ECN 117 Economics of international migrations Midterm Exam 2 (Version A) | Complete Questions and Answers | Latest Spring 26/27.

University of California, Davis Santiago Pérez Economics Spring 2025 Name: _____________________________ ID:_________________________________ 1 ECN 117 Economics of international migrations Midterm Exam 2 (Version A) In general, the highest grades will be given to clear and precise answers. Use your time wisely, and good luck. You will be given 70 minutes to complete the exam. There are total of 47 points in the exam, so you should allocate about 1 minute per point. A. [2 points each, total of 26 points]. Multiple choice questions. Miami Control cities 1979 (Before Marielitos) 0.3 0.3 1981 (After Marielitos) 0.2 0.1 1. The table above shows crime rates (total number of crimes divided by total number of people) in Miami and a set of control cities, before and after the arrival of Marielitos in 1980. Assume you use these data and the difference-in-differences method to measure the impact of the arrival of Marielitos (in 1980) on crime rates in Miami. In that case, you would find that: a. Marielitos had no effect on crime rates in Miami b. Marielitos decreased crime rates in Miami by 0.2 c. Marielitos decreased crime rates in Miami by 0.1 d. Marielitos increased crime rates in Miami by 0.1 e. Marielitos increased crime rates in Miami by 0.2 Miami Control Cities Wages Arrival of Marielitos (1980)University of California, Davis Santiago Pérez Economics Spring 2025 Name: _____________________________ ID:_________________________________ 2 2. The figure above shows the evolution of the average wages of US-born workers in Miami and a set of “control cities”, before and after the arrival of Marielitos in 1980. Consider two potential research strategies to analyze the impacts of Marielitos: (1) a comparison of wages of US-born workers in Miami “before and after” the arrival of Marielitos, and (2) a “difference-indifferences” approach comparing changes in wages in Miami to changes in wages in the control cities. Which of the following is true? a. The measured impacts of Marielitos will be more positive when using the “before and after” approach than when using the “difference-in-differences” approach b. The measured impacts of Marielitos will be more positive when using the “difference-indifferences” approach than when using the “before and after” approach c. The measured impacts of Marielitos will be the same under both approaches d. The measured impacts of Marielitos will be positive when using the “difference-indifferences” approach but negative when using the “before and after” approach e. The measured impacts of Marielitos will be negative when using the “difference-indifferences” approach but positive when using the “before and after” approach 3. According to our discussion in lecture, which of these groups is the one whose labor market outcomes are more likely to be negatively affected by immigration? a. Natives in communication-intensive occupations b. Natives in manual-intensive occupations c. Previous immigrants in communication-intensive occupations d. Previous immigrants in manual-intensive occupations e. Refugees 4. Which of the following are problems when using the spatial correlations approach to measure the labor market impacts of immigration? a. Native born workers are more likely to move out of cities that receive more immigration b. Migrants might be more likely to move to cities experiencing an economic boom c. Migrants might be more likely to move to cities experience an economic decline since rents in these locations will be cheap d. a and b e. a, b and c 5. Assume that there are two types of workers, skilled and unskilled. Consider the potential impacts of an inflow of unskilled immigrants. Under which conditions would such impact be potentially more negative for the average native worker? a. Unskilled natives and unskilled immigrants are perfect substitutes with each other; natives are 80% skilled and 20% unskilled b. Unskilled natives and unskilled immigrants are imperfect substitutes with each other; natives are 80% skilled and 20% unskilled c. Unskilled natives and unskilled immigrants are perfect substitutes with each other; natives are 50% skilled and 50% unskilled d. Unskilled natives and unskilled immigrants are imperfect substitutes with each other; natives are 50% skilled and 50% unskilledUniversity of California, Davis Santiago Pérez Economics Spring 2025 Name: _____________________________ ID:_________________________________ 3 e. Unskilled natives and unskilled immigrants are imperfect substitutes with each other; natives are 100% skilled and 0% unskilled With immigrant coauthor Without immigrant coauthor Average number of patents 4.6 2.1 6. The table above shows the average number of patents of US-born researchers depending on whether they have an immigrant coauthor or they do not. Just based on the information available from this table, which of the following is true? a. US-born researchers who collaborate with immigrants are more productive than those who do not b. Working with an immigrant coauthor has a positive causal impact on the productivity of US-born researchers c. Based on this table, it is not possible to know whether working with an immigrant coauthor increases the productivity of US-born researchers d. a and b e. a and c 7. Assume you want to measure the effects of