*Robustness Checks*

The robustness check conducted here employs state-by-year fixed effects in lieu of state and year effects. This new specification is used to account for state specific policies such as changes in state-level educational systems, state-level minimum wages, and labor regulations enforcement (Almeida et al. 2022). Moreover, since commodity (iron ore or soybeans, for instance) production in Brazil is geographically concentrated in a few states, these state-by-year effects can also pick-up the effects of the increased Chinese demand for these primary commodities. Table 7 reports the Probit and IV Probit estimates of the selection equation using this new set of fixed effects. We can see that the results are very similar to those in Table 3. Table 8 reports the IV estimates for the average wage of informal workers in columns (1)–(3) and for the average wage of formal workers in columns (4)–(6). The estimates in columns (1)–(3) are very similar to those in columns (4)–(6) in Table 6. Moreover, these new estimates are more statistically significant in many cases. The specifications for the average wage of formal workers present estimated coefficients that are very similar in magnitude to those in columns (4)–(6) in Table 5. Nevertheless, their statistical significance has slightly declined. In sum, these estimates obtained using a different specification of fixed effects corroborates the results obtained with the main specifications.

**Table 7.** Worker-level IVProbit estimates of the effects of industry-level import penetration on the informal status indicator using Equation (1) and state, year and industry fixed effects.



**Table 7.** *Cont.*

Notes: Number of observations is 671,134. \*\*\*, \*\*, and \* indicate statistical significance at the 1%, 5%, and 10% levels, respectively. Standard errors clustered at the industry level. Sample weights from PNAD/Census used. The excluded instruments used in all IV estimates are the Latin American countries' Chinese share of imports, the Latin American countries' high-income countries share of imports, and their interactions.

**Table 8.** Worker-level IV estimates of the effects of industry-level import penetration on the wages of informal and formal workers using Equations (2) and (3), respectively.



**Table 8.** *Cont.*

Notes: \*\*\*, \*\*, and \* indicate statistical significance at the 1%, 5%, and 10% levels, respectively. Standard errors are bootstrapped with 500 repetitions. Sample weights from PNAD/Census used.

#### **5. Conclusions**

China is one of the most populous countries in the world, and it entered the 21st century not only as one of the largest and fast-growing economies but also as a major player in world trade. This rather swift ascension together with its cost advantage in manufacturing production prompted several concerns in developing countries as to whether they would still be able to sustain a dynamic manufacturing sector in view of this Chinese competitive edge. Such concern is built on the fact that many observers perceive a strong manufacturing sector as a key driver of economic growth and as a provider of higher wage jobs relative to those available in agriculture and services.

A good case study to assess such concerns is the increase in import competition experienced by the Brazilian manufacturing sector in 2000–2012. In this period, the import penetration increased by more than 20 percent and the Chinese share of such imports increased from 3 to 20 percent. Brazil is also the largest economy of Latin America and has a large and diverse manufacturing sector with ubiquitous informal employment.

This study employed Brazilian household data to examine the impacts of the increasing Chinese and rest of the world import penetrations on the likelihood of informal employment and on the average wage of formal and informal workers in the manufacturing sector for 2000–2012. The empirical methodology employs a switching regressions model as in Paz (2014), which accounts for worker self-selection into formal and informal jobs and for the potential endogeneity of trade policy.

The empirical results indicate that greater industry-level Chinese and ROW import penetrations increase the informal job likelihood at different intensities. Furthermore, these effects are heterogeneous and modulated by the unskilled labor intensity of the industry and by the degree of industrialization of the Brazilian states. Indeed, the ROW import penetration has a negative effect, and the Chinese import penetration has no effect on the informality likelihood in unskilled-labor intensive industries. In contrast, both import penetrations have positive impact on informality likelihood in the remaining industries. An increase in the Chinese import penetration reduces informality while ROW import penetration increases it in manufacturing states. Nevertheless, both forms of import penetration positively affect informality in non-manufacturing states.

The effects on the average formal and informal wages are more nuanced. An increase in Chinese import penetration raises the average formal wage, except in manufacturing states. Greater ROW import penetration decreases the average formal wage, except for in manufacturing states or in non-unskilled-labor intensive industries. For the average informal wage, both the Chinese and the ROW import penetration have positive effects of different magnitudes, and negative effects for informal workers located in non-manufacturing states.

The evidence amassed in this study suggests that the effects of international trade on labor market outcomes are moderated by the country of origin of imports and, at the same time, by the unskilled labor intensity of industries and by regional characteristics. The important nuances uncovered by this study should not be overlooked in the design of public policies to address potential harmful effects of increased import competition, especially because the most vulnerable workers seem to experience a negative impact from this trade.

Moreover, this study's estimates are at variance with those of the extant literature for different countries and periods of time. This strongly suggests that such trade effects are highly heterogeneous. Unfortunately, data scarcity that plagues the entire literature such as the lack of employee–employer matched data covering informal jobs—is a major limitation of this study because it precludes an investigation of the role of either the unobservable characteristics of workers or firm characteristics. Given the available data, a promising avenue for future research is the use of cross-country data to investigate whether country-specific institutions are behind these disparate results.

**Funding:** This research received no external funding.

**Data Availability Statement:** The data used in this article can be found at www.ibge.gov.br and ipeadata.gov.br.

**Conflicts of Interest:** The author declares no conflict of interest.
