A Response To Michael Hiltzik On Reforming SSI

I recently participated in a vitally important project that produced the YG Network's new reformocon policy book, Room to Grow. I contributed a chapter on antipoverty policy, in which I made the case for work-promoting reforms to means-tested programs and a voucher-based early childhood program. My not-so-secret agenda is encouraging conservatives to develop a robust upward mobility program. Indeed, all of the reformocons in the book have the not-so-secret agenda of moving conservative policy in a direction that offers more help to the poor and middle class without abandoning conservative principles and while promoting market-leveraging solutions.

But L.A. Times columnist Michael Hiltzik can see right through all that to the “contempt for the underprivileged — especially the disabled” that my co-contributors and I bear. He uses his latest column not only to shine a light on my cruelty, but to expose Robert Stein's nefarious plan to give middle-class families tax cuts for its real goal: making people have kids to support their retirement, so that senior entitlements can be abolished, or something. Even The New Republic's Danny Vinik is exposed as “desperate” for thinking the book has “valid” ideas. Oooo . . . kayyyy . . .

Let me take a moment to defend my own contribution against Hiltzik's unhinged critique, which substitutes righteous indignation for solid evidence. I'll stick to the Supplemental Security Income (SSI) program here and save the rest for another post. SSI serves poor elderly, blind, and disabled children and adults. I argue in my essay that families have increasingly tried to receive SSI benefits for their children rather than the less generous and less open-ended benefits from the main cash assistance program for poor families with kids. States have encouraged these efforts for their own reasons, as I will discuss below. The result is that the beneficial work-promoting reforms to welfare that have been implemented over the last 20 years have failed to help these families. In fact, many of these SSI kids graduate right onto the adult SSI rolls when they become adults, which will hamper their upward mobility as young adults and beyond.

Hiltzik says my claim that “[i]ncreasingly, families seek and obtain disability status for children with comparatively minor (and often dubious) learning disabilities or behavioral problems” in order to receive SSI benefits is a “fact-free assertion retailed by a succession of appallingly lazy new reporters and repeatedly debunked by disability experts.” Well, glad it's not just me!

He concedes that I actually cite some research, specifically a book by Cornell economist Richard Burkhauser and Federal Reserve Bank of San Francisco economist Mary Daly, both of them “disability experts.” The problem, Hiltzik says, is that he “couldn't find any statement in that study that children on disability characteristically had only ‘minor' or ‘dubious' impairments.” A Google Books search confirms that “dubious” does not appear in the book, while “minor” appears once on page 33 in an unrelated context. Of course, those words are mine, not the book's, which is why there are no quotation marks around them in my essay. But the point I make and attribute to Burkhauser and Daly is theirs in Chapter 6 of the book. I'd encourage Hiltzik to reach out to Burkhauser, who I count as a friend, if he thinks I'm misattributing the point. And I'd encourage him to review the chapter if he wants their empirical case.

Hiltzik says that by themselves, learning disabilities and behavioral problems don't qualify kids for SSI “as any expert” could have told me. That would be a devastating rebuttal had I argued that they did. In theory, parents of children with these problems have to show “marked and severe” functional limitations and that the condition is likely to persist for at least a year (or to kill them, and of course, there are income and asset requirements). The questions for anyone interested in SSI policy — as opposed to feeling good about one's pure-heartedness — are these: (1) whether the share of children with learning disabilities and behavioral problems has increased in a way that suggests that children with relatively minor impairments increasingly receive SSI instead of less generous cash assistance, and (2) whether this is a problem for those children.

Here is a chart that took me about 15 minutes to create, using data from the Social Security Administration's SSI Annual Statistical Report and U.S. Department of Health and Human Services data compiled by the respected nonpartisan group Child Trends:

In 1990 there was one child on SSI for every 25 on Aid to Families with Dependent Children (AFDC), the primary welfare program for poor families. By 2010, there was one child on SSI for every 2.7 children on Temporary Assistance for Needy Families (TANF), the post-1996 version of AFDC.

