Social Security

According to Greg Mankiw,

Concern about social security's future comes not from decades of scare-mongering by conservative ideologues but from decades of dispassionate analysis by some of the best policy economists.

He cites a 1998 statement by Bill Clinton and another by President Clinton’s Advisory Council on Social Security. Greg certainly has a point that Paul Krugman is stretching by using the word “decades,” since 1998 was less than a decade ago, and there were, at the time, clearly many relatively liberal policy analysts who were concerned about the future of Social Security (though I think the most vocal expressions of concern came from conservatives). But I think Greg is also being a bit disingenuous here.

Though I know little about the details of Social Security projections, I know something about the assumptions that go into them, and those assumptions, I’m pretty sure, have changed dramatically between 1997 (when the Advisory Council published its report) and 2007. The title of the report is “Report of the 1994-1996 Advisory Council on Social Security,” which suggests that the analysis was done before 1997, at a time when the US productivity slowdown that began in the 1970s still appeared to be an ongoing process. When productivity grows slowly, the outlook for Social Security looks bad.

Starting in the mid-1990s, but not fully apparent in available statistics until the decade was drawing to a close, US productivity accelerated to growth rates not seen since the 1960s. Productivity in the early 2000s appeared to accelerate even further. Over the past couple of years, productivity has appeared to decelerate again, but this deceleration is at least partly a cyclical phenomenon that is not expected to last (and, for the last two quarters, I might add, productivity has accelerated again, although that acceleration is also suspect). Certainly the average expectation of economists today would call for much faster productivity growth in the future than did the average expectation in 1996. When productivity grows quickly, the outlook for Social Security looks fine.

One could, however, make the point that, if we want to be honest with ourselves, we really don’t have much of a clue whether the Social Security system is in trouble or not. Any expectation – high, low, or in between – about the future rate of productivity growth is scarcely more than a slightly educated guess. To be truly conservative, we should make the worst reasonable assumption (based still on only a slightly educated guess as to what range of assumptions is reasonable), and use that assumption in the analysis, which will then tell us that Social Security is in trouble. So on this issue at least, the conservatives (and Barrack Obama) really are being conservative.

But I still have a problem with Senator Obama’s conservative position. As I understand it, the Medicare system fails even under fairly optimistic assumptions about productivity. If you make the assumptions bad enough to make Social Security require significant changes, you’ve made them so bad that the Medicare system requires a complete overhaul and damn near goes broke anyway. Given our limited analytic and political resources in coming up with and implementing solutions to these problems, doesn’t it make sense to spend those resources in such a way that we at least have a chance of coming out OK – that is, spend them on a Medicare overhaul that is almost surely necessary, rather than on a Social Security overhaul that may or may not be necessary?

Health Care Problems Exaggerated?

I’m a bit confused by Greg Mankiw’s latest blog post on the subject of health care. He seems to be arguing that, aside from redistribution issues and the perception of rising prices, the problem is relatively minor. (“...the magnitude of the problems we face are often exaggerated by those seeking more sweeping reforms...”) I suppose Greg regards the actuarial insolvency of the Medicare system as a problem of smaller magnitude than those alleged by reformers, or perhaps as a purely demographic problem that is only nominally related to the health care issue. But it seems to me, if the government has made a commitment to pay for certain things, the fact that the prices of those things are rising rapidly – regardless of whether quality is rising faster than prices – is a big problem.

I agree with Greg’s contention (in his New York Times piece) that it can be perfectly rational to spend a larger and larger fraction of our income on health care, but that doesn’t change the fact that, under current institutional arrangements, figuring out how to pay for it is a big, big problem. To put my point a little differently, those “pundits of the left” who pretend to be concerned about the health care system but really have a redistribution agenda, they would seem to be holding some pretty good cards right now, given that the government has already promised more free health care than it will be able to deliver under current fiscal arrangements.

When Greg asks the question, “What health reform would you favor if the reform were required to be distribution-neutral?” it is impossible to answer without making an assumption about how the distributions will be worked out under the current system. One possible assumption is that the Medicare problem will be solved by means testing. If so, one objection I have to the current system is that it will distort saving incentives by means-testing away the wealth that people have saved. That is an efficiency problem, not a distribution problem, but it’s hard to think how one might address that problem in a way that is both distribution-neutral and politically feasible. I believe (though Greg may disagree) that taxing rich workers is more efficient than taxing middle-class capitalists, but clearly that change is not distribution-neutral. I also believe (and Greg will probably agree) that taxing middle-class workers is more efficient than taxing middle-class capitalists, but…like that’s gonna happen.

I suspect that Greg is wrong about the motivation of radical health reform advocates. Redistribution, I would argue, is not the reason for health reform but the way to sell it. Somebody’s going to have to pay for Americans’ future health care, and if you say you’ll make the rich pay for it, the non-rich majority will be more willing to listen to the rest of your ideas.

I also suspect that Greg is wrong about why Americans are unhappy with the current system. I personally don’t mind rising prices, but I am unhappy with the current system. What makes me most unhappy (and has ever since I graduated college during a recession and had to apply for individual health insurance because I didn’t have a job yet) is the insecurity of it. Group health insurance (which most Americans get through their own or their spouse’s employer) is expensive but not prohibitively so. Individual health insurance is on average somewhat more expensive, but the problem is not the mean; the problem is the variance. If you don’t have access to group health insurance, there is no guarantee that you can be insured for any price.

There’s a distributional issue that’s important to me, too, but it’s not the rich vs. everyone else distribution that Greg talks about. And it isn’t the poor (in general) vs. everyone else either: the poor already have Medicaid. The category of people that I worry about are those who are poor, or who become poor, specifically because they (or people in their families) are sick. In some cases, it is probably their own fault for passing up health insurance when it was available. In other cases, I imagine, they never had a chance to become insured, or their insurance was cancelled.

No doubt the breadth of both of these problems – the problem of insecurity and the problem of people who are poor because of large health expenses – is exaggerated in my mind, but they make me very uncomfortable with the current system. And I don’t sense that the virtues of the current system (compared to those in other industrialized countries) are sufficient to justify the existence of these problems.

Da mihi castitatem, sed noli modo

If you look through my July archives (the early part of the month, which is lower on the page), you can see discussions of various arguments for cutting the US federal deficit and why I don’t find those arguments convincing. I’ve come up with an argument now that I do find convincing. That is, I would have found it convincing a few months ago, but now I think it is outweighed by other considerations. Here’s the gist of it:

Under current law, Medicare is going to become prohibitively expensive in another 10 or 20 years. The government will have to find a solution, and the solution will almost certainly involve either means testing or taxes. Economically, means testing is equivalent to a tax. Therefore, high taxes in the future are a virtual certainty. Optimal taxation theory says that, the higher a tax is already, the more damage is done by increasing it, and the more advantage there is to reducing it. Since taxes in the future will be high, there is a great advantage to anything we can do to reduce those taxes. One thing we can do to reduce those high future taxes is to cut the deficit today so as to reduce the debt burden that will have to be paid out of those taxes.

In general, I can’t argue with this logic, but the thing is, there is a good chance the US will go into a recession some time in the next year, and it could get quite ugly. Based on recent experience, I wouldn’t rule out a liquidity trap. So I have rather reversed my earlier position. My earlier view was, “All my logic says the arguments against the deficit are unconvincing, yet I still favor deficit cuts, because it seems intuitively like the right thing to do.” Now I would say, “I have a solid logical argument against the deficit, but I nonetheless oppose cutting it. Save the fiscal responsibility until the danger of recession has passed.”

Some people would say (1) a recession is the Fed’s problem, and the Fed can compensate for fiscal tightening, and (2) as for a liquidity trap, we can cross that bridge when we come to it, so (3) we should cut the deficit now, which will give us more of a chance to increase it later when it may be really necessary. But that strikes me as just the kind of timing mistake that has given macroeconomic fine tuning a bad name. Except for the dumb luck of the 2001 tax cut, fiscal stimulus during the post-World-War II period has always been applied much too late and succeeded only in accelerating already strong recoveries. This time around, why not try to prevent a recession? Or at least don’t deliberately make it worse. The Fed may need to bring the US to the brink of recession to maintain its credibility, but Congress has no credibility to lose. Somebody has to be the good cop.

That Which We Call a Rose

In a TCS Daily article, Jeff Miron proposes a set of budget cuts that “every economist should endorse, regardless of party affiliation…Democratic economists, and all other economists, should use their blogs, and their op-eds to highlight the enormous scope for welfare-enhancing cuts in government expenditure.” As a Democratic economist (albeit not an influential one), I hereby do use my blog to endorse Jeff Miron’s proposals. (I’m not so sure about some of the stuff he identifies as “pork”, but I’d go along with those cuts if his proposed farm subsidy cuts were also part of the package.) I make this endorsement specifically as an economist with the “Democratic” tendency to be skeptical about the importance of the disincentive effects associated with high marginal tax rates.

What puzzles me, though, is why Republican economists – who typically argue against high marginal tax rates – would support Miron’s proposals (except perhaps because they are sensible deficit hawks and would support any reasonable proposal to reduce the deficit). The heavy lifting in his list of cuts is done through means testing of government benefit programs – Social Security, Medicare, and higher education subsidies. Economically, this is precisely equivalent to increasing marginal tax rates. Think about it: if richer people were to get the same benefits as they get now but pay those benefits back to the government in the form of higher taxes, wouldn’t that be just like not getting the benefits in the first place? If higher marginal taxes discourage people from saving and investing, won’t they also be discouraged by the prospect that their future income will reduce their Social Security benefits?

Politically, “getting the rich off welfare” may be an easier sell with both parties than “raising taxes.” And there may be some substantive sense in which decreasing the amount of money that passes through government programs constitutes “reducing the size of government.” But as far as economics goes, this proposal looks like a tax increase, walks like a tax increase, and quacks like a tax increase. As I said, I support it.