Monday, August 01, 2005

It's Been Mostly Dead All Day

Like Mark Thoma and Brad DeLong, I am left scratching my head after reading Kevin Hassett's column in Bloomberg about the "death" of Social Security reform. I'll focus on this passage:

Until two weeks ago, Social Security reform was sort of dead. But now it seems to be all dead. The breakdown occurred when the administration backed away from a proposal making its way through the House of Representatives that would have introduced personal accounts without specifically restoring solvency to the system.

Ben Bernanke, chairman of President George W. Bush's Council of Economic Advisers, publicly signaled the White House's displeasure with such an approach.
Asked if restoring solvency was an inviolable condition, Bernanke said, "Yes, I think the president will insist on maintaining the long-term solvency of the Social Security system.''

The word from the other side of town, up on Capitol Hill, is that this signal from the president sucked the remaining life out of the House measure. There appears to be nothing left for even Miracle Max to work with.
The Administration should have backed away from a plan that would introduce personal accounts without specifically restoring solvency. From the outset (e.g. the President's 2001 Commission) personal accounts have been the sugar to make the medicine go down. The medicine is restoring solvency. And we particularly applaud Ben Bernanke for making it clear that taking the medicine is something to insist on. If that sucked the life out of the House measure, so be it. Nothing prevents them from starting over and making improvements.

Let's hope they do, or that the President puts enough specificity on his proposal so that it can be scored by the Chief Actuary at SSA and be used as the basis for legislation.

Other blogs commenting on this post


JG said...

The DeMint proposal to put SS bonds in private accounts without addressing solvency has been scored at ...
... for any who are interested.

As to "solvency" -- if by that we mean long-term actuarial balance over 75 years or some such period -- I'll take the contrary position that it has never mattered to Social Security in the past, doesn't now, won't in the future, and even if we restore solvency to SS today it will do very little to solve SS's real big problems that will hit in about 20 years.

Although FDR's original SS program was designed using a "funded" scheme with actuarial balance in mind, the very first Congress to sit with payroll taxes coming in intentionally ended its solvency by turning SS "paygo", spending the taxes through benefit increases and tax hikes.

Later, in SS's glory years, its champions bragged that it was insolvent on a long-term basis, such as Samuelson in his famous Newsweek column calling it "A Ponzi scheme that works": "The beauty of social insurance is that it is actuarially unsound".

Clearly, long-term solvency never mattered a whit to SS's tenders in the past -- which is why it went broke in 1983, when a solvent program would never have been allowed to approach that point and would have had many billions in the trust fund. What always mattered was cash flow solvency.

And when that cash flow crunch hit in 1983 Congress didn't fix the situation by making SS long-term solvent, but by fixing cash flow for a generation only, by cutting benefits and increasing taxes 50-50.

Now look at the future. The SS Trustees say that by 2030 income taxes will have to increase 60% from today's levels to fund promised entitlements, by 34% just to finance Trust Fund operations, and rising from there. [data]

That's a cash-flow crunch, there's never been anything like it. And a long-term solvency fix enacted this afternoon that counts the trust fund bonds as SS assets would do zero, nothing to close it. (While one that doesn't count the bonds would do very little.)

It is inconceivable that to close a cash flow gap much, much, larger than 1983's, SS benefits won't be cut one way or another by more than in 1983. If they are, then 75+ year actuarial balance calculations from today will become totally moot then. And if we know today such will be the case in the future, then we know long-term actuarial calculations made today are an academic exercise that will have zero (or near zero) practical effect on the policies needed to deal with the cash flow crunch of the 2030s.

So why all the bother over them? They don't matter any more today than they did when Samuelson wrote.

That cash flow crisis is what should be the subject of the debate today. Democrats should by defending SS by saying: "Yes, we should raise income taxes 34% to finance the Trust Fund", while reformers should be answering "Really? while we're raising them another 30+% for Medicare?"

