Clusters and Employment Growth in Mexico: Insights from a Panel Data Model

Clusters and Employment Growth in Mexico: Insights from a Panel Data Model | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Clusters and Employment Growth in Mexico: Insights from a Panel Data Model Rafael Garduño Rivera, Neil Reid, Haoying Wang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7227009/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 4 You are reading this latest preprint version Abstract This paper examines the impact of industrial clusters on employment growth in Mexican municipalities from 1986 to 2019 using a spatial panel data model. Contrary to cluster theory’s expectation of positive employment effects, our results show that greater industrial specialization, measured by the Krugman Specialization Index, reduces employment overall, with the strongest declines in manufacturing, services, and mining. Commerce is less affected. By contrast, higher education, access to electricity, and foreign direct investment significantly promote employment growth, with education exerting the largest positive effect. Spatial spillovers further reveal that employment dynamics extend beyond municipal boundaries. These findings suggest that while clustering may enhance productivity and innovation, excessive specialization undermines job creation. Policymakers should therefore emphasize diversification strategies and adapt cluster initiatives to local economic conditions when designing employment-focused regional development policies. JEL codes: O12; R11; O47 Clusters Employment Growth Regional Economic Growth Productivity 1. Introduction Following the work of Harvard economist Michael Porter ( 1990 ), the concept of industrial clusters quickly gained currency and became popular among economic development practitioners in cities, regions, and countries around the world (Sölvel, 2008; Lindqvist et al., 2003 ). Porter ( 2000 , 15) defined clusters as “geographic concentrations of interconnected companies, specialized suppliers, service providers, firms in related industries, and associated institutions (e.g., universities, standards agencies, trade associations) in a particular field that compete but also cooperate”. Thus, clusters are both place-based and industry-focused. As the concept of industrial clusters gained popularity, research on place-based cluster initiatives focused on specific industries came to the fore. These included the racing car industry in Oxfordshire (UK), the wine industry in the Napa Valley (USA), and the ceramic tile industry in Castellón province (Spain) (Henry and Pinch 2001 , Taplin 2010 , Molina-Morales et al. 2017 ). Accompanying the growing popularity of industrial clusters, a whole host of related concepts emerged (and in some cases reemerged) to bolster the idea that places (particularly regions) were “a key element of the new age of global-based capitalism” (Florida 1995 , p. 528). These concepts include, but are not limited to, social networks, tacit knowledge, the triple helix, local buzz, global pipelines, knowledge spillovers, and absorptive capacity (Porter 1990 , Howells 2002 , Gertler 2003 , Bathelt et al. 2004 , Etzkowitz & Zhou 2017 , Fritsch and Kauffeld-Monz 2008, Iammarino and McCann 2008, Ter Wal and Boschma 2009 , Spithoven et al. 2011 ). In many respects, cluster-based economic development heralded a more holistic view of regional development in which the strategic imperative was to enhance both the performance and resilience capacity of an entire regional economy. While widely adopted by communities across the world, cluster-based economic development strategies were not without their critics. Taylor ( 2010 , p. 276) referred to the cluster concept as a “mesmerizing mantra” which, among other things, “fetishises proximity, treats entrepreneurship simplistically, promotes the chaotic concept of institutional thickness, and is limited by the chaotic concept of social capital”, while Martin and Sunley ( 2003 , p. 5), who were skeptical of the efficacy of the approach, suggested that “the cluster concept should carry a public health warning”. According to Porter (1988), clusters are a useful framework for designing strategies that promote economic growth. Similarly, such regional concentrations represent advantages for companies from different sectors and, therefore, generate benefits and positive externalities in and across regions (Nolan et al., 2011 ). These benefits occur because clusters can affect competitiveness and productivity inside and outside of their regions (Martin & Sunley, 2003 ). Therefore, the interest of economic geographers, economists, economic development practitioners, and politicians in cluster policies has increased. However, there are risks in the implementation of cluster policies. Policymakers must compare the potential benefits and risks according to the characteristics of the region and make a rigorous analysis before promoting a local industrial cluster policy as a strategy for economic development (Palazuelos, 2005 ). The analysis of specialized industrial clustering is relevant to designing policies that promote relations and strengthen a region's economy (Monge, 2012 ). Therefore, governments and companies create the conditions necessary for economic growth (Porter, 1998 ). Based on a review of existing research on trade-driven clusters, described in the following two sections, we investigate the impact of clusters on Mexican municipal employment growth. Building upon the findings of previous researchers, the main hypothesis is that regions with more industrial clusters in various sectors have higher levels of employment growth. At this point, it is important to clarify the difference between clustering, specialization and diversification in regional economic development. Regional economic clustering, specialization, and diversification are distinct yet interconnected concepts. Clustering refers to the geographic concentration of related industries and firms, creating a network of interconnected businesses and institutions (Mo et al., 2020 ; Shin & Hwang, 2022 ). Specialization is to focus on producing a limited range of goods or services, often within a specific industry or sector, leading to efficiency gains (Kemeny and Storper, 2014 ). Diversification involves expanding the range of goods and services produced or offered, either within a region or by a specific entity, to reduce risk and create new opportunities (Alibekova et al., 2023 ). This research will focus mainly on regional economic clustering, although, due to its interconnection with specialization and diversification, we will mention the other two concepts. Several case studies in Mexico analyze the effects of different factors (i.e., wage increase, economic activity [GVA growth], agglomeration, convergence, immigration) in specialized regions (Unger, 2003 ; Unger and Chico, 2004 ; Davila Flores, 2008; Pérez and Palacio, 2009; and Monge, 2012 ). However, very few studies have studied its impact on the economic growth of the region in terms of people employed. In addition, an analysis of the impact on employment growth in the country and its different micro-regions is needed. Employment growth is crucial for regional economic growth in Mexico as it directly impacts income levels, living standards, and overall economic activity within a region. Increased employment translates to higher household income, leading to increased consumption and investment, which further stimulates local economies. Additionally, a strong labor market can attract further investment and talent, contributing to long-term sustainable growth (Wang & Garduno-Rivera, 2023 ; Ahmad & Chongvilavan, 2024). Consequently, this investigation develops a model that measures the impact of industrial clustering on the employment growth of the municipalities in Mexico. The period analyzed goes from 1986 to 2019. The model aims to identify if clustering, as an economic strategy, has resulted in local employment growth. To answer this question, the proposed spatial econometric model examines municipal-level employment during the 1986–2019 period to identify whether industrial clusters have benefited local employment in Mexico. This model explains the variation in employment with respect to clustering (measured by the Krugman Specialization index) and other independent variables, including access to electricity, road distance to the US-Mexico border, foreign direct investment (FDI), and population educational level. Our paper is structured as follows: First, we present an overview of the current debate about clusters' advantages and disadvantages. For context purposes, we then outline the evolution of the Mexican government’s approach to the country’s economic growth. Next, we show the data used and the empirical methodology developed. In the following section, we present our results and findings. Finally, we provide a conclusion, in which we briefly discuss the significance of our findings with respect to the design and implementation of public policies of regional economic development. 2. Clusters and Economic Development There is a large extant literature on the benefits of cluster-based economic development. From the perspective of the firm, these include having access to specialized inputs and knowledge spillovers (Pandit and Cook 2003 ) and being able to engage in mutually beneficial cooperative actions such as the joint purchasing of material inputs and joint marketing (Newlands 2003 , Reid and Carrol 2008, Felzensztein et al. 2012 ). Clusters allow better access to employees and suppliers, given that firms can access a concentration of qualified, employed people, which lowers recruitment costs. In addition, a group of companies provides effective means for obtaining other important inputs, as a cluster provides a deep and specialized base of suppliers that reduces transfer costs (Porter, 1998 ). All these benefits can contribute to increased profits and plant-level productivity (Arimoto et al. 2013). The constant interaction between the firms in a cluster facilitates the learning of new techniques, concepts of service, and commercialization. Hence, a cluster can provide a firm with access to the most sophisticated forms and advanced technologies (Porter, 1998 ). The reduction in the asymmetry of information between firms favors the creation of new and more sophisticated products and projects. With respect to employees, working for a cluster firm can generate higher incomes (Gibbs and Bernat 1997 , Spencer et al. 2010 , Wennberg and Lindqvist 2010 ). Clusters can facilitate both vertical and horizontal job mobility and create an environment where it is easier to become an entrepreneur (de Blasio and Di Addario 2005 ). Communities and regions can benefit from industrial clusters as the latter can contribute to increased new business formation (Feser et al. 2008 ), job creation and employment growth (Spencer et al. 2010 , Wennberg and Lindqvist 2010 , Delgado et al. 2014 ), lower unemployment rates (Spencer et al. 2010 ), higher tax revenues (Wennberg and Lindqvist 2010 ), reduced poverty rates (Fowler and Kleit 2014 ), and increased levels of innovation (Delgado et al. 2014 ). While there are studies that provide contradictory findings to those articulated above, “on balance, industries tend to perform better when they are situated within a cluster” (Spencer et al, 2010 , 709). One of the key debates with respect to industrial clustering is the respective importance of localization and urbanization economies in driving economic growth within a region. Localization economies refer to the benefits of being part of a geographic cluster of firms that are part of the same industry. In contrast, urbanization economies refer to the benefits of being part of a large agglomeration of firms regardless of the industries to which they belong (Moomaw 1988 , Beaudry and Schiffauerova 2009 ). As far back as 1890, Alfred Marshall extolled the benefits of localization economies, arguing that being part of such a cluster of businesses provided access to a specialized labor market and supplier base, while facilitating knowledge spillovers. In contrast, Jane Jacobs ( 1969 ) emerged as a champion of urbanization economies, arguing that variety and diversification were critical to growth. Glaeser et al. ( 1992 ) meshed the ideas of Marshall (1989) with those of Arrow ( 1962 ) and Roemer (1986) to establish the Marshall-Arrow-Roemer (MAR) model. In introducing his concept of industrial clusters, Porter agreed with the MAR model but argued that competition and not a local monopoly foster and facilitate innovation. Numerous studies have explored the relative benefits of specialization, competition, and diversification on employment growth. In their examination of growth in large industries across 170 U.S cities between 1956 and 1987, Glaeser et al ( 1992 , 1150–1151) concluded that "at the city-industry level, specialization hurts, competition helps, and city diversity helps employment growth”. With respect to Indonesia, Isnaeni and Khoirunurrofk (2018) found that specialization constrained employment growth, while diversity (although not statistically significant) had a positive effect on employment growth. Using a lifetime growth model approach to a study of clusters in the United Kingdom, Beaudry and Swann ( 2009 ) found that firms grew faster when surrounded by firms in the same industry (localization economies) compared to firms surrounded by firms in other industries (urbanization economies). In the case of 171 counties of Iran from 1996 to 2006, Farahmand et al ( 2012 ) found that urbanization, but not localization economies, contributed to employment growth. Given that they have a common market, specialization can foster competitiveness among firms in the same sector. This specialization can be a challenge for new firms whose competitive disadvantages make them vulnerable to large firms that already have established markets, suppliers and customers. In this context, small firms face high entry barriers. In contrast, clustering may cause market saturation and, thus, a drop in price. If a region specializes in a certain product, market saturation can result in competition between producers in terms of prices. The increase in competition (and thus the decline in prices) may cause the market exit of those companies that exceed the market price, which could lead to the formation of an oligopolistic market. Consequently, over-saturation of the market may lead to price competition between companies without focusing on achieving innovation or improvement of the product (Pacheco-Vega, 2007 ). In addition, there is the risk of over-specialization. This may occur when most of the regional production is concentrated in a sector, and that industry experiences an economic decline, and the risk of the region's economy being negatively affected increases (Palazuelos, 2005 ). Recent methodological advances highlight the need to account for the temporal dimension in clustering. Jaya & Folmer ( 2021 ), for instance, extend Agglomerative Hierarchical Clustering to identify spatiotemporal clusters, illustrating how regional patterns can evolve dynamically over time. With respect to the focus of this paper, Mexico, Mendoza-Velazquez (2017) examined the impact of industrial specialization, competition, and diversity on employment growth in 41 traded clusters in Mexico from 2004 to 2008. He found that industrial specialization generally exerted a negative impact on employment growth within both states and clusters. He suggested that this may be reflective of low levels of innovation, the high costs of employment reassignation, low levels of adaptability, and the early-stage nature of the clusters with respect to their position in the cluster life cycle. Six clusters experienced employment growth because of specialization: business services, processed foods, petroleum and gas products, construction machinery, transportation and logistics, and metal manufacturing. In contrast, competition positively impacted