Trade Liberalization and Regional Dynamics∗ Rafael Dix-Carneiro† Duke University and BREAD

Brian K. Kovak‡ Carnegie Mellon University and NBER

February 2017

Abstract We study the evolution of trade liberalization’s effects on local labor markets, following Brazil’s early 1990s trade liberalization. Regions that initially specialized in industries facing larger tariff cuts experienced prolonged declines in formal sector employment and earnings relative to other regions. The impact of tariff changes on regional earnings 20 years after liberalization was three times the size of the effect 10 years after liberalization. These findings are robust to a variety of alternative specifications and to controlling for a wide array of postliberalization shocks. The pattern of increasing effects on regional earnings is not consistent with conventional spatial equilibrium models, which predict that effect magnitudes decline over time due to spatial arbitrage. We investigate potential mechanisms, finding empirical support for a mechanism involving imperfect interregional labor mobility and dynamics in labor demand, driven by slow capital adjustment and agglomeration economies. This mechanism gradually amplifies the initial labor demand shock resulting from liberalization. We show that the mechanism explains the slow adjustment path of regional earnings and quantitatively accounts for the magnitude of the long-run effects.

∗

This project was supported by an Early Career Research Grant from the W.E. Upjohn Institute for Employment Research. The authors would like to thank Peter Arcidiacono, Penny Goldberg, Gustavo Gonzaga, Walker Hanlon, Guilherme Hirata, Joe Hotz, Joan Monras, Enrico Moretti, Nina Pavcnik, Mine Senses, Juan Carlos Suarez Serrato, Lowell Taylor, Gabriel Ulyssea, Eric Verhoogen, and participants at various conferences and seminars for helpful comments. Ekaterina Roshchina provided excellent research assistance. Dix-Carneiro thanks Daniel Lederman and the Office of the Chief Economist for Latin America and the Caribbean at the World Bank for warmly hosting him while part of the paper was written. Remaining errors are our own. † [email protected] ‡ [email protected]

1

Trade Liberalization and Regional Dynamics

1

Dix-Carneiro and Kovak

Introduction

Prominent theories of international trade typically focus on long-run equilibria in which the reallocation of resources across economic activities is achieved without frictions. These models have traditionally given little attention to the adjustment process in transitioning from one equilibrium to another, creating a tension between academic economists advocating trade liberalization and policy makers concerned with the labor market outcomes of workers employed in contracting sectors or firms (Salem and Benedetto 2013, Hollweg, Lederman, Rojas and Ruppert Bulmer 2014). While theory tends to focus on long-run outcomes, empirical studies of the labor market effects of trade liberalization typically emphasize short- or medium-run effects. Frequently changing designs of cross-sectional household surveys forced researchers to focus on relatively short intervals to guarantee consistency over the periods analyzed (Goldberg and Pavcnik 2007). Thus, although many countries underwent major trade liberalization episodes throughout the 1980s and 1990s (e.g. Brazil, Mexico, and India, among others), we still know very little about the evolution of the effects of these policy reforms on labor markets. We fill this gap in the literature by using 25 years of administrative employment data from Brazil to study the dynamics of local labor market adjustment following the country’s trade liberalization in the early 1990s. We exploit variation in the tariff declines across industries and variation in the industry mix of local employment across Brazilian regions to measure changes in local labor demand induced by liberalization. We then compare formal employment and earnings growth between regions facing larger and smaller tariff declines, while controlling for pre-existing trends in these outcomes.1 This approach allows us to observe the ensuing regional labor market dynamics for 20 years following the beginning of liberalization. The results are striking. We find large and steadily increasing effects on regional earnings and employment. Regions facing larger tariff declines experience deteriorating formal labor market outcomes compared to other regions. These effects grow for more than a decade before beginning to level off in the late 2000s. This pattern is robust to a wide variety of alternative measurement strategies, weighting schemes, and controls for pre-existing trends across multiple decades. The growing effects are not driven by post-liberalization shocks such as later tariff changes, exchange rate movements, privatization, or the commodity price boom of the 2000s. We conclude that liberalization’s effects on regional earnings and employment grew substantially over time. This pattern challenges the conventional wisdom that labor mobility gradually arbitrages away spatial differences in local labor market outcomes (Blanchard and Katz 1992, Bound and Holzer 2000). If that were the case, one would observe declining regional effects of liberalization on 1

In this paper, we focus on formal labor market outcomes, covering workers with a signed work card providing access to the benefits and labor protections afforded by the legal employment system. See Dix-Carneiro and Kovak (2015b) for an analysis covering the informal labor market, which includes the self-employed and employees without signed work cards.

2

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

earnings, such that the short- and medium-run estimates of trade exposure in prior work would be an upper bound on the long-run effects.2 Instead, we document increasing effects of liberalization; the effect on regional earnings 20 years after the start of liberalization is more than 3 times larger than the effect after 10 years. Liberalization’s long-run effects on regional labor market outcomes are therefore much larger than initially supposed. This surprising finding leads us to evaluate a variety of alternative mechanisms that might account for the growth in liberalization’s effects on regional earnings. The evidence rules out mechanisms based on slow urban decline (as in Glaeser and Gyourko 2005), changing worker composition (based on observable or unobservable characteristics), and slow responses of trade quantities to tariff changes. Instead, we find strong evidence for a mechanism involving imperfect interregional labor mobility and dynamics in labor demand, driven by a combination of slow regional capital adjustment and agglomeration economies. Intuitively, as capital slowly reallocates away from harder-hit regions, workers’ marginal products steadily fall. Similarly, with agglomeration economies, a negative local labor demand shock decreases local economic activity, reducing regional productivity, and further decreasing the marginal product of labor. We find minimal responses of regional workingage population to regional tariff declines, suggesting imperfect worker mobility across regions. In this setting, dynamic labor demand, driven by slow capital adjustment or agglomeration economies, can rationalize the steady relative decline in wages in regions facing larger tariff declines. We present a wide array of evidence in support of this mechanism. Regions facing larger tariff reductions experience steady declines in the number of formal establishments and declining average establishment size, suggesting that capital stocks slowly reallocate away from negatively affected regions. Capital investment shifts away from these regions on impact, with immediate declines in establishment entry and job creation. In contrast, establishment exit and job destruction increase slowly over time, consistent with firm owners waiting for installed capital to depreciate before contracting or closing down regional establishments. Supporting the presence of agglomeration economies, we show that employment in a given industry × region pair falls more when other industries in the region face larger tariff cuts. Regional labor market equilibrium would suggest the opposite in the absence of agglomeration economies (Helm 2016). Finally, we extend the specific-factors model of regional economies in Kovak (2013) to incorporate slow factor adjustment and agglomeration economies. Within this framework, we show that a proxy for regional capital adjustment quantitatively accounts for a substantial portion of the long-run earnings effects that we observe. Standard magnitude agglomeration economies and perfect long-run capital mobility quantitatively account for all of the long-run earnings effects. In contrast to the other alternative mechanisms that we considered, this dynamic labor demand mechanism is both qualitatively and 2

Papers documenting short- and medium-run regional effects of trade exposure include Autor, Dorn and Hanson (2013), Costa, Garred and Pessoa (2016), Edmonds, Pavcnik and Topalova (2010), Hakobyan and McLaren (forthcoming), Hasan, Mitra and Ural (2006), Hasan, Mitra, Ranjan and Ahsan (2012), Kondo (2014), Kovak (2013), McCaig (2011), Topalova (2010), and many others.

3

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

quantitatively consistent with the observed earnings responses. Only recently have researchers begun measuring reallocation costs and the dynamics of labor market adjustment following trade policy reforms. The papers in this literature calibrate or estimate small open economy models in order to study their quantitative implications for welfare and the implied transitional dynamics when facing hypothetical changes in trade policy.3 We contribute to this literature by describing empirical transitional dynamics in response to a real-world trade liberalization. We document the importance of dynamic labor demand in the evolution of liberalization’s effects on labor markets and suggest that incorporating this mechanism into quantitative models is an important task for future work. A growing empirical literature finds substantial differences in the effects of trade exposure across local labor markets with different industry structures.4 Each of these papers measures the effects of trade shocks over a fixed time window of 7 to 10 years. We contribute to this literature by placing the single-year estimates from prior work into a dynamic context, documenting the evolution of trade liberalization’s regional effects over time. This exercise is possible because our data provide complete yearly coverage of the formal labor market, even at fine geographic levels, and because Brazilian liberalization represents a discrete shock occurring during a well-defined time period. A similar analysis would be much more challenging when studying shocks that continually evolve over time, such as Chinese export growth, because it is difficult to separate the influence of dynamics from the effects of newly arriving shocks.5 Our paper proceeds as follows. Section 2 describes the history and institutional context of Brazil’s early 1990s trade liberalization. Section 3 describes the data sources, local labor market definition, and empirical approach. Section 4 presents i) our main results for liberalization’s effects on regional earnings and employment, ii) a wide array of robustness tests, and iii) analyses ruling out the influence of post-liberalization shocks. Section 5 evaluates potential mechanisms that could account for the growing earnings effects of liberalization. Section 6 concludes.

2

Trade Liberalization in Brazil

Brazil’s trade liberalization in the early 1990s provides an excellent setting in which to study the labor market effects of changes in trade policy. The unilateral trade liberalization involved large declines in average trade barriers and featured substantial variation in tariff cuts across industries. Many papers have examined the labor market effects of trade liberalization in the Brazilian context 3 Examples include Artu¸c, Chaudhuri and McLaren (2010), Caliendo, Dvorkin and Parro (2015), Co¸sar (2013), Dix-Carneiro (2014), Kambourov (2009), Traiberman (2016), and many others. 4 See footnote 2 for citations. 5 Autor, Dorn, Hanson and Song (2014) discuss this point in their study of the effects of Chinese export growth across U.S. industries.

4

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

to take advantage of this variation.6 In the late 1980s and early 1990s, Brazil ended nearly one hundred years of extremely high trade barriers imposed as part of an import substituting industrialization policy.7 In 1987, nominal tariffs were high, but the degree of protection actually experienced by a given industry often deviated substantially from the nominal tariff rate due to i) a variety of non-tariff barriers such as suspended import licenses for many goods and ii) a system of “special customs regimes” that lowered or removed tariffs for many transactions (Kume, Piani and de Souza 2003).8 In 1988 and 1989, in an effort to increase transparency in trade policy, the government reduced tariff redundancy by cutting nominal tariffs and eliminating certain special regimes and trade-related taxes, but there was no effect on the level of protection faced by Brazilian producers (Kume 1990). Liberalization effectively began in March 1990, when the newly elected administration of President Collor suddenly and unexpectedly abolished the list of suspended import licenses and removed nearly all of the remaining special customs regimes (Kume et al. 2003). These policies were replaced by a set of import tariffs providing the same protective structure, as measured by the gap between prices internal and external to Brazil, in a process known as tariffication (tarifica¸c˜ ao) (de Carvalho, Jr. 1992). In some industries, this process required modest tariff increases to account for the lost protection from abolishing import bans.9 Although these changes did not substantially affect the protective structure, they left tariffs as the main instrument of trade policy, such that tariff levels in 1990 and later provide an accurate measure of protection. The main phase of trade liberalization occurred between 1990 and 1995, with a gradual reduction in import tariffs culminating with the introduction of Mercosur. Tariffs fell from an average of 30.5 percent to 12.8 percent, and remained relatively stable thereafter.10 Along with this large average decline came substantial heterogeneity in tariff cuts across industries, with some industries such as agriculture and mining facing small tariff changes, and others such as apparel and rubber facing declines of more than 30 percentage points. We measure liberalization using long-differences in the log of one plus the tariff rate from 1990 to 1995, shown in Figure 1. During this time period, tariffs accurately measure the degree of protection faced by Brazilian producers, and tariff changes from 6

Examples include Arbache, Dickerson and Green (2004), Goldberg and Pavcnik (2003), Gonzaga, Filho and Terra (2006), Kovak (2013), Krishna, Poole and Senses (2014), Menezes-Filho and Muendler (2011), Pavcnik, Blom, Goldberg and Schady (2004), Paz (2014), Schor (2004), and Soares and Hirata (2016) among many others. 7 Although Brazil was a founding signatory of the General Agreement on Tariffs and Trade (GATT) in 1947, it maintained high trade barriers through an exemption in Article XVIII Section B, granted to developing countries facing balance of payments problems (Abreu 2004). Hence, trade policy changes during the period under study were unilateral. 8 These policies were imposed quite extensively. In January 1987, 38 percent of individual tariff lines were subject to suspended import licenses, which effectively banned imports of the goods in question (Authors’ calculations from Bulletin International des Douanes no.6 v.11 supplement 2). In 1987, 74 percent of imports were subject to a special customs regime (de Carvalho, Jr. 1992). 9 Appendix Figure A1 shows the time series of tariffs. Note the tariff increases in 1990 for the auto and electronic equipment industries. 10 Simple averages of tariff rates across N´ıvel 50 industries, as reported in Kume et al. (2003). See Appendix A.1 for details on tariff data.

5

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

1990 to 1995 reflect the full extent of liberalization faced by each industry. We do not rely on the timing of tariff cuts between 1990 and 1995, because this timing was chosen to maintain support for the liberalization plan, cutting tariffs on intermediate inputs earlier and consumer goods later (Kume et al. 2003). As discussed below, along with regional differences in industry mix, the cross-industry variation in tariff cuts provides the identifying variation in our analysis. Following the argument in Goldberg and Pavcnik (2005), we note that the tariff cuts were nearly perfectly correlated with the preliberalization tariff levels (correlation coefficient = -0.90). These initial tariff levels reflected a protective structure initially imposed in 1957 (Kume et al. 2003), decades before liberalization. This feature left little scope for political economy concerns that might otherwise have driven systematic endogeneity of tariff cuts to counterfactual industry performance. To check for any remaining spurious correlation between tariff cuts and other steadily evolving industry factors, we regress pre-liberalization (1980-1991) changes in industry employment and average monthly earnings on the 1990-1995 tariff reductions, with detailed results reported in Appendix B.1. We attempted a variety of alternative specifications and emphasize that the results should be interpreted with care, as they include only 20 tradable-industry observations. Most specifications exhibit no statistically significant relationship, but heteroskedasticity-weighted specifications place heavy weight on agriculture and find a positive relationship. Agriculture was initially the least protected industry, and it experienced approximately no tariff reduction. It also had declining wages and employment before liberalization, driving the positive relationship with tariff reductions. Consistent with earlier work, when omitting agriculture, tariff cuts are unrelated to pre-liberalization earnings trends (Krishna, Poole and Senses 2011). Given these varying results, we include controls for pre-liberalization outcome trends in all of the analyses presented below, to account for any potential spurious correlation. Consistent with the notion that the tariff changes were exogenous in practice, these pre-trend controls have little influence on the vast majority of our results.

3 3.1

Data and Empirical Approach Data

Our main data source for regional labor market outcomes is the Rela¸c˜ ao Anual de Informa¸c˜ oes Sociais (RAIS), spanning the period from 1986 to 2010. This is an administrative dataset assembled yearly by the Brazilian Ministry of Labor, providing a high quality census of the Brazilian formal labor market (De Negri, de Castro, de Souza and Arbache 2001, Saboia and Tolipan 1985). Accurate information in RAIS is required for workers to receive payments from several government benefits programs, and firms face fines for failure to report, so both agents have an incentive to provide accurate information. RAIS includes nearly all formally employed workers, meaning those 6

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

with a signed work card providing them access to the benefits and labor protections afforded by the legal employment system. It omits interns, domestic workers, and other minor employment categories, along with those without signed work cards, including the self-employed.11 These data have recently been used by Dix-Carneiro (2014), Helpman, Itskhoki, Muendler and Redding (forthcoming), Krishna et al. (2014), Lopes de Melo (2013), and Menezes-Filho and Muendler (2011), though these papers utilize shorter panels. The data consist of job records including worker and establishment identifiers, allowing us to track workers and establishments over time. We utilize the establishment’s geographic location (municipality) and industry, and worker-level information including gender, age, education (9 categories), and December earnings.12 These data have various advantages relative to previous work on the effects of trade on local labor markets. First, relative to Kovak (2013) and Autor et al. (2013), we can analyze the dynamics of adjustment to the trade liberalization shock, as RAIS data are available every year. Second, RAIS is a census rather than a sample, so it is representative at fine geographic levels.13 Third, a rich set of labor market outcomes can be analyzed with such data, including how liberalization affected job creation and job destruction rates, the number of active establishments, and the establishment size distribution. Fourth, the ability to follow workers over time allows us to control for both observable and unobservable worker characteristics. As is typically the case in administrative employment datasets, the limitation of RAIS is a lack of information on workers who are not formally employed, making it impossible to tell whether a worker is out of the labor force, unemployed, informally employed, or self-employed. This is important in the Brazilian context, with informality rates often exceeding 50 percent of all employed workers during our sample period.14 When we need information on individuals who are not formally employed, or information before 1986, we supplement the analysis using the decennial Brazilian Demographic Census, covering 1970-2010. While these data provide much smaller samples and do not permit following individuals over time, they cover the entire population, including the informally employed, unemployed, and those outside the labor force.15 When possible, we corroborate results from RAIS using the Demographic Census, finding very similar results across datasets. Throughout the analysis, we limit our sample to include working-age individuals, aged 18-64. When studying employed individuals, we omit those working in public administration and those 11

See Appendix B.2 for summary statistics on the informal sector, and Dix-Carneiro and Kovak (2015b) for analyses covering the informal labor market. 12 RAIS reports earnings for December and average monthly earnings during employed months in the reference year. We use December earnings to ensure that our results are not influenced by seasonal variation or month-to-month inflation. See Appendix Section A.2 for more detail on the RAIS database. 13 The National Household Survey (Pesquisa Nacional por Amostra de Domic´ılios - PNAD) would be a natural alternative data source for a yearly analysis, but it only provides geographic information at the state level, does not allow one to follow individual workers over time, and provides a much smaller sample. 14 Authors’ calculations using Brazilian Demographic Census. 15 See Appendix A.3 for more detail on the Demographic Census data and Dix-Carneiro and Kovak (2015b) for analyses covering the informal labor market.

7

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

without valid information on industry of employment.16 To analyze outcomes by local labor market, we must define the boundaries of each market. We use the “microregion” definition of the Brazilian Statistical Agency (IBGE), which groups together economically integrated contiguous municipalities (counties) with similar geographic and productive characteristics (IBGE 2002), closely paralleling an intuitive notion of a local labor market. When necessary, we combine microregions whose boundaries changed during our sample period, to ensure that we consistently define local labor markets over time. This process leads to a set of 475 consistently identifiable local labor markets for analyses falling within 1986-2010 and 405 markets for analyses using data from 1980 and earlier.17

3.2

Empirical Approach

Our empirical analysis follows the literature on the regional effects of trade by comparing the evolution of labor market outcomes in regions facing large tariff declines to those in regions facing smaller tariff declines. Intuitively, regions experience larger declines in labor demand when their most important industries face larger liberalization-induced price declines (Topalova 2007). Kovak (2013) presents a specific-factors model of regional economies that captures this intuition (a generalization of this setup appears below in Section 5.4.1). In this model, the regional labor demand shock resulting from liberalization is X

βri Pˆi ,

where

i

λri ϕ1i βri ≡ P 1 , j λrj ϕj

(1)

hats represent proportional changes, r indexes regions, i indexes industries, ϕi is the cost share of non-labor factors, and λri is the share of regional labor initially allocated to tradable industry i. Pˆi is the liberalization-induced price change facing industry i, and (1) is a weighted average of these price changes across tradable industries, with more weight on industries capturing larger shares of initial regional employment.18 Thus, although all regions face the same vector of liberalization16

We exclude public administration because the labor market in this field operates quite differently from the rest of the market. This choice has no substantive effect on any of our results. 17 This geographic classification is a slightly aggregated version of the one in Kovak (2013), accounting for additional boundary changes during the longer sample period. Related papers define local markets based on commuting patterns (e.g. Autor et al. (2013)). Our local market definition performs well based on this standard as well – only 3.4 and 4.6 percent of individuals lived and worked in different markets in 2000 and 2010, respectively. The main regional definition is shown in Figure 2. The analysis omits 11 microregions, shown with a cross-hatched pattern the figure. These include i) Manaus, which was part of a Free Trade Area and hence not subject to tariff cuts during liberalization; ii) the microregions that constitute the state of Tocantins, which was created in 1988 and hence not consistently identifiable throughout our sample period; and iii) a few other municipalities that are omitted from RAIS in the 1980s. The inclusion or exclusion of these regions when possible has no substantive effect on the results. We also implemented the main analyses using a more aggregate local labor market definition, “mesoregions” defined by IBGE, and results are nearly identical. 18 Following Kovak (2013), we drop the nontradable sector, based on the assumption that nontradable prices move with tradable prices. We confirm this assumption by calculating a measure of local nontradables prices in Section

8

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

induced price changes, differences in the regional industry mix generate regional variation in labor demand shocks. We operationalize this shock measure by defining the “regional tariff reduction” (RT R), which utilizes only liberalization-induced variation in prices, replacing Pˆi with the change in log of one plus the tariff rate. RT Rr = −

X

βri d ln(1 + τi )

(2)

i

τi is the tariff rate in industry i, and d represents the long difference from 1990-1995, the period of Brazilian trade liberalization. We calculate tariff changes using data from Kume et al. (2003), λri using the 1991 Census, and ϕi using 1990 National Accounts data from IBGE.19 Together, these allow us to calculate the weights, βri . Note that RT Rr is more positive in regions facing larger tariff reductions, which simplifies the interpretation of our results, since nearly all regions faced tariff declines during liberalization. Figure 2 maps the spatial variation in RT Rr . Regions facing larger tariff reductions are presented as lighter and yellower, while regions facing smaller cuts are shown as darker and bluer. The region at the 10th percentile faced a tariff reduction of 0.2 percentage points, while the region at the 90th percentile faced a 10.7 percentage point decline. Hence, in interpreting the regression estimates below, we compare regions whose values of RT Rr differ by 10 percentage points, closely approximating the 90-10 gap of 10.5 percentage points. Note that there is substantial variation in the tariff shocks even among local labor markets within the same state. As we include state fixed effects in our analyses, these within-state differences provide the identifying variation in our study.20 We use the following specification to compare the evolution of labor market outcomes in regions facing large tariff reductions to those in regions facing smaller tariff declines. yrt − yr,1991 = θt RT Rr + αst + γt (yr,1990 − yr,1986 ) + rt

(3)

We estimate this equation separately for each year t ∈ [1992, 2010], as reflected by the t subscripts. yrt is the value of a regional outcome such as earnings or employment, θt is the cumulative effect of liberalization on outcomes by year t, αst are state fixed effects (allowed to differ across years), and (yr,1990 − yr,1986 ) is a pre-liberalization trend in the outcome variable. While the change in outcome varies with the year t under consideration, the liberalization shock, RT Rr , does not. 4.1. 19

See Appendix A.4 for more detail on the construction of (2). We use the Census to calculate λri because the Census allows for a more detailed industry definition than what is available in RAIS (see Appendix A.1) and because the Census allows us to calculate weights that are representative of overall employment, rather than just formal employment. That said, shocks using formal employment weights yield very similar results (Appendix Table B6, Panel D). 20 A regression of RT Rr on state fixed effects yields an R2 of 0.36; i.e. 64% of the variation in RT Rr is not explained by state effects. Our main conclusions are unaffected by the inclusion or exclusion of state fixed effects.

9

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

Instead, it always reflects the regional measure of tariff reductions during liberalization, from 1990 to 1995. Using this strategy, each year’s θt represents one point on the empirical impulse response function describing the cumulative local effects of liberalization as of each post-liberalization year. This methodology captures only relative effects across regions, as does the rest of the literature examining the regional or sectoral effects of trade. We use 1991 as the base year for outcome changes, and include state fixed effects to account for any state-specific policies that might commonly affect outcomes for all regions in the same state, such as state-specific minimum wages, introduced in 2002 (Neri and Moura 2006).21 We control for pre-liberalization changes in outcomes (yr,1990 − yr,1986 ) to address the possibility of confounding pre-existing trends, and consider longer pre-liberalization trends as a robustness test. For our main outcomes, we present results with and without state fixed effects and pre-trends, with little effect on the coefficients of interest. Since many of our dependent variables are themselves estimates, we weight regressions based on the inverse of their standard error to account for heteroskedasticity. We also cluster standard errors at the mesoregion level to account for potential spatial correlation in outcomes across neighboring regions. To consistently estimate θt , rt must be uncorrelated with RT Rr , conditional on the state fixed effects and outcome pre-trend. For this identification assumption to be violated, there would need to be an omitted variable that i) drives wage or employment growth across regions within a state and ii) is correlated with RT Rr but iii) is not captured by pre-liberalization outcome trends. While such a feature is unlikely to exist, in Section 4.2 we confirm that our results are robust to a wide variety of potential confounders and alternative specifications. Our empirical approach is similar to prior studies examining the local effects of trade liberalization, but we make two important contributions to that literature. First, the RAIS data allow us to calculate changes in regional outcomes in each year following liberalization. We trace out the dynamic regional response to liberalization as it evolves over time, rather than observing liberalization’s local effect in only one post-shock period, as in the prior literature (e.g. Topalova (2007), Autor et al. (2013), or Kovak (2013)). The RAIS data also allow us to control for pre-liberalization trends that might otherwise confound the analysis. Second, we study a discrete, well-defined trade policy shock that was complete by 1995. This contrasts with Autor et al. (2014), who use U.S. panel data to study the effects of growing trade with China. They emphasize that the continuously evolving nature of Chinese trade confounds their ability to study the dynamic response to a trade shock at any given point in time. 21

Using 1991 as the base year allows us to take advantage of more detailed industry information in the 1991 Census when calculating the industry distribution of regional employment (λri ), and makes our results comparable with Kovak (2013).

