Click here for search results
Online Media Briefing Cntr
Embargoed news for accredited journalists only.
Login / Register

Six Questions on the Cost of Corruption with World Bank Institute Global Governance Director Daniel Kaufmann

1. Given the scale of corruption in developing countries, why should governments increase aid levels?

First, as shown clearly by the data, the scale of corruption varies significantly from country to country. While many developing countries face the challenge of addressing systemic corruption, some emerging economies in fact have lower levels of corruption than some wealthy countries. Second, even within the group of countries with very high levels of corruption, there are two distinct categories: those with leadership and resolve to do something about it, and those without. We should be wary of sweeping generalizations.

Third, it is well known by now that the current levels of aid required to meet the major developmental challenges, as set out in the Millennium Development Goals (MDGs) for instance, are inadequate. A significant scaling up in aid is obviously required. Yet this is a necessary precondition; in itself it will not be sufficient without the global community, multinationals, domestic firms and the national leaders doing their part on the challenge of good governance and corruption control.

The high level of country variation in the quality of governance and control of corruption indicated in the first two points above suggests therefore that a combination of a constructive yet tough approach is warranted: for their part the rich world ought to commit to larger aid levels _ which are still such a small fraction of their national incomes_ yet such 'good' additional money should not go after 'bad' money. In other words, those countries that have already made inroads in lowering corruption, or are making progress - with committed leadership - should be particularly rewarded.

Aid should also flow to institutions outside the government. And obviously it does make sense to have clear limits on amounts of aid flowing to governments where corruption and misgovernance is very high, as the World Bank and other institutions do have.

2. How large is the 'Global Corruption Industry', measured in terms of financial amounts per year worldwide?

Corruption is a very large 'industry'. Yet until very recently, it was virtually impossible to venture an estimate of the extent of corrupt annual transactions. In fact, only a few years ago, corruption was regarded as impossible to measure. Thanks to the 'explosion' in measurement approaches and actual data in this field, at least it is now possible to estimate rough orders of magnitude. Our focus is on measuring the extent of bribery from the private sector (firms and individuals) to the public sector.

A conservative approach to such measurement gives an estimate for annual worldwide bribery of about US $1 trillion dollars (US $1,000 billion.). We obtain figures on bribes from worldwide surveys of enterprises, which ask questions about bribes paid for the operations of the firm (licenses, regulations, etc.), as well as bribes paid to get favorable decisions on public procurement. Further, an estimate on bribes paid by household users of public services is derived from governance and anti-corruption diagnostic surveys.

There is a margin of error in all these estimates, so we should regard them as preliminary orders of magnitude. But the main point is that this is not a relatively small phenomena of a few billion dollars - far from it.

3. But can this estimate of bribery be equated with overall corruption worldwide?

A comprehensive estimate of worldwide corruption would exceed the estimate of bribery alone. This estimate of annual worldwide bribery of about US $1 trillion does not include the extent of embezzlement of public funds (from central and local budgets), or from theft (or misuse) of public assets. We do not have an estimate for the worldwide extent of embezzlement in the public sector. Consequently, the above bribery estimate (between firms and the public sector) does not cover every form of corruption (such as embezzlement).

While there are no worldwide estimates of embezzled funds, we do know that it is a very serious issue in many settings. We do know, for instance, that in countries where there is widespread corruption, embezzlement can be a serious problem. For instance, focusing exclusively on the top leadership, Transparency International estimates that Suharto embezzled anywhere between US$ 15-35 billions in Indonesia, while Marcos in the Philippines, Mobutu in Zaire, and Abacha in Nigeria may have embezzled up to US$ 5 billion each.

Furthermore, the $1 trillion estimate does not include the full extent of 'tainted procurement', but only the bribe fees associated with such procurement. While, for instance, the estimates of bribery exchanging hands for public procurement bids can be estimated in the vicinity of US$200 billion per year, the overall annual volume estimate of the 'tainted' procurement projects, where such bribes take place, may be close to US$1.5 trillion or so. Finally, there is no attempt to include the extent of fraud within the private sector, but only bribery transactions between the private and public sector.

It should be noted, at the same time, that the $1 trillion worldwide annual bribery estimate does include bribes between firms and public officials or politicians in the industrialized world, and also between multinational corporations from industrial countries bribery to the public sector in emerging economies. It also includes bribery within emerging economies.

4. Can we then say that the developmental cost of corruption is at least US$ 1 trillion per year? If not, are there estimates of the developmental cost of corruption?

