Premium
Using Unemployment Rates as Instruments to Estimate Returns to Schooling
Author(s) -
Arkes Jeremy
Publication year - 2010
Publication title -
southern economic journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.762
H-Index - 58
eISSN - 2325-8012
pISSN - 0038-4038
DOI - 10.4284/sej.2010.76.3.711
Subject(s) - unemployment , counterintuitive , quarter (canadian coin) , economics , ordinary least squares , econometrics , instrumental variable , unemployment rate , census , labour economics , demographic economics , demography , macroeconomics , geography , population , sociology , philosophy , archaeology , epistemology
I use state unemployment rates during a person's teenage years to estimate the returns to schooling. A higher unemployment rate reduces the opportunity costs of attending school. Using the same 1980 Census data set that Angrist and Krueger (1991) use, I also estimate returns to schooling with a modified version of their quarter‐of‐birth instrument. The estimates from the two‐stage least squares (2SLS) model using the unemployment rate and the model using the quarter‐of‐birth instruments are almost identical. In addition, these 2SLS estimates are larger than the ordinary least squares (OLS) estimates, supporting this counterintuitive, yet prevalent, result in the literature.