employment growth, suggesting that competitive environments foster innovation and job creation. From an employment growth perspective, these findings suggest that it is better to have a larger number of industries/clusters competing in the same geographic region, as opposed to a smaller number competing in an environment of greater industrial concentration (Mendoza-Velazquez 2017). Juárez-Torres et al. ( 2022 ) simulated the impact of an exogenous shock amounting to one billion Mexican pesos on the final demand of 24 clusters in Mexico. In doing so, they calculated both within-cluster spillover effects (direct effect) and the effect on the other 23 clusters (indirect effect). The impact varies across clusters. For example, the oil and gas sector generates 400 new jobs, while the retail and eating services sector generates 7,200 new jobs. 3. The Mexican Context In the period following World War II and the early 1980s, Mexico’s economic development strategy relied heavily upon state-led industrialization and the fostering of manufacturing growth via import substitution. Tariffs and strict controls on foreign direct investment were cornerstones of this policy. As a result, Mexico was “one of the most protected economies in Latin America” (Vernon 1992 , 721). It was during this period that the Maquiladora program was introduced. These policies demonstrated some success, with Mexico’s per capita real GDP growing at an average annual rate that exceeded 3%. At the same time, however, Mexico failed to develop a strong capital goods industry and became heavily dependent upon oil revenues and external debt. Indeed, it was the oil crisis of 1981 coupled with rising US interest rates that prompted Mexico to declare a moratorium on its external debt payments, leading to a fundamental shift in its development strategy (Moreno-Brid et al. 2009 ). This shift meant a reduced role for the state and a neo-liberal agenda focused on trade and financial liberalization. Market deregulation, privatization, free trade, and foreign direct investment came to the fore. Mexico became a signatory to the General Agreement on Tariffs and Trade (GATT) in 1986 and the North American Free Trade Agreement (NAFTA) in 1994 (Vernon 1992 ). Bilateral free trade agreements with other countries (e.g., Japan) followed. Mexico’s foreign trade increased significantly following the introduction of the new policies. Mexico’s export revenue was heavily dependent upon oil up until the mid-1980s. Thereafter, an increasing share of Mexico’s export earnings came from the manufacturing sector (Moreno-Brid et al. 2009 ). In 2024, manufactured goods accounted for 89.8% of Mexico’s export earnings (INEGI, 2025). Sector-specific policies (trying to pick winners) were abandoned and replaced with ‘horizontal’ policies that were applied across all sectors of the economy. Thus, specific sectors no longer received preferential treatment (e.g., tax breaks); rather, the burdens of bureaucracy were simplified to the benefit of all industries and economic sectors. One goal of this new strategy was the creation of between 800,000 and 1 million new jobs annually, a goal that has not been met. Moreno-Brid et al. ( 2009 , 155) argue “the downsizing of the Mexican state led to a contraction of public investment that was not compensated for by the private sector”, with the result that “neo-liberal reforms failed to put the economy on a path of sustained and robust expansion” (Moreno et al, 2009, 156). While Mexico’s primary exports shifted from oil to manufactured goods such as auto parts, clothing, and electronics, fixed capital investment was insufficient to generate enough new employment opportunities. In exposing industries to international competition, trade liberalization had differential impacts on the behavior and performance of clusters in Mexico. For example, Mexican shoe manufacturers found themselves ill-equipped to compete with the influx of imported footwear in terms of price, quality, and fashion content. While the Mexican government did reinstate some protective measures (e.g., increased tariffs on shoes from China and devaluation of the peso), Mexican shoe manufacturers in Guadalajara who increased their levels of cooperation (e.g., information exchange with suppliers) with each other enjoyed improved performance (Rabellotti 1999 ). Following the signing of the North American Free Trade Agreement (NAFTA) in 1994, Mexico’s industrial policy was rebalanced towards identifying and supporting targeted industries. For example, under President Zedillo (1994–2000), the Program for Industrial Policy and Foreign Trade (PROPICE) was launched, and industries with greater export potential (e.g., automobiles and electronics) were favored. Subsequent administrations continued sector-specific initiatives (Flores et al. 2018 ). A topic of significant debate with respect to Mexico is income inequality, which has been a critical challenge throughout history. Artal-Tur et al. ( 2025 ) found that economic growth exacerbated income inequalities, while periods of economic slowdown did the opposite. It represents a problem with social, political, and economic implications. According to the World Bank, the Gini index in Mexico was 43.5 in 2022 1 . Also, Esquivel-Hernández (2015) notes that Mexico is among the 25% of countries with the highest rate of inequality in the world. Given this background of inequality in Mexico, it is necessary to promote public policies that reduce the gap between the rich and the poor. A cluster policy can be used to reduce the asymmetry between regions through the promotion of economic growth (employment) and an increase in the productivity of less developed regions. In the literature on clusters in Mexico, several papers studied the impact of factors such as wages, trade integration, and location in different clusters. However, there is a need to analyze the impact of clusters on regional employment growth. It can be concluded that, in the Mexican case, clusters have had a positive impact on their regions since they have managed to promote productive innovation, improve the level of specialization, lower transaction costs, and encourage foreign trade. From this review of the literature, the central hypothesis of this work arises. It aims to test the hypothesis that regions with a greater variety of clusters have higher employment growth in various sectors. The objectives of developing a model that tests this hypothesis are, first, to highlight the impact that cluster policies have had on the employment growth of Mexico and, second, to implement such policies as a tool for employment growth in the less developed regions of the country. 4. Research design We assembled a spatial balanced panel data set of the 2,377 Mexican municipalities from 1986 to 2020. The data comes from different sources. We use data collected by INEGI's Mexican Economic Censuses for 1986, 1989, 1994, 1999, 2004, 2009, 2014, and 2019 and complement it with INEGI's Mexican Population Censuses of 1980, 1990, 2000, 2010, and 2020, including two conteos (smaller-scale censuses) in 1995 and 2005. The statistics for each census record information from the previous year, i.e., the economic census of 2009 reflects the 2008 data. The municipal level is used because clusters are generated with industrial proximity, making the municipal level more representative than the state level. Furthermore, the aim is to avoid the problem of a state-level analysis since it would not consider the impact of intra-region heterogeneity on each municipality's economic growth. In total, we use 19,016 observations obtained in accordance with the 1986 municipal boundaries (Garduño-Rivera, 2014). The model seeks to explain employment as a proxy of economic development in terms of employed people in the municipal population (during the years 1986 to 2019). This variable was selected because it is a good measure of the spillover effect that regional specialization and clusters generate across a region (Brenner & Kauermann, 2016 ). Given data limitations, the following independent variables were included: industrial clustering indicator, education, access to electricity, and foreign direct investment (FDI). The aim is to estimate the impact of industrial clustering on the employment of Mexican municipalities. The following Spatial Autocorrelation Regression (SAR) model for panel data with fixed effects is proposed: Equation (1) $$\:{L}_{it}={\lambda\:}_{1}{WL}_{it}+{\beta\:}_{1}{K}_{it-5}+{\beta\:}_{2}{HEdu}_{it-5}+\:{\beta\:}_{3}{Electricity}_{it-5}+{{\beta\:}_{4}{FDI}_{it-5}+{c}_{i}+e}_{it}$$ where we use employment as our outcome variable and lag the explanatory variables for one period (5 years). \(\:{c}_{i}\) is a vector of municipality-level effects and \(\:{e}_{it}\) is a vector of spatially lagged error. The model seeks to explain the influence of several key factors on employment ( \(\:{L}_{it}\) ) in municipality i. W is a standardized spatial weight matrix based on the inverse distance between municipalities. And \(\:{\lambda\:}_{1}\) specifies how much potential spillover effect there is of the dependent variable ( \(\:{L}_{it}\) ) from the surrounding municipalities into municipality i . This means that the employment of one municipality interacts with the employment in its neighborhood. The main hypothesis is that industrial clustering is important to economic development at the municipality level. The Krugman Specialization Index ( \(\:{K}_{it-1}\) ) of each municipality i at time t-5 is used to measure industrial specialization and clustering. The Krugman specialization index (K) calculates the difference in the productive structure of one municipality against a reference municipality (the average municipality) (Gordo et al. 2023). This index can take values between zero and two. If the index is zero (perfect diversification), the economic structure of the municipality resembles the economic structure of the reference level. A highly specialized municipality will receive a high K value, while a municipality specialized in similar industries to the reference municipalities will receive a low K value (Palan, 2010 ). The education variable ( \(\:{HEdu}_{it-1}\) ) is the percentage of the population that has a university degree in each municipality. The justification for using the education variable is that higher education increases the probability of skilled employment and, therefore, of more people being employed in the region (Arends-Kuenning et al., 2019 & 2024 ; Zheng & Du, 2020 ). Following this proposition, it is expected that this variable would have a positive effect on the percentage of employed people in the municipalities. The electricity access variable ( Electricity t−1 ) measures regional access to electricity. It is the percentage of households with electricity at the municipal level. The main idea behind households' access to electricity is that it is a proxy for economic development, which affects personal productivity. The expected estimate of this variable is positive, given that better access to electricity increases firms' productivity and the standard of living of the region's population (Alam et al., 2018 ). Foreign direct investment (FDI) by municipality is a variable that indicates the level of foreign investment going to a municipality, which could benefit economic growth. This measures the FDI influence in the economic and social sphere, serving as an indicator for the creation or spurring of an economic cluster via an increase in economic activity. This data was obtained from the “Data Mexico” initiative provided by the Economic Minister of Mexico (Secretaría de Economía 2024). We use a SAR model for panel data with fixed effects to unpack the spatial heterogeneity of the employment activity across different sectors, regions and time. The Municipalities fixed-effects controls for the unobserved characteristics of individual municipalities. This research endeavor is necessary because economic activity across sectors, regions, and time relies heavily on local resources, the U.S. market and economic development policies. Our methodological approach is consistent with recent advances in cluster analysis. In particular, Jaya & Folmer ( 2021 ) propose an extension of hierarchical clustering that incorporates both spatial and temporal dimensions, reinforcing the relevance of considering dynamic regional interactions in our analysis of Mexico. Table 1 provides summary statistics for all variables. Table 1 Variable Summary Statistics Variable Obs Mean Std. dev. Min Max Labor 19,016 14188.97 138946.8 0 6,251,389 Labor_com 19,016 1963.864 10464.19 0 418,124 Labor_mfng 19,016 2791.271 25577.87 0 1,730,249 Labor_ming 19,016 57.5102 773.6278 0 77,134 Labor_serv 19,016 4215.198 48084.99 0 2,590,086 K 19,016 0.889805 1.197304 0 112 HEdu 16,639 0.052001 0.0592948 0 1 Electricity 19,016 0.803175 0.1892603 0 1 FDIall 19,016 55564.96 785254.8 0 46,700,000 5. Results Table 2 summarizes the results of the SAR model for panel data with fixed effects. Since the variables on both sides of Eq. 1 were log-transformed, all estimated parameters can be interpreted as elasticities. The results were obtained from 16,639 observations that correspond to all 2,377 municipalities in 1989, 1994, 1999, 2004, 2009, 2014, and 2019 for the SAR fixed-effects regression with the balanced panel data. Since we lagged the explanatory variables by one period, we lost all that period´s observations, which reduced our panel data from 19,016 observations to 16,639. The first column shows the results obtained using the entire employed population as the dependent variable. The second to fifth columns use only industry-level employment , going from retail/wholesale commerce, manufacturing, and mining to the service sector, as the dependent variable. Table 2 Regression results measuring the impact of clustering on employment Variables (1) lnpersonnel (2) lnper_com (3) lnper_mfng (4) lnper_ming (5) lnper_serv L5lnksi -0.188*** 0.0425 -0.387*** -0.138** -0.219*** (-7.78) (1.38) (-11.23) (-2.92) (-7.38) L5lnhigh_edu 9.399*** -0.0290 4.835*** -3.911*** 10.38*** (48.14) (-0.17) (18.21) (-13.22) (44.70) L5lnelectr 1.244*** 0.927*** 0.974*** -0.0383 1.588*** (24.98) (13.83) (13.54) (-0.41) (24.34) L5lnFDI 0.00629** 0.0314*** 0.00983** -0.0190*** 0.00617* (2.99) (11.64) (3.27) (-4.61) (2.39) \(\:{\varvec{\lambda\:}}_{1}\) 0.721*** 0.974*** 0.811*** 0.838*** 0.671*** (85.90) (134.22) (54.27) (19.84) (61.30) N 16,639 16,639 16,639 16,639 16,639 Standard errors in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1 The regression results suggest five main findings. First, clustering has a negative effect on total employment. A one percent increase in the Krugman index, an increase in specialization, leads to a 0.19 percent decrease in total employment, a 0.22 percent decrease in services employment, a 0.14 percent decrease in mining employment, and a 0.39 percent decrease in manufacturing employment. The manufacturing sector has experienced the most significant decline as a consequence of increased specialization. In contrast, commerce has remained relatively unaffected. Nevertheless, it is essential to emphasize that specialization does not inherently result in market concentration or economic dependence on a single regional activity. Put differently, the prioritization of particular industries does not necessarily entail the dominance of a limited number of firms within the regional economy, nor, ultimately, does it imply an excessive reliance on a single form of economic activity. While specialization can lead to increased efficiency and productivity, it doesn't automatically translate into market dominance or a lack of diversification in the broader economy. The actions of the State, as discussed above, must consider that specialization may lead to lower economic growth and lower employment (Delechat et al., 2024 ; Kemeny & Storper, 2014 ). Therefore, specialization in one sector should not be encouraged without considering the local economic context. In this sense, specialization