VoxEU Column Global governance Politics and economics

Is it worth it: Does corruption influence firm productivity?

Does it pay to be corrupt? This column presents evidence from 22 emerging economies in Europe and the former Soviet Union on the effects of corruption on firm productivity. It finds that in a highly corrupt country, bribing officials actually has a negative effect on productivity, whereas in countries with strong institutions, it can open doors that competitors dare not touch.

In April, two former Siemens executives received suspended sentences for their part in the biggest bribery scandal in German history. Between 2000 and 2006, €1.3 billion was dished out to foreign officials in exchange for the German conglomerate securing lucrative contracts. In case we needed reminding, corruption is not just isolated to backward countries and banana republics – it happens everywhere.

There are several studies dealing with the causes and effects of corruption. Yet most of these studies focus on the questions at the country level such as why countries have higher rates of corruption than others and what this means for aggregate growth (e.g. Svensson 2005).

In comparison, the micro-level mechanisms regarding corruption have received far less attention. What are the incentives for firms like Siemens to pay bribe money – and, perhaps even more important, what are the effects on the firm? In principle a firm wants to enhance its productivity, but does corruption pay off?

Theory offers several explanations of how productivity is influenced by bribe money. One can say that corruption raises opportunity costs, since money could have been spent more usefully, hence corruption lowers productivity. Otherwise, it is also possible that grease money – as the name suggests – could accelerate the tedious procedures of bureaucracy (Kaufmann and Wei 1999, Fishman and Svensson 2007).

Identifying the effects of firm-level bribery

In a recent paper (De Rosa et al. 2010), we analyse the connection between corruption and productivity by using comparable firm-level data for 22 emerging countries in central and eastern Europe as well as former Soviet states (De Rosa 2010). The data come from the recent 2008 wave of the World Bank / EBRD Business Environment and Enterprise Performance Survey (BEEPS). Our analysis compares corruption proper (bribery) with the effects of red tape per se, as reflected by the time that managers are forced to spend dealing with bureaucracy.

Identifying the effect of corruption on firm productivity is not a trivial task, particularly in the cross-sectional data used. The causality can go both ways. On the one hand, bribery may affect productivity in the way theory suggests. On the other, it may be the case that firm productivity (or some variables correlated with it) affects the firm’s decision to pay grease money in the first place. It may, for example, be the case that less productive firms rely heavily on bribes as a means for them to improve their competitiveness and gain access to markets. If this is not properly controlled for in the econometric analysis, estimation results and, therefore, policy conclusions would be heavily biased.

In order to deal possible reverse causality, we use instrumental variables estimation. From the wealth of data available in the BEEPS on firms’ perception of investment barriers and institutional constraints, we select a number of variables that may be good instrumental variables candidates.

We regress productivity at the firm level on a number of firm characteristics, including information on whether a firm pays bribes (labelled “bribe tax”) and the percentage of time managers spend with bureaucrats and politicians (“time tax”).1 We find that grease money has a noticeable and negative effect on firm productivity, while there is no measurable link between the time managers spend with public officials and productivity.

One extension is to investigate whether bribes are more effective if managers spend more time with bureaucrats. Interacting the bribe tax and time tax variables does not yield results supporting this hypothesis.

Given the multi-country dimension of the data, it is also possible to take country-specific perspectives by using additional international data. Therefore an enlarged model investigates whether the effect of bribe money on the productivity of individual firms correlates with the overall prevalence of corruption in a country – i.e. with peer incentives to behave corruptly – and with the ability of the legal system to deter and sanction corrupt behaviour.

For this purpose we use data from Transparency International to classify the grade of corruption and data from the Global Competitiveness Report to determine the quality of the legal system. These two indicators are depicted in Figures 1 and 2, ranked by the respective index. In both cases, high values of the index mean “better” country profiles, i.e., lower incidence of corruption or better quality of the legal system.

Figure 1. Corruption index for sample countries, 2008

Source: Own calculation based on Transparency International data.