What happened over this period to expand the SSI rolls? The landmark 1990 Sullivan v. Zebley decision broadened eligibility of children for SSI, and program administrators were left with more discretion in determining eligibility in specific cases. Meanwhile, welfare reform began in the early 1990s when Presidents Bush and Clinton began issuing waivers to states to encourage work and discourage dependency on the Aid to Families with Dependent Children program. In 1996, the federal overhaul of welfare tightened up SSI eligibility somewhat but left it looser than before Sullivan. (This is all covered by Burkhauser and Daly, but the points are well understood, perhaps even by Hiltzik.) At the same time, welfare reform made TANF a much less attractive program than AFDC was.

These changes gave families strong incentives to attempt to go on SSI instead of AFDC or TANF. SSI benefits were more generous and came with no work requirements or time limits (which tightened considerably after 1996). States had equally strong incentives to direct families onto SSI when they could. SSI is entirely federally funded, making it cheaper for states to serve a family, and after 1996 states came under great pressure to reduce their TANF rolls and show that those receiving TANF were enrolled in welfare-to-work activities.

After 1990 children became a bigger share of SSI recipients, and enrollment rates among low-income children rose steadily (Figure 2). This expansion involved children with comparatively minor mental impairments rather than physical disability or mental retardation. Children with such impairments have grown from about 10 percent of SSI kids in 1990 to a majority, while those with mental retardation have fallen from about 40 percent to 10 percent (Figure 6-4 from Burkhauser and Daly, based on Social Security Administration data). Burkhauser and Daly cite research showing that SSI enrollment is sensitive to AFDC/TANF generosity (also briefly reviewed here).

Against all of this evidence, Hiltzik offers no reason to believe that the increase in SSI enrollment among children with mental impairments reflects an increase in need. Poverty rates among children have declined, as I show in my piece, despite the fact that the AFDC/TANF rolls declined much more than the SSI rolls increased. It is implausible that the rate of mental impairment has grown as dramatically as the SSI rolls. In other columns, Hiltzik has cited Harold Pollack, who notes a University of Michigan study finding that just 8 percent of participants in a local survey of TANF recipients begun in 1997 ended up receiving SSI benefits in the subsequent six years. However, Pollack is citing figures for SSI receipt by the mothers in the study (who may qualify on the basis of a disability) rather than receipt of child SSI (see footnote 12). It is entirely inapt.

Pollack and others also cite the high denial rate of child applications nationally as evidence that few children are on SSI illegitimately. Two in five child applicants for SSI ultimately receive awards (Table 69). That is a weak argument for the obvious reason that it says nothing about how many should have been denied, which might be a higher number still. Given that the rise in SSI receipt was not simply a one-time jump immediately after Sullivan and that there is no reason to think poverty or mental impairment rose, we might presume that the number of illegitimate applications rose. In that case, the denial rate should have risen too. But while it rose in the early 1990s, it hasn't really increased since 1996.

Finally, Pollack and others have said that there's little evidence that dubiously qualified child SSI recipients graduate into adult SSI. Seventeen-year-old SSI recipients with “mental and behavioral disorders” other than mental retardation have lower rates of entry into adult SSI at age 19. This is true, and it would be shocking to find otherwise. But nearly half of these kids still do move directly into adult SSI when they age out of child SSI (Table 5).

Hiltzik seems to believe that I (and maybe the other reformocons) don't care about poor kids, or maybe it's just poor disabled kids I hate. If he could have read the piece without prejudice, he would have seen my insistence that poor children questionably on SSI and their families “would be better served on TANF, even if the adults in them remained non-working.” At least then fewer children would graduate into adult SSI and a near guarantee of immobility. I prefer keeping child SSI a program for families whose disabled children require intense demands on their time and create large expenses. I would keep everyone else in TANF, where I would require those who can work to eventually do so but retain an enduring safety net during recessions and for families with deep employability challenges during expansions.

But when you are assured of your own righteousness, like Hiltzik is, that's all you need to know, apparently.

I'll take on Hiltzik and his ilk in a second post about Social Security Disability Insurance — the primary programs for disabled adults — which I argued should be reformed to promote work among less-skilled able-bodied men who in the past would have been working.

This piece originally appeared in National Review Online

This piece originally appeared in National Review Online