10

Trade Liberalization and Regional Dynamics

4

Dix-Carneiro and Kovak

Results

4.1

Main Findings

We begin by examining the effects of liberalization on formal sector earnings and employment in local labor markets. First, we calculate “regional earnings premia,” which reflect average log monthly earnings for workers in a given region, controlling for the composition of the regional workforce.22 For each year t, we regress log December earnings for worker j on flexible controls for age, sex, and education (Xjt ); industry fixed effects (φit ); and region fixed effects (µrt ).23 ln(earnjrit ) = Xjt Γt + φit + µrt + ejrit

(4)

We estimate this equation separately for each year t ∈ [1991, 2010], allowing the regression coefficients (Γt ) and fixed effects (φit and µrt ) to differ across years. The region fixed effect estimates from these regressions, µ ˆrt , represent the regional log earnings premia for the relevant year. By estimating these regressions separately in each year, we allow for changes in the regional composition of workers (X) and changes in the returns to worker characteristics (Γ) over time.24 This approach ensures that our earnings estimates are not driven by changes in observable worker composition, changing discrimination, changes in the returns to schooling, or any other changes in the returns to observable characteristics that operate at the national level. Our dependent variable when studying earnings is then the change in regional log earnings premium from 1991 to each subsequent year, 1992 to 2010. Table 1 presents summary statistics for this and other main dependent variables throughout the paper. Table 2 shows the results of estimating (3) for regional formal sector log earnings premia and formal log employment. All estimates for the coefficient on RT Rr are negative, indicating that regions facing larger tariff reductions experience relative declines in earnings or employment. Consider Panel A, which presents liberalization’s effect on regional earnings. Columns (1) to (3) examine changes in earnings from 1991 to 2000, while columns (4) to (6) examine changes from 1991 to 2010, such that the effects cumulate over time. Columns (2) and (4) add state fixed effects, and columns (3) and (6) add pre-trend controls for the change in the regional outcome from 1986 to 1990. The coefficient estimate of -0.529 in column (3) indicates that a region facing a 10 percentage point larger tariff reduction (approximately the 90-10 gap in RT Rr ) experienced a 5.29 percentage point larger proportional decline (or smaller increase) in formal earnings from 1991 to 2000. This 22

Estimating the regional earnings premia for each year separately from the effects of liberalization on regional earnings reduces the computational demands relative to pooling across years and estimating both steps jointly. 23 We use monthly earnings rather than hourly wages because RAIS only provides hours from 1994 onward. Census results using hourly wages are similar. 24 Appendix B.3 presents the coefficient estimates from (4) for 1991, 2000, and 2010. In Section 5.2, we control for observable and unobservable worker heterogeneity by pooling across years and including individual fixed effects. The results are very similar.

11

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

magnitude is similar in size to the corresponding estimate in Kovak (2013) (-0.439), which used a different data source (Census of Population) and covers all workers rather than restricting attention to the formally employed. The estimate of -1.594 in column (6) indicates that the gap in earnings growth expanded to 15.94 percentage points by 2010.25 This increase in liberalization’s effect on earnings from 2000 to 2010 is a striking feature of Table 2. It indicates that the divergent earnings growth in regions facing different tariff reductions continued well beyond the liberalization period. Figure 3 confirms this pattern by plotting the coefficients on RT Rr (θt ) for each year. The points for 2000 and 2010 correspond to the RT Rr coefficients in columns (3) and (6) of Table 2. The vertical lines indicate that liberalization began in 1991 and was complete by 1995. We present coefficient estimates for 1992-94, but these should be interpreted with care, as liberalization was still ongoing.26 The local earnings effects of liberalization appear just after liberalization and steadily grow for more than a decade, before leveling off in the late 2000s, a pattern that is very robust to details of the specification.27 Figure 3 also shows pre-liberalization coefficients, in which the dependent variable is the change in regional earnings premium from 1986 to the year listed on the x-axis, and the independent variable is RT Rr . If anything, the relative earnings declines in regions facing larger tariff reductions represent a reversal of the pre-liberalization trend. Recall that all post-liberalization results control for pre-liberalization trends, as shown in (3). It is likely that the prices of local nontradable goods change in response to the regional shocks to the prices of traded goods (Kovak 2013, Monte 2016). If this is the case, the relative decline in nominal earnings in regions facing larger tariff reductions may be partly offset by declines in the local price index. To empirically evaluate this possibility, we construct local price indexes using housing rents information in the Census, following the approach of Moretti (2013).28 Only the 1991 and 2010 Censuses included rent questions, so we can only calculate the change in rental prices for 1991-2010. Our local price index uses consumption weights from the Brazilian Consumer Price Index system (IPC) and accounts for the fact that the prices of non-housing nontradables tend to move with housing prices. See Appendix A.5 for details on constructing the index. We then calculate the change in log real earnings as the change in log nominal earnings minus the change in log local price level. Panel B of Table 2 shows the effect of regional tariff reductions on the change in real regional earnings for 1991-2010. The effect on real earnings in column (6) is smaller than the 25 Appendix B.4 presents an alternative research design at the industry × region level finding similar growth in the regional earnings effects of liberalization and confirming the importance of cross-industry regional equilibrium in driving the main earnings effects discussed here. 26 However, the tariff cuts were almost fully implemented by 1993, so these early coefficients are still informative regarding liberalization’s short-run effects. When regressing RT Rr on an alternate version measuring tariff changes from 1990-93, the R2 is 0.93. 27 See Section 4.2 for a variety of robustness tests. Appendix B.5 shows the underlying scatterplots, confirming our choice of linear estimating equation and showing that the results are not driven by outliers. Appendix B.6 shows that the same pattern appears when estimating formal earnings or formal hourly wages using Census data. 28 As in the U.S., the Brazilian government does not produce local price indexes outside a few large cities.

12

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

effect on nominal earnings by about 21 percent. This difference confirms that regional nontradable prices move with tradable prices, falling more in places facing larger tariff reductions. However, the long-run effects of liberalization on real regional earnings are still large and statistically significant. Table 2 Panel C and Figure 4 both examine liberalization’s effects on regional log formal employment. The year 2000 estimate of -3.533 shows that a region facing a 10 percentage point larger tariff reduction experienced a 35.3 percentage point larger proportional decline (smaller increase) in formal employment from 1991 to 2000. As with earnings, the employment effect grew substantially from 2000 to 2010, indicating that employment growth continued to diverge for regions facing different regional tariff changes. Most of this divergence was complete by 2004, after which the estimates level off.29 Note that since Table 2 and Figure 4 examine formal employment, there are two channels through which formal employment might decline in regions facing more negative shocks. Formally employed workers may migrate away from negatively affected places to more favorably affected places, or existing residents of the region may shift out of formal employment and into nonemployment or informal employment. Table 3 rules out the interregional migration mechanism, showing that a region’s working-age population did not respond to RT Rr .30 We measure workingage population using Census data, so we can observe individuals outside formal employment, and control for 1980-1991 and 1970-1980 population pre-trends. None of the population estimates are significantly different from zero, and the point estimates with extensive pre-trend controls (columns (3) and (6)) are quantitatively small. These results suggest that workers losing formal employment in harder hit regions did not leave the region, but transitioned out of formal employment. Using Census data on informal workers, Panel A of Table 4 confirms that in regions facing larger tariff reductions informal employment increased relative to the national average. For example, the estimate 1.196 in column (6) implies that on average a region facing a 10 percentage point larger tariff reduction experienced an 11.96 percentage point larger increase in informal employment by 2010. Rather than migrating away, many workers who lose formal employment in negatively affected regions appear to transition into the informal sector in the same region.31 Panel B of Table 4 implements a similar exercise for regional earnings premia in the informal sector. Somewhat unexpectedly, there is no significant relationship between regional tariff reductions and earnings in the informal sector. A potential explanation for the lack of effect on informal wages is that 29 To assess the scale of our long-run estimates, consider Dix-Carneiro (2014), which studies a very similar setting with slow adjustment of labor across Brazilian industries rather than regions. After estimating the model’s parameters using RAIS data, he simulates the economy’s response to a price shock when capital is mobile across industries (see his Figures 4 and 6). The long-run wage elasticity in the adversely affected sector (High-Tech Manufacturing) is -1.56. This is exceedingly close to our 2010 earnings estimate of -1.594. The long-run employment elasticity in Dix-Carneiro (2014) is -3.2. Although this is somewhat smaller than our 2010 employment estimate of -4.663, the two effects are similar in magnitude, suggesting that our findings are reasonable in the context of this type of model. 30 Similarly, Autor et al. (2013) find little evidence for population responses to trade shocks in the U.S. 31 See Dix-Carneiro and Kovak (2015b) for a more extensive discussion of the various margins of labor market adjustment following Brazilian liberalization.

13

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

consumers in harder-hit regions experience declining incomes and shift toward lower-priced, lowerquality goods produced in the informal sector, offsetting wage declines for informally employed workers.32 Together, these results are quite surprising, particularly compared to the conventional wisdom from the literature studying local labor demand shocks. The standard framework predicts initially large wage effects of local labor demand shocks, as labor supply is approximately fixed in the very short run, after which employment adjustment arbitrages away spatial wage differences, and observed wage effects fall in magnitude (Blanchard and Katz 1992, Bound and Holzer 2000). This mechanism is consistent with the steadily growing employment effects in Figure 4, but is at odds with the growing earnings effects in Figure 3. It predicts large negative coefficients shortly after liberalization, but then declining magnitude effects as arbitrage partly equalizes earnings growth across regions. Even in the absence of equalizing migration, as shown in Table 3, one would expect constant effects over time. Instead, we find continuing divergence in earnings growth for 14 years following the end of liberalization, with earnings growth in regions facing larger tariff reductions lagging further and further behind other regions. This pattern means that the local labor market effects of trade estimated in prior work for a single post-liberalization year actually understate the longer run effects. The remainder of the paper focuses on examining and explaining this surprising result.

4.2

Robustness

We first establish that the steadily growing earnings effects are robust to alternative measurement and specification choices and that they were not driven by confounding effects from other shocks to Brazilian local labor markets. Detailed analyses appear in Appendix Sections B.7–B.9, and we summarize the results here. Appendix B.7 shows that the growing earnings estimates are robust to alternative pre-trend controls, RT Rr shock measures, earnings premium measures, and weighting. We use Census data to construct longer pre-liberalization earnings trends, from 1970-1980 and from 1980-1991, and control for these alongside the 1986-1990 RAIS pre-liberalization trends present in our main specification.33 We construct alternative RT Rr measures, i) using industry weights, λri , reflecting only formal employment, ii) using effective rates of protection, which account for the effects of tariffs on inputs 32

Burstein, Eichenbaum and Rebelo (2005) show that lower quality goods gain market share in recessions, while McKenzie and Schargrodsky (2011) make a similar argument in the context of the 2002 economic crisis in Argentina. While there is little direct evidence on the relative quality of goods produced by formal and informal firms, it is well known that informal firms are significantly smaller than formal firms (LaPorta and Schleifer 2014, Meghir, Narita and Robin 2015, Ulyssea 2014), and Kugler and Verhoogen (2011) show that larger firms produce higher quality goods than small firms, on average. Moreover, LaPorta and Schleifer (2008) show that informal firms use lower quality inputs and speculate that they produce lower quality outputs as a result. 33 Because 1991 is the base year for our post-liberalization earnings growth outcome, 1980-1991 pre-liberalization trends are subject to mechanical endogeneity. We resolve this problem by calculating an alternative earnings growth measure with 1992 as the base year. See Panel C of Table B6.

14

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

and outputs for each industry, and iii) including a zero price change for the nontradable sector. We also construct alternative earnings premium measures. The first is calculated without controlling for industry fixed effects, maintaining national industry-level earnings variation in the regional earnings premia.34 The second measure simply uses mean log earnings, without controlling for any worker characteristics. Finally, we present results weighting regions equally or weighting by the region’s 1991 formal employment. In all cases, our main results are confirmed, finding steady growth in liberalization’s effects on regional earnings. The employment effects are similarly robust to these alternatives. Although our findings are robust to these specification and measurement changes, the effects of liberalization could appear to grow over time because of correlated shocks occurring after trade liberalization. To explain the smooth growth of the effects in Figures 3 and 4, such confounders would need to affect industries or regions similarly to liberalization and would need to grow steadily over time or occur quite regularly. Although these circumstances are unlikely, in Appendix B.8 we construct controls for a wide variety of salient economic shocks in the post-liberalization period, demonstrating that they cannot account for the growing earnings effects. If tariff changes after 1995 were correlated with those occurring during liberalization (199095), they might drive the apparently increasing effects of liberalization, although this is unlikely since post-1995 tariff changes were very modest. We calculate post-liberalization regional tariff reductions as in (2), but use tariff reductions between 1995 and each year t > 1995, and include these post-liberalization tariff reductions as additional controls alongside RT Rr . Other potential confounders are the Brazilian Real devaluations that occurred in 1999 and 2002. If these exchange rate movements affected industries differently, they might have been correlated with tariff changes during liberalization. We construct industry-specific real exchange rates as import- or exportweighted averages of real exchange rates between Brazil and its trading partners. We then take the change in log real exchange rate from 1990 to year t > 1995, and calculate regional shocks using weighted averages as in (2). There was also a substantial wave of privatization during our sample period. We address privatization by controlling flexibly for the 1995 share of regional employment at state-owned firms or the change in this share from 1995 to t. Controlling for each of these postliberalization shocks has little effect on the earnings results, which continue to exhibit substantial post-liberalization growth in all cases. The global commodity price boom of the late 2000s is another potential post-liberalization confounder that might explain our growing earnings results, particularly since agricultural products faced the most positive tariff change during liberalization. In Appendix B.8.4, we provide extensive evidence ruling out this possibility. First, the timing of the commodity price boom does not 34

By omitting the industry fixed effects, these regional earnings measures include both direct industry effects and local general equilibrium effects. As shown in Appendix B.7, the associated estimates are only a bit larger than the main results, indicating that local equilibrium effects account for the majority of the overall effects of liberalization on regional earnings.

15

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

correspond to the timing of our effects. Commodity prices were flat or declining between 1991 and 2003, during which our earnings and employment results grew substantially. Commodity prices then grew sharply after 2004, when our results began to level off.35 We show that the substantial growth in earnings effects remains when i) dropping regions most exposed to commodity price growth, by restricting the region sample to include only those with below-median or bottomquartile employment shares in agriculture and mining or ii) when restricting our regional earnings measure only to workers in manufacturing. Finally, we use three approaches to directly control for the regional effects of the commodity price boom. We control for the share of workers in agriculture and mining and for changes in regional commodity prices using the measure introduced by Ad˜ ao (2015). We also use more detailed commodity price data from the IMF Primary Commodity Price Series to construct similar regional controls for commodity price changes. Finally, we control for China’s effects on commodity markets using the import and export quantity measures and instruments from Costa et al. (2016).36 All of these controls have little influence on the observed earnings effects of RT Rr . As a final set of robustness tests, Appendix B.9 presents results when splitting the sample by tradable and nontradable sector and by skill. We find growing earnings effects in all of these subsamples. This pattern is particularly noteworthy for the nontradable sector, as it confirms that regional labor market equilibrium transmits the effects of liberalization from the tradable sector to the nontradable sector, as predicted in the model of Kovak (2013), which is the basis for the RT Rr shock. The earnings effects for more skilled workers are a bit larger than those for less skilled workers, while the employment effects are larger for less skilled workers. However, these results should be interpreted with care, as the RT Rr shocks are derived from a model with a single type of labor.37 Together, the results in this section demonstrate the robustness of our main findings to alternative measures and estimation approaches and rule out a wide variety of salient post-liberalization shocks as potential confounders. We conclude that the earnings and employment profiles shown in Figures 3 and 4 reflect growing causal effects of liberalization over time. In the next section, we consider a variety of potential mechanisms that could drive this growth in liberalization’s effects on local labor market outcomes. 35

A similar argument applies to Bustos, Caprettini and Ponticelli (2016), who study the effects of genetically modified crops in Brazil. Genetically modified crops were outlawed before 2003 and only permanently authorized in 2005, so this channel cannot explain the substantial growth in earnings effects before 2005. 36 Special thanks to Rodrigo Ad˜ ao for providing commodity price data and code, and to Francisco Costa for providing the shock and instrument measures from Costa et al. (2016). 37 For a more general model with two skill types, see Dix-Carneiro and Kovak (2015a).

16

Trade Liberalization and Regional Dynamics

5

Dix-Carneiro and Kovak

Mechanisms

As mentioned above, the conventional model of local labor markets predicts large effects of liberalization just after the tariff change and smaller effects as labor reallocation arbitrages away spatial differences in earnings growth. Our findings contradict this prediction, instead exhibiting increasing differences in earnings growth for 15 years after liberalization between regions facing larger and smaller tariff reductions. In this section, we consider a variety of potential mechanisms that might explain these growing earnings effects, finding strong empirical support for mechanisms involving imperfect interregional labor mobility and dynamic labor demand, particularly slow capital adjustment and agglomeration economies.

5.1

Urban Decline

Glaeser and Gyourko (2005) and Notowidigdo (2013) present models of urban decline in which the slow depreciation of housing stocks drives slow adjustment in local labor markets facing permanent negative labor demand shocks. In their models, the price of housing falls sharply in depressed markets, incentivizing individuals to remain in the city in spite of nominal earnings losses following the demand decline. As housing slowly depreciates, this incentive dissipates, so population and therefore employment steadily decline. This mechanism could therefore rationalize the slowly growing employment effects we document in Figure 4. However, as in the conventional model of local labor markets, this mechanism predicts the opposite of what we find for earnings in Figure 3. Although wages fall on impact in regions facing negative shocks, they recover slowly over time as workers leave the market due to housing depreciation.38 Moreover, the mechanism depends on declining population in cities facing negative shocks. In Brazil, overall population growth was large enough during our sample period that out of 475 local labor markets, only 11 experienced population decline between 1991 and 2000, and only 6 did so between 1991 and 2010.39 Table 3 also finds no response of local working-age population to RT Rr . Thus, while the slow housing depreciation mechanism is quite relevant for rust-belt cities in the U.S., it does not appear to apply in the Brazilian context.

5.2

Changing Composition of Worker Unobservables

Liberalization might cause average earnings to slowly decline in regions facing larger tariff reductions relative to other regions because of worker selection. Higher-earning workers may be more likely 38

Note that Glaeser and Gyourko (2005) do not model a production side and instead directly shock wages or amenities. However, a simple extension of their model to include labor market equilibrium would have the features cited here, as in Notowidigdo (2013). Glaeser and Gyourko (2005) also argue that local average wages will decline over time in negatively shocked markets because the most productive workers have the strongest incentive to leave. As shown in Section 5.2, since we control for worker characteristics when calculating regional earnings premia, selection effects of this kind are not driving our results. 39 Authors’ calculations using Census data.

17

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

to leave the formal labor market in harder-hit regions, and this selective worker reallocation may increase over time. Although we flexibly control for detailed worker characteristics including age, sex, and education when calculating regional earnings premia in our main specifications, worker composition may also adjust along unobservable dimensions. To examine this possibility, we calculate alternative earnings premia, pooling the RAIS data across years and controlling for worker-level fixed effects, which capture time-invariant worker characteristics, including unobservables.40 We implement this procedure in two ways. First, we use a straightforward worker fixed-effects regression. ln(earnjairt ) = αj + ψa + φit + µrt + jairt ,

(5)

where αj are worker fixed effects, ψa are age effects (indicators for falling within each age bin shown in Table B3), and φit are time-varying industry effects. We then calculate the change in log regional earnings premium using the regional earnings estimates, µ ˆrt , and examine their response to RT Rr . As shown in Panel B of Table 5, when controlling for worker unobservables in this fashion, the earnings effects continue to grow over time. A limitation of (5) is that it restricts the returns to worker characteristics to be constant over time. Since the returns to observable characteristics change substantially over time (see Appendix B.3), we allow for time-varying returns (δt ) to worker characteristics (αj ) in the following specification. ln(earnjairt ) = δt αj + ψa + φit + µrt + jairt .

(6)

δt can vary arbitrarily over time, but does so identically for all workers. This restriction distinguishes δt αj from worker × time fixed effects, which would absorb all variation in the data. We estimate (6) using the iterative algorithm described by Arcidiacono, Foster, Goodpaster and Kinsler (2012) and calculate standard errors using the wild bootstrap method suggested by Davidson and MacKinnon (2006), with 500 iterations. Panel C of Table 5 presents earnings estimates using the resulting regional earnings premia. The growth in earnings effects remains, and the results from the more flexible earnings premium specification in Panel C are quantitatively very close to the baseline specification in Panel A. These findings rule out worker selection as a mechanism driving the observed growth in the earnings effects of liberalization.

5.3

Slow Response of Imports or Exports

Although trade liberalization was complete by 1995, it is possible that trade quantities were slow to respond to the sharp change in trade policy, perhaps because of difficulty in forming new trade links 40

For computationally tractability, we draw a 3 percent random sample of all valid individual IDs that appear in RAIS with a positive earnings observation between 1986 and 2010. This procedure yields 450 microregions with formally employed workers earning labor income in December for all years in our sample.

18

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

with firms abroad. Prices faced by Brazilian producers may evolve slowly in response to tariff cuts if import quantities respond slowly to liberalization. If so, the slow evolution of imports in response to the tariff cuts could potentially explain the slow growth in the effects of liberalization over time. To examine this possibility, we i) study the relationship between regional tariff reductions and trade quantity measures to determine whether such a slow trade response occurred in practice and ii) control for changes in trade quantities to see whether they mediate the relationship between tariff changes and earnings. We follow Autor et al. (2013) by constructing changes in imports and exports per worker for each industry from 1991 to each subsequent year, using Comtrade data.41 We then form regional weighted averages of these changes in trade flows, weighting by the industry’s initial share of regional employment. See Appendix A.6 for details on the construction of these measures. We first examine the effect of regional tariff reductions on these regional measures of import, export, and net export growth, looking for evidence of slow growth in trade quantities that might drive the slow growth in earnings effects. We do so using the trade growth measures as dependent variables in (3). Figure 5 plots the effects of RT Rr on each trade flow measure.42 First, consider the effects on regional imports (blue circles). As expected, regions facing larger tariff reductions experienced larger increases in the regional import measure. These import increases occurred immediately after liberalization, with large positive coefficients already present in 1995. Because we measure trade flows in $100,000 units, the 1995 coefficient of 0.144 implies that a region facing a 10 percentage point larger tariff reduction experienced a $1,440 larger increase in imports per worker. These import effects actually decrease on average until 2003 (coefficient estimate = 0.070), in sharp contrast to the earnings effects, which grew to more than two-thirds of their long-run level during the same time period. After 2003, the import effects increase, but this coincides with a leveling-off in the earnings and employment effects. This timing is inconsistent with slow import growth driving our results. The sign of the export effects (red triangles) is positive, indicating that industries experiencing larger export increases were on average located in the same regions as industries facing larger tariff reductions.43 This effect works against the hypothesis that slow trade quantity growth drove relative earnings declines in these regions. After 2003, both the import and export effects grow quite substantially, following the overall trends in Brazilian imports and exports. Note, however, that the relationship between RT Rr and net exports (green diamonds) falls from 2005 to 2010, again a time period with substantial growth in the earnings effects. Overall, the evolution of import and export quantities is not consistent with the hypothesis that slow trade quantity growth explains our results. 41

Appendix B.10 shows results for an ad-hoc alternative functional form using the change in log trade, yielding the same conclusions. 42 In Figure 5 we do not have pre-liberalization trends for trade flows because Comtrade data for Brazil begin in 1989. 43 The positive sign for the export effect is not driven by any particular industry or industries and is robust to dropping agriculture and/or natural-resource industries.

19

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

To confirm this point, we include controls for regional import and export growth when examining the effect of RT Rr on regional earnings premia. If the growing earnings effects remain when including these controls, we can be confident that a different mechanism is at play. We examine the relationship between earnings growth and RT Rr , as in (3), including controls for the regional growth in imports (RegImprt ) and exports (RegExprt ) from 1990 to year t. yrt − yr,1991 = θt RT Rr + β1 RegImprt + β2 RegExprt + αst + γt (yr,1990 − yr,1986 ) + rt

(7)

The import and export coefficients, β1 and β2 , are constant over time, allowing us to test whether the slow evolution of trade flows explains the evolution of earnings growth (since RegImprt and RegExprt change over time, unlike RT Rr ). Panel B of Table 6 shows that the effect of RT Rr on regional earnings still grows steadily over time when controlling for changes in regional imports and exports, implying that slow trade quantity growth is not driving the relationship between the tariff reductions and earnings. A remaining concern is that if regional imports and exports are endogenous to regional earnings growth, then the coefficients on RT Rr will be biased along with the trade flow coefficients, invalidating the analysis just described. Panels C and D address this issue following the strategy of Autor et al. (2013), instrumenting for Brazilian trade flows using trade flows for other countries.44 We consider instruments based on the combination of Argentina, Chile, Colombia, Paraguay, Peru, and Uruguay (“Latin America”) and on Colombia alone, which liberalized during the same time period as Brazil and imposed similar tariff cuts across industries (Paz 2014). In each case, we measure imports and exports between these countries and the rest of the world, excluding Brazil.45 Panels C and D show the results. In both cases, the effects of RT Rr continue to grow over time, with a similar magnitude to the main results, shown in Panel A. These and the preceding results in this section rule out slow import or export responses as the mechanism driving the slowly growing earnings effects.

5.4

Dynamic Labor Demand

A remaining potential mechanism driving the growing effects of liberalization on earnings and employment involves dynamics in labor demand. If labor is imperfectly mobile across regions and an initial labor demand shock is followed by a dynamic process that amplifies the shock’s effects over time, one will observe the growing regional earnings and employment effects we document. We consider two potential sources of these dynamics: agglomeration economies (e.g. Kline and Moretti 2014) and slow adjustment of capital stocks (e.g. Dix-Carneiro 2014). As we will show, both appear to play important roles in explaining our findings. 44

We also include regional measures of commodity price growth from Ad˜ ao (2015) in the set of instruments. Due to Comtrade data availability, the changes in trade flows for Latin America are calculated from 1994 to each subsequent year and those for Colombia alone are calculated from 1991 to each subsequent year. 45

20

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

5.4.1

Evidence on the Importance of Dynamic Labor Demand

To study these mechanisms and formalize our argument, we generalize the specific-factors model in Kovak (2013) to include agglomeration economies and slow adjustment of labor and capital. We focus on the formal economy, consisting of many regions, indexed by r, which may produce goods in many industries, indexed by i. Production in each industry uses Cobb-Douglas technology with constant returns to scale and three inputs: labor, a fixed factor, and capital. Formal labor, Lr , is assumed to be perfectly mobile between industries within a region. The fixed factor, Tri , is usable only in its respective region and industry and is fixed over time. This factor represents inputs such as natural resources, land, or very slowly depreciating infrastructure and capital that are effectively fixed over the time horizons we consider. Capital, Kri , is also usable only in its respective region and industry but may change slowly over time through depreciation and investment decisions.46 Output of industry i in region r is  ϕi 1−ϕi 1−ζi Yri = Ari Lri Triζi Kri

(8)

where ϕi , ζi ∈ (0, 1). Goods and factor markets are perfectly competitive, and producers face exogenous prices Pi , common across regions and fixed by world prices and tariffs. To allow for the possibility of agglomeration economies, we allow productivity, Ari , to vary with the amount of local economic activity. We also allow for factor adjustment by letting Lr and Kri change over time. Recall that changes in Lr primarily reflect workers entering or leaving the formal workforce rather than other channels such as interregional migration, as shown in Table 3. We assume that changes in Kri reflect depreciation and firms’ investment decisions rather than physical mobility via secondary markets for installed capital. As shown in Appendix C, factor market clearing, zero profits, and cost minimization imply the following equilibrium relationship, in which hats represent proportional changes. ! w ˆr =

X i

βri Pˆi +

X

βri Aˆri − δr

i

ˆr − L

X

ˆ ri λri (1 − ζi )K

(9)

i

λri ϕ1i where βri ≡ P 1 >0 j λrj ϕj

1 1 > 0. j λrj ϕj

and δr ≡ P

w ˆr is the proportional change in the regional wage, and λri is the initial share of regional employment in industry i. This is an equilibrium relationship because the factor supplies and productivity levels may respond endogenously to the liberalization shock reflected by Pˆi . 46

We separate fixed factors and variable capital for two reasons. First, our research design is based on regional differences in industry mix, which are driven by fixed factors. Second, including fixed factors in each region ensures that all regions maintain some economic activity even when faced with very negative shocks. Hence, this formulation is common in the literature on agglomeration economies (e.g. Helm (2016) and Kline and Moretti (2014)).

21

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

As a thought exercise, suppose we were to hold productivity and factor supplies constant (Aˆri = ˆr = K ˆ ri = 0). In that case, the wage change equals the simple weighted average price shock in L (1). In this restricted model, there is no scope for dynamic effects of liberalization, and one would observe a substantial wage effect of liberalization on impact, with no changes thereafter. More realistically, if productivity or factor supplies evolve over time in response to the liberalizationinduced price shocks, then the effects of liberalization on regional wages can change over time as well. First, we consider factor supply responses. Imagine that only regional labor supply responds ˆ ri = 0. Immediately following liberalization, wages to liberalization, while maintaining Aˆri = K decline more in regions facing larger tariff reductions, and formal employment falls more in these regions, as in Figure 4. Equation (9) shows that this change in employment partly offsets the wage losses experienced on impact, since δr > 0. If employment adjusts slowly, then the observed wage effects of liberalization get smaller over time. In other words, with labor adjustment only, the model reflects the conventional prediction that liberalization’s effects on local wages decline over time. If we allow both regional employment and regional capital stocks to vary in response to liberalization, complex patterns can emerge, depending on the relative speed of labor and capital adjustment. For example, if regional labor is held fixed and capital stocks contract more in regions facing larger tariff declines (as we show below), the marginal product of labor will fall, and relative wages will decline even further in harder hit regions, as seen in Figure 3. More generally, the model can qualitatively rationalize growing earnings effects of liberalization if the labor supply elasticity is finite and capital adjusts more quickly than labor. Now consider changes in productivity, Aˆri . We assume that these result from agglomeration economies in which changes in the amount of local economic activity drive changes in the productivity of local firms. There is little agreement on the specific source of agglomeration economies, with various papers arguing that they result from changes in population, overall employment, or employment in particular industries (Melo, Graham and Noland 2009).47 For agglomeration economies to be relevant in our context, we must observe effects of regional tariff reductions on at least one of these agglomeration sources. In Table 3 and Appendix B.11, we show that neither working-age population nor overall employment (sum of formal and informal) respond substantially to RT Rr , while Figure 4 shows that liberalization substantially affected formal employment. For agglomeration economies to be relevant in our context, agglomeration must apply to regional formal employment, since other potential sources of agglomeration do not significantly respond to liberalization. This is plausible, as labor market pooling and knowledge spillovers are more likely to apply in formal employment than in informal employment, which disproportionately includes agricultural production. In this case, a negative labor demand shock decreases wages on impact, which 47

Many papers argue that population or employment density is the relevant quantity, but since we utilize regions with fixed boundaries, the change in log population or employment density is identical to the change in log population or employment level.

22

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

endogenously decreases formal employment and therefore decreases regional productivity through agglomeration economies. As shown in (9), this productivity decline amplifies the wage decline from the initial shock, leading to further reductions in local formal employment and productivity, etc. If this amplification occurs slowly over time, perhaps due to slow labor supply responses or slow responses of productivity to formal employment (Kline and Moretti 2014), then the observed effects of liberalization may also grow over time. Therefore, given imperfect labor mobility across regions, both capital adjustment and agglomeration economies could qualitatively explain the earnings and employment patterns in Figures 3 and 4. To provide evidence for the relevance of dynamic labor demand, we rearrange (9) to infer the labor demand shifts needed to rationalize the changes in earnings with the observed regional tariff reductions and changes in formal employment. For consistency with the agglomeration literature, we assume identical factor cost shares across industries (ϕi = ϕ ∀i and ζi = ζ ∀i, which implies δr = ϕ).48 The economy-wide value of ϕ is 0.544 (see Appendix A.4), and we discuss the value of ζ in Section 5.4.3. X i

βri Aˆri + ϕ(1 − ζ)

X

ˆ ri = w λri K ˆr −

i

X

ˆr βri Pˆi + ϕL

(10)

i

|

{z

observed

}

The left hand side of (10) captures the overall shifts in labor demand resulting from agglomeration economies and capital adjustment, which we can measure as a residual using the observable quantities on the right hand side. We measure w ˆr as the change in regional earnings premium, P ˆ ˆ − i βri Pi as RT Rr , and Lr as the change in regional formal employment. Figure 6 (solid blue circles) shows the relationship between this inferred labor demand measure and regional tariff reductions in each year following the start of liberalization. We can infer that labor demand steadily declined in regions facing larger tariff reductions and that these dynamics were complete by the late 2000s. Given this evidence for dynamic labor demand in general, we examine evidence for the two specific sources of dynamics: agglomeration economies and slow capital adjustment. 5.4.2

Evidence for Agglomeration Economies and Capital Adjustment

To examine these mechanisms in more detail, we follow the literature by imposing additional longrun assumptions that allow us to compare our results to prior work and to quantify the roles of agglomeration and slow capital adjustment. We assume a constant elasticity long-run agglomeration function.49 ˆr, Aˆri = κL

κ≥0

(11)

48 When assuming identical factor cost shares across industries, our production function is identical to those in Kline and Moretti (2014) and Helm (2016). Hanlon and Miscio (2016) use a slightly different Cobb-Douglas production function, but also assume constant cost shares across industries. 49 Kline and Moretti (2014) provide empirical support for a constant agglomeration elasticity.

23

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

Table 3 shows that working-age population does not substantially respond to liberalization, indicating that the main margin of labor supply adjustment is workers’ choice of whether to pursue formal employment within a given region. Table 4 shows informal sector earnings do not substantially ˆ r ) depend upon respond to liberalization. Therefore, we assume that changes in formal labor (L changes in the regional formal wage (w ˆr ), and assume a constant elasticity long-run local formal labor supply function. ˆr = 1 w L ˆr , η

η≥0

(12)

ˆr = R ˆ ∀r, where R is the price of Finally, we assume perfectly mobile capital in the long run (R capital).50 We take 2010 to be the long run (20 years following the start of liberalization), consistent with the flat earnings and employment responses by the late 2000s. Imposing these assumptions on the model yields the following expressions for the long-run regional wage change and the change in employment in a given region × industry combination (derived in Appendix C). w ˆr =

X η ϕ(1 − ζ)η ˆ R βri Pˆi − η[1 − ϕ(1 − ζ)] − κ + ϕζ η[1 − ϕ(1 − ζ)] − κ + ϕζ

(13)

i

η[1 − ϕ(1 − ζ)] − κ X ϕ(1 − ζ) ˆ ˆ ri = 1 Pˆi − 1 · βri Pˆi − R L ϕζ ϕζ η[1 − ϕ(1 − ζ)] − κ + ϕζ η[1 − ϕ(1 − ζ)] − κ + ϕζ

(14)

i

We test for the presence of agglomeration economies using the change in employment in each region × industry combination, following an approach similar to that of Helm (2016). As shown in (14), in the absence of agglomeration (κ = 0), holding fixed an industry’s own price decline, larger regional tariff reductions increase local industry employment. Intuitively, if other industries in the same region face larger tariff cuts, more laborers will locally transition into the reference industry in equilibrium. However, in a setting with agglomeration economies (κ > 0), price reductions in other industries in the same region reduce the local productivity of the reference industry. If agglomeration forces are strong enough, larger regional tariff reductions can reduce local industry employment conditional on the industry’s own price change. We therefore estimate the following specification. ˆ ri = γ0 + γ1 Pˆi + γ2 RT Rr + ri L

(15)

ˆ which does not vary across This expression is the reduced form of (14). γ0 captures the term for R, industries or regions, and γ2 < 0 implies the presence of agglomeration economies.51 We measure ˆ ri using changes from 1991 to 2010 to capture long-run adjustment. We control for industry price L changes either directly using tariff reductions (−d ln(1 + τi )), or with industry fixed effects. Since the nontradable sector does not directly experience a tariff change, we use RT Rr to measure its 50

Perfect long-run capital mobility is a standard assumption in this literature (Hanlon and Miscio 2016, Helm 2016, Kline and Moretti 2014). P η[1−ϕ(1−ζ)]−κ 51 Recall that RT Rr ≡ − i βri Pˆi , so γ2 < 0 implies η[1−ϕ(1−ζ)]−κ+ϕζ < 0 in (14), which in turn implies κ > 0.

24

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

price change, following the arguments in (Kovak 2013). The results of estimating (15) appear in Table 7. In all cases, the coefficient on RT Rr is negative and highly significant; an industry’s local employment actually falls when other industries in the same region face larger tariff reductions, implying the presence of agglomeration economies. This finding is robust to including state fixed effects and outcome pre-trends, to using direct industry price change controls or industry fixed effects, and to restricting attention to tradable industries.52 We also find evidence for slow capital adjustment. Although regional capital stock measures are unavailable, we can observe changes in the number of formal establishments in a given region, which are likely to approximate changes in regional capital stocks.53,54 Figure 7 shows that regions facing larger tariff reductions experienced steady relative declines in the number of formal establishments, with the effect increasing most quickly in the early 2000s and leveling out later in the sample period. It is possible that capital simply reallocated from smaller exiting establishments to larger continuing establishments in harder-hit locations. If this were the case, the change in number of establishments would not be particularly informative about the change in regional capital stock. However, the decline in the number of establishments was not offset by increases in the average size of remaining establishments; if anything these establishments shrank on average. Moreover, Appendix B.13 shows that larger tariff declines drove increases in exit rates throughout the establishment size distribution. These results strongly support the interpretation that trade shocks induced a gradual reallocation of capital away from harder hit locations. To reinforce this conclusion, we present evidence on the margins of capital adjustment. We expect investment to respond immediately following liberalization, with new investment directed toward more favorable markets and away from markets facing larger tariff reductions. In contrast, depreciation takes time to erode the capital stock in a negatively affected region. We confirm these patterns using measures of regional establishment entry and exit and job creation and destruction. We measure cumulative entry, exit, job creation, and job destruction by observing changes from 1991 to each subsequent year, and calculate each measure following Davis and Haltiwanger (1990).55 We then examine the relationship between the log of each measure and RT Rr . Figure 8 reports the results for entry and exit, and Figure 9 shows the results for job creation and de52

Because we use RT Rr to measure the industry-specific price change for nontradable industries, it is not possible to separately identify the effects of industry-specific and regional tariff reductions for nontradable industries alone. 53 It is not possible to construct regional capital stocks in Brazil during our sample period. Capital investment in manufacturing firms could in principle be constructed from the Annual Manufacturing Survey (PIA) beginning in 1996, but the Brazilian Statistical Agency (IBGE) has a strict policy against constructing PIA variables at the regional level. Moreover, with investment data beginning in 1996, we would not have credible capital stock measures until well after liberalization. Data sources covering non-manufacturing sectors also begin well after liberalization. 54 Regional capital could slowly reallocate from firms in the formal sector to firms in the informal sector, but this is unlikely, as firms in the informal sector are much less capital intensive than those in the formal sector (LaPorta and Schleifer 2014, Fajnzylber, Maloney and Montes-Rojas 2011) 55 For establishment entry and exit, the Davis and Haltiwanger (1990) measure reduces to the number of establishments that entered or exited between 1991 and year t as a share of active establishments in year t. See Appendix A.7 for details.

25

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

struction. New investment, as observed in establishment entry and job creation, falls immediately in negatively affected regions and stays low throughout the sample period. In contrast, the exit and job destruction effects grow slowly over time as existing establishments in regions facing larger tariff cuts allow their installed capital stocks to erode through depreciation, directing investment elsewhere. Together, these results support the conclusion that capital slowly reallocated away from regions facing larger tariff declines, steadily amplifying the earnings effects of liberalization. 5.4.3

Quantification

The preceding results provide evidence that both slow capital adjustment and agglomeration economies play qualitatively important roles in driving the evolution of liberalization’s effects on earnings and employment. We now investigate the extent to which these mechanisms can quantitatively explain the long-run labor market effects we observe. We begin by examining the share of the long-run change in labor demand that can be explained by regional capital adjustment. In (10), capital’s contribution to overall adjustment is given by P ˆ ri using the change in log number of regional formal ˆ ri . We proxy for P λri K ϕ(1 − ζ) i λri K i establishments (as discussed above) and measure ζ (fixed-factors’ share of non-labor input costs), using estimates of equipment, structures, and land cost shares from Valentinyi and Herrendorf (2008).56 We consider three alternative values for ζ, defining fixed factors as i) land only (ζ = 0.152), ii) land and structures (ζ = 0.545) and iii) land and half of structures (ζ = 0.349).57 Figure 6 shows the evolution of liberalization’s effect on these capital adjustment measures compared to the overall labor demand adjustment inferred from (10). Although the shapes of the capital adjustment and overall adjustment profiles are not identical, they both grow over time and have similar scales. Depending on the value of ζ, capital adjustment can account for between 47 and 88 percent of the inferred labor demand adjustment in 2010. While this is a somewhat wide range, it is clear that capital adjustment accounts for an important share of overall long-run labor demand adjustment, but that it is unlikely to account for all of the adjustment in the absence of agglomeration. To quantify the strength of agglomeration economies needed to rationalize the data, we first need to estimate the inverse labor supply elasticity, η. We do so following (12) by regressing the 1991-2010 change in log formal employment on the change in log regional earnings premium with RT Rr serving as an instrument for w ˆr . The resulting estimate of 0.363 is shown in Panel A of Table 8. Given this value for η, we estimate κ using non-linear least squares based on long-run changes in ˆ term is regional earnings in (13) or long-run changes in employment in (14). In both cases, the R captured by the intercept, and the regional weighted average price shocks are measured by RT Rr . 56

Agglomeration estimation exercises regularly require cost share calibrations along these lines, e.g. Kline and Moretti (2014). 57 While i) is likely an underestimate because there are fixed inputs other than land (e.g. heavy infrastructure), ii) is likely an overestimate, because some structures depreciate substantially at a 15 year time horizon. Thus, the intermediate value, iii), is our preferred estimate.

26

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

When estimating equation (14), we include all industries and control for industry price changes using tariff changes, as in column (3) of Table 7, though the results are nearly identical when using the alternative approaches in columns (4)-(6) of Table 7. We show estimates for each value of ζ and bootstrap the entire estimation procedure when calculating standard errors to account for potential correlation between the η and κ estimates. The resulting estimates of κ appear in Panel B of Table 8. All of the estimates are positive and fall within the range of the prior literature (Melo et al. 2009). For example, Kline and Moretti (2014) find an estimate of 0.2, which is quite close to our wage-based estimate of 0.188 for the intermediate value of ζ. The value of ζ is important in determining the magnitude of the agglomeration elasticity, which is unsurprising since Figure 6 showed that capital adjustment explains a smaller share of overall adjustment for higher values of ζ, leaving a larger role for agglomeration economies. The estimates in Table 8 and the patterns in Figure 6 show that capital adjustment and standard agglomeration economies can quantitatively account for the long-run behavior of regional earnings in response to liberalization. Along with this long-run evidence, Figures 4, 7, and 8 show that regional labor and capital evolved slowly over time following liberalization and did so in a way consistent with growing earnings effects of liberalization. In contrast to the other mechanisms that we considered, dynamic labor demand, driven by slow capital adjustment and agglomeration economies, is both qualitatively and quantitatively consistent with the earnings responses in Figure 3.

6

Conclusion

This paper documents regional labor market dynamics following the Brazilian trade liberalization of the early 1990s. Using 25 years of administrative employment data, we find large and growing effects of trade liberalization on regional formal earnings and employment. Contrary to conventional wisdom, which assumes wage-equalizing labor adjustment, the regional effects of liberalization grow for more than a decade before leveling off. This pattern is not driven by post-liberalization economic shocks and is robust to a wide variety of alternative specifications. After ruling out a number of potential mechanisms that could generate these growing effects over time, we find strong evidence in support of a combination of imperfect interregional labor mobility and dynamic labor demand, driven by slow capital adjustment and agglomeration economies. Our results have important implications for our thinking about the labor market effects of trade liberalization. A growing literature has shown in a variety of contexts that trade and trade policy have heterogeneous effects across regions in the short-run. However, most researchers, ourselves included, generally assumed that these effects would be upper bounds on the long-run effects, as labor reallocation would arbitrage away regional differences. This paper finds precisely the opposite. Short-run effects vastly underestimate the long-run effects, indicating that the costs and benefits of liberalization remain sharply unevenly distributed across geography, even twenty years after the

27

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

policy began. Our empirical results also inform a large and growing literature using structural models of the labor market to study trade-induced transitional dynamics. We document the importance of regional adjustment to trade liberalization, even in the long run, and highlight margins of adjustment that have received little attention by this line of work.58 We find evidence for slow capital adjustment in response to trade liberalization, reinforcing the message of Dix-Carneiro (2014) that jointly quantifying mobility frictions for labor and other factors such as capital is key to understanding trade adjustment.59 We also find that agglomeration economies are quantitatively important in accounting for the magnitudes of trade’s effects on regional earnings, suggesting another feature for inclusion in models examining the effects of trade shocks on labor markets.

58 59

With the exception of Caliendo et al. (2015), this literature has abstracted from geography. Artu¸c, Bet, Brambilla and Porto (2014) take an initial step in this direction.

28

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

References Abreu, Marcelo de Paiva, “The Political Economy of High Protection in Brazil before 1987,” Inter-American Development Bank Special Initiative on Trade and Integration Working Paper, 2004, (SITI-08A). Acemoglu, Daron, David Autor, David Dorn, Gordon H. Hanson, and Brendan Price, “Import Competition and the Great US Employment Sag of the 2000s,” Journal of Labor Economics, January 2016, 34 (S1), S141–S198. Ad˜ ao, Rodrigo, “Worker Heterogeneity, Wage Inequality, and International Trade: Theory and Evidence from Brazil,” Unpublished, 2015. Arbache, Jorge Saba, Andy Dickerson, and Francis Green, “Trade Liberalisation and Wages in Developing Countries,” Economic Journal, 2004, 114 (493), F73–F96. Arcidiacono, Peter, Gigi Foster, Natalie Goodpaster, and Josh Kinsler, “Estimating Spillovers Using Panel Data, with an Application to the Classroom,” Quantative Economics, 2012, 3 (3), 421–470. Artu¸ c, Erhan, G´ erman Bet, Irene Brambilla, and Guido Porto, “Trade Shocks and Factor Adjustment Frictions: Implications for Investment and Labor,” unpublished, 2014. , Shubham Chaudhuri, and John McLaren, “Trade Shocks and Labor Adjustment: A Structural Empirical Approach,” American Economic Review, 2010, 100 (3), 1008–45. Autor, David, David Dorn, and Gordon Hanson, “The China Syndrome: Local Labor Market Effects of Import Competition in the United States,” American Economic Review, 2013, 103 (6). , , , and Jae Song, “Trade Adjustment: Worker Level Evidence,” Quarterly Journal of Economics, 2014, 129 (4), 1799–1860. Blanchard, Olivier Jean and Lawrence F. Katz, “Regional Evolutions,” Brookings Papers on Economic Activity, 1992, (1), 1–75. Bound, John and Harry J. Holzer, “Demand Shifts, Population Adjustments, and Labor Market Outcomes during the 1980s,” Journal of Labor Economics, 2000, 18 (1), 20–54. Burstein, Ariel, Martin Eichenbaum, and Sergio Rebelo, “Large Devaluations and the Real Exchange Rate,” Journal of Political Economy, August 2005, 113 (4), 742–784. Bustos, Paula, Bruno Caprettini, and Jacopo Ponticelli, “Agricultural Productivity and Structural Transformation. Evidence from Brazil,” American Economic Review, 2016, 106 (6), 1320–1365. Caliendo, Lorenzo, Maximiliano Dvorkin, and Fernando Parro, “Trade and Labor Market Dynamics,” Unpublished, 2015. Co¸sar, Kerem, “Adjusting to Trade Liberalization: Reallocation and Labor Market Policies,” unpublished, 2013. Costa, Francisco J.M., Jason Garred, and Jo˜ ao Paulo Pessoa, “Winners and Losers from a Commoditiesfor-Manufactures Trade Boom,” Journal of International Economics, 2016, 102, 50–69. Davidson, Russell and James G. MacKinnon, “Bootstrap Methods in Econometrics,” in Terence C. Mills and Kerry Patterson, eds., Palgrave Handbook of Econometrics, Vol. 1, Basingstoke: Palgrave MacMillan, 2006, pp. 812–838. Davis, Steven J. and John Haltiwanger, “Gross Job Creation and Destruction: Microeconomic Evidence and Macroeconomic Implications,” NBER Macroeconomics Annual, 1990, 5. de Carvalho, Jr., M´ ario C., “Alguns Aspectos da Reforma Aduaneira Recente,” FUNCEX Texto Para Discuss˜ ao, 1992. De Negri, Jo˜ ao Alberto, Paulo Furtado de Castro, Natalia Ribeiro de Souza, and Jorge Saba Arbache, “Mercado Formal de Trabalho: Compara¸ca ˜o entre os Microdados da RAIS e da PNAD,” IPEA Texto Para Discuss˜ ao, 2001, (840). Dix-Carneiro, Rafael, “Trade Liberalization and Labor Market Dynamics,” Econometrica, 2014, 82 (3). and Brian K. Kovak, “Trade Liberalization and the Skill Premium: A Local Labor Markets Approach,” American Economic Review - Papers and Proceedings, 2015, 105 (5), 551–557. and , “Trade Reform and Regional Dynamics: Evidence From 25 Years of Brazilian Matched EmployerEmployee Data,” NBER Working Paper, 2015, (w20908).