in several different economic sectors in the same region can be beneficial. Clustering in different economic sectors would transform a municipality or region into a more economically resilient area. The second finding is the effects of higher education on employment. As expected, its effect is positive and significant in total employment, manufacturing, and service sectors. A one-percent increase in higher education provides a 9.4% increase in total employment overall. Note that the effect varies drastically among sectors: In manufacturing, the effect is 4.8%, whereas for service, where education is a necessity, it is above 10%. It is important to note that the mining sector does not depend on higher education - quite the opposite. For mining, a one percent increase in higher education reduces their employment by almost 4 percent. This result suggests that mining tends to locate where the unskilled labor is, not necessarily where the college-educated are found. The third finding is related to “access to electricity”, which is a proxy for infrastructure readiness. The estimate is positive and significant in most of the specifications. For the mining sector, “access to electricity” is negative, although it is not significant, which means that places with high access to electricity may not have mining. This result is also expected since all the mining is done far away from the urban areas, where access to electricity is higher. It is important to note that the service sector benefits the most from it, and it is strongly dependent on access to electricity. A one-percent increase in access to electricity boosts employment in this sector by 1.6%. These results concur with existing studies, such as Alam et al. ( 2018 ), which report that the higher the access to electricity, the higher the employment and labor productivity in urban areas. The fourth finding relates to the distance to the US-Mexico border, which is significant and negative only for the (retail/wholesale) commerce sector. All the other sectors and total employment are not affected by the distance to the US-Mexico border. For the (retail/wholesale) commerce sector, it is important to be close to the border since a one-percent increase in the distance to the border reduces employment by 3.28%. Finally, as expected, FDI is positive and significant in all the specifications except for the service sector. The results suggest that a one percent increase in FDI 2 can increase employment by 0.0086% in the lowest case (total employment) and 0.05% in the highest case (agriculture). These results concur with previous studies that increasing FDI boosts economic activity, which improves employment and wages as a result (i.e., Jordan and Garduño, 2024; Ibarra-Olivo & Rodriguez-Pose, 2022). 6. Conclusion According to much of the literature we reviewed, clustering impacts the development of innovation, productivity, the creation of new businesses, and employment growth in the industries that are part of a cluster. With this in mind, we sought to assess the impact of clustering on employment growth across the municipalities of Mexico. It can be hypothesized that clusters provoke a greater production of goods and services, which translates into employment growth in the region. However, the results we obtained are mixed. To test the hypothesis that regions with greater clustering in different sectors have higher employment growth, we developed a SAR panel data model to assess whether clustering has promoted employment growth across Mexican municipalities from 1986 to 2019. The contribution of this work is two-fold. On the one hand, this is the first study that focuses on analyzing the impact of clustering on employment growth in Mexico. On the other hand, this research has implications for public policies as it sheds light on how to develop industrial policies to promote job growth in an emerging market economy. Consistent with previous studies, the results obtained from the regression analysis suggest that the Krugman specialization index has a significant negative impact overall (in most of the sectors, except in the commerce sector). The greater the specialization in a region, the lower its employment growth. Hence, industry clustering policies throughout the country have had a negative effect on employment growth. Meanwhile, industrial clustering has positive benefits such as increased productivity, innovation, and competitiveness, as well as enhanced resilience to economic shocks. Clusters foster collaboration, knowledge sharing, efficient resource allocation, and ultimately attract investment. However, the results found in this research point out that clustering did not increase employment in certain sectors. It actually reduces it in most Mexican sectors, apart from commerce. Consequently, government policymakers should not promote clusters but industry diversification in those sectors. They should study the region's characteristics before adopting clustering as a strategy for economic growth if the objective is to increase local employment. Declarations Author Contribution Rafael Garduno-Rivera: Writing – review & editing, Writing – original draft, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Haoying Wang: Writing – review & editing, Writing – original draft, Methodology, Data curation, Conceptualization. Neil Reid: Writing – review & editing, Writing – original draft, Conceptualization. References Ahmad, E., & Chongvilaivan, A. (2024). Digital Transformation of Multilevel Tax Policies and Administration for Resilience and Sustainable Growth. Asian Development Bank . Alam, M. S., Miah, M. D., Hammoudeh, S., & Tiwari, A. K. (2018). The nexus between access to electricity and labour productivity in developing countries. Energy policy , 122 , 715-726. Alibekova, G., Alzhanova, F., Osmanov, Z., & Omarov, A. (2023). Regional specialization and diversification of industries in Kazakhstan. Journal of Eastern European and Central Asian Research (JEECAR) , 10 (5), 898-906. Alvarez, A., Garduño‐Rivera, R., & Nuñez, H. M. (2017). Mexico's North‐South divide: The regional distribution of state inefficiency 1988–2008. Papers in Regional Science , 96 (4), 843-859. Arends-Kuenning, M., Baylis, K., & Garduño-Rivera, R. (2019). The effect of NAFTA on internal migration in Mexico: a regional economic analysis. Applied Economics , 51 (10), 1052-1068. Arends-Kuenning, M.P., Baylis, K. & Garduño-Rivera, R. (2024). The effect of NAFTA on Mexico's wage gap. Ann Reg Sci 73 , 1241–1267. https://doi.org/10.1007/s00168-024-01287-w Arimoto, Y., Nakajima, K., & Okazaki, T. (2014). Sources of productivity improvement in industrial clusters: The case of the prewar Japanese silk-reeling industry. Regional Science and Urban Economics , 46 , 27-41. Arrow, K. J. (1962). The economic implications of learning by doing. The review of economic studies , 29 (3), 155-173. Artal-Tur, A., Osorio-Caballero, M. I., & Nuñez, R. (2025). Growth and Inequality Linkages of the Mexican States in the New Century: A Panel Data Approach with Spatially Lagged Variables. Economies , 13 (2), 54. Baylis, K., Garduño-Rivera, R., & Piras, G. (2012). The distributional effects of NAFTA in Mexico: Evidence from a panel of municipalities. Regional Science and Urban Economics , 42 (1-2), 286-302. Bathelt, H., Malmberg, A., & Maskell, P. (2004). Clusters and knowledge: local buzz, global pipelines and the process of knowledge creation. Progress in Human Geography , 28 (1), 31-56. Beaudry, C., & Schiffauerova, A. (2009). Who's right, Marshall or Jacobs? The localization versus urbanization debate. Research Policy , 38 (2), 318-337. Beaudry, C., & Swann, G. P. (2009). Firm growth in industrial clusters of the United Kingdom. Small Business Economics , 32 , 409-424. Brenner, T., & Kauermann, G. (2016). Specialization and convergence of industry-specific employment in Germany: A linear mixed-model approach with spatial components. Regional Studies , 50 (2), 326-341. Dávila Flores, A. (2008). Los clústeres industriales del noreste de México (1993-2003): Perspectivas de desarrollo en el marco de una mayor integración económica con Texas. Región y sociedad , 20 (41), 57-88. De Blasio, G., & Di Addario, S. (2005). Do workers benefit from industrial agglomeration? Journal of Regional Science , 45 (4), 797-827. Delgado, M., Porter, M. E., & Stern, S. (2014). Clusters, convergence, and economic performance. Research Policy , 43 (10), 1785-1799. Delechat, M. C. C., Melina, M. G., Newiak, M. M., Papageorgiou, M. C., & Spatafora, M. N. (2024). Economic Diversification in Developing Countries: Lessons from Country Experiences with Broad-Based and Industrial Policies . International Monetary Fund. Ezquivel-Hernández, G. E. (2015). Desigualdad extrema en México. Concentración del poder económico y político . Etzkowitz, H., & Zhou, C. (2017). The triple helix: University–industry–government innovation and entrepreneurship . Routledge. Farahmand, S., Akbari, N., & Abootalebi, M. (2012). Spatial effects of localization and urbanization economies on urban employment growth in Iran. Journal of Geography and Regional Planning , 5 (4), 115. Felzensztein, C., Gimmon, E., & Aqueveque, C. (2012). Clusters or unclustered industries? Where inter‐firm marketing cooperation matters. Journal of Business & Industrial Marketing , 27 (5), 392-402. Feser, E., Renski, H., & Goldstein, H. (2008). Clusters and economic development outcomes: An analysis of the link between clustering and industry growth. Economic Development Quarterly , 22 (4), 324-344. Flores, M., Medellín, S., & Villarreal, A. (2018). Global markets and the role of geographical proximity in Mexico's employment growth. Growth and Change , 49 (3), 548-568. Florida, R. (1995). Toward the learning region. Futures , 27 (5), 527-536. Fowler, C. S., & Kleit, R. G. (2014). The effects of industrial clusters on the poverty rate. Economic Geography , 90 (2), 129-154. Fritsch, M., & Kauffeld-Monz, M. (2010). The impact of network structure on knowledge transfer: an application of social network analysis in the context of regional innovation networks. The Annals of Regional Science , 44 , 21-38. Gertler, M. S. (2003). Tacit knowledge and the economic geography of context, or the undefinable tacitness of being (there). Journal of Economic Geography , 3 (1), 75-99. Gibbs, R. M., & Bernat Jr, G. A. (1997). Rural industry clusters raise local earnings. Rural America/Rural Development Perspectives , 12 (3), 18-25. Glaeser, E. L., Kallal, H. D., Scheinkman, J. A., & Shleifer, A. (1992). Growth in cities. Journal of Political Economy , 100 (6), 1126-1152. Gordo, E., & Pérez, M. (2002). Los efectos de la integración económica sobre la especialización y distribución geográfica de la actividad industrial en los países de la UE. Boletín Económico , (DEC), 23-34. Henry, N., & Pinch, S. (2001). Neo-Marshallian nodes, institutional thickness, and Britain's ‘Motor Sport Valley’: thick or thin? Environment and Planning A , 33 (7), 1169-1183. Howells, J. R. (2002). Tacit knowledge, innovation, and economic geography. Urban studies , 39 (5-6), 871-884. Iammarino, S., & McCann, P. (2006). The structure and evolution of industrial clusters: Transactions, technology, and knowledge spillovers. Research policy , 35 (7), 1018-1036. Ibarra-Olivo, J. E., & Rodriguez-Pose, A. (2022). FDI and the growing wage gap in Mexican municipalities. Papers in Regional Science, 101 (6), 1411–1439. Instituto Nacional de Estadística y Geografía (INEGI). 2025. Balanza Comercial de Mercancías de México (BCMM). Accessed April 14, 2025. https://www.inegi.org.mx/contenidos/saladeprensa/boletines/2025/comext_o/balcom_o2025_01.pdf. Isnaeni, N. F., & Khoirunurrofik, K. (2021). The effect of knowledge spillovers and human capital through technological intensity on employment growth in Indonesia. Asia-Pacific Journal of Regional Science , 5 (1), 21-39. Jacobs, Jane. 1969. The Economies of Cities. Random House. Jordaan, J. A., & Garduño-Rivera, R. (2024). Municipality manufacturing agglomerations unveiled: Exploring spatial structural transformation in Mexico under import substitution and trade liberalisation. Applied Geography , 168 , 103317. Juárez-Torres, M., Puigvert-Angulo, J., & Zazueta-Borboa, F. (2022). The role of clusters in the performance of the Mexican economy (No. 2022-06). Working Papers. Kemeny, T., & Storper, M. (2014). Is Specialization Good for Regional Economic Development? Regional Studies , 49 (6), 1003–1018. https://doi.org/10.1080/00343404.2014.899691 Lindqvist, Göran, Christian Ketels, and Örjan Sölvell. 2003. The Cluster Initiative Greenbook 2.0. Ivory Tower Publishers. http://www.cluster-research.org/dldocs/GreenbookSep03.pdf. Marshall, Alfred. 1890. Principles of Economics. Macmillan. Martin, R., & Sunley, P. (2003). Deconstructing clusters: chaotic concept or policy panacea? Journal of Economic Geography , 3 (1), 5-35. Mendoza-Velazquez, A. (2017). The effect of industrial competition on employment: A Porter’s approach to the study of industrial clusters in Mexico. Competitiveness Review: An International Business Journal , 27 (4), 410-432. Mo, C., He, C., & Yang, L. (2020). Structural characteristics of industrial clusters and regional innovation. Economics Letters , 188 , 109003. Molina-Morales, F. X., Martínez-Cháfer, L., & Valiente-Bordanova, D. (2017). Disruptive technological innovations as new opportunities for mature industrial clusters. The case of digital printing innovation in the Spanish ceramic tile cluster. Investigaciones Regionales-Journal of Regional Research , (39), 39-57. Monge, M. F. (2012). Análisis de la cadena productiva de Tequila: el caso de Jalisco. D isertación Licenciatura. Departamento de Economía. UAM-Azcapotzalco . Moomaw, R. L. (1988). Agglomeration economies: localization or urbanization? Urban Studies , 25 (2), 150-161. Moreno-Brid, J. C., Carpizo, J. E. P., & Bosch, J. R. (2009). Economic development and social policies in Mexico. Economy and Society , 38 (1), 154-176. Newlands, D. (2003). Competition and cooperation in industrial clusters: the implications for public policy. European Planning Studies , 11 (5), 521-532. Nolan, C., Morrison, E., Kumar, I., Galloway, H., & Cordes, S. (2011). Linking industry and occupation clusters in regional economic development. Economic Development Quarterly , 25 (1), 26-35. Pacheco-Vega, Raúl. 2007. Una crítica del paradigma de desarrollo industrial mediante clústeres forzados”. Estudios Sociológicos , 25(75): 683-707. https://doi.org/10.24201/es.2007v25n75.432. Palan, Nicole (2010). Measurement of Specialization - The Choice of Indices, FIW Working Paper , No. 62, FIW - Research Centre International Economics, Vienna Palazuelos, M. (2005). Clusters: Myth or realistic ambition for policymakers? Local Economy , 20 (2), 131-140. Pandit, N. R., & Cook, G. (2003). The benefits of industrial clustering: Insights from the British financial services industry at three locations. Journal of Financial Services Marketing , 7 (3), 230-245. Pérez, S.S.L. and M.V. Palacio. 