VoxEU Column Labour Markets Migration Productivity and Innovation

Job-related mobility and plant performance in Sweden

While job-related mobility is key to knowledge sharing, it may also undermine on-the-job training through labour poaching, and assessing its overall impact on productivity and growth is not straightforward. This column uses data on nearly 2.7 million new hires in Sweden to analyse the impact of labour mobility on plant performance. The greatest positive impact is seen in the country’s three largest cities, while firms in other large urban and university regions emerge as the biggest losers from job mobility.

The ability to attract and retain human capital is a fundamental issue in regional policy, especially in less developed regions. This is because human capital has long been considered as a main – if not the main – driver of regional development. Human capital facilitates knowledge spillovers and localised learning that can be transformed into productivity and growth by firms (e.g. Malmberg and Power 2005). Job-related migration is key in this process (Fratesi 2014), as it is the mechanism through which embodied knowledge circulates both within and between regions (Eriksson and Lindgren 2009, Saxenian and Sabel 2008).

Does job-related mobility lead to improvements in plant performance everywhere?

Despite the agreement on the relevance of job mobility, the exact importance of mobility for productivity and growth remains controversial, as mobility may also undermine on-the-job training through labour poaching (McCann and Simonen 2005, Combes and Duranton 2006). Given these potentially contradictory effects, empirically analysing the impact of labour mobility on plant performance is of capital importance.

We address this issue in a recent study (Eriksson and Rodríguez-Pose 2017). We make an explicit distinction between types of job-related mobility according to work-experience and distinguish between inflows of graduates (inexperienced) and experienced workers. Our aim is to assess the extent to which hiring recent graduates matters in relation to acquiring experience by poaching workers from other firms. We also pay special attention to the difference between low- and high-skilled labour flows across different types of regions. This represents a new dimension, as previous studies (e.g. Boschma et al. 2009) have primarily focused on the mobility of the high-skilled to the detriment of the mobility of workers with lower levels of qualification (Maskell et al. 1998, Eriksson and Lindgren 2009). Moreover, since both the supply and demand of labour differ over the urban hierarchy – with large and diverse regions often considered capable of absorbing greater numbers of migrants (Partridge and Rickman 2003) and having the greatest potential for effective matching (Puga 2010) – we examine the geography of mobility. The objective here is to assess the extent to which plant performance is affected by the knowledge and skills acquired by workers in different areas. We also take into account the functionality and size of the region of origin and destination of the migrant. This is important as labour flows across different types of regions may reinforce already existing regional disparities (Faggian and McCann 2009).

Counting 2.7 million hires in Sweden

Our analysis relies on a longitudinal micro-database containing matched information on all workers (workplace, education, work experience, place of residence) and on the features of all plants (sector, location, performance) in Sweden between 2002 and 2006. We document the geographical mobility of new recruits across 72 functional regions in, focusing on migration between four types of regions: metropolitan regions, large urban centres, smaller urban centres, and smaller locations. We then apply regression analysis to examine how the origin and type of 2,696,909 new hires influence plant performance (defined as annual productivity growth) in a total of 69,932 Swedish plants.

The productivity impact of worker mobility in Sweden

Our analysis provides a number of interesting results.

