Inspired by their 2019 book Good Economics for Hard Times, I decided to cover here one of the academic papers that Abhijit Banerjee and Esther Duflo referred to when delving in the subject of immigration.
While the question of how immigration impacts labour market outcomes is very popular among policymakers, answering it is not a straightforward endeavour: by just looking at the correlation between the proportion of immigrants and the wage rate on a given location, we are certainly not viewing the whole story. There are a number of considerations that makes it challenging to study the causal relationship between immigration and labour market outcomes:
Workers typically choose to migrate to locations where they anticipate better economic opportunities. This behaviour would indicate that immigration is correlated to higher wages.
Immigration policy tends to become more strict when a country is experiencing an economic downturn. This pattern might lead to the interpretation that less immigration is correlated with a poor economic environment.
Firms might decide to move to cities where they expect an influx of immigrants in the near future. In this case market expectations play a dynamic role that is not factored in when only looking at the correlation between wages and immigration.
In order to address the concerns above, we would need to observe a sharp change in the supply of immigrants that is not directly related to the wages or unemployment levels in that particular location. Also, workers and firms from other locations should not be able to anticipate and respond to this change.
The Case Study
On the 20th of April in 1980, Fidel Castro proclaimed that Cubans who wanted to leave the country were able to do so from the port of Mariel. This political announcement led to the arrival of 125,000 Cubans, also referred to as the ‘Mariels’, between April and October on the same year of the announcement. The historical event is known as the Mariel Boatlift and it led to a 7% increase on the Miami labour force, predominantly in ‘low-skilled’ professions. If we assume that the ‘Mariels’ chose to migrate to Miami because of geographic proximity rather than the city’s specific employment circumstances relative to other places in the United States, the Mariel Boatlift offers a suitable natural experiment to evaluate the causal relationship between immigration and wage rates.
In his paper, David Card studies the effects of immigration on wages and unemployment rates of Miami’s labour market. Card compares the trend of wages and unemployment rate of non-Cuban residents in Miami before and after the arrival of the ‘Mariels’ to the trend in Atlanta, Los Angeles, Houston and Tampa. Since these four cities followed a similar economic growth pattern to Miami before the Mariel Boatlift, they provide an appropriate counterfactual state of what Miami’s wages and unemployment rate would have been if the Mariel Boatlift had not happened.
Results
The author found out that that the influx of Mariel immigrants had no effect on the wage rate or the unemployment levels of ‘low-skilled’ non-Cuban workers. The data also reveals that the Mariel boatlift had essentially no impact on the wage rate of Cuban workers that resided in Miami before the boatlift. These results suggest that there was a rapid absorption of the Mariel immigrants into Miami's labour force.
The paper also explores two potential explanations as to why an increase in labour supply didn’t lead to adverse effects on labour market outcomes as we would suspect if the market would operate under the traditional demand and supply economic framework. One of them is that ‘Mariels’ displaced other immigrants who would have moved to Miami in the early 1980s had the Boatlift not happened. The data seems to support this hypothesis to some extent as net migration of natives and non-Mariel immigrants slowed down after the Boatlift. Another explanation comes from the industry distribution in Miami and how it was suitable to welcome in a large number of ‘low-skilled’ immigrants. Evidence suggests that industries such as Apparel and Textile, which demand relatively ‘low-skilled’ labour, were predominant in Miami in the late 1970s.
Finally, the effect of immigration on labour market opportunities has been widely studied and the results of Card’s paper challenges the traditional economic view. Evidently, the claim that immigration always leaves unaffected labour market outcomes cannot be made by just looking at one single case study, and further research is required to understand the channels through which immigration affects the economy and if there are any distributional effects that exist within a wider context. However, experiences as the one portrayed by the Mariel boatlift encourage us to keep our minds open and not always assume that the theory we were taught will always hold in real labour markets.
Comments