2009. Desarrollo regional y concentración industrial: Impacto en el empleo (1994-2004). Observatorio de la Economía Latinoamericana , no. 117. Porter, M. E. (1990). The Competitive Advantage of Notions. Harvard Business Review . Porter, M. E. (2000). Location, competition, and economic development: Local clusters in a global economy. Economic Development Quarterly , 14 (1), 15-34. Porter, M. E. (1998). Clusters and the new economics of competition (Vol. 76, No. 6, pp. 77-90). Boston: Harvard Business Review. Rabellotti, R. (1999). Recovery of a Mexican cluster: devaluation bonanza or collective efficiency? World development , 27 (9), 1571-1585. Reid, N., & Carroll, M. C. (2016). Creating trust through branding: The case of Northwest Ohio's greenhouse cluster. In Agri-Food Commodity Chains and Globalizing Networks (pp. 175-187). Routledge. Romer, P. M. (1986). Increasing returns and long-run growth. Journal of Political Economy , 94 (5), 1002-1037. Secretaría de Economía (SE). Foreign Direct Investment. Data México. Accessed on April 13, 2025. https://www.economia.gob.mx/datamexico/. Shin, H., & Hwang, J. (2022). Illusions of clustering: a systematic evaluation on the effects of clusters on regional economic performance in Korea. International Regional Science Review , 45 (2), 135-160. Spencer, G. M., Vinodrai, T., Gertler, M. S., & Wolfe, D. A. (2010). Do clusters make a difference? Defining and assessing their economic performance. Regional Studies , 44 (6), 697-715. Spithoven, A., Clarysse, B., & Knockaert, M. (2011). Building absorptive capacity to organize inbound open innovation in traditional industries. Technovation , 31 (1), 10-21. Taplin, I. M. (2010). From co‐operation to competition: market transformation among elite Napa Valley wine producers. International Journal of Wine Business Research , 22 (1), 6-26. Taylor, M. (2010). Clusters: A mesmerizing mantra. Tijdschrift voor economische en Sociale Geografie , 101 (3), 276-286. Ter Wal, A. L., & Boschma, R. A. (2009). Applying social network analysis in economic geography: framing some key analytic issues. The Annals of Regional Science , 43 , 739-756. Trejo-Nieto, A. (2024). Place-based strategies for sustainable and inclusive regional development in the south of Mexico. Review of Regional Research , 44 (3), 1-25. Unger, K., & Chico, R. (2004). La industria automotriz en tres regiones de México. Un análisis de clústeres. El Trimestre Económico , 909-941. Unger, Kurt. 2003. “Los Clústeres industriales en México: especializaciones regionales y la política industrial”. CIDE, Economics Division. Working Paper 278. https://repositorio-digital.cide.edu//handle/11651/6050. Vernon, J. M. (1992). Mexico's Accession to the GATT: A Catalyst at Odds with the Outcome. Mary's LJ , 24 , 717. Wang, H., & Garduno‐Rivera, R. (2023). The economic impact of trade openness on Mexico's indigenous peoples. Regional Science Policy & Practice , 15 (7), 1612-1625. Wennberg, K., & Lindqvist, G. (2010). The effect of clusters on the survival and performance of new firms. Small Business Economics , 34 , 221-241. World Bank, Data. "The Gini index" http://datos.bancomundial.org/indicador/SI.POV.GINI (Accessed: May 2nd, 2025). Zheng, S., & Du, R. (2020). How does urban agglomeration integration promote entrepreneurship in China? Evidence from regional human capital spillovers and market integration. Cities , 97 , 102529. Footnotes "The Gini index measures to what extent the distribution of income among individuals or households within an economy deviates from a perfectly equitable distribution. It measures the area between the Lorenz curve and a hypothetical line of absolute equality; it is expressed as a percentage of the maximum area under the line. Thus, a 0 Gini index represents a perfect balance, while an index of 100 represents a perfect inequality”. See more at World Bank: http://datos.bancomundial.org/indicador/SI.POV.GINI Around $ 555 million MXN, or USD 27.8 million. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 10 Aug, 2025 Editor assigned by journal 05 Aug, 2025 Submission checks completed at journal 05 Aug, 2025 First submitted to journal 27 Jul, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7227009","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":498283004,"identity":"730e47e9-d8db-4463-a5ac-041b1a2e622c","order_by":0,"name":"Rafael Garduño Rivera","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAvElEQVRIiWNgGAWjYLACHiDmJ12LZAPJWgwOEKvavH/xsw9v2+7IGd/IPfiZh8HGnqAWmRvPjGfObXtmbHYjL1mahyEtkaADJSQOGDPzth1O3HYjx0ByBsPhBIK2SEgc/wzWsnlGjvHPGQz/CTtMgr8HYssGiRwziQ8MBxiJcBhPMeOcc4eNJc68MbP4YJBMhF/4j29meFN2WI6/Pcf4RkKFHWGHMUgkIPMMCGsAJpQDxKgaBaNgFIyCEQ0ALBo5RQUCeygAAAAASUVORK5CYII=","orcid":"","institution":"University of Navarra","correspondingAuthor":true,"prefix":"","firstName":"Rafael","middleName":"Garduño","lastName":"Rivera","suffix":""},{"id":498283005,"identity":"d443bf02-c45a-460c-b4b7-1133b1b3f062","order_by":1,"name":"Neil Reid","email":"","orcid":"","institution":"University of Toledo","correspondingAuthor":false,"prefix":"","firstName":"Neil","middleName":"","lastName":"Reid","suffix":""},{"id":498283006,"identity":"f5327411-7310-4267-a12f-01f54c01308a","order_by":2,"name":"Haoying Wang","email":"","orcid":"","institution":"New Mexico Institute of Mining and Technology","correspondingAuthor":false,"prefix":"","firstName":"Haoying","middleName":"","lastName":"Wang","suffix":""}],"badges":[],"createdAt":"2025-07-27 15:08:10","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7227009/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7227009/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":91165463,"identity":"1eec9b6c-e80f-4c39-a9f3-1df6cb5df1f2","added_by":"auto","created_at":"2025-09-12 10:17:14","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":593966,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7227009/v1/dcb3388b-a3c9-4fbb-8984-0af922496de6.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Clusters and Employment Growth in Mexico: Insights from a Panel Data Model","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eFollowing the work of Harvard economist Michael Porter (\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e1990\u003c/span\u003e), the concept of industrial clusters quickly gained currency and became popular among economic development practitioners in cities, regions, and countries around the world (S\u0026ouml;lvel, 2008; Lindqvist et al., \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2003\u003c/span\u003e). Porter (\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2000\u003c/span\u003e, 15) defined clusters as \u0026ldquo;geographic concentrations of interconnected companies, specialized suppliers, service providers, firms in related industries, and associated institutions (e.g., universities, standards agencies, trade associations) in a particular field that compete but also cooperate\u0026rdquo;. Thus, clusters are both place-based and industry-focused. As the concept of industrial clusters gained popularity, research on place-based cluster initiatives focused on specific industries came to the fore. These included the racing car industry in Oxfordshire (UK), the wine industry in the Napa Valley (USA), and the ceramic tile industry in Castell\u0026oacute;n province (Spain) (Henry and Pinch \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2001\u003c/span\u003e, Taplin \u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e2010\u003c/span\u003e, Molina-Morales et al. \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Accompanying the growing popularity of industrial clusters, a whole host of related concepts emerged (and in some cases reemerged) to bolster the idea that places (particularly regions) were \u0026ldquo;a key element of the new age of global-based capitalism\u0026rdquo; (Florida \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e1995\u003c/span\u003e, p. 528). These concepts include, but are not limited to, social networks, tacit knowledge, the triple helix, local buzz, global pipelines, knowledge spillovers, and absorptive capacity (Porter \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e1990\u003c/span\u003e, Howells \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2002\u003c/span\u003e, Gertler \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2003\u003c/span\u003e, Bathelt et al. \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2004\u003c/span\u003e, Etzkowitz \u0026amp; Zhou \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2017\u003c/span\u003e, Fritsch and Kauffeld-Monz 2008, Iammarino and McCann 2008, Ter Wal and Boschma \u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e2009\u003c/span\u003e, Spithoven et al. \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). In many respects, cluster-based economic development heralded a more holistic view of regional development in which the strategic imperative was to enhance both the performance and resilience capacity of an entire regional economy.\u003c/p\u003e\u003cp\u003eWhile widely adopted by communities across the world, cluster-based economic development strategies were not without their critics. Taylor (\u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e2010\u003c/span\u003e, p. 276) referred to the cluster concept as a \u0026ldquo;mesmerizing mantra\u0026rdquo; which, among other things, \u0026ldquo;fetishises proximity, treats entrepreneurship simplistically, promotes the chaotic concept of institutional thickness, and is limited by the chaotic concept of social capital\u0026rdquo;, while Martin and Sunley (\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2003\u003c/span\u003e, p. 5), who were skeptical of the efficacy of the approach, suggested that \u0026ldquo;the cluster concept should carry a public health warning\u0026rdquo;.\u003c/p\u003e\u003cp\u003eAccording to Porter (1988), clusters are a useful framework for designing strategies that promote economic growth. Similarly, such regional concentrations represent advantages for companies from different sectors and, therefore, generate benefits and positive externalities in and across regions (Nolan et al., \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). These benefits occur because clusters can affect competitiveness and productivity inside and outside of their regions (Martin \u0026amp; Sunley, \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2003\u003c/span\u003e). Therefore, the interest of economic geographers, economists, economic development practitioners, and politicians in cluster policies has increased. However, there are risks in the implementation of cluster policies. Policymakers must compare the potential benefits and risks according to the characteristics of the region and make a rigorous analysis before promoting a local industrial cluster policy as a strategy for economic development (Palazuelos, \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2005\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe analysis of specialized industrial clustering is relevant to designing policies that promote relations and strengthen a region's economy (Monge, \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). Therefore, governments and companies create the conditions necessary for economic growth (Porter, \u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e1998\u003c/span\u003e). Based on a review of existing research on trade-driven clusters, described in the following two sections, we investigate the impact of clusters on Mexican municipal employment growth. Building upon the findings of previous researchers, the main hypothesis is that regions with more industrial clusters in various sectors have higher levels of employment growth.\u003c/p\u003e\u003cp\u003eAt this point, it is important to clarify the difference between clustering, specialization and diversification in regional economic development. Regional economic clustering, specialization, and diversification are distinct yet interconnected concepts. Clustering refers to the geographic concentration of related industries and firms, creating a network of interconnected businesses and institutions (Mo et al., \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Shin \u0026amp; Hwang, \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Specialization is to focus on producing a limited range of goods or services, often within a specific industry or sector, leading to efficiency gains (Kemeny and Storper, \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). Diversification involves expanding the range of goods and services produced or offered, either within a region or by a specific entity, to reduce risk and create new opportunities (Alibekova et al., \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). This research will focus mainly on regional economic clustering, although, due to its interconnection with specialization and diversification, we will mention the other two concepts.\u003c/p\u003e\u003cp\u003eSeveral case studies in Mexico analyze the effects of different factors (i.e., wage increase, economic activity [GVA growth], agglomeration, convergence, immigration) in specialized regions (Unger, \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e2003\u003c/span\u003e; Unger and Chico, \u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e2004\u003c/span\u003e; Davila Flores, 2008; P\u0026eacute;rez and Palacio, 2009; and Monge, \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). However, very few studies have studied its impact on the economic growth of the region in terms of people employed. In addition, an analysis of the impact on employment growth in the country and its different micro-regions is needed. Employment growth is crucial for regional economic growth in Mexico as it directly impacts income levels, living standards, and overall economic activity within a region. Increased employment translates to higher household income, leading to increased consumption and investment, which further stimulates local economies. Additionally, a strong labor market can attract further investment and talent, contributing to long-term sustainable growth (Wang \u0026amp; Garduno-Rivera, \u003cspan citationid=\"CR76\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Ahmad \u0026amp; Chongvilavan, 2024). Consequently, this investigation develops a model that measures the impact of industrial clustering on the employment growth of the municipalities in Mexico. The period analyzed goes from 1986 to 2019. The model aims to identify if clustering, as an economic strategy, has resulted in local employment growth.\u003c/p\u003e\u003cp\u003eTo answer this question, the proposed spatial econometric model examines municipal-level employment during the 1986\u0026ndash;2019 period to identify whether industrial clusters have benefited local employment in Mexico. This model explains the variation in employment with respect to clustering (measured by the Krugman Specialization index) and other independent variables, including access to electricity, road distance to the US-Mexico border, foreign direct investment (FDI), and population educational level.\u003c/p\u003e\u003cp\u003eOur paper is structured as follows: First, we present an overview of the current debate about clusters' advantages and disadvantages. For context purposes, we then outline the evolution of the Mexican government\u0026rsquo;s approach to the country\u0026rsquo;s economic growth. Next, we show the data used and the empirical methodology developed. In the following section, we present our results and findings. Finally, we provide a conclusion, in which we briefly discuss the significance of our findings with respect to the design and implementation of public policies of regional economic development.