Column 2005-2-8 Commentary

In his latest column, David Brooks makes a return to reality, and even rises to the point of displaying a certain amount of low animal cunning. In fact, this column is more dangerous than many of his recent columns, which were too obviously stupid, naive, and divorced from reality to actually influence the opinions of anybody with half a brain. But in this one Brooks strives, with a modicum of success, to appear reasonable. He acknowledges facts, talks about spreading the wealth, and even suggests that Bush's tax cuts be allowed to expire. That doesn't sound so bad, does it? But in fact, even as Brooks reaches out to Democrats with one hand, he prepares to stab them in the back with the other. The whole column is a classic misdirection ploy: watch closely, Brooks says, as I talk about the need for asset ownership among the poor, throw around the names of centrist Democrats, and generally extend the hand of bipartisanship. You can ignore my other hand, it's not doing anything important. It's certainly not gutting Social Security. No, it's definitely not doing that.

The fact is, though, that gutting Social Security is exactly what Brooks is proposing. In the first sentence he says "President Bush said he was open to other people's ideas on how to fix Social Security, so I hope he'll listen to mine." And he follows this up by a paragraph in which he makes "a blunt political observation": namely, that Bush's plan for phasing out Social Security is in trouble. This strongly suggests that he's actually going to discuss Social Security in this column, and the fact that he doesn't mention it again for nine paragraphs should set off alarm bells. Of course, as Brooks plans, by then most readers have been lulled into complacency by his distancing himself from Bush by pointing out that the latter is going to have trouble passing his plan to destroy Social Security and his talking about asset ownership for the poor and the necessity to fight poverty, not to mention his invoking the names of four current and former Senate Democrats: Bob Kerrey, Joe Lieberman, Daniel Patrick Moynihan, and John Breaux. The upshot is that when Brooks gets to Social Security, he can toss off a one-line proposal and quickly move on, confident that few readers will be alert enough to figure out what's really going on. This is Brooks's plan to fix Social Security (the rest of it being so much window dressing, or possibly protective camouflage): "We could start by indexing Social Security benefits to prices, not wages, so the system wouldn't go broke." He doesn't explain anything how this would fix Social Security or what the consequences would be (other than a reference to "reduced benefits" in the next sentence), because if he did so it would become obvious that this is not a mere tweak to ensure the system's solvency but a deep, deep cut.

First of all, let me point out the obvious: Social Security will never go "broke" (unless, of course, the conservatives have their way). Under the CBO's current projection, if no changes are made, then in 2053 Social Security will only be able to pay out 78% of promised benefits, which will still be a higher level of benefits than people recieve today, even after adjusting for inflation. This hardly seems to describe a system that is "broke". Those of you who have been paying attention may remember the year 2052 as being cited as the year when benefit levels would have to drop. And indeed, that was the predicted year, last year. This year, it's a year later. It doesn't take much thought to see what happens if that trend continues. (For more on Social Security and Republican fearmongering, see www.dailyhowler.com). However, Brooks's proposal would certainly mean that Social Security would be able to meet all of its obligations for the foreseeable future. (In this, he has the advantage over Bush's plan, which wouldn't even fix the funding problems: see here for more.) But first, an explanation for those of you who don't know what the difference is between wage indexing and price indexing, or even what either of those mean. Well, I didn't either, but it's really not that complicated. Wage indexing is the way benefits are calculated now: the Social Security administration calculates the mean wage of the whole workforce for each of your 35 highest-earning years before you turn sixty. Then it calculates the mean wage when you turn sixty and divides this number by the mean wage in each of the 35 selected years, obtaining a multiplier for each year. Then your wage in each of the 35 selected years is multiplied by the multiplier for that year, and the resulting wage-indexed amounts are added and divided by 35 years*12 months to obtain, essentially, your monthly social security payment. Actually, that number is massaged further, but for the purposes of understanding wage indexing, this is all you need to know (for a possibly more lucid explanation, with tables and things, you can go here). By contrast, price indexing would simply adjust each year's wages for inflation. Wage indexing reflects the overall increase in standard of living during the retirees lifetime (since standard of living goes up when wages increase faster than prices); price indexing would drop retirees back to the standard of living of forty years ago. For some solid numbers, you can go here: I'll just give you an example. If price indexing had replaced wage indexing in 1959, a worker who had earned an average wage over the course of his or her lifetime and retired in 2003 would recieve a benefit 40% lower than under current rules, which would live him or her 5% under the poverty line. This is why Brooks doesn't want to go into detail about his plan to fix Social Security.

