Corruption and Middlemen

105 0

By Mukta Kaushik

LUCKNOW : Factors affecting the “Relative Value” of middlemen- a middlemen’s fee as a percentage of bribe,which increases with demand for their services.

Important factors influencing it are:
1. The extent to which corrupt actions are repeated i.e. Frequency.
2. The degree to which actors involved in corruption know each other i.e. Familiarity.

Considering first the “Frequency”- Potential middlemen will collect valuable information about many characteristics of a corrupt act, like the nature of bureaucratic processes, specific individuals willing to be bribed and the amount of money required to attain those ends. For example, if there is a demand for a certain thing and it is also in supply but there is no market where the potential buyer can meet the potential seller, then the cost of searching would increase the transaction cost considerably. Here the role played by the middlemen is important as it functions as a facilitator between the two. Frequent instances of contact of middlemen with prospective parties create opportunities for cultivating relationships with state agents, who are open to taking bribes. Thus, frequency increases the quality of intermediation.

Second characteristic Familiarity of potential participants- for maintaining the relevancy of middlemen contracting parties should be absolute strangers. A middleman is valuable only when one party remains the same and either briber or recipient changes with each deal. Familiarity is more relevant than frequency because even if middlemen may have contacts but if potential parties are in touch there is no relevance of middlemen.

Corruption is a major concern across the country. Cross-national studies estimate the economic cost of bribery at up to $ 2 trillion, or 2% of global domestic product (International Monetary Fund, 2016). But till now the role of middlemen in corruption has not been adequately studied.

What is the role of middlemen in this corruption? A recent work by Dr. Jennifer Bussell tries to analyze this role of middlemen in these corrupt transactions.
For testing these conceptual and theoretical claims data from surveys of politicians and bureaucrats from 3 states of India i.e. Bihar, Jharkhand and Uttar Pradesh was collected. These states were selected because they provide the major reference for theoretical development and also due to the persistence of corruption in a range of state activities (Bussell, 2012; Chandra, 2004; Sukhtankar & Vaishnav, 2015; TII & CMS, 2005; Transparency International, 2008). An analysis of corruption in public service delivery found that more than 60% of Indian households paid a bribe to receive government service, amounting to more than 210 billion rupees each year (TII & CMS, 2005).

Surveys were conducted by Morsel Research and Development Private Limited, with all levels of elected officials from village councilors to Members of Parliament and the three levels of bureaucrats active in subnational governance. Here vignette-based questions were used to eliminate potential social desirability bias in response and the need for symmetric comparisons across corruption types.

Across all vignettes, it was found that respondents allocated rents to middlemen 57% of the time and allocated 60.4% of the bribes suggesting that when they do participate in corrupt acts, they extract substantial portion.

Fig 1: Participation, and rents received,by middlemen both decline in occasional, familiar forms of corruption. The x-axis is the vignette that was described to the respondent. For the gray solid line, the y-axis is the proportion of respondents answering the question who allocated something to the middleman. For the black dotted line, the y-axis is the proportion of the bribe described in the vignette that was allocated to the middleman. Secondly, middlemen extract a larger proportion of rents from frequent, unfamiliar corruption than from occasional familiar corruption. In unfamiliar transactions, middlemen receive 81% and 69% of the rents whereas only 43% and 11% of rents in familiar corruption.


Fig 2: Middlemen participate more and take a greater share of rents in corruption that is frequent and between unfamiliar parties. The graph displays the results of difference-of-proportions tests comparing expectations that middlemen participate in, and receive benefits from, different types of corruption. In each test, I subtract from the outcome for frequent and unfamiliar corruption vignette (Ration). The circle denotes the difference between occasional and familiar corruption (Policy); the triangle, frequent and familiar corruption (Road); and the square, occasional and unfamiliar corruption (Land). Hollow markers show differences in the presence of middlemen, while solid markers show differences in the proportion of rents middlemen extract. Vertical lines give 95% confidence intervals for the differences (Ration minus other vignettes).

Thus anti-corruption measures focus on increasing bureaucratic salaries but the important role of middlemen is completely overlooked. Thus, to fight corruption in areas prone to middlemen influence, policymakers may be advised to devise strategies for legal employment options for these intermediaries.

1. International Monetary Fund. (2016, May). Corruption: Costs and mitigating strategies. Staff
Discussion Note. Notes/Issues/2016/12/31/Corruption-Costs-and-Mitigating-Strategies-43888
2. Bussell, J. (2012). Corruption and reform in India: Public services in the digital age. New York
Cambridge University Press.

3. Sukhtankar, S., & Vaishnav, M. (2015). Corruption in India: Bridging research evidence and  policy options. India Policy Forum 2014–15. Retrieved from policy-options-pub-61186
4. Transparency International India & Centre for Media Studies. (2005). India corruption study. New
Delhi: Transparency International India.
5. Transparency International. (2008). Corruption perceptions index. Retrieved from
6. Bussell, J. (2017). When do middlemen matter? Evidence from variation in corruption in India.

Related Post

Leave a comment

Your email address will not be published. Required fields are marked *