Raising the bar (6)

ABSTRACT

Raising the bar (6). Spatial Economic Analysis. This editorial summarizes and comments on the papers published in issue 12(4) so as to raise the bar in applied spatial economic research and highlight new trends. The first paper addresses the question of whether ‘jobs follow people’ or ‘people follow jobs’. The second paper develops a new methodology to determine functional regions. The third paper is a major contribution to the growing literature on new modelling approaches and applications of disaster impact models. The fourth paper focuses on the costs and benefits of higher education. The fifth paper develops a two-step procedure to identify endogenously spatial regimes in the first step using geographically weighted regression, and to account for spatial dependence in the second step. Finally, the sixth paper estimates a dynamic spatial panel data model to explain house prices and to show that restricted housing supply in the city of Cambridge, UK, has some undesirable labour market effects.

摘要

提高标准（六）。Spatial Economic Analysis. 此一编辑评论概要并评析第十二期第四辑所发表的文章，从而提高应用空间经济研究的标准，并强调崭新的趋势。第一篇文章处理有关“人随工作”或“工作随人”的问题。第二篇文章发展定义功能区域的崭新方法。第三篇文章对于灾害影响模型的崭新模式化方法及应用此类成长中的文献作出重大贡献。第四篇文章聚焦高等教育的成本与效益。第五篇文章发展一种指认内生空间体制的二阶段方法：第一阶段运用地理加权的迴归，并在第二阶段考量空间依赖。最后，第六篇文章评估动态空间面板数据模型，以解释住房价格，并展现英国剑桥城市中的有限住房供给，对于劳动市场产生若干不利的影响。

RÉSUMÉ

Hausser la barre (6). Spatial Economic Analysis. Cet éditorial cherche à résumer et commenter les articles parus dans le numéro 12(4) afin de relever la barre de la recherche économique spatiale appliquée et de souligner les nouvelles tendances. Le premier article aborde la question suivante: est-ce que ‘les emplois tendent à suivre les gens’ ou est-ce que ‘les gens tendent à suivre les emplois’? Le deuxième article développe une méthodologie pour délimiter les régions fonctionnelles. Le troisième article apporte une contribution importante à la documentation de plus an plus abondante au sujet des nouvelles façons de modéliser et à propos de l’application des méthodes d’évaluation de l’impact des catastrophes. Le quatrième article porte sur les coûts et les avantages de l’enseignement supérieur. Le cinquième article élabore une procédure en deux temps, dans un premier temps pour identifier à partir d’une régression pondérée géographiquement des régimes spatiaux endogènes, et dans un deuxième temps afin de tenir compte de la dépendance spatiale. Pour terminer, la sixième article cherche à estimer un modèle dynamique à données de panel spatiales pour expliquer les prix du logement et pour démontrer que l’offre limitée de logements à Cambridge, au R-U, a des effets néfastes sur le marché du travail.

RESUMEN

Levantando la barra (6). Spatial Economic Analysis. Este editorial es un resumen y una observación acerca de los artículos publicados en el número 12(4) con la finalidad de elevar el listón en la investigación económica espacial aplicada y resaltar las nuevas tendencias. En el primer artículo se analiza la cuestión de si ‘los trabajos siguen a las personas’ o ‘las personas siguen a los trabajos’. En el segundo artículo se desarrolla una nueva metodología para determinar las regiones funcionales. El tercer artículo es una importante contribución a la creciente bibliografía sobre los nuevos enfoques de modelos y las aplicaciones de los modelos para el impacto de desastres. El cuarto artículo se centra en los costes y los beneficios de la educación superior. En el quinto artículo se desarrolla un procedimiento bifásico para identificar los regímenes espaciales endógenos en la primera fase mediante una regresión ponderada geográficamente, y explicar la dependencia espacial en la segunda fase. Finalmente, en el sexto artículo se calcula un modelo de datos de panel dinámico y espacial para explicar los precios de la vivienda y mostrar que la oferta limitada de nuevas viviendas en la ciudad de Cambridge (Reino Unido) tiene algunos efectos adversos en el mercado laboral.

