An impact of corruption on growth economics essay

Bachelor Thesis

Diva Tertia Almira, 10110860

Thesis supervisor: dr. KBT Thio

Date: 2nd February 2013

University of AmsterdamFaculty of Economics and BusinessBSc Economics and BusinessSpecialization: EconomicsCourse: General Economics / International Economics

Table of Contents

I. Introduction…………………………………………………………………..……………3II. Causes of Corruption……………………………………………………………..……….. 5III. Measures of Corruption…………………………………………………………………… 8IV. Literature Review…………………………..……………………………………………. 11V. Conclusion…………………………………….…………………………………………16VI. Reference List……………………………………………………………………………18

I. Introduction

Bureaucratic corruption has been widely condemned by the society because of the harmful consequences. It is commonly blamed for the failures of many developing countries. These criticisms are essentially based on moral ground as corruption reflects corrosive bureaucracy of a country. However, to be unbiased the emphasis should only be on the economic aspect and therefore disregards the ethical aspect. Bureaucratic corruption refers to the misuse of public office for private gain. The misuse of public office usually practiced in a form of bribery, nepotism, patronage or embezzlement. But the discussion is limited to a corruption of a particular form, which is bribery. Bribes can be perceived as tax, therefore, like tax, it increases costs for a transaction and in turn are associated with low economic activity. Bribes can also be perceived as a form of lobbying to influence or to deter a certain decision implemented by the government. As the practice of bribing are widespread across countries, the discussion of corruption sparks interest in the empirical literatures. The study in the field of corruption has a long history in economics literatures. The relationship between bureaucratic corruption and economic corruption has been a topic of discussion over the last 50 years. Unfortunately academics expressed their difficulty in measuring corruption. Empirical evidence in this field has also been limited due to difficulty in quantifying efficiency of governments. It is not surprising as corruption usually practiced in secrecy and silence. Fortunately recent study has started to use regression analysis and construct indexes to assess the level of corruption in a country. The study of cause and consequences of corruption attract academics to explore the relationship of corruption with economic growth. Most academics approves that corruption would have an adverse effect on economic growth. Nevertheless corruption could conceivably be beneficial to economic growth. To extensively discuss the association between economic growth and corruption, understanding the cause of corruption would be a preeminent first step. The source of corruption might be too varied and complicated to identify empirically. However there are few characteristics that every corrupt country has in common. This phenomenon could be an evidence to discover the characteristic of a country’s political, economic and social systems that might affect the cost or benefit of corruption. And thus would ultimately find a country characteristic that might cause a tendency toward corruption. The most obvious characteristic would be the economic development of the country. Though there are few other characteristics that society would find it surprising. To extensively discuss the matter, it is plausible to start after the origin of corruption. The second chapter discusses the main characteristic that would cause corruption and the empirical evidence. Issues regarding the direction of causality would also be discussed briefly. The following chapter elucidates the procedure, competency and description of important measure of corruption. In literature review, both opposing opinions regarding the relationship between economic growth and corruption would be discussed comprehensively. To finish, the conclusion would summarize the focal issues from all of the preceding chapters. A few suggestions regarding a solution in minimizing corruption would also be presented in the last chapter. To this extent, the thesis will attempt to examine the negative relationship between corruption and economic growth. The research question can be formulated as: Does corruption hinder economic growth? This thesis concludes that literatures reach no agreement regarding the relationship between corruption and economic growth. However to contrast both ideas, extensive discussion of two opposing opinions on the relationship between corruption and growth would shine a light to review both arguments. To date the study of corruption remains one of the most interesting and difficult subject to study.

