1.1 Interpreting Convergence in Europe: A Public Policy Approach
The issue of the European regional disparities was a discussion topic already among the debates of the founding fathers of the European Economic Community (EEC ). In fact, considering the inequalities present in its six Member States (MSs ), that is, Belgium, France, West Germany, Italy, Luxemburg, and Netherlands, they strongly supported the strengthening of the social and economic cohesion as a necessary condition for the establishment of the common market. For these reasons, the Treaty of Rome (1957) set itself the goal to reduce the ādisparities between the levels of development of the various regions and the backwardness of the least favoured regions or islands, including rural areasā (Art. 158), in order āto promote [ā¦] a harmonious development of economic activities, a continuous and balanced expansion, an increase in stability, an accelerated raising of the standard of living and closer relations between the States belonging to itā (Art. 2).
As several scholars pointed out (Armstrong and Taylor 2000; Boldrin and Canova 2001; Leonardi 1995; Molle et al. 1980; DziembaÅa 2016; Crescenzi and Giua 2020; Williams and Varghese 2018), convergence in Europe should be realized by reducing socio-economic disparities among all regions. This goal should not be achieved by reducing the level of the most developed areas, that is, by downward convergence. On the contrary, by upward convergence the goal should be to increase economic performance and social well-being of the less developed areas faster than the most developed ones.
Moreover, convergenceās measurement had historically represented a crucial issue of discussion and debate, since its estimation requires the use of indicators with a longitudinal and a territorial dimension (Arbolino et al. 2018; Ferrara et al. 2017). For a long time, gross domestic product (GDP ) per capita was used from the official reports of the European Community (EC ) (CEC 1994, 1996; Sapir et al. 2004) and academic studies (Boldrin and Canova 2001; Mohl and Hagen 2010; Paci 1997; Sala-i-Martin 1996) in order to assess economic cohesion. In fact, as compared other indicators, such as the gross national product (GNP) and income inequalities, it has historically been available at the Nomenclature of Units for Territorial Statistics (NUTS ) Level II. Nevertheless, the use of GDP per head limits a comparability of the local economies across time, since its calculation is to a large extent conditioned by variations in the rate of exchange of the national currencies. For this reason, in 1986, the Single European Act (SEA) officially provided to also use Purchasing Power Standards (PPS) in order to estimate economic convergence. In fact, as compared to GDP per head, it is able to guarantee a more objective evaluation of the development of a specific NUTS II area, since it āadjusts for differences between countries in purchasing power due to differences in price levelsā (CEC 2014, p. 1).
In the last 50 years, several scholars have tried to understand whether economic convergence is taking place in Europe, as well as its intensity around the single regions (Parr 2015; Crescenzi et al. 2017). Using advanced econometrics, these studies arrived at different results. In fact, while some of them indicated that convergence is taking place (Armstrong and Taylor 2000 ; Cappelen et al. 2003; Leonardi 1995, 2005), other showed a stationary situation (Rodriguez-Pose 1998), or an absence of this process (Boldrin and Canova 2003; Neven and Gouymte 1995). These different results are to a large extent conditioned by the complexity in measuring convergence, which is in turn connected with both technical and political-historic issues.
Technical issues are first of all conditioned by the presence of different definitions of the concept of convergence that in several cases corresponds to different convergence concepts themselves. Second, there is no measure able to capture all the dominant aspects of the same process of convergence, since many types of variables contribute to its determination and each of them could have a different weight. Finally, the impact of both specific public interventions and more structured development policies on growth do not necessary emerge immediately. On the contrary, it may take time, therefore limiting the possibility to clearly compare the result achieved, across time and space.
Political-historic issues are above all connected with two events that more or less directly conditioned convergence measurement, that is, the different enlargements had from the Treaty of Rome to the present and the ongoing economic crisis. Either altered possible comparisons across the decades. In the former case, it was to a large extent conditioned by the inclusion in the calculation of new regions with a GDP much lower than the old ones, such as the Eastern areas. On the contrary, since 2008 the tragic impact of the global economic crisis in regional GDP growth decline is arresting the long-run convergence process (Berkowitz et al. 2015; Pontarollo 2015).