  • First, job-related mobility in Sweden has its biggest positive impact in the three largest cities: Stockholm, Gothenburg, and Malmö. Firms in these cities benefit from hiring new workers. Most of these new workers are already settled in the big cities of Sweden, as about 55% of all job moves actually occur within or between the three metropolitan regions. The results also indicate that universities play a capital role in supplying skills to the private sector. This benefits the three large metropolitan areas, but not firms in large urban and university regions – the second regional category – as they emerge as the biggest losers from job mobility.
  • Second, job-related mobility in Sweden follows patterns akin to those of the ‘metropolitan escalator’ described by Fielding (1993) for London or by de la Roca and Puga (2017) for Spanish cities. Swedish large cities seem to “provide the greatest opportunities for those with ambition and capacity to learn” (Gordon 2015). Large cities act as a magnet to people in the early stages of their career, when their motivation to contribute to the economy is greatest.
  • Third, as highlighted by Eriksson (2011), the results of the analysis underline the importance of acknowledging the geographical dimension of labour flows for plant performance. The benefits of hiring workers from outside the region where the plant is located have been far greater than those of relying on local talent. One of the main exceptions is the recruitment of high-skilled experienced workers from other regions, which is, however, only positively significant in relatively small, mostly rural regions. This relatively moderate effect could be due to the profile of the migrants trading down from metropolitan areas in Sweden to smaller urban centres. As shown by de la Roca and Puga (2017), while these workers may gain both in salary and quality of life, they may lack the same ambition and capacity to learn behind moves by workers trading up in big cities. Ambition and motivation are, according to Gordon (2015), crucial factors for improvements in plant performance. Hence, higher competition for jobs in metropolitan areas leads to a pronounced spatial sorting, leaving more experienced workers to make a greater contribution to the economies of smaller, more peripheral regions.

Our analysis has highlighted that, when analysing job-related mobility, the experience, level of education, and geography of flows are all extremely useful in order to better understand the impact of labour mobility, as the effects of mobility flows are far from evenly distributed over the regional system. In the case of a country like Sweden, with relatively low inter-regional mobility rates and with only a few large diverse regions that together account for almost half of the plants, geographical mobility is a fundamental factor for improving the productivity and economic performance of firms – although the degree of impact of cross-regional job-related migration varies across categories of workers and according to the type of region of origin and destination.


Boschma, R, R Eriksson, and U Lindgren (2009), “How does labour mobility affect the performance of plants? The importance of relatedness and geographical proximity”, Journal of Economic Geography, 9 (2), 169–190.

Combes, P, and G Duranton (2006), “Labour pooling, labour poaching and spatial clustering”, Regional Science and Urban Economics, 36, 1–28.

De la Roca, J, and D Puga (2017), “Learning by working in big cities”, Review of Economic Studies, 84 (1), 106-142.

Eriksson, R H (2011), “Localized spillovers and knowledge flows: How does proximity influence the performance of plants?”, Economic Geography, 87 (2), 127–152.

Eriksson, R and U Lindgren (2009), “Localized mobility clusters: Impact of labour market externalities on firm performance”, Journal of Economic Geography, 9 (1), 33–53.

Eriksson, R, and A Rodríguez-Pose (2017), “Job-related mobility and plant performance in Sweden”, Geoforum, 83, 39-49.

Faggian, A, and P McCann (2009), “Human capital, graduate migration and innovation in British regions”, Cambridge Journal of Economics, 33 (2), 317-333.

Fielding, A J (1993), “Migration and the metropolis: an empirical and theoretical analysis of Inter-regional migration to and from South East England”, Progress in Planning 39, 71–166.

Fratesi, U (2014), “The Mobility of High-skilled Workers – Causes and Consequence”, Regional Studies, 48 (10), 1587-1591.

Gordon, I R (2015), “Ambition, human capital acquisition and the metropolitan escalator”, Regional Studies, 49, 1042-1055.

Malmberg, A and D Power (2005), “(How) Do (Firms in) Clusters Create Knowledge?”, Industry and Innovation, 12, 409-431.

Maskell, P, H Eskilinen, I Hannibalson, A Malmberg, and E Vatne (1998), Competitiveness, Localised Learning and Regional Development Specialisation and prosperity in small open economies, London: Routledge

McCann, P and J Simonen (2005), “Innovation, knowledge spillovers and local labour markets”, Papers in Regional Science, 84, 465-485.

Partridge, M and D S Rickman (2003), “The Waxing and Waning of Regional Economies: The Chicken-Egg Question of Jobs versus People” Journal of Urban Economics 53, 76-97.

Puga, D (2010), “The magnitude and causes of agglomeration economies”, Journal of Regional Science, 50, 203–219.

Saxenian, A, and C Sabel (2008), “Roepke Lecture in Economic Geography Venture capital in the ‘periphery’: the new Argonauts, global search, and local institution building”, Economic Geography 84, 379–394.

630 Reads