How Do Mental Disability Benefits Affect Veterans’ Well-Being?

The study asks whether and how Veterans Affairs Disability Compensation for mental health conditions affects veterans’ economic stability, healthcare engagement, and overall well-being. Using administrative data on 867,000 veterans and leveraging quasi-random assignment to examiners with differing rating tendencies, the authors estimate the causal effects of additional benefits. An extra $1,000 per year in disability income over five years reduced food insecurity by 4.1 percent and homelessness by 1.3 percent, while lowering debt collections by 6.4 percent. Healthcare use increased by 2.5 percent, with improved preventive care and medication adherence, though there were precise null effects on most health outcomes and mortality.

Full Citation and Link to Article

David Silver and Jonathan Zhang. Invisible Wounds: How Mental Disability Benefits Shape Veteran Well-Being. American Economic Journal: Economic Policy (Forthcoming). Available at: https://www.aeaweb.org/articles?id=10.1257/pol.20230811 

Extended Summary

Central Research Question

This study investigates whether and how disability compensation for mental health conditions provided by the U.S. Department of Veterans Affairs (VA) influences the economic stability, healthcare engagement, and overall well-being of veterans. While policymakers and clinicians recognize the importance of supporting veterans with service-connected mental health conditions such as post-traumatic stress disorder (PTSD) and depression, there has been limited causal evidence on the broader impacts of these benefits. The central question is whether additional, durable income provided through VA Disability Compensation (VA DC) leads to measurable improvements in veterans’ economic outcomes, health behaviors, and mortality. By focusing on mental health conditions—a rapidly growing share of VA claims—the study seeks to illuminate how income transfers affect vulnerable populations with “invisible wounds of war.”

Previous Literature

Prior research on VA Disability Compensation has largely concentrated on labor market effects or program costs, particularly for Vietnam-era veterans. For example, Angrist et al. (2010) and Autor et al. (2016) examined impacts on labor supply and fiscal sustainability but offered little on health and well-being. More recent work has begun to address health outcomes. Trivedi et al. (2022) studied hospitalization and mortality following expanded eligibility for Vietnam veterans with diabetes, while Huynh and Chan (2024) built on examiner-based approaches to estimate treatment effects on mortality. Outside the VA setting, research on income transfer programs like Supplemental Security Income (SSI) and Social Security Disability Insurance (SSDI) has found mixed results, with some studies showing improvements in financial stability but relatively imprecise or null effects on health and mortality (Gelber et al. 2022; Black et al. 2024). The broader literature on income and health suggests positive correlations, but causal estimates are difficult due to confounding. This study adds to a small but growing body of work by directly examining how mental health disability benefits shape both economic security and health-related outcomes, providing unusually precise estimates through a large sample and quasi-random design.

Data

The authors use linked administrative microdata from the Department of Veterans Affairs, covering approximately 867,000 veterans who filed a first-time disability claim for a mental health condition between 2004 and 2019. These data include claims histories, examiner assignments, disability ratings, and resulting compensation amounts. The dataset is further linked to electronic health records, VA-administered surveys, debt management records, and the National Death Index.

Economic and financial well-being is measured through annual food insecurity screens, indicators of homelessness (diagnosis codes, VA homeless bed usage, rental assistance participation), and debt collections referred to the Treasury. Health measures include diagnoses of major depressive disorder, substance use, body mass index, blood pressure, glucose levels, pain scores, suicide attempts, overdoses, and mortality. Healthcare engagement is assessed through utilization data, preventive care screenings, prescription adherence, and VA patient satisfaction surveys. This rich dataset allows the authors to track multiple dimensions of veterans’ well-being over one- and five-year horizons.

Methods

The main empirical challenge is that higher disability payments are correlated with more severe underlying conditions. To overcome this, the study exploits quasi-random assignment of veterans to examiners with varying tendencies to assign higher or lower disability ratings. This “examiner design” is similar to judge designs used in studies of SSDI and criminal justice. Because veterans are largely assigned to examiners based on availability, examiner stringency is plausibly exogenous to veteran characteristics.

The authors construct an examiner “tendency score” using leave-out means of awarded benefits, adjusted for claim-year and location fixed effects. They then estimate impacts using two-stage least squares (2SLS), with examiner tendency as the instrument for benefit amount. This approach isolates the causal effect of additional compensation, expressed in terms of an extra $1,000 per year, on various outcomes. The design yields durable income variation: a one standard deviation increase in examiner leniency increases annual compensation by $1,445 in the first year and $1,230 per year on average over the next five years.

