Wednesday, December 8, 2010

"Topping Off"

Casey Mulligan wrote an interesting piece at the Economix blog about whether the poor should prefer big government. Near the end of his article, he says:

The phenomenon of “topping off” illustrates how higher-income people derive more value from various types of government spending than low-income people do. High-income people seem to value pensions, which is why they “top off” Social Security: they accumulate pensions above and beyond what the government forces them to accumulate though the Social Security system. You might say that pensions are easy to accumulate when your income is high — and that’s the point.

Low-income people, on the other hand, often accumulate no pension beyond what Social Security forces them to accumulate.

This excerpt might be a little mysterious. After all, what is "topping off" and why does Mulligan state that this phenomenon means Social Security is less harmful to wealthy individuals? Here's my explanation in the context of a contrived example.

Imagine a world where $20,000 is enough to have an average quality life in retirement. In this world, suppose that the following facts are true:

  1. A wealthy individual would like to save $30,000 for his golden years.
  2. A poor individual would like to save $10,000 for his golden years.
  3. You are either wealthy or poor and half of people are poor.
  4. There are no interest rates, there is no inflation and no one dies until after retirement (to make life simple).

At some point, the government notices a disturbing trend: half of individuals in society have below average quality of life in retirement. From a policy perspective, this simply won't do.

The government concludes that poor people have problems saving money for retirement. They claim the poor suffer from lack of foresight. For this reason, the government institutes a "Social Security" program, where each individual is taxed during his working years enough to guarantee an average quality of life in retirement (in our example, $20,000).

This program acts like a forced savings regime. Over the course of each worker's career, the government collects $20,000. Upon retirement, everyone receives the money the government saved on their behalf, which they use for retirement consumption. Individuals cannot use the government savings to pay down debt.

For the wealthy, retirement consumption is unaffected. They want more than the government is forcing them to save. Hence, wealthy individuals top off their retirement fund by contributing an extra $10,000 to a private savings plan. If there are zero collection costs,* Social Security of this form neither helps nor hurts the wealthy.

For the poor, retirement consumption may be different under the program. If the poor cannot use the government savings to pay down debt, the poor would be forced to consume $20,000 in retirement instead of the $10,000 they would have chosen in absence of the policy. For the government to force $10,000 more savings in retirement for poor people, they must induce poor people to spend $10,000 less during their working years.

Poor people don't top off their retirement savings precisely because the program distorts their consumption. This contrived program of "Social Security" makes poor individuals worse off, while not affecting wealthy individuals.

The system I described is not technically how Social Security works in the United States (that's why I called it "Social Security" above). In my example, the government was forcing retirement savings. In the real world, the current working population pays taxes that support the current retired population (who received a promise of retirement benefits during their working years). That caveat aside, there may be something to this intuitive idea of topping off.

One final note: My example is not an adequate description of how economists think about the overall desirability of Social Security programs. If you are interested, the right way to think about Social Security in the real world is to think about the savings decisions in an "overlapping generations" setting. In short, you get different results on the desirability and optimal form of Social Security depending on how quickly the population grows (or shrinks), how per capita wealth grows over time, how interest rates evolve over time and what is the appropriate interest rate to use for discounting.

*I am assuming income taxes for this "Social Security" program do not discourage working and the government can collect and distribute these taxes at zero cost. Relaxing each of these assumptions would create a cost of the program that would make it less desirable to enact.

No comments:

Post a Comment

Please feel free to share your ideas about this post in the open forum. Be mindful that comments in this blog are moderated. Please keep your comments respectful and on point.