Figure 2. Qualify of Legal Framework index, 2008

Source: Own calculation based on Global Competitiveness Report data.

We use this country-level information in order to allow the effect of firm-level bribe taxes to depend on the institutional characteristics of the country in which the firm operates. For example, for the case of the Transparency International corruption perception index, the empirical model interacts the firm-level bribe tax with the country-level variable on the level of corruption. The estimation results show that, in countries that are classed as exceptionally corrupt by Transparency International, bribe money has a remarkably negative effect on productivity.

Profiting from low corruption

A possible explanation is that, in countries where corruption is common practice, firms cannot get any additional benefit from also paying bribes. In fact, the more a firm is willing to pay in terms of bribes, the less it can spend in a more efficient way. This induces negative effects on productivity. We find a similar effect in the interaction of bribe tax and quality of the legal system. Firms that operate in countries that have a legal system with a low quality experience strong negative effects of bribes on productivity.

In countries that are classified as not corrupt or as having an adequate legal system, the story is the reverse: firms can benefit from paying grease money. An explanation could be that firms operating in countries with low corruption rates can profit because bribing is an unusual practice that helps them gain an advantage over their competitors.

Of course, the list of countries with a high corruption rate does not completely match with the countries that have an inadequate legal system, as shown in the diagrams above. According to these indicators, there are, however, some countries classed as highly corrupt and with a low quality legal system, where firms experience negative productivity effects. Similarly, there are countries that have both a good legal system and a low corruption rate yielding positive productivity effects for firms. These countries are listed in Table 1 below.

The Czech Republic and Hungary, for example, are not very corrupt and have high quality legal systems. Our findings would suggest that corruption – on average – can lead to productivity gains for the paying firms in those countries. The opposite applies to Russia, Armenia, and the Kyrgyz Republic. In these countries bribe money results in negative productivity effects for the firms involved.

Table 1. Countries with positive and negative productivity effects

Low corruption and good legal system; Positive productivity effect
High corruption and low quality legal system; Negative productivity effect
Macedonia, FYR
Russia
Estonia
Armenia
Czech Republic
Kyrgyz Republic
Hungary
 
Latvia
 
Lithuania
 
Slovakia
 
Slovenia
 
Montenegro
 

Source: based on own calculations, see De Rosa et. al. (2010)

Cautious conclusions

So does this analysis draw the conclusion that corruption is profitable for some countries? No! Definitely not. Our analysis only focuses on the short-term microeconomic scales for firms that paid grease money. Long-term effects and general equilibrium effects are not assessed. So, for example, it is possible that corrupt practices prevent resources and skills to be used in an optimal macroeconomic sense. Thus the macroeconomic entrepreneurial activity of a country suffers. The macroeconomic costs of corruption can thus be substantial, as shown in Dreher and Schneider (2010), for example. The importance of country effects for microeconomic performance also suggests that piecemeal anti-corruption measures may be ineffective in the absence of a comprehensive approach to reforming state institutions.

Note: This column reflects the views of the authors writing in their personal capacity.

References

De Rosa, Donato, Nishaal Gooroochurn, and Holger Görg (2010), “Corruption and productivity: Firm-level evidence from the BEEPS survey”, World Bank Policy Research Paper 5348.

Dreher, A and F Schneider (2010), “Corruption and the Shadow Economy: An Empirical Analysis”, Public Choice, 144:215-238.

Fisman, Raymond and Jakob Svensson (2007), “Are corruption and taxation really harmful to growth? Firm level evidence”, Journal of Development Economics, 83 (1): 63-75.

Kaufmann, Daniel and Shang-Jin Wei (1999), “Does grease money speed up the wheels of commerce?”, NBER Working Paper No. 7093.

Svensonn, Jakob (2005), “Eight Questions about Corruption”, Journal of Economic Perspectives, 19, 3:3-28


1 To be more precise, the BEEPS does not ask firms whether they pay bribes themselves, as this is unlikely to generate honest responses. Rather, the question is framed around the firm’s perception of whether bribes are paid “by a typical firm in your area of business”.

1,470 Reads