*2.4. Theoretical Framework*

The analysis of the effects of the changes in the trade openness of the Brazilian economy on the formal and informal labor market outcomes is guided by the theoretical model and empirical framework developed by Paz (2014) 1. This is a heterogeneous firm model based on Melitz (2003), in which there is monopolistic competition and firms are heterogeneous in two dimensions: productivity and wages. Firms decide whether to enter the market or not considering their intrinsic characteristics and the current level of market competitiveness. If they decide to do so, what type of labor do firms offer to workers: formal or informal? By hiring formal workers, firms are subject to a fixed cost per worker to comply with labor regulations and a variable cost related to social security contributions. An informal labor contract has no such costs, but firms incur the risk of being audited (with positive probability) and fined if they are caught employing informal workers. In equilibrium, larger (more productive and higher wage) firms will hire only formal workers. In contrast, smaller (less productive and lower wage) firms will hire informal workers.

An increase in import penetration fosters competitiveness in domestic markets. This makes the smallest firms—who typically employ informal workers and pay lower wages experience negative profits and exit the market. This leads to a reduction in the employment

of informal workers and an increase in the average informal average wage. Yet the firms that were previously indifferent about hiring either formal or informal workers switch from formal to informal employment in response to this increased competitiveness. This raises the level of informal employment. Such a change may increase or decrease average formal and informal wages depending on the specific firm's joint distribution of productivity and wages. In sum, an increase in import penetration leads to ex ante ambiguous effects of import penetration on informality and on average formal and informal wages.

An emerging literature has uncovered evidence of more nuanced effects of import competition depending on the source country of the imports and on the impacted industry characteristics. Facchini et al. (2010) found that the elasticity of substitution between Chinese and Brazilian manufacturing products was higher than that between Brazilian and high-income country products. This finding implies that imports from China are closer substitutes for Brazilian made goods. As a result, Chinese imports exert a stronger competitive pressure on Brazilian producers than imports from the rest of the world (especially those from high-income countries). This leads to the first testable prediction:
