How does a region react to different types of shocks? The case of a Brazilian T&C industrial region

Regional resilience, shocks and regional economic subsystems: a brief introduction

One can understand resilience in the regional economic field as “the capacity of a regional or local economy to withstand or recover from the market, competitive and environmental shocks that have occurred or are likely to occur in its developmental growth path” (Martin & Sunley, 2015, p. 15). Such a recovery can occur through the “changing ability of a regional economy – its firms, workers, and institutions – to adapt, over time, to what pervades that economy with the needed resilience to minimize its vulnerability and ability to recover successfully from shocks” (Martin, 2018, p. 844). The concept of regional economic resilience can take on two primary meanings. The first is when there is (1) maintenance or restoration of the region’s previous developmental path, and the second is when there is (2) a change to a new sustainable regional path (Evenhuis, 2017).

In the first case, the preservation or restoration of the previous path of regional economic development emphasizes the resilience approach known as the equilibrium-centered view (Gunderson & Holling, 2002) based on deterministic and linear assumptions (Folke, 2006). Within the equilibrium-centered view, there is also an alternative approach to the idea of singular equilibrium (Folke, 2006) derived from studies on socio-ecological systems (Holling, 1973). This approach is known as ecological resilience and emphasizes multiple equilibrium points (Pendall et al., 2010). Thus, regional resilience is seen as a cycle of an adaptive renewal, and critical events are opportunities for innovation and development (Folke, 2006).

One of the main points of the field of study on regional resilience is to identify which elements determine its resilience (Martin & Sunley, 2015). In general, these studies adopt the premise that there are transversal predictive variables of regional resilience. On the one hand, some researchers argue that a region’s resilience is a function of its specific resources. Therefore, their studies shed light on the goal of “knowing how” such resources lead regions to resilience (e.g. Clark et al., 2010). This current study generally focuses on the in-depth research of regions (e.g. Hervas-Oliver et al., 2011). On the other hand, some researchers seek to “know which” the determinants that lead regions to resilience are. This line of studies generally suggests macroeconomic variables (Evenhuis, 2017) as predictors of regional resilience (e.g. Crescenzi et al., 2016). Some examples of these variables are local culture (Huggins & Thompson, 2015), innovative capacity (Clark et al., 2010), economic specialization (Lagravinese, 2015), local entrepreneurial capacity (Spigel & Vinodrai, 2020) and local knowledge networks (Crespo, Suire, & Vicente, 2016).

The framework proposed by Martin and Sunley (2015) suggests that regional resilience is determined by the dynamics of four main economic, interactive and non-exclusive subsystems: (1) structural and business subsystem, (2) labor market subsystem, (3) financial subsystem and (4) governance subsystem. There are two considerations related to the model, according to Martin and Sunley (2015): (1) a single economic subsystem does not entirely determine the economic responses and (2) the subsystems interact with agency and decision-making. In addition to the framework outlined by Martin and Sunley (2015) and Miranda Júnior and Hoffmann (2021) suggest the so-called economic subsystem of “Territorial elements.”

Regarding shocks, the regional economic resilience literature has used financial shocks as a basis for empirical research, especially the 2008–2009 global financial crisis (Gong, Hassink, Tan, & Huang, 2020). However, non-financial shocks also have the potential to cause economic losses (Martin & Sunley, 2015), such as natural disasters (Boschma, 2015), technological disruptions (Østergaard & Park, 2013), and, more recently, the COVID-19 pandemic (Sutton & Arku, 2022).

Through empirical evidence from Billington, Karlsen, Mathisen, and Pettersen (2017), we know that firms use regional resources to deal with shocks. The resilience resulting from this firm–region interaction is built by contributing to the region’s economic and social systems and vice versa (Billington et al., 2017). From the above, we extract the general proposition of this study, which complements the two considerations by Martin and Sunley (2015) about their theoretical model: the responses coming from the economic resilience subsystems are related to the types of shocks faced by regional actors (such as firms, institutions, community). Based on this, we derive three specific propositions:

A brief history of the origin of the T&C Brusque region

This research deals with the textile and clothing (T&C) sector in a Brazilian industrial cluster in the Brusque region, located in Santa Catarina, south of Brazil. The city has a territorial extension of 283.4 km², an estimated population of around 132,000 individuals, and a Human Development Index (HDI) of 0.80. The emergence of T&C industrial activities in the region dates back to the arrival of Polish and German immigrants to Brusque at the end of the 19th century. These immigrants (Glatz, 2018; Seyferth, 1974) influenced the opening of the region’s three large anchor textile factories – Renaux, Buettner and Schlösser – at the end of the 19th century. Historically, the region of Brusque is known as the “cradle of spinning” because the first spinning factory’s emergence in the state of Santa Catarina took place there in 1908. From the 1930s to the 1970s, the textile cluster increased in the number of new firms, especially spin-offs from the pioneers. During this period, the pioneers founded several support institutions. The centennial existence and economic representativeness of the Brusque region in the state of Santa Catarina justify its choice for the establishment of the T&C sector. According to Yin’s typology (2018), it is a representative or typical case.

Research design

We used a mixed-method approach applied to Brusque’s empirical case study of the Brazilian T&C industrial region to answer our research questions about how regions respond to different shocks. The case study is appropriate for the theory-building purpose (Eisenhardt & Graebner, 2007) on the phenomenon of interest that cannot be studied outside its natural setting (Yin, 2018). The case study method is also recommended in regional resilience literature. According to Martin and Sunley (2015, p. 22):

Ultimately, determining the extent and nature of regional economic adaptation in response to shocks requires an in-depth study of individual industries, firms, and workers, — only case study analysis, whether regional or local, can reveal such an extent and nature.

Data collection and sample

We used two data sources and two stages for data collection.

The first stage involved documentary research and the second in-depth interviews. This variety of sources allowed data triangulation, a reliability factor in qualitative research (Jick, 1979). We carried out documentary research at the municipal museum “Casa de Brusque” where there are historical books on the region’s T&C industrial origins, dating from the late 19th century, and the collection of the regional newspaper “O Município,” founded in the 1950s and still in circulation today. Through exploratory reading, we selected 882 newspaper reports dating from the 1980s to 2021. After the circular process of data interpretation (Denzin & Lincoln, 2005), we selected 670 newspaper reports as sampling material using as selection criteria those that addressed simultaneously (1) shocks faced by the Brusque T&C industrial region and (2) its responses to these shocks.

The second stage of data collection was to conduct in-depth interviews with 20 regional actors (e.g. companies, local supporting institutions, government agencies, labor unions, universities and others), considering the level of new insights expected for this research (Glaser & Strauss, 1967). Thirteen regional actors were identified by reading the sampling newspaper reports, and the other seven used the snowball technique. Table 1 provides an overview of the actors interviewed and their roles in the Brusque T&C industrial region.

We developed a semi-structured interview guide with the same questions used previously in a document collection: (1) What were the shocks faced by the Brusque T&C industrial region? and (2) What were the region’s responses to these shocks? Each interview lasted about one hour, and all were tape-recorded and transcribed in full.

Analysis techniques

Correspondence analysis

In the second stage, our objective was to verify whether the categorical variables “types of shocks” and “types of regional resilience economic subsystems” were statistically correlated. We applied the exploratory statistical technique of simple correspondence analysis using the symmetric normalization method. The assumptions for applying the simple correspondence analysis technique were satisfied, according to Doey and Kurta (2011). Through the R software, we adopted the criteria of two dimensions in solution and chi-square as a distance measure. According to Greenacre (2007), the application of the chi-square distance was justified because it satisfies many favorable properties of correspondence analysis, such as the principle of distributive equivalence.

Shocks and regional resilience economic subsystems

We identified different shocks and reactions to them in our raw qualitative data. Based on Knaflic (2015), the narrative in Figure 1 summarizes the main ones. Like Martin and Sunley (2015) and Diodato and Weterings (2015), we considered recessions in the level of employment in the Brusque T&C industrial region (from 1985 to 2019) as events suggestive of shocks.

After categorizing the events according to our codebook, we tabulated the respective category frequencies. The balloon plot in Figure 2 shows the absolute frequencies observed between each pair composed by the cross-tabulation between the categories: “shocks” and “economic subsystems.” The tabulations in each cell do not necessarily correspond to the same shock–reaction pair; there is a combination of distinct pairs. The assumption adopted was that a maximum of 20% of the expected values were less than 5 (Greenacre, 2007).

