We use cookies on our website to give you the best online experience. Please know that if you continue to browse on our site, you agree to this use. You can always block or disable cookies using your browser settings. To find out more, please review our privacy policy.


Policy Hub: Macroblog provides concise commentary and analysis on economic topics including monetary policy, macroeconomic developments, inflation, labor economics, and financial issues for a broad audience.

Authors for Policy Hub: Macroblog are Dave Altig, John Robertson, and other Atlanta Fed economists and researchers.

Comment Standards:
Comments are moderated and will not appear until the moderator has approved them.

Please submit appropriate comments. Inappropriate comments include content that is abusive, harassing, or threatening; obscene, vulgar, or profane; an attack of a personal nature; or overtly political.

In addition, no off-topic remarks or spam is permitted.

April 12, 2005

Inequality and Social Security

A nice article on the topic from Greg Ip appears in yesterday's Wall Street Journal (page A2 in the print edition).

The debate over Social Security has managed to drown out other longstanding issues in American society, including the widening gap between rich and poor and surging health-care costs. Yet these two phenomena play an important, though little appreciated, role in Social Security's problems. That is because they are eroding the base of taxable wages available to support Social Security benefits.

The reason is that taxable payroll is expected to expand more slowly than is GDP and, by 2080, to equal just 33% of GDP, compared with 38% now. Stephen Goss, Social Security's chief actuary, says there are two main reasons why. One is that a "somewhat increasing share of all the earnings in the economy [is] above our taxable maximum," and the second is that a growing share of "employee compensation...is going not to wages but to fringe benefits, which are not included in our tax base," Mr. Goss says.

Social Security payroll taxes are levied on wages up to a certain cap, currently $90,000 a year, which rises annually with the average wage. In the past 25 years, a growing share of income has been paid to people who earn more than the cap.

This increasing concentration of income at the upper strata of society is an important reason why, from 1980 through 2000, taxable payroll fell to 83% of wages of contributing workers from 90%...

Even if inequality stopped widening, Social Security's tax base probably would continue to be eroded because of rising health-care costs. Since 1996, health-care costs have risen to 7.3% of employee compensation from 6.3%, and Social Security's actuaries expect it to keep rising. This is a big problem for Medicare, the federal health program for the elderly, but it also affects Social Security, because payroll taxes aren't levied on health-care insurance premiums. Mr. Goss says that is the main reason for taxable payroll's shrinking share of GDP.

Wholesale prices for popular brand-name prescription drugs rose an average 7.1% in 2004, more than twice the general inflation rate, a new study commissioned by the nation's largest seniors lobby says.

The increase is the biggest in the five years that AARP, with 35 million members, has sponsored the study. It's just slightly higher than the 7% price rise in 2003.

... although the results are, apparently, under some dispute:

Ken Johnson of Pharmaceutical Research and Manufacturers of America, the trade group of brand-name drugmakers, called the study "exaggerated and misleading." He said the use of wholesale prices excludes factors such as rebates that could cut retail costs.

"Price data clearly shows prescription-drug prices have increased about 4% a year," Johnson said. He called that in line with growth in other health costs.

That observation, of course, doesn't diminish Ip's point.

As if to underscore the point, there was this report from this morning's USA Today...

March 26, 2005

Barro on Social Security

With an assist from Tyler Cowen, pgl at Angry Bear notes a new Business Week article by Robert Barro, wherein Barro repudiates his previous support for introducing private accounts into the Social Security system.  pgl likes Barro's piece because he emphasizes again the fact that the differential between current internal rates of return from the Social Security program and returns on capital is not prima facie evidence that private accounts dominate.  And he is right -- Barro is cogent, as usual, on this point.  Although they arrive at a similar spot, however, I don't think the boys at Angry Bear entirely endorse the reason Barro no longer favors the private account option. 

Contributions that fund just the minimum cannot go into a meaningful personal account. People would opt for too much risk, knowing they would be bailed out if they fell short. Also, contributions that cover the minimum provide no individual
return and, therefore, amount to a tax that discourages work.

Personal accounts have to supplement the minimum payout. But then why have a public program at all, rather than relying on individual choices on saving? I think there is no good reason to go beyond the minimum standard; that is why I view personal accounts as a mistake -- they enlarge a Social Security program that already promises too much.

What's the minimum payout?  Barro provides an example.

Knowing this, some people will save too little and rely on public support when old. Thus, there is reason to require workers to save for retirement. How much depends on what is viewed as a minimal standard of living; suppose it is $1,000 per person per month. (Currently, a person with the median earnings history gets $1,200 from Social Security.)

Barro goes on to propose some of usual fix-ups...

...baseline Social Security benefits should be indexed to prices. The practice of indexing initial benefits to past wages should be eliminated. Moreover, the price index should be an accurate one, such as the Bureau of Labor Statistics' chain-weighted consumer price index, rather than the standard flawed index. In addition, the normal retirement age should effectively be indexed for changes in life expectancy, going beyond the current plan to raise the age to 67. These three adjustments are all-important.