Robustness checks confirm the validity of the instrument. Pre-claim characteristics are not correlated with examiner assignment, monotonicity assumptions hold across subgroups, and sentiment analysis of examiner notes suggests no systematic differences in bedside manner. The exclusion restriction—that examiner tendency only affects outcomes through assigned benefits—is deemed plausible given the strictly forensic nature of evaluations. With a large sample and strong first stage, the design produces highly precise estimates, often ruling out even modest effect sizes.

Findings/Size Effects

The results show substantial improvements in economic stability, modest increases in healthcare engagement, and limited impacts on downstream health outcomes and mortality.

Economic stability:

An additional $1,000 per year in benefits reduces food insecurity by 4.1 percent over five years, from a baseline of 27 percent. Homelessness decreases by 1.3 percent from a baseline of 10 percent. Debt collections fall by 6.4 percent, reflecting fewer delinquent balances referred to Treasury. These results highlight the role of disability compensation in meeting basic needs for food and shelter and in easing financial strain. Importantly, there is no evidence of increased spending on alcohol or “sin goods,” despite the prevalence of substance use disorders among veterans.

Healthcare utilization and engagement:

Higher benefits increase VA outpatient utilization by 2.5 percent, with no effect on inpatient care. Preventive care improves: flu vaccination rates rise, hepatitis C screening increases by 0.31 percentage points, and medication adherence—particularly for cardiovascular drugs—strengthens. Veterans schedule more appointments and report higher satisfaction with care, particularly around communication and trust. These effects are not driven by lower copayments, since most veterans already face minimal out-of-pocket costs, but instead reflect reduced financial and psychological barriers to engaging with care.

Health outcomes:

Despite increased engagement, the study finds no detectable effects on most health outcomes. Rates of major depressive disorder, substance use, body mass index, blood pressure, glucose levels, and suicide attempts remain unchanged. Self-reported pain scores decline modestly by 0.3–0.5 percent, possibly reflecting reduced stress or psychosocial distress. The authors estimate precise null effects on mortality: five-year all-cause mortality does not fall by more than 0.14 percent, ruling out substantial survival gains. These findings suggest that while income improves intermediate behaviors and stability, it does not translate into measurable improvements in physical or mental health over a medium horizon.

Heterogeneity:

The benefits are strongest for economically vulnerable groups, particularly veterans with baseline incomes below $30,000 and younger veterans from recent service eras. Veterans whose claims are denied experience outcomes as poor as those rated 100% disabled, suggesting that eligibility rules may exclude individuals with substantial unmet needs. Correlated random coefficients models indicate that veterans less likely to receive benefits may in fact experience larger gains, underscoring the importance of targeting and eligibility criteria.

Comparisons to other programs:

The estimated cost to eliminate food insecurity via VA compensation is roughly $24,000 per veteran per year, and to eliminate homelessness about $78,000. These costs are higher than those of targeted in-kind programs, such as rent subsidies or food assistance, suggesting that VA DC is less cost-effective when evaluated on single outcomes. However, VA DC provides broad, unconditional support, reaching a wider population and improving multiple aspects of well-being simultaneously.

Conclusion

This study provides the most comprehensive causal evidence to date on the impacts of VA Disability Compensation for mental health conditions. It demonstrates that additional, durable income significantly improves veterans’ economic security, reduces homelessness and debt, and enhances engagement with preventive healthcare and patient trust. However, these gains do not extend to measurable improvements in physical or mental health outcomes or survival within a five-year period, aside from modest reductions in self-reported pain.

The findings highlight both the strengths and limitations of cash transfers as a policy tool. On one hand, VA mental health disability benefits demonstrably help veterans meet basic needs, reduce financial strain, and engage more with healthcare, thereby improving quality of life. On the other hand, they do not appear to resolve deeper health challenges or extend longevity, at least in the medium term. The results also reveal that the greatest potential gains may lie among denied or marginal applicants, who experience poor outcomes but are excluded under current rules.

For policymakers, these results suggest that while VA disability compensation provides critical support, it may be less cost-effective than targeted programs for specific problems like homelessness or food insecurity. At the same time, the broad coverage and unconditional nature of VA DC create value beyond any single outcome, particularly for vulnerable populations struggling with multiple, interconnected challenges. Future policy reforms might consider expanding eligibility, especially to groups at risk of denial but facing severe disadvantage, and integrating disability compensation with complementary, targeted supports.