Provisión de servicios públicos locales en México: La influencia de municipios vecinos y concentración de la población

Resumen:En este trabajo se analiza empíricamente el efecto de la densidad de la población sobre la provisión de bienes públicos locales en México. Asimismo, medimos espacialmente la influencia de los municipios vecinos. Para ello, desarrollamos un modelo Durbin de rezago espacial, tomando en consideración las características socioeconómicas, demográficas y políticas. Encontramos una fuerte correlación espacial en la provisión de servicios públicos en México, aún más, encontramos una relación positiva entre la concentración de la población y la provisión de servicios públicos que sugiere la presencia de economías de escala.Abstract:The supply of public goods and services is one of the primary obligations of local governments, since it constitutes what the ordinary citizen can expect from their offices, independently of any party ideology (Zhang, et al., 2004; Fernández, 2002). According to Blonch and Zenginobuz (2007), since the 1960s, efforts have been considerably broadened to design norms and institutions in order to obtain information about agents' preferences for public goods, or to encourage private contributions to public goods. Tiebout (1956) explains that local public goods are associated to the mobility of people, by producing a model in which agents reveal their preferences for public goods by choosing the jurisdiction whose fiscal structure (taxes and expenses) suits your tastes better. From this perspective, elections constitute baskets that combine taxes to pay and public goods to be consumed; the result of this process is an efficient balance. In this way, a key factor that could have some influence on the capacity of local governments to provide basic services to residents is demographic and spatial. For several decades, it has been studied whether city size and urban sprawl are related to the efficiency in the provision of public goods. Some authors, such as Buettner and Holm-Hadulla (2013), have found that large cities are more efficient in providing public goods by presenting economies of scale and reducing per-capita costs. In fact, the geographical distribution of people in the municipal territory makes it more expensive to supply water, sanitation and drainage in isolated areas than in those countries with more concentrated. On the other hand, an element to face the inefficient provision of public goods has been the implementation of consolidation programs. According to Di Porto, Parenti, Paty and Abidi (2017), in the last decades, many countries of the world have introduced inter-municipal cooperation plans. These authors argue that the decision of cooperation of the municipalities depends largely on what the neighbors decide. If no municipality has a close cooperative relationship with its contiguous neighbors, the likelihood of joining an intermunicipal community will be less than if close neighbors already cooperate. This research aims to empirically analyze the effect of the escalation economies and spatially measure the influence of neighboring municipalities on the coverage of local public services in Mexico. In particular, we evaluate the effect of population concentration on the provision of local public goods. We argue that the provision of local public goods has spatial effects not considered in the existing literature. The existence of positive externalities with the interaction of economies of scale provides arguments in favor of inter-municipal cooperation in Mexico. Using cross-section information from 2,456 municipalities, We estimated a Durbin spatial lag model that allows us to measure, at the same time, the spatial dependence of the provision of public services and the influence of the characteristics of neighboring municipalities, i.e. the effect of the explanatory variables of the municipality i on the provision of the municipality j. We establish a strong spatial correlation in the provision of public services in Mexico, even more; the formation of groups of municipalities with high and low coverage of public services is detected. We found a positive relationship between the concentration of the population and the provision of public services, which suggests the presence of climbing results. To examine whether the supply of public services has spatial dependence, the Moran spatial correlation test applies to the translogarithmic transformation of the public goods index (dependent variable). The null hypothesis is that the correlation is zero, which entails that the indicator that is being analyzed is randomly distributed in the study region; against the alternative hypothesis of non-zero spatial correlation. The results of this test indicate that the correlation coefficient of Moran is 0.7011, which indicates that the global dependence is positive, and according to which the p-value is less than 0.05, it is accepted that the coefficient of correlation is statistically different from zero. On the other hand, the results show that the use of the Durbin model is appropriate, because it provides more information to the classical spatial model, the parameters of the explanatory variables are statistically significant: one of the benefits of this model is the measurement of the influence of the characteristics of neighboring municipalities. In particular, results show that the economic, political and socio-demographic conditions of neighboring municipalities have a strong effect on the provision of public services. In relation to direct effects, one of the most important results that corroborates the importance of economies of scale is the positive effect of population density on the coverage of public services. In addition, the results presented are consistent with some works discussed in advance and with international studies. This evidence supports the hypothesis of economies of scale, which suggests that the concentration of the population generates a decrease in per capita costs in the provision of public services. On the other hand, the results also suggest the existence of negative externalities derived from the concentration of the population in the neighboring municipalities, an increase in the concentration of the population in 1% of the municipality i, reduces the provision of public services by 0.072% in the municipality j. Thus, the total impact is the cumulative effect of the direct and indirect effect. A 1% increase in population density will increase approximately 0.0678% the public provision with a statistical significance of 1%. In this way, the different specifications show the density of the population has a direct positive impact on the coverage of public services, while the indirect impacts are negative. Since the direct impact is greater than the indirect one, the total impact is also positive in most specifications. Particularly, we find evidence that supports the works developed by Carruthers and Ulfarsson (2003), Nakamura and Tahira (2008) and Hortas and Solé-Ollé (2010), who point out that the per capita cost is lower in compact communities than in those areas with greater geographic dispersion. We provide evidence of the presence of economies of scale in the provision of public services, which together with the formation of clusters, generate arguments in favor of the compaction of municipalities and the creation of intermunicipal bodies responsible for the provision of public services.