II. Causes of Corruption

Why is corruption more widespread in some countries than others? Before discussing even further the relationship between corruption and economic growth, recognizing the cause of corruption is essential. A couple of decades ago there are not many literatures discussing the cause of corruption as it is difficult to study corruption empirically. Fortunately, academics in recent studies have found variety of factors that may correlate countries with high perceived corruption. This can be achieved by regressing corruption indexes on several potential explanatory variables. Economic development and transparency of political institutions is a few of the factors that are believed to contribute to the level of perceived corruption of a country. The strongest and most consistent factor is economic development. Treisman (2007) have found that lower perceived corruption correlates to higher economic development. For instance, in 2005 the correlation between log of per capita GDP and the probability for bribes is -0. 72. Furthermore the correlations between log of per capita GDP and the corruption perception indexes by Transparency International and World Bank are also robust, 0. 79 and 0. 81 respectively. The correlation also survived the inclusion of various controls including trade and inequality. These evidences verified that there is indeed a negative correlation between corruption and economic development. On the contrary, there are a few countries that were perceived as corrupt but can still grow rapidly. This higher economic development in turn increased democracy in some countries. Therefore it can be concluded that democracy or higher political rights can also be the key in decreasing corruption. As mentioned above, a transparent political institution is necessary to reduce the level of corruption. This is supported by empirical findings which confirm that higher political freedom will increase the World Bank corruption index. Political rights are measured by the Freedom House index, higher political rights and civil liberties will consequently lead to lower index. Decreasing the index by one unit is expected to decrease the World Bank corruption index by 0. 13 in 2005. Unfortunately the results are less significant by using Transparency International corruption Index. And the results are not always significant if substituting Freedom House index with other political freedom index. Treisman (2007) also argued that besides political rights, press freedom, the duration of democratization and how the President is elected in the country each has an effect on the corruption index. The higher the press freedom, the higher the corruption index. The press freedom index is also constructed by FH. An increase of press freedom by one unit is expected to increase the corruption index by 0. 012. A last interesting discussion is the Presidential system. A country which elect their President directly has lower corruption index by 0. 11 compared to those countries whose President is elected by the legislature. In support, his previous works, Treisman (2000) also found that countries which have been a democratic country without interruption since 1950 have a significant impact on corruption. This study investigated if democracy has been established for a longer time will positively affect the corruption index. If a country has been a democracy since the year 1950, the corruption index is expected to increase by 0. 56. He argued that the current state of political system is not significant, but a long exposure to democracy would lower corruption. Other notable factors are economic rents and market competition. Ades & Di tella (1999) presented evidence that corruption have a relationship with economic rents available for civil servants to capture. Corruption tends to be greater where there are more economic red tapes in the country especially if the economy is oriented toward natural resources. They explain corruption as industrial policy function which shows subsidies in manufactures are correlated to corruption indexes. At the same time, the degree of openness to trade will reduce monopoly power of domestic firms. Thus diminishing the profits available for corrupt civil servants. In the same direction, Treisman (2000) found that the share of imports in GDP is correlated to Transparency International Index from 1996 to 1998. Svensson (2005) also found that highly corrupt countries are less open and regulate their market more extensively. From his results highly corrupt countries indeed have a significantly different policy characteristic from countries with low corruption rates. Unfortunately the above factors of corruption that has been presented can be ambiguously categorized as either consequences or causes of corruption. The direction of casualty is blurred in few cases. For instance, openness to trade and the extent of red tapes are highly endogenous in the direction of casualty with corruption. Treisman (2000) found that high corruption could cause a high level of bureaucratic regulation but not the other way around. The data could not be concluded that government regulatory interventionism would leads to higher perceived corruption or corruption can be reduced by merely limiting red tapes. Even more alarming, corruption may also both be the consequence and cause of a variable at the same time. This would eventually trapped countries in a vicious circle. Where by intervening this variable would only escalate corruption even further. There are several attempts to identify the correct causality of corruption. But the issues of casualty have not been fully resolved, since causality may also works in both directions.

