No problem here

By Megan McArdle

Is Social Security really just all fine, its alleged demise a crazy Republican talking point? Matt's accusation echoes a number of the ones that I've seen on liberal blogs:


The Post even includes bonus inaccuracy:

Because Social Security increases are pegged to wages, rather than inflation, economic growth alone won't solve the problem. Fiscal responsibility first is fine; fiscal responsibility only is an irresponsible dodge, as Ms. Clinton well knows.


This is just wrong. Social Security benefit increases are, indeed, partially tied to wage rates but it's still true that the faster the economy grows the more affordable promised benefits become. Indeed, that's the basic premise of pay-as-you-go financing of social insurance schemes. The relatively poor present borrows from the relatively rich future. All the Post would need to do is to look back at past SSA Trustees' Reports and they would see that when the economy grows faster, the outlook for Social Security's finances gets bigger. They would also see that if the SSA updated its projections of likely future productivity growth to reflect the post-1995 return to pre-1973 levels of high productivity growth, that the alleged financial problems would substantially diminish.

It may (or may not) turn out that, in fact, the economy does not grow fast enough to close the financing gap. But this isn't a logical fact about the nature of the program, it's a contingent hypothesis that the Post seems to be subscribing to even though its editorial writers don't appear to understand what the hypothesis is or how Social Security works.



But I don't think this is right. Start with the unfortunately common meme that the Social Security administration's claims have historically varied. Yes, they have, but not lately. The SSI trust fund figures did grow for a few years in the late 1990s, due to a combination of changes in assumptions about labor force participation, productivity, birthrates, and average wages. But since 2002, the dates for the trust fund exhaustion have fluctuated around a rough mean of 2040.

And though the long term projections have gotten somewhat better as a result of improving fertility, the short-term projections have gotten slightly worse. Ten years ago we thought that program income would fall below program outlays in 2019; now the projected date is 2017. Meanwhile, the estimated year in which the social security surplus will peak has also moved steadily backwards, to 2010. In 2011, a small hole will appear in the general fund budget as OASDI revenues start to decline, a hole that will become a gaping wound by 2020. From the Social Security Administration's perspective this is fine, but from the taxpayers perspective, this is a huge problem that needs to be, well, fixed--and from which economic growth is unlikely to rescue us.

More broadly, Matt's criticism of the "economic growth won't save us" argument seems to misstate the underlying changes that drove the improved predictions. If you actually look at the difference between the assumptions in the 1997 trustee's report and those in the 2001 report, you'll see that the improvement in assumptions came only in part from real wage growth (a decent proxy for productivity). More of it came from changing assumptions about birthrates, unemployment rates, and the interest rate that the SSA collects on its assets. But unemployment rates can't just keep falling forever, particularly since the Baby Boomers are our largest generation, and older workers tend to stay unemployed longer. Likewise, birthrates could shoot up, but demographics is a slow-motion train wreck; to fix things in 2041, you'd have to be having the babies to fund it right now, so they have time to get out of college and get some work experience. We aren't. Indeed, the latest data from the census bureau show that the birthrate for child-bearing women fell slightly in the early part of this decade. Meanwhile, death rates never seem to reach that much promised plateau, making the problem worse.

If economic growth were going to save us, it should have while we were paying for the unusually small age cohort that preceded the boomers. If productivity growth could save Social Security's finances, we should have seen it push back the date at which the government starts paying out more in OASDI benefits than it takes in in taxes. Instead, the reverse is true.

This article available online at:

http://www.theatlantic.com/business/archive/2007/10/no-problem-here/2038/