The Global Crisis and regional employment in Europe: The performance of sheltered economies

Ugo Fratesi, Andrés Rodríguez-Pose 16 April 2017



The Great Recession has hit Europe harder than any other crisis in the post-WWII period, but its impact, especially in terms of employment, has been very uneven geographically (European Commission 2013 and Figure 1). First, capital regions and large cities were - with few exceptions - able to create more (or destroy fewer) jobs. Second, the evolution of employment during the crisis has reinforced pre-existing disparities. This has been the case in Germany, Italy, Poland, Spain, and the UK where previous north-south (Italy and Spain), west-east (Germany and Poland) and south-north (UK) economic divisions have become more evident. Third, the crisis hit hard many regions which had been lagging behind economically. Employment destruction in most of the periphery of Europe was generally larger than in the core, regardless of whether we consider the core of Europe or the core of each country. Much of the convergence in employment generation experienced in the boom years has been rapidly reverted since the beginning of the crisis.

Figure 1. Regional employment change in European regions relative to the national average (2008-2012)

Source: Authors’ calculations using Eurostat data.

Have sheltered economies played a part in the destruction of employment?

What are the factors behind these employment trends? The economic structure of regions, their degree of specialisation and their resilience (Pike et al 2010, Martin 2012) have been the object of considerable scrutiny. In a recently published paper, we examine the evolution of employment trends in the regions of Europe since the outbreak of the Great Recession in 2008 (Fratesi and Rodriguez-Pose 2016). We assess the extent to which the emergence of ‘protected’ or ‘sheltered’ economies in the years prior to the crisis -- that is, economies that are “more impervious to changes in the business cycle” (Rodríguez-Pose and Fratesi 2007, 624) -- affected the performance of European regions in terms of employment generation once the crisis set in. The hypothesis driving the analysis is that the economic history of a given European region may determine the extent to which it has been capable of adapting and reacting to the conditions generated by crisis. Regions that traditionally were more open to changes in the economic cycle may have reacted differently compared to more protected regions. Protected regions, by definition, may have been structurally less able to take advantage of the high phases of the economic cycle but also generally better able to weather the lower phases of the cycle. These protected regions often, but not necessarily always, coincide with the weak regions within their respective countries.

In order to identify which regions had become sheltered before the outbreak of the crisis, we resort to a methodology developed by Rodríguez-Pose and Fratesi (2007). The method involves detecting regions with a counter-cyclical behaviour. We do this by comparing the average growth rates in periods of high and low growth, relative to the country where they are located and over extended periods of time. We also classify regions according to their economic dynamism before the outbreak of the economic crisis (1995-2007).

Sheltered regions were mainly found in central and eastern European countries such as Bulgaria, Poland, Romania, and Slovakia, although some ‘old’ member states - such as Greece and the UK - exhibit wide internal differences in regional trajectories relative to the economic cycle. Moreover, in most European countries the regional open/sheltered economy gap generally coincided with the regional more/less developed gap. By contrast, open economies coincided with national capitals and core regions. This was the case across Europe where capital cities like London, Paris, Madrid, Amsterdam, Brussels, Copenhagen, Athens, Warsaw, Budapest, Prague or Dublin have been more open than the average region in their country. The same applies for other large cities such as Barcelona, Frankfurt, Munich, Rotterdam, Manchester, Gothenburg, Malmö, Birmingham, Manchester, Leeds, Porto or Marseilles.

How did differences in protection and economic dynamism prior to the downturn affect the employment trajectories of European regions after the crisis set in? Table 1 reports regional employment growth, depending on past protection and dynamism, over the first four years of the crisis (2008-2012). The beginning of this period witnessed the peak of total employment in the European Union.

Table 1. Regional employment growth rates in 2008-2012 depending on protection and dynamism

Source: authors’ calculations using Eurostat data

The table and the different analyses in the paper indicate that the most important divide in terms of employment trajectories was between regions which were dynamic and those which were less dynamic before the crisis. Prior to the outbreak of the crisis, open dynamic regions outperformed open and less dynamic ones by about 1.7%, while protected and dynamic regions outperformed protected and less dynamic ones by about 2.1%. The divide between protected and non-protected regions was weaker and worked in the opposite direction: open dynamic regions have had a better post-crisis employment trajectory than protected dynamic regions by about 0.5%, while open less dynamic were less affected by the crisis than protected less dynamic regions by about 0.9%. Hence, regions which were dynamic before the crisis were also better able to adapt to the changing conditions and have weathered the crisis better. By contrast, those regions which were more protected in the past from the vagaries of the economic cycle have suffered more in recent years.

The development of sheltered economies in the years of economic expansion made more protected regions less capable of fending off the crisis than more open economies. Long periods of protection from market forces affected negatively regions’ adaptability and economic resilience. Protected economies which were able to better withstand the economic troughs of the pre-crisis cycle have been less capable of escaping unscathed from the troughs of a longer and deeper economic crisis. The severity of the crisis has rendered the standard protection mechanisms ineffective. The past, therefore, matters: regions -- sheltered or not -- which were better equipped for generating employment growth have also been able to withstand the Great Recession.

Instead of continuing to shelter lagging regions from the market, regional policy should focus on making them more open, dynamic, and competitive. Protecting regions, often in the periphery of Europe, has made them more impervious to standard downturns, while also hampering growth and employment generation in periods of economic recovery. Sheltered regions have become much more vulnerable to prolonged economic crises and transforming many of the sheltered regions in the periphery of Europe into open and competitive regions will build greater resilience when it comes to employment generation. 


European Commission (2013), The urban and regional dimension of the crisis: Eighth progress report on economic, social and territorial cohesion, Luxembourg: Publications Office of the European Union.

Fratesi, U and A Rodríguez-Pose (2016),  “The crisis and regional employment in Europe: what role for sheltered economies?”, Cambridge Journal of Regions, Economy and Society 9(1): 33-57.

Martin, R (2012) “Regional economic resilience, hysteresis and recessionary shocks”, Journal of Economic Geography, 12(1):1-32.

Pike, A, S Dawley and J Tomaney (2010), “Resilience, adaptation and adaptability”, Cambridge Journal of Regions, Economy and Society, 3(1):59-70.

Rodríguez-Pose, A and U Fratesi  (2007) “Regional business cycles and the emergence of sheltered economies in the southern periphery of Europe”, Growth and Change, 38(4): 621-648.



Associate Professor of Regional Economics, Politecnico di Milano

Professor of Economic Geography, London School of Economics; Research Fellow, CEPR