Regional Convergence in the European Union, 1985- 1999: A Spatial Dynamic Panel Analysis

Abstract

[1985][1986][1987][1988][1989][1990][1991][1992][1993][1994][1995][1996][1997][1998][1999]. So far there is no direct estimator available for dynamic panels with strong spatial dependencies. We propose a two-step procedure, which involves first spatial filtering of the variables to remove the spatial correlation, and application of standard GMM estimators for dynamic panels in a second step. Our results show that ignorance of the spatial correlation leads to potentially misleading results. Applying a system GMM estimator on the filtered variables, we obtain a speed of convergence of 6.9 per cent and a capital elasticity of 0.43. JEL No.: C23, O00, R11

Problems with Regional Production Functions and Estimates of Agglomeration Economies: A Caveat Emptor for Regional Scientists

“…There are now numerous studies that have estimated absolute and conditional beta convergence for the regions, with the more recent ones mainly concentrating on conditional beta convergence. See, for example, Barro and Sala-i- Martin (1991Martin ( , 2004, Armstrong (1995), Evans and Karras (1996), Rey and Montouri (1997), Garofalo and Yamarik (2002), Badinger et al (2004), Egger et al (2006), Benos and Karagiannis (2008), Esposti and Bussoletti (2008), and Tselios (2009). See also the survey of Martin and Sunley (1998).…”

Dinâmica espacial e temporal da inovação no estado de São Paulo: uma análise das externalidades de diversificação e especialização

“…Na literatura, Badinger et al (2004) foram os primeiros autores a proporem um procedimento indireto para estimar modelos de painel de dados dinâmicos espaciais. O procedimento é simples e baseia-se em duas etapas, que envolvem primeiro a filtragem espacial das variáveis para remover a correlação espacial e a aplicação tradicional dos estimadores MGM para modelos dinâmicos na segunda etapa (Badinger et al, 2004).…”

“…O procedimento proposto por Badinger et al (2004) adota a filtragem espacial desenvolvida por Getis e Griffith (2002). Para exemplificar a filtragem espacial, considera-se qualquer variável sob estudo, quer dizer y:…”

Estimation Strategies for a Spatial Dynamic Panel using GMM. A New Approach to the Convergence Issue of European Regions

“…The accurate number of years required to avoid short-run variations is still under discussion in the literature (see Temple, 1999, for an analysis). Temple (1999) recommends 5 or 10 years long periods, but we preferred to follow the approach from Badinger et al (2004) k -nearest neighbours' weight matrix has the most fitted to representing spatial interaction of our sample: this specification lead to each region has the same number of neighbouring regions ( k ) including islands on our sample and reduce the heterogeneity problem of regional superficies (Anselin, 2002). Table 1 Descriptive statistics during the whole period.…”

“…Working on a dynamic panel specification, Badinger et al (2004) applied a GMM estimator to spatially filtered variables; Elhorst (2005) suggests a maximum likelihood estimation of models that are dynamic both in space and time for regional analysis; Piras and Arbia (2007) extend panel-data models with spatial error autocorrelation for a convergence analysis of EU regions. More precisely the main argument of applying the extended GMM in a spatial context is that it corrects for the endogeneity of the spatial lagged dependent variable and other potentially endogenous explanatory variables.…”