The latter reflection is inevitably connected with the instruments historically used to estimate convergence, that is, beta and sigma convergence. Both were introduced, in 1991, by Barro and Sala-i-Martin. While āthe first [type of convergence] relates to poor economies growing faster than rich ones,ā the second one āinvolves a decline over time in the cross-sectional dispersion of per capita income or productā (ibidem, 112). Considering these two indicators, several studies conducted before the start of the actual economic crisis pointed out that only beta convergence is taking place among EU regions, at both EU-15 and EU-27 (Cuadrado-Roura 2001; Leonardi 2005; LĆ³pez-Bazo 2003), as well as āthe speed of convergence is not constant in time, with low values being generally found during the eighties and higher values being detected for periods before and after that decadeā (DG Regio 2008, p. 5). Moreover, taking into account the poorer six macro regions in Europe, that is, Greece, Spain, Ireland, Portugal, the six eastern German LƤnder, and the Mezzogiorno in Italy, the Sapir Report (Sapir et al. 2004, p. 59) arrived at the conclusions that beta convergence was taking place only between MSs but not necessarily within the single countries.
More recently, the Commission of the European Communities (CEC) quantified the impact of economic crisis in regional economic growth, comparing the level of convergence in Europe before 2008 and in the period 2008ā2011. Its āanalysis of changes in GDP per head between 2000 and 2011 confirm[ed] that, in the long run, convergence is mostly a result of the least developed regions catching up rather than growth declining in the more developed onesā (CEC 2014, p. 6, emphasis added). For example, the number of NUTS II regions with a GDP per head below 50% of the EU average shifted from 37 in 2000 to 20 in 2011. Moreover, 16 regions increased, in the same period, their GDP per head to between 50% and 75% of the EU average. In addition, considering both the coefficient of variation of regional GDP per head and the real growth rates of GDP per head, it pointed out that the crisis is arresting the convergence tendency in Europe. Nevertheless, it also provided a positive expectation for the future, supporting the idea that
there are grounds for believing that the long-run convergence process in the EU will continue after the crisis comes to an end. Since the process is driven in part by less developed regions adopting technology and methods of working developed and tested in other regions, it means that they tend to catch up in terms of productivity. This process [ā¦] is likely to see growth in less developed regions return to a higher rate than in the more developed parts of the EU in the years to come, just as over the period 2003ā2008. [ā¦] GDP per head grew faster in real terms in the less developed Member States over the period 2000ā2013 and is forecast to continue to do so in 2014 and 2015 [ā¦]. The rate of growth in the moderately developed Member States, however, fell below that in the highly developed Member States in 2010 and continued to be lower in 2011ā2013 but is forecast to be slightly higher by 2015. (ibidem, pp. 6ā7)
Assuming that convergence is taking place in the two previous mentioned ways, that is, as a result of the growth of the less developed regions and happening above all between MSs and not necessary within them, it is more and more interesting to understand not only the āif,ā the āhow,ā and the āwhereā of this phenomena, but also the āwhy,ā that is, which factors can determine and/or can facilitate the achievement of different economic performance across the old and the new MSs and their respective regions. In fact, since the first years of the EEC, upward convergence should be achieved through specific domestic and/or European policy interventions.
Therefore, the convergence issue has long been at the center of the international academic debate, involving scholars from different disciplines and different lines of thought. At the same time all studies have pointed out that the task of objectively interpreting convergence is arduous since several economic, social, political, and institutional variables could intervene, inevitably limiting the possibility of monitoring all the causal mechanisms. For this reason, limiting the field of action, several scholars analyzed the way in which single variables (or a set of them) can have a more or less direct influence on economic performance and convergence in specific geographical areas.
This book fits into this debate proposing a novel interpretation of policy analysis. In fact, it is believed that its characteristic interdisciplinarity allows for the joint use of variables with a different nature, that is, political, institutional, and administrative, and, consequently, to be able to frame the same problem from different points of view. Moreover, one of its main advantages concerns the plurality of methods shared by scholars across the world and that every single researcher can decide to use, adapting them at his specific research question(s), as well as at the collective problem(s) he/she is trying to analyze. As previously discussed, this is particularly valid for a complex collective problem such as convergence, since it simultaneously involves problems of various types. In fact, even if the public policy analysis approach can vary depending on the methods used (more or less indu...