Abstract

The literature on inter‑municipal cooperation (IMC) focusing on the characteristics of its members and factors driving a decision to start cooperating is abundant. Various studies indicate that small municipalities are particularly vulnerable to economies of scale and scope, hence they are more likely to start cooperating than bigger units. On the other hand, small municipalities face incentives to free‑ride on bigger local governments due to spillovers. However, it is unclear if there exists a nexus between the size of IMC entities (measured by population) and types of tasks performed jointly by their partners (often of a different number). This paper aims to fill the existing gap by testing whether a share of expenditures on one of the three categories of tasks (‘economies of scale and scope tasks,’ ‘spillover tasks’ and ‘multi tasks’) in total expenditures incurred jointly differs significantly depending on the size of an IMC entity. For that purpose, the Kruskal‑Wallis rank test was used. To pinpoint which specific medians are statistically different from the others in each year of analysis, Dunn’s multiple comparison test with the Bonferroni adjustment was performed. The research is based on Polish inter‑municipal unions (IMC‑unions) and their financial statements over the period 2003–2018 and covers 2,541 observations. The results show that the vast majority of statistically significant differences were observed in the share of expenditures on ‘economies of scale and scope tasks’ in total expenditures, suggesting that very small IMC‑unions spent more of their budget on these tasks than medium‑sized IMC‑unions and small IMC‑unions (over several years). No significant differences were noted in the share of ‘spillover’ expenditures. Significant differences in ‘multi’ expenditures occurred only in 2017 and 2018, indicating that small IMC‑unions spent more than very small IMC‑unions. These first results add to the existing literature by driving a conclusion that smaller IMC‑unions concentrate more on reducing per capita spending.

Highlights

  • In many economic and social studies, researchers are interested in features, attributes and characteristics that are inherently sensitive

  • The method is called mirrored question design. Nowadays it is one of many techniques belonging to a larger group of methods named randomised response techniques (RRTs)

  • Randomised response techniques usually cannot be used in telephone or internet surveys

Read more

Summary

Introduction

In many economic and social studies, researchers are interested in features, attributes and characteristics that are inherently sensitive. The first statistical method of indirect questioning intended for dealing with sensitive features was proposed by Warner (1965). The method is called mirrored question design Nowadays it is one of many techniques belonging to a larger group of methods named randomised response techniques (RRTs). Randomised response techniques usually cannot be used in telephone or internet surveys To avoid this drawback, Yu, Tian, and Tang (2008) introduced crosswise and triangular models that resemble RRTs but do not need any randomising device.

Non‐randomised response techniques
New proposed generalised crosswise model
Properties of the generalised crosswise model
Conclusions
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call