Arnold Kling  

Progressive Implications of Social Security Privatization

PRINT
Health Insurance Regulation... Housing Bubble?...

In this essay, I wonder why the Left and the Right do not reverse their positions on Social Security privatization.


However, there is one important difference between keeping Social Security as it is and switching to privatization. Under the current system, Social Security's liabilities will continue to be funded by payroll taxes. However, under privatization, the transitional debt would be repaid using -- guess what? General revenues! In other words, privatization is a vehicle for changing Social Security's medium-term funding mechanism from payroll taxes to income taxes. It is exactly what the Left presumably wants, and what the Right presumably opposes.

The Left complains about "diverting" payroll taxes into private accounts. But that means that other taxes, mostly income taxes, would be needed to cover current benefits. From the standpoint of the supporters of progressive taxation, that would seem to be an improvement. I honestly do not believe that either side has really thought this through.

UPDATE: Valued Econlog commenter Jim Glass has his own post on the Social Security Trust Fund. He cites a paper showing that the effect of the trust fund is to increase government's propensity so spend, and thereby reduce saving.

UPDATE 2: Andrew Samwick says that income taxes do not have to rise to pay for the transition. That is true, as those familiar with the zero-transition-cost argument will understand. But assuming that Congress suffers from cash-flow illusion, they may end up cutting spending or raising taxes in the near term, and then face a lower deficit than they would have otherwise in the long term.

For Discussion. Has there been an analysis undertaken of the distributional effects of a shift toward privatization?


Comments and Sharing


CATEGORIES: Social Security



TRACKBACKS (3 to date)
TrackBack URL: http://econlog.econlib.org/mt/mt-tb.cgi/167
The author at Vox Baby in a related article titled Progressivity in Social Security Reform writes:
    Actually, this need not be the case. For example, suppose the reform turned out to be exactly Commission Model 2 with all near-term cash flow deficits covered by borrowing. All of the debt issued to cover these deficits is repaid out of reduced futur... [Tracked on December 10, 2004 6:54 PM]
COMMENTS (12 to date)
Boonton writes:

From Arnold's Article:

In the 1930's, relatively few people lived significantly past the retirement age of 65. In those days, it would have been foolhardy to save enough to last until you were 80. But if everyone contributed to a collective pool, then we could insure that the few who lived long past retirement would not be destitute.

Since the 1930's, longevity has increased by more than a decade. However, the Social Security retirement age has been raised only a few years. As a result, Social Security no longer represents insurance for the unusually long-lived. It is now an "entitlement" for everyone.

We already have privatized SSI accounts, they are called 401K's, IRA's, Roth-IRA's and so on. Imagine auto insurance in an environment where more cars get stolen and into accidents. No one would say that auto insurance companies ripped us of by not investing the premiums in higher yielding investment. It would be quite straight forward, premiums were higher because more bad things were happening to cars.

SSI is an insurance against a good thing (living old) that has a bad thing coupled with it (being too old to work and not having the savings necessary). We can afford this luxury because we are prosperous society. When you hear that the ratio to workers to retired is going from 16 down to 2 what you are really hearing is that we are getting so rich that we can afford having a lot of people not working.

This multiple nature of SSI is just one reason why it is silly to debate returns. SSI is part insurance, part welfare for those who need it most, part redistribution to those with lower incomes and part return. (Another reason it is silly is the assumption that the historical returns we have seen for the past 50 years would have remained exactly the same if we instituted some type of private account system in, say, 1950).

The solution is quite simple, modestly return SSI to where it was. Encourage working among older people who still can & index the retirement age to lifespan. Do it properly you can hold the line on tax increases and perhapd even cut the payroll tax down a bit over time.

Steve writes:
The Left complains about "diverting" payroll taxes into private accounts. But that means that other taxes, mostly income taxes, would be needed to cover current benefits.
The Bush administration has discovered another option - cover current benefits by taking on more debt. This is clearly what would happen if social security privitization were implemented by the current administration. Of course, taking on more debt now just means that you need more taxes later to pay it off. Or that you need to pay out less in benefits. Which is of course the unstated plan of the Republican party.

Cutting benefits is not politially possible now. But it will be much more politically palatable if the public is led to believe that the system will be bankrupt by the time they retire anyway (which of course it would be thanks to creating individual accounts without a tax increase).

Boonton writes:

Benefit cuts were done once before in real time when SSI benefits were 'frozen' against cost of living increases. It can easily be done again because the cuts can be structured so they will not happen far into the future.

1. Phase in retirement age increases so people have a long time to prepare. You can market such a campaign with something like "increase your 401K contribution by $25 a month and you'll break even!".

2. Create an incentive to keep people who can work working past 65. For example, why not refund 80% of SSI taxes for post 65 yr old employees to both workers and employers? Even though the SSI fund will receive only a small amount of additional revenue the general tax revenue created by added workers should be positive. Plus the economy will be stronger in general from added income generators.