III. Measures of Corruption

Another difficult task for academics is to measure the magnitude of corruption. Not until a couple of decades ago, corruption has not been measured quantitatively. As most corruption is surreptitious, empirical work on quantifying corruption is extremely problematic and challenging. Measuring government efficiency is not necessarily labeled as an exact science either. Consequently, empirical research on this field is fairly rare and the empirical results are mostly ambiguous or inconclusive. As it is difficult to measure corruption, perception index is one of the best alternative methods to measure corruption. Perception index measures corruption by expert assessments and public surveys regarding corruption. Fortunately more and more establishments published a perception index towards corruption. A widely recognized index is referred as the Corruption Perception Index, henceforth CPI. Countries are graded annually by an index from 0 to 10 by the Transparency International. A 0 indicates that a country is highly corrupt. And if a country scores a 10, its government can be perceived as clean. The method used to construct the index is by public assessments through more than ten different institutions. The assessments compile questionnaire regarding country’s abuse of public power exclusively in form of bribery and the effectiveness of eradicating corruption in the public sector. Consequently it will cover both political and administrative parts. It also aims to reduce measurement error by averaging different resources. Similarly, World Bank also published a rating control of corruption. The two ratings are highly correlated with correlation of 0. 98 in 2004. Both have slightly different research methods. But the most striking difference is that Transparency International only covers countries that have three components rating. This will lead into smaller country coverage but will indeed increase precision. Another index that has been used by the earliest empirical study on the relationship between corruption and economic growth is the Business International Index, henceforth BII. This index is founded by the firm BI and is based on a questionnaire answered by correspondents in 70 countries. The index comprises of nine sub- indexes. Including Political Change–institutional, Political Stability—social, Probability of Opposition Group Takeover, Stability of Labor, relationship with neighboring countries, bureaucracy and red tape, terrorism, legal system, and corruption. Even more complex than previous indexes, International Country Risk Guide (ICRG) formulates an index compromising more than 20 variables. The index includes both descriptive assessment and economic data. It divides into 3 subcategories of risk including political, economic and financial risk rating. Each subcategory compromised into several risk components and weights. As corruption is one of the most significant components to political stability, it is assigned as one of the variables. The measurement is based on points, 6 as the uppermost grade and 0 as the least desirable one. The foremost advantage of ICRG is that ratings are available since 1980 annually. This index also has a high correlation with Transparency International and World Bank index, with correlation of 0. 88 and 0. 84 respectively. Although these indexes have been used frequently in numerous leading journals, the reliability has been questioned from the start. Knack (2006) argues that the data do not measure corruption itself but only opinions and perception towards the level of corruption. Such opinions are usually not based on personal experience and therefore could be biased. Moreover expert assessment might be influenced by their western preconceptions and familiarity with certain cultures. Other concerns focus on the sources of constructing index as different index measures different subjects. For instance, both Transparency International and World Bank have different source of survey. Some sources are evaluations by foreign experts; others are a poll for local citizen or international investors. As a result, many wonder what exactly the average is measuring and question the competence. Even if identical survey questions were asked, it would be interpreted differently between different nationalities. For instance, a small amount of bribe is considered to be a minor inconvenience in a country with rampant organized crimes but not in developed countries. Thus comparability of surveys across countries cannot be compared fairly. In defense, Lambsdorff (2004) argues that there is no better alternative. This is especially true in the early study of corruption, as there were only a few sources can be found in this very sensitive subject. Rates of prosecution or conviction for corruption might be an alternative but these are likely only reflect the competency of police and judiciary not corruption as a whole. Another alternative is by comparing the value of existing infrastructure stock to past infrastructure spending. Unfortunately this approach is clearly difficult to extend the research cross-nationally. Aside from those two there are other plausible alternative. Recently, Transparency International began to ask respondents about their own experiences of corruption. Such data would be less likely to be biased and distorted by impressions or false news. Another defense to support the use of perception index is the fact that all of these ratings are highly correlated. Treisman (2007) explains that even though each index has a slightly different methodology to conduct their researches, these different indexes are aimed at a common target. Even the correlation between domestic and international business executives surveys by World Economic Forums in 2005 is correlated at r= 0. 87. The ratings reflected no Western or culture bias, as both experts’ surveys are highly correlated.

