The policy rationale for disability insurance is sensible. If you are disabled (either in the short-run or long-run) and unable to work, government support is highly beneficial in reducing inequality. Whereas other policies to redistribute income (e.g., welfare, unemployment insurance) may reduce inequality, they also may have the unintended consequences of reducing incentives to work. In an individual is disabled an unable to work, however, there is no such moral hazard. Further, the the marginal value of an additional dollar of income is likely high for disabled individuals since–assuming decreasing marginal utility of consumption–income levels are likely low for the disabled.
On the other hand, identifying people who are disabled is imperfect in practice. For instance, some conditions–such as back pain or schizophrenia–can be debilitating but do not have quantifiable biomarkers that allow administrators to classify individuals as disabled with 100% certainty. How does this imperfect identification of disabled individuals affect the social welfare of US disability programs?
This is the question that Deshpande and Lockwood (2021) attempt to answer in their NBER working paper. The authors claim that the value of disability insurance–Social Security Disability Insurance (SSDI) and Supplemental Security Income (SSI) programs in the US–depend on two factors: (i) the marginal utility of income and (ii) the counterfactual earnings (i.e., what would earnings have been if benefits were not received).
- Marginal utility. Marginal utility is estimated based on consumption levels in the Panel Study of Income Dynamics (PSID) data; various utility functional forms are used but the baseline approach assumes individuals are risk averse–implying decreasing marginal utility of consumption–under a constant relative risk aversion (CRRA) functional form with relative risk aversion of 2.
- Counterfactual earnings. The authors use French and Song’s (2014) quasi-experimental estimates of the impact of disability benefits on the earnings of individuals with different types of health conditions three years after the disability decision. Specifically, the authors use observed (i.e., with disability benefit) earnings from the PSID plus thee estimated mean reduction in earnings due to disability benefits by detailed health condition from French and Song’s (2014).
Based on this approach, the authors find:
…disability recipients, especially those with less-severe health conditions, are much more likely to have experienced a wide variety of non-health shocks than non-recipients. Selection into disability receipt on the basis of non-health shocks is so strong among individuals with less severe health conditions that by many measures less-severe recipients are worse off than more-severe recipients. As a result, under baseline assumptions, benefits to less severe recipients have an annual surplus value (insurance benefit less efficiency cost) over cost-equivalent tax cuts of $7,700 per recipient, about three-fourths that of benefits to more-severe recipients ($9,900). Insurance against non-health risk accounts for about one-half of the value of U.S. disability programs.
In other words, even if the disability program is not 100% perfect in identifying who is disabled, those who have a less severe health condition but quality for disability likely have low counterfactual income regardless, so the disability program still does a good job of redistributing income to those who need it most.
Interesting throughout.
- Deshpande M, Lockwood L. Beyond Health: Non-Health Risk and the Value of Disability Insurance. National Bureau of Economic Research; 2021 May 31.