3. Remove the indexing of SSI benefits to average wages and index to just price instead.

If you shrink SSI benefits gradually you can reach a point where you return the program to more of a income guarantee/subsidy rather than a full fledged retirement program. On the flip side, you have 401K's/IRA for individuals to save for 'extravagent' retirements.

Lawrance George Lux writes:

Alternate Plan to forestall the great conflict between Left and Right. Needs for the Plan: one-rate monthly SS benefit, coupled with a yearly limitation on Medicare and Medicaid benefits.

Plan features:
1) introduce Flat tax of 17.5%
2) integrate FICA taxes into Income taxes by:
a) setting the first set Dollar amount to go to the Social Security Fund.
b) This set Dollar amount should be multplied by the number of Workers for Enterprise Income, then placed in the SS Fund.
3) The Dollar amounts set to be sent to the SS Fund should be of such largesse as to equal about forty percent of the Revenues generated.

Crunching numbers could bring up the likelihood that the SS Fund shortfall has disappeared, the Fund is secure, the Tax base is secure, and Business gains from ease of taxation ensuring potential growth, and who needs Private Accounts.
No one will complain about contributions to the SS Fund, because they would be paying the Tax anyway, especially as the Flat Tax will allow no Deductions, Exemptions, tax credits, tax exclusions, or investment credits.

Now if We could only get Bush to quit spending. lgl

Dave Altig writes:

Arnold --

I'm not convinced of your premise. Here's what I have to say over at macroblog. http://macroblog.typepad.com/macroblog/

What I can't buy is the statement "Under the current system, Social Security's liabilities will continue to be funded by payroll taxes." The beginning of the end for that property of the current system began the minute we invented the idea of the "trust fund." As I explained in a previous post:

The "Trust Fund," of course, is nothing but IOUs that the government writes to itself. It, in fact, does not represent assets accumulated by the government, but rather the "promise" that we will spend the surplus payroll taxes today, and pay for future benefits out of general federal revenues (i.e. the income tax).

Privatization or not, we will be paying social security benefits out of general revenues just as soon as outgo exceeds payroll tax collections sometime in the next decade.

Boonton writes:
The "Trust Fund," of course, is nothing but IOUs that the government writes to itself. It, in fact, does not represent assets accumulated by the government, but rather the "promise" that we will spend the surplus payroll taxes today, and pay for future benefits out of general federal revenues (i.e. the income tax).

I'm curious as to the hostility towards any effort made by SSI to save in light of the coming generational buldge. Gore's 'lockbox' was mocked and now the surplus is endlessly attacked even though it was thought up by Alan Greenspan.

As a vehicle for savings the trust fund surplus pays down debt held by the public today. This makes borrowing in the future easier as well as making tomorrow's economy stronger. The trust fund is also an accounting device useful for segregating funds between SSI and general revenue.

There is nothing magical about private accounts paid for with taxpayer debt nor is there any way to get around trusting the intergenerational 'promise'. SSI benefits can be abolished by an act of congress. Guess what, so can the tax favored status of 401K accounts. Just as baby boomers will put pressure on SSI benefits, there will be political pressure over 'tax free Roth IRA millionaires' while regular workers will be paying income and payroll taxes.

Whether income taxes are raised or not, the redistributive effect you write of occurs. It has to if we switch from a regressive payroll tax to a progressive income tax for meeting the need for revenue.

"As a vehicle for savings the trust fund surplus pays down debt held by the public today."

Doesn't seem to be having that effect.

Does your spending cash at a restaurant "pay down" your Visa card balance?

Edge writes:

If the purchase of treasury bonds by the Social Security trust fund does in fact, lead the government to spend any surplusses, a simple fix is to give the trustees independence, and allow the trust to make other investments, at least in the form of marketable securities, rather than the special US treasury securities which some people believe have poltical risks.

If the goal of privatization is to increase returns, so that we don't need to increase taxes or cut benefits, the same goal could be achieved by giving the trust the same sort of flexibility as other large conservative pension funds, setting standards to be however conservative or risk-tolerant as deemed appropriate.

But let's examine the premise that Congress spends any surplus generated in the trust funds by examining historical data.

We find two periods when trust fund assets were increasing as a fraction of GDP: 1940 through 1947, and 1984 through today and beyond.

Presumably, though congress spent those trust fund surplusses and six times more during WW II, we can discount any causality of the proposition that congress squandered the surplusses to fund government spending gone wild. The national debt did more than double as a fraction of GDP during that time, but it won't go down in history as the second worst abuse of the Social Security trust fund.

Then there is the period from 1984 through today and beyond, when the Social Security trust has built a significant reserve.

From 1981 through 1993, the federal debt doubled as a fraction of GDP. In 1984, Social Security went from a pay as you go system to a partially funded system. Did the reform of Social Security lead to that explosion of debt, and if so, how?

From 1994 through 2001, the gross federal debt shrank from 67% of GDP to 57% of GDP (gross debt INCLUDES the trust fund liabilities), while the trust fund surplus grew from 15% of GDP to 25% of GDP. That's a significant bit of contrary data to contend with.