\u003c/p\u003e"},{"header":"2. Clusters and Economic Development","content":"\u003cp\u003eThere is a large extant literature on the benefits of cluster-based economic development. From the perspective of the firm, these include having access to specialized inputs and knowledge spillovers (Pandit and Cook \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2003\u003c/span\u003e) and being able to engage in mutually beneficial cooperative actions such as the joint purchasing of material inputs and joint marketing (Newlands \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2003\u003c/span\u003e, Reid and Carrol 2008, Felzensztein et al. \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). Clusters allow better access to employees and suppliers, given that firms can access a concentration of qualified, employed people, which lowers recruitment costs. In addition, a group of companies provides effective means for obtaining other important inputs, as a cluster provides a deep and specialized base of suppliers that reduces transfer costs (Porter, \u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e1998\u003c/span\u003e). All these benefits can contribute to increased profits and plant-level productivity (Arimoto et al. 2013). The constant interaction between the firms in a cluster facilitates the learning of new techniques, concepts of service, and commercialization. Hence, a cluster can provide a firm with access to the most sophisticated forms and advanced technologies (Porter, \u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e1998\u003c/span\u003e). The reduction in the asymmetry of information between firms favors the creation of new and more sophisticated products and projects. With respect to employees, working for a cluster firm can generate higher incomes (Gibbs and Bernat \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e1997\u003c/span\u003e, Spencer et al. \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2010\u003c/span\u003e, Wennberg and Lindqvist \u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e2010\u003c/span\u003e). Clusters can facilitate both vertical and horizontal job mobility and create an environment where it is easier to become an entrepreneur (de Blasio and Di Addario \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2005\u003c/span\u003e). Communities and regions can benefit from industrial clusters as the latter can contribute to increased new business formation (Feser et al. \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2008\u003c/span\u003e), job creation and employment growth (Spencer et al. \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2010\u003c/span\u003e, Wennberg and Lindqvist \u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e2010\u003c/span\u003e, Delgado et al. \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2014\u003c/span\u003e), lower unemployment rates (Spencer et al. \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2010\u003c/span\u003e), higher tax revenues (Wennberg and Lindqvist \u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e2010\u003c/span\u003e), reduced poverty rates (Fowler and Kleit \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2014\u003c/span\u003e), and increased levels of innovation (Delgado et al. \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). While there are studies that provide contradictory findings to those articulated above, \u0026ldquo;on balance, industries tend to perform better when they are situated within a cluster\u0026rdquo; (Spencer et al, \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2010\u003c/span\u003e, 709). One of the key debates with respect to industrial clustering is the respective importance of localization and urbanization economies in driving economic growth within a region.\u003c/p\u003e\u003cp\u003eLocalization economies refer to the benefits of being part of a geographic cluster of firms that are part of the same industry. In contrast, urbanization economies refer to the benefits of being part of a large agglomeration of firms regardless of the industries to which they belong (Moomaw \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e1988\u003c/span\u003e, Beaudry and Schiffauerova \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2009\u003c/span\u003e). As far back as 1890, Alfred Marshall extolled the benefits of localization economies, arguing that being part of such a cluster of businesses provided access to a specialized labor market and supplier base, while facilitating knowledge spillovers. In contrast, Jane Jacobs (\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e1969\u003c/span\u003e) emerged as a champion of urbanization economies, arguing that variety and diversification were critical to growth. Glaeser et al. (\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e1992\u003c/span\u003e) meshed the ideas of Marshall (1989) with those of Arrow (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e1962\u003c/span\u003e) and Roemer (1986) to establish the Marshall-Arrow-Roemer (MAR) model. In introducing his concept of industrial clusters, Porter agreed with the MAR model but argued that competition and not a local monopoly foster and facilitate innovation.\u003c/p\u003e\u003cp\u003eNumerous studies have explored the relative benefits of specialization, competition, and diversification on employment growth. In their examination of growth in large industries across 170 U.S cities between 1956 and 1987, Glaeser et al (\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e1992\u003c/span\u003e, 1150\u0026ndash;1151) concluded that \"at the city-industry level, specialization hurts, competition helps, and city diversity helps employment growth\u0026rdquo;. With respect to Indonesia, Isnaeni and Khoirunurrofk (2018) found that specialization constrained employment growth, while diversity (although not statistically significant) had a positive effect on employment growth. Using a lifetime growth model approach to a study of clusters in the United Kingdom, Beaudry and Swann (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2009\u003c/span\u003e) found that firms grew faster when surrounded by firms in the same industry (localization economies) compared to firms surrounded by firms in other industries (urbanization economies). In the case of 171 counties of Iran from 1996 to 2006, Farahmand et al (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2012\u003c/span\u003e) found that urbanization, but not localization economies, contributed to employment growth.\u003c/p\u003e\u003cp\u003eGiven that they have a common market, specialization can foster competitiveness among firms in the same sector. This specialization can be a challenge for new firms whose competitive disadvantages make them vulnerable to large firms that already have established markets, suppliers and customers. In this context, small firms face high entry barriers. In contrast, clustering may cause market saturation and, thus, a drop in price. If a region specializes in a certain product, market saturation can result in competition between producers in terms of prices. The increase in competition (and thus the decline in prices) may cause the market exit of those companies that exceed the market price, which could lead to the formation of an oligopolistic market. Consequently, over-saturation of the market may lead to price competition between companies without focusing on achieving innovation or improvement of the product (Pacheco-Vega, \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2007\u003c/span\u003e). In addition, there is the risk of over-specialization. This may occur when most of the regional production is concentrated in a sector, and that industry experiences an economic decline, and the risk of the region's economy being negatively affected increases (Palazuelos, \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2005\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eRecent methodological advances highlight the need to account for the temporal dimension in clustering. Jaya \u0026amp; Folmer (\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), for instance, extend Agglomerative Hierarchical Clustering to identify spatiotemporal clusters, illustrating how regional patterns can evolve dynamically over time.\u003c/p\u003e\u003cp\u003eWith respect to the focus of this paper, Mexico, Mendoza-Velazquez (2017) examined the impact of industrial specialization, competition, and diversity on employment growth in 41 traded clusters in Mexico from 2004 to 2008. He found that industrial specialization generally exerted a negative impact on employment growth within both states and clusters. He suggested that this may be reflective of low levels of innovation, the high costs of employment reassignation, low levels of adaptability, and the early-stage nature of the clusters with respect to their position in the cluster life cycle. Six clusters experienced employment growth because of specialization: business services, processed foods, petroleum and gas products, construction machinery, transportation and logistics, and metal manufacturing. In contrast, competition positively impacted employment growth, suggesting that competitive environments foster innovation and job creation. From an employment growth perspective, these findings suggest that it is better to have a larger number of industries/clusters competing in the same geographic region, as opposed to a smaller number competing in an environment of greater industrial concentration (Mendoza-Velazquez 2017).\u003c/p\u003e\u003cp\u003eJu\u0026aacute;rez-Torres et al. (\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) simulated the impact of an exogenous shock amounting to one billion Mexican pesos on the final demand of 24 clusters in Mexico. In doing so, they calculated both within-cluster spillover effects (direct effect) and the effect on the other 23 clusters (indirect effect). The impact varies across clusters. For example, the oil and gas sector generates 400 new jobs, while the retail and eating services sector generates 7,200 new jobs.\u003c/p\u003e"},{"header":"3. The Mexican Context","content":"\u003cp\u003eIn the period following World War II and the early 1980s, Mexico\u0026rsquo;s economic development strategy relied heavily upon state-led industrialization and the fostering of manufacturing growth via import substitution. Tariffs and strict controls on foreign direct investment were cornerstones of this policy. As a result, Mexico was \u0026ldquo;one of the most protected economies in Latin America\u0026rdquo; (Vernon \u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e1992\u003c/span\u003e, 721). It was during this period that the Maquiladora program was introduced. These policies demonstrated some success, with Mexico\u0026rsquo;s per capita real GDP growing at an average annual rate that exceeded 3%. At the same time, however, Mexico failed to develop a strong capital goods industry and became heavily dependent upon oil revenues and external debt. Indeed, it was the oil crisis of 1981 coupled with rising US interest rates that prompted Mexico to declare a moratorium on its external debt payments, leading to a fundamental shift in its development strategy (Moreno-Brid et al. \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2009\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThis shift meant a reduced role for the state and a neo-liberal agenda focused on trade and financial liberalization. Market deregulation, privatization, free trade, and foreign direct investment came to the fore. Mexico became a signatory to the General Agreement on Tariffs and Trade (GATT) in 1986 and the North American Free Trade Agreement (NAFTA) in 1994 (Vernon \u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e1992\u003c/span\u003e). Bilateral free trade agreements with other countries (e.g., Japan) followed. Mexico\u0026rsquo;s foreign trade increased significantly following the introduction of the new policies. Mexico\u0026rsquo;s export revenue was heavily dependent upon oil up until the mid-1980s. Thereafter, an increasing share of Mexico\u0026rsquo;s export earnings came from the manufacturing sector (Moreno-Brid et al. \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2009\u003c/span\u003e). In 2024, manufactured goods accounted for 89.8% of Mexico\u0026rsquo;s export earnings (INEGI, 2025).\u003c/p\u003e\u003cp\u003eSector-specific policies (trying to pick winners) were abandoned and replaced with \u0026lsquo;horizontal\u0026rsquo; policies that were applied across all sectors of the economy. Thus, specific sectors no longer received preferential treatment (e.g., tax breaks); rather, the burdens of bureaucracy were simplified to the benefit of all industries and economic sectors. One goal of this new strategy was the creation of between 800,000 and 1\u0026nbsp;million new jobs annually, a goal that has not been met. Moreno-Brid et al. (\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2009\u003c/span\u003e, 155) argue \u0026ldquo;the downsizing of the Mexican state led to a contraction of public investment that was not compensated for by the private sector\u0026rdquo;, with the result that \u0026ldquo;neo-liberal reforms failed to put the economy on a path of sustained and robust expansion\u0026rdquo; (Moreno et al, 2009, 156). While Mexico\u0026rsquo;s primary exports shifted from oil to manufactured goods such as auto parts, clothing, and electronics, fixed capital investment was insufficient to generate enough new employment opportunities.\u003c/p\u003e\u003cp\u003eIn exposing industries to international competition, trade liberalization had differential impacts on the behavior and performance of clusters in Mexico. For example, Mexican shoe manufacturers found themselves ill-equipped to compete with the influx of imported footwear in terms of price, quality, and fashion content. While the Mexican government did reinstate some protective measures (e.g., increased tariffs on shoes from China and devaluation of the peso), Mexican shoe manufacturers in Guadalajara who increased their levels of cooperation (e.g., information exchange with suppliers) with each other enjoyed improved performance (Rabellotti \u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e1999\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eFollowing the signing of the North American Free Trade Agreement (NAFTA) in 1994, Mexico\u0026rsquo;s industrial policy was rebalanced towards identifying and supporting targeted industries. For example, under President Zedillo (1994\u0026ndash;2000), the Program for Industrial Policy and Foreign Trade (PROPICE) was launched, and industries with greater export potential (e.g., automobiles and electronics) were favored. Subsequent administrations continued sector-specific initiatives (Flores et al. \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2018\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eA topic of significant debate with respect to Mexico is income inequality, which has been a critical challenge throughout history. Artal-Tur et al. (\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) found that economic growth exacerbated income inequalities, while periods of economic slowdown did the opposite. It represents a problem with social, political, and economic implications. According to the World Bank, the Gini index in Mexico was 43.5 in 2022\u003csup\u003e1\u003c/sup\u003e. Also, Esquivel-Hern\u0026aacute;ndez (2015) notes that Mexico is among the 25% of countries with the highest rate of inequality in the world. Given this background of inequality in Mexico, it is necessary to promote public policies that reduce the gap between the rich and the poor. A cluster policy can be used to reduce the asymmetry between regions through the promotion of economic growth (employment) and an increase in the productivity of less developed regions.\u003c/p\u003e\u003cp\u003eIn the literature on clusters in Mexico, several papers studied the impact of factors such as wages, trade integration, and location in different clusters. However, there is a need to analyze the impact of clusters on regional employment growth. It can be concluded that, in the Mexican case, clusters have had a positive impact on their regions since they have managed to promote productive innovation, improve the level of specialization, lower transaction costs, and encourage foreign trade. From this review of the literature, the central hypothesis of this work arises. It aims to test the hypothesis that \u003cem\u003eregions with a greater variety of clusters have higher employment growth in various sectors.\u003c/em\u003e The objectives of developing a model that tests this hypothesis are, first, to highlight the impact that cluster policies have had on the employment growth of Mexico and, second, to implement such policies as a tool for employment growth in the less developed regions of the country.\u003c/p\u003e"},{"header":"4. Research design","content":"\u003cp\u003eWe assembled a spatial balanced panel data set of the 2,377 Mexican municipalities from 1986 to 2020. The data comes from different sources. We use data collected by INEGI's Mexican Economic Censuses for 1986, 1989, 1994, 1999, 2004, 2009, 2014, and 2019 and complement it with INEGI's Mexican Population Censuses of 1980, 1990, 2000, 2010, and 2020, including two \u003cem\u003econteos\u003c/em\u003e (smaller-scale censuses) in 1995 and 2005. The statistics for each census record information from the previous year, i.e., the economic census of 2009 reflects the 2008 data.