(De)centralised Pricing for Public Services: Evidence from Tanzania

How to cite

Blum, Florian. 2016. "(De)centralised Pricing for Public Services: Evidence from Tanzania." AEA RCT Registry. AEARCTR-0000979. https://doi.org/10.1257/rct.979-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Florian Blum*

* Primary investigator

Registry ID
AEARCTR-0000979
DOI
https://doi.org/10.1257/rct.979-1.0
First registered
Last updated
Status
On going
Trial dates
2015-12-08 – 2016-03-31
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/979

Abstract

Public service delivery programs in developing countries typically perform poorly. A common feature of such programs is that implementation is delegated to frontline agents who operate in environments that are difficult to monitor. Existing research that investigates this information asymmetry primarily focuses on effort-related moral hazard to explain inefficiencies in service delivery: In the absence of performance incentives and when effort is unobservable, agents will exert lower effort, and hence serve fewer recipients, than is socially optimal.
This project investigates an alternative source of moral hazard relating to rent-seeking behaviour. As existing research has shown that rent-seekers use sophisticated pricing strategies when deciding about side-payments (Olken & Barron, 2009), it is possible that agents knowingly withhold quantities in order to extract rents from recipients with a high willingness to pay. In the terminology of Bandiera et. al. (2009), such behaviour leads to active waste in public service provision and reducing it requires increased administrative control in the form of price regulation.
On the other hand, increasing control can also reduce the outreach of public service delivery, especially when agents aren’t using discretion for rent seeking. For example, agents typically hold superior information about local demand and cost conditions, which allows them to use discretion in order to distribute co-payments more efficiently among program recipients.

Given these countervailing forces, understanding the objectives and pricing mechanisms used by frontline public service providers, and evaluating the effectiveness of theory-based regulation interventions, becomes an empirical question. This research project will address those questions in the context of public delivery of chicken vaccinations to smallholder farmers in Tanzania. The Tanzanian government maintains a vaccination program that delivers a novel Newcastle Disease vaccine for chicken to smallholder farmers that don’t have access to private markets. This setting is ideally suited to explore the aforementioned questions, as during such campaigns frontline staff is tasked with collecting co-payments from farmers to cover last-mile delivery costs. Currently there is no formal service charge schedule that regulates the size of those payments, giving rise to a status quo that allocates full discretion to the frontline agent. During the upcoming vaccination campaign in early 2016, agents will be randomly exposed to price regulation in order to assess the effect of increased control on service provision levels.

Experimental design

Service providers will be subject to two orthogonally assigned treatments:

Treatment 1: Price-Cap Regulation - Service providers will not be allowed to charge more than 80 TSh per chicken served
Treatment 2: Fixed Cost - Service providers are responsible to pay a fixed cost of 25'000 TSh to participate in the campaign. To avoid challenges relating to liquidity constraints this charge will be collected post-intervention.

Design details

Randomization method
Randomisation done in office by a computer
Randomization unit
To avoid spillovers, the unit of randomisation was chosen at the ward level, which is the level at which livestock officers are organised. There are, on average, 1.8 livestock officers per ward.
Clusters
400 wards
Observations
720 livestock officers
Arms
180 control, 180 price cap only, 180 fixed cost only, 180 fixed cost and price cap
Primary outcomes
- Number of farmers served - Chickens vaccinated - Prices charged by service providers - Immunity data and disease outbreak data in chicken (if the budget post intervention allows to collect such data) - Types of farmers served

Cited by

No citation data available for this registration yet.

Keywords