remittances on the families of Argentine migrants left behind. Half of Argentine migrants live in the US and the remaining half lives in Europe. Assume that, because of an economic boom in the Eurozone, the Euro apreciated relative to the Argentine peso but the US dollar neither appreciated nor depreciated. Following the approach in Yang (2004), which groups of families would you compare to measure the effects of remittances on families left behind? a. Argentine families with a family member in Europe, to Argentine families with no family members abroad b. Argentine families with a family member in the US, to Argentine families with no family members abroad c. Argentine Families with family members in the US, to Argentine families with family members in Europe d. Argentine Families with family members in Europe, to Brazilian families with family members in Europe e. Argentine Families with family members in Europe, before and after the emigration of a family member 8. Abarcar and Theoharides (2024) studies the impact in the Philippines of a US policy expanding the number of visa slots for nurses willing to migrate the United States. The paper findings can be summarized as follows: a. The policy led to an increase in the migration of nurses out of the Philippines b. The policy reduced the overall number of nurses in the Philippines c. The policy increased the overall number of nurses in the Philippines d. a and b e. a and cUniversity of California, Davis Santiago Pérez Economics Spring 2025 Name: _____________________________ ID:_________________________________ 4 9. The figure above shows evidence from Italy on the relationship between the time in which prospective employers submitted an application for immigrants’ employment permit (x-axis, negative numbers imply that the application arrived in time and positive numbers imply that it arrived too late) and the probability of being arrested for a crime (y-axis). The evidence from this figure indicates the following: a. Access to legal job opportunities increases immigrants’ crime rates b. Access to legal job opportunities reduces immigrants’ crime rates c. Access to legal job opportunities has no effect on immigrants’ crime rates d. It is not possible to know whether access to legal job opportunities reduces or increases immigrants’ crime rates: documented and undocumented immigrants differ along many dimensions beyond legal status e. None of the above 10. Abramitzky, Boustan, Jácome, Pérez and Torres (2024) study differences in incarceration rates between immigrants and the US born over the last 150 years. The paper findings can be summarized as follows: a. Immigrants’ incarceration rates were similar to those of the US born in the past but are higher today b. Immigrants’ incarceration rates were lower to those of the US born in the past but are higher today c. Immigrants’ incarceration rates were higher to those of the US born in the past and are still higher today d. Immigrants’ incarceration rates were similar to those of the US born in the past but are lower today e. Immigrants’ incarceration rates were higher to those of the US born in the past but are similar today 11. Assume that non-citizen immigrants who committed crimes in the US were immediately deported without serving their prison sentence in the US. A group of researchers was unaware of this issue and decided to compare incarceration rates of immigrants to incarceration rates of the US-born in order to understand which of the two groups is more likely to commit a crime. Which of the following is true? where g(K) is the distribution of an individual characteristic K (see, e.g., Angrist et al. 2016). The right-hand side of equation (3) is easily estimated by the 2SLS regression of g(K)L on L using Z as an instrument and controlling parametrically or nonparametrically for differences in X. The fraction of previous offenders—arguably the best proxy for the average propensity to commit crimes—is nearly identical between compliers and the total sample (1.2 and 1.1 percent, respectively). The same is true for the age distribution, whereas compliers are more likely to come from poorer countries relative to the rest Figur% 3. Numb%r of Crim%* p%r App,i-./0 B%for% ./1 Af0%r C,i-2 D.3*, Co/1i0io/., o/ 04% Timi/g of App,i-.0io/ Notes: The graphs show the average number of crimes per applicant committed before and after click days, conditional on the timing of application. The scatterplots are averages within 5ve-minute bins, and the solid line and shaded areas are the predicted outcomes and associated con5dence intervals, respectively, based on quadratic polynomial regressions. The graphs in the right and left columns show the crime rates in the year before and after click days (2007 and 2008, respectively). The top, middle, and bottom rows show the results for all applicants, the subsample of type-A applicants, and the subsample of type-B applicants, respectively. 