950-964.
Cite this work
Our articles and data visualizations rely on work from many different people and organizations. When citing this topic page, please also cite the underlying data sources. This topic page can be cited as:
Esteban Ortiz-Ospina and Max Roser (2019) - “Corruption” Published online at OurWorldinData.org. Retrieved from: 'https://ourworldindata.org/corruption' [Online Resource]BibTeX citation
@article{owid-corruption,
author = {Esteban Ortiz-Ospina and Max Roser},
title = {Corruption},
journal = {Our World in Data},
year = {2019},
note = {https://ourworldindata.org/corruption}
}
Reuse this work freely
All visualizations, data, and code produced by Our World in Data are completely open access under the Creative Commons BY license. You have the permission to use, distribute, and reproduce these in any medium, provided the source and authors are credited.
The data produced by third parties and made available by Our World in Data is subject to the license terms from the original third-party authors. We will always indicate the original source of the data in our documentation, so you should always check the license of any such third-party data before use and redistribution.
All of our charts can be embedded in any site.