KEYWORDS:

• labour market areas
• disasters
• education
• house prices
• spatial econometrics

关键词:

• 劳动市场区域
• 灾害
• 教育
• 住房价格
• 空间计量经济学

MOTS-CLÉS:

• bassins d’emplois
• catastrophes
• éducation
• prix du logement
• économétrie spatiale

PALABRAS CLAVES:

• áreas del mercado laboral
• desastres
• educación
• precios de la vivienda
• econometría espacial

JEL:

• C
• D57
• J61
• O18
• Q54

Spatial Economic Analysis is a pioneering journal dedicated to the development of theory and methods in spatial economic analysis. This issue contains six papers contributing to these theoretical and empirical developments.

In the previous special double issue on spatial econometrics (Fingleton & Pirotte, 2017), we commemorated Raymond Florax who passed away on 1 March this year. The first paper of the present issue is of his hand together with two co-authors from the University of Groningen (Hoogstra, van Dijk, & Florax, 2017). They perform a meta-analysis of 64 studies (mostly published in the main journals in regional science and spatial economics) employing the Carlino–Mills (CM) methodology to address the question of whether ‘jobs follow people’ or ‘people follow jobs’. In addition, the authors try to identify the key explanatory factors of the differences found between these studies. Overall, the work offers a very thorough and extremely informative review and analysis of previous research findings, and points out that although the findings are somewhat mixed, the majority of the results suggests that ‘jobs follow people’ rather than ‘people follow jobs’.

To answer questions such as whether jobs follow people or people follow jobs empirically, geographical data are needed. However, the choice of units is not as simple as it seems, as several studies have pointed out, including several recent papers in this journal (Arbia, Espa, Giuliani, & Dickson, 2017; Bonneu & Thomas-Agnan, 2015; Day, Chen, Ellis, & Roberts, 2016), as well as in other journals (Bhattacharjee, Castro, Maiti, & Marques, 2016; Pryce, 2013). The second contribution to this issue, by Casado-Díaz, Martínez-Bernabéu, and Rowe (2017), is a methodological one aiming to develop a set of local labour market areas (LLMAs) for Chile. To this end, it applies and further develops an evolutionary computation approach taken from Martínez-Bernabeu, Flórez-Revuelta, and Casado-Díaz (2012), termed the grouping evolutionary algorithm (GEA). The GEA defines LLMAs through an optimization process in which their internal cohesion is maximized subject to reaching a certain level of external self-containment in terms of commuting flows. The paper proposes two improvements to the GEA methodology. First, a modification of the interaction index used in the problem’s fitness function; and second, the adoption of a spatially structured population model in the evolutionary algorithm. To make this proposed methodology accessible to new users in the field, the authors have made their code available so interested readers should consult the paper’s supplemental data online.

The third paper in this issue, by Oosterhaven and Többen (2017), is also partly a methodological paper, though on a different area. It determines the impact of disasters measured at the regional level and criticizes many existing models, including the input–output, inoperability input–output and spatial computable equilibrium models for their fixed coefficients and focus on demand-driven shocks. In addition, the supply-driven input–output model is said to have shortcomings. For these reasons, the authors switch to multi-regional supply-use tables. The reviewers of this paper were full of praise, commenting that it is well-written due to its simple and clear structure, a major contribution to the growing literature on new modelling approaches and applications of disaster impact models, has a sound theory, and, finally, an interesting application. In line with this, the second author (note that the paper is part of the second author’s doctorate) received a series of questions about this particular paper during his defence (Többen, 2017). It indicates that it is a challenging piece of work that can potentially evolve into a distinct genre of work, further confirmed by the fact that it has already received a lot of views.

The fourth paper in this issue, by Hermannsson, Lecca, and Swales (2017), combines a computable general equilibrium framework with elements from growth accounting to analyse the impact of the increase in human capital generated by a single cohort from further education colleges (FECs) for (the open economy of) Scotland. The paper outlines a detailed economic–theoretical framework, including associated endogenous investment, employment and competitiveness effects, to arrive at the impact assessments. These assessments are accompanied by a general quantification of costs and benefits associated with this education policy. The authors find that one year’s output from Scottish FECs generates a 0.126% increase in gross domestic product (GDP) over a number of decades, equivalent to a present value of just under ₤2.3 billion. Researchers interested in the modelling and measurement of these impacts should also consult the recent paper by Haapanen and Böckerman (2017).

