disparity

Journal of the Vietnam Veterans Institute 3(1): 22-36, 1994
Online at: http://www.v-r-a.org/docs/DisAVets.htm#_edn12

[RWT1] Discrimination Against Veterans by the Federal Agency
Charged with Protecting Veterans' Rights

. . . A 1987 article by Crane and Wise[12] establishes the broad framework for this assertion. These economic researchers made use of the National Longitudinal Study of the High School Class of 1972, along with ensuing surveys up to 1979, to determine the effect of military service on civilian earnings. Although the sample size in the national study on which the article was based is not overwhelming large (roughly 23,000 high school seniors from 1,300 schools), the conclusions are profound. As stated by Crane and Wise: “After controlling for other attributes of high school graduates, the weekly earnings on civilian jobs of those who served in the military were in 1979 approximately 12 percent less than the earnings of those who worked in the civilian sector.” Twelve percent less, seven years after graduating from high school.

One needs only to construct a simple, hypothetical case study to understand how this might occur; it’s obvious. Consider two high school graduates in 1972, one goes in the military, the other goes to work for a large manufacturing corporation. Then, after completing military service, the veteran goes to work for the same corporation. The postulated implications are illustrated in Table 1.

Table 1. Effect of Military Service on Civilian Earnings: A Hypothetical Case Study.⁽¹⁾

⁽¹⁾ Assuming a three-year tour of duty and an annual increase in wages of 4% each year.

Using the simple assumptions for this theoretical case study (three years on active duty and wage increases of 4% annually), what are the differences in annual earnings seven years after graduation, at age 25? Approximately 12 percent; just what was found by Crane and Wise.[13] What are the differences in annual earnings at age 50? Approximately 12 percent. And what are the differences in annual earnings at retirement, at age 65? Approximately 12 percent. It’s not a difficult scenario to follow, but the long-term cost of military service can become staggering, over $180,000 in this example. And if one uses $20,000 as the initial wage, the total difference in lifetime earnings jumps to over $300,000.

Furthermore, one need not rely on the Crane and Wise study to support the simplified civilian earnings model in Table 1. In one of the most methodologically sound analyses to date, J.D. Angrist reported a similar - even greater - effect of military service on civilian earnings.[14] In his 1990 paper, Angrist set out “to measure the long-term labor market consequences of military service during the Vietnam era.” He used the draft lotteries of 1970, 1971, and 1972 to assign risk of induction parameters to individuals born between 1950 and 1953, restricting his investigation “to men who turned 19 in the year they were at risk of induction.” By this means, he was able to identify “draft eligible” and “draft ineligible” subjects. Angrist then used Social Security records to monitor the earnings of these individuals from 1964 to 1984, with the longest period for assessing the effect of military service on earnings being fourteen years, from 1970 to 1984.

Angrist’s investigation established that prior to the draft lottery, there were no significant differences in the earnings of the two groups, a valuable control for his study. However, the situation changed dramatically for those who became eligible for the draft; their annual earnings dropping appreciably compared to the draft ineligible group. And by analyzing the Social Security records to 1984, Angrist was able to conclude that “long after their service in Vietnam was ended, the earnings of white veterans were approximately 15 percent less than the earnings of comparable nonveterans.” Fifteen percent in this case.[15] Taken in conjunction with the Crane and Wise study, it would appear that the rationale behind Table 1 is sound, only the magnitude of the career losses would be in doubt.

One could argue, of course, that seven years (the time frame for the Crane and Wise study) or even fourteen years (the maximum time frame in the Angrist study) is not sufficient to draw such sweeping conclusions; perhaps the individual veteran just hasn’t had enough time to catch up. But, how would the veteran go about catching up? By going to college on the GI Bill, back when it was still available? Possibly, but the college campus was not a friendly place for veterans during the Vietnam era. It still isn’t.[16] And as might be expected in that hostile environment, fewer Vietnam-era veterans completed their undergraduate studies than nonveterans,[17] and those who did would still be behind their nonveteran peers ... other college graduates.[18]

[12] A National Bureau of Economic Research Project Report from the John F. Kennedy School of Government, Harvard University, entitled: Military service and civilian earnings of youths, by J.R. Crane and D.A. Wise. Published in: Public Sector Payrolls (David A. Wise, Editor), University of Chicago Press, Chicago, pp. 119-145, 1987.

[14] A report from the Department of Economics, Harvard University, entitled: Lifetime earnings and the Vietnam era draft lottery: Evidence from Social Security Administration records, by J.D. Angrist. Published in: The American Economic Review, 80: 313-336, 1990. A number of earlier studies that attempted to assess the effect of military service on civilian earnings (some showing positive effects, some negative) are cited in this report; the shortcomings of the earlier works are addressed. Also see: Errata, The American Economic Review, 80: 1284-1286, 1990. A typographer’s error is reported in the earlier article whereby the titles and captions for Figures 1 and 3 were inverted.

[15] The differential in annual earnings in the early 1980s (for veterans and nonveterans in their early 30s) was reported to be approximately $3,500, well in excess of the difference projected in Table 1.

[16] Many reports on the problems veterans face on the college campus in the 1990s have originated from analyses in Ohio. See, for example: E. Holland’s A peacetime war, Ohio State Quest, 14(1): 5-7, 1992; A.H. Miller and J.A. Stever’s Anti-vet bias at OSU: Love beads vs. dog tags, Campus Report, 7(4): 3-8, 1992; J.A. Stever and R.W. Trewyn’s Veterans and the campus war, The Ohio AMVET, 7(1): 14-15, 1992; and A.H. Miller’s Political correctness in state universities: What state legislators need to know, The Heritage Lectures, published by the Heritage Foundation, Washington, DC, 1992.

[18] While a loss in seniority similar to that depicted in Table 1 would occur when comparing veteran and nonveteran college graduates, that in no way negates the fact that higher education should increase one’s civilian earning potential. As a result, it is possible that those veterans who were able to make use of their GI Bill benefits could exhibit more earning potential over their lifetime than nonveterans with a high school education. The actual benefits derived by veterans were quantified in another article by J.D. Angrist entitled: The effect of veterans benefits on education and earnings, Industrial and Labor Relations Review, 46: 637-652, 1993. The 1987 Survey of Veterans was employed by Angrist to determine that for each year of additional education, annual earnings of veterans increased by 4.3%. For those veterans who received the benefits, 1.4 years of additional education were accrued, on average. However, the 6% (1.4 years times 4.3%) increase in annual earnings would not overcome the 15% loss in earnings experienced by white Vietnam-era veterans (Angrist, 1990; footnote 14).

Age	Nonveteran	Veteran	Difference	Total
18	$12,000
19	12,480
20	12,979
21	13,498	$12,000	$1,498	$1,498
25	15,791	14,038	1,753	8,116
30	19,212	17,080	2,132	17,989
40	28,439	25,282	3,157	44,618
50	42,097	37,424	4,673	84,037
60	62,313	55,396	6,917	142,383
65	$75,814	$67,398	$8,416	$181,347