\u003c/p\u003e\u003cp\u003eThe municipal level is used because clusters are generated with industrial proximity, making the municipal level more representative than the state level. Furthermore, the aim is to avoid the problem of a state-level analysis since it would not consider the impact of intra-region heterogeneity on each municipality's economic growth. In total, we use 19,016 observations obtained in accordance with the 1986 municipal boundaries (Gardu\u0026ntilde;o-Rivera, 2014).\u003c/p\u003e\u003cp\u003eThe model seeks to explain employment as a proxy of economic development in terms of employed people in the municipal population (during the years 1986 to 2019). This variable was selected because it is a good measure of the spillover effect that regional specialization and clusters generate across a region (Brenner \u0026amp; Kauermann, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). Given data limitations, the following independent variables were included: industrial clustering indicator, education, access to electricity, and foreign direct investment (FDI). The aim is to estimate the impact of industrial clustering on the employment of Mexican municipalities. The following Spatial Autocorrelation Regression (SAR) model for panel data with fixed effects is proposed:\u003c/p\u003e\u003cp\u003eEquation (1)\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$\\:{L}_{it}={\\lambda\\:}_{1}{WL}_{it}+{\\beta\\:}_{1}{K}_{it-5}+{\\beta\\:}_{2}{HEdu}_{it-5}+\\:{\\beta\\:}_{3}{Electricity}_{it-5}+{{\\beta\\:}_{4}{FDI}_{it-5}+{c}_{i}+e}_{it}$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003ewhere we use employment as our outcome variable and lag the explanatory variables for one period (5 years). \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{c}_{i}\\)\u003c/span\u003e\u003c/span\u003e is a vector of municipality-level effects and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{e}_{it}\\)\u003c/span\u003e\u003c/span\u003e is a vector of spatially lagged error. The model seeks to explain the influence of several key factors on employment (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{L}_{it}\\)\u003c/span\u003e\u003c/span\u003e) in municipality \u003cem\u003ei. W\u003c/em\u003e is a standardized spatial weight matrix based on the inverse distance between municipalities. And \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\lambda\\:}_{1}\\)\u003c/span\u003e\u003c/span\u003e specifies how much potential spillover effect there is of the dependent variable (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{L}_{it}\\)\u003c/span\u003e\u003c/span\u003e) from the surrounding municipalities into municipality \u003cem\u003ei\u003c/em\u003e. This means that the employment of one municipality interacts with the employment in its neighborhood.\u003c/p\u003e\u003cp\u003eThe main hypothesis is that industrial clustering is important to economic development at the municipality level. The Krugman Specialization Index (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{K}_{it-1}\\)\u003c/span\u003e\u003c/span\u003e) of each municipality \u003cem\u003ei\u003c/em\u003e at time \u003cem\u003et-5\u003c/em\u003e is used to measure industrial specialization and clustering. The Krugman specialization index (K) calculates the difference in the productive structure of one municipality against a reference municipality (the average municipality) (Gordo et al. 2023). This index can take values between zero and two. If the index is zero (perfect diversification), the economic structure of the municipality resembles the economic structure of the reference level. A highly specialized municipality will receive a high K value, while a municipality specialized in similar industries to the reference municipalities will receive a low K value (Palan, \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2010\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe education variable (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{HEdu}_{it-1}\\)\u003c/span\u003e\u003c/span\u003e) is the percentage of the population that has a university degree in each municipality. The justification for using the education variable is that higher education increases the probability of skilled employment and, therefore, of more people being employed in the region (Arends-Kuenning et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2019\u003c/span\u003e \u0026amp; \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Zheng \u0026amp; Du, \u003cspan citationid=\"CR79\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Following this proposition, it is expected that this variable would have a positive effect on the percentage of employed people in the municipalities.\u003c/p\u003e\u003cp\u003eThe electricity access variable (\u003cem\u003eElectricity\u003c/em\u003e\u003csub\u003e\u003cem\u003et\u0026minus;1\u003c/em\u003e\u003c/sub\u003e) measures regional access to electricity. It is the percentage of households with electricity at the municipal level. The main idea behind households' access to electricity is that it is a \u003cem\u003eproxy\u003c/em\u003e for economic development, which affects personal productivity. The expected estimate of this variable is positive, given that better access to electricity increases firms' productivity and the standard of living of the region's population (Alam et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2018\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eForeign direct investment (FDI) by municipality is a variable that indicates the level of foreign investment going to a municipality, which could benefit economic growth. This measures the FDI influence in the economic and social sphere, serving as an indicator for the creation or spurring of an economic cluster via an increase in economic activity. This data was obtained from the \u0026ldquo;Data Mexico\u0026rdquo; initiative provided by the Economic Minister of Mexico (Secretar\u0026iacute;a de Econom\u0026iacute;a 2024).\u003c/p\u003e\u003cp\u003eWe use a SAR model for panel data with fixed effects to unpack the spatial heterogeneity of the employment activity across different sectors, regions and time. The Municipalities fixed-effects controls for the unobserved characteristics of individual municipalities. This research endeavor is necessary because economic activity across sectors, regions, and time relies heavily on local resources, the U.S. market and economic development policies. Our methodological approach is consistent with recent advances in cluster analysis. In particular, Jaya \u0026amp; Folmer (\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) propose an extension of hierarchical clustering that incorporates both spatial and temporal dimensions, reinforcing the relevance of considering dynamic regional interactions in our analysis of Mexico. Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e provides summary statistics for all variables.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eVariable Summary Statistics\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"6\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVariable\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eObs\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eMean\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eStd. dev.\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eMin\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eMax\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLabor\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e19,016\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e14188.97\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e138946.8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e6,251,389\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLabor_com\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e19,016\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1963.864\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e10464.19\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e418,124\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLabor_mfng\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e19,016\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e2791.271\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e25577.87\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e1,730,249\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLabor_ming\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e19,016\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e57.5102\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e773.6278\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e77,134\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLabor_serv\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e19,016\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4215.198\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e48084.99\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e2,590,086\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eK\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e19,016\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.889805\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.197304\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e112\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHEdu\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e16,639\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.052001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.0592948\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eElectricity\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e19,016\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.803175\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.1892603\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFDIall\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e19,016\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e55564.96\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e785254.8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e46,700,000\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e"},{"header":"5. Results","content":"\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e summarizes the results of the SAR model for panel data with fixed effects. Since the variables on both sides of Eq.\u0026nbsp;1 were log-transformed, all estimated parameters can be interpreted as elasticities. The results were obtained from 16,639 observations that correspond to all 2,377 municipalities in 1989, 1994, 1999, 2004, 2009, 2014, and 2019 for the SAR fixed-effects regression with the balanced panel data. Since we lagged the explanatory variables by one period, we lost all that period\u0026acute;s observations, which reduced our panel data from 19,016 observations to 16,639. The first column shows the results obtained using the entire employed population as the dependent variable. The second to fifth columns use only industry-level employment\u003ca class=\"FNLink\" href=\"#Fn2\" id=\"#FNLinkFn2\"\u003e\u003c/a\u003e, going from retail/wholesale commerce, manufacturing, and mining to the service sector, as the dependent variable.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eRegression results measuring the impact of clustering on employment\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"6\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVariables\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003e(1)\u003c/p\u003e\u003cp\u003e\u003cem\u003elnpersonnel\u003c/em\u003e\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003e(2)\u003c/p\u003e\u003cp\u003e\u003cem\u003elnper_com\u003c/em\u003e\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003e(3)\u003c/p\u003e\u003cp\u003e\u003cem\u003elnper_mfng\u003c/em\u003e\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003e(4)\u003c/p\u003e\u003cp\u003e\u003cem\u003elnper_ming\u003c/em\u003e\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003e(5)\u003c/p\u003e\u003cp\u003e\u003cem\u003elnper_serv\u003c/em\u003e\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eL5lnksi\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-0.188***\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.0425\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-0.387***\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e-0.138**\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e-0.219***\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e(-7.78)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e(1.38)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e(-11.23)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e(-2.92)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e(-7.38)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eL5lnhigh_edu\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e9.399***\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-0.0290\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e4.835***\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e-3.911***\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e10.38***\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e(48.14)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e(-0.17)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e(18.21)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e(-13.22)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e(44.70)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eL5lnelectr\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.244***\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.927***\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.974***\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e-0.0383\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1.588***\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e(24.98)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e(13.83)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e(13.54)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e(-0.41)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e(24.34)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eL5lnFDI\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.00629**\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.0314***\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.00983**\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e-0.0190***\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.00617*\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e(2.99)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e(11.64)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e(3.27)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e(-4.61)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e(2.39)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{\\varvec{\\lambda\\:}}_{1}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.721***\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.974***\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.811***\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.838***\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.671***\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e(85.90)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e(134.22)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e(54.27)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e(19.84)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e(61.30)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eN\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e16,639\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e16,639\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e16,639\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e16,639\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e16,639\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eStandard errors in parentheses. *** p\u0026thinsp;\u0026lt;\u0026thinsp;0.01, ** p\u0026thinsp;\u0026lt;\u0026thinsp;0.05, * p\u0026thinsp;\u0026lt;\u0026thinsp;0.1\u003c/p\u003e\u003cp\u003eThe regression results suggest five main findings. First, clustering has a negative effect on total employment. A one percent increase in the Krugman index, an increase in specialization, leads to a 0.19 percent decrease in total employment, a 0.22 percent decrease in services employment, a 0.14 percent decrease in mining employment, and a 0.39 percent decrease in manufacturing employment. The manufacturing sector has experienced the most significant decline as a consequence of increased specialization. In contrast, commerce has remained relatively unaffected. Nevertheless, it is essential to emphasize that specialization does not inherently result in market concentration or economic dependence on a single regional activity. Put differently, the prioritization of particular industries does not necessarily entail the dominance of a limited number of firms within the regional economy, nor, ultimately, does it imply an excessive reliance on a single form of economic activity. While specialization can lead to increased efficiency and productivity, it doesn't automatically translate into market dominance or a lack of diversification in the broader economy. The actions of the State, as discussed above, must consider that specialization may lead to lower economic growth and lower employment (Delechat et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Kemeny \u0026amp; Storper, \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). Therefore, specialization in one sector should not be encouraged without considering the local economic context. In this sense, specialization in several different economic sectors in the same region can be beneficial. Clustering in different economic sectors would transform a municipality or region into a more economically resilient area.