Absence of conflict of interest.
Citation
Angrist, J. D., & Chen, S. H. (2011). Schooling and the Vietnam-era GI bill: Evidence from the draft lottery. American Economic Journal: Applied Economics, 3(2), 96-118.
Highlights
- The study’s objective was to examine the impact of the G.I. Bill on education, employment, earnings.
- The authors used a nonexperimental comparison group analysis to estimate the impact of the Vietnam-era G.I. Bill on outcomes. The approach compared the outcomes of Vietnam-era veterans and non-veterans in 1999 using the 1-in-6 long form sample of the 2000 U.S. Census.
- The authors found that, compared with White male non-veterans, Vietnam-era White male veterans eligible for the draft and the G.I. Bill had significantly more years of schooling and were significantly more likely to have earned secondary and postsecondary degrees. The study found no statistically significant relationships between veteran status and employment or earnings.
- The quality of causal evidence presented in this report is moderate because it was based on a well-implemented nonexperimental design. This means we are somewhat confident that the estimated effects are attributable to the G.I. Bill and not to other factors.
Intervention Examined
The WWII G.I. Bill
Features of the Intervention
Congress passed the WWII G.I. Bill in 1944. All veterans who served in the military from September 1940 through July 1947 were eligible for the benefits. Benefits covered at least one year of education for all eligible people and one additional month of coverage for each month of active duty, up to 48 months. The benefits covered three or four full years of postsecondary education for most participants. Participants’ received full payment for tuition, books, and supplies at any U.S. higher education institution, along with a stipend to cover living expenses that varied by size of the veteran’s family. The Vietnam-era G.I. Bill stipend for full-time students had about the same value as the WWII G.I. Bill, adjusted for inflation. The benefits were nearly twice the typical cost of tuition, room, and board at 4-year public universities at the time.
Features of the Study
The authors used a nonexperimental comparison group analysis to estimate the impact of the Vietnam-era G.I. Bill. They examined a sample of men who were born in the United States between 1950 to 1952 using data from the 1-in-6 long form sample of the 2000 U.S. Census. This sample included about 700,000 White men and about 96,000 non-White men. About a quarter of this sample served in the Vietnam war, and roughly 38 percent were eligible for the draft. About 12 percent of the sample was non-White. Average schooling was 13.8 years for White men and 12.6 years for non-White men. At the time of follow-up, the average age of men in the sample was about 48.
The authors used an instrumental variables statistical approach to estimate the impact of veteran status, and hence eligibility for the G.I. Bill, on men’s education, employment, and earnings outcomes. The approach compares the outcomes of White male Vietnam-era veterans and non-veterans in 1999, taking into account the potential influence of their draft lottery numbers and date of birth on the likelihood of serving in the military. The analysis took into account men’s year, month, and state of birth.
Findings
Employment
- The study found no statistically significant relationships between veteran status and employment.
Earnings
- The study found no statistically significant relationships between veteran status and earnings.
Education
- Compared with White male non-veterans, Vietnam-era White male veterans eligible for the draft had significantly more years of schooling (0.34 relative to the sample mean of 13.8) and more years of college (0.26 relative to the sample mean of 1.7).
- Compared with White male non-veterans, Vietnam-era White male veterans eligible for the draft were significantly more likely to earn secondary and postsecondary degrees. Specifically, they were 2 percentage points more likely to be a high school graduate or higher (relative to the sample mean of 91 percent), were 8 percentage points more likely to have an associate’s degree (relative to the sample mean of 41 percent), were 5 percentage points more likely to have a bachelor’s degree (relative to the sample mean of 33 percent), and were 2 percentage points more likely to have a master’s degree (relative to the sample mean of 14 percent).
Considerations for Interpreting the Findings
The study examined the impact of the G.I. Bill indirectly, attributing the impact of veteran status to the impact of the G.I. Bill. The authors used the random draft lottery numbers that served to draft young men into the military from 1970 to 1972 to address a challenge in estimating impacts of military service: the typically voluntary nature of enlistment.
The study authors estimated multiple related impacts on outcomes related to education. Performing multiple statistical tests on related outcomes makes it more likely that some impacts will appear statistically significant purely by chance and not because they reflect the program’s effectiveness. The authors did not perform statistical adjustments to account for the multiple tests, so the number of statistically significant findings in these domains could be overstated.
Causal Evidence Rating
The quality of causal evidence presented in this report is moderate because it was based on a well-implemented nonexperimental design. This means we are somewhat confident that the estimated effects are attributable to the GI Bill and not to other factors.