IV. Literature Review

Corruption has been widely criticized by the society because of the negative implications. Apart from criticism based on moral grounds and incompetence, corruption is assumed to have important prejudicial effects on the economic growth. Here corruption is defined as a misuse of public resources for private gain. Surprisingly, corruption might have been beneficial for economic growth at some levels. Leff (1964) argued that corruption is commonly perceived negatively and this becomes an important obstacle to an unbiased re-examination of the subject. He reasons that corruption can be favorable; especially as the economic policies in many underdeveloped countries may be based on priorities other than global economic development. Thus with this reasoning, corruption might have induce government to have favorable view toward activities that would improve economic development. It also provides incentive for government officials to have more energetic action towards entrepreneurs. According to Leff (1964), corruption also reduces uncertainty and increases investment. As investment always take place around risk and uncertainty, the magnitude of risk is magnified around political environment. In short by enabling investors to take more control, reduce uncertainty and influence their environment, corruption can increase the rate of investment. He also perceived corruption as a hedge against bad policy. Even though government intends to promote growth, there is no assurance that its policies are successful to achieve the goal. In this case, corruption can reduce the losses from such mistakes. For instance, the agricultural producers whose graft sabotage Juan Peron’s – a former President of Argentina – economic policies were later thanked for having maintained Argentina’s capacity to import. There are also several other arguments concerning competition and efficiency. Since license and other favors from the government officials are in limited supply, the bidding for such favors are competitively allocated among entrepreneurs. And because payment of the highest bribes is the most principal criteria for allocation, the importance to produce revenue become even more significant. As efficiency of production is one of the dependent variable to producing revenue, a tendency toward competition and efficiency is introduced into the system. From all of these arguments, it can be concluded that ” Efficient Corruption” may allow entrepreneurs to perform better in an economy overwhelmed by bureaucratic hold-ups and bad rigid laws. In support to arguments by Leff (1964) toward competition and efficiency, Lui (1985) examined the context of a queue where customers having different values of time are ranked by their bribe payments to the queue servers. The literature revolves around the Queuing Model examining Myrdal’s hypothesis. Myrdal (1968) hypothesized that corrupt bureaucrats have tendencies to intentionally cause administrative delays to attract more bribes. Whilst Queuing Model is defined by Lui (1985) as a queue where customers come and wait to one end of the line to receive prizes distributed by a server at the other end. This study assumed that an individual arrives in the queue and decides to either bribe or stay put to receive a gift. The amount of bribe will determine the time the individual will spend in the queue. Lui (1985) mentioned that the desired socially optimal solution is consistent with the individual optimization strategies. This implies that there is a Nash Equilibrium in the model. Thus if there is a system built around bribery for allocating favors from bureaucrats, it may lead to a favorable outcome where the most efficient firms will be able to afford to pay the highest bribes. On the other hand, Mauro (1995) concluded that there is a link between corruption and growth. The journal firstly defined what is meant by corruption used throughout the study. High bureaucracy or red tape, inefficient judicial system and unstable political system are regarded to be under the umbrella of corruption. To measure the aforementioned forms of corruption, Mauro (1995) used the Business International (BI) index between 1980 and 1983. This index is founded by the firm BI and is based on a questionnaire answered by correspondents in 70 countries. The index comprised of nine sub- indexes. The study took three sub-indexes that were most relevant to his study to derive a bureaucratic efficiency level. In other words, the bureaucratic efficiency level in his study comprises of the judiciary system, red tape and corruption sub-indexes. Each index was given a grade between 0 and 10. Grade 10 on judiciary system is better than a 5; grade 5 on corruption and red tape is worse than a 10. To no surprise, these three indexes were correlated to one another. The correlation between corruption and red tape was 0. 79 and to judiciary system was 0. 78. The correlation between red tape and judiciary system was 0. 78. This is because in countries where red tape hinders bureaucratic efficiency, there is bound to be corruption. In countries where corruption in the form of bribery becomes a tradition, government officials tend to cause bureaucratic delay unless given a sum of money. There are times in the study where he used institutional efficiency, which he derived it from using all nine indexes from BI and the variable political stability. Then he introduced an ethnolinguistic fractionalization (ELF) index, which measures the probability that two randomly selected individuals from a country will not belong to the same ethnolingustic group. A high ELF index implies that a country is highly diversified in terms of ethnic and spoken language. He found that ELF is negatively, and more importantly, correlated with institutional efficiency and bureaucratic efficiency. The correlation between ELF and institutional efficiency was -0. 