... and oppose others:

A mistake even greater than personal accounts would be, in addition, to raise the maximum earnings taxed by Social Security above the current $90,000, a proposal that President Bush seems to welcome. This change increases marginal tax rates by about 10 percentage points on a productive group that already faces high marginal rates.

As I noted in a previous post, this last method is the only one that appears to be politically popular (if you believe the polls, anyway.)  This is where I find the puzzle.  It seems to me that Barro's position leads to to the implication that Social Security be means-tested to the extreme.  Figure out what the "minimum payout" should be, and provide it only to  those who, for whatever reason, find themselves with retirement income flows that fall below that level.  In other words, make it look like the welfare system that Barro seems to think it should be.  (That may mean some phase-out of benefit payments so that there are not large jumps in effective marginal tax rates for those just above the minimum payout level.  But this would presumably still look very much different than the system we have today.)

I'm only guessing, but my presumption is that Barro did not spell it out this way because he believes that such an extreme position is politically infeasible.   But if that is so, then it seems quite likely that squeezing the program down to the minimum payout is infeasible as well.  And if that is the case, Barro's reasons for previously preferring private accounts come back into play.

UPDATE: Arnold Kling notices Barro's fiendish plot too.

March 20, 2005

Spinning Polls

As I noted in my previous post, today's New York Times kicks off an article on social security with the observation that "Judging from the polls, President Bush's plan to transform Social Security... isn't going very well."  I know that this is revisiting some old news, but the Times lead hearkens back to this CNNMoney headline from a few weeks back:

Poll: Bush slips on Social Security
Number of those who say retirement plan needs fixing in the next two years falls to 38% from 49%.

And there is this, from the Gallup Organization:

Majority Disapproval for Bush on Social Security

That latter headline is based on this information:

George W. Bush's ratings for handling Social Security have dropped in recent weeks as debate on the issue has intensified, with a majority now disapproving. Additionally, Americans today perceive less urgency for Social Security reform: fewer say major changes are needed to the system within the next two years than did so in January.

Note that neither of these results has much to do whether the public generally supports some form of reform involving private accounts.  The first question pertains only to whether people think the president is doing a good sales job. (Surprise! They don't.) The second question refers only to the timing of the reform.

If you've heard this before I apologize, but let me suggest a few alternative headlines, based on a larger universe of poll results.  These come from a Gallup report dated Thursday. (You will have to subscribe to gain access, but it's free for a 30 day trial.)  To avoid confusion, I'll put the made-up headlines in red.

Majority Approve of Private Accounts

That would seem justified by these results:   

Would you favor or oppose a plan in which people who choose to could invest some of their Social Security contributions in the stock market? (ABC News/Washington Post poll, March 10-13, 2005)

Favor        Oppose        No Opinion
    56%           41                  3

A question from Gallup's March 7-10 poll used the following language, and received a similarly positive response.

Suppose Congress and the president do pass legislation to make changes to the Social Security system this year. Do you think that legislation should -- or should not -- include a provision that would allow people who retire in future decades to invest some of their Social Security taxes in the stock market and bonds?

   Yes, Should      No, Should Not   No Opinion
        58%                   37                        5

Another question, from a Feb. 16-21 Pew Research Poll, using language along these same lines, yielded a solidly positive balance of opinion, although it also found a higher number of people expressing no opinion (probably because the question was preceded by one asking how much they had heard or read about the proposal).

[Generally, do you favor or oppose] a proposal which would allow younger workers to invest a portion of their Social Security taxes in private retirement accounts, which might include stocks or mutual funds? (Pew Research Poll, Feb. 16-21, 2005)

  Favor        Oppose        No Opinion
46%             38                  16

Or how about this...

Majority Say Social Security Should Be Fixed Now

... supported by this data:

Do you think it is -- or is not -- necessary for Congress and the president to pass legislation this year to make changes to the Social Security system?

      Yes, is      No, is not        No Opinion
51%             46                       3

OK, full disclosure time.  When asked if there should be a trade-off between allowing diversion of funds into private accounts and future benefit receipts, poll respondents say "no" by a large margin, suggesting this headline:

When Asked, Majority Say They Want Free Lunch

As to the difference between the positive response to whether there ought to be reform this year and the oft-reported result noted in the CNNMoney headline above:  The latter question makes reference to "major reform" and the former does not.  Let me suggest, then, this headline:

Majority Get Nervous When You Ask Them A Question That Suggests Life As They Know It Is About To Come To An End

Wait. There's more interesting things in the Gallup report.  This is based on Feb. 4-6 polling:

Assuming there would be no change in Social Security benefits for those who are now age 55 or older, do you think each of the following would be a good idea or a bad idea to address concerns with the Social Security system? How about -- [RANDOM ORDER]?         

                                                                   Good idea    Bad idea    No Opinion

Limiting benefits for wealthy retirees                   68%            29                3

Requiring higher income workers to pay
Social Security taxes on ALL of their wages           67             30                3

Further reducing the total amount of benefits
a person would receive if they retire early            40             57                 3

Increasing Social Security taxes for all workers     37             60                3

Increasing the age at which people are
eligible to receive full benefits                                 35             63                2

Reducing retirement benefits for people
who are currently under age 55                               29             67               4

I'm kind of enjoying my tryout, so how about this one out for that last list of responses: 

Majority Say Tax That Guy Behind The Tree

Here's one final set of results from Gallup.