\u003c/p\u003e\u003cp\u003eThe second finding is the effects of higher education on employment. As expected, its effect is positive and significant in total employment, manufacturing, and service sectors. A one-percent increase in higher education provides a 9.4% increase in total employment overall. Note that the effect varies drastically among sectors: In manufacturing, the effect is 4.8%, whereas for service, where education is a necessity, it is above 10%. It is important to note that the mining sector does not depend on higher education - quite the opposite. For mining, a one percent increase in higher education reduces their employment by almost 4 percent. This result suggests that mining tends to locate where the unskilled labor is, not necessarily where the college-educated are found.\u003c/p\u003e\u003cp\u003eThe third finding is related to \u0026ldquo;access to electricity\u0026rdquo;, which is a proxy for infrastructure readiness. The estimate is positive and significant in most of the specifications. For the mining sector, \u0026ldquo;access to electricity\u0026rdquo; is negative, although it is not significant, which means that places with high access to electricity may not have mining. This result is also expected since all the mining is done far away from the urban areas, where access to electricity is higher. It is important to note that the service sector benefits the most from it, and it is strongly dependent on access to electricity. A one-percent increase in access to electricity boosts employment in this sector by 1.6%. These results concur with existing studies, such as Alam et al. (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2018\u003c/span\u003e), which report that the higher the access to electricity, the higher the employment and labor productivity in urban areas.\u003c/p\u003e\u003cp\u003eThe fourth finding relates to the distance to the US-Mexico border, which is significant and negative only for the (retail/wholesale) commerce sector. All the other sectors and total employment are not affected by the distance to the US-Mexico border. For the (retail/wholesale) commerce sector, it is important to be close to the border since a one-percent increase in the distance to the border reduces employment by 3.28%. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFinally, as expected, FDI is positive and significant in all the specifications except for the service sector. The results suggest that a one percent increase in FDI\u003csup\u003e2\u003c/sup\u003e can increase employment by 0.0086% in the lowest case (total employment) and 0.05% in the highest case (agriculture). These results concur with previous studies that increasing FDI boosts economic activity, which improves employment and wages as a result (i.e., Jordan and Gardu\u0026ntilde;o, 2024; Ibarra-Olivo \u0026amp; Rodriguez-Pose, 2022).\u003c/p\u003e"},{"header":"6. Conclusion","content":"\u003cp\u003eAccording to much of the literature we reviewed, clustering impacts the development of innovation, productivity, the creation of new businesses, and employment growth in the industries that are part of a cluster. With this in mind, we sought to assess the impact of clustering on employment growth across the municipalities of Mexico. It can be hypothesized that clusters provoke a greater production of goods and services, which translates into employment growth in the region. However, the results we obtained are mixed. To test the hypothesis that regions with greater clustering in different sectors have higher employment growth, we developed a SAR panel data model to assess whether clustering has promoted employment growth across Mexican municipalities from 1986 to 2019.\u003c/p\u003e\u003cp\u003eThe contribution of this work is two-fold. On the one hand, this is the first study that focuses on analyzing the impact of clustering on employment growth in Mexico. On the other hand, this research has implications for public policies as it sheds light on how to develop industrial policies to promote job growth in an emerging market economy.\u003c/p\u003e\u003cp\u003eConsistent with previous studies, the results obtained from the regression analysis suggest that the Krugman specialization index has a significant negative impact overall (in most of the sectors, except in the commerce sector). The greater the specialization in a region, the lower its employment growth. Hence, industry clustering policies throughout the country have had a negative effect on employment growth.\u003c/p\u003e\u003cp\u003eMeanwhile, industrial clustering has positive benefits such as increased productivity, innovation, and competitiveness, as well as enhanced resilience to economic shocks. Clusters foster collaboration, knowledge sharing, efficient resource allocation, and ultimately attract investment. However, the results found in this research point out that clustering did not increase employment in certain sectors. It actually reduces it in most Mexican sectors, apart from commerce. Consequently, government policymakers should not promote clusters but industry diversification in those sectors. They should study the region's characteristics before adopting clustering as a strategy for economic growth if the objective is to increase local employment.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eRafael Garduno-Rivera: Writing \u0026ndash; review \u0026amp; editing, Writing \u0026ndash; original draft, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Haoying Wang: Writing \u0026ndash; review \u0026amp; editing, Writing \u0026ndash; original draft, Methodology, Data curation, Conceptualization. Neil Reid: Writing \u0026ndash; review \u0026amp; editing, Writing \u0026ndash; original draft, Conceptualization.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAhmad, E., \u0026amp; Chongvilaivan, A. (2024). Digital Transformation of Multilevel Tax Policies and Administration for Resilience and Sustainable Growth. \u003cem\u003eAsian Development Bank\u003c/em\u003e.\u003c/li\u003e\n\u003cli\u003eAlam, M. S., Miah, M. D., Hammoudeh, S., \u0026amp; Tiwari, A. K. (2018). The nexus between access to electricity and labour productivity in developing countries. \u003cem\u003eEnergy policy\u003c/em\u003e, \u003cem\u003e122\u003c/em\u003e, 715-726.\u003c/li\u003e\n\u003cli\u003eAlibekova, G., Alzhanova, F., Osmanov, Z., \u0026amp; Omarov, A. (2023). Regional specialization and diversification of industries in Kazakhstan. \u003cem\u003eJournal of Eastern European and Central Asian Research (JEECAR)\u003c/em\u003e, \u003cem\u003e10\u003c/em\u003e(5), 898-906.\u003c/li\u003e\n\u003cli\u003eAlvarez, A., Gardu\u0026ntilde;o‐Rivera, R., \u0026amp; Nu\u0026ntilde;ez, H. M. (2017). Mexico\u0026apos;s North‐South divide: The regional distribution of state inefficiency 1988\u0026ndash;2008. \u003cem\u003ePapers in Regional Science\u003c/em\u003e, \u003cem\u003e96\u003c/em\u003e(4), 843-859.\u003c/li\u003e\n\u003cli\u003eArends-Kuenning, M., Baylis, K., \u0026amp; Gardu\u0026ntilde;o-Rivera, R. (2019). The effect of NAFTA on internal migration in Mexico: a regional economic analysis. \u003cem\u003eApplied Economics\u003c/em\u003e, \u003cem\u003e51\u003c/em\u003e(10), 1052-1068.\u003c/li\u003e\n\u003cli\u003eArends-Kuenning, M.P., Baylis, K. \u0026amp; Gardu\u0026ntilde;o-Rivera, R. (2024). The effect of NAFTA on Mexico\u0026apos;s wage gap. \u003cem\u003eAnn Reg Sci\u003c/em\u003e \u003cstrong\u003e73\u003c/strong\u003e, 1241\u0026ndash;1267. https://doi.org/10.1007/s00168-024-01287-w\u003c/li\u003e\n\u003cli\u003eArimoto, Y., Nakajima, K., \u0026amp; Okazaki, T. (2014). Sources of productivity improvement in industrial clusters: The case of the prewar Japanese silk-reeling industry. \u003cem\u003eRegional Science and Urban Economics\u003c/em\u003e, \u003cem\u003e46\u003c/em\u003e, 27-41.\u003c/li\u003e\n\u003cli\u003eArrow, K. J. (1962). The economic implications of learning by doing. \u003cem\u003eThe review of economic studies\u003c/em\u003e, \u003cem\u003e29\u003c/em\u003e(3), 155-173.\u003c/li\u003e\n\u003cli\u003eArtal-Tur, A., Osorio-Caballero, M. I., \u0026amp; Nu\u0026ntilde;ez, R. (2025). Growth and Inequality Linkages of the Mexican States in the New Century: A Panel Data Approach with Spatially Lagged Variables. \u003cem\u003eEconomies\u003c/em\u003e, \u003cem\u003e13\u003c/em\u003e(2), 54.\u003c/li\u003e\n\u003cli\u003eBaylis, K., Gardu\u0026ntilde;o-Rivera, R., \u0026amp; Piras, G. (2012). The distributional effects of NAFTA in Mexico: Evidence from a panel of municipalities. \u003cem\u003eRegional Science and Urban Economics\u003c/em\u003e, \u003cem\u003e42\u003c/em\u003e(1-2), 286-302.\u003c/li\u003e\n\u003cli\u003eBathelt, H., Malmberg, A., \u0026amp; Maskell, P. (2004). Clusters and knowledge: local buzz, global pipelines and the process of knowledge creation. \u003cem\u003eProgress in Human Geography\u003c/em\u003e, \u003cem\u003e28\u003c/em\u003e(1), 31-56.\u003c/li\u003e\n\u003cli\u003eBeaudry, C., \u0026amp; Schiffauerova, A. (2009). Who\u0026apos;s right, Marshall or Jacobs? The localization versus urbanization debate. \u003cem\u003eResearch Policy\u003c/em\u003e, \u003cem\u003e38\u003c/em\u003e(2), 318-337.\u003c/li\u003e\n\u003cli\u003eBeaudry, C., \u0026amp; Swann, G. P. (2009). Firm growth in industrial clusters of the United Kingdom. \u003cem\u003eSmall Business Economics\u003c/em\u003e, \u003cem\u003e32\u003c/em\u003e, 409-424.\u003c/li\u003e\n\u003cli\u003eBrenner, T., \u0026amp; Kauermann, G. (2016). Specialization and convergence of industry-specific employment in Germany: A linear mixed-model approach with spatial components. \u003cem\u003eRegional Studies\u003c/em\u003e, \u003cem\u003e50\u003c/em\u003e(2), 326-341.\u003c/li\u003e\n\u003cli\u003eD\u0026aacute;vila Flores, A. (2008). Los cl\u0026uacute;steres industriales del noreste de M\u0026eacute;xico (1993-2003): Perspectivas de desarrollo en el marco de una mayor integraci\u0026oacute;n econ\u0026oacute;mica con Texas. \u003cem\u003eRegi\u0026oacute;n y sociedad\u003c/em\u003e, \u003cem\u003e20\u003c/em\u003e(41), 57-88.\u003c/li\u003e\n\u003cli\u003eDe Blasio, G., \u0026amp; Di Addario, S. (2005). Do workers benefit from industrial agglomeration? \u003cem\u003eJournal of Regional Science\u003c/em\u003e, \u003cem\u003e45\u003c/em\u003e(4), 797-827.\u003c/li\u003e\n\u003cli\u003eDelgado, M., Porter, M. E., \u0026amp; Stern, S. (2014). Clusters, convergence, and economic performance. \u003cem\u003eResearch Policy\u003c/em\u003e, \u003cem\u003e43\u003c/em\u003e(10), 1785-1799.\u003c/li\u003e\n\u003cli\u003eDelechat, M. C. C., Melina, M. G., Newiak, M. M., Papageorgiou, M. C., \u0026amp; Spatafora, M. N. (2024). \u003cem\u003eEconomic Diversification in Developing Countries: Lessons from Country Experiences with Broad-Based and Industrial Policies\u003c/em\u003e. International Monetary Fund.\u003c/li\u003e\n\u003cli\u003eEzquivel-Hern\u0026aacute;ndez, G. E. (2015). Desigualdad extrema en M\u0026eacute;xico. \u003cem\u003eConcentraci\u0026oacute;n del poder econ\u0026oacute;mico y pol\u0026iacute;tico\u003c/em\u003e.\u003c/li\u003e\n\u003cli\u003eEtzkowitz, H., \u0026amp; Zhou, C. (2017). \u003cem\u003eThe triple helix: University\u0026ndash;industry\u0026ndash;government innovation and entrepreneurship\u003c/em\u003e. Routledge.\u003c/li\u003e\n\u003cli\u003eFarahmand, S., Akbari, N., \u0026amp; Abootalebi, M. (2012). Spatial effects of localization and urbanization economies on urban employment growth in Iran. \u003cem\u003eJournal of Geography and Regional Planning\u003c/em\u003e, \u003cem\u003e5\u003c/em\u003e(4), 115.\u003c/li\u003e\n\u003cli\u003eFelzensztein, C., Gimmon, E., \u0026amp; Aqueveque, C. (2012). Clusters or unclustered industries? Where inter‐firm marketing cooperation matters. \u003cem\u003eJournal of Business \u0026amp; Industrial Marketing\u003c/em\u003e, \u003cem\u003e27\u003c/em\u003e(5), 392-402.\u003c/li\u003e\n\u003cli\u003eFeser, E., Renski, H., \u0026amp; Goldstein, H. (2008). Clusters and economic development outcomes: An analysis of the link between clustering and industry growth. \u003cem\u003eEconomic Development Quarterly\u003c/em\u003e, \u003cem\u003e22\u003c/em\u003e(4), 324-344.\u003c/li\u003e\n\u003cli\u003eFlores, M., Medell\u0026iacute;n, S., \u0026amp; Villarreal, A. (2018). Global markets and the role of geographical proximity in Mexico\u0026apos;s employment growth. \u003cem\u003eGrowth and Change\u003c/em\u003e, \u003cem\u003e49\u003c/em\u003e(3), 548-568.\u003c/li\u003e\n\u003cli\u003eFlorida, R. (1995). Toward the learning region. \u003cem\u003eFutures\u003c/em\u003e, \u003cem\u003e27\u003c/em\u003e(5), 527-536.\u003c/li\u003e\n\u003cli\u003eFowler, C. S., \u0026amp; Kleit, R. G. (2014). The effects of industrial clusters on the poverty rate. \u003cem\u003eEconomic Geography\u003c/em\u003e, \u003cem\u003e90\u003c/em\u003e(2), 129-154.\u003c/li\u003e\n\u003cli\u003eFritsch, M., \u0026amp; Kauffeld-Monz, M. (2010). The impact of network structure on knowledge transfer: an application of social network analysis in the context of regional innovation networks. \u003cem\u003eThe Annals of Regional Science\u003c/em\u003e, \u003cem\u003e44\u003c/em\u003e, 21-38.\u003c/li\u003e\n\u003cli\u003eGertler, M. S. (2003). Tacit knowledge and the economic geography of context, or the undefinable tacitness of being (there). \u003cem\u003eJournal of Economic Geography\u003c/em\u003e, \u003cem\u003e3\u003c/em\u003e(1), 75-99.\u003c/li\u003e\n\u003cli\u003eGibbs, R. M., \u0026amp; Bernat Jr, G. A. (1997). Rural industry clusters raise local earnings. \u003cem\u003eRural America/Rural Development Perspectives\u003c/em\u003e, \u003cem\u003e12\u003c/em\u003e(3), 18-25.\u003c/li\u003e\n\u003cli\u003eGlaeser, E. L., Kallal, H. D., Scheinkman, J. A., \u0026amp; Shleifer, A. (1992). Growth in cities. \u003cem\u003eJournal of Political Economy\u003c/em\u003e, \u003cem\u003e100\u003c/em\u003e(6), 1126-1152.