Step by Step

Author: Eduardo Porter

Published: April 12, 2005

EMPLOYERS are slashing their fixed-benefit pension plans. Cuts in Social Security are being proposed. Americans understand that the only way to retire comfortably is to save much more money. So why don't we?

This behavioral quirk has long stumped mainstream economists, who tend to assume that people are generally rational beings who have read the Aesop fable about the ant and the grasshopper and understand the virtues of thrift.

Yet over the last couple of years, researchers from fields as disparate as economics, psychology and neuroscience have come together to shed new light on the problem. Armed with brain scans and economic models, they are finding that the reluctance to set aside for tomorrow may be rooted in traits that were carved by evolution, hard-wiring our brain to prefer immediate gratification. The research also suggests that, when properly re-engineered, our behavioral peculiarities can be harnessed to make us save more.

We save remarkably little for a society that views a golden-age retirement as a right. Fewer than two-thirds of Americans own a tax-deferred savings account like an individual retirement account or a job-based 401(k) plan. Of those who do, many contribute much less than they need.

In an appearance before the Senate Special Committee on Aging in March, Alan Greenspan, the Fed chairman, said that absent a raise in payroll taxes, by the middle of this century Social Security benefits would amount to 30 percent of a typical worker's last wage -- down from 42 percent today. To replace 80 percent of pre-retirement income, as financial advisers recommend, workers will therefore need much more in the piggy bank.

Yet by 2001, the last time the Federal Reserve Board conducted its broad survey of consumer finances, a typical worker 55 to 64 years old had $42,000 in his 401(k) and I.R.A.'s, about enough for an annuity payment of $200 a month.

Some people do better than others, of course. William G. Gale of the Brookings Institution argues that about a third of workers are on track, salting away enough for a comfortable retirement. Yet another third have clearly not saved enough for a penury-free retirement. The middle third, mostly younger workers, could go either way.

We know we are doing something wrong. In one survey of workers, the economists David Laibson and James J. Choi of Harvard and Brigitte C. Madrian of the Wharton School of the University of Pennsylvania found that while 68 percent of respondents reported saving too little, only 24 percent said they would increase contributions to their 401(k), and only 3 percent actually did it within four months.

''It is not a question of education,'' said Stephen Utkus, director of the Vanguard Center for Retirement Research. ''It is a question of acting on the knowledge you have.''

What's wrong with everyone? Many people are simply befuddled by the arithmetic of retirement saving, clueless about how much they need to save. But the complexity of the decision cannot alone explain such procrastination. Indeed, new research suggests that we are just less rational than we have been led to believe.

The life-cycle theory of savings, a mainstay of standard economics, says that we act rationally to even out our consumption -- saving when we're young to consume when we're old -- coolly maximizing our economic well-being over our lifetimes.

But many people do not really behave that way. In fact, we exhibit lots of foibles that make little sense.

For instance, we value the $100 that we may lose more highly than the $100 we may win, a concept that economists call loss aversion. This helps explain why we can hate losses, though be willing to suffer opportunity costs -- that is, foregone future earnings -- even when the two are economically equal.

This eccentricity on its own could cut into our retirement savings, making us reluctant to lose income today in exchange for a future reward. But our economic inconsistencies run deeper. If given a choice today between doing 10 hours of awful work on May 1, or 12 hours on May 15, virtually everyone would prefer the 10 hours. But come April 30, we pick the 12 hours, exhibiting an overweighting of immediate rewards and costs that scholars call hyperbolic discounting. Last year, Mr. Laibson and his fellow economist George Loewenstein of Carnegie Mellon University, together with Jonathan D. Cohen and Samuel M. McClure of Princeton's Center for the Study of Brain, Mind and Behavior studied brain scans of people choosing between rewards at different points in time. The scans suggested that decisions were made in two areas of the brain. One is the lateral prefrontal cortex, a system that developed as humans began to split off from apes, which evaluates trade-offs between abstract rewards. The other is the more primitive limbic system, designed to respond to immediate stimuli, which humans share with other mammals.