0 0.005 0.01 0.015 0.02 0.025 −30 −20 − 0 0.005 0.01 0.015 0.02 0.025 −30 −20 − 0 0.005 0.01 0.015 0.02 0.025 −30 −20 − 0 0.005 0.01 0.015 0.02 0.025 −30 −20 − 0 0.005 0.01 0.015 0.02 0.025 −30 −20 − 0 0.005 0.01 0.015 0.02 0.025 −30 −20 − 2007: all applicants 2007: type-A applicants Timing of the application, T (cutoff: T = 0) 2007: type-B applicants 2008: all applicants 2008: type-A applicants 2008: type-B applicantsUniversity of California, Davis Santiago Pérez Economics Spring 2025 Name: _____________________________ ID:_________________________________ 5 a. Using incarceration rates will tend to overestimate immigrants’ true crime rate. The overestimation will be bigger in magnitude the larger the proportion of immigrants who are citizens. b. Using incarceration rates will tend to overestimate immigrants’ true crime rate. The overestimation will be bigger in magnitude the larger the proportion of immigrants who are non-citizens. c. Using incarceration rates will tend to underestimate immigrants’ true crime rate. The underestimation will be bigger in magnitude the larger the proportion of immigrants who are citizens. d. Using incarceration rates will tend to underestimate immigrants’ true crime rate. The underestimation will be bigger in magnitude the larger the proportion of immigrants who are non-citizens. e. Using incarceration rates will correctly estimate immigrants’ true crime rate 12. In class, we discussed the article “Immigration and Entrepreneurship in the United States” by Azoulay, Jones, Kim and Miranda (2022). The findings of this paper can be summarized as follows: a. Immigrants are more likely to start firms than natives, but this is explained by their higher propensity to start small (with less than 5 employees) firms b. Immigrants are more likely to start firms than natives, but this is explained by their higher propensity to start large (with more than 100 employees) firms c. Immigrants are more likely to start firms than natives, and this is explained by their higher propensity to start both small and large firms d. Immigrants are less likely to start firms than natives e. Immigrants are similarly likely than natives to start firms 13. In class we discussed the article “The Contribution of High-Skilled Immigrants to Innovation in the United States”. The findings of this paper can be summarized as follows: a. Immigrants hold more per capita patents than natives, but their average quality is lower b. Immigrants hold more per capita patents than natives, and their average quality is higher c. Immigrants are more likely be cited by researchers who are based outside of the US d. Immigrants are more likely to coauthor with other immigrants as they spend time in the US e. b and c B. [5 points]. Short answer question. Consider the model of the labor market impacts of immigration with two types of labor, skilled and unskilled. Assume that in the short run the skills of natives are fixed (that is, there is a fixed number of skilled and unskilled natives). However, in the long run natives can accumulate skills in response to increases in the “skill premium” (the ratio between the wages of skilled to unskilled workers). What would be the short-run effect of an inflow of unskilled immigrants on the skill premium? What would be the long-run effect on the skill premium? Explain (it might be helpful to use a graph). You can assume that immigrants and natives of the same skill level are perfect substitutes with each other.University of California, Davis Santiago Pérez Economics Spring 2025 Name: _____________________________ ID:_________________________________ 6 C. [16 points]. Exercise Labor demand in city A is initially given by: Wa=12-La Labor demand in city B is given by: Wb=12-Lb where La and Lb indicate the number of workers in cities A and B, respectively. Initially, there are 5 native workers in city A and 5 native workers in city B. a. [1 point]. What would be the wage in cities A and B in this situation?University of California, Davis Santiago Pérez Economics Spring 2025 Name: _____________________________ ID:_________________________________ 7 Assume that, due to an oil discovery, labor demand in city A becomes: Wa=16-La whereas labor demand in city B remains the same. This increase in demand causes an inflow of 2 immigrant workers (from outside the country) to city A. b. [4 points] You are a researcher interested in measuring the impacts of immigration on the wages of natives. To do so, you decide to implement the “spatial correlations” approach comparing the wages of natives in city A (which has 2 immigrants) to the wages of natives in city B (which has zero immigrants). What would be the measured impact of the 2 immigrants on the wages of natives according to the spatial correlations approach? For this part of the question, assume that natives in city B are not allowed to move to city A. c. [4 points]. What is the true impact of the 2 immigrants on the wages of natives in city A? For this part of the question, you can continue assuming that natives in city B are not allowed to move to city A.University of California, Davis Santiago Pérez Economics Spring 2025 Name: _____________________________ ID:_________________________________ 8 Next, assume that native workers can instead freely change their city of residence. d. [4 points]. What would be the measured impact of immigration on the wages of natives when using the spatial correlations approach? Finally, assume that natives can change their city of residence but need to pay a cost in order to do so. e. [3 points]. Explain the following statement: “the higher the cost of moving between cities, the higher the measured impact of immigration when using the spatial correlations approach”University of California, Davis Santiago Pérez Economics Spring 2025 Name: _____________________________ ID:_________________________________ 9