Estimating demand relation for municipality supplied services and testing the paternalistic distance on the validity of non-users about their preferences for improvements in eldercare services

From an economic perspective, many if not most of the supplied public services from the municipalities to the consumer living in the area can be characterized as non-market goods. The services are thus typically not traded on a regular market, but instead offered in fixed amount to typically a zero or relatively low costs for the consumer.  In the pursuit of improving these services, this is a problem, since the non-market characteristic makes it impossible to derive demand relations for the service per see and for potential changes in the services.

As a tool to solve this problem, a number of economic valuation methods have been developed to elicit the latent preferences for hypothetical changes in the level of the service in focus. Contingent on a stipulated market including a description of the services to be evaluated, to who and how the services will be supplied and how a hypothetical payment should be viewed, the respondents in the economic valuation survey are asked to state their preferences, i.e. the methods are called stated preference (SP) methods.

One of the fundamental assumptions in SP surveys is that the respondents behave rationally and that their stated hypothetical preferences are in line with the preferences they would express had the service been available on a real market under the same conditions as in the SP survey. Though this potentially is a problem in some types of surveys, evidence seems to point towards that market experience potentially reduce or even eliminates some of the anomalies in the stated behavior (List, Ladenburg 2009).

However, in some cases it is not possible to target respondents with experience. The reasons for this can be that the type of good /service it either completely new or not known by the respondents. On the other hand, another issue could be that the group of consumer, which the specific service is aimed at, might not be able to respondent to a SP survey due to for example cognitive restrictions. In that case, how do we identify the demand for changes in services? One way around the problem could be to ask other citizens (non-users of the service) about their preference for the level of supply of the services, which the specific group of consumers benefits from. Such a framework would impose a paternalist preference structure, as the respondents in the survey would express preferences on the behalf of others. However can we expect to obtain rational preferences when the respondents have no or limited experience with the services in question? And if not, does the paternalistic distance influence this propensity to express irrational preferences in the survey.

Based on a study focusing on preferences for eldercare among citizens (non-eldercare-receiving person), the present paper tests for two types of biases in stated preference surveys i.e. Zero Protest (ZP) bias and Infinitive Protest (IP) bias.  Furthermore, putting the stated preferences into a paternalistic altruism framework, the sensitivity to the above mentioned type of biases are modeled using different definitions of the a paternalistic distance between the individual respondent and the group people that receives elder care.