Thede, S. and Gustafson, N. (2009), «The multifaceted impact of corruption on international trade», International Trade Working Papers, Lund University. Estimates of the impact of corruption on trade by income level and ACR membership A survey conducted by KPMG in the UK found that while 80% of respondents agreed that the UK`s anti-corruption law was an admirable attempt to tackle the problem of corruption, 58% felt that the law was impractical and ignored reality. that corruption is an accepted way of doing business in many countries. Other similar studies have found that widespread international criticism of the U.S. anti-corruption law is hypocritical in the face of the U.S. business practice of giving gifts to potential customers (p. e.g., trips to conferences, golf excursions, tickets for entertainment and sporting events, use of luxury facilities such as spas, condos and country clubs, etc.). The next step is to econometrically use the nature of the panel data in our samples. As is known, there are two basic methods for this: random effects and fixed effects. Random effects assume that individual effects are not correlated with regressors, while fixed effects assume that they are.

The Hausman test allows us to distinguish between the two in order to select the impartial and efficient estimator. In all three samples, the Hausman test rejects the zero value of equal coefficients, implying that fixed effects are the preferred specification, as is common in commercial gravity models. As explained, the PFEC controls for unobserved bilateral heterogeneity. This does not affect the estimated parameters of the variables of interest to the CPI. In the case of CCI, the parameter of difference in the bilateral level of corruption loses its statistical significance. However, the biggest change occurs with IBS. Now, the estimated parameter for average corruption in the pair is positive and significant at 1%, which means that corruption boosts trade. This indicates that the parameter estimated in column 1 of Table V is biased. As regards the coefficient of differences in levels of corruption, it remains negative and very significant.

Despite these important deterrents, experts report that global corruption in the economy shows little sign of abating. Transparency International (TI), a leading anti-corruption organization based in Berlin, estimates that one in four people worldwide paid bribes in 2009. It seems that the total number of bribes continues to increase every year. The World Economic Forum calculated the cost of corruption in 2011 at more than five percent of global GDP ($2.6 trillion), with more than $1 trillion paid in bribes each year.1 Transparency International`s Corruption Perceptions Index (CPI) ranks countries and territories based on their perceived level of public sector corruption. It is an aggregated indicator that combines different sources of information on corruption and allows for comparison between countries. Perceptions are used because corruption is usually a hidden activity that is difficult to measure. The CPI confirms that corruption remains a global problem and that it takes place even in the richest countries.5 Studies by Austrian economist Friedrich Schneider in 2012 alone estimated the annual loss of the German economy at €250 billion.6 If we focus on the results achieved so far, we see a positive effect of corruption on trade when we use SCI[17]. This surprising result (but not new using other measures of corruption and other dependent variables) is plausible under the «wheel lubrication» hypothesis. However, this assumption makes sense in countries where markets are highly regulated, usually associated with the existence of weak institutions.

Therefore, it is important to analyze this point. In addition, we found a significant negative effect of bilateral differences in corruption in the use of SCs (not for CPI and CCI). This is most likely the result of a much more accurate measurement of the elimination of corrupt activities between countries. [18] We included two pair-specific corruption variables in the regression[4]. The first (corrupijt) is the geometric mean of the values of the respective corruption variables for the exporter and importer[5]. This will give us an average of the level of corruption in which the respective country pair is involved. We use a geometric mean because it is more suited to the characteristics of the available data, especially for the CPI and CCI, since these indices classify countries into levels of corruption. In addition, the geometric mean gives less weight to extreme values in the data. In Table II, we present descriptive statistics for selected measures of corruption used in the regressions. An important aspect of international trade is the proliferation of RTAs over the past two decades.

Then, in the table, we also gave the median (which is quite close to the average) to see if there is a statistical correlation between the level of corruption in the two countries and the fact that they are partners in an RTA.