38 and with bureaucratic efficiency was -0. 31, both significant at 1 per cent. In a country where different ethnic groups are present, government officials may treat different ethnic groups differently, making corruption worse. The focus of Mauro’s (1995) paper is the effect of corruption on investment and, ultimately, economic growth. He found that higher corruption index, which implies less corruption, is associated with an increase in investment rate. Specifically, an increase in standard deviation of one of the bureaucratic efficiency index is associated with an increase in investment/ GDP of 4. 75 per cent. This is obtained by multiplying the standard deviation of bureaucratic efficiency in the data and the slope coefficient of bureaucratic efficiency in a simple regression by OLS estimation. For instance, if Indonesia could improve it bureaucratic efficiency (2. 25) to the level of Malaysia (7. 00), its investment rate could increase by almost 15 per cent. If ethnolinguistic fractionalization is instrumented in the 2SLS, a one standard deviation increase implies a 9. 3 per cent rise in investment/ GDP. In addition, Svensson (2005) update Mauro’s (1995) calculations and also find similar results. Mauro’s (1995) explanation for this is: ethnolinguistic fractionalization slows down the diffusion of ideas and technological innovations. Investment level is one of the most important elements of economic growth. The higher the investment rate, the higher the economic growth, all things equal. Mauro (1995) have evidence that the bureaucratic efficiency has an impact on growth. A one standard deviation increase in bureaucratic efficiency implies a 1. 3 per cent increase in GDP per capita growth. Again, taking the example of Indonesia and Malaysia, Indonesian GDP per capita growth can increase by 4. 04 per cent if it can improve its bureaucracy to their Malaysian counterparts. He also finds that other window, such as productivity and cost, has much weaker evidence compared to investment. This proved that corruption reduced economic growth via investment window. Mendez and Sepulveda (2006) used a similar framework to find the relationship between GDP per capita and corruption level. One key difference between this research to the one by Mauro (1995) is the source of the corruption index. Mauro (1995) used the index issued by Business International whereas Mendez and Sepulveda (2006) referenced three institutions namely the Political Risk Services that publishes the International Country Risk Guide, Institute for Management that publishes their own index of corruption and lastly the Transparency International that publishes the Corruption Perceptions Index. Another difference is how they separate countries according to their political freedom. Using the index of freedom from the Freedom House International, they differentiate countries by ‘ free’ countries and ‘ not free’ countries. Taking the whole list of countries, corruption does not appear to be a significant determinant of growth. The results are the same for all three corruption indexes. The estimated coefficient on corruption is negative, stating a negative relationship between economic growth and corruption. Even so, the finding is not significantly different from zero. However, the free countries sample showed that a positive but minute level of corruption can influence economic growth positively. Thus the empirical evidence proved that political regime is an important determinant of the relationship between corruption and economic growth. The results they obtained were significant at a 1 per cent level. When taking the sample of not free countries, their results were inconclusive as each corruption index showed different conclusion. The interpretation of the not free country results is as follows. The boundary between corrupt and legal practices is blurred and the corruption indexes might not reflect these differences. Moreover the fund grabbing corruption over speed money corruption might be higher in not free countries and is likely to go unnoticed in the data. For corruption to influence growth, Mo (2001) discussed extensively the channel on which corruption directly affect economic growth. The journal developed an analytical framework to estimate the effects of corruption and the channels through which it affects the rate of GDP growth. The transmission channels included are investment, human capital and political stability. Contrary to the findings of Mauro (1995) and Mendez and Sepulveda (2006), the study by Mo (2001) has evidence that higher corruption hinders economic growth. He constructed a model for which total productivity is a function of corruption level, initial per capita GDP and human capital stock. In turn real GDP growth rate is a function of total productivity, investment over output ratio and growth of labor. The regression analysis showed that an improvement of the corruption index by one unit implies an increase of the growth rate by 0. 545 percentage points. The novelty of his literature is that he broke down and described the channels as to which corruption affects economic growth. The four channels are as follows: the direct channel, the investment channel, the human capital channel and the political stability channel. He concluded that the channels contributed 11. 8, 21. 4, 14. 8 and 53 per cent respectively. Even though it was seldom discussed in similar literatures, his empirical finding shows that political instability channel is the most important channel. As Mo (2001) explained it, corruption creates the opportunity for increased inequality and makes rent seeking activities more desirable. The increased inequality reduces productivity growth, investment and job opportunities. These in turn create sociopolitical instability.

