The Agenda

NRO’s domestic-policy blog, by Reihan Salam.

Charles Blahous on the Social Security Retirement Age and Dylan Matthews on the Disability Benefit


Dylan Matthews makes an excellent point: the Social Security disability program is in desperate need of reform. Incentives are badly misaligned, and eligibility standards vary widely across jurisdictions. Dylan points to a new proposal from the Hamilton Project as a possible solution:

So David Autor and Mark Duggan at the Center for American Progress and the Hamilton Project have proposed (PDF) mandating employer-provided disability insurance. If people acquire a disability on the job, they can use this insurance until they can return to the workforce. If that isn’t possible within 21 months, they can apply for Social Security disability insurance. That way, fewer people will drop out of the workforce, the Social Security program’s rolls will shrink, and program costs will fall, without failing to provide support to disabled workers. Deficit reduction has few win-wins, but this is one of them. 

The beauty of this approach is that it does a better job of encouraging continued attachment to the labor force:

The proposed universal PDI [private disability insurance] coverage would shield employers from catastrophic insurance costs. Under our proposal, PDI policies would hand off responsibility for ongoing disability coverage to the traditional SSDI program two years following the onset of disability. While employers will face a financial incentive to reduce PDI claims, they will not be exposed to the long-term costs of a permanent, work-limiting disability. [Emphasis added]

Federal officials who determine eligibility for disability benefits don’t feel an immediate pinch when it grants disability benefits. But an employer that fails to accommodate work-capable workers will feel a pinch if she allows them to go on disability, because it will raise the cost of her PDI coverage.

Under our proposal, private disability insurers, contracted by employers, will ensure that employers comply with the ADA in the event of a work-limiting disability, pay for reasonable accommodations, provide limited and temporary income replacement to workers with disabilities, and assist employers to limit the cost of disability claims through cost-effective interventions.  Employers also will have the option to apply for the right to self-insure, as many large firms currently do for health insurance, rather than contract with a private insurer for this coverage.

These insurer and employer costs will be time-limited and not prohibitively expensive, either on average or in any specific case. Employers will, however, face an incentive to minimize the economic costs of disability to keep their insurance premiums low. As we shall discuss, preemptively terminating workers with disabilities will not be a cost-effective way to minimize costs. This is because a worker’s employer-based PDI policy will remain in force for up to a year following the date of job termination. 

Moreover, universal PDI mitigates the adverse selection problem:

It reduces the adverse selection problem that employers currently face when considering whether to offer PDI coverage to their workers. An employer that offers PDI coverage at present faces some risk that this coverage will differentially attract workers with disabilities, which will in turn lead to correspondingly higher claims rates and policy premiums. This adverse selection may be one reason that employer-based PDI coverage is only offered to one-third of U.S. workers at this time. Requiring all employers to offer such coverage, as is currently the case with unemployment insurance and workers’ compensation insurance, will eliminate this adverse selection problem.

As the authors explain, the fundamental problem with SSDI is that disability is not necessarily all-or-nothing. When SSDI was created, it was assumed that workers would only go on disability if they had no other choice. That has changed:

Workers coping with a work-limiting disability face a choice between striving to remain in the labor force while receiving no income or medical support from the public disability insurance system or effectively forgoing efforts at maintaining employment and self-sufficiency so as to be eligible for SSDI consideration. This potentially creates a catch-22 for workers who develop health limitations that are significant but not necessarily career-ending: it may be difficult or infeasible for them to remain employed and economically self-sufficient absent disability assistance, but it is not possible for them under the current SSDI program to obtain this assistance without first leaving the labor force. …

During the applications and appeals process, claimants receive no income support, workplace accommodations, or medical benefits from the SSDI program. But they face strong incentives against participating in the labor force—even on a trial basis—since evidence of gainful employment would disqualify their claim.  

Indeed, this is one reason why male labor force participation has declined in recent decades.

My only disagreement with Dylan is over his choice of title: “The part of Social Security that does need reform.” The clear implication is that the rest of the program doesn’t need reform. The fashionable view among left-of-center activists and opinion journalists is that the only problem with Social Security is that it isn’t sufficiently generous, and that the retirement age should remain untouched. Charles Blahous has addressed this arguments at Economics 21, my home institution, and in a follow-up post.

Blahous offers a great deal of historical context that is important for understanding how Social Security has changed over the years, e.g., COLAs were enacted in 1972, early retirement was enacted in 1956 for women and in 1961 for men, and the early retirement is far more common now that it was at its inception, despite increases in life expectancy at age 65.


Sign up for free NRO e-mails today:

Subscribe to National Review