That should be the public debate -- but we're not getting a word of it.

Bruce Webb said...

What should be the public debate is the productivity assumptions of Intermediate Cost. Why exactly should we be accepting projections that call for 2.0% productivity growth THIS year? And for productivity growth to NEVER exceed 2.0% in the future. But that is exactly what Intermediate Cost calls for 2005 Report: Principal Economic Assumptions Where did you possibly get "Yes, we should raise income taxes 34% to finance the Trust Fund" when the current payroll gap is 1.92%. (2005 Report p. 16)?

I see a lot of outrage but little attention to the fact that tiny changes in productivity assumptions or minor changes in tax rates backfills the entire "gap". Payroll tax vs productivity: what would it take. And when it comes down to it the long range trends in this EPI chart are irreversible Social Security Trustees: Changes in projections over time

There is no cash flow crisis, and given ordinary economic growth there never will be a cash flow crisis. At least until you can explain away the productivity asumptions of Table V.B1.

Democrats defend SS by bringing numbers to the table. What do you got?

Bruce Webb said...

I followed your link. Your 34% lumped Medicare in with Social Security. Either you knew that and were deliberately distorting the debate over the specific question of Social Security solvency , or you didn't know that at all. Neither is very flattering to you.

And Medicare was both in the headline and the first paragraph of your [data]. There is a debate to be had on Medicare, and I am itching to engage. But lumping it in silently with Social Security is a troll trick. Particularly egregrious when posting on the site of a sympathizer of private accounts. Shame on you.

Andrew Samwick said...

I blogged about the productivity assumptions here. I came to the conclusion that we would need 3.5 percent real GDP growth, with 3.3 percent productivity growth, to plug the 75-year gap holding all other factors constant. That strikes me as too high for a productivity growth assumption for SSA. I am also skeptical that the Trustees projections factor in enough improvements in longevity over this period. See this post for some discussion.

On Medicare, I'm still vexed about how to restore solvency. So much of its long-term imbalances (and I count the federal contributions to Parts B & D in this concept) are driven by the assumed growth in health expenditures per capita. But I have no magic bullet there. So, as a starting point, my instinct would be to phase-in (possibly steep) increases in the age of initial eligibility from 65 to a level that we could afford. But there are a lot of problems with shifting 65 - 70 year olds from universal public insurance to existing private markets. Perhaps a compromise would be to start charging age-specific premiums to buy into Medicare, declining to zero in very old age, but sufficient to improve solvency even if phased in over time. See this post for some similar thoughts last December.

Bruce Webb said...

"I came to the conclusion that we would need 3.5 percent real GDP growth, with 3.3 percent productivity growth, to plug the 75-year gap holding all other factors constant. That strikes me as too high for a productivity growth assumption for SSA."

You are probably right, 3.3% strikes me as pretty high. But this ignores a couple things. One we are turning in 3.3% right as we speak. Two we have this from the President's 2006 Budget Analytical Perspectives p.191 "Trend productivity growth is assumed, conservatively, to be 2.6 percent per year. That pace is noticeably below the average since the business cycle peak in the first quarter of 2001 (4.2 percent per year). It is, however, close to the pace during 1996–2000 (2.5 percent) and not far from the average since the official productivity series began in 1947 (2.3 percent)." If 2.6% is good enough for the Budget, and is characterized as "conservative", why not use it as our neutral touchstone for evaluating the assumptions of Intermediate Cost? And Three, the Office of the Chief Actuary states flat out that you never need producitivity above 2.2% in any future year to ensure solvency What is the Low Cost Alternative: what does it mean What did he get wrong?

At a minimum we ought to use the President's 2.6% as an input, at least to generate an alternative projection. This stubborn insistance many have of using numbers and dates derived from an economic model for 2005 that is clearly inoperative (we will get a better number than 2.0% this year, we are probably already there) is not advancing the dialog. It is time to start using some real world numbers.