V. Conclusion

The main aim of this thesis was to investigate the relationship between corruption and economic growth. This field of social studies is relatively new therefore past studies are not found in abundance. As the empirical study of corruption keeps evolving and advancing, more viable empirical works are published. This is remarkable considering corruption is a frustratingly difficult subject for a social scientist to study. In recent developments, academics have found number of factors that may link economies with high perceived corruption. This is attained by regressing corruption indexes on several potential explanatory variables. As previously explained; countries that is perceived by citizens and experts to be less corrupt are the one which are highly developed, transparent government, minimum economic rents and open to international trade. However the direction of causality of these factors still cannot be classified as either consequences or causes of corruption. Some literatures claim that corruption can also be a consequence and cause of a variable at the same time. As a result countries would eventually be trapped in a vicious circle. One major hindrance for studies about corruption is the issue of how corruption is quantified. Past studies tried many variations on the measurement of corruption. Fortunately researchers are getting better at constructing indexes of perceived corruption. The most widely known measure up to today is the CPI. It is a simple measurement conducted by Transparency International which gives a grade 0 to a completely corrupt government and a 10 for a completely clean government. More complex indexes are the Business International Index and the ICRG. The BII measures corruption for 9 different elements. It relies on the survey response in which it asks business establishments in a country. The ICRG uses 20 variables to quantify corruption, making it a very complex measure. More important variables are weighted more heavily than other ‘ less important’ variables. Knack (2006) pointed out that the indexes discussed are based on opinions of individuals and businesses. It cannot be measured precisely as other economic indicators such as economic growth. However, Lambsdorff (2004) explains that there is no other way to indicate the magnitude of corruption and these indexes are sufficient to reflect the level of corruption. Lastly, Treisman (2007) argues these indexes have a common goal even though the methodologies are different across each one. Past academics are split on the discussion of the role of corruption in the economy. The literature by Leff (1964) justifies corruption in a sense that it improves the efficiency of an economy. It may reduce uncertainty and thus boost investment which encourages entrepreneurs to be more competitive. Through the Queuing Model by Lui (1985) also supports this side of the debate. An individual standing in a queue is likely to bribe the authorities in pursue of a better position in line. If individuals are figuratively representing firms, then the most efficient firms will stay in the market as the most efficient players. However empirical studies by Mauro (1995), Mendez and Sepulveda (2006), Svensson (2005) and Mo (2001) claim corruption have negative impacts on the economy. Mauro (1995) found lower perceived corruption increases the investment rate. In turn, higher investment can result in higher economic growth. Svensson (2005) found a negative relationship between corruption economic growth and it is statistically significant. Mo (2001) also found that corruption is deterrent to economic growth through four different channels namely the direct channel, the investment channel, the human capital channel and the political stability channel. It was concluded that an improvement of the corruption index by one unit may boost economic growth by 0. 545 per cent. In conclusion, it is in the best interest of every country to reduce corruption. It is evident from past studies that countries with less corruption is more rich and more advanced than corrupt ones. There are numerous attempts by government to eradicate corruption although some policies are not as effective as others. To achieve low corruption, countries can focus internally by on improving political institutions. Secondly, governments can establish an independent anti- corruption agency which is directed at supervising business transactions and ensure the transparency of government institutions. As corruption is closely related to country’s legal system it would be advisable to improve the competence of every part of the legal system including police and judiciary. Lastly hiring private firms to manage public sectors such as import inspection on custom duties would also decrease the opportunity for officials to receive bribes. Not only it would improve the performance of the sectors but also would decrease corruption.

VII. Reference List

Ades & Di Tella (1999). Rents, Competition and Corruption. The American Economic Review, Vol. 89, No. 4, pp. 982-993. Aidt, Dutta and Sena. (2008). Governance Regimes, Corruption and Growth: Theory and Evidence. Journal of Comparative Economics 36, pp. 195–220. Bardhan, Pranab. (1997) Corruption and Development: A Review of Issues. Journal of Economic Literature, Vol. 35, No. 3, pp. 1320-1346Knack, Stephen (2006) Measuring Corruption in Eastern Europe and Central Asia: A Critique of the Cross-Country Indicators. World Bank Policy Research Working Paper 3968. Johann & Lambsdorff (2005) Consequences and causes of corruption: What do we know from a cross-section of countries? Passauer Diskussionspapiere: Volkswirtschaftliche Reihe, Vol. 34, No. 05. Leff, Nathaniel H. (1964). Economic Development through Bureaucratic Corruption. American Behavioural Scientist 8, pp. 8 –14. Lui, Francis T. (1985). An Equilibrium Queuing Model of Bribery. Journal of Political Economy Vol. 93, No. 4, pp. 760-781Mauro, Paolo. (1995). Corruption and Growth. The Quarterly Journal of Economics, vol. 110, pp. 681– 712Mendez and Sepulveda. (2006). Corruption, Growth and Political Regimes: Cross Country Evidence. European Journal of Political Economy Vol. 22, pp. 82– 98Mo, Pak Hung. (2001). Corruption and Economic Growth. Journal of Comparative Economics 29, pp. 66–79Myrdal, Gunnar. (1968). Asian Drama: An Inquiry into the Poverty of Nations. New York: Pantheon, Chapter 20. Svensson, Jakob. (2005). Eight Questions about Corruption. Journal of Economic Perspectives 19, pp. 19–42. Treisman, Daniel. (2000). The Causes of Corruption : a cross-national study. Journal of Public Economics 76, pp. 399-457. Treisman, Daniel. (2007). What Have We Learned About the Causes of Corruption from Ten Years of Cross-National Empirical Research?. Annual Revision of Political Science 10, pp. 211-244.