Special Issue Monográfico 2011 'Contributions to spatial econometrics'

This introduction summarizes the main contributions of the papers selected to be published in this special issue. These papers were presented (among others) in the Fourth Seminar Jean Paelinck hold in Oviedo (Spain) in 2010 and their quality justified the edition of this Special Issue. As members of the organizing committee, we are pleased with the results of this Seminar which it is considered as a reference in the spatial econometric field.
This special issue consists of two types of contributions. On the one hand, papers focus on the development of new methodologies linked to the concept of causality in spatial econometrics and, on the other hand, applied contributions where different economic problems are analyzed from a spatial or spatio-temporal perspective.

In Getis and Paelinck (L’Espace Géographique, 2004, No 1) some analytical indices for geographic patterns were proposed; one of them was a Chaitin conditional complexity index, c, based on the observed coordinates. This index was reanalyzed, and showed a large variability as a function of those coordinates. A new index of «peakiness», p, is proposed, tested, and applied to French data relating to «upper» employment in 37 areas of the Rhône-Alpes region (centered around Lyons).

Accounting for spatial structures in econometric studies is becoming an issue of special interest, given the presence of spatial dependence and spatial heterogeneity problems arising in data. Generally, researchers have been employing parametric tests for detecting spatial dependence structures: Moran’s I and LM tests in spatial regressions are the most popular approaches employed in literature.However, this approach remains misleading in the presence of nonlinear spatial structures, inducing important biases in the estimation of the parameters of the model. In this paper we illustrate that issue by applying three non-parametrical proposals when testing for spatial structure in data. Empirical findings for the regions of the European Union show important failures of traditional parametric tests if nonlinearities characterise geo-referenced data. Our results clearly recommend employing new families of tests, beyond parametrical ones, when working in such environments.

The Spatial Durbin model occupies an interesting position in Spatial Econometrics. It is the reduced form of a model with cross-sectional dependence in the errors and it may be used as the nesting equation in a more general approach of model selection. Specifically, in this equation we can obtain the Likelihood Ratio test of Common Factors (LRCOM). This test has good properties if the model is correctly specified, as shown in Mur and Angulo (2006). However, as far as we know, there is no literature in relation to the behaviour of the test under non-ideal conditions, which is the purpose of the paper. Specifically, we study the performance of the test in the case of heteroscedasticity, non-normality, endogeneity, dense weighting matrices and non-linearity. Our results offer a positive view of the Likelihood Ratio test of Common Factors, which appears to be a useful technique in the toolbox of spatial econometrics.

The classical approach to estimate spatial models lays on the choice of a spatial weights matrix that reflects the interactions among locations. The rule used to define this matrix is supposed to be the most similar to the «true» spatial relationships, but for the researcher is difficult to elucidate when the choice of this matrix is right and when is wrong. This key step in the process of estimating spatial models is a somewhat arbitrary choice, as Anselin (2002) pointed out, and it can be seen as one of their main methodological problems. This note proposes not imposing the elements of the spatial matrix but estimating them by cross entropy (CE) econometrics. Since the spatial weight matrices are often row-standardized, each one of their rows can be approached as probability distributions. Entropy Econometrics (EE) techniques are a useful tool for recovering unknown probability distributions and its application allows the estimation of the elements of the spatial weights matrix instead of the imposition by researcher. Hence, the spatial lag matrix is not a matter of choice for researcher but of empirical estimation by CE. We compare classical with CE estimators by means of Monte Carlo simulations in several scenarios on the true spatial effect. The results show that Cross Entropy estimates outperform the classical estimates, especially when the specification of the weights matrix is not similar to the true one. This result points to CE as a helpful technique to reduce the degree of arbitrariness imposed in the estimation of spatial models.

The paper sets up a nesting spatial regression model incorporating heteroskedastic shocks, and discusses hypothesis testing in both nested and nonnested cases in a quasi-likelihood framework, suggesting directions for future research effort.

