THE IMPACT OF CORRUPTION ON GDP PER CAPITA

The paper studies the impact of corruption on the GDP per capita. A panel data covers the period between 2003 and 2011. The objective is to test the hypothesis that there is a strong negative impact of corruption on the GDP per capita. Three tests were conducted, the pool OLS, The Fixed Effect and the Random effect estimations. The main result of this study is that all three tests had shown strong statistically significant negative impact of corruption on the GDP per capita.


INTRODUCTION
There are mixed results in the literature in the debate over the impact of corruption on economic growth.While The efficient grease' hypothesis argues that corruption enhances efficiency in the economy , where bribes serves as a lubricant that reduces delays and transaction costs and thus leads to growth (Leff (1964) and Huntington (1968)), other academic work show that corruption negatively impacts economic growth.Kaufman and Wei (1999); Aidt (2009); Mauro (1995) and Mauro (1997), Shleifer and Vishny (1993), Blackburn et al (2009), Barreto (1996); Tanzi (1997)) The purpose of this study is to examine the impact of corruption on countries' GDP per capita.Transparency international publishes yearly a corruption index for all countries around the world.Countries which are very "clean" like Scandinavian countries are represented by a small number in the index (in 2011, Denmark and Finland have each a score of 2, Sweden's score is 4), while countries that are highly corrupt like North Korea, Somalia have large number in the index (in 2011 both are at the bottom of the list with a score of 182).
Examining the index, it is worth noting that corruption in Eastern European countries, that have completed accession the European Union, has decreased dramatically, while those countries in Eastern Europe that are not part of the European Union continues to have a high level of corruption.For example, in 2011 Belarus had a score of 143 on the index, Tajikistan' score is 152, Kyrgyzstan's is 164.On the other hand Estonia's score is 29, Slovenia's is 35 and Poland score is 41.The reason for that is the European Union had set rigid requirements for the accessing countries; "country must have stable institutions guaranteeing democracy, the rule of law, human rights, and a functioning market economy"and it is known that countries that enjoy the above characteristics have lower level of corruption.
The objective of this study is to test the hypothesis that there is a strong negative impact of corruption on the GDP per capita.Three tests were conducted, the pool OLS, The Fixed Effect and the Random effect estimations.
The main result of this study is that all three tests had shown strong statistically significant negative impact of corruption on the GDP per capita, which proves the hypothesis.LITERATURE REVIEW Leff (1964) argues that corruption enhances efficiency in the economy."Corrupt bureaucrats can force the government to become involved in promoting economic activities and can make a higher rate of investment possible because investors need assurance of noninterference in their affairs" Kaufman and Wei (1999) examine the relationship between bribe payment, management time wasted with bureaucrats, and cost of capital.Their results contradict the efficient grease' hypothesis.The authors find "that firms that pay more bribes are also likely to spend more, not less, management time with bureaucrats negotiating regulations, and face higher, not lower, cost of capital" Aidt (2009) tackling the topic whether corruption sands or greases the wheels wrote "While corruption in a very narrow sense can be seen as a lubricator that may speed things up and help entrepreneurs getting on with wealth creation in specific instances, in a broader sense, corruption must be considered as an obstacle to development".The author finds a strong negative correlation between growth and corruption.Mauro (1995 and1997) find that corruption lead to a decrease of economic growth by decreasing investment.Kisunko and Kapoor (2000), building on the work of Mauro, use data for Bangladesh during the 1990's investigate the relationships between corruption and growth and between corruption and investment.
The authors' findings suggest "that countries serious about improving governance and reducing corruption should redefine the role of government, overhaul the system of incentives, and strengthen domestic institutions to make sure the necessary checks and balances are in place" Mauro (2004) questions the reasons why corruption persists despite the negative impact on growth.To answer this question, he uses two models involving strategic complementarities and multiple equilibria.His explanation is that "when corruption is widespread, individuals do not have incentives to fight it even if everybody would be better off without it".Igwike, Hussain, and Noman (2012) study the link between corruption and economic development.They employed the annual growth rate of the gross domestic product to measure economic development.The authors find that corruption has a negative impact on economic growth.
Li and Wu (2010) examine the puzzle that some countries experience high economic growth despite rampant corruption.They studied the role of trust in the corruption-efficiency relationship.The authors use a pooled data set of 65 countries.Their results support their hypothesis that trust mitigates the negative effect of corruption on economic growth.Blackburn et al (2009) study the issue of why corruption in some countries has less economic impact than in other.By developing a dynamic general equilibrium model, the authors predict "that countries with organized corruption networks are likely to display lower levels of bribes, higher levels of research activity and higher rates of growth than countries with disorganized corruption arrangements".Rock et al (2004) examine the robustness of the relationship between corruption and economic growth by using four different corruption datasets.The authors find that the impact of corruption on growth depends on the economic development of the country."corruption slows growth and/or reduces investment in most developing countries, particularly small developing countries, but increases growth in the large East Asian newly industrializing economies.The author explains this Asian paradox" in terms of stable and mutually beneficial exchanges of government promotional privileges for bribes and kickbacks".

