Economic View: An Automatic Solution for the Retirement Savings Problem

But some problems are frustrating in another way: we know how to fix them and we can afford to fix them, but we drop the ball. That’s the situation with a crisis facing many Americans: saving enough for retirement.

Here is one measure of the problem: A Boston College economist, Alicia H. Munnell, and her colleagues have estimated that more than half of Americans are saving too little to support an adequate lifestyle if they plan to retire at 65. Why is the situation so serious? One reason is that traditional pension plans — in which employees have almost no decisions to make — are being supplanted by defined-contribution plans like 401(k)’s. In these plans, employees have to decide for themselves how much to save and how to invest their money. For many people, being asked to solve their own retirement savings problems is like being asked to build their own cars.

To fix this, we need to do two things. First, make payroll retirement savings plans available to everyone. Then, add empirically proven design features to them, making it easier for workers to make good choices. In other words, improve the plans’ choice architecture.

Payroll savings plans are vital because they are essentially the only way that middle-class Americans reliably save for retirement. Your grandmother probably knew that the best way to save is to put money aside before you have a chance to spend it. That approach has always worked — and is a core idea embedded in these plans.

In the past, homeowners used another form of forced saving, building home equity by paying off their mortgages. But the ease of refinancing has eroded the norm that people should pay off these loans by the time they reach retirement age. Among households with someone over 60, mortgage debt has grown drastically in recent decades. (Here’s a savings tip: If you are over 45, use today’s low interest rates to refinance with a 15-year mortgage.)

Given the importance of payroll savings, it’s alarming that only about half of the American work force has access to a retirement savings plan in the workplace — and that number falls to 42 percent in the private sector, according to Boston College research.

The Obama administration has proposed a simple solution to this problem: the automatic I.R.A. This plan, originally proposed by scholars at the Brookings Institution, would require any employer that doesn’t offer its own plan to enroll workers automatically into individual retirement accounts, with the option to opt out. The burden on employers would be tiny, and the benefit to workers could be life-changing.

The concept puts to work part of what we behavioral economists, along with industry experts, have learned about effective 401(k) retirement plans over the past couple of decades. The operative word common to many best practices is “automatic.”

When employees are first eligible for a retirement savings plan, they should be enrolled unless they choose to opt out. This solves the procrastination problem that keeps roughly a fifth of workers who are eligible for a plan from joining, even when the employer is matching some of their contributions. Companies that adopt automatic enrollment find that few employees opt out initially, or later.

But we should move beyond automatic enrollment alone. That’s because most companies set a low default savings rate for new enrollees, often at just 3 percent of their income. Of course, employees can choose a different, higher rate, but many just accept the default percentage and stay with it indefinitely.

My colleague Shlomo Benartzi, a business professor at the University of California, Los Angeles, and I devised a successful solution to this problem that we call Save More Tomorrow. Under it, a worker can join a plan in which their savings contributions are increased, say, one to two percentage points a year, each time the employee gets a raise. In the first company that tried this plan, the savings rate more than tripled in three years.

Some companies find that linking savings increases to pay increases puts a burden on their payroll and human resource departments. Such companies can avoid this problem by using a generic version of the plan, called automatic escalation, that steps up savings rates each year until the employee hits a predetermined maximum.

In a recent report in Science magazine, Professor Benartzi and I estimated more than four million people were using some form of automatic escalation, and had collectively increased annual savings in the United States by over $7 billion a year. This is significant progress, but we could do much better.

Many employees don’t know that their workplace has such an option, and finding out how to enroll can be hard. This helps explain why only 11 percent of eligible workers have signed up for it.  Promoting automatic escalation and making sign-up easy, or even automatic (with the ability to opt out, of course), could greatly increase enrollment. In our original study, in which a financial adviser was available to explain the plan and, importantly, fill out the appropriate forms, nearly 80 percent of workers who were offered the plan took it.

THE third piece of the automatic plan involves investments. Retirement savers tend to be relatively passive investors, often sticking with whatever asset allocation they selected on the day they joined the plan. But those who do make changes often do so at exactly the wrong time: they buy high and sell low.

Although the stock market has doubled in the past few years, 401(k) investors have collectively been selling stocks to buy bonds during this period. (Note that in January this year, there was a sharp reversal, with investors pouring money into stocks. This might make some trend watchers nervous about future stock market returns!)

A solution to bad market timing is to offer a default investment vehicle, like a target-date mutual fund, that automatically rebalances an investor’s portfolio, both cyclically as the market rises and falls, and as the