VoxEU Column Industrial organisation

Plants’ resilience and clusters: Evidence from the Canadian textile and clothing industry

Policymakers strive to encourage resilience among firms. We often assume that industry clustering creates resistance to shocks. This column uses the evidence from Chinese imports in the Canadian textile industry to show that firms in clusters were in fact no more resilient to the ‘China shock’.

Everywhere in the world, policymakers are trying to build more resilient economies (IMF 2017, Caldera Sánchez et al. 2015).  As there is an academic consensus that geographic clustering leads to productivity gains (e.g. Duranton and Puga 2004, Combes and Gobillon 2014), it is tempting to associate clusters with other good outcomes, such as more ‘resilient’ firms, industries, or regions.

We do not have much empirical evidence to back this up. Delgado and Porter (2017) is an exception. They studied the relationship between resilience and clusters of related industries in the US during the Global Crisis in the US using aggregate data, and found that industries experience a higher employment growth when located in a region where other related industries are represented. On the other hand, Martin et al. (2013) did not find evidence that French exporters in clusters had been more resilient in 2008 and 2009. 

In recent work (Behrens et al. 2017), we ask whether firms in geographic clusters are more resilient than others. More specifically, we look at the Canadian textile and clothing (T&C) sector to investigate how plants inside and outside geographic clusters reacted to the profound changes the sector has undergone since the turn of the century.

The disruptive impact of Chinese competition

The Canadian textile and clothing sector recently experienced a series of negative shocks. The number of plants active in this sector halved between 2005 and 2013 as a consequence. Figure 1 depicts the evolution of textile and clothing employment, and shows that the sector also lost about two-thirds of its workforce between 2005 and 2013.  

Figure 1 Employment in the Canadian textile and clothing industry, 2001-2013

Source: Behrens et al. (2017).

Meanwhile, textile and clothing imports from China increased by approximately 200% in a decade. A significant part of this increase was due to quota removals in 2005 at the end of the Multifibre Arrangement (MFA). The shift in trade protection was stronger in Canada than the US or the EU, because it did not limit the influx of textile imports (Audet 2007: 270). We investigate whether these shocks affected the plants located inside and outside clusters equally.

A bottom-up approach

Existing research commonly uses industry-level aggregate data to evaluate the influence of geographic clustering on economic performance or resilience to shocks (Delgado et al. 2014). We look at the location of individual plants instead. Within a sector, plants operate in different locations – some belong to large geographic clusters, others to smaller but more specialised ones, and many plants are isolated.

We develop a bottom-up procedure that builds on point-pattern techniques from spatial statistics to identify clusters of geographically proximate plants from geocoded data. Figure 2 shows the location of textile and clothing clusters in Québec in 2001, the province with the highest share of textile and clothing employment in Canada.

Figure 2 Location of textile and clothing clusters in Québec in 2001

Source: Behrens et al. (2017).

As can be seen from the figure, there is substantial variation in the geography of individual plants and geographic clusters. With this procedure, we are thus able to compare plants' outcomes depending on whether they were inside or outside clusters.

A framework for resilience

Resilience may refer to the ability of a system either to recover from, to absorb, or to adapt to a shock, and has become a buzzword in policy and academic circles (Martin and Sunley 2015). We develop a simple framework to articulate these different notions of resilience. A naive concept of resilience is simply a plant’s chance of survival after an adverse shock, but this might be the result of two forms of resilience:

  • Strong resilience: The plant can remain active in her main line of activity by absorbing the shock without changing its structure, identity, and function.
  • Weak resilience: The plant can switch out of textile manufacturing into some other industry – either manufacturing or services.

Plants in clusters are no more resilient than others

We assess the resilience of clusters to adverse shocks using this severe exogenous shock to the textile and clothing sector with three measures of plants' resilience, and a map of plants and clusters in Canada.

We compare how clustered and unclustered plants resisted or adapted to the surge in Chinese imports between 2001 and 2013. We use a difference-in-difference strategy exploiting the removal of import quotas on 1 January 2005. We deal with the potential endogeneity of clusters’ definition using detailed historical information on the location of textile and clothing industries in Canada in the 19th century.

Our findings all lead to the same conclusion – in the face of a severe shock, plants in clusters were neither more likely to absorb the shock, nor to adapt and switch to another industry, than unclustered plants. Plants inside clusters were no more resilient than plants outside clusters. 

There is a notable side result. We find that switching patterns are different for plants inside and outside clusters. Plants in clusters – particularly large, urban clusters – were more likely to switch to services. Plants outside clusters tended to switch to other manufacturing activities.

Effect on communities 

In the face of major disturbances – such as the ‘China shock’ – it does not matter whether firms belong to a cluster or not. But local communities that host large clusters of firms will tend to suffer more, as they would experience more closures and exits. Therefore, to anticipate the consequences of a negative industrial shock, knowing the exact location of these clusters would be at least as useful as knowing whether an industry was concentrated nationally.


Audet, D (2007), “Smooth as silk? A first look at the post MFA textiles and clothing Landscape”, Journal of International Economic Law 10(2): 267–284.

Behrens, K, B Boualam and J Martin (2017), “Are clusters resilient? Evidence from Canadian textile industries”. CEPR Discussion Paper 12184.

Caldera Sánchez, A, M Rasmussen and O Röhn (2015), "Economic resilience: what role for policies?" OECD Economics Department working paper 1251.

Combes, P-P and L Gobillon (2014), “The empirics of agglomeration economies”, in G Duranton, J V Henderson and W C Strange (eds), Handbook of Regional and Urban Economics, Vol. 5: 247–348.

Delgado, M and M E Porter (2017), “Clusters and the Great Recession”, Technical Report.

Delgado, M, M E Porter and S Stern (2014), “Clusters, Convergence, and Economic Performance”, Research Policy 43(10): 1785–1799.

Duranton, G and D Puga (2004), “Micro-foundations of urban agglomeration economies”, in J V Henderson and J-F Thisse (eds), Handbook of Regional and Urban Economics, Vol. 4: 2063–2117.

IMF (2017), “Building a More Resilient and Inclusive Global Economy” Speech by Christine Lagarde, Managing Director.

Martin, P, T Mayer and F Mayneris (2013), “Are clusters more resilient in crises? Evidence from French exporters in 2008-2009”, CEPR Discussion Paper 9667.

Martin, R and P Sunley (2015), “On the notion of regional economic resilience: Conceptualization and explanation”, Journal of Economic Geography 15(1): 1–42.

945 Reads