No to the first part of the question; Yes to the second. The $1 trillion estimate refers to the annual worldwide transactions only - and of a particular type _ and does not account for the significant losses in investment, private sector development, and economic growth to a country, or to the increases in infant mortality, poverty and inequality _ all resulting from corruption and misgovernance.

Thus one needs a different type of data and analysis to assess the real developmental cost of corruption and poor governance. Such analysis is currently available. For instance, in our research we have found that there is a '400% governance dividend' of good governance and corruption control: countries that improve on control of corruption and rule of law can expect (on average), in the long run, a four-fold increase in incomes per capita. Thus, a country with an income per capita of US$ 2,000 could expect to attain US$ 8,000 in the long run by making strides to control corruption. Similarly, such a country could expect, on average, a 75% reduction in child mortality.

We have also found that the business sector grows significantly faster where corruption is lower and property rights and rule of law is safeguarded. On average, it can make a difference of about 3% per year in annual growth for the enterprises.

At the national level, other researchers also find a very significant difference in income growth rates. The difference can be between 2 and 4 percent per annum in the country's annual income growth rates between countries with a different extent of corruption control (moderate vs. poor, or good vs. moderate - when the difference is one standard deviation in control of corruption, controlling for other factors). It has also found that corruption constitutes a very substantial 'tax' on investment. For instance, one study on Foreign Direct Investment estimates that corruption is the equivalent of a 20% tax to foreign investors.

In the in-depth country diagnostic work, we also find that corruption and bribery is a regressive tax. Not only smaller enterprises pay a higher share of their revenue in bribes than their larger counterparts, but also poorer households bear a disproportionate share of the bribery burden, paying a much higher share of their incomes than higher income households - often for public services that were expected to be provided for free. Researchers at the International Monetary Fund, utilizing worldwide data on income distribution, also find that corruption is associated with increased income inequality.

5. Does the data suggest that it is the responsibility of Governments to find the solution to the corruption problem? To what extent are firms most often extorted?

Just as it has become clear that the challenge of addressing corruption is not only one for emerging (or poor) countries and their governments, the data and research carried out at the Bank, based on enterprise surveys, also indicates the extent to which some powerful corporations exert undue influence on state institutions, its laws, regulations and policies, often through illicit means. Thus while extortion of smaller and weaker firms by politicians and public officials is prevalent in many settings, the evidence also indicates that there is an important phenomena of undue influence and capture by powerful firms _ which we label as 'state capture'. We find that the extent of monopolistic vested interests and state capture varies significantly across countries. Where state capture is a major constraint, the private sector as a whole suffers by growing much more slowly - in contrast to the firms that are 'purchasing' laws and regulations, which do benefit privately. Some of the firms engaging in capture are multinationals, others are part of the domestic elite. At the same time, there are many multinational and domestic corporations which are making a commitment to good governance and integrity, and prepared to work with governments to have an investment climate with less corruption.

In synthesis, the data shows that good governance and addressing corruption ought to be central elements of an improved investment climate and business environment. But because of the particular interface between the private and public sectors, the corporate sector also has a key responsibility in an improved investment climate.

6. Can the data also be helpful in providing feedback to the World Bank to improve its work on governance and anti-corruption?

Definitely. The data is very important for evaluating the effectiveness of our and other donor efforts in this field. For instance, a recently concluded Global Poll commissioned by the World Bank indicated that, while the Bank had made strides on many development assistance areas, it needed to do even more on anti-corruption. Analysis of the responses from emerging economies suggest, in fact, that where corruption is higher, the likelihood of development loans being wasted is also much higher. Similarly, responses from an internet survey on questions about corruption - including many respondents in government, civil society and the private sector in developing countries - tell us that their view that without a tough commitment by the country to control corruption, donor funds should not flow freely. And yet another set of surveys, based on responses from enterprises, indicates that on average the firms evaluate poorly the multilateral organizations efforts to promote private sector development where there is high corruption and misgovernance. By contrast, where governance and corruption control is better, the firms evaluate significantly better the efforts by our international organizations to promote business development.

Related Links
$1,000,000,000,000 and counting…
How the Bank Helps Countries Fight Corruption
Tackling Corruption In Indonesia
10 Things You Did Not Know About the World Bank and Anti-Corruption
The Data Revolution: Measuring Governance and Corruption
Comment on $1 Trillion - The Cost of Corruption
Corruption - Related Links
Corruption Issue Brief


Permanent URL for this page: