The production of welfare goods and its financing: an experiment on alternative mechanisms and their effects on motivations

Virginia Cecchini Manara (Università di Milano)
Marco Faillo (Università di Trento)
Kaludjio Klaser (Università di Trento)
Lorenzo Sacconi (Università di Milano)

Abstract

In this work we consider the production of goods and services that have a significant social component, in terms of generating benefits for others – such as health, education, or social services.
Given an endowment of funds, the production of such goods or services usually entails a tradeoff between quantity, in terms of outputs (number of patients treated, of teaching hours or students enrolled in a program, ...), and their quality, in terms of outcomes that produce a benefit in the beneficiaries’ lives. Typically, the provision of high quality is linked with ideal or intrinsic motivations of the agents who provide such goods or services (health workers, teachers, social workers).
Starting from a stylized model of social service provision, where a worker is entrusted with carrying out a choice about a production plan of goods that generate a benefit for a beneficiary, we run an experimental study that compares several financing structures in order to test how the choices of the worker in terms of quantity and quality are affected by the availability of additional funds, the presence of a financier and the different schemes for remunerating the financier.
Our results show that the financing mechanisms have an impact on the intrinsic motivations of agents and generate effects on quality provision, depending on the performance metrics that are linked with the payment structures.
In particular, we show the effect on the workers’ behavior of different schemes for repaying their financier, and we observe how “indirect” incentives that are not directly targeted on workers, but mediated by the mechanism that generates financial returns for the investors, can determine behavior changes in the investee workers. When funding is based on measurable outputs, we observe a shift towards quantity production; the opposite holds for funding based on successful outcomes, suggesting a good reason for linking finance to outcomes evaluation. Nonetheless, in a final treatment we show that whenever outcome measurement is not perfect and depends on proxy measures, such mechanisms may generate dysfunctional behaviors that produces negative impact on the beneficiaries.