Inelastic connections: Bureaucratic corruption suggestions and the demand for intermediaries

The literature on intermediary agents in the procurement of bureaucratic services suggests that they enable corruption, as they separate the client from bureaucrats' demands for bribes. This logic is inconsistent with anecdotal findings in developing countries, where agents often publicize their "contacts" with bureaucrats as they market their services through informal networks. We develop an experimental survey to assess the effects of priming the existence of illicit bureaucratic connections on the willingness-to-pay for agency intermediation services. We do this in the context of apostille certification for professional degrees in Venezuela - a country with high rates of emigration of young professionals. We will draw our sample from College students in Universidad Católica Andrés Bello in Caracas, Venezuela. Participants will be placed in a hypothetical "take-it-or-leave-it" (TIOLI) situation in which, after graduation, they need to certify their degrees promptly in order to migrate, and are evaluating whether to hire agency services. In the hypothetical conversation with the agent, participants will receive signals priming the existence of illicit "contacts" within the bureaucracy at random. Similarly, we will randomize the agent's fees, and whether her contact was obtained from a trusted reference. With all this information, the participant will decide whether to procure the service or not. We hypothesize that corruption priming will have a positive effect on the proportion of participants deciding to take the service. Similarly, we hypothesize that corruption priming will make this proportion less sensible to higher agent fees - that is, that the demand of agency services becomes more inelastic to price changes once participants are reaffirmed about the existence of illicit bureaucratic connections. Finally, we hypothesize that the effect of corruption suggestions is contingent to agents contacted through trusted references. We will benchmark our estimates on the effect of corruption priming with those of higher agent experience and lower delivery times. If our hypotheses are confirmed, our findings will contradict the standard view about how intermediaries enable corruption, and we will develop an alternative theoretical framework to understand their role in the delivery of public services in institutionally underdeveloped settings.