A researcher wishes to investigate the impact of immigration on the Canadian labour market. She uses time-series data on the number of immigrants the national unemployment rate from 1961 to 2000 to estimate the following relationship:
dur(t) = a + yd ln imm(t) = e(t)
where dur(t) is the change (from one year to the next) in the unemployment rate, and d ln imm(t) is the change in the logarithm of the number of immigrants arriving in Canada.
a. Explain what economic theory suggests should be the sign of y.
b. She obtains the following regression estimates (with standard errors in parentheses):
Intercept 0.015 (0.147)
d ln imm(t) -1.56 (0.75)
She interprets her results as showing that, far from adversely affecting the Canadian labour (and increasing the unemployment rate), increases in immigration actually reduce unemployment. Critically evaluate her interpretation. In your answer, you should discuss:
i. Under what conditions her interpretation may be correct
ii. The reasons her interpretation may be flawed
iii. Any evidence that you know sheds light on these issues