As can be seen, reactions to shocks are concentrated in the category “competitive shock,” with economic shocks as a subcategory. We infer, therefore, that economic shocks are the most mentioned in the regional resilience literature. However, they are not the only ones negatively impacting a region’s economy (Martin & Sunley, 2015). Based on this cross-tabulation, we applied the bivariate exploratory technique of simple correspondence analysis to investigate the association between the categories “shocks” and “economic subsystems” using the chi-square method (Table 4). This method determines whether the rows and columns are independent of each other or not.

The p-value of 0.000 (χ² 29.637, d.f. 8) at a significance level of 5% tells us that there is a minimal probability that the frequencies observed in Table 4 can be reconciled with the assumption of homogeneity. It is possible to conclude that there is a statistically significant association between the shock and categorical variables of the economic subsystems. Regarding the correspondence analysis, the total inertia was 20%, a percentage considered adequate for the simple correspondence analysis technique (Doey & Kurta, 2011). The eigenvalue values were 13.4% for dimension 1 (67.3% of inertia) and 6.5% for dimension 2 (32.7% of inertia). The perceptual map (Figure 3) illustrates these associations on a two-dimensional plane.

From 1988 to 1992

During the 1980s and mid-1990s, the national government implemented seven economic plans to curb domestic inflation. When these economic plans were implemented in 1986, the inflation rate was double digits. Among the measures taken by some of these plans was the taxation of products considered “inflation feeders,” such as clothing products, to discourage their domestic consumption. The consequence of this measure for the Brusque T&C industry was a decrease in revenue from clothing sales and an increase in inventories.

Brazilian parliamentarians promulgated, in 1988, the seventh Constitution of Brazil. As reported in 1990 and also mentioned in the interviews, the new labor rights guaranteed by the 1988’s Constitution increased labor-intensive industries’ production costs by 50%. As a result, textile firms in the Brusque region laid off many of their workers (see 1988–1992 period in Figure 1).

The Brusque T&C industrial region, previously characterized by medium and large-sized industries (Renaux, 2010), gradually became an industrial hub of micro, small and medium-sized enterprises (MSMEs). According to Glatz (2018), unemployment in the Brusque region and the expertise in T&C manufacturing activities influenced micro-entrepreneurship in the clothing sector – mainly due to the low entry costs in this industry (circular looms instead of flat looms and knitwear production instead of plain fabrics). From Eraydin (2016) and Glonti, Manvelidze, and Surmanidze (2021), we know that small-size firms contribute to regional economic resilience mainly due to their adaptability and flexibility capabilities before the effects of external shocks.

While the pioneering textile industries dispersed throughout the region, the new clothing manufacturing MSMEs concentrated on the avenue called “Rua Azambuja.” Since Brusque founding, this avenue has been a mono-structure with an intense flow of tourists attracted by the church and the religious sanctuary. The clothing trade along Rua Azambuja Avenue contributed to the gradual economic reorientation of the Brusque region (Betta, 2016). The avenue has established itself as the largest open-air shopping center in southern Brazil (Glatz, 2018). However, the inadequate structure of Rua Azambuja motivated the migration of some MSMEs – and new entrants from other regions – to large shopping malls in Brusque downtown (Betta, 2016). It has gradually contributed to the shift of commerce to the city center (Belli, Cassol, Alberton, & Marinho, 2013).

The national economic policy based on opening the market to foreign trade in the early 1990s intensified new entrants and their low-cost T&C manufacturing in the national competitive market (Meyer-Stamer, 1998). As a reaction to this market shock, among other factors, clothing MSMEs founded a local industrial and traded association called AICA. Its member firms worked in collaborative networks dedicated to internationalizing their clothing trade to South American countries. According to Witek-Hajduk (2014) and Diodato and Weterings (2015), the export orientation implies less vulnerability to demand contractions, contributing to regional economic resilience.

Previously in the Brusque T&C region, the supply chain stages were concentrated in large vertical firms. To be more competitive, the T&C firms have specialized in specific stages. Moreover, the diversity of firms operating in the same activities and at all stages of the T&C industry has made the supply chain less vulnerable to interruptions (Oxborrow & Brindley, 2012). Finally, according to Yu and Wang (2016), political instability negatively influences regional economic resilience. In this regard, it is essential to mention the political instability in the national political sphere due to the Brazilian president’s impeachment in 1992.

From 1994 to 1999

Another recession in employment in the Brusque T&C industrial region happened from 1994 to1999 (see Figure 1). During this period, the impacts of the 1994 Mexican peso crisis and the national governance economic deflation policies to maintain high domestic interest rates and overvaluation of the national currency stand out. Although this strategy controlled domestic inflation, on the other hand, the result was an increase in imported manufactured goods, increased competition and national deindustrialization (Meyer-Stamer, 1998).