Mr. Laibson said that this research was in its early stages, but he noted that it was compelling. It suggests that this primitive brain structure, which developed to deal with immediate contingencies like a charging rhinoceros, tends to drive many of our decisions.

''It seems to drive a lot of indulgent behaviors,'' Mr. Laibson said.

Among them is the procrastinating, spend-now-don't-worry-about-tomorrow ethos that guides our approach to saving. Understanding this psychological undertow is helping economists create new saving methods that take advantage of our behavioral quirks.

''The biggest driver of savings behavior is inertia,'' said Brian Tarbox, president of the Tarbox Group, a firm that helps fund providers devise pension plans. ''Inertia, engineered in the right way, can be used to the benefit of participants.''

The required engineering is straightforward: it just requires flipping people's choices. Rather than offering workers the opportunity to save, companies are automatically enrolling them in savings plans and offering them the chance to opt out.

Ms. Madrian and Dennis F. Shea of the UnitedHealth Group studied savings patterns in one company that switched from optional to automatic enrollment. Before the switch, enrollment among newly hired workers was 37 percent. After the switch, it jumped to 86 percent. The change was most drastic among lower-income workers, who need to increase their savings the most. The 401(k) enrollment among newly hired workers earning $20,000 to $29,000 rose to 83 percent from 25 percent.

Automatic enrollment has drawbacks. Procrastinating workers will stick to the default savings rate set by the company, even workers who would have saved more on their own. They will also stay with the default asset allocation -- even when the company merely dumped the savings into a money-market fund offering very low returns.

''The key to defaults is that defaults are sticky,'' Ms. Madrian said. ' 'People are going to stick at the default.''

One alternative is to force people to make a choice. Ms. Madrian, Mr. Laibson and Mr. Choi found that forcing workers to choose whether they want to join a 401(k) plan, instead of just giving them the option to do so, also increased participation rates radically.

Two economists, Richard H. Thaler of the University of Chicago and Shlomo Benartzi of the University of California, Los Angeles, have a different method, which they call Save More Tomorrow, or Smart. The idea is to offer workers the option to start saving in the future and increase their contribution rate by a set amount every year, say, 1 or 2 percent of their wages, to a preset limit.

This device helps overcome hyperbolic discounting. Just like the chunkier among us have an easier time signing up for a diet at a later time rather than starting one immediately, workers are more likely to agree to bite the savings bullet if it is deferred.

In one pilot tracked by the economists, workers' average contribution to their 401(k) plans rose from 3.5 percent of their wage before joining the Smart plan to 13.6 percent four years after signing up to the new method.

These ideas are gaining traction in the financial industry.

The fund manager Vanguard, for instance, has about 180 corporate 401(k) plans on a new program that lets workers sign up for automatic annual contribution increases; it also has 50 plans on automatic enrollment and three companies that have chosen both.

One that is signing on to the full package is BHP Billiton, a mining company that is headquartered in Melbourne, Australia. According to Dan Helman, who is responsible for retirement plans in the United States and Canada, BHP Billiton signed up first for automatic enrollment, then it started offering employees Vanguard's managed account option, under which Vanguard actively manages the asset allocation. This month it should begin making the automatic annual increase available, too.

The results, so far, are encouraging. ''Participation rates in the 401(k) were about 70 percent before,'' Mr. Helman said. ''Now there are only two people out of more than 2,000 who are not on the 401(k).''

It is too soon to say whether this tinkering with our limbic system will truly make us close the savings gap. For one, only half of all workers have a company savings plan. Tinkering with 401(k) structures does nothing for the other half.

''How much of a solution this is will depend on how soon people take this up,'' Mr. Thaler said. ''If we reach people early enough this would be a pretty good solution.''

But if this rejiggering of incentives proves insufficient, the new behavioral insights could spur a wave of interest in finding other solutions. ' 'I can see enormous interest in the pharmaceutical sector to design a neurochemical intervention mechanism to avoid temptation,'' Mr. Laibson said.