In June 2016, Donald Trump justified the protectionist policies during his first administration as “the quickest way to bring our jobs back to our country”. This column shows that while the trade war with China failed to create jobs at home, it did boost employment in Brazil. Regions in the country specialised in industries targeted by China’s trade policy towards the US experiencing a relative increase in formal employment and wage bills between 2016 and 2021. With tensions between major economies rising, policymakers in third countries should be aware not only of the risks of global fragmentation, but also of the opportunities it can create.
In June 2016, Donald Trump justified his protectionist views as “the quickest way to bring our jobs back to our country” (Trump 2016). The first US-China trade conflict involved the imposition of tariff increases on hundreds of billions of dollars’ worth of bilateral trade. Between 2018 and 2019, the average US tariff on Chinese products jumped from 2.9% to 24.9%, while China’s average tariff on US goods rose from 9.8% to 28.2% (Bown 2021). However, despite Trump’s motivation, research shows that the conflict failed to generate employment gains in either the US or China (Waugh 2019, Benguria and Saffie 2020, Chor and Li 2024).
In a recent paper (Cavalcanti et al. 2025), we show that the US-China trade war did create jobs, but not in the US or China – instead, it created jobs in Brazil. Specifically, Brazilian regions specialised in industries targeted by China’s retaliatory tariffs on US goods experienced a relative increase in formal employment and wage bills between 2017 and 2021. These findings contribute to a recent but growing literature on the global spillovers of major-power trade tensions, including evidence of trade deflection (Evenett and Martín 2025) and strategic realignment (Crowley 2019).
Third-country spillovers from the US-China trade war
The 2018-2019 US-China trade war represented a major escalation in commercial tensions between the world’s two largest economies. The US imposed steep tariffs on a wide range of Chinese goods, and China responded with its own tariffs on US exports. Although the measures were aimed at each other, their effects spilled over internationally. Countries not directly involved, especially those with similar export profiles, faced a reshaped global trading environment, with new risks, but also opportunities.
Brazil was one such country. The country has strong trade ties to both China and the US. In 2017, China accounted for 22% of Brazil’s exports and the US for 12%. However, the composition of Brazil’s exports was asymmetrically correlated with those of the two trade war participants: it aligned far more closely with the US than with China. In particular, Brazil’s exports were highly correlated with US exports to China (53.5%), but showed almost no correlation with Chinese exports to the US (1.3%).
This asymmetry in export patterns created the conditions for trade diversion (Fajgelbaum et al. 2024). As China imposed high tariffs on American products, it began turning to alternative suppliers, including Brazil.
Using detailed product-level export data from Brazil to both China and the US between 2017 and 2021, we exploit variation in the timing of tariff impositions to implement a staggered difference-in-differences approach, following the methodology developed by Callaway and Sant’Anna (2021). This allows us to examine the product-level impact of the discriminatory tariffs imposed by China and the US. The results, displayed in Figure 1, are striking: Brazil’s exports of goods targeted by China’s tariff imposition increased within a few months of the tariff implementation, whereas exports to the US showed no significant response.
Figure 1 Impact of trade war tariffs on Brazilian exports
Notes: Estimates based on monthly data on the value and weight of Brazilian exports to China and the US, disaggregated by product at the 6-digit HS level, from January 2017 to December 2021. In Panel A, the dependent variable is the logarithm of one plus the value exported (in dollars of 2016, deflated using the US CPI) to China (Panel A.1) or the US (Panel A.2). In Panel B, the dependent variable is the logarithm of one plus the quantity exported (in kilograms) to China (Panel B.1) or the US (Panel B.2). Bars represent 95% confidence intervals. Standard errors are clustered at the HS 4-digit level.
But are these trade effects merely the result of exporters adjusting to different destinations, or could they generate broader economic consequences, including real impacts on the labour market?
Measuring regional exposure
To assess whether the diversion of exports translated into meaningful changes in employment and wages within Brazil, we combine Brazilian employer-employee matched microdata with detailed tariff schedules and trade data. This allows us to construct a measure of regional exposure to the trade war, based on each region’s employment structure and the global significance of the affected trade flows.
Figure 2 displays how regional exposure to trade war tariff increases varies spatially in Brazil. We standardise the Chinese and American trade war tariff change variables to capture regional variation relative to the national mean. Darker regions represent those most exposed to tariff increases imposed by China (Panel A) and the US (Panel B).
Figure 2 Regional trade war tariff changes (standardised)
The regions most affected by Chinese tariffs on American goods (Panel A) are in the North, Centre West and South, where agricultural and commodity-based industries are concentrated. In contrast, the regions most exposed to US tariffs (Panel B) are primarily located in the Southeast, Brazil’s most industrialised area. Exposure to Chinese and American tariff increases is distributed differently across Brazilian regions and is negatively correlated (correlation = –0.35). This feature enables us to disentangle the effects of these two shocks on local labour market outcomes in Brazil.
Labour market effects
We then use data from 2012 to 2021 to evaluate the effect of the trade war tariffs on the evolution of labour market outcomes across Brazil’s microregions, focusing on the changes in formal employment and the wage bill relative to 2016. Our results are clear. Brazilian regions specialised in industries targeted by China’s trade policy toward the US experienced a relative increase in formal employment and wage bills between 2016 and 2021. In contrast, regions more exposed to US tariffs on China did not exhibit differential labour market outcomes relative to other Brazilian regions (Figure 3).
Quantitatively, a one standard deviation increase in exposure corresponds to a 2.1% rise in employment and a 2.3% increase in the wage bill between 2016 and 2021. Since Brazil’s formal employment growth was slightly negative over the decade, these gains are substantial.
Figure 3 US-China trade war and Brazilian local labour markets
Panel A: Formal employment
Panel B: Formal wage bill
Implications
This evidence highlights a broader message: while trade wars often fail to achieve their stated domestic goals, they can reshape global trade in powerful and sometimes unintended ways. Trade wars risk creating a fragmented world economy, by clashing different economic systems (Crowley 2019).
Our study complements this view by showing that these geopolitical clashes also have domestic implications for third countries. In Brazil’s case, the conflict created regional winners.
These lessons are particularly relevant today. With tensions between major economies rising, policymakers in third countries should be aware not only of the risks of global fragmentation, but also of the opportunities it can create.
Source : VOXeu