\u003c/li\u003e\n\u003cli\u003eGordo, E., \u0026amp; P\u0026eacute;rez, M. (2002). Los efectos de la integraci\u0026oacute;n econ\u0026oacute;mica sobre la especializaci\u0026oacute;n y distribuci\u0026oacute;n geogr\u0026aacute;fica de la actividad industrial en los pa\u0026iacute;ses de la UE. \u003cem\u003eBolet\u0026iacute;n Econ\u0026oacute;mico\u003c/em\u003e, (DEC), 23-34.\u003c/li\u003e\n\u003cli\u003eHenry, N., \u0026amp; Pinch, S. (2001). Neo-Marshallian nodes, institutional thickness, and Britain\u0026apos;s \u0026lsquo;Motor Sport Valley\u0026rsquo;: thick or thin? \u003cem\u003eEnvironment and Planning A\u003c/em\u003e, \u003cem\u003e33\u003c/em\u003e(7), 1169-1183.\u003c/li\u003e\n\u003cli\u003eHowells, J. R. (2002). Tacit knowledge, innovation, and economic geography. \u003cem\u003eUrban studies\u003c/em\u003e, \u003cem\u003e39\u003c/em\u003e(5-6), 871-884.\u003c/li\u003e\n\u003cli\u003eIammarino, S., \u0026amp; McCann, P. (2006). The structure and evolution of industrial clusters: Transactions, technology, and knowledge spillovers. \u003cem\u003eResearch policy\u003c/em\u003e, \u003cem\u003e35\u003c/em\u003e(7), 1018-1036.\u003c/li\u003e\n\u003cli\u003eIbarra-Olivo, J. E., \u0026amp; Rodriguez-Pose, A. (2022). FDI and the growing wage gap in Mexican municipalities. \u003cem\u003ePapers in Regional Science, 101\u003c/em\u003e(6), 1411\u0026ndash;1439. \u003c/li\u003e\n\u003cli\u003eInstituto Nacional de Estad\u0026iacute;stica y Geograf\u0026iacute;a (INEGI). 2025. Balanza Comercial de Mercanc\u0026iacute;as de M\u0026eacute;xico (BCMM). Accessed April 14, 2025. https://www.inegi.org.mx/contenidos/saladeprensa/boletines/2025/comext_o/balcom_o2025_01.pdf.\u003c/li\u003e\n\u003cli\u003eIsnaeni, N. F., \u0026amp; Khoirunurrofik, K. (2021). The effect of knowledge spillovers and human capital through technological intensity on employment growth in Indonesia. \u003cem\u003eAsia-Pacific Journal of Regional Science\u003c/em\u003e, \u003cem\u003e5\u003c/em\u003e(1), 21-39.\u003c/li\u003e\n\u003cli\u003eJacobs, Jane. 1969. The Economies of Cities. Random House.\u003c/li\u003e\n\u003cli\u003eJordaan, J. A., \u0026amp; Gardu\u0026ntilde;o-Rivera, R. (2024). Municipality manufacturing agglomerations unveiled: Exploring spatial structural transformation in Mexico under import substitution and trade liberalisation. \u003cem\u003eApplied Geography\u003c/em\u003e, \u003cem\u003e168\u003c/em\u003e, 103317.\u003c/li\u003e\n\u003cli\u003eJu\u0026aacute;rez-Torres, M., Puigvert-Angulo, J., \u0026amp; Zazueta-Borboa, F. (2022). \u003cem\u003eThe role of clusters in the performance of the Mexican economy\u003c/em\u003e (No. 2022-06). Working Papers.\u003c/li\u003e\n\u003cli\u003eKemeny, T., \u0026amp; Storper, M. (2014). Is Specialization Good for Regional Economic Development? \u003cem\u003eRegional Studies\u003c/em\u003e, \u003cem\u003e49\u003c/em\u003e(6), 1003\u0026ndash;1018. https://doi.org/10.1080/00343404.2014.899691\u003c/li\u003e\n\u003cli\u003eLindqvist, G\u0026ouml;ran, Christian Ketels, and \u0026Ouml;rjan S\u0026ouml;lvell. 2003. The Cluster Initiative Greenbook 2.0. Ivory Tower Publishers. http://www.cluster-research.org/dldocs/GreenbookSep03.pdf.\u003c/li\u003e\n\u003cli\u003eMarshall, Alfred. 1890. Principles of Economics. Macmillan. \u003c/li\u003e\n\u003cli\u003eMartin, R., \u0026amp; Sunley, P. (2003). Deconstructing clusters: chaotic concept or policy panacea? \u003cem\u003eJournal of Economic Geography\u003c/em\u003e, \u003cem\u003e3\u003c/em\u003e(1), 5-35.\u003c/li\u003e\n\u003cli\u003eMendoza-Velazquez, A. (2017). The effect of industrial competition on employment: A Porter\u0026rsquo;s approach to the study of industrial clusters in Mexico. \u003cem\u003eCompetitiveness Review: An International Business Journal\u003c/em\u003e, \u003cem\u003e27\u003c/em\u003e(4), 410-432.\u003c/li\u003e\n\u003cli\u003eMo, C., He, C., \u0026amp; Yang, L. (2020). Structural characteristics of industrial clusters and regional innovation. \u003cem\u003eEconomics Letters\u003c/em\u003e, \u003cem\u003e188\u003c/em\u003e, 109003.\u003c/li\u003e\n\u003cli\u003eMolina-Morales, F. X., Mart\u0026iacute;nez-Ch\u0026aacute;fer, L., \u0026amp; Valiente-Bordanova, D. (2017). Disruptive technological innovations as new opportunities for mature industrial clusters. The case of digital printing innovation in the Spanish ceramic tile cluster. \u003cem\u003eInvestigaciones Regionales-Journal of Regional Research\u003c/em\u003e, (39), 39-57.\u003c/li\u003e\n\u003cli\u003eMonge, M. F. (2012). An\u0026aacute;lisis de la cadena productiva de Tequila: el caso de Jalisco. D\u003cem\u003eisertaci\u0026oacute;n Licenciatura. Departamento de Econom\u0026iacute;a. UAM-Azcapotzalco\u003c/em\u003e.\u003c/li\u003e\n\u003cli\u003eMoomaw, R. L. (1988). Agglomeration economies: localization or urbanization? \u003cem\u003eUrban Studies\u003c/em\u003e, \u003cem\u003e25\u003c/em\u003e(2), 150-161.\u003c/li\u003e\n\u003cli\u003eMoreno-Brid, J. C., Carpizo, J. E. P., \u0026amp; Bosch, J. R. (2009). Economic development and social policies in Mexico. \u003cem\u003eEconomy and Society\u003c/em\u003e, \u003cem\u003e38\u003c/em\u003e(1), 154-176.\u003c/li\u003e\n\u003cli\u003eNewlands, D. (2003). Competition and cooperation in industrial clusters: the implications for public policy. \u003cem\u003eEuropean Planning Studies\u003c/em\u003e, \u003cem\u003e11\u003c/em\u003e(5), 521-532.\u003c/li\u003e\n\u003cli\u003eNolan, C., Morrison, E., Kumar, I., Galloway, H., \u0026amp; Cordes, S. (2011). Linking industry and occupation clusters in regional economic development. \u003cem\u003eEconomic Development Quarterly\u003c/em\u003e, \u003cem\u003e25\u003c/em\u003e(1), 26-35.\u003c/li\u003e\n\u003cli\u003ePacheco-Vega, Ra\u0026uacute;l. 2007. Una cr\u0026iacute;tica del paradigma de desarrollo industrial mediante cl\u0026uacute;steres forzados\u0026rdquo;. \u003cem\u003eEstudios Sociol\u0026oacute;gicos\u003c/em\u003e, 25(75): 683-707. https://doi.org/10.24201/es.2007v25n75.432.\u003c/li\u003e\n\u003cli\u003ePalan, Nicole (2010). Measurement of Specialization - The Choice of Indices, \u003cem\u003eFIW Working Paper\u003c/em\u003e, No. 62, FIW - Research Centre International Economics, Vienna \u003c/li\u003e\n\u003cli\u003ePalazuelos, M. (2005). Clusters: Myth or realistic ambition for policymakers? \u003cem\u003eLocal Economy\u003c/em\u003e, \u003cem\u003e20\u003c/em\u003e(2), 131-140.\u003c/li\u003e\n\u003cli\u003ePandit, N. R., \u0026amp; Cook, G. (2003). The benefits of industrial clustering: Insights from the British financial services industry at three locations. \u003cem\u003eJournal of Financial Services Marketing\u003c/em\u003e, \u003cem\u003e7\u003c/em\u003e(3), 230-245.\u003c/li\u003e\n\u003cli\u003eP\u0026eacute;rez, S.S.L. and M.V. Palacio. 2009. Desarrollo regional y concentraci\u0026oacute;n industrial: Impacto en el empleo (1994-2004). \u003cem\u003eObservatorio de la Econom\u0026iacute;a Latinoamericana\u003c/em\u003e, no. 117. \u003c/li\u003e\n\u003cli\u003ePorter, M. E. (1990). The Competitive Advantage of Notions. \u003cem\u003eHarvard Business Review\u003c/em\u003e.\u003c/li\u003e\n\u003cli\u003ePorter, M. E. (2000). Location, competition, and economic development: Local clusters in a global economy. \u003cem\u003eEconomic Development Quarterly\u003c/em\u003e, \u003cem\u003e14\u003c/em\u003e(1), 15-34.\u003c/li\u003e\n\u003cli\u003ePorter, M. E. (1998). \u003cem\u003eClusters and the new economics of competition\u003c/em\u003e (Vol. 76, No. 6, pp. 77-90). Boston: Harvard Business Review.\u003c/li\u003e\n\u003cli\u003eRabellotti, R. (1999). Recovery of a Mexican cluster: devaluation bonanza or collective efficiency? \u003cem\u003eWorld development\u003c/em\u003e, \u003cem\u003e27\u003c/em\u003e(9), 1571-1585.\u003c/li\u003e\n\u003cli\u003eReid, N., \u0026amp; Carroll, M. C. (2016). Creating trust through branding: The case of Northwest Ohio\u0026apos;s greenhouse cluster. In \u003cem\u003eAgri-Food Commodity Chains and Globalizing Networks\u003c/em\u003e (pp. 175-187). Routledge.\u003c/li\u003e\n\u003cli\u003eRomer, P. M. (1986). Increasing returns and long-run growth. \u003cem\u003eJournal of Political Economy\u003c/em\u003e, \u003cem\u003e94\u003c/em\u003e(5), 1002-1037.\u003c/li\u003e\n\u003cli\u003eSecretar\u0026iacute;a de Econom\u0026iacute;a (SE). Foreign Direct Investment. Data M\u0026eacute;xico. Accessed on April 13, 2025. https://www.economia.gob.mx/datamexico/.\u003c/li\u003e\n\u003cli\u003eShin, H., \u0026amp; Hwang, J. (2022). Illusions of clustering: a systematic evaluation on the effects of clusters on regional economic performance in Korea. \u003cem\u003eInternational Regional Science Review\u003c/em\u003e, \u003cem\u003e45\u003c/em\u003e(2), 135-160.\u003c/li\u003e\n\u003cli\u003eSpencer, G. M., Vinodrai, T., Gertler, M. S., \u0026amp; Wolfe, D. A. (2010). Do clusters make a difference? Defining and assessing their economic performance. \u003cem\u003eRegional Studies\u003c/em\u003e, \u003cem\u003e44\u003c/em\u003e(6), 697-715.\u003c/li\u003e\n\u003cli\u003eSpithoven, A., Clarysse, B., \u0026amp; Knockaert, M. (2011). Building absorptive capacity to organize inbound open innovation in traditional industries. \u003cem\u003eTechnovation\u003c/em\u003e, \u003cem\u003e31\u003c/em\u003e(1), 10-21.\u003c/li\u003e\n\u003cli\u003eTaplin, I. M. (2010). From co‐operation to competition: market transformation among elite Napa Valley wine producers. \u003cem\u003eInternational Journal of Wine Business Research\u003c/em\u003e, \u003cem\u003e22\u003c/em\u003e(1), 6-26.\u003c/li\u003e\n\u003cli\u003eTaylor, M. (2010). Clusters: A mesmerizing mantra. \u003cem\u003eTijdschrift voor economische en Sociale Geografie\u003c/em\u003e, \u003cem\u003e101\u003c/em\u003e(3), 276-286.\u003c/li\u003e\n\u003cli\u003eTer Wal, A. L., \u0026amp; Boschma, R. A. (2009). Applying social network analysis in economic geography: framing some key analytic issues. \u003cem\u003eThe Annals of Regional Science\u003c/em\u003e, \u003cem\u003e43\u003c/em\u003e, 739-756.\u003c/li\u003e\n\u003cli\u003eTrejo-Nieto, A. (2024). Place-based strategies for sustainable and inclusive regional development in the south of Mexico. \u003cem\u003eReview of Regional Research\u003c/em\u003e, \u003cem\u003e44\u003c/em\u003e(3), 1-25.\u003c/li\u003e\n\u003cli\u003eUnger, K., \u0026amp; Chico, R. (2004). La industria automotriz en tres regiones de M\u0026eacute;xico. Un an\u0026aacute;lisis de cl\u0026uacute;steres. \u003cem\u003eEl Trimestre Econ\u0026oacute;mico\u003c/em\u003e, 909-941.\u003c/li\u003e\n\u003cli\u003eUnger, Kurt. 2003. \u0026ldquo;Los Cl\u0026uacute;steres industriales en M\u0026eacute;xico: especializaciones regionales y la pol\u0026iacute;tica industrial\u0026rdquo;. CIDE, Economics Division. Working Paper 278. https://repositorio-digital.cide.edu//handle/11651/6050.\u003c/li\u003e\n\u003cli\u003eVernon, J. M. (1992). Mexico\u0026apos;s Accession to the GATT: A Catalyst at Odds with the Outcome.\u003cem\u003e Mary\u0026apos;s LJ\u003c/em\u003e, \u003cem\u003e24\u003c/em\u003e, 717.\u003c/li\u003e\n\u003cli\u003eWang, H., \u0026amp; Garduno‐Rivera, R. (2023). The economic impact of trade openness on Mexico\u0026apos;s indigenous peoples. \u003cem\u003eRegional Science Policy \u0026amp; Practice\u003c/em\u003e, \u003cem\u003e15\u003c/em\u003e(7), 1612-1625.\u003c/li\u003e\n\u003cli\u003eWennberg, K., \u0026amp; Lindqvist, G. (2010). The effect of clusters on the survival and performance of new firms. \u003cem\u003eSmall Business Economics\u003c/em\u003e, \u003cem\u003e34\u003c/em\u003e, 221-241.\u003c/li\u003e\n\u003cli\u003eWorld Bank, Data. \u0026quot;The Gini index\u0026quot; http://datos.bancomundial.org/indicador/SI.POV.GINI (Accessed: May 2nd, 2025).\u003c/li\u003e\n\u003cli\u003eZheng, S., \u0026amp; Du, R. (2020). How does urban agglomeration integration promote entrepreneurship in China? Evidence from regional human capital spillovers and market integration. \u003cem\u003eCities\u003c/em\u003e, \u003cem\u003e97\u003c/em\u003e, 102529.\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Footnotes","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003e \"The Gini index measures to what extent the distribution of income among individuals or households within an economy deviates from a perfectly equitable distribution. It measures the area between the Lorenz curve and a hypothetical line of absolute equality; it is expressed as a percentage of the maximum area under the line. Thus, a 0 Gini index represents a perfect balance, while an index of 100 represents a perfect inequality\u0026rdquo;. See more at World Bank: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://datos.bancomundial.org/indicador/SI.POV.GINI\u003c/span\u003e\u003cspan address=\"http://datos.bancomundial.org/indicador/SI.POV.GINI\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003e Around \u003cspan\u003e$\u003c/span\u003e555\u0026nbsp;million MXN, or USD 27.8\u0026nbsp;million.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"letters-in-spatial-and-resource-sciences","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"lsrs","sideBox":"Learn more about [Letters in Spatial and Resource Sciences](http://link.springer.com/journal/12076)","snPcode":"12076","submissionUrl":"https://submission.springernature.com/new-submission/12076/3","title":"Letters in Spatial and Resource Sciences","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Clusters, Employment Growth, Regional Economic Growth, Productivity","lastPublishedDoi":"10.21203/rs.3.rs-7227009/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7227009/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThis paper examines the impact of industrial clusters on employment growth in Mexican municipalities from 1986 to 2019 using a spatial panel data model. Contrary to cluster theory’s expectation of positive employment effects, our results show that greater industrial specialization, measured by the Krugman Specialization Index, reduces employment overall, with the strongest declines in manufacturing, services, and mining. Commerce is less affected. By contrast, higher education, access to electricity, and foreign direct investment significantly promote employment growth, with education exerting the largest positive effect. Spatial spillovers further reveal that employment dynamics extend beyond municipal boundaries. These findings suggest that while clustering may enhance productivity and innovation, excessive specialization undermines job creation. Policymakers should therefore emphasize diversification strategies and adapt cluster initiatives to local economic conditions when designing employment-focused regional development policies.\u003c/p\u003e\n\u003cp\u003eJEL codes: O12; R11; O47\u003c/p\u003e","manuscriptTitle":"Clusters and Employment Growth in Mexico: Insights from a Panel Data Model","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-12 10:09:07","doi":"10.21203/rs.3.rs-7227009/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-08-10T11:53:02+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-08-05T09:40:08+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-08-05T09:40:06+00:00","index":"","fulltext":""},{"type":"submitted","content":"Letters in Spatial and Resource Sciences","date":"2025-07-27T14:56:22+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"letters-in-spatial-and-resource-sciences","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"lsrs","sideBox":"Learn more about [Letters in Spatial and Resource Sciences](http://link.springer.com/journal/12076)","snPcode":"12076","submissionUrl":"https://submission.springernature.com/new-submission/12076/3","title":"Letters in Spatial and Resource Sciences","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"ad9ddfe9-25c2-4afc-89e1-826f174aa500","owner":[],"postedDate":"September 12th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-04-29T11:24:14+00:00","versionOfRecord":[],"versionCreatedAt":"2025-09-12 10:09:07","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7227009","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7227009","identity":"rs-7227009","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}