Home Mail Articles Stats/current Supplements Subscriptions Links
The following article appeared in Left Business Observer #87, December 1998. It retains its copyright and may not be reprinted or redistributed in any form - print, electronic, facsimile, anything - without the permission of LBO.
The annual report of the Trustees
of the Social Security System, the source of the projections quoted
below, is here.
The worst thing about the Lewinsky affair is that it's interfered with the proper hatred of Bill Clinton. He's been indicted by the House for all the wrong crimes -- not for bringing about the end of welfare, or promoting mass incarceration, or killing Iraqis with sanctions and ordnance, or his terrible taste in music. But his enemies are so repellent that you don't want them to win anything.
But one of the best things about the Lewinsky affair is that it's keeping Clinton and Congress from doing what they'd like to do -- privatize Social Security. Elite debate has now come down just to the details, and the public seems to accept this as inevitable. A sizeable minority, charmed by the stock market's extraordinary performance over the last few years, even looks forward to the prospect, convinced they've found the key to wealth. LBO hasn't scrutinized the privatizers' arguments in four years; it's time for another look.
Folks on all sides of the debate accept the basic premise, that Social Security faces certain insolvency when the baby boomers retire. Compelling "facts" are trotted out to prove the point: the system will start running a deficit in 2013, and will have spent down all its reserves by 2032. There will be just 1.8 workers for every retiree in 2075, compared with 3.4 today. We could all do better by buying stock in Amazon.com than contracting with Social Security anyway.
Taking these specious points in reverse order: there's no need to say anything about the stock market approach other than to point to the article on p. 5 [in the print edition; not on this website]. The dependency argument needs a bit of attention though. Yes, it's a fact that the population will age, and the number of retirees per worker will rise. Pointing just to those numbers, though, is a selective rendition of history, since the history of capitalism has been to bring an ever-larger share of the population into working for pay. In 1900, there were almost four nonworkers for every paid worker; as almost everyone left the farm and as most women came to draw paychecks, that number fell steadily to just over one today. As the chart shows, the boomer retirement will raise this number a bit, but not by much in the scheme of things. And these, it should be emphasized, are fairly conservative projections. In 2050 there should be a larger share of the population working for pay than in 1950, when mom was at home and the earliest boomers were in kindergarten.
But the question isn't how many retirees we'll have -- it's can the U.S. economy afford them? On official projections, the answer is famously no, but those projections are oddly gloomy. The Trustees of the Social Security System project economic growth over the next 75 years will be less than half that of the last 75 years. LBO is certainly no fan of growth for its own sake, but were the U.S. economy to grow for decades at near-depression rates, then we're going to face lots of social and fiscal strains. For an economy dependent on high consumer and foreign borrowing, a sharp growth slowdown would be disastrous for us, and probably disastrous for the rest of the world too, which has come to rely on the U.S. as the world's final source of demand. So if our rulers expect that, they should tell us some other way than in obscure tables in government reports.
And if they don't expect that, then why are they projecting it? The official answer is that the ceaseless growth in the working share of the population -- driven in recent decades by women's entry into paid labor -- will finally cease, and labor force growth will slow. On top of that, productivity growth will continue to be weak. Combine the two trends and you have a snail's-pace economy -- a projected average yearly GDP growth rate of 1.4% for the next 75 years, compared with 3.3% for the last 75. The implied per capita GDP growth rate of 1.0% (1.4% GDP growth less 0.4% population growth) is well below Western Europe's 1.7% rate for this decade, a level that is regarded by the American opinion-making class as the stuff of crisis.
Those assumptions are at odds with the evidence. Countries with low labor force growth, like Western Europe, typically see higher productivity growth than those with rapid labor force growth, like the U.S. (You could argue that U.S. capital has long relied on cheap labor and cheap raw materials, allowing it to stint on capital investment.) It's unlikely that both productivity and labor force growth rates will limp along together.
But at every turn there's a bearish assumption in the Trustees' numbers. At 0.4% a year, the projected population growth through 2075 represents quite a deceleration from the 1.2% average of the last 75, and well under the Census Bureau's 0.7% projection for the next 50 years. The workforce is slated to grow even less -- by just 0.2%. Not only is the youthful share of the population expected to decline, the Trustees project that fewer of them will be working: the share of the population aged 20-64 at work (the employment-population ratio) is projected to decline, a violation of all historical precedent.
The Trustees' growth projections have been trending steadily downward since the early 1980s, so much so that you'd almost think there was an intention behind the trajectory (though the system's actuaries deny any political pressure to emit bearish forecasts to grease the privatization agenda). As is typically the case with official projections, there are three scenarios -- a gloomy one, an optimistic one, and an official, moderate one. In 1981, the Trustees projected a long-term growth rate of 3.1% in their middle scenario and 2.1% in their gloomy one. In 1986, the numbers were 2.5% and 1.4%. And this year, they're 1.4% and 0.6%. The Trustees' optimistic prediction for 1998 -- 2.1% -- matches their most bearish forecast from 1981. Aren't lowered expectations a banished relic of the Carter years?
Rerun the projections with more reasonable -- though still conservative -- projections and the "crisis" largely or fully disappears. If the employment-population ratio for those aged 20-64 remains constant, a third of the projected shortfall for 2020 disappears; if it rises, because the share of women employed approaches that of men, then two-thirds of the projected deficit disappears. As the nearby chart shows, if the economy grows at a modest 2.5% rate, red ink will turn to black. And even if the official bearish projections turn out to be true, the shortfall could be made up easily by subjecting investment income to Social Security taxes, and by eliminating the cap that exempts wage income above a certain maximum ($68,400 in 1998). The reason for sparing such income from Social Security tax is that the program is supposed to be financed solely by labor, with no contribution from capital, capital already being so burdened. The "crisis" of Social Security is a political one, not an economic one.
But let's take the 1.4% growth projection and apply it to the stock market, that cornucopia of wealth that's supposed to replace the public pension system. Privatizers typically assume an average real stock market return of 7% a year, from the combined increase in prices and the cash dividends. How the stock market is supposed to grow five times as fast as the economy is a mystery that's rarely noted, much less investigated. To achieve those stock returns, the ratio of stock prices to underlying corporate profits -- price/earnings (P/E) ratios, a fundamental measure of whether stocks are "expensive" or "cheap" which goes out of fashion at enthusiastic times like these -- would have to rise to astronomic levels.
The stock of all U.S. corporations is worth roughly 26 times their collective profits -- the highest since 1945, and twice the historical average, meaning that stocks are very expensive by stodgy old measures. But today's P/Es would be nothing compared to tomorrow's, if stocks truly were to return that 7% for the next 77 years (4% in price appreciation and 3% in cash dividends) while the economy was growing at 1.4%. If profits grew in line with GDP while the stock market were growing at five times that pace, P/E ratios would rise to 33 by 2010, 71 by 2040, and 178 by 2075. But wait! Dividends can't grow any faster than the economy, can they? Constraining those to the 1.4% assumption too means that stock prices will have to rise even faster to compensate for the lower growth in dividends over time. That would give us a P/E of 138 by 2040 and 764 by 2075. When you work for the Cato Institute, you're never called upon to defend your preposterous reasoning.
Privatizers often make financial arguments against Social Security, presenting absurdly precise dollar estimates of just how bad a deal it is. Typically, a worker's contributions today are measured against the projected benefits in retirement, and the conclusion is that the stock market would be a far better deal. But that's a poor way to measure a public pension system's value. Social Security represents a deal across generations, with today's workers financing today's retirees on the assumption that the same deal will be offered when they retire. It offers not only an approximation of a decent retirement income -- Social Security keeps 42% of the elderly population above the poverty line -- it also includes disability benefits and coverage for survivors on the insured's death. Its formulas favor the poor over the rich, and compensate women some for their lower and more volatile lifetime incomes. A privatized scheme is almost certain not to match the disability insurance, and is certain to amplify, not offset, distinctions between rich and poor, and men and women. It's a program based on extremely unfashionable notions of solidarity and universality. Years ago, only right-wing ideologues wanted to privatize Social Security; now it's the entire American establishment.
Leading the charge is Wall Street, which would make a fortune out of privatization; no wonder financiers have been discreetly showering money on the privatization campaign. A few quick numbers will explain Wall Street's enthusiasm. Chile's privatized pension system, the enthusiast's favorite model, devotes about 30% of revenues to administrative costs, which means everything from paperwork to brokers' fees to marketing expenses. The U.S. life insurance industry is a bit more efficient, devoting about 10% of premium income to administration, which includes everything from paperwork to profits. Social Security's overhead is under 1%. About $430 billion flowed into the system's coffers this year; 10% of that would be a very pleasing $43 billion, and 30% would yield $130 billion, a windfall even by Wall Street's standards. Higher fees, lower benefits, greater gender inequity, and more risk -- no wonder privatization has to be sold with a cooked-up scare campaign.
Back to Clinton and his scandal. Max Sawicky of the Economic Policy Institute says that Clinton -- whom Sawicky describes as "the product of crossbreeding a polecat and a jellyfish" -- seems far more willing to appease the privatizers than he's been in the past, in an effort to keep his job. The AARP is absent on the matter, and forces to defend the system are thin on the ground. The right wants full privatization with everyone having an individual account, free to invest as he or she chooses (and if you want disability insurance, buy it yourself!); the "liberal" position is for partial privatization with the government investing a portion of the Social Security reserve in the stock market. The AFL-CIO has no official line, but informally supports the "liberal" option. Right now, the thing most likely to save Social Security in its present form is a bear market, if we ever have another one of those.
Home Mail Articles Stats/current Supplements Subscriptions Links