1

MacArthur Foundation Network on Mental Health Policy Research —Sheldon Danziger, Richard Frank and Ellen Meara

MENTAL ILLNESS, WORK, AND INCOME SUPPORT PROGRAMS

Improved treatment makes it easier and more Number of Recipients with practical for individuals with mental illness to Mental Illness Grows Steadily participate in mainstream society, including the To limit the risk of enrolling those who are labor force. However, a diagnosis of a severe capable of working, both ssi and di require ample and persistent mental illness or a substance use documentation of a disability. Only one–third of disorder is still often accompanied by a drop in initial applications are accepted, and even after employment and lower earnings. appeals, most applicants, according to the Social Security Administration ( ssa ), are not awarded The Social Security Disability Insurance program benefi ts. If the individuals show improvement, (di ) and the Supplemental Security Income ( ssi ) benefi ts are often terminated. program provide income support for increasing numbers of individuals with mental illness. More For many years, the ssi and di programs served only than  percent of the current caseload receives those with physical disabilities or the most disabling benefi ts based on a diagnosed mental disorder. mental disorders. However, beginning in the late A growing share of a third program, Temporary  s the eligibility criteria were restructured to Assistance for Needy Families ( tanf ), which address conditions such as mental disorders and offers cash support to low–income single mothers, chronic pain. Between  and  , according to is composed of individuals with mental illness, as ssa data, the share of the ssi (aged  to  ) disabled new work requirements result in faster exits of due to a mental disorder rose from  percent to those without mental health conditions.  percent. Likewise, di awards for mental illness were fewer than  percent of new awards in  . In These programs have come under increasing  , individuals with a mental illness represented scrutiny as the shares of recipients with mental nearly  percent of all benefi ciaries and  percent illness increase. Like most illnesses, there are many of those under age  . variations and gradations of mental illness. Some may completely prevent individuals from working, Mixed Evidence for Perverse Incentives while others may be less debilitating. Many, in fact, Some argue that the reason for this rise in question whether these programs serve only the ssi caseloads in the  s was the loosening “truly needy” and not also those who are capable eligibility criteria. However, as Danziger and of working and surviving on their own. Are these coauthors argue, the question remains: were more programs enticing individuals to choose them over “undeserving” applicants fl ooding the rolls or work, even when they are capable of working? were the prior eligibility rules in the  s simply too exclusionary? Sheldon Danziger, Richard Frank, and Ellen Meara examine the research on this issue in their Others argue that the screening process is broken, paper “Mental Illness, Work, and Income Support and point to the parallel rise and fall of disability Programs” for the Funda mental Policy – Spotlight applications and unemployment rates during the on Mental Health Conference.  s. In other words, if individuals can work, they 2

do, but when employment tightens, they apply for completely understand the dynamics of the mental disability benefi ts. health population when designing policies for the di and ssi programs. Although di applications do follow the unemployment rates in the  s, Danziger, Frank, and Meara fi nd The authors raise three questions for policy that the award rate, or the share of applications research and practice and suggest that the policy approved for benefi ts, is counter–cyclical over response to addressing income support for the entire  – period. They also fi nd a those with mental illness will differ dramatically continuous rise in ssi rates after  for those depending on answers to each. with mental conditions while the trend for those with disorders other than mental disorders closely First, are the criteria for eligibility into ssi and di mirrors the unemployment rate. too stringent, too lax, or both? Barriers should be removed or reconfi gured accordingly. Within the tanf program, the authors fi nd little evidence of misplaced incentives among a large Second, are current rules and regulations to prevent group of Michigan recipients between  and the able–bodied from using limited resources  . Slightly more than one–third of this group appropriate? Given the unique nature of mental met the diagnostic screening criteria for major disorders, program design features to reduce this depression, posttraumatic stress disorder, and risk should differ as well. generalized anxiety disorders. Yet only  percent received ssi . Another  percent applied for but Finally, what are the mechanisms driving the rising did not receive benefi ts. Further evidence that caseload for mental disorders in each program eligibility criteria may exclude those in need is the over time? The current focus on eligibility continued vulnerability of those rejected for the criteria is misplaced if, for example, the mental program. Greater proportions of those rejected disorder caseload in fact reflects shifting labor for ssi had unstable housing and homelessness markets from manufacturing toward services. than those who applied for and received ssi . Also, People with mental illness might have been able those who were rejected were much less likely to be to work on a factory line but may be less able to working (or to have worked in the past) than those interact with the public in a service job, which who had never applied for ssi . could in turn affect employment rates of those with mental disorders. Areas for Future Research As ssi and di caseloads rise, policymakers are Addressing these questions will allow policymakers increasingly concerned that the eligibility criteria to improve the effi ciency and effectiveness of public are too lax. The evidence that ssi and di are serving support systems by ensuring the appropriateness more than the “truly needy,” however, is mixed. It of enrollees and maintaining a manageable fl ow is therefore imperative that policymakers more of enrollees.

Sheldon Danziger, PhD, Ellen Meara, PhD, The MacArthur Foundation Network on Mental Gerald R. Ford School of Public Policy, Department of Health Care Policy, Health Policy Research has worked to develop a University of Michigan Harvard Medical School knowledge base linking mental health policies, financing, and organization to their effects on Richard Frank, PhD, The authors acknowledge the support access to quality care. www.macfound.org Department of Health Care Policy, of the Robert Wood Johnson Foundation Harvard Medical School in the preparation of this material.