For spatial data with a sufficiently long time dimension, the concept of «global» cointegration has been recently introduced to the econometrics research agenda. Global cointegration arises when non-stationary time series are cointegrated both within and between spatial units. In this paper, we analyze the role of globally cointegrated variable relationships using German regional data (NUTS1 level) for GDP, trade, and FDI activity during the period 1976-2005. Applying various homogeneous and heterogeneous panel data estimators to a Spatial Panel Error Correction Model (SpECM) for regional output growth allows us to analyze the short- and long-run impacts of internationalization activities. For the long-run cointegration equation, the empirical results support the hypothesis of export- and FDI-led growth. We also show that for export and outward FDI activity positive cross-regional effects are at work. Likewise, in the short-run SpECM specification, direct and indirect spatial externalities are found to be present.

The aim of this paper is to analyze the effect of the accessibility to employment offices on local unemployment rates according to the distribution of three different types of municipalities: large urban, small urban and non-urban. We built a new accessibility measure taking into account the number of employment offices together with the distance and size of their catchment area. We propose an empirical model with spatial regimes that allows including simultaneously spatial heterogeneity and spatial autocorrelation. The results suggest that the accessibility to employment offices is especially important in non-urban areas where employment opportunities are limited. Employment services are important because bridge the gap between unemployed workers and employers where job opportunities are unclear.

Cross-border regional development is one of the EU current major concerns. These regions are usually less dynamic socio-economically. Some of them have recently benefited from new roads, which have mainly been funded through the European financial program of Transnational Transport Networks, TEN-T. Using socioeconomic data from the Portugal/Spain cross-border area a model able to measure the relation between accessibility and development in this region is being calibrated. This paper reflects an initial study using Portuguese and Spanish geographical units in the border area for the period 1991-2001 and giving special efforts to the building of similar spatial units in both countries.

Road traffic noise is one of the main concerns of large cities. Most of them have classified their territory in acoustic areas and have constructed strategic noise maps. From both sources we have elaborated seven types of acoustic neighbourhoods according to both their noise gap in regard to the legal standard and the percentage of population exposed to noise. A spatial Durbin model has been selected as the strategy that best models the impact of noise on housing prices. However, results for Madrid do not confirm the hedonic theory and indicate, as one of the possibilities, that the official acoustic areas in Madrid could be incorrectly designed.

This paper analyses the causes of deforestation for a representative set of Bolivian municipalities. The literature on environmental economics insists on the importance of physical and social factors. We focus on the last group of variables. Our objective is to identify causal mechanisms between these factors of risk and the problem of deforestation. To this end, we present a testing strategy for spatial causality, based on a sequence of Lagrange Multipliers. The results that we obtain for the Bolivian case confirm only partially the traditional view of the problem of deforestation. Indeed, we only find unequivocal signs of causality in relation to the structure of property rights. 

Recently, a significant share of the empirical analysis on the impact of public capital on regional growth has used multivariate time-series frameworks based on vector autoregressive (VAR) models. Nevertheless, not as much attention has been dedicated to the analysis of the long-run determinants of regional growth processes using multi-region panel data and applying panel integration and co-integration techniques. This paper estimates the dynamic domestic effects of public infrastructure using a structural vector autoregressive (S-VAR) methodology for the Spanish regions. From a methodological point of view, the paper contains several features that can be viewed as a contribution to the existing empirical literature. First, the important issues of the stationarity of the data and the existence and estimation of cointegrating relationships in the long-run are addressed in the context of the analysis of panel data. Secondly, the long-run cointegrating production function is embedded within structural vector error correction (S-VEC) shortrun models to produce consistent estimates of impulse responses, contrary to many researchers who have estimated unrestricted VAR models in levels or VAR models in first differences. The estimates reveal new results with respect to the previous empirical evidence. 

This study revisits the utility of gravity models in the analysis of the principal determinants of exports. Traditional cross-sectional models are improved by considering the effect of omitted variables and/or the dynamic of trade flows through the use of spatial econometric techniques and panel data specification. This proposal is applied to the Tunisian olive oil exports during the period 2001-2009. The results provide evidence of the inertia found in export volumes, with trade relations anchored in the past likely to continue in the future. Also, we obtain evidence on the existence of a clear similarity in flows between neighbouring importing countries. On the other hand, the results show a positive, significant relationship between the importing country’s income level and imported olive oil volume. The effect of importers’ human development index is positive. The distance between countries has a negative impact on trade flow. On the contrary, sharing a common language increases olive oil trade flows. Finally, trade figures and results reflect a strong dependence of Tunisian olive oil production on precipitations