From 2002 through FY 2005, the federal debt grew, from 57% of GDP, to 67% of GDP, while the trust contined to plod along in its steady growth.

Well, OK, here in the most recent years, we have the first example where the presence of a Social Security surplus clearly contributed to an increase of the gross federal debt. Some people would say that it was a good thing for the federal government to at least be neutral during the most recent timeframe, rather than paying down the federal debt. For example, our own federal reserve chairman, who also headed the commission which proposed that Social Security should become a partially funded system invested in government securities, the guy who's normally supposed to take the punch bowl AWAY from the party. And that was before we knew there was a recession, before 9-11, and before Bernanke openly speculated about unconventional means of fighting deflation.

But doesn't this latest example, where a trickle of fiscal easing might have been reasonable, but a flood has become profligate, suggest that the current batch of politicians can't be trusted to touch the Social Security trust, given that they've gone from a balanced "on budget" position to one in which we're only bringing in sixty-eight cents for every dollar spent in just four years?

Why should we trust these politicians to manage a a problem that won't manifest for 40 years when they can't come close to figuring out a short term budget?

The basic problem with Social Security is that we aren't saving enough. Borrowing another 20% of GDP now in the hope that the gamble will pay off 75 years from now isn't a prudent move.

In answer to your original question, progressives consider the SS trust to hold risk free bonds, and consider the $Trillions of surplusses paid in to those bonds to be a moral obligation for the treasury to repay.

If the treasury issues junk debt to finance a privatization scheme largely crafted to phase out Social Security over time, progressives have every reason to believe that the government might renign on current obligations, rather than raising revenues from the general fund to cover those obligations.

To progressives, the idea of borrowing a whole bunch of money, using the SS trust as a sort of collateral, puts the assets in the trust at risk, and subjects current retirees and near-retirees to probable benefit cuts.

George Bush seems to acknowledge the political importance of maintaining promised benefits to current retirees and near-retirees, but the plan proposed would accelerate any crisis sufficiently that it is quite probable current beneficiaries would see benefit cuts.

I suspect that the only way to get ANY progressives to embrace market-based reforms for social security are either to go with something like Kotlikoff's plan (which increases national savings), or set up an independent SS trustee able to invest in a broader range of assets.

Boonton writes:
Well, OK, here in the most recent years, we have the first example where the presence of a Social Security surplus clearly contributed to an increase of the gross federal debt. Some people would say that it was a good thing for the federal government to at least be neutral during the most recent timeframe, rather than paying down the federal debt.

Your questions are good but the element that makes this story a lot more complicated is what caused what? A person inclined towards my point of view would argue that gov't debt would have increased more without the social security surplus. A person inclined towards Jim Glass's POV would argue that when the gov't was paying down the debt it would have paid off even more if the surplus wasn't there.

Empirically I did some regressions a while back and posted the results here. I couldn't find a positive relationship between tax revenue (including Social Security) as a % of GDP and spending as a % of GDP. I did find some evidence for a slightly negative relationship, as revenue goes up spending actually goes down.

Theoritical reasons for this are:

1. Much of the Federal Budget is on auto-pilot. Many spending programs are directly linked not to how much Congress wants to spend but to the underlying economy. Examples include things like unemployment benefits, welfare, Medicaid and even things like Social Security benefits (in a booming economy, some people will delay retirement thereby decreasing benefit payouts...in a sour economy more people will opt to retire earlier).

2. Federal taxes are largely on auto-pilot as well. Rates are often tinkled with at the margin but the bulk of tax revenue depends on the economy. A strong economy brings in tax dollars, a weak economy brings in less.

If the goal of privatization is to increase returns, so that we don't need to increase taxes or cut benefits, the same goal could be achieved by giving the trust the same sort of flexibility as other large conservative pension funds, setting standards to be however conservative or risk-tolerant as deemed appropriate.

Here's a serious problem with the line that privatization advocates have been giving us. Returns are not magically increased. In fact, market based economics tells us that returns are all the same! What do I mean?

Take CAPM (the capital asset pricing model). Basically every return is exactly the same adjusted for risk. If some asset had a return that was different rational investors would buy or short that asset until its return falls in line with all the other returns. What this and other market oriented models tell us is that there are no 'better returns' out there.

If stocks historically have earned more than bonds (adjusted for risk) then that is a fluke which it would be foolish to depend upon into the future. If, going forward, stocks continue to have better returns than bonds (adjusted for risk again) then we have a major problem with economic theory.

Charles Nicholson writes:

The most fascinating non-discussion regarding Social Security is the non-existance of the so-called Social Security Trust Fund. In reality, what is touted as a Cash Asset has in fact been invested in Government Securities since Lyndon Johnson's time. I.E. the Fund is mortgaged to the hilt, so much so that in real accounting terms it is a Zero asset. As Greg Brown sings so poignantly, "All the Money's Gone."

Comments for this entry have been closed
Return to top