But what about the rest of the column? Giving everybody a tax-deferred savings account of $3500 which could be invested in a few selected mutual funds but not accessed until retirement? Brooks says "by the time workers retired, they would each have a substantial nest egg, over $100,000, waiting for them" (assuming that they didn't invest in the wrong mutual funds), and it sounds pretty good, frankly. However, there are a couple of important points to be made here. First of all, Social Security doesn't only provide for retirees: it also provides disability and survivor benefts. Where would those be coming from under the Brooks plan? Also, what about the people who are too old right now to have had their tax-deferred savings account working for them since birth? Do they just get screwed? Amazingly, Brooks actually proposes a way to pay for these accounts: letting the Bush tax cuts expire. (Of course, according to Paul Krugman, it would only take 1/4 of the revenue from letting the tax cuts expire to put Social Security on a sound financial footing). But the final thing to consider is, will Brooks's plan even replace the Social Security benefits that he is cutting by switching to price indexing? A lump sum of $100000 or more sounds pretty good, but consider that Social Security pays out an inflation-adjusted amount every month for years. Under the current system, the average monthly payment is $1158 (according, again, to the socsec.org link above). Say you live for 20 years after retirement: that's more than 270,000 dollars, without even adjusting for inflation. Now consider someone born this year and thus retiring in 2072. If the Brooks plan is implemented right away, this person will have, say, $200000 (and that's being quite generous) in 2005 dollars from his Brooks account. Say this person was an average wage-earner throughout his life. The question is, can this $200000 replace the 46% or so of his monthly Social Security benefit that he lost? Well, let's say he lives for 25 years, a not unreasonable assumption for 2072. Then that $200000 pays out $667 a month until he dies. Yes, it will keep earning, but unless it's earning at considerably higher than the inflation rate, the new amount will be negligible. So the question is, what will his Social Security payout be? Well, according to this article (which is also a good explanation of why price indexing is bad), between 1988 and 2003 wages outperformed prices by 22%. Extrapolating using this number, between 2003 and 2072 wages will outperform prices by about 77%. Therefore, the promised benefit in 2072 will be 77% higher than it is today (in today's dollars), or about $2050. 46% of this is $942. Therefore, the private account that Brooks proposes will be able to replace lost Social Security benefits only if the recipient doesn't live that long past retirement: anything longer than 17 years or so, and he stands to lose. And the problem will only get worse as time goes by. Of course, inflation could be higher, reducing the value of the promised benefit, but then that will start to cut into the value of the savings account: if inflation is sufficiently high, the savings account will actually lose value, making it an even worse proposition.

At this point, I should insert a caveat: I am not an economist. Nor have I double-checked all these numbers. But I feel fairly confident in the general outline of the argument. And note that the $200000 figure I used is considerably higher than Brooks's.

Also, Brooks could point out at this point that by 2072 the system will only be paying out 80% or so of benefits. Possibly true (the CBO's predictions are very pessimistic), but even so there's a simple way to return solvency: raising the cap on payroll taxes. Income above $90000 is not taxed by the payroll tax. Raising or eliminating this cap would result in a lot of extra revenue. Need more money? Bring back the estate tax on all estates worth more than $500000, or a million. Or, as Brooks himelf has proposed letting Bush's tax cuts lapse, we could use that money.

All this is bad enough, but the real killer has yet to be exposed. I should say that I am indebted to this article in The New Republic for pointing this out: even I had been sufficiently lulled by Brooks to miss his attempt to privatize Social Security through the back door. That attempt comes out here: "or individuals could divert some of their payroll taxes into their KidSave accounts, trading guaranteed benefits for more ownership." And here we are, back at privatized accounts, only with some government seed money. Incidentally, since the accounts cannot be accessed until retirement, it's unclear how any more "ownership" would result. But the salient point here is that any movement of payroll taxes out of Social Security and into personal accounts is the beginning of the destruction of the system, because that money is needed to pay for people's retirements now. And that means that if your private savings accounts don't do quite as well as you had hoped, you will be in serious trouble when you retire, which is the whole reason that there is so much opposition to Bush's plan.

So, it turns out that Brooks's plan is basically just like Bush's, only with a lot of talk about spreading the wealth thrown on top of it. This isn't particularly surprising, really -- after all, how much original thought is Brooks capable of? What is surprising is the cunning with which Brooks presents his proposal, burying the key points behind a mountain of pointless verbiage about the ownership society. Perhaps we've been underestimating him all this time? Well . . . . Nah. But the moral is that when Brooks appears to be at his most reasonable (not a very high standard, admittedly), he is also at his most dangerous. We can only hope that the Times will be reasonable and get rid of him.