پیچیدگی بوروکراتیک و اثرات فساد اداری در تاسیسات آب و برق

This paper explores how the relationship between bureaucratic complexity and corruption affects the performance in utilities. We observe considerable variation in the performance of the utilities across countries, even across countries which appear to be relatively similar. Our hypothesis is that corruption plays an important role in explaining this observed difference in performance. In particular, corruption coupled with a complex regulatory structure can have negative effects on performance. The analysis points at the importance of considering the institutional framework and institutional quality when introducing new bureaucratic procedures, as the same set of policy advice will work differently in different countries. We measure bureaucratic complexity by the number of procedures needed for starting a business from the Doing Business Database provided by the World Bank.

The negative impacts of corruption and poor quality of institutions on economic development are well established. Institutional improvement is a slow process, and simply changing formal procedures will not necessarily have the desired effect. Institutional reform often imply that more concerns and political motivations are taken into consideration, and the demand on administrative procedures increases. The introduction of more administrative procedures to ensure consumer welfare may result in greater bureaucratic complexity. Greater complexity, in turn, can provide new opportunities to hide corruption, however. Thus, rather than protecting consumer interests, more procedures may result in higher levels of corruption and inferior sector performance.1 Whether more complexity works to secure consumer welfare or is exploited as a tool for hiding corruption may depend upon the general quality of institutions and governance. The same set of procedures might result in different outcomes across countries depending on levels of income, credibility of legal systems, and human capacity.2 Bureaucratic procedures are particularly relevant for the utilities sectors, which are often natural monopolies, characterized by economies of scale and low marginal costs. There are often few firms, and the incentives and opportunities to benefit from market power are higher than in other sectors. Thus there is a stronger need for regulation through bureaucracy. The variation across countries in how complexity works might therefore be expressed more strongly in utilities compared to many other industries. There is limited empirical information about how the establishment of comprehensive administrative procedures fulfils the goal of higher consumer welfare in terms of improved sector performance. Even if greater complexity increased welfare in higher-income countries with well functioning institutions and a credible legal system, it might raise the level of corruption where institutions are weaker. There are, however, a few studies that address the linkage between various forms of governance and infrastructural performance. Estache et al. (2006) find that regulatory reforms offset the effects of corruption on performance indicators only to a limited extent and that the effectiveness of reforms is reduced when there is a higher level of corruption. Gasmi et al. (2006) find the positive effect of political accountability on the performance of regulation to be stronger in developing countries. This study explores how performance across the utility sectors, such as electricity, water and telecoms, is affected by the connection between the level of corruption and the level of bureaucratic complexity. We apply data from the Doing Business and Enterprise Surveys Database collected by the World Bank. In Section 2 of the paper we first look at whether performance in the electricity, water and telecom sectors in developing countries is systematically affected by the levels of corruption and complexity. We go on to test more thoroughly the relationship between bureaucratic complexity, corruption, and levels of income. We find that in general, service delivery in the utilities functions significantly better in countries with few procedures and low levels of corruption. We also find that the pattern with respect to corruption, complexity and sector performance differs across income levels. In general, complexity (the number of procedures) is a significant determinant of corruption levels, but the effect depends on the income level. Higher confidence in the judiciary also tends to reduce corruption. For developing countries only, the effect of fewer procedures is smaller but still significant. Our findings support the notion that simple procedures, by lessening the problem of corruption, holds potential for improving performance.

For most of the performance measures there are few observations, and it is difficult to apply regression analysis. We have therefore divided our sample into several categories of countries based on the level of corruption, complexity and income in order to assess the effect of complexity and corruption on the sector performance of utilities. We find that for the full sample of 84 developing countries, the level of corruption and complexity coincides as 53 countries fall into categories I and IV, where we have less corruption and less complexity (I) and more corruption and more complexity (IV), respectively. With respect to performance, the general rule seems to be that performance is better when corruption and complexity levels are lower. When we split the countries into two sub-groups based on income levels, this still holds true for the higher-income half of the sample. As many as 30 of the 42 countries fall into categories I and IV, in which the levels of corruption and complexity correspond. Most of the countries in the higher-income half are characterized as being less corrupt (29 of 42). The situation is different when we consider the lower-income half of the sample, however. Then we see that 29 countries are perceived as being more corrupt, and there is no longer a correspondence between corruption and complexity levels. For this group performance is also not generally better when corruption and complexity levels are lower. This lends support to a hypothesis that having a large number of administrative rules may function differently in rich and poor countries. In countries with strong institutions the procedures may function as intended and guarantee a thorough administrative process, hence serving as a barrier against ad hoc solutions and favoritism. In poorer countries with weaker institutions a large number of procedures, established with the same intention, may have the reverse effect and increase the risk of corruption. Further this could suggest that when designing instruments of regulation to tackle corruption, for example in terms of simplifying procedures, these might work differently according to income level. Income levels are highly correlated both to the corruption level, as we have see here, but also to the general level of institutional quality. If fewer and simpler procedures are meant to be a counter-measure against corruption demands are made on institutional quality as well. If fewer procedures result in more transparency, there is still a need for information to be disseminated and for some form of control of violations. A well functioning judiciary and a free press that is willing and able to report abuses of power would be supplements to simplifying procedures. Thus, countries need to reach a certain level of institutional quality to benefit from less complexity.