Let's not save Social Security!

Let's not save Social Security!

Let's not save Social Security!

Political commentary and more.
May 5 2005 1:46 PM

Let's Not Save Social Security

Once it's fixed, we can't fix it again!

(Continued from Page 1)

The Pozen Plan, which Bush has now endorsed, is different. It reduces the benefits of those who make high wages in their working years by indexing them to prices instead of wages. Low-wage workers would get full wage-indexed benefits. But unless you assume we live in a rigid, static meritocracy, the wages people make will not correspond exactly to their financial status when they reach retirement age. Some low-wage workers will be quite affluent when they retire. (Maybe they owned suddenly-valuable homes, or they've saved, or inherited, or had a few really good years, or they're not poor at all but rather lazy coupon clippers who lived off investment income not subject to the payroll tax.) Some high-wage workers will retire penniless.

The Pozen plan doesn't care--it pays them all benefits based on how well-off it looked like they were going to be when they were paying in the system.  The Pozen plan is really not a means test at all but rather an adjustment in Social Security's existing benefit schedule, gradually lowering the benefits to which higher-wage earners are entitled, whether they are wealthy or imporverished when they get older (although high-wage workers would still get slightly higher benefits than low-wage workers). 

That doesn't make it a bad plan. Means tests are said to be humiliating--you have to prove you're poor--and Pozen avoids that. Plus, unlike true means-testing, his plan encourages recipients to save for their senior years. (Means testing would punish them with benefit cuts if they save too much.)


But once we recognize that Bush and Pozen are actually just fiddling with the benefit schedules, we can acknowledge that ... 

5. Democrats would fiddle with the benefit schedules too! The most frequently cited Democratic alternative plan is the one proposed by Peter Diamond and Peter Orszag--call it the Two Peters Plan. They estimate the total 75 year Social Security shortfall at 1.9 percent of taxable payroll, and would meet it by a combination of tax increases and benefit cuts. Specifically, as best I can make out,, they make three changes in the benefit schedules: 

1) They cut benefits for the top tier of wage earners, reducing by a third the amount of wages over about $44,,000 that get replaced by Social Security at retirement. This change alone saves almost 10 percent of Social Security's shortfall (.18 percent of payroll).

2) In addition, they cut overall benefits gradually to compensate for greater longevity--accounting for half of another .55 percent of payroll--or .27 percent. It's not clear how these cuts would be allocated.

3) In addition, they would impose a "legacy charge" of .97 percent of payroll, more than half of which is financed by benefit cuts, or another .50 percent of payroll.

You get the impression that Two Peters chop up these benefit cuts into little parts because they'd rather you didn't add them all up. When you do you get a benefit cut totalling .95 percent of payroll, or about 50 percent of the total shortfall. In the liberal-endorsed Two Peters' plan', as in the Bush-endorsed Pozen plan, "reductions are smaller for lower earners, and larger for higher ones." 

Sure, the Two Peters plan is a gentler than the Pozen plan, which uses benefit cuts to make up 70 percent of the shortfall. But is that what all the fuss is about--70 percent versus 50 percent? It would appear to be a question of degree, and not all that much degree neither.  

6. Indeed, the Democrats' solvency plans are grimmer than you'd think: Democrats charge that Bush's proposed benefit cuts are vicious, and that they don't fix Social Security's solvency problem. But if even vicious cuts don't fix the problem, doesn't that mean the problem is a bit bigger than the Dems have been letting on? I believed Social Security wasn't in much trouble at all--just needed a few "tweaks"--until I looked at the tweaks the Two Peters were proposing. In addition to the benefit cuts outline above, Diamond and Orszag have the current 12.4 percent Social Security payroll tax rising to 15.4 percent in 2078 and continuing to rise "slowly over time thereafter." Even if the Medicare tax is kept at its current 2.9 percent (a seeming impossibility) that means total FICA  payroll taxes in excess of 18%. You want to try to finance universal health care on top of that? I don't.

7. We could radically cut the cost of Social Security:  The alternative to raising taxes is cutting benefits more. As long as we're going to be accused of making large cuts, we might as well actually make large cuts and radically reduce the amount of national income eaten up by Social Security--freeing up that money for other purposes. That's what abandoning a universal program and instituting a true means-test would do.

The Democrats--and Republicans-- now whining about the Pozen plan claim it will turn Social Security into a welfare program. That's silly. For one thing,as noted, Pozen would keep paying high-wage recipients benefits--higher benefits than lower earners get. If Democrats want to know what a benefit cut that really zings the rich looks like, they should go to Australia, where (last time I checked) the top quarter of recipients gets no benefits at all. Zero. The bottom half gets full benefits. The people in between get in between. Now that's a means test! Not coincidentally, after means-testing was introduced in the 1980s, Australia's pension system cost a little more than half what ours costs, in terms of GDP.

8. Even a radical means test wouldn't turn Social Security into welfare: Why? Traditional "welfare" programs--most obviously the old Aid to Families with Dependent Children (AFDC)--help the poor whether or not they work or try to find work. But Social Security is "work-tested." You can't get Social Security checks unless you've worked and paid in contributions. That means Social Security will never be stigmatized the way AFDC was stigmatized. NYT columnist Paul Krugman, criticizing Pozen, repeats the old saw that "programs for the poor always turn into poor programs." But even if Pozen did make Social Security a program for the poor, which it doesn't, the old saw isn't true. The disproof: the Earned Income Tax Credit. It's a program for the poor--it goes only to people making less than $35,000. But it's a good program! It works. It's popular. Congress after Congress has supported and indeed expanded it. It's popular because, like Social Security, it's work tested. As its name implies, it only goes to people who've earned some income. If means testing makes Social Security as unpopular as the EITC, Democrats have nothing to fear. 

9. Do we really want to save Social Security now? We're told, by both the President and the editorial boards, that the responsible course is to a) put Social Security on a path to solvency and b) do it sooner rather than later. Both aspects of this responsible advice are suspect.

Sure, some gradual changes--like raising the retirement age--are easier to swallow if they're started early. All other things being equal its better to give people lots of advance notice of any changes, so they can plan. But most of the changes that now being talked about are changes that happen well into the future. As Kevin Drum argues, we could wait a decade and not lose too much ground. 

And there is a big risk associated with saving Social Security now--the risk that we'll save too much. Nobody--neither the President, nor Pozen, nor the Democrats--is talking about radically reducing the size of the program, Australia style. They're all talking about saving a program that consumes somewhere from 12 to 17 percent of the national payroll.  Once its financial imbalance is solved it will be virtually impossible to restructure. Voters will consider the problem taken care of. The system will be solvent-a tub resting firmly on its own bottom, funded by a dedicated payroll tax-so why talk about changing it?

Before they stabilize the system at this high level of GDP consumption--cementing it in place,in effect--Democrats may want time to think about whether they want to devote such a large part of society's resources to a universal check-mailing scheme. My guess is Democrats will need at least some of those GDP points for health care. It will be easier to get them if Social Security is still perceived as a progam in need of reform, as opposed to a program that got fixed back in 2005.  

The reason to postpone the solution, in other words, isn't Drum's reason--that if the American economy goes better than expected Social Security might not need cutting back in future decades after all. The reason to postpone the solution is that if American politics goes better than expected we may want to cut Social Security back in future decades even more radically than anyone is now contemplating, in order to pay for more important (and more deeply equalizing) government programs. Would you take a deal that gave us universal Medicare-style health insurance if the price was cutting down Social Security into a mere program of earned insurance against poverty? It seems like a no-brainer to me. But it's only possible if Social security is perceived as in need of fixing--even better, if it's in crisis!   

As Kinsley notes, Bush's private account plan lacks economic logic (if it's not funded with new taxes it actually makes the solvency problem worse) but it has a political logic, which is this: it's something positive to offer voters in exchange for the pain of solvency-producing reforms. Otherwise the future of Social Security is all Europe-style negativity--cuts and taxes, with noisy fights over which it will be. It might be worth a small private account scheme, Kinsley argues, if that's the necessary spoonful of sugar that helps the medicine go down.

But this is medicine we don't need to take right now. And the Democrats will have a better spoonful down the road--health care--that will require a bigger dose of medicine. Let's take it then.