DATA METHODOLOGY
Three tests were conducted, the pool OLS, The Fixed Effect and the Random effect estimations.
The independent variable in this study is the GDP per capita.The dependent variables are Corruption index, foreign investment and bank loans.Included also dummy variables for the years from 2004 to 2011.The base year is 2003, an interaction term y11crp (year2011*corruption).The coefficient on this dummy variable will measure how the GDP per capita has changed over the eight-year period to an additional unit increase in corruption.
The data consists of 20 countries; 10 with low level of corruption on the index and 10 with high level of corruption in the index.The time period is between 2003 and 2011.
The variables "GDP per capita", "foreign investment" and "bank loans as a percent of GDP" were collected from the World Bank website.The Corruption index data was collected from the transparency international website.

Variables
Where is the unobserved effect.
The main objective of this study is to test the hypothesis that there is strong negative impact of corruption on the GDP per capita.
The hypothesis is: First the model was estimated using Pooled OLS.Below are the results.

Table 2 Pooled OLS results
As we see from table 2 which means that the higher the level of corruption in a country the less is the GDP per capita .The results are very statistically significant.A country's 10 points increase on the corruption index will lead to a decrease in the amount of $2849 of the GDP per capita.
Foreign investment data is in millions of dollars; .That means for every 10 billion increase in foreign investment the GDP per capita will increase by $287.The impact has the right sign but doesn't seem to be too strong.As for bank loans coefficient, the sign is expected to be positive, which is not the case, although the coefficient is not statistically significant.
What is interesting in the results of table 2 is that the coefficients on the dummy variables show a sharp increase over the years.For example the coefficient on y11 implies that, holding foreign investment, bank loans, corruption index, and other factors fixed, the GDP per capita on average worldwide has increased by $15,731 in 2011 from what it was in 2003.This can be explained by the tremendous growth in the BRICS countries and the emerging markets in general.
To further test the significance of the dummy variables and since the dummies are individually quite significant a test was conducted to check whether these dummies as a group are jointly significant.
Below are the results. .reg gdppercapita corruptionindex bankloansasapercentofgdp forinv y11crp Table 3. Test of the joint significance of the dummy variables As we see in Table 3: F(7, 147) = 8.24 and .So the year dummies are jointly statistically significant.
Next the model was estimated using the Fixed Effect and the random effect.This way the pooled OLS can be compared with the other two estimations.Below are the results.

Table 4 Fixed Effect estimation results
In Table 4 the coefficients on the corruption index and on foreign investment show the same sign as with the pooled OLS and they are statistically significant at the 5% level; as for bank loan coefficient, although it shows the expected positive sign, but it still statistically insignificant.
Next the model was estimated using the Random Effect.Below are the results.
Table 5 shows that all the coefficients, including the bank loan coefficient, have the expected signs.Furthermore all the coefficients are statistically significant at the 5 % level.

CONCLUSION
The objective of this study is to test the hypothesis that there is a strong negative impact of corruption on the GDP per capita.Three tests were conducted, the pool OLS, The Fixed Effect and the Random effect estimations.The main result of this study is that all three tests had shown strong statistically significant negative impact of corruption on the GDP per capita.( 5) y08 = 0 ( 4) y07 = 0 ( 3) y06 = 0 ( 2) y05 = 0 ( 1) y04 = 0 . test y04 y05 y06 y07 y08 y09 y10 y11 For the Fixed Effect and the Random Effect models: