The Economics of Social Security

Economists view social security differently than does the public at large. Many journalists and politicians speak of social security as if it were a defined-contribution pension plan. They speak as if the benefits that someone receives from social security reflect that person's contributions into the program.

Economists instead view social security as an ongoing intergenerational transfer mechanism. At any point in time, the working-age population is being taxed to support retirees. This view of social security as an intergenerational transfer has some interesting corollaries.

  1. Holding tax rates constant and longevity constant, the ability to expand benefits in social security depends on the growth rate of the population plus the growth rate of productivity. If the younger generation is 10 percent larger than the older generation and 10 percent more productive, then benefits for the older generation can be 20 percent higher.

  2. In general, a transfer from the young to the old is a transfer from people who are relatively rich to people who are relatively poor. That is because of the phenomenon of economic growth, which makes each generation better off than the previous generation.

  3. Some of the involuntary, impersonal transfers that take place within social security are reversed by voluntary personal transfers. Within familes, the old tend to give money to the young.

Policies and the Social Security Crisis

The looming decline in the ratio of workers to beneficiaries is sometimes referred to as a crisis for Social Security. Economists do not agree on how to deal with this crisis.

Some economists are not convinced that the crisis will appear. This is not a matter of ideology, but of optimism. There are both liberals and conservatives who believe that productivity growth in the next twenty years may be two percent or more per year. If this proves to be the case, then social security transfers as a percent of GDP will increase little, if at all.

On the other hand, there are economists who are concerned about the demographic outlook for social security. Among such economists, liberals tend to support having the Federal government run a budget surplus over the next twenty years, in order to increase national saving. A higher saving rate will raises the capital/output ratio, which in turn they hope will increase GDP, which would make it easier to afford high security outlays.

Conservative economists, myself included, would like to see the retirement age raised and then indexed so that it increases along with longevity. Such a policy would serve to limit social security outlays as a percent of GDP.

Liberal politicians have proposed something called a "lockbox" for social security. Their thinking is that while the baby boom population is still working, the social security system should be able to take in more in revenues than it pays in benefits. If these surpluses can be "locked up" and put away, they suggest, then when the baby boomers retire these funds can be drawn down.

Inside the "lockbox" are claims on future output. If the "lockbox" works, then the Baby Boom generation will have the legal means to collect its Social Security benefits, regardless of how high this makes the tax burden on those who are then working.

Conservative politicians have proposed something called "partial privatization," which would allow individuals to direct some of their social security contributions to the stock market. Their thinking is that the high returns from the stock market might make it easier to pay social security benefits.

Like the lockbox, partial privatization does nothing to lower the ratio of social security benefits to GDP. In fact, if stock prices do well, then baby boomers will be able to consume an even higher share of total output than they would if we stuck with the current social security system. Conversely, if the stock market does poorly, it seems likely that politicians would use tax money for a bailout. Thus, neither the lockbox nor partial privatization address the economics of social security.

The Case for Raising the Retirement Age

I favor raising the retirement age and indexing it to longevity. Indexing would affect people who are far away from retirement. Once you reach age 50, your retirement age would be frozen. Up until that point, the statutory retirement age would increase on a year-for-year basis with observed increases in longevity. Ten years from now, if longevity has increased by two years, then the retirement age will have gone up by two years as well.

An increase in the retirement age should apply to people currently aged 50 and under. It need not apply to people in their 50's and older, who already are planning for their retirement. Also, no one would be forced to wait until the statutory retirement age to retire. You can always retire earlier and live off of savings, with Social Security kicking in at the statutory retirement age.

Instead, by keeping the retirement age constant in spite of higher longevity, we are automatically expanding the role of social security in the economy. We are creating a situation in which improvements in health and medicine lead people to spend a large and growing fraction of their lives dependent on the government for income.

The proportion of the population that obtains benefits from social security might be an arbitrary political decision if the process of collecting taxes to pay social security benefits did not impose any additional cost on the economy. However, as we will see when we study the microeconomics of markets, in addition to collecting revenue, taxes impose a "deadweight loss," meaning that they reduce the economy's output. In other words, using taxes and transfers to try to redistribute the pie has its limits, because the pie gets smaller when tax rates are higher. The social security tax punishes work and thrift, so that we get less of those two activities the more we raise social security taxes.

The social security tax is a particularly evil tax, because it is regressive, meaning that high earners pay a lower proportion of their income for social security taxes than do low earners. Some of the seniors receiving Social security checks are quite affluent, while some of the workers paying taxes to fund those checks may be lower down on the wealth scale. This is not the sort of transfer scheme that we should be leaving on autopilot to grow at an exponential rate.