The problem set is due at the beginning of class on November 16, 2022. You should submit your assignment on Quercus. Please submit your written answers in a Word document or PDF; your
code should be submitted separately as a .do or .txt file.
If you have technical difficulties, please email Steven Ryan at steven.ryan@mail.utoronto.ca.Please read Jacob Bastian (2019) “The Rise of Working Mothers and the 1975 Earned Income
Tax Credit” American Economic Journal: Economic Policy, 12(3): 44–751.
a) Describe the 1975 introduction of the Earned Income Tax Credit (EITC). Why was the EITC introduced? How does it differ from a guaranteed annual income? Who is eligible to receive it? How does the 1975 version of the EITC differ from the EITC as it existed in 2013?
b) What does economic theory predict about the labour supply response to the
introduction of the EITC, both on the extensive margin (the decision about whether or not to work) and on the intensive margin (the decision about how many hours to work conditional on working)? Do the incentives to supply labour vary as a function of earnings at different parts of the EITC schedule at the time of its introduction in 1975?
To be concrete, consider three individuals whose household earnings prior to the EITC’s introduction were, respectively, $0, $10,000 in constant 2013 USD, and $30,000 in constant 2013 USD. How would the introduction of the EITC affect labor supply incentives for each of these individuals? Would these incentives offered through the EITC additionally vary depending on whether or not a mother is married?2 You can refer to Figure 4.B in Bastian (2020), specifically the blue 1975 EITC schedule, to help you
answer this question.
c) Describe the Difference–in–Difference (DiD) methodology that Bastian uses to estimate the causal effect of the introduction of the EITC in 1975 on mothers’ labour supply? How does he define the treatment group and the control group? Describe the assumption that must hold for the DiD to produce an unbiased estimate of the causal effect of the EITC on labour supply? How can one test this assumption?
d) Consider the DiD estimator to identify the causal effect of the introduction of the EITC on mothers’ labour supply. First, show how one can implement the DiD using raw means when there are no covariates. Second, show how you can equivalently use a regression model to estimate the DiD. Be sure to discuss which coefficient (or combination of coefficients) in the regression model corresponds to the DiD estimate of causal impact. Note that we are considering the “reduced–form” effect of the EITC’s introduction—the percentage point change in mothers’ employment due to the EITC—rather than estimating elasticities. What are the advantages of using a regression model to estimate the DiD?