An extremely popular method in regional science and spatial economics is geographically weighted regression (GWR), or more generally locally weighted regressions, pioneered by the work of Brunsdon, Fotheringham, and Charlton (1996), McMillen (1996), and Fotheringham, Brunsdon, and Charlton (2002). According to Bivand (2017, p. 1):

Geographically weighted regression (GWR) is an exploratory technique mainly intended to indicate where non-stationarity is taking place on the map, that is where locally weighted regression coefficients move away from their global values. Its basis is the concern that the fitted coefficient values of a global model, fitted to all the data, may not represent detailed local variations in the data adequately – in this it follows other local regression implementations.

Spatial Economic Analysis has published several papers applying GWR and related approaches for this reason. Two recent examples are Salinas-Pérez, Rodero-Cosano, García-Alonso, and Salvador-Carulla (2015) and Bhattacharjee, Cai, and Maiti (2017). In addition, the journal receives many submissions making the point that GWR outperforms the ordinary least squares (OLS) approach in that it provides a better model fit, especially in hedonic housing price studies. The fifth paper in this issue, by Billé, Benedetti, and Postiglione (2017), stands out in that it is trying to make a major methodological step forward. It proposes a two-step procedure to identify endogenously spatial regimes in the first step and then to account for spatial dependence in the second step. In the first step GWRs are run, not to observations assigned to predetermined groups but to groups that reveal similar beta-coefficients. This methodology may also be used to investigate club convergence, another important topic in regional science and spatial economics. In the second step, spatial econometric models are run, including the impact of neighbouring units, since we know from several studies that individual observations in the cross-sectional domain may not be treated as being independent. This also holds for hedonic housing price studies (see the next paper in this issue by Szumilo, Laszkiewicz, & Fuerst, 2017). Where several studies stop, having found that GWR outperforms OLS, the study by Billé et al. (2017) continues by showing that spatial dependence also needs to be accounted for when applying GWR. Their code is made available in R.

The above-mentioned paper by Szumilo et al. (2017), the last in this issue and volume, estimates a dynamic spatial panel data model to explain house prices in Cambridge (UK). The authors choose this city since it is characterized by rapid economic development relative to housing supply, and they investigate the consequences of this. Instead of adopting one spatial weight matrix in their spatial econometric model, several are considered and tested against each other based on different numbers of nearest neighbours, following recent work by Ezcurra and Rios (2015); related research published previously in this journal includes Bouayad-Agha and Védrine (2010) and Bhattacharjee, Castro, and Marques (2012). Importantly, the adopted spatial econometric model is underpinned by an economic-theoretical model á la Glaeser (Glaeser, Gyourko, & Saks, 2006). The study concludes that labour shortages may occur due to restricted housing supply, and that companies may be forced to hire and retain only the most productive employees. These employees in turn may see their level of welfare decrease since the wage they receive is not proportional to the increase in living costs in the area.

Turning to other matters, co-editors John McCombie (University of Cambridge) and Gwilym Pryce (University of Sheffield) recently finished their terms on the journal. This summer we welcome Francesco Quatraro (University of Turin) and Justin Doran (University College Cork) as new co-editors. We thank the departing co-editors for all the work they have done for the journal, and we wish the two new ones all the best with their new position.

In addition to this editorial, this issue also contains an editorial introducing a virtual special issue of Spatial Economic Analysis on urban development (Jordan, Monastiriotis, & Elhorst, 2017). This was compiled to mark the keynote lecture at the 47th Annual Conference of the Regional Science Association International – British and Irish Section in Harrogate (UK), given by Professor Bob Stimson of the University of Queensland, Australia. Cities provide significant opportunities for economic growth and development as long as urban design models are not only effective but also sustainable, inclusive and equitable. This virtual special issue draws together 10 articles from earlier volumes of Spatial Economic Analysis, which inform a successful urban design agenda. The papers comprising this virtual special issue are freely downloadable until the end of the year at http://bit.ly/rsea-urban-dev.