Which is riskier for average American workers today: investing some of their Social Security taxes in stocks and bonds, or relying on the Social Security system to pay them the current level of benefits when they retire?      

Investing in        Relying on          No Opinion
stocks and          the system
bonds is             to pay current
riskier                level of benefits
                          is riskier

  46%                        50                            4

I've hit on this objection of treating benefits under the current system as if they are riskless before. If I believed for a second that this poll stuff was informative, I'd say I told you so.

March 20, 2005

The Volatility Problem

The New York Times picks up a theme explored in a series of LA Times articles last year.

After mining data from the Panel Study of Income and Dynamics, a database produced by the University of Michigan that tracks the incomes of the same families over a 40-year period, scholars have concluded that incomes are much less stable - i.e., much more volatile - today than they have been in the past. "There has unequivocally been general upward-trend income volatility since at least 1975," said Bruce A. Moffitt, the Krieger-Eisenhower professor of economics at Johns Hopkins University, who, with Professor Gottschalk, wrote one of the first papers on income volatility in the 1990's. "It accelerated in the 1980's, turned down in the early 1990's, and then accelerated into the end of the 1990's."

According to a measure of volatility constructed by Jacob S. Hacker, a Yale political scientist, which tracks the five-year moving average of family incomes, income volatility rose 88 percent between 1978 and 2000...

"The problem in the past few decades," Professor Moffitt said, "is that volatility has risen while real incomes haven't risen." What's more, income volatility has grown significantly for those who can afford it least. A series of articles last year in The Los Angeles Times, written by Peter G. Gosselin, who worked closely with Professor Moffitt and other scholars, reported that in the 1970's, income for middle-class Americans tended to fluctuate by 16 percent a year. But in the 1980's and 1990's, middle-class incomes fluctuated an average of 30 percent. For those whose earnings placed them in the bottom fifth, income volatility rose from 25 percent in the early 1970's to 50 percent in recent years.

Part of the solution?  Maintain the existing defined-benefit social security system, of course.

As we learn more about income volatility in the information age, some scholars say, Social Security - an insurance program designed for the industrial age - may be even more essential...

"Social Security provides a vital kind of insurance," Professor Hacker said. "The real issue lurking behind this debate is whether we should have a program that provides the bedrock protection against economic risk."

I'm not going to dispute any of this, but it as at least worth pointing out that volatility is not the same thing as risk.  That is obvious in the case where increased ups and downs in income are perfectly predictable, but that isn't really what I have in mind.  What I have in mind is the perfectly straightforward observation that volatility of returns in financial assets can actually reduce the risk associated with fluctuating labor income by providing a means of diversifying total income.

It is true that an appeal to income diversification is not all that helpful to those of us who favor a private account option for social security reform.  Steve Davis and Paul Willen have a very nice, and accessible, paper on the issue, and find that broad equity returns do not appear to be highly correlated with occupation-level income shocks.  Furthermore, what benefits there might be would likely accrue to higher-income individuals.

Still, not a lot of research has been done on this issue, and I would ask this question: If muting the impact of labor-income volatility is a motivating factor for supporting the social security system, is a program intrinsically tied to both aggregate and individual labor-income fortunes the sensible way to go?

There is also this, from the Times article:

Because of other longstanding trends in the economy, strong income volatility can wreak greater havoc now than it did in the past...

Why? Many families already rely on two incomes. What's more, fixed commitments have risen as a percentage of total income.

Again, maybe so.  But can't we be just a bit skeptical?  Working more because we consume more is a choice.  I make that choice all the time.  So do you, I bet.  What's the problem?  (The same would go for endogenous income volatility associated with the choice to move in and out of the labor force, change employment circumstances as a life-style choice, and so on.)

Here's a simple smell test I can't help but employ.  Any time a learned study seems to conclude that economic welfare has declined relative to the 1970s, it's time to start asking digging deeper.  The facts may be right.  What they mean is much trickier proposition.

A couple of other random points about the Times piece.

-- I meant to mention this before, but I was slightly amused that one of the poster-boys for the LA Times first article on the problem of income volatility was one Paul Fredo, a financial analyst, whose income fluctuated a lot, between $200,000 and $120,000.  It looks to me that it is people in this income class that are likely to bear most of the burden of closing the social security financing gap.  (Think removing the ceiling on income subject to payroll taxes, reducing benefit indexation for "high-income" folks, and so on.)  So much for the helping hand of government in his case.

-- The "Bruce A. Moffitt" referenced in the NYT article is actually Robert A. Moffitt.

-- The NYT article leads with this sentence:

Judging by the polls, President Bush's plan to transform Social Security from an insurance program that guarantees a minimum income into something more closely resembling a 401(k) investment program isn't going very well.

That is a highly selective reading of the polls.  More on that to follow.