Among Brusque’s local reactions, we highlight the foundation of the local technology center for clothing manufacturers, in 1994, as a strategy to improve the quality of clothing production and, thus, make the region more competitive compared to imported T&C clothing. Considering the contribution of public investment expenditures to regional resilience (Psycharis, Panori, & Athanasopoulos, 2021), among the financial arrangements category, we highlight the fiscal incentive policies of Brusque Municipality.

The employment recession in 1998 was related to the new wave of MSMEs bankrupting in Rua Azambuja due to financial insolvency caused by unsecured credit. According to a report, due to insufficient check funds, financial ruin reached a high percentage of sales in the clothing sector in 1997. On the macroeconomic sphere, domestic inflation returned to a single-digit level in 1995 due to the final national economic plan, “Plano Real.”

In addition, we highlight two other vital responses that the Brusque region gave to the recessive period of 1994–1999. First, the construction by the local government of an effluent treatment plant for the T&C industry in 1996 considerably influenced the migration of firms that operate in the clothing dyeing stage from other Brazilian industrial clusters to Brusque (Figure 4). Second, the creation of the trade fair to promote Brusque’s T&C manufacturers in 1997, called Pronegócio, in partnership with Sebrae, a non-state institute that supports institutions aimed at MSMEs.

From 2008 to 2009

According to some reports, the employment drop in Brusque in 2009 (see Figure 1) is related to the effects of the 2008–2009 global financial crisis. Although the impacts of this crisis on the Brazilian economy were less pronounced in the short term, in the long term, it caused instability in the political environment, risk aversion and a preference for liquidity in the economic environment (Domingues, Magalhães, Betarelli, Carvalho, & Santiago, 2015). In addition, according to interviewees, the drop in employment in Brusque in 2010 was related to deindustrialization associated with the decline of the large anchor textile firms Buettner, Schloesser and Renaux.

From 2010 to present

According to the interviewed entrepreneurs, the drop in employment in Brusque related to T&C firms in this period was mainly due to the reduction in sales caused by the increased competition related to new entrants installed in points of sale built along the access roads to the city of Brusque. It negatively influenced shopping tourism in the city center and, consequently, T&C sales. As a strategy for not depending on shopping tourism, firms in the region adhered to the private label business model in 2010. Contrary to the evidence of Pal, Torstensson, and Mattila (2014), retailers’ control over all aspects of T&C manufacturing led to several innovations by Brusque firms. Some examples of such innovations are the changes in the functional layout, quality control of clothing, compliance with labor laws, formalization of company registration with government agencies, the fight against slave labor and compliance with environmental protection laws.

Other events related to this period that deserve mention are (1) the record increase in cotton prices due to lower domestic supply and low inventories in the international market and (2) one of the most significant floods in the history of the Brusque region. The economic consequences for the local T&C firms were both intra-organizational (destruction of machinery and inputs) and inter-organizational (supply chain disruption). T&C firms suspended their activities in the short term and anticipated their employees’ vacations.

During the 2010s, there was a shortage of T&C labor force in the Brusque industrial region, mainly due to the domestic economic recession in 2014 and the consequent political instability arising from the impeachment of the Brazilian president in 2016. As an example of entrepreneurship in industrial and business structure in the region, medium-sized T&C firms, in partnership with local support institutions, mobilized resources to create production cells within the firms, inspired by Toyota, to overcome this market shock.

Another market shock related to suppliers was a national truck drivers’ strike, called the Diesel crisis, in protest against the constant increase in fuel prices in 2018. The consequence for the Brusque T&C firms and other industrial sectors was the total suspension of T&C manufacturing activities. An immediate reaction of firms to this shock was the adoption of non-cotton inputs in their production.

Finally, although numerical data on employment in the Brusque T&C industrial region has yet to be available, we have collected qualitative data on the reaction of Brusque’s T&C firms to the initial effects of the COVID-19 pandemic in the region. A study carried out in Chinese regions, Gong et al. (2020), show evidence that the institutional experience dealing with previous pandemic crises, government support and regional industrial structures can affect regional resilience. However, we found that T&C firms in Brusque used labor market conditions, such as holidays for all employees and flexibility in salaries and working hours, to react to the lockdown.