29

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

Edmonds, Eric, Nina Pavcnik, and Petia Topalova, “Trade Adjustment and Human Capital Investment: Evidence from Indian Tariff Reform,” American Economic Journal: Applied Economics, 2010, 2 (4), 42–75. Fajnzylber, Pablo, William F. Maloney, and Gabriel V. Montes-Rojas, “Does formality improve micro-firm performance? Evidence from the Brazilian SIMPLES program,” Journal of Development Economics, 2011, 94 (2), 262 – 276. Glaeser, Edward L. and Joseph Gyourko, “Urban Decline and Durable Housing,” Journal of Political Economy, 2005, 113 (2), 345–375. Goldberg, Pinelopi and Nina Pavcnik, “Trade, Wages, and the Political Economy of Trade Protection: Evidence from the Colombian Trade Reforms,” Journal of International Economics, 2005, 66 (1), 75–105. and , “Distributional Effects of Globalization in Developing Countries,” Journal of Economic Literature, 2007, XLV, 39–82. Goldberg, Pinelopi Koujianou and Nina Pavcnik, “The response of the informal sector to trade liberalization,” Journal of Development Economics, 2003, 72 (3), 463–496. Gonzaga, Gustavo, Naercio Menezes Filho, and Cristina Terra, “Trade liberalization and the evolution of skill earnings differentials in Brazil,” Journal of International Economics, 2006, 68 (2), 345–367. Hakobyan, Shushanik and John McLaren, “Looking for Local Labor Market Effects of NAFTA,” Review of Economics and Statistics, forthcoming. Hanlon, W. Walker and Antonio Miscio, “Agglomeration: A Long-Run Panel Data Approach,” Unpublished, 2016. Hasan, Rana, Devasish Mitra, and Beyza P. Ural, “Trade Liberalization, Labor-Market Institutions, and Poverty Reduction: Evidence from Indian States,” India Policy Forum, 2006, 3. , , Priya Ranjan, and Reshad N. Ahsan, “Trade Liberalization and Unemployment: Theory and Evidence from India,” Journal of Development Economics, 2012, 97 (2). Helm, Ines, “National Industry Trade Shocks, Local Labor Markets, and Agglomeration Spillovers,” Unpublished, 2016. Helpman, Elhanan, Oleg Itskhoki, Marc-Andreas Muendler, and Steven J. Redding, “Trade and Inequality: From Theory to Estimation,” Review of Economic Studies, forthcoming. Hollweg, Claire H., Daniel Lederman, Diego Rojas, and Elizabeth Ruppert Bulmer, Sticky feet : how labor market frictions shape the impact of international trade on jobs and wages., Washington D.C.: The World Bank, 2014. IBGE, Censo Demogr´ afico 2000: Documenta¸ca ˜o das Microdados da Amostra, Instituto Brasileiro de Geografia e Estat´ıstica, 2002. , Sistema Nacional de ´ Indices de Pre¸cos ao Consumidor, Vol. 34 of S´erie Relat´ orios Metodol´ ogicos, Instituto Brasileiro de Geografia e Estat´ıstica, 2005. Kambourov, Gueorgui, “Labor Market Regulations and the Sectoral Reallocation of Workers: The Case of Trade Reforms,” Review of Economic Studies, 2009, 76 (4), 1321–1358. Kline, Patrick and Enrico Moretti, “Local Economic Development, Agglomeration Economies and the Big Push: 100 Years of Evidence from the Tennessee Valley Authority,” Quarterly Journal of Economics, 2014, 129 (1), 275–331. Kondo, Illenin O., “Trade Reforms, Foreign Competition, and Labor Market Adjustments in the U.S.,” unpublished, 2014. Kovak, Brian, “Regional Efects of Trade Reform: What is the Correct Measure of Liberalization?,” American Economic Review, 2013, 103 (5), 1960–1976. Krishna, Pravin, Jennifer P. Poole, and Mine Zeynep Senses, “Wage Effects of Trade Reform with Endogenous Worker Mobility,” NBER Working Paper, 2011, (17256). , , and , “Wage Effects of Trade Reform with Endogenous Worker Mobility,” Journal of International Economics, 2014, 93 (2), 239–252. Kugler, Maurice and Eric Verhoogen, “Prices, Plant Size, and Product Quality,” Review of Economic Studies, 2011, 79 (1), 307–339.

30

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

´ Kume, Hon´ orio, “A Pol´ıtica Tarif´ aria Brasileira no Per´ıodo 1980-88: Avalia¸ca ˜o e Reforma,” S´erie Epico, March 1990, (17). , Guida Piani, and Carlos Frederico Br´ az de Souza, “A Pol´ıtica Brasileira de Importa¸ca ˜o no Per´ıodo 1987-1998: Descri¸ca ˜o e Avalia¸ca ˜o,” in Carlos Henrique Corseuil and Honorio Kume, eds., A Abertura Comercial Brasileira nos Anos 1990: Impactos Sobre Emprego e Sal´ ario, Rio de Janiero: MTE/IPEA, 2003, chapter 1, pp. 1–37. LaPorta, Rafael and Andrei Schleifer, “The Unofficial Economy and Economic Development,” Brookings Papers on Economic Activity, 2008, 47 (1), 123–135. and

, “Informality and Development,” Journal of Economic Perspectives, Summer 2014, 28 (3), 109–126.

Lopes de Melo, Rafael, “Firm Wage Differentials and Labor Market Sorting: Reconciling Theory and Evidence,” unpublished, 2013. MacKinnon, James G., “Thirty Years of Heteroskedasticity-Robust Inference,” Queen’s Economics Department Working Paper, 2011, (1268). McCaig, Brian, “Exporting Out of Poverty: Provincial Poverty in Vietnam and US Market Access,” Journal of International Economics, 2011, 85 (1). McKenzie, David and Ernesto Schargrodsky, “Buying less, but shopping more: Changes in consumption patterns during a crisis,” Econom´ıa, 2011, 11 (2), 1–35. Meghir, Costas, Renata Narita, and Jean-Marc Robin, “Wages and Informality in Developing Countries,” American Economic Review, 2015, 105 (4), 1509–1546. Melo, Patricia C., Daniel J. Graham, and Robert B. Noland, “A Meta-Analysis of Estimates of Urban Agglomeration Economies,” Regional Science and Urban Economics, 2009, 39, 332–342. Menezes-Filho, Naercio and Marc-Andreas Muendler, “Labor Reallocation in Response to Trade Reform,” NBER Working Paper, 2011, (17372). Monte, Ferdinando, “The Local Incidence of Trade Shocks,” Unpublished, 2016. Moretti, Enrico, “Real Wage Inequality,” American Economic Journal: Applied Economics, 2013, 5 (1), 65–103. Neri, Marcelo and Rodrigo Moura, “Brasil: La institucionalidad del salario m´ınimo,” in Andr´es Marinakis and Juan Jacobo Velasco, eds., Para qu´e sirve el salario m´ınimo?, Organizaci´ on Internacional del Trabajo, 2006, pp. 105–158. Notowidigdo, Matthew J., “The Incidence of Local Labor Demand Shocks,” Unpublished, 2013. Pavcnik, Nina, Andreas Blom, Pinelopi Goldberg, and Norbert Schady, “Trade Liberalization and Industry Wage Structure: Evidence from Brazil,” World Bank Economic Review, 2004, 18 (3), 319–334. Paz, Louren¸ co, “The impacts of trade liberalization on informal labor markets: an evaluation of the Brazilian case,” Journal of International Economics, 2014, 92 (2), 330–348. Saboia, Jo˜ ao L. M. and Ricardo M. L. Tolipan, “A rela¸ca ˜o anual de informa¸co ˜es sociais (RAIS) e o mercado formal de trabalho no Brasil: uma nota,” Pesquisa e Planejamento Economico, 1985, 15 (2), 447–456. Salem, Samira and John Benedetto, “The USITC’s Roundtable on the Labor Market Effects of Trade: Discussion Summary,” Journal of International Commerce and Economics, August 2013. Schor, Adriana, “Heterogeneous productivity response to tariff reduction. Evidence from Brazilian manufacturing firms,” Journal of Development Economics, 2004, 75 (2), 373–396. Soares, Rodrigo R. and Guilherme Hirata, “Competition and the Racial Wage Gap: Testing Becker’s Model of Employer Discrimination,” IZA Discussion Paper, Februray 2016, (9764). Stock, James H and Motohiro Yogo, “Testing for Weak Instruments in Linear IV Regression,” in Donald W. K. Andrews and James H. Stock, eds., Identification and Inference for Econometric Models: Essays in Honor of Thomas Rothenberg, Cambridge University Press, 2005, pp. 80–108. Topalova, Petia, “Trade Liberalization, Poverty, and Inequality: Evidence from Indian Districts,” in Ann Harrison, ed., Globalization and Poverty, University of Chicago Press, 2007, pp. 291–336. , “Factor Immobility and Regional Impacts of Trade Liberalization: Evidence on Poverty from India,” American Economic Journal: Applied Economics, 2010, 2 (4).

31

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

Traiberman, Sharon, “Occupations and Import Competition,” Unpublished, 2016. Ulyssea, Gabriel, “Firms, Informality and Development: Theory and evidence from Brazil,” unpublished, 2014. ´ Valentinyi, Akos and Berthold Herrendorf, “Measuring Factor Income Shares at the Sectoral Level,” Review of Economic Dynamics, 2008, 11, 820–835.

32

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

0.00 -0.05 -0.10 -0.15 -0.20

Petroleum, Gas, Coal

Rubber

Petroleum Refining

Pharma., Perfumes, Detergents

Plastics

Other Manuf.

Machinery, Equipment

Auto, Transport, Vehicles

Electric, Electronic Equip.

Chemicals

Mineral Mining

Footwear, Leather

Paper, Publishing, Printing

Nonmetallic Mineral Manuf

Textiles

Food Processing

Wood, Furniture, Peat

Metals

Apparel

-0.25

Agriculture

Change in ln(1+tariff), 1990-95

Figure 1: Tariff Changes

Tariff data from Kume et al. (2003), aggregated to allow consistent industry definitions across data sources. See Appendix Table A1 for details of the industry classification. Industries sorted based on 1991 national employment (largest on the left, and smallest on the right)

33

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

Figure 2: Regional Tariff Reductions

Belém Belém

Manaus Manaus

Fortaleza Fortaleza

Recife Recife

Salvador Salvador Brasília Brasília

8% to 15% 4% to 8%

Belo Horizonte Horizonte Belo

3% to 4% 1% to 3%

São Paulo Paulo São Curitiba Curitiba

-1% to 1%

Porto Alegre Alegre Porto

mean 0.044

10 0.002

25 0.012

percentile 50 75 0.031 0.066

90 0.107

Local labor markets reflect microregions defined by IBGE, aggregated slightly to account for border changes between 1986 and 2010. Regions are colored based on the regional tariff reduction measure, RT Rr , defined in (2). Regions facing larger tariff reductions are presented as lighter and yellower, while regions facing smaller cuts are shown as darker and bluer. Dark lines represent state borders, gray lines represent consistent microregion borders, and crosshatched migroregions are omitted from the analysis. These microregions were either i) part of a Free Trade Area ii) part of the state of Tocantins and not consistently identifiable over time, or iii) not included in the RAIS sample before 1990.

34

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

Figure 3: Regional log Formal Earnings Premia - 1987-2010 1.0  

Pre-­‐liberaliza6on   (chg.  from  1986)  

Liberaliza6on                              Post-­‐liberaliza6on   (chg.  from  1991)    

0.5  

0.0   1987   1988   1989   1990   1991   1992   1993   1994   1995   1996   1997   1998   1999   2000   2001   2002   2003   2004   2005   2006   2007   2008   2009   2010  

-­‐0.5  

-­‐1.0  

-­‐1.5  

-­‐2.0  

Each point reflects an individual regression coefficient, θˆt , following (3), where the dependent variable is the change in regional log formal earnings premium and the independent variable is the regional tariff reduction (RT Rr ), defined in (2). Note that RT Rr always reflects tariff reductions from 1990-1995. For blue circles, the earnings changes are from 1991 to the year listed on the x-axis. For purple diamonds, the changes are from 1986 to the year listed. All regressions include state fixed effects, and post-liberalization regressions control for the 1986-1990 outcome pre-trend. Negative estimates imply larger earnings declines in regions facing larger tariff reductions. Vertical bars indicate that liberalization began in 1991 and was complete by 1995. Dashed lines show 95 percent confidence intervals. Standard errors adjusted for 112 mesoregion clusters.

35

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

Figure 4: Regional log Formal Employment - 1987-2010 1.5  

Pre-­‐liberaliza6on   (chg.  from  1986)  

Liberaliza6on                              Post-­‐liberaliza6on   (chg.  from  1991)    

0.5  

1987   1988   1989   1990   1991   1992   1993   1994   1995   1996   1997   1998   1999   2000   2001   2002   2003   2004   2005   2006   2007   2008   2009   2010   -­‐0.5  

-­‐1.5  

-­‐2.5  

-­‐3.5  

-­‐4.5  

-­‐5.5  

-­‐6.5  

Each point reflects an individual regression coefficient, θˆt , following (3), where the dependent variable is the change in regional log formal employment and the independent variable is the regional tariff reduction (RT Rr ), defined in (2). Note that RT Rr always reflects tariff reductions from 1990-1995. For blue circles, the employment changes are from 1991 to the year listed on the x-axis. For purple diamonds, the changes are from 1986 to the year listed. All regressions include state fixed effects, and post-liberalization regressions control for the 1986-1990 outcome pre-trend. Negative estimates imply larger employment declines in regions facing larger tariff reductions. Vertical bars indicate that liberalization began in 1991 and was complete by 1995. Dashed lines show 95 percent confidence intervals. Standard errors adjusted for 112 mesoregion clusters.

36

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

Figure 5: Regional Imports, Exports, and Net Exports Per Worker - 1991-2010

1  

Liberaliza6on                              Post-­‐liberaliza6on   (chg.  from  1991)    

0.8  

0.6  

Exports   0.4  

Imports   0.2  

Net  Exports   0   1991   1992   1993   1994   1995   1996   1997   1998   1999   2000   2001   2002   2003   2004   2005   2006   2007   2008   2009   2010  

-­‐0.2  

-­‐0.4  

Each point reflects an individual regression coefficient, θˆt , following (3), where the dependent variable is the change in regional imports per worker (blue circles), exports per worker (red triangles), or net exports per worker (green diamonds), measured in $100,000 units. The independent variable is the regional tariff reduction (RT Rr ), defined in (2). Note that RT Rr always reflects tariff reductions from 1990-1995. All regressions include state fixed effects, but do not include pre-liberalization trends due to a lack of Comtrade trade data before 1989. Positive estimates imply larger increases in trade flow per worker in regions facing larger tariff reductions. Vertical bar indicates that liberalization was complete by 1995. Dashed lines show 95 percent confidence intervals. Standard errors adjusted for 112 mesoregion clusters.

37

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

Figure 6: Inferred Adjustment and Capital Adjustment Quantification - 1992-2010 1.5  

Liberaliza6on                              Post-­‐liberaliza6on   Pre-­‐liberaliza6on   (chg.  from  1991)     (chg.  from  1986)  

1.0  

0.5  

0.0   1992   1993   1994   1995   1996   1997   1998   1999   2000   2001   2002   2003   2004   2005   2006   2007   2008   2009   2010   -­‐0.5  

Capital  (establishments)  adjustment   ζ  =  0.545    

-­‐1.0  

ζ  =  0.349    

-­‐1.5  

ζ  =  0.152     -­‐2.0  

Inferred  Adjustment  

-­‐2.5  

-­‐3.0  

Each point reflects an individual regression coefficient, θˆt , following (3). For the blue profile with solid circles, the dependent variable is the inferred labor demand shifts from agglomeration and capital adjustment, defined in (10). For the gray profiles with hollow markers, the dependent variable is capital’s contribution to overall adjustment, P using ˆ ri . the change in the number of regional formal establishments as a proxy for the change in regional capital, i λri K We present profiles for three values of ζ, specific factors’ share of non-labor inputs, based on Valentinyi and Herrendorf (2008). The independent variable is the regional tariff reduction (RT Rr ), defined in (2). Note that RT Rr always reflects tariff reductions from 1990-1995. All regressions include state fixed effects, and post-liberalization regressions control for the 1986-1990 outcome pre-trend. Negative estimates imply larger declines in residual labor demand or the number of establishments in regions facing larger tariff reductions. Vertical bar indicates that liberalization was complete by 1995. Dashed lines show 95 percent confidence intervals. Confidence intervals for capital adjustment profiles shown in Appendix B.12. Standard errors adjusted for 112 mesoregion clusters.

38

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

Figure 7: Regional log Number of Formal Establishments and log Average Formal Establishment Size (Number of Workers) - 1987-2010 2  

1  

Pre-­‐liberaliza5on   (chg.  from  1986)  

Liberaliza5on                              Post-­‐liberaliza5on   (chg.  from  1991)    

Establishments   Pretrend  

0   1987   1988   1989   1990   1991   1992   1993   1994   1995   1996   1997   1998   1999   2000   2001   2002   2003   2004   2005   2006   2007   2008   2009   2010  

-­‐1  

Establishment  Size  

Estab.  Size   Pretrend    

-­‐2  

-­‐3  

Establishments   -­‐4  

-­‐5  

Each point reflects an individual regression coefficient, θˆt , following (3), where the dependent variable is the change in regional log number of formal establishments or the change in regional log average formal establishment size. The independent variable is the regional tariff reduction (RT Rr ), defined in (2). Note that RT Rr always reflects tariff reductions from 1990-1995. For blue circles and red triangles, the changes are from 1991 to the year listed on the x-axis. For purple diamonds and orange squares, the changes are from 1986 to the year listed. All regressions include state fixed effects, and post-liberalization regressions control for the 1986-1990 outcome pre-trend. Negative estimates imply larger declines in the number of establishments or average establishment size in regions facing larger tariff reductions. Vertical bars indicate that liberalization began in 1991 and was complete by 1995. Dashed lines show 95 percent confidence intervals. Standard errors adjusted for 112 mesoregion clusters.

39

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

Figure 8: Regional log Cumulative Formal Establishment Entry and Exit - 1987-2010

5  

Pre-­‐liberaliza5on   (chg.  from  1986)  

Liberaliza5on                              Post-­‐liberaliza5on   (chg.  from  1991)    

4  

Exit   3  

2  

1  

Entry   Pretrend  

0   1987   1988   1989   1990   1991   1992   1993   1994   1995   1996   1997   1998   1999   2000   2001   2002   2003   2004   2005   2006   2007   2008   2009   2010  

-­‐1  

Exit   Pretrend  

Entry  

-­‐2  

-­‐3  

Each point reflects an individual regression coefficient, θˆt , following (3). The dependent variable is the log cumulative formal establishment entry or exit from 1991 to the year listed on the x-axis (blue circles and red triangles) or from 1986 to the year listed (purple diamonds and orange squares), calculated as in Davis and Haltiwanger (1990). The independent variable is the regional tariff reduction (RT Rr ), defined in (2). Note that RT Rr always reflects tariff reductions from 1990-1995. All regressions include state fixed effects, and post-liberalization regressions control for log cumulative establishment entry or exit during 1986-1990. Positive exit estimates and negative entry estimates imply larger rates of exit and smaller rates of entry in regions facing larger tariff reductions. Vertical bars indicate that liberalization began in 1991 and was complete by 1995. Dashed lines show 95 percent confidence intervals. Standard errors adjusted for 112 mesoregion clusters.

40

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

Figure 9: Regional log Cumulative Job Creation and Destruction - 1987-2010

5  

Pre-­‐liberaliza5on   (chg.  from  1986)  

Liberaliza5on                              Post-­‐liberaliza5on   (chg.  from  1991)    

4  

Job  Destruc)on   3  

2  

1  

Job  Crea)on   Pretrend   0   1987   1988   1989   1990   1991   1992   1993   1994   1995   1996   1997   1998   1999   2000   2001   2002   2003   2004   2005   2006   2007   2008   2009   2010   -­‐1  

-­‐2  

-­‐3  

Job  Crea)on    

Job  Destruc)on   Pretrend  

-­‐4  

Each point reflects an individual regression coefficient, θˆt , following (3). The dependent variable is the log cumulative job creation or destruction rate from 1991 to the year listed on the x-axis (blue circles and red triangles) or from 1986 to the year listed (purple diamonds and orange squares), calculated as in Davis and Haltiwanger (1990). The independent variable is the regional tariff reduction (RT Rr ), defined in (2). Note that RT Rr always reflects tariff reductions from 1990-1995. All regressions include state fixed effects, and post-liberalization regressions control for log cumulative job creation or destruction during 1986-1990. Positive job destruction estimates and negative job creation estimates imply larger rates of job destruction and smaller rates of job creation in regions facing larger tariff reductions. Vertical bars indicate that liberalization began in 1991 and was complete by 1995. Dashed lines show 95 percent confidence intervals. Standard errors adjusted for 112 mesoregion clusters.

41

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

Table 1: Descriptive Statistics Panel A: Liberalization Shock Regional tariff reductions (RTRr)

Panel B: Main Outcome Variables Change in log formal earnings premium Change in log informal earnings premiuma Change in log formal employment

1991-1995 0.044 (0.039) 1991-1995

1991-2000

1991-2005

1991-2010

0.258 (0.161)

0.305 (0.174) -0.050 (0.135) 0.599 (0.549) 0.269 (0.162) 0.728 (0.318) -0.260 (0.387) -0.824 (0.421) -0.099 (0.270) -1.092 (0.285) 0.052 (0.175) 0.198 (0.103)

0.401 (0.189)

0.712 (0.201) 0.161 (0.197) 1.308 (0.614) 0.291 (0.228) 1.271 (0.444) -0.128 (0.384) -1.135 (0.516) 0.299 (0.176) -1.305 (0.405) 0.366 (0.114) 0.388 (0.178)

1991

1995

2000

2005

2010

755.98 (273.08) 30,466 (152267) 0.397 (0.194) 0.113 (0.077)

1,105.83 (394.71) 34,929 (161657)

944.18 (323.97) 40,100 (163917)

939.93 (480.00) 51,631 (197206)

1,152.40 (469.95) 70,170 (269602)

0.268 (0.377)

Change in log informal employmenta Change in log num. formal establishments Change in log avg. formal establishment size Change in log formal job destruction Change in log formal job creation Change in log formal establishment exit Change in log formal establishment entry Change in log working-age populationa

Panel C: Region Characteristics Average Formal Earnings (2010 R$) Formal Employment Share Agriculture/Mininga Share Manufacturinga

0.358 (0.230) -0.180 (0.279) -1.014 (0.398) -0.608 (0.387) -1.206 (0.226) -0.397 (0.287)

0.976 (0.576)

1.055 (0.389) -0.220 (0.375) -0.966 (0.466) 0.143 (0.221) -1.183 (0.343) 0.251 (0.139)

475 microregion observations. See the text for descriptions of all measures. estimates are available only for 1991, 2000, and 2010.

42

a

Calculated using the Census, so

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

Table 2: Regional log Formal Earnings Premia and Employment - 2000, 2010 Change in outcome: (1) Panel A: log Formal Earnings Premia Regional tariff reduction (RTR)

-0.451*** (0.152)

1991-2000 (2)

-0.638*** (0.154)

Formal earnings pre-trend (86-90) ✓

State fixed effects (26) R-squared

0.040

0.225

(3)

-0.529*** (0.141) -0.312** (0.149) ✓ 0.268

Panel B: log Formal Real Earnings Premia (regional deflators following Moretti (2013)) Regional tariff reduction (RTR)

(4)

-1.885*** (0.316)

1991-2010 (5)

-1.736*** (0.184)

✓ 0.320

-1.594*** (0.306)

0.501

-1.382*** (0.180)

Formal earnings pre-trend (86-90) ✓

State fixed effects (26) R-squared Panel C: log Formal Employment Regional tariff reduction (RTR)

0.238

-3.748*** (0.516)

-3.545*** (0.563)

Formal employment pre-trend (86-90) ✓

State fixed effects (26) R-squared

0.072

0.291

-3.533*** (0.582) -0.0331 (0.147) ✓ 0.291

-6.059*** (0.560)

0.449

-4.675*** (0.660)

✓ 0.149

0.409

(6)

-1.594*** (0.169) -0.418*** (0.144) ✓ 0.537

-1.260*** (0.168) -0.359*** (0.133) ✓ 0.477

-4.663*** (0.679) -0.0319 (0.156) ✓ 0.410

Negative coefficient estimates for the regional tariff reduction imply larger declines in formal earnings or employment in regions facing larger tariff reductions. Microregion observations: Panels A and C, 475; Panel B, 456 (omits a few sparsely populated locations with insufficient data to calculate regional price deflators). Regional earnings premia calculated controlling for age, sex, education, and industry of employment. Panels A and B: efficiency weighted by the inverse of the squared standard error of the estimated change in log formal earnings premium. Pre-trends computed for 1986-1990. Standard errors (in parentheses) adjusted for 112 mesoregion clusters. *** Significant at the 1 percent, ** 5 percent, * 10 percent level.

43

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

Table 3: Regional log Working-Age Population - 2000, 2010 Change in log Working-Age Population: (1) Regional tariff reduction (RTR) Population pre-trend (80-91)

0.333 (0.243) 0.406** (0.164)

Population pre-trend (70-80) State fixed effects (26) R-squared

✓ 0.654

1991-2000 (2) -0.061 (0.330)

0.297*** (0.072) ✓ 0.557

(3) 0.018 (0.204) 0.328* (0.171) 0.137*** (0.047) ✓ 0.678

(4) 0.392 (0.319) 0.632*** (0.225)

✓ 0.666

1991-2010 (5) -0.175 (0.473)

0.445*** (0.087) ✓ 0.554

(6) -0.059 (0.294) 0.531** (0.235) 0.190** (0.073) ✓ 0.685

Positive (negative) coefficient estimates for the regional tariff reduction imply larger increases (decreases) in population in regions facing larger tariff reductions. Outcomes calculated using Census data. 405 microregion observations. Efficiency weighted by the inverse of the squared standard error of the dependent variable estimate. Pre-trends computed for 1980-1991 and 1970-1980. Standard errors (in parentheses) adjusted for 90 mesoregion clusters. *** Significant at the 1 percent, ** 5 percent, * 10 percent level.

44

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

Table 4: Regional log Informal Employment and Earnings Premia - 2000, 2010 Change in outcome: (1) Panel A: log Informal Employment Regional tariff reduction (RTR) Informal employment pre-trend (80-91)

2.017*** (0.431) 0.069 (0.115)

All employment pre-trend (70-80) State fixed effects (26) R-squared Panel B: log Informal Earnings Premia Regional tariff reduction (RTR) Informal earnings pre-trend (80-91)

R-squared

1.706*** (0.344)

✓

0.121** (0.056) ✓

0.579

0.589

-0.027 (0.161) -0.191*** (0.049)

All workers' earnings pre-trend (70-80) State fixed effects (26)

1991-2000 (2)

✓ 0.676

0.654

1.593*** (0.532) 0.050 (0.114) 0.110** (0.044) ✓ 0.592

-0.217 (0.160)

0.008 (0.064) ✓

(3)

-0.034 (0.163) -0.193*** (0.048) -0.016 (0.060) ✓ 0.676

(4)

2.122*** (0.468) 0.149 (0.132)

✓ 0.524

0.352 (0.256) -0.288*** (0.086)

1991-2010 (5)

1.448*** (0.491)

0.263*** (0.080) ✓ 0.552

0.054 (0.298)

✓

0.001 (0.109) ✓

0.690

0.667

(6)

1.196* (0.705) 0.109 (0.126) 0.239*** (0.063) ✓ 0.562

0.338 (0.251) -0.291*** (0.084) -0.035 (0.102) ✓ 0.690

Positive (negative) coefficient estimates for the regional tariff reduction imply larger increases (declines) in informal earnings or employment in regions facing larger tariff reductions. Outcomes calculated using Census data. 405 microregion observations. Regional earnings premia calculated controlling for age, sex, education, and industry of employment. Efficiency weighted by the inverse of the squared standard error of the dependent variable estimate. Pre-trends computed for 1980-1991 and 1970-1980. Standard errors (in parentheses) adjusted for 112 mesoregion clusters. *** Significant at the 1 percent, ** 5 percent, * 10 percent level.

45

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

Table 5: Mechanisms: Changing Worker Composition - 1995, 2000, 2005, 2010 Change in log Formal Earnings Premia:

1991-1995 (1)

Panel A: Main specification Regional tariff reduction (RTR)

1991-2000 (2)

-0.096 -0.529*** (0.120) (0.141) Panel B: Earnings premia controlling for individual fixed effects (fixed returns) Regional tariff reduction (RTR) -0.193* -0.514*** (0.115) (0.144) Panel C: Earnigns premia controlling for individual fixed effects (time-varying returns) Regional tariff reduction (RTR) -0.230** -0.551*** (0.093) (0.098) Formal earnings pre-trend (86-90) ✓ ✓ State fixed effects (26) ✓ ✓

1991-2005 (3)

1991-2010 (4)

-1.294*** (0.139)

-1.594*** (0.169)

-1.119*** (0.147)

-1.271*** (0.172)

-1.322*** (0.094) ✓ ✓

-1.454*** (0.119) ✓ ✓

Negative coefficient estimates for the regional tariff reduction (RT Rr ) imply larger declines in formal earnings in regions facing larger tariff reductions. Microregion observations: Panel A, 475; Panels B and C, 450 (omits regions with insufficient observations to identify region-year fixed effects in any particular year). Regional earnings premia: Panel A: calculated controlling for age, sex, education, and industry of employment; Panels B and C: controlling for individual fixed effects. Standard errors (in parentheses) adjusted for 112 mesoregion clusters. Efficiency weighted by the inverse of the squared standard error of the estimated change in log formal earnings premium. See text for detailed description of each panel. *** Significant at the 1 percent, ** 5 percent, * 10 percent level.

46

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

Table 6: Mechanisms: Slow Response of Imports or Exports - 1995, 2000, 2005, 2010 Change in log Formal Earnings Premia:

Panel A: Main specification Regional tariff reduction (RTR)

Panel B: Controls for trade quantities (OLS) Regional tariff reduction (RTR)

1991-1995 (1)

-1.294*** (0.139)

-1.594*** (0.169)

-0.089 (0.112)

-0.521*** -1.287*** (0.138) (0.181) -0.382 (2.242) 0.142 (3.355)

-1.562*** (0.221)

-0.129 (0.106)

-0.569*** -1.342*** (0.129) (0.173) 1.668 (2.631) -0.149 (3.861) 93.04

-1.757*** (0.212)

-0.049 (0.108)

-0.488*** -1.372*** (0.132) (0.161) -3.489 (2.427) 5.379* (3.268) 876.2

-1.502*** (0.213)

Import quantity control Export quantity control First-stage F (Kleibergen-Paap)

Import quantity control Export quantity control First-stage F (Kleibergen-Paap) Formal earnings pre-trend (86-90) State fixed effects (26)

1991-2010 (4)

-0.529*** (0.141)

Export quantity control

Panel D: Colombia IV Regional tariff reduction (RTR)

1991-2005 (3)

-0.096 (0.120)

Import quantity control

Panel C: Latin America IV Regional tariff reduction (RTR)

1991-2000 (2)

✓ ✓

✓ ✓

✓ ✓

✓ ✓

Negative coefficient estimates for the regional tariff reduction (RT Rr ) imply larger declines in formal earnings in regions facing larger tariff reductions. Panel A replicates the earnings results in columns (3) and (6) of Table 2. Panels B-D include regional import and export quantity controls as in (7). We instrument for the potentially endogenous import and export controls using regional measures of commodity price growth from Ad˜ ao (2015) and with regional trade flows for other countries. “Latin America” consists of Argentina, Chile, Colombia, Paraguay, Peru, and Uruguay. We measure imports and exports between Latin America or Colombia and the rest of the world excluding Brazil. Due to Comtrade data availability, changes in Colombian trade flows are measured from 1991 to each subsequent year and Latin American trade flows from 1994. We allow for time-varying first-stage coefficients, so we have 2 endogenous variables (RegImprt and RegExprt ) and 57 instruments for Colombia (3 instruments × 19 years) and 48 instruments for Latin America (3 instruments × 16 years). First-stage Kleinbergen-Paap F statistics are compared to the Stock and Yogo (2005) critical value of 21 to reject 5 percent bias relative to OLS. Standard errors (in parentheses) adjusted for 112 mesoregion clusters. Efficiency weighted by the inverse of the squared standard error of the estimated change in log formal earnings premium. *** Significant at the 1 percent, ** 5 percent, * 10 percent level.

47

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

Table 7: Test for Agglomeration Economies Change in log Region × Industry Employment: Regional tariff reduction (RTR) Industry tariff reduction

(1) -7.751*** (0.625) -1.790*** (0.294)

All Industries (2) (3) -6.084*** (0.623) -1.666*** (0.290)

Formal employment pre-trend (86-90) Industry fixed effects (20) State fixed effects (26)

✓

-6.183*** (0.631) -1.669*** (0.291) -0.106*** (0.036) ✓

(4) -6.333*** (0.646)

Tradable Industries (5) (6)

-0.147*** (0.032)

-6.708*** (0.675) -2.017*** (0.332) -0.110*** (0.037)

✓ ✓

✓

-6.704*** (0.694) -0.150*** (0.032) ✓ ✓

Negative coefficient estimates for the regional tariff reduction imply the presence of agglomeration economies, following (15). Observations represent region × industry pairs. The dependent variable is the change in log formal employment in a given region × industry pair from 1991 to 2010. Columns (1) - (4) cover all industries, including the nontradable sector, while columns (5) and (6) restrict attention to tradable industries. For tradable industries, industry tariff reductions are given by the decline in the log of one plus the tariff rate. For the nontradable sector, the industry tariff reduction is measured using RT Rr . Standard errors (in parentheses) adjusted for 112 mesoregion clusters. *** Significant at the 1 percent, ** 5 percent, * 10 percent level.

48

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

Table 8: Agglomeration Elasticity Estimates Panel A: Inverse labor supply elasticity (η)

0.363*** (0.060)

Panel B: Agglomeration elasticity (κ) (1) low (0.152)

(2) mid (0.349)

(3) high (0.545)

Wage-based agglomeration elasticity (κ)

0.042* (0.023)

0.188*** (0.023)

0.333*** (0.025)

Employment-based agglomeration elasticity (κ)

0.215*** (0.032)

0.330*** (0.038)

0.461*** (0.043)

Specific factors' share of non-labor inputs (ζ):

Labor supply elasticity, η, estimated from (12) using RT Rr as an instrument for the change in regional log earnings premium. The first-stage partial F-statistic (Kleibergen-Paap) for this regression is 59.14. Given the estimate of η, the agglomeration elasticity, κ, is estimated using two alternative methods. The earnings-based approach estimates (13), and the employment-based approach estimates (14), both using nonlinear least squares, and both including 1986-1990 pre-liberalization outcome trends and state fixed effects. The employment-based estimates control for industry price changes as in column (3) of Table 7, and results using other approaches are very similar. We present estimates for three different values of ζ, specific factors’ share of non-labor inputs, based on Valentinyi and Herrendorf (2008). See text for details. Standard errors (in parentheses) bootstrapped by regional resampling. *** Significant at the 1 percent, ** 5 percent, * 10 percent level.

49

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

Online Appendices (Not for publication) A Data and Definitions A.1 Tariffs . . . . . . . . . . . . . . . . . . . . . . . A.2 RAIS Data . . . . . . . . . . . . . . . . . . . . A.3 Demographic Census . . . . . . . . . . . . . . . A.4 Regional Tariff Changes . . . . . . . . . . . . . A.5 Local Price Indexes . . . . . . . . . . . . . . . . A.6 Regional Change in Imports and Exports . . . A.7 Entry, Exit, Job Creation, and Job Destruction

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

B Supplemental Empirical Results B.1 Industry-Level Outcome Pre-Trends vs. Tariff Reductions B.2 Informal Employment . . . . . . . . . . . . . . . . . . . . B.3 Regional Earnings Premium Regressions . . . . . . . . . . B.4 Industry-Region Earnings Results . . . . . . . . . . . . . . B.5 Formal Earnings Regression Scatterplots . . . . . . . . . . B.6 Census Earnings, Wage, and Employment Results . . . . B.7 Robustness Tests . . . . . . . . . . . . . . . . . . . . . . . B.8 Potential Confounders . . . . . . . . . . . . . . . . . . . . B.9 Earnings and Employment Sample Splits . . . . . . . . . . B.10 Regional Change in log Imports and Exports . . . . . . . B.11 Overall Employment . . . . . . . . . . . . . . . . . . . . . B.12 Capital Adjustment Confidence Intervals . . . . . . . . . . B.13 Exit by Establishment Size . . . . . . . . . . . . . . . . .

. . . . . . .

. . . . . . . . . . . . .

. . . . . . .

. . . . . . . . . . . . .

. . . . . . .

. . . . . . . . . . . . .

. . . . . . .

. . . . . . . . . . . . .

. . . . . . .

. . . . . . . . . . . . .

. . . . . . .

. . . . . . . . . . . . .

. . . . . . .

. . . . . . . . . . . . .

. . . . . . .

. . . . . . . . . . . . .

. . . . . . .

. . . . . . . . . . . . .

. . . . . . .

. . . . . . . . . . . . .

. . . . . . .

. . . . . . . . . . . . .

. . . . . . .

. . . . . . . . . . . . .

. . . . . . .

. . . . . . . . . . . . .

. . . . . . .

. . . . . . . . . . . . .

. . . . . . .

51 51 54 54 55 58 59 59

. . . . . . . . . . . . .

61 61 63 66 68 70 72 73 76 86 88 94 95 98

C Model 100 C.1 Baseline Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 C.2 Agglomeration Economies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101

50

Trade Liberalization and Regional Dynamics

A A.1

Dix-Carneiro and Kovak

Data and Definitions Tariffs

Tariff data come from Kume et al. (2003), who report nominal tariffs and effective rates of protection from 1987 to 1998 using the Brazilian industry classification N´ıvel 50. We aggregate these tariffs slightly to an industry classification that is consistent with the Demographic Census data used to construct local tariff shock measures. The classification is presented in Table A1. In aggregating, we weight each N´ıvel 50 industry by its 1990 industry value added, as reported in IBGE National Accounts data. Figure A1 shows the evolution of nominal tariffs from 1987 to 1998 for the ten largest industries. The phases of Brazilian liberalization are visible (see Section 2 for a discussion and citations). Large nominal tariff cuts from 1987-1989 had little effect on protection, due to the presence of substantial nontariff barriers and tariff exemptions. In 1990, the majority of nontariff barriers and tariff exemptions were abolished, being replaced by tariffs providing equivalent protection; note the increase in tariffs in some industries in 1990. During liberalization, from 1990 to 1994, tariffs fell in all industries, then were relatively stable from 1995 onward. In Section 4.2 we calculate post-liberalization tariff changes using UNCTAD TRAINS. See Appendix B.8 for details.

51

Tradable

52

Transportation and Communications

Private Services

Public Administration

98

Financial Institutions Real Estate and Corporate Services

94 95

97

Mineral Mining (except combustibles) Petroleum and Gas Extraction and Coal Mining Nonmetallic Mineral Goods Manufacturing Iron and Steel, Nonferrous, and Other Metal Production and Processing Machinery, Equipment, Commercial Installation Manufacturing, and Tractor Manufacturing Electrical, Electronic, and Communication Equipment and Components Manufacturing Automobile, Transportation, and Vehicle Parts Manufacturing Wood Products, Furniture Manufacturing, and Peat Production Paper Manufacturing, Publishing, and Printing Rubber Product Manufacturing Chemical Product Manufacturing Petroleum Refining and Petrochemical Manufacturing Pharmaceutical Products, Perfumes and Detergents Manufacturing Plastics Products Manufacturing Textiles Manufacturing Apparel and Apparel Accessories Manufacturing Footwear and Leather and Hide Products Manufacturing Food Processing (Coffee, Plant Products, Meat, Dairy, Sugar, Oils, Beverages, and Other) Miscellaneous Other Products Manufacturing Utilities Construction Wholesale and Retail Trade

2 3 4 5 8 10 12 14 15 16 17 18 20 21 22 23 24 25 32 91 92 93

96

Industry Name Agriculture

Industry 1

42

39, 43

36, 37

38 40, 41

2 3 4 5-7 8 10-11 12-13 14 15 16 17,19 18 20 21 22 23 24 25-31 32 33 34 35

Nível 50 1

2000, 2010 Census (CNAE-Dom) 1101-1118, 1201-1209, 1300, 1401, 1402, 2001, 2002, 5001, 5002 050, 053-059 12000, 13001, 13002, 14001-14004 051-052 10000, 11000 100 26010, 26091, 26092 110 27001-27003, 28001, 28002 120 29001 130 29002, 30000, 31001, 31002, 32000, 33003 140 34001-34003, 35010, 35020, 35030, 35090 150, 151, 160 20000, 36010 170, 290 21001, 21002, 22000 180 25010 200 23010, 23030, 23400, 24010, 24090 201, 202, 352, 477 23020 210, 220 24020, 24030 230 25020 240, 241 17001, 17002 250,532 18001, 18002 190, 251 19011, 19012, 19020 260, 261, 270, 280 15010, 15021, 15022, 15030, 15041-15043, 15050, 16000 300 33001, 33002, 33004, 33005, 36090, 37000 351, 353 40010, 40020, 41000 340, 524 45001-45005 410-424, 582, 583 50010, 50030, 50040, 50050, 53010 ,53020, 53030, 53041, 53042, 53050, 53061-53068, 53070, 53080, 53090, 53101, 53102, 55020 451-453, 585, 612 65000, 66000, 67010, 67020 461-464, 543, 552, 571-578, 584, 589 63022, 70001, 71020, 72010, 74011, 74012, 74021, 74022, 74030, 74040, 74050, 74090, 92013, 92014, 92015, 92020 471-476, 481, 482, 588 60010, 60020, 60031, 60032, 60040, 60091, 60092, 61000, 62000, 63010, 63021 ,64010 ,64020, 91010 511, 512, 521-523, 525, 531, 533, 541, 542. 544, 1500, 50020, 53111, 53112, 53113, 55010, 55030, 63030, 545, 551, 577, 586, 587, 613-619, 622-624, 632, 901, 70002, 71010, 71030, 72020, 73000, 74060, 80011, 80012, 902 80090, 85011, 85012, 85013, 85020, 85030, 90000, 91020, 91091, 91092, 92011, 92012, 92030, 92040, 93010, 93020, 93030, 93091, 93092, 95000 354, 610, 611, 621, 631, 711-717, 721-727 75011-75017, 75020

1970, 1980, 1991 Census (atividade) 011-037, 041, 042, 581

Consistent industry classification used in generating local tariff shocks from N´ıvel 50 tariff data in Kume et al. (2003) and Decennial Census data.

Nontradable

Table A1: Consistent Industry Classification Across Censuses and Tariff Data

Trade Liberalization and Regional Dynamics Dix-Carneiro and Kovak

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

Figure A1: Tariffs - 1987-1998 90  

Tex$les  

80  

70  

Auto,  Transport,  Vehicles   Food  Processing   Nonmetallic  Mineral  Manuf  

60  

Electric,  Electronic  Equip.  

50  

Machinery,  Equipment   Metals   Agriculture  

40  

30  

Chemicals   Petroleum  Refining  

20  

10  

0   1987  

1988  

1989  

1990  

1991  

1992  

1993  

1994  

1995  

1996  

1997  

1998  

Nominal tariffs from Kume et al. (2003), aggregated to the industry classification presented in Table A1. The ten largest industries by 1990 value added are shown.

53

Trade Liberalization and Regional Dynamics

A.2

Dix-Carneiro and Kovak

RAIS Data

The Rela¸c˜ ao Anual de Informa¸c˜ oes Sociais (RAIS) is a high quality census of the Brazilian formal labor market. Originally, RAIS was created as an operational tool for the Brazilian government to i) monitor the entry of foreign workers into the labor market; ii) oversee the records of the FGTS (Fundo de Garantia do Tempo de Servi¸co) program, a national benefits program consisting of employers’ contributions to each of its employees; iii) provide information for administering several government benefits programs such as unemployment insurance; and iv) generate statistics regarding the formal labor market. Today it is the main tool used by the government to enable the payment of the ”abono salarial ” to eligible workers. This is a government program that pays one additional minimum wage at the end of the year to workers whose average monthly wage was not greater than two times the minimum wage, and whose job information was correctly declared in RAIS, among other minor requirements. Thus, workers have an incentive to ensure that their employer is filing the required information. Moreover, firms are required to file, and face fines until they do so. Together, these requirements ensure that the data in RAIS are accurate and complete. Observations in the data are indexed by a worker ID number, the Programa de Integra¸c˜ ao Social (PIS), and an establishment registration number, the Cadastro Nacional da Pessoa Jur´ıdica (CNPJ). Both of these identifiers are consistent over time, allowing one to track workers and establishments across years. Establishment industry is reported using the Subsetor IBGE classification, which includes 12 manufacturing industries, 2 primary industries, 11 nontradable industries, and 1 other/ignored.60 Worker education is reported using the following 9 education categories (listing corresponding years of education in parentheses): illiterate (0), primary school dropout (1-3), primary school graduate (4), middle school dropout (5-7), middle school graduate (8), high school dropout (9-10), high school graduate (11), college dropout (12-14), and college graduate (≥ 15). In each year, and for each job, RAIS reports average earnings throughout the year, and earnings in December.61 We focus on labor market outcomes reported in December of each year. This choice ensures that earnings and formal employment status are measured at the same time for all workers and all jobs. It avoids the potential confounding effects on average yearly earnings that might arise in situations where some workers begin working in early in the year and others begin late in the year.

A.3

Demographic Census

We utilize information from the long form of the Demographic Censuses (Censo Demogr´ afico) for 1970, 1980, 1991, 2000, and 2010. The long form micro data reflect a 5 percent sample of the population in 1970, 1980, and 2010, a 5.8 percent sample in 1991, and a 6 percent sample in 2000. The primary benefit of the Census for our purposes is the ability to observe those outside formal employment, who are not present in the RAIS database. Although our main analysis focuses on monthly earnings, following the information available in RAIS, the Census provides weekly hours information from 1991-2010, allowing us to calculate hourly wages as monthly earnings divided by 4.33 times weekly hours. Census results for monthly earnings and hourly wages are very similar. In 1970 and 1980, hours information is presented in 60 A less aggregate industry classification (CNAE) is available from 1994 onward, but we need a consistent classification from 1986-2010, so we use Subsetor IBGE. 61 From 1994 onward, RAIS reports hours, making it possible to calculate hourly wages. However, since we need a consistent measure from 1986-2010, we focus on monthly earnings.

54

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

5 rough bins. Thus, when calculating pre-liberalization trends using data from 1970 and 1980, we use monthly earnings even when examining hourly wage outcomes. In 1991-2010, the Census asks whether each worker has a signed work card. This is the standard definition of formal employment, and is necessary for a worker to appear in the RAIS sample. Thus, we use this as our primary definition of formal employment. In 1980 and 1991, there is an alternative proxy for formal employment, reporting whether the worker’s job includes contributions to the national social security system. When calculating pre-liberalization outcome trends for 19801991, we use this alternative measure to identify formally employed workers. The social security contributions proxy appears to be a good one; in 1991, when both measures are available, 95.9 percent of workers would be classified identically when using either measure. In 1970, there is no information on formality, so pre-liberalization outcome trends for 1970-1980 are calculated for all workers. The definition of employment changes across Census years. In 1970 it includes those reporting working or looking for work during August 1970 (the questionnaire does not separately identify working vs. looking for work). In 1980 it includes those who report working during the year prior to September 1, 1980. In 1991 it includes those reporting working regularly or occasionally during the year prior to September 1, 1991. In 2000 and 2010 it includes those who report paid work, temporary leave, unpaid work, or cultivation for own consumption during the week of July 23-29 in 2000 and July 25-31 in 2010. Note that the employment concept changes substantially across years. This highlights yet another benefit of using RAIS as our primary data source, since the employment concept in RAIS is consistent throughout the sample. Yet, while the changes complicate the interpretation of Census-based employment rates over time, there is no reason to expect systematic differences across regions to result from the changing employment concept. Thus, our cross-region identification strategy should be valid when using the Census to measure employment in spite of these measurement issues.

A.4

Regional Tariff Changes

Regional tariff reductions, defined in (2), are constructed using information from various sources. Tariff changes come from Kume et al. (2003), and are aggregated from the N´ıvel 50 level to the industry classification presented in Table A1 using 1990 value-added weights from the IBGE National Accounts. Figure 1 shows the resulting industry-level variation in tariff changes. The weights, βri in (2) depend upon the initial regional industry distribution (λri ) and the specific-factor share in production (ϕi ). We calculate the λri using the 1991 Census. We use the Census because it provides a less aggregate industry definition than what is available in RAIS, and because the Census allows us to calculate weights that are representative of overall employment, rather than just formal employment. However, note that shocks using formal employment weights yield very similar results (see Panel D of Table B6). We calculate the ϕi using data from the Use Table of the 1990 National Accounts from IBGE. The table “Componentes do Valor Adicionado” provides the wagebill (Remunera¸c˜ oes) and gross operating surplus (Excedente Operacional Bruto Inclusive Rendimento de Autˆ onomos), which reflects the share of income earned by capital. We define ϕi as capital’s share of the sum of these two components. When imposing equal cost shares across industries (see Section 5.4.2), we calculate ϕ using the economy-wide wagebill and gross operating surplus, yielding a value of ϕ = 0.544. Because Brazilian local labor markets differ substantially in the industry distribution of their

55

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

employment, the weights βri vary across regions. Figure A2 demonstrates how variation in industry mix leads to variation in RT Rr . The figure shows the initial industry distribution of employment for the regions facing the largest tariff reduction (Rio de Janeiro) the median tariff reduction (Alfenas in southwestern Minas Gerais state), and the smallest tariff reduction (actually a small increase, Mata Grande in northwest Alagoas state). The industries on the x-axis are sorted from the most negative to the most positive tariff change. Rio de Janeiro has more weight on the left side of the diagram, by virtue of specializing in manufacturing, particularly in apparel and food processing industries, which faced quite large tariff reductions. Thus, its regional tariff reduction is quite large. Alfenas is a coffee growing and processing region, which also has some apparel employment, balancing the large tariff declines in apparel and food processing against the small tariff increase in agriculture. Mata Grande is located in a sparsely populated mountainous region, and is almost exclusively agricultural, leading it to experience a small tariff increase overall. Thus, although all regions faced the same set of tariff reductions across industries, variation in the industry distribution of employment in each region generates substantial variation in RT Rr .

56

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

Figure A2: Variation Underlying Regional Tariff Reduction

Industry Weight

0.50

Rio de Janeiro, RJ (.15)

Alfenas MG (.03)

Mata Grande, AL (-.01)

0.75 0.97

0.40 0.30 0.20

Agriculture

Petroleum, Gas, Coal

Mineral Mining

Footwear, Leather

Metals

Wood, Furniture, Peat

Paper, Publishing, Printing

Chemicals

Textiles

Petroleum Refining

Food Processing

Machinery, Equipment

Electric, Electronic Equip.

Auto, Transport, Vehicles

Nonmetallic Mineral Manuf

Plastics

Other Manuf.

Pharma., Perfumes, Detergents

Rubber

0.00

Apparel

0.10

Industries sorted from most negative to most positive tariff change

Industry distribution of 1991 employment in the regions facing the largest (Rio de Janeiro, RJ), median (Alfenas, MG) and smallest (Mata Grande, AL) regional tariff reduction. Industries sorted from the most negative to the most positive tariff change (see Figure 1). More weight on the left side of the figure leads to a larger regional tariff reduction, and more weight on the right side leads to a smaller regional tariff reduction.

57

Trade Liberalization and Regional Dynamics

A.5

Dix-Carneiro and Kovak

Local Price Indexes

Moretti (2013) calculates local price indexes for the U.S. using the change in monthly rents for 2 or 3 bedroom apartments. We adjust this approach to the Brazilian context in a few ways. First, we focus on 1 or 2 bedroom apartments, which are far more common in the Brazilian setting, accounting for more than 85 percent of the stock of rental units in 1991 and 2010. Many Brazilian cities include favelas with somewhat improvised structures, and rural areas often feature less formal dwellings. We restrict the sample to include only units with modern construction materials (masonry or wood framing), with at least one bathroom, and with modern sanitation (sewer or septic tank). These restrictions allow us to avoid comparing modern apartments to informal dwellings. Using this sample of apartments, we calculate the change in log average monthly rent in each region. 19 very sparsely populated microregions do not have observations for any rental units satisfying these characteristics in either 1991 or 2010, so we have rent indexes for 456 microregions in our sample. We then need to transform the change in rental prices into a regional price index. Given the cross-sectional nature of our analysis, we only need to be concerned with prices that vary at the local level, i.e. nontradables, since tradable goods prices move together across regions, and thus do not affect this exercise. Using local Consumer Price Indexes produced by the Bureau of Labor Statistics for 23 U.S. metropolitan areas, Moretti (2013) shows that, as expected, local non-housing nontradables’ prices move with local rental prices. He estimates a slope of 0.35 for the effect of ´ housing prices on non-housing nontradables’ prices. The Brazilian Consumer Price Index (Indices de Pre¸cos ao Consumidor - IPC) system reports that in 2002-03, housing’s share of consumption was 16.24 percent and that the share for other nontradable goods was 39.94 percent (IBGE 2005). Together, these figures imply that the effective weight on housing prices in the consumer price index is 0.1624 + 0.3994 · 0.35 = 0.3022. Our local price deflator is therefore 0.3022 times the change in log rental prices in the region.

58

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

A.6

Regional Change in Imports and Exports

Import and export data between Brazil and the rest of the world come from Comtrade, at the 6-digit HS level. We map from HS codes to the industries presented in Table A1 and calculate total Brazilian trade flows by industry and year. In the main text, we follow Autor et al. (2013) (ADH) by generating regional weighted averages of changes in imports and exports per worker. For each industry, we calculate the change in imports (Mit ) and exports (Xit ) from 1990 to each subsequent year t. These trade flows are measured in $100,000 units. We then generate the regional change in imports and exports per worker as follows. RegImprt =

X Lrit ∆1990−t Mit 0 Lrt0 Lit0

(16)

X Lrit ∆1990−t Xit 0 Lrt0 Lit0

(17)

i

RegExprt =

i

The rightmost ratios in these expressions measure the change in imports or exports per worker initially employed in the industry, in year t0 = 1991. The preceding ratios represent industry weights for each region, reflecting industry i’s share of tradable employment in region r in 1991. These weights are equivalent to λri in (1). We then generate weighted averages by summing these terms over tradable industries. Finally, we construct regional net exports as the difference in regional exports and imports. RegN etExprt = RegExprt − RegImprt In Appendix B.10, we present an alternative set of results based on the change in log trade flows rather than the change in trade flows per worker. RegLnImprt =

X Lrit

0

Lrt0

i

RegLnEmprt =

X Lrit

∆1990−t ln(Mit )

0

Lrt0

i

∆1990−t ln(Xit )

(18)

(19)

We emphasize that this measure is presented only for descriptive purposes and as a statistical robustness test since it does not have the same theoretical underpinnings as the measures following Autor et al. (2013).

A.7

Entry, Exit, Job Creation, and Job Destruction

We calculate cumulative job creation and job destruction following Davis and Haltiwanger (1990). X

job creationrt ≡

e∈Ert , get >0

xet get , Xrt

X

job destructionrt ≡

e∈Ert , get <0

59

xet |get |, Xrt

(20)

(21)

Trade Liberalization and Regional Dynamics

where

get ≡

Let − Le,1991 , xet

1 xet ≡ (Let + Le,1991 ), 2

Dix-Carneiro and Kovak

Xrt ≡

X

Let ,

e∈Ert

Let is employment at establishment e in year t and Ert is the set of active establishments in region r in year t. Note that employment growth, get , is calculated from 1991 to year t. The dependent variables for the regressions underlying Figure 9 are ln(job creationrt ) and ln(job destructionrt ). Entry and exit are calculated analogously, replacing establishment employment with an indicator for the establishment being active in the relevant year, ιet . entry rt ≡

x ˜et g˜ , ˜ rt et X g˜et >0

(22)

x ˜et |˜ g |, ˜ rt et X g˜et <0

(23)

X e∈Ert ,

exitrt ≡

X e∈Ert ,

where

g˜et ≡

ιet − ιe,1991 , x ˜et

1 x ˜et ≡ (ιet + ιe,1991 ), 2

˜ rt ≡ X

X

ιet ,

e∈Ert

These definitions yield intuitive results. entry rt is equivalent to the share of active firms in region r in year t that were not active in 1991. exitrt is equivalent to the number of firms in region r that were active in 1991 and not in year t, divided by the number of firms active in year t. The dependent variables for the regressions underlying Figure 8 are ln(entry rt ) and ln(exitrt )

60

Trade Liberalization and Regional Dynamics

B B.1

Dix-Carneiro and Kovak

Supplemental Empirical Results Industry-Level Outcome Pre-Trends vs. Tariff Reductions

Along with regional variation in the industrial composition of employment, our analysis relies on variation in tariff cuts across industries. Here we analyze the relationship between tariff cuts during liberalization (1990-1995) and trends in industry wages and employment before liberalization, 19801991. We calculate these pre-liberalization outcome trends using the Demographic Census, to provide a longer pre-liberalization period than what is available in RAIS, which starts in 1986. We implemented a variety of specifications, with results reported in Table B1. In all specifications, the independent variable is the proportional reduction in one plus the tariff rate (∆1990−95 ln(1 + τi )). In panels A-C the dependent variable is the change in log industry earnings. Panel A uses average log earnings; Panel B uses average log earnings residuals controlling for individual age, sex, education, and formal status; and Panel C uses average log earnings residuals controlling for these individual characteristics and region fixed effects. In Panel D, the dependent variable is the change in industry log employment. Column (1) weights industries equally, and presents standard errors based on pairwise bootstrap of the t-statistic, to improve small sample properties with only 20 tradable industry observations. Column (2) uses the same estimator, but drops agriculture. Column (3) uses heteroskedasticity weights and presents heteroskedasticityrobust standard errors, which are likely understated in this small sample (MacKinnon 2011). Column (4) uses the same estimator, but drops agriculture. In all cases, the results should be seen primarily as suggestive, because the analysis uses only 19 or 20 observations. Nearly all of the earnings estimates are positive, indicating larger tariff reductions in industries experiencing more positive wage growth prior to liberalization. The majority of the estimates are insignificantly different from zero, with the exception of weighted results in Panels A and B. These specifications heavily weight agriculture, which exhibited negative wage growth prior to liberalization and experienced essentially no tariff decline during liberalization, driving the strong negative relationship. By dropping agriculture, Column (4) confirms that the significant relationship is driven by agriculture. The employment estimates are larger, and change sign across columns. Given the diversity of findings across earnings and employment specifications, this exercise is somewhat inconclusive. Tariff cuts may or may not have been substantially correlated with pre-liberalization outcome trends. These findings motivate us to control for pre-liberalization outcome trends whenever possible throughout the paper. This ensures that our results are robust to potential spurious correlation between liberalization-induced labor demand shocks and ongoing trends.

61

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

Table B1: Pre-Liberalization Industry Trends - 1980-1991 unweighted, bootstrapped 1980-1991 change in log:

(1)

unweighted, bootstrapped, omitting agriculture (2)

Panel A: average earnings Industry tariff reduction

0.345 (0.322) Panel B: earnings premia (with individual controls) Industry tariff reduction 0.203 (0.273) Panel C: earnings premia (with individual and region controls) Industry tariff reduction 0.135 (0.177) Panel D: employment Industry tariff reduction

Observations

-1.624 (1.272) 20

weighted

weighted, omitting agriculture

(3)

(4)

0.111 (0.354)

1.029*** (0.139)

0.510 (0.582)

-0.017 (0.311)

0.610*** (0.157)

-0.235 (0.350)

0.044 (0.209)

0.184 (0.158)

0.018 (0.222)

-2.696** (1.361)

0.687 (0.417)

-1.651 (1.894)

19

20

19

Decennial Census data. 20 industry observations (19 omitting agriculture). See text for details of dependent and independent variable construction. Column (1) weights industries equally, and presents standard errors based on pairwise bootstrap of the t-statistic. Column (2) uses the same estimator as Column (1), but drops agriculture. Column (3) uses heteroskedasticity weights and presents heteroskedasticity-robust standard errors. Column (4) uses the same estimator as Column (3), but drops agriculture. *** Significant at the 1 percent, ** 5 percent, * 10 percent level.

62

Trade Liberalization and Regional Dynamics

B.2

Dix-Carneiro and Kovak

Informal Employment

The following results provide some descriptive evidence on the informal sector in Brazil. Informality is defined as working without a signed work card (Carteira de Trabalho e Previdˆencia Social ), which entitles workers to benefits and labor protections afforded them by the legal employment system. Table B2 shows that the overall rate of informality increased from 1991 to 2000, before decreasing substantially from 2000 to 2010. Rates of informality are highest in agriculture and much lower in manufacturing. Table B1 breaks out informality rates in the manufacturing sector into individual industries. Finally, Table B2 focuses on the year 2000 and shows the industry distribution of formal and informal employment. There is very substantial overlap in the industry distributions of formal and informal employment. The biggest differences occur in agriculture, which comprises a much larger share of informal employment, and food processing and metals, which comprise larger shares of formal employment. In contrast, the nontradable share is nearly identical for formal and informal employment.

63

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

Table B2: Informal Share of Employment - 1991-2010 1991

2000

2010

Overall

0.58

0.64

0.49

Agriculture Mining Manufacturing Nontradable

0.89 0.61 0.28 0.55

0.86 0.45 0.39 0.64

0.83 0.21 0.29 0.48

Author’s calculations using Brazilian Demographic Census data for workers age 18-64. Informality defined as not having a signed work card.

Figure B1: Informal Share of Employment by Industry - 1991-2010 1991  

0.9  

2000  

0.8  

2010  

0.7   0.6   0.5   0.4   0.3   0.2  

Nontraded  

Agriculture  

Petroleum,  Gas,  Coal  

Mineral  Mining  

Footwear,  Leather  

Metals  

Paper,  Publishing,  PrinBng  

Wood,  Furniture,  Peat  

Chemicals  

TexBles  

Petroleum  Refining  

Machinery,  Equipment  

Food  Processing  

Electric,  Electronic  Equip.  

Nonmetallic  Mineral  Manuf  

Auto,  Transport,  Vehicles  

PlasBcs  

Pharma.,  Perfumes,  Detergents  

Other  Manuf.  

0  

Apparel  

0.1  

Rubber  

Informal  Share  of  Industry  Employment  

1  

Authors’ calculations using Brazilian Demographic Census data for workers age 18-64. Informality defined as not having a signed work card. Industries sorted from most negative to most positive tariff change (with the exception of the nontraded sector).

64

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

Figure B2: Industry Distribution of Formal and Informal Employment - 2000    0.23      0.70  0.69  

formal  

0.09  

informal  

0.08   0.07   0.06   0.05   0.04   0.03   0.02  

Nontraded  

Agriculture  

Petroleum,  Gas,  Coal  

Mineral  Mining  

Footwear,  Leather  

Metals  

Paper,  Publishing,  Prin8ng  

Wood,  Furniture,  Peat  

Chemicals  

Tex8les  

Petroleum  Refining  

Machinery,  Equipment  

Food  Processing  

Electric,  Electronic  Equip.  

Nonmetallic  Mineral  Manuf  

Auto,  Transport,  Vehicles  

Plas8cs  

Pharma.,  Perfumes,  Detergents  

Other  Manuf.  

0.00  

Apparel  

0.01  

Rubber  

Industry  Share  of  Sector  Employment  

0.10  

Authors’ calculations using year 2000 Brazilian Demographic Census data for workers age 18-64. Informality defined as not having a signed work card. Industries sorted from most negative to most positive tariff change (with the exception of the nontraded sector).

65

Trade Liberalization and Regional Dynamics

B.3

Dix-Carneiro and Kovak

Regional Earnings Premium Regressions

As discussed in Section 4.1, we calculate regional earnings premia by regressing workers’ log December earnings on flexible demographic and educational controls, industry fixed effects, and region fixed effects, separately in each year. Table B3 shows the coefficient estimates from these earnings premium regressions for 1991, 2000, and 2010. The region fixed effect estimates provide average log earnings for formally employed workers in the region, controlling for the age, sex, education, and industry composition of the region’s employment. These regional premia then form the outcome variable in our earnings analyses. Note that the coefficient estimates on the controls conform to expectations. Women are paid less than otherwise similar men, and this earnings gap declines over time. Workers exhibit an inverted U-shaped wage profile as they age, as is standard in Mincerian regressions. The returns to education are monotonically positive.

66

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

Table B3: Regional Earnings Premium Regressions - 1991, 2000, 2010 dependent variable: log monthly earnings (1) (2) 1991 2000 Female Age 25-29 30-39 40-49 50-64 Education (years) Primary School Dropout (1-3) Primary School Graduate (4) Middle School Dropout (5-7) Middle School Graduate (8) High School Dropout (9-11) High School Graduate (12) College Dropout (13-15) College Graduate (≥16)

-0.349*** (0.000)

-0.295*** (0.000)

-0.261*** (0.000)

0.210*** (0.000) 0.407*** (0.000) 0.527*** (0.001) 0.435*** (0.001)

0.209*** (0.000) 0.374*** (0.000) 0.525*** (0.000) 0.507*** (0.001)

0.148*** (0.000) 0.273*** (0.000) 0.382*** (0.000) 0.474*** (0.000)

0.015*** (0.001) 0.111*** (0.001) 0.188*** (0.001) 0.297*** (0.001) 0.454*** (0.001) 0.711*** (0.001) 0.967*** (0.002) 1.374*** (0.001)

0.007*** (0.001) 0.075*** (0.001) 0.129*** (0.001) 0.181*** (0.001) 0.305*** (0.001) 0.523*** (0.001) 0.902*** (0.001) 1.384*** (0.001)

0.130*** (0.001) 0.182*** (0.001) 0.206*** (0.001) 0.236*** (0.001) 0.289*** (0.001) 0.430*** (0.001) 0.792*** (0.001) 1.368*** (0.001)

X X

X X

X X

13,582,443 0.858

17,733,492 0.842

30,662,075 0.759

Fixed Effects Industry (24) Region (475) Observations R-squared

(3) 2010

Individual worker observations from RAIS. Earnings premium regressions were run for each year from 1986-2010. Here we show three years as examples. The region fixed effect estimates provide average log earnings for formally employed workers in the region, controlling for the age, sex, education, and industry composition of the region’s employment. These regional premia then form the outcome variable in our regional earnings analyses. The omitted category is a male, age 18-24, with 0 years of education (illiterate). Robust standard errors in parentheses. *** Significant at the 1 percent, ** 5 percent, * 10 percent level.

67

Trade Liberalization and Regional Dynamics

B.4

Dix-Carneiro and Kovak

Industry-Region Earnings Results

In this appendix, we pursue an alternative research design for studying the earnings effects of liberalization, with the unit of analysis at the industry × region level. Equations (13) and (14) suggest a similar industry × region research design, but in the context of our model this approach only applies in the long run. The results in this section are therefore supplementary. Similar approaches appear in Hakobyan and McLaren (forthcoming) and Acemoglu, Autor, Dorn, Hanson and Price (2016). We must first estimate earnings premia at the industry × region level. Because there are 4,773 such combinations, it is not feasible to directly estimate industry × region fixed effects (and obtain their standard errors) in an earnings regression like (4). Instead, we estimate the following earnings regression separately in each year t, absorbing the industry × region fixed effects, µirt ln(earnjrit ) = Xjt Γt + µirt + ejrit

(24)

ˆ t , and average these residuals within industry We then calculate residuals u ˆjrit ≡ ln(earnjrit )−Xjt Γ × region bins to recover the fixed-effect estimates, µ ˆirt . Because this procedure does not directly yield standard errors for these fixed-effect estimates, we implement 100 bootstrap repetitions and calculate bootstrap standard errors. We then calculate the change in these industry × region earnings premia and use them as the dependent variable in the following regression specification. µ ˆirt − µ ˆir,1991 = θt RT Rr + δt d ln(1 + τi ) + αst + γt (ˆ µir,1990 − µ ˆir,1986 ) + εirt

(25)

We estimate this regression separately in each year t > 1991. Note that the unit of observation is the industry × region, and we include both the region-level tariff reduction, RT Rr , and the industry-level tariff change, d ln(1 + τi ). In the most stringent specifications, we control for state fixed effects and both region and industry earnings pre-trends. The results for 2000 and 2010 appear in Panel A of Table B4. First, note that the estimated regional and industry-level effects of liberalization are quite consistent across specifications. We continue to find large increases in the regional effects of liberalization between 2000 and 2010, while the industry-level results are quite constant over time. The industry-level tariff change estimates reflect the direct effect of liberalization on workers in the affected industry, irrespective of their region of residence. The RT Rr coefficients capture the regional general-equilibrium effects operating across industries. In 2010, the coefficient on RT Rr is more than three times larger than the industry effects. This finding makes clear the central role of regional labor market equilibrium in affecting workers outcomes following trade liberalization. Panel B replaces the controls for the industry-level tariff change with industry fixed effects, finding similar results.

68

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

Table B4: Industry × Region log Formal Earnings 2000, 2010 Change in industry × region earnings (1) Panel A: Industry Tariff-Change Controls Regional tariff reduction (RTR) Industry tariff reduction (-d ln(1+τi))

-0.618** (0.270) -0.683*** (0.215)

1991-2000 (2)

-0.878*** (0.276) -0.654*** (0.216)

Formal earnings pre-trend (86-90) ✓

State fixed effects (26) R-squared Panel B: Industry Fixed Effects Regional tariff reduction (RTR)

0.082

-0.528** (0.229)

0.116

-0.777*** (0.217)

Formal earnings pre-trend (86-90) ✓

State fixed effects (26) R-squared

0.313

0.345

(3)

-0.762*** (0.254) -0.661*** (0.215) -0.196** (0.080) ✓ 0.143

-0.724*** (0.212) -0.243*** (0.034) ✓ 0.375

(4)

-2.567*** (0.582) -0.684*** (0.254)

1991-2010 (5)

-2.485*** (0.266) -0.573** (0.251)

✓ 0.212

-2.199*** (0.538)

0.323

-2.109*** (0.248)

✓ 0.437

0.530

(6)

-2.252*** (0.216) -0.590*** (0.215) -0.393*** (0.075) ✓ 0.390

-2.036*** (0.244) -0.277*** (0.035) ✓ 0.554

Negative coefficient estimates for the regional tariff reduction imply larger declines in formal earnings in regions facing larger tariff reductions. Negative coefficients for the industry tariff reduction imply larger declines in formal earnings in industries facing larger tariff reductions. 4,733 industry × region observations in each year. Earnings premia calculated controlling for age, sex, and education. Panel A controls for industry tariff reductions, while Panel B uses industry fixed effects to capture liberalization’s direct effect on each industry. Efficiency weighted by the inverse of the squared standard error of the estimated change in log formal earnings premium (see text for description of bootstrap standard error calculation). Pre-trends computed for 1986-1990. Standard errors (in parentheses) adjusted for 112 mesoregion clusters. *** Significant at the 1 percent, ** 5 percent, * 10 percent level.

69

Trade Liberalization and Regional Dynamics

B.5

Dix-Carneiro and Kovak

Formal Earnings Regression Scatterplots

Figure B3 shows scatter plots underlying the formal earnings regression estimates in Figure 3 for 1995, 2000, 2005, and 2010. Each marker represents a microregion, and microregions in each major region are shown with a separate type of marker. The size of each marker is proportional to the weight the relevant microregion receives in the estimation. The mean value of the dependent variable is normalized to zero in each year to focus attention on the slope. These scatter plots make clear three important points about the earnings estimates. First, as shown in Figure 3, the magnitude of the slope increases substantially and steadily as time passes following liberalization. Second, the relationship between changes in formal earnings premia and regional tariff reductions is approximately linear in all time periods, justifying our choice of functional form. Third, the increasing magnitude slope is driven by shifts in earnings across large numbers of microregions in various parts of the country, rather than by a few outliers.

70

1

.5

0

.05

.1

1995

.15

0

.05

.1

2005

.15

.05

.05

.1

2010

.1

.15

.15

Scatter plots of regressions behind Figure 3 estimates. 475 microregion observations. Y-axis is the change in log formal earnings premium in a given region from 1991 to the year listed in the title. Dependent variable demeaned in each year. X-axis is the regional tariff reduction, RT Rr . The size of each marker represents the inverse of the variance of the dependent variable estimate, used to weight the regressions. Major regions shown with different symbols. Southeast: dark blue circles, Northeast: orange diamonds, Center-West: red triangles, South: light blue Xs, North: green squares.

0

0

2000

Figure B3: Formal Earnings Premia vs. Regional Tariff Reductions - 1995, 2000, 2005, 2010

-.5

-.5

-1

0

.5

1

-1

0

1 .5 0 -.5 -1 1 .5 0 -.5

71 -1

Trade Liberalization and Regional Dynamics Dix-Carneiro and Kovak

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

B.6

Census Earnings, Wage, and Employment Results

Table B5 estimates versions of equation (3) using formal sector outcomes calculated using Census data. Because the Census includes hours information, we are able to examine both earnings and wage premia. In all cases, we find negative coefficients on RT Rr , indicating that regions facing larger tariff reductions experienced relative declines in monthly earnings, hourly wages, or employment. We also find substantial growth in the magnitude of these effects, corroborating the results in Table 2, which uses RAIS outcomes.

Table B5: Census Regional log Formal Earnings, Wages, and Employment - 2000, 2010 (1)

1991-­‐2000 (2)

-­‐0.397 (0.335)

-­‐0.293** (0.120)

Change  in  outcome:  

Panel  A:  log  Formal  Earnings  Premia Regional  tariff  reduction  (RTR)

(3)

(4)

-­‐1.384** (0.572)

✓

-­‐0.261** (0.116) -­‐0.0896* (0.0528) ✓

0.579

0.583

0.156

Formal  earnings  pre-­‐trend  (86-­‐90) State  fixed  effects  (26) R-­‐squared Panel  B:  log  Formal  Wage  Premia Regional  tariff  reduction  (RTR)

0.031

-­‐0.630* (0.355)

-­‐0.533*** (0.124)

Formal  earnings  pre-­‐trend  (86-­‐90) State  fixed  effects  (26) R-­‐squared Panel  B:  log  Formal  Employment Regional  tariff  reduction  (RTR)

✓ 0.071

-­‐2.478*** (0.487)

0.605

-­‐1.756*** (0.281)

Formal  employment  pre-­‐trend  (86-­‐90) State  fixed  effects  (26) R-­‐squared

✓ 0.317

0.612

-­‐0.495*** (0.118) -­‐0.108* (0.0555) ✓ 0.609

-­‐1.619*** (0.258) 0.211*** (0.0704) ✓ 0.630

1991-­‐2010 (5)

-­‐0.890*** (0.198)

✓

-­‐1.320** (0.525)

0.718

-­‐0.765*** (0.173)

✓ 0.136

-­‐3.913*** (0.758)

0.718

-­‐2.865*** (0.443)

✓ 0.319

0.629

(6)

-­‐0.855*** (0.186) -­‐0.0994 (0.0728) ✓ 0.720

-­‐0.721*** (0.163) -­‐0.124* (0.0642) ✓ 0.721

-­‐2.725*** (0.417) 0.227** (0.0926) ✓ 0.638

Outcomes calculated using Census data. Negative coefficient estimates for the regional tariff reduction imply larger declines in formal earnings, wages, or employment in regions facing larger tariff reductions. 475 microregion observations. Regional earnings premia calculated controlling for age, sex, education, and industry of employment. Efficiency weighted by the inverse of the squared standard error of the estimated outcome. RAIS pre-trends computed for 19861990. Standard errors (in parentheses) adjusted for 112 mesoregion clusters. *** Significant at the 1 percent, ** 5 percent, * 10 percent level.

72

Trade Liberalization and Regional Dynamics

B.7

Dix-Carneiro and Kovak

Robustness Tests

Table B6 presents the various earnings robustness tests summarized in Section 4.2. For comparison, Panel A shows the main specification, corresponding to the estimates in Table 2 and Figure 3. All panels in the table control for pre-liberalization earnings growth from 1986-1990, calculated using RAIS. Panel B additionally controls for longer pre-liberalization earnings trends calculated using the Census. The 1980-1991 control reflects the growth in formal earnings premium, where formality is defined based on whether the worker’s job included social security contributions. The 1970-1980 control reflects growth in the earnings premium for all workers, since there is no formality information in the 1970 Census. See Appendix A.3 for more detail on Census data. A potential problem with Panel B is mechanical endogeneity, because the 1980-1991 pre-trend and the 1991-t earnings growth dependent variable both include the 1991 earnings premium. Panel C resolves this issue by using earnings growth from 1992 to year t as the dependent variable, while including the Census pre-trend controls. Panel D calculates the regional tariff reduction (RT Rr ) in (2) using weights based on the initial industry distribution of regional formal employment, rather than overall employment. Panel E calculates RT Rr using effective rates of protection rather than nominal tariffs. Effective rates of protection capture the overall effect of liberalization on producers in a given industry, accounting for tariff changes on industry inputs and outputs. Kume et al. (2003) provide effective rates of protection along with the nominal tariffs used in our main analysis. The magnitude of the changes in effective rates of protection is larger than for nominal tariffs, so the coefficients in Panel 3 smaller by the same proportion. Since versions of RT Rr based on effective rates of protection and nominal tariffs are nearly perfectly correlated (correlation = 0.993), the variation in earnings growth explained by both versions is nearly identical. Panel F calculates RT Rr including the nontradable sector with a tariff change value of 0. This measure ignores the fact that nontradable prices move with tradable prices (see Appendix A.5 and Panel B of Table 2), and in doing so underestimates the magnitude of the average liberalization-induced price change faced by each region. Because the magnitude of RT Rr is reduced, the coefficient estimates are inflated by the same proportion. Panel G omits industry fixed effects when calculating regional earnings premia. This maintains the national industry-level variation in earnings in the outcome measure, rather than restricting attention to the regional equilibrium earnings used in our main specifications. Panel H omits all controls from the earnings premium regressions, using simple average log earnings for workers in the relevant region. While the main analysis weights by the inverse of the squared standard error of the estimated growth in regional wage premium, Panel I weights all regions equally, and Panel J weights by 1991 formal employment. In all cases, the effects grow substantially over time, as in our main specification. In fact, in all but two of these robustness tests, the long run effect of liberalization on earnings is larger than it is in the main specification. Table B7 shows that the formal employment results in Table 2 and Figure 4 are similarly robust. The panel labels correspond to Table B6, so see above for descriptions of each specification. Note that Panels G and H do not apply to employment, since they relate to earnings premia. Panel I also does not apply, because the main employment specification is unweighted, since RAIS contains the population of formally employed workers.

73

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

Table B6: Robustness: Regional log Formal Earnings Premia - 1995, 2000, 2005, 2010 Change in log Formal Earnings Premia:

1991-1995 (1)

Panel A: Main specification Regional tariff reduction (RTR)

-0.096 (0.120) Panel B: Long pre-trends (Census: 1970-80, 1980-91, RAIS: 1986-90) Regional tariff reduction (RTR) -0.243* (0.130) Panel C: Long pre-trends, earnings growth from 1992 to t Regional tariff reduction (RTR) -0.478*** (0.115) Panel D: RTR using formal employment industry weights Regional tariff reduction (RTR) -0.089 (0.189) Panel E: RTR using effective rates of protection Regional tariff reduction (RTR) -0.047 (0.076) Panel F: RTR including zero nontradable price change Regional tariff reduction (RTR) -0.798 (0.489) Panel G: Earnings premium without industry fixed effects Regional tariff reduction (RTR) 0.131 (0.148) Panel H: Earnings premium with no controls (mean log earnings) Regional tariff reduction (RTR) 0.317* (0.171) Panel I: Unweighted (equally weighted) Regional tariff reduction (RTR) -0.244 (0.169) Panel J: Weighted by 1991 formal employment Regional tariff reduction (RTR) -0.020 (0.128) Formal earnings pre-trend (86-90) ✓ State fixed effects (26) ✓

1991-2000 (2)

1991-2005 (3)

1991-2010 (4)

-0.529*** (0.141)

-1.294*** (0.139)

-1.594*** (0.169)

-0.770*** (0.184)

-1.498*** (0.186)

-1.814*** (0.199)

-1.009*** (0.197)

-1.737*** (0.195)

-2.039*** (0.215)

-0.358 (0.217)

-1.270*** (0.246)

-1.665*** (0.298)

-0.328*** (0.091)

-0.823*** (0.090)

-1.017*** (0.107)

-1.758*** (0.577)

-3.350*** (0.643)

-4.625*** (0.696)

-0.422*** (0.151)

-1.420*** (0.163)

-1.895*** (0.209)

0.046 (0.214)

-1.192*** (0.147)

-1.905*** (0.182)

-0.490** (0.197)

-1.074*** (0.215)

-1.546*** (0.224)

-0.345** (0.147) ✓ ✓

-1.217*** (0.143) ✓ ✓

-1.631*** (0.175) ✓ ✓

Negative coefficient estimates for the regional tariff reduction (RTR) imply larger declines in formal earnings in regions facing larger tariff reductions. 475 microregion observations, except Panels B and C, which use a more aggregate region definition with 405 observations for consistency with 1970 and 1980 Census data. Regional earnings premia calculated controlling for age, sex, education, and industry of employment except in Panels G and H. Standard errors (in parentheses) adjusted for 112 mesoregion clusters, except Panels B and C with 90 mesoregion clusters. Efficiency weighted by the inverse of the squared standard error of the estimated change in log formal earnings premium except in Panels I and J. See text for detailed description of each panel. *** Significant at the 1 percent, ** 5 percent, * 10 percent level.

74

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

Table B7: Robustness: Regional log Formal Employment - 1995, 2000, 2005, 2010 Change in log Formal Employment:

1991-1995 (1)

Panel A: Main specification Regional tariff reduction (RTR)

-1.900*** (0.422) Panel B: Long pre-trends (Census: 1970-80, 1980-91, RAIS: 1986-90) Regional tariff reduction (RTR) -1.157 (0.787) Panel C: Long pre-trends, earnings growth from 1992 to t Regional tariff reduction (RTR) -0.722 (0.804) Panel D: RTR using formal employment industry weights Regional tariff reduction (RTR) -1.728*** (0.598) Panel E: RTR using effective rates of protection Regional tariff reduction (RTR) -1.201*** (0.274) Panel F: RTR including zero nontradable price change Regional tariff reduction (RTR) -5.677*** (1.571) Panel G: Not applicable Panel H: Not applicable Panel I: Not applicable Panel J: Weighted by 1991 formal employment Regional tariff reduction (RTR) Formal earnings pre-trend (86-90) State fixed effects (26)

-1.195*** (0.195) ✓ ✓

1991-2000 (2)

1991-2005 (3)

1991-2010 (4)

-3.533*** (0.582)

-4.517*** (0.685)

-4.663*** (0.679)

-3.393*** (0.930)

-4.687*** (1.019)

-4.537*** (1.007)

-2.957*** (0.972)

-4.252*** (1.084)

-4.102*** (1.070)

-2.690*** (0.793)

-4.491*** (0.782)

-4.362*** (0.789)

-2.336*** (0.369)

-2.959*** (0.438)

-3.074*** (0.430)

-8.574*** (2.441)

-10.874*** (2.752)

-12.507*** (2.750)

-2.119*** (0.358) ✓ ✓

-3.406*** (0.340) ✓ ✓

-2.842*** (0.397) ✓ ✓

Negative coefficient estimates for the regional tariff reduction (RTR) imply larger declines in formal employment in regions facing larger tariff reductions. 475 microregion observations, except Panels B and C, which use a more aggregate region definition with 405 observations for consistency with 1970 and 1980 Census data. Panel labels correspond to Table B6, so Panels G and H, which relate to earnings premia, are not applicable here, nor is Panel I, since the main specification is unweighted. Standard errors (in parentheses) adjusted for 112 mesoregion clusters, except Panels B and C with 90 mesoregion clusters. See text for detailed description of each panel. *** Significant at the 1 percent, ** 5 percent, * 10 percent level.

75

Trade Liberalization and Regional Dynamics

B.8 B.8.1

Dix-Carneiro and Kovak

Potential Confounders Post-Liberalization Tariff Reductions

We calculate post-liberalization regional tariff reductions as in (2), but use tariff reductions between 1995 and year t > 1995. Because the Kume et al. (2003) data end in 1998, we use UNCTAD TRAINS to construct post-liberalization tariff reductions. The TRAINS data are reported by 6digit HS codes. In order to maintain as much industry variation as possible, we created an industry mapping from HS codes to Census industry codes, which yields 44 consistently identifiable tradable industries. This provides more industry detail than the main industry definition in Table A1. The concordance is available upon request. Panel B of Table B8 includes these post-liberalization tariff reduction controls in the regional earnings growth regression. The post-liberalization control has the expected negative coefficient, but its inclusion has very little effect on the liberalization coefficient. B.8.2

Real Exchange Rates

We construct regional real exchange rate shocks as follows. We begin with real exchange rates between Brazil and its trading partners, calculated from Revision 7.1 of the Penn World Tables. We then calculate each country’s 1989 shares of Brazil’s imports and exports in each industry using Comtrade. As in the prior section, we use the industry definition mapping from HS codes to Census industries. Industry-specific real exchange rates are weighted averages of country-specific real exchange rates, weighting either by the 1989 import share or export share. We define industrylevel real exchange rate shocks as the change in log industry real exchange rate from 1990 to each subsequent year. Finally we create regional real exchange rate shocks as weighted averages of industry real exchange rate shocks, where the region’s industry weights are given by the 1991 industry distribution of employment. Panel C of Table B8 includes both the import-weighted and export-weighted real exchange rate controls. With these controls, the earnings effects grow even more than in the main specification. B.8.3

Privatization

Substantial privatization in Brazil began in 1991 with the administration of President Collor, but significantly increased during President Cardoso’s administration (1995-2002). Beginning in 1995, the RAIS data allow us to identify as state-owned any firm at least partly owned by the government. In panels D and E of Table B8, we include different controls for the regional effects of privatization. Panel D includes quartile indicators for the 1995 share of regional employment in state-owned firms, controlling flexibly for the initial share of employment subject to potential privatization. Panel E controls for the change in state-owned firm employment share from 1995 to subsequent year t. In both cases, the privatization controls have no meaningful effect on the RT Rr coefficients.

76

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

Table B8: Potential Confounders: Regional log Formal Earnings Premia - 1995, 2000, 2005, 2010 Change in log Formal Earnings Premia:

Panel A: Main specification Regional tariff reduction (RTR)

1991-1995 (1)

1991-2000 (2)

1991-2005 (3)

1991-2010 (4)

-0.096 (0.120)

-0.529*** (0.141)

-1.294*** (0.139)

-1.594*** (0.169)

-0.096 (0.120) n/a

-0.542*** (0.137) -3.705 (3.273)

-1.234*** (0.171) -2.415 (3.872)

-1.809*** (0.218) -2.124 (1.425)

-0.113 -0.482*** (0.118) (0.161) Import-weighted real exchange rate 0.136 0.570* (0.133) (0.327) Export-weighted real exchange rate 0.051 -0.164 (0.160) (0.280) Panel D: Privatization: initial state-owned employment share Regional tariff reduction (RTR) -0.090 -0.490*** (0.134) (0.163) Quartile indicators, 1995 state-owned ✓ ✓ employment share distirbution Panel E: Privatization: change in state-owned employment share, 1995 to t Regional tariff reduction (RTR) -0.096 -0.514*** (0.120) (0.149) Change in state-owned employment share 0.095 (0.178) Formal earnings pre-trend (86-90) ✓ ✓ State fixed effects (26) ✓ ✓

-1.475*** (0.184) 0.243* (0.141) 0.084 (0.384)

-1.728*** (0.207) 0.374* (0.201) -0.166 (0.355)

-1.235*** (0.172) ✓

-1.580*** (0.205) ✓

-1.243*** (0.146) 0.286 (0.214) ✓ ✓

-1.558*** (0.182) 0.176 (0.227) ✓ ✓

Panel B: Post-liberalization tariff reductions Regional tariff reduction (RTR) Post-liberalization (1995 to t) regional tariff reductions Panel C: Exchange rates Regional tariff reduction (RTR)

Negative coefficient estimates for the regional tariff reduction (RTR) imply larger declines in formal earnings in regions facing larger tariff reductions. 475 microregion observations. Regional earnings premia calculated controlling for age, sex, education, and industry of employment. Standard errors (in parentheses) adjusted for 112 mesoregion clusters. Efficiency weighted by the inverse of the squared standard error of the estimated change in log formal earnings premium. See text for detailed description of each panel and for control construction. *** Significant at the 1 percent, ** 5 percent, * 10 percent level.

77

Trade Liberalization and Regional Dynamics

B.8.4

Dix-Carneiro and Kovak

Commodity Price Boom

Figures B4 - B6 show price indexes for major Brazilian export commodities from 1991 to 2010, using data from the Primary Commodity Price Series produced by the International Monetary Fund. Although there is some variability across commodities, there is little evidence of the commodity price boom prior to 2004. For example, in Figure B4, aggregate agricultural commodity prices were nearly identical in 1991 and 2003, and Brazil’s main agricultural exports (coffee, sugar, soy, and cotton) all have lower average prices in 2003 than in 1991. Natural resource and meat commodities in Figures B5 and B6 exhibit similar patterns, with relatively flat series prior to 2003 and substantial growth for many commodities starting in 2004. Note the contrast between this time series pattern and the effects of regional tariff reductions on formal earnings and employment growth in Figures 3 and 4. The earnings effects grow steadily from 1996 to 2003, in spite of the fact that most commodity prices actually fell a bit during that time span. Commodity prices start growing very quickly in 2004 and later, during which the earnings effects start to level off. A similar argument applies to the employment effects, which grow from 1994 to 2004 and level off subsequently. Thus, the timing of the commodity price boom does not conform with the timing of the earnings and employment effects, making it very unlikely that commodity prices drive our results, particularly before 2004. To reinforce this time-series evidence, in Table B9 we implement a wide variety of tests to rule out the commodity price boom as a potential confounder. Panel A reproduces the main specification for comparison. Panels B and C respectively restrict the sample of regions to those with below median and bottom quartile employment shares in agriculture and mining, the sectors affected by the commodity price boom. Note that mining includes fuel extraction. When focusing on regions with minimal exposure to commodity sectors, we find even larger growth in the effects of liberalization on earnings than in the entire sample. In Panel D, we maintain all regions, but only calculate earnings premia for workers employed in the manufacturing sector, omitting workers in commodity and nontradables sectors. Once again, the earnings effects continue to grow substantially over time given this restriction. As an aside, note that earnings in the manufacturing sector, which most directly experienced the effects of trade liberalization, exhibit significant effects on impact, in 1995. This finding suggests that the very short-run effects of liberalization were concentrated in the industries facing the largest tariff cuts, but that the earnings effects spread out to other sectors over time through labor market equilibrium. In Panels E and F, rather than restricting the sample of regions or workers, we control for the commodity price boom directly. We utilize the regional commodity price shocks constructed by Ad˜ao (2015). Special thanks to Rodrigo Ad˜ao for sharing his data and code. See Appendix C in Ad˜ao (2015) for details on the data source and his equation (16) for the shock construction. To summarize, he calculates commodity-specific changes in log price from 1991 to each subsequent year using data from the Commodity Research Bureau, and constructs regional weighted averages of these commodity price shocks. The weights reflect each commodity sector’s share of total labor payments in all commodity sectors in the region in 1991. Because this measure does not incorporate regional variation in overall exposure to commodity price changes, we also flexibly control for the regional importance of commodities by including quartile indicators for the region’s 1991 share of regional employment in agriculture and mining. As seen in Panel E, controlling directly for these commodity price movements has little influence on the increasing profile of earnings effects, either before or after the beginning of commodity price boom.

78

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

We implement a similar exercise in Panel F of Table B9 using more detailed commodity price information from the IMF Primary Commodity Price Series, a subset of which is shown in Figures B4 - B6. While Ad˜ ao’s measure uses only 6 aggregate commodity indexes, our alternative measure uses 19 separate indexes. As an example, this detail allows us to distinguish between the commodity price changes faced by regions specialized in coffee vs. sugar, which are grouped together under Ad˜ao’s classification. Table B10 shows our mapping from commodity industry codes in the 1991 Census to the IMF price indexes. We calculate the change in log price index from 1991 to each subsequent year for each IMF commodity and then generate regional weighted averages of these price changes, where weights reflect the relevant commodity’s share of regional employment in 1991. As seen in Panel F, including these commodity price controls has no substantive effect upon the relationship between regional tariff reductions and regional earnings, either before or after the beginning of commodity price boom. Because our regional commodity price shock measure already incorporates the overall share of commodity industries in regional employment, we do not additionally control for that share, though doing so has no substantive effect on the results. The rise of China appears to have played a substantial role in driving up commodity prices in the late 2000s. As a final test of the commodity price boom hypothesis, we follow Costa et al. (2016), who study the regional effects of import competition from China and Chinese demand for exports. Rather than focusing on commodity prices, Costa et al. study the effects of import and export quantity shocks, along the lines of Autor et al. (2013). They construct industry-level Chinese import supply (IS) and export demand (XD) shocks as the growth in industry imports from or exports to China from 2000 to 2010, divided by Brazilian employment in the industry in 2000. They then generate regional weighted average shocks using the year 2000 industry distribution of employment in each region. Finally, they instrument for these shocks using similar measures based on the growth in Chinese trade to countries other than Brazil. Special thanks to Francisco Costa for providing us with their shock and instrument measures. Because Costa et al. examine shocks and outcomes between 2000 and 2010, in Panel A of Table B11 we provide our baseline earnings estimate for this time period, with a base year of 2000 rather than 1991. We use a slightly more aggregate region definition to match theirs, yielding 405 region observations. The coefficient estimate of -1.068 in column (1) is nearly identical to the difference between the estimates for 2000 and 2010 in columns (3) and (6), respectively, of Panel A in Table 2 of -1.065. The slight difference results from the difference in region definitions between the two tables. In columns (2)-(4) of Table B11, we introduce the Chinese import supply and export demand shocks, instrumented following Costa et al.. The two shocks have the expected sign, with increased import competition lowering regional earnings and increased export demand increasing them (very slightly), though only the import supply shock is statistically different from zero. This result might seem surprising, given that Costa et al. find significant effects of export demand on wages. However, when they control for the regional composition of workers and for outcome pretrends, as we do here, the export result loses statistical significance (see their Table 2, Panel B, column (5) in their paper). When we include these controls, they have only a very small effect on our coefficient of interest, further confirming that the divergence in earnings growth between regions facing larger and smaller tariff reductions was not driven by China’s effects on commodity markets.

79

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

Figure B4: Agricultural Commodity Prices - 1991-2010 3.5   Coffee  

Sugar  

Soy  

Co7on  

Agriculture  

3  

2.5  

2  

1.5  

1  

Monthly price series from IMF Primary Commodity Price Series, rescaled to equal 100 in January 1991.

80

2010  

2009  

2008  

2007  

2006  

2005  

2004  

2003  

2002  

2001  

2000  

1999  

1998  

1997  

1996  

1995  

1994  

1993  

1992  

0  

1991  

0.5  

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

Figure B5: Resource Commodity Prices - 1991-2010 8   Wood  

Fuel  

Metals  

7  

6  

5  

4  

3  

2  

2010  

2009  

2008  

2007  

2006  

2005  

2004  

2003  

2002  

2001  

2000  

1999  

1998  

1997  

1996  

1995  

1994  

1993  

1992  

0  

1991  

1  

Monthly price series from IMF Primary Commodity Price Series, rescaled to equal 100 in January 1991, except the fuel index which begins in 1992 and is rescaled to 1 in January 1992.

81

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

Figure B6: Meat Commodity Prices - 1991-2010 4   Beef  

Fish  

Poultry  

3.5  

3  

2.5  

2  

1.5  

1  

Monthly price series from IMF Primary Commodity Price Series, rescaled to equal 100 in January 1991.

82

2010  

2009  

2008  

2007  

2006  

2005  

2004  

2003  

2002  

2001  

2000  

1999  

1998  

1997  

1996  

1995  

1994  

1993  

1992  

0  

1991  

0.5  

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

Table B9: Commodity Price Boom: Regional log Formal Earnings Premia - 1995, 2000, 2005, 2010 Change in log Formal Earnings Premia:

1991-1995 (1)

Panel A: Main specification Regional tariff reduction (RTR)

1991-2000 (2)

-0.096 -0.529*** (0.120) (0.141) Panel B: Below-median agriculture/mining employment share (238 obs) Regional tariff reduction (RTR) -0.017 -0.534*** (0.148) (0.165) Panel C: Bottom quartile agriculture/mining employment share 118 obs) Regional tariff reduction (RTR) 0.006 -0.347 (0.267) (0.262) Panel D: Manufacturing sector earnings Regional tariff reduction (RTR) -0.501*** -0.965*** (0.158) (0.192) Panel E: Direct commodity price controls per Adao (2015) Regional tariff reduction (RTR) -0.052 -0.290 (0.259) (0.257) Regional commodity price shocks 0.033 -0.039 (Adao 2015) (0.207) (0.167) Quartile indicators, 1991 agriculture/mining ✓ ✓ employment share distribution Panel F: Direct commodity price controls using detailed commodity price data (IMF) Regional tariff reduction (RTR) 0.023 -0.591*** (0.143) (0.188) Regional commodity price shocks 0.179 0.160 (IMF data) (0.120) (0.266) Formal earnings pre-trend (86-90) ✓ ✓ State fixed effects (26) ✓ ✓

1991-2005 (3)

1991-2010 (4)

-1.294*** (0.139)

-1.594*** (0.169)

-1.424*** (0.163)

-1.829*** (0.213)

-1.379*** (0.280)

-2.153*** (0.373)

-1.878*** (0.214)

-2.252*** (0.262)

-1.269*** (0.276) 0.118 (0.092) ✓

-1.926*** (0.372) 0.045 (0.127) ✓

-1.210*** (0.137) 0.421 (0.277) ✓ ✓

-1.718*** (0.330) -0.069 (0.149) ✓ ✓

Negative coefficient estimates for the regional tariff reduction (RTR) imply larger declines in formal earnings in regions facing larger tariff reductions. 475 microregion observations unless otherwise noted (Panels B and C). Regional earnings premia calculated controlling for age, sex, education, and industry of employment. Standard errors (in parentheses) adjusted for 112 mesoregion clusters. Efficiency weighted by the inverse of the squared standard error of the estimated change in log formal earnings premium. See text for detailed description of each panel and for control construction. *** Significant at the 1 percent, ** 5 percent, * 10 percent level.

83

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

Table B10: Mapping from Commodity Industries to IMF Price Indexes 1991 Census Industry (atividade) 11 Agave 12 Cotton 13 Rice 14 Banana 15 Cocoa 16 Coffee 17 Sugar cane 18 Tobacco 19 Cassava 20 Corn 21 Soybeans 22 Wheat 23 Horticulture and floriculture 24 Forestry 25 Other agricultural products 26 Livestock 27 Aviculture 28 Beekeeping and Sericulture 29 Other livestock 31 Rubber 32 Yerba mate 33 Plant fibres 34 Fruits, oilseeds, and waxes 35 Wood 36 Charoal 37 Other harvesting activities 41 Fishing 42 Aquaculture 51 Oil and natural gas mining 52 Coal mining 55 Metallic mineral panning and deposition 56 Radioactive mineral mining 58 Metallic mineral mining (except those in other categories) 581 Agriculture and livestock services

84

IMF Index PRAWM PCOTTIND PRICENPQ PBANSOP PCOCO PCOFFOTM PSUGAISA PRAWM PRAWM PMAIZMT PSOYB PWHEAMT PRAWM PSAWORE PRAWM PBEEF PPOULT PRAWM PRAWM PRUBB PRAWM PRAWM PRAWM PSAWORE PCOALAU PRAWM PFISH PFISH PNRG PCOALAU PMETA PURAN PMETA PRAWM

Agricultural Raw Materials Cotton Rice Bananas Cocoa beans Coffee Sugar Agricultural Raw Materials Agricultural Raw Materials Maize (corn) Soybeans Wheat Agricultural Raw Materials Soft Sawnwood Agricultural Raw Materials Beef Poultry (chicken) Agricultural Raw Materials Agricultural Raw Materials Rubber Agricultural Raw Materials Agricultural Raw Materials Agricultural Raw Materials Soft Sawnwood Coal Agricultural Raw Materials Fishmeal Fishmeal Fuel (Energy) Coal Metals Uranium Metals Agricultural Raw Materials

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

Table B11: Regional log Formal Earnings Premia with Costa et al (2015) controls Change in log Formal Earnings Premia, 2000-2010:

Regional tariff reduction (RTR) Formal earnings pre-trend (86-90)

(1) OLS -1.068*** (0.111) -0.077 (0.053)

China import supply (Costa et al. 2015)

(2) IV -0.929*** (0.123) -0.064 (0.055) -0.034*** (0.010)

China export demand (Costa et al. 2015)

-1.069*** (0.107) -0.076 (0.051)

✓

✓

0.001 (0.002) ✓

0.733

0.738 22.16

0.733 441.6

State fixed effects (26) R-squared 1st stage F (Kleibergen-Paap)

(3) IV

(4) IV -0.931*** (0.122) -0.063 (0.055) -0.034*** (0.010) 0.001 (0.002) ✓ 0.739 11.31

Negative coefficient estimates for the regional tariff reduction (RTR) imply larger declines in formal earnings in regions facing larger tariff reductions. See text for description of China import supply and export demand controls and associated instruments from Costa et al. (2015). First stage partial F-statistics reported in brackets. 405 microregion observations. Regional earnings premia calculated controlling for age, sex, education, and industry of employment. Standard errors (in parentheses) adjusted for 90 mesoregion clusters. Efficiency weighted by the inverse of the squared standard error of the estimated change in log formal earnings premium. *** Significant at the 1 percent, ** 5 percent, * 10 percent level.

85

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

B.9

Earnings and Employment Sample Splits

Tables B12 and B13 present earnings results splitting the sample of workers into those employed in the tradable sector (Panel B), those employed in the nontradable sector (Panel C), more educated (Panel D), and less educated (Panel E). Note that earnings and employment effects grow for all subsamples. The employment effects are concentrated in the tradable sector and among less skilled workers, though panels D and E in both tables should be interpreted with care, as the regional tariff reduction shocks are derived from a model with a single type of labor. For a more general model with two skill types, see Dix-Carneiro and Kovak (2015a).

Table B12: Sample Splits: Regional log Formal Earnings Premia - 1995, 2000, 2005, 2010 Change in log Formal Earnings Premia:

Panel A: Full sample Regional tariff reduction (RTR) Panel B: Tradable sector workers Regional tariff reduction (RTR) Panel C: Nontradable sector workers Regional tariff reduction (RTR)

1991-1995 (1)

1991-2000 (2)

1991-2005 (3)

1991-2010 (4)

-0.096 (0.120)

-0.529*** (0.141)

-1.294*** (0.139)

-1.594*** (0.169)

-0.287* (0.149)

-0.754*** (0.184)

-1.623*** (0.203)

-1.934*** (0.254)

-0.060 (0.150)

-0.389** (0.179)

-1.143*** (0.169)

-1.401*** (0.183)

-0.539*** (0.173)

-1.611*** (0.192)

-2.053*** (0.224)

-0.626*** (0.143) ✓ ✓

-1.354*** (0.137) ✓ ✓

-1.758*** (0.180) ✓ ✓

Panel D: More educated workers (high school or more) Regional tariff reduction (RTR) 0.310* (0.160) Panel E: Less educated workers (less than high school) Regional tariff reduction (RTR) -0.218* (0.121) Formal earnings pre-trend (86-90) ✓ State fixed effects (26) ✓

Negative coefficient estimates for the regional tariff reduction (RTR) imply larger declines in formal earnings in regions facing larger tariff reductions. 475 microregion observations. Regional earnings premia calculated controlling for age, sex, education, and industry of employment. Standard errors (in parentheses) adjusted for 112 mesoregion clusters. Efficiency weighted by the inverse of the squared standard error of the estimated change in log formal earnings premium. *** Significant at the 1 percent, ** 5 percent, * 10 percent level.

86

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

Table B13: Sample Splits: Regional log Formal Employment - 1995, 2000, 2005, 2010 Change in log Formal Employment:

Panel A: Full sample Regional tariff reduction (RTR) Panel B: Tradable sector workers Regional tariff reduction (RTR) Panel C: Nontradable sector workers Regional tariff reduction (RTR)

1991-1995 (1)

1991-2000 (2)

1991-2005 (3)

1991-2010 (4)

-1.900*** (0.422)

-3.533*** (0.582)

-4.517*** (0.685)

-4.663*** (0.679)

-5.790*** (0.850)

-8.416*** (0.993)

-10.097*** (1.101)

-10.156*** (1.140)

0.726 (0.455)

-0.733 (0.664)

-1.500* (0.778)

-1.600** (0.749)

-1.219* (0.644)

-1.637** (0.772)

-2.195*** (0.733)

-4.556*** (0.626) ✓ ✓

-6.328*** (0.770) ✓ ✓

-6.910*** (0.787) ✓ ✓

Panel D: More educated workers (high school or more) Regional tariff reduction (RTR) 0.141 (0.450) Panel E: Less educated workers (less than high school) Regional tariff reduction (RTR) -2.507*** (0.465) Formal earnings pre-trend (86-90) ✓ State fixed effects (26) ✓

Negative coefficient estimates for the regional tariff reduction (RTR) imply larger declines in formal employment in regions facing larger tariff reductions. 475 microregion observations. Standard errors (in parentheses) adjusted for 112 mesoregion clusters. *** Significant at the 1 percent, ** 5 percent, * 10 percent level.

87

Trade Liberalization and Regional Dynamics

B.10

Dix-Carneiro and Kovak

Regional Change in log Imports and Exports

This section presents versions of Figure 5 and Table 6 using alternative trade quantity measures reflecting regional weighted averages of the change in log imports or exports rather than the change in imports or exports per worker. These alternative measures are presented only for descriptive purposes and as a statistical robustness test. We use the change in trade per worker in the main text both because it is theoretically justified, as shown by Autor et al. (2013), and because it intuitively captures the effects of changing trade quantities on local labor market outcomes. Figures B7 and B8, which show scatter plots relating the industry-level change in trade per worker and the change in log trade for imports and exports in 2000 and 2010. Markers are proportional to 1991 employment, and industry labels correspond to Table A1. These plots show that the change in log trade often deviates substantially from the change in trade per worker. This deviation occurs primarily in industries with relatively small trade flows and relatively large employment. As an example, consider the Wood, Furniture, and Peat industry (code 14) in Panel A of Figure B7. In this industry, Brazil imported R$71 million in 1990 and R$249 million in 2000 (all values in year 2005 Reais). This very large proportional growth in imports corresponds to the large value for the change in log imports of 1.25. However, initial employment in this industry was also quite large, 822,579, so the change in imports per worker was only R$216, much smaller than the values in the thousands or tens of thousands in other industries. Therefore, although the amount of imports increased very much in proportional terms, it was still insignificant compared to the number of workers in the industry. The change in trade per worker captures the relative scale of trade and employment, while the change in log trade does not. Nonetheless, figure B9 shows the relationships between RT Rr and the regional change in log imports and exports (paralleling Figure 5). See (18) and (19) in Appendix A.6 for details on constructing the change in log trade measures. Regions facing larger tariff reductions experience larger increases in log imports and larger declines in log exports. The magnitude of each effect grows over time, suggesting that perhaps slow trade quantity responses could explain the slow growth of regional earnings and employment effects in Figures 3. We demonstrate that this is not the case by directly controlling for the regional change in log import and export measures when estimating the effect of RT Rr on regional earnings growth. The specifications in Table B14 parallel those in Table 6, using the alternative change-in-log measures of trade flows for both the controls and the instruments. In all cases, the earnings effects of liberalization grow even more when controlling for import and export quantity growth than in the baseline specification in Panel A. The relevant Stock and Yogo (2005) critical value for the first-stage F-statistic is 21, so Panel C exhibits a potential weak instruments problem. We therefore present two additional sets of results in Table B15, using the change in trade per worker measure when calculating the instruments rather than the change in log trade flow measure. The weak instrument issue is not longer present, and the effects of RT Rr on regional earnings still increase substantially over time. As with the standard trade-per-worker trade flow measures considered in the main text, the analysis using the alternative change-in-log trade flow measure rules out slow import or export responses as the mechanism driving the slowly growing earnings effects. This is to be expected given the discussion at the beginning of this section, since the change in log trade measure does not well capture the effects of changing trade flows on workers’ labor market outcomes.

88

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

Figure B7: Change in Trade Per Worker vs. Change in log Trade - 1990-2000

20000

30000

32

12

17

10000

10 8

21

0

change in imports per worker, 1990-2000

40000

Panel A: Imports

2

24

4

25

-.5

0

5

15

22

23 1

.5

14

1

1.5

change in log imports, 1990-2000 Omits the 4 smallest industries, with less than 150,000 employees in 1991.

10000

15000

12

5000

32

8

17

15

22

-.5

23

5 25 2

10 14

24

0

change in exports per worker, 1990-2000

20000

Panel B: Exports

421 1

0

.5

1

1.5

change in log exports, 1990-2000 Omits the 4 smallest industries, with less than 150,000 employees in 1991.

Each point is an industry, with labels corresponding to Table A1. The y-axis measures the change in trade per worker initially employed in the industry (in 1991) and the x-axis measures the change in log trade. The figure omits the 4 smallest industries in terms of 1991 employment, which often fall well outside the scale shown. Because they are small, they receive very little weight in the regional analysis that forms this paper’s main analysis. Marker size proportional to 1991 employment.

89

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

Figure B8: Change in Trade Per Worker vs. Change in log Trade - 1990-2010

60000

80000

32

12

8

40000

17

20000

10

21 5

0

change in imports per worker, 1990-2010

100000

Panel A: Imports

15 1

2 25

24

.5

22

4

1

23

14

1.5

2

2.5

change in log imports, 1990-2010 Omits the 4 smallest industries, with less than 150,000 employees in 1991.

60000

2

40000

12

20000

change in exports per worker, 1990-2010

Panel B: Exports

8

1725 15

32 5

10 144

0

24 23

-1

21 1

22

0

1

2

change in log exports, 1990-2010 Omits the 4 smallest industries, with less than 150,000 employees in 1991.

Each point is an industry, with labels corresponding to Table A1. The y-axis measures the change in trade per worker initially employed in the industry (in 1991) and the x-axis measures the change in log trade. The figure omits the 4 smallest industries in terms of 1991 employment, which often fall well outside the scale shown. Because they are small, they receive very little weight in the regional analysis that forms this paper’s main analysis. Marker size proportional to 1991 employment.

90

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

Figure B9: Regional Imports and Exports, Change in log Measure - 1992-2010 10  

Liberaliza5on                              Post-­‐liberaliza5on   (chg.  from  1991)    

Imports  

5  

0   1991   1992   1993   1994   1995   1996   1997   1998   1999   2000   2001   2002   2003   2004   2005   2006   2007   2008   2009   2010  

-­‐5  

-­‐10  

Exports  

-­‐15  

-­‐20  

Each point reflects an individual regression coefficient, θˆt , following (3), where the dependent variable is the change in regional imports (blue circles) or exports using the change in log measures described in (18) and (19) in Appendix A.6. The independent variable is the regional tariff reduction (RTR), defined in (2). Note that the RTR always reflects tariff reductions from 1990-1995. All regressions include state fixed effects, but do not include pre-liberalization trends due to a lack of Comtrade trade data before 1989. Positive (negative) estimates imply larger increases in trade flow in regions facing larger (smaller) tariff reductions. Vertical bar indicates that liberalization was complete by 1995. Dashed lines show 95 percent confidence intervals. Standard errors adjusted for 112 mesoregion clusters.

91

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

Table B14: Slow Response of Imports or Exports, Change in log Measure - 1995, 2000, 2005, 2010 (Part 1 of 2) Change in log Formal Earnings Premia:

Panel A: Main specification Regional tariff reduction (RTR)

Panel  B:  Controls  for  trade  quantitities  (OLS) Regional tariff reduction (RTR)

1991-1995 (1)

-1.294*** (0.139)

-1.594*** (0.169)

-0.142 (0.124)

-0.581*** -1.476*** (0.142) (0.151) 0.025** (0.012) -0.009 (0.008)

-1.887*** (0.212)

-0.236 (0.145)

-0.608*** -1.479*** (0.150) (0.343) -0.009 (0.062) -0.052 (0.039) 19.71

-2.158*** (0.579)

-0.227 (0.146)

-0.570*** -1.316*** (0.160) (0.306) -0.044 (0.049) -0.059* (0.032) 61.45

-1.985*** (0.509)

Import  quantity  control  (change  in  log) Export  quantity  control  (change  in  log) First-­‐stage  F  (Kleibergen-­‐Paap)

Import  quantity  control  (change  in  log) Export  quantity  control  (change  in  log) First-­‐stage  F  (Kleibergen-­‐Paap) Formal earnings pre-trend (86-90) State fixed effects (26)

1991-2010 (4)

-0.529*** (0.141)

Export  quantity  control  (change  in  log)

Panel  D:  Colombia  IV   Regional tariff reduction (RTR)

1991-2005 (3)

-0.096 (0.120)

Import  quantity  control  (change  in  log)

Panel  C:  Latin  America  IV Regional tariff reduction (RTR)

1991-2000 (2)

✓ ✓

✓ ✓

✓ ✓

✓ ✓

Negative coefficient estimates for the regional tariff reduction (RTR) imply larger declines in formal earnings in regions facing larger tariff reductions. Panel A replicates the earnings results in columns (3) and (6) of Table 2 and in Figure 3. Panels B-D control for regional import and export quantities, calculated using the change in log trade flows; see Appendix A.6 for details. These panels stack the data across years, allowing the effect of RT Rr to vary over time but fixing the import and export quantity coefficients over time. We instrument for the potentially endogenous import and export controls using regional measures of commodity price growth from Ad˜ ao (2015) and with regional trade flows for other countries. We consider the combination of Argentina, Chile, Colombia, Paraguay, Peru, and Uruguay (“Latin America”) Colombia alone. In each case, we measure imports and exports between these countries and the rest of the world, excluding Brazil. This gives us 2 endogenous variables and 57 instruments (= 3 instruments × 19 years). First-stage Kleinbergen-Paap F statistics are shown, for comparison to the Stock and Yogo (2005) critical value of 21 to reject 5 percent bias relative to OLS. Standard errors (in parentheses) adjusted for 112 mesoregion clusters. Efficiency weighted by the inverse of the squared standard error of the estimated change in log formal earnings premium. *** Significant at the 1 percent, ** 5 percent, * 10 percent level.

92

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

Table B15: Slow Response of Imports or Exports, Change in log Measure - 1995, 2000, 2005, 2010 (Part 2 of 2) Change in log Formal Earnings Premia:

1991-1995 (1)

Panel  E:  Latin  America  IV  (change  in  trade  per  worker) Regional tariff reduction (RTR) -0.063 (0.147) Import  quantity  control  (change  in  log) Export  quantity  control  (change  in  log) First-­‐stage  F  (Kleibergen-­‐Paap) Panel  F:  Colombia  IV  (change  in  trade  per  worker) Regional tariff reduction (RTR)

-0.096 (0.120)

1991-2000 (2)

-0.487*** -1.143*** (0.156) (0.333) -0.022 (0.050) 0.005 (0.031) 31.44

-0.529*** (0.141)

Import  quantity  control  (change  in  log)

-1.294*** (0.139)

1991-2010 (4)

-1.364** (0.561)

-1.594*** (0.169)

-0.045 (0.055) -0.026 (0.026) 61.58

Export  quantity  control  (change  in  log) First-­‐stage  F  (Kleibergen-­‐Paap) Formal earnings pre-trend (86-90) State fixed effects (26)

1991-2005 (3)

✓ ✓

✓ ✓

✓ ✓

✓ ✓

See Table B14 for general notes. Because the instrument in Panel C of Table B14 was marginally weak, Panels E and F present versions using instruments based on the change in imports per worker, while the trade quantity controls are calculated using the change in log. Both specifications reject the weak instrument concern.

93

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

B.11

Overall Employment

Table B16 shows the effect of liberalization on overall regional employment, including both formally and informally employed workers. We use Census data to capture informally employed individuals, and control for 1980-1991 and 1970-1980 outcome pre-trends. The estimates vary substantially across specifications and all but one are insignificantly different from zero. These results provide little evidence in favor of overall employment as a potential source of agglomeration economies.

Table B16: Regional log Overall Employment - 2000, 2010 Change in log Overall Employment: (1) Regional tariff reduction (RTR) Overall employment pre-trend (80-91)

0.203 (0.209) 0.329** (0.136)

Overall employment pre-trend (70-80) State fixed effects (26) R-squared

✓ 0.563

1991-2000 (2) -0.419 (0.450)

0.221*** (0.071) ✓ 0.479

(3) -0.272 (0.260) 0.281** (0.130) 0.120*** (0.035) ✓ 0.585

(4) 0.657** (0.314) 0.538*** (0.202)

✓ 0.574

1991-2010 (5) -0.478 (0.683)

0.385*** (0.093) ✓ 0.484

(6) -0.265 (0.410) 0.454** (0.189) 0.230*** (0.052) ✓ 0.609

Positive (negative) coefficient estimates for the regional tariff reduction imply larger increases (decreases) in overall employment in regions facing larger tariff reductions. Outcomes calculated using Census data. 405 microregion observations. Efficiency weighted by the inverse of the squared standard error of the dependent variable estimate. Pre-trends computed for 1980-1991 and 1970-1980. Standard errors (in parentheses) adjusted for 112 mesoregion clusters. *** Significant at the 1 percent, ** 5 percent, * 10 percent level.

94

Trade Liberalization and Regional Dynamics

B.12

Dix-Carneiro and Kovak

Capital Adjustment Confidence Intervals

Figures B10 - B12 show the capital adjustment profiles in Figure 6, including 95-percent confidence intervals, which were omitted from Figure 6 for clarity.

Figure B10: Capital Adjustment Quantification - Low ζ - 1992-2010 1.5  

Liberaliza6on                              Post-­‐liberaliza6on   Pre-­‐liberaliza6on   (chg.  from  1991)     (chg.  from  1986)  

1.0  

0.5  

0.0   1992   1993   1994   1995   1996   1997   1998   1999   2000   2001   2002   2003   2004   2005   2006   2007   2008   2009   2010   -­‐0.5  

-­‐1.0  

-­‐1.5  

-­‐2.0  

Capital  (establishments)  adjustment,  ζ  =  0.152    

-­‐2.5  

-­‐3.0  

Each point reflects an individual regression coefficient, θˆt , following (3). The dependent variable is capital’s contribution to overall adjustment, using the number of regional formal establishments as a proxy for regional capital. This figure shows the profile using the low estimate of ζ = 0.152. The independent variable is the regional tariff reduction (RTR), defined in (2). Note that the RTR always reflects tariff reductions from 1990-1995. All regressions include state fixed effects, and post-liberalization regressions control for the 1986-1990 outcome pre-trend. Negative estimates imply larger declines in the number of establishments in regions facing larger tariff reductions. Vertical bar indicates that liberalization was complete by 1995. Dashed lines show 95 percent confidence intervals. Standard errors adjusted for 112 mesoregion clusters.

95

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

Figure B11: Capital Adjustment Quantification - Mid ζ - 1992-2010 1.5  

Liberaliza6on                              Post-­‐liberaliza6on   Pre-­‐liberaliza6on   (chg.  from  1991)     (chg.  from  1986)  

1.0  

0.5  

0.0   1992   1993   1994   1995   1996   1997   1998   1999   2000   2001   2002   2003   2004   2005   2006   2007   2008   2009   2010   -­‐0.5  

-­‐1.0  

-­‐1.5  

Capital  (establishments)  adjustment,  ζ  =  0.349    

-­‐2.0  

-­‐2.5  

-­‐3.0  

Each point reflects an individual regression coefficient, θˆt , following (3). The dependent variable is capital’s contribution to overall adjustment, using the number of regional formal establishments as a proxy for regional capital. This figure shows the profile using the low estimate of ζ = 0.349. The independent variable is the regional tariff reduction (RTR), defined in (2). Note that the RTR always reflects tariff reductions from 1990-1995. All regressions include state fixed effects, and post-liberalization regressions control for the 1986-1990 outcome pre-trend. Negative estimates imply larger declines in the number of establishments in regions facing larger tariff reductions. Vertical bar indicates that liberalization was complete by 1995. Dashed lines show 95 percent confidence intervals. Standard errors adjusted for 112 mesoregion clusters.

96

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

Figure B12: Capital Adjustment Quantification - High ζ - 1992-2010 1.5  

Liberaliza6on                              Post-­‐liberaliza6on   Pre-­‐liberaliza6on   (chg.  from  1991)     (chg.  from  1986)  

1.0  

0.5  

0.0   1992   1993   1994   1995   1996   1997   1998   1999   2000   2001   2002   2003   2004   2005   2006   2007   2008   2009   2010   -­‐0.5  

-­‐1.0  

Capital  (establishments)  adjustment,  ζ  =  0.545    

-­‐1.5  

-­‐2.0  

-­‐2.5  

-­‐3.0  

Each point reflects an individual regression coefficient, θˆt , following (3). The dependent variable is capital’s contribution to overall adjustment, using the number of regional formal establishments as a proxy for regional capital. This figure shows the profile using the low estimate of ζ = 0.545. The independent variable is the regional tariff reduction (RTR), defined in (2). Note that the RTR always reflects tariff reductions from 1990-1995. All regressions include state fixed effects, and post-liberalization regressions control for the 1986-1990 outcome pre-trend. Negative estimates imply larger declines in the number of establishments in regions facing larger tariff reductions. Vertical bar indicates that liberalization was complete by 1995. Dashed lines show 95 percent confidence intervals. Standard errors adjusted for 112 mesoregion clusters.

97

Trade Liberalization and Regional Dynamics

B.13

Dix-Carneiro and Kovak

Exit by Establishment Size

Here we examine the relationship between establishment exit and RT Rr , separately by initial establishment size. We run the following specification at the establishment-year level, using the sample of all active establishments in 1991. Exitirt =

6 X k=1

βtk Sizeki

· RT Rr +

6 X

φk Sizeki + γt N T i + ϑt P reExitr1986−1990 + irt

(26)

k=1

where Sizeki is an indicator for whether establishment i fell into size bin k in 1991, N T i is an indicator for establishments in the nontradable sector, and P reExitr1986−1990 is a pre-trend control for the share of regional establishments in 1986 that shut down between 1986 and 1990. Figure B13 plots the βtk coefficients, with the relevant initial employment bin definitions shown on the right side. Although there is some variation across establishment sizes, with more exit among larger establishments than smaller establishments, it is clear that exit rates increased throughout the size distribution for establishments whose regions faced larger tariff declines.

98

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

Figure B13: Regional log Cumulative Formal Establishment Exit, by Establishment Size - 19922010 Liberaliza6on                              Post-­‐liberaliza6on   (chg.  from  1991)     0.9  

50-­‐99   20-­‐49  

0.7  

100+  

0.5  

10-­‐19   1-­‐4  

0.3  

5-­‐9   0.1  

1991   1992   1993   1994   1995   1996   1997   1998   1999   2000   2001   2002   2003   2004   2005   2006   2007   2008   2009   2010   -­‐0.1  

Plots the βtk coefficients in (26), estimated using the sample of all active establishments in 1991. The size range, indexed by k, is reported at the right side of each profile. Vertical bar indicates that liberalization was complete by 1995.

99

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

C

Model

C.1

Baseline Model

This section generalizes the specific-factors model of regional economies from Kovak (2013) to allow for changes in regional productivity (agglomeration economies) and changes in labor and capital inputs. The economy consists of many regions, indexed by r, which may produce goods in many industries, indexed by i. Production in each industry uses Cobb-Douglas technology with constant returns to scale and three inputs: labor, a fixed industry-specific factor, and capital. Labor, Lr , is assumed to be perfectly mobile between industries within a region. The industry-specific factor, Tri , is usable only in its respective region and industry and is fixed over time. Capital, Kri , is usable only in its respective region and industry but may change over time. Output of industry i in region r is  ϕi 1−ϕi 1−ζi Yri = Ari Lri Triζi Kri , (C1) where ϕi , ζi ∈ (0, 1). To allow for the possibility of agglomeration economies and factor adjustment, we allow Ari , Lr , and Kri to change over time. Goods and factor markets are perfectly competitive, and producers face exogenous prices Pi , common across regions and fixed by world prices and tariffs. Consider a particular region r, and suppress the region subscript. Let aLi , aT i , and aKi be the respective amounts of labor, specific factor, and capital used in producing one unit of Yi . Regional factor market clearing implies X aLi Yi = L, (C2) i

aT i Yi = Ti

∀i,

(C3)

aKi Yi = Ki

∀i.

(C4)

Perfect competition implies that the price equals factor payments, aLi w + aT i si + aKi Ri = Pi

∀i,

(C5)

where w is the wage, si is the specific-factor price, and Ri is the price of capital. Define x ˆ as the proportional change in x, and differentiate (C5). ˆ i = Pˆi + Aˆi (1 − ϕi )w ˆ + ϕi ζi sˆi + ϕi (1 − ζi )R

∀i,

(C6)

∀i.

(C7)

which uses the fact that, from cost minimization, (1 − ϕi )ˆ aLi + ϕi ζi a ˆT i + ϕi (1 − ζi )ˆ aKi = −Aˆi Differentiate the factor market clearing conditions. X ˆ λi (ˆ aLi + Yˆi ) = L,

(C8)

i

Yˆi = −ˆ aT i ˆi − a Yˆi = K ˆKi 100

∀i, ∀i,

(C9) (C10)

Dix-Carneiro and Kovak

Trade Liberalization and Regional Dynamics

where λi ≡ LLi is the share of regional labor allocated to industry i, and we use the fact that Tˆi = 0. With Cobb-Douglas production, the elasticity of substitution is one, so ˆi, a ˆKi − a ˆT i = sˆi − R

(C11)

ˆ i − w, a ˆLi − a ˆKi = R ˆ

(C12)

Combining (C8), (C10), and (C12) yields X X ˆi − w ˆ− ˆ i. λi R ˆ=L λi K i

(C13)

i

Combine (C9), (C10), and (C11) to yield ˆi + K ˆi sˆi = R

∀i.

(C14)

Plug this into (C6) and simplify. ˆ ˆ ˆi ˆ − ϕi ζi K ˆ i = Pi + Ai − (1 − ϕi )w R ϕi

(C15)

Finally, plug this into (C13), solve for w, ˆ and restore regional subscripts to yield the equilibrium relationship for regional wage changes, equation (9) in the main text. ! X X X ˆr − ˆ ri w ˆr = βri Pˆi + βri Aˆri − δr L λri (1 − ζi )K (C16) i

i

i

λri ϕ1i where βri ≡ P 1 j λrj ϕj

C.2

1 1 . j λrj ϕj

and δr ≡ P

Agglomeration Economies

As discussed in the main text, when examining agglomeration economies and quantifying the longrun effects of slow capital adjustment and agglomeration, we assume perfectly mobile capital in the ˆr = R ˆ ∀r), and identical technology across industries (ϕi = ϕ ∀i and ζi = ζ ∀i). The long run (R ˆ ri for the assumption of perfectly mobile capital allows us to substitute out the change in capital, K ˆ which is constant across industries and regions. change in its price, R, Start with the labor market clearing condition in (C8), substitute in the specific-factors clearing condition in (C9) and the Cobb-Douglas conditions in (C11) and (C12) to yield X ˆ λi sˆi − w ˆ = L. (C17) i

Rearrange the zero-profit condition in (C6) to solve for sˆi , sˆi =

1 ˆ ϕ(1 − ζ) ˆ 1 − ϕ (Pi + Aˆi ) − R− w, ˆ ϕζ ϕζ ϕζ

101

(C18)

Trade Liberalization and Regional Dynamics

Dix-Carneiro and Kovak

and plug it into (C17). Solving for w ˆ and restoring regional subscripts yields the following expression. X 1 ϕζ ˆ r − ϕ(1 − ζ) R ˆ w ˆr = βri (Pˆi + Aˆri ) − L (C19) 1 − ϕ(1 − ζ) 1 − ϕ(1 − ζ) 1 − ϕ(1 − ζ) i

where

λri ϕ1 βri ≡ P 1 = λLri j λrj ϕ

To incorporate agglomeration economies, we assume a constant elasticity agglomeration function, (11), and a constant labor supply elasticity, (12). Substituting these into (C19) and simplifying yields the following expression for the regional wage change, equation (13) in the main text, which we use to estimate the agglomeration elasticity, κ. w ˆr =

X ϕ(1 − ζ)η η ˆ R βri Pˆi − η[1 − ϕ(1 − ζ)] − κ + ϕζ η[1 − ϕ(1 − ζ)] − κ + ϕζ

(C20)

i

We also use an alternative employment-based approach to estimate κ. Start by noting that employment in a region × industry pair is given by Lri ≡ aLri Yri . Differentiating this definition and plugging in the specific-factor market clearing condition, (C9), and the Cobb-Douglas substitution conditions, (C11) and (C12), we have ˆ i = sˆi − w. L ˆ (C21) Substitute in sˆi from (C18) and simplify. ϕ(1 − ζ) ˆ ˆ i = 1 (Pˆi + Aˆi ) − 1 − ϕ(1 − ζ) w L ˆ− R ϕζ ϕζ ϕζ

(C22)

Plug in the labor supply and agglomeration equations, (12) and (11). ϕ(1 − ζ) ˆ ˆ i = 1 Pˆi − η[1 − ϕ(1 − ζ)] − κ w L ˆ− R ϕζ ηϕζ ϕζ

(C23)

Finally, plug in the equilibrium wage change in (C20), combine terms, and restore regional subscripts to yield equation (14) in the main text. η[1 − ϕ(1 − ζ)] − κ X ϕ(1 − ζ) ˆ ri = 1 Pˆi − 1 · ˆ L βri Pˆi − R ϕζ ϕζ η[1 − ϕ(1 − ζ)] − κ + ϕζ η[1 − ϕ(1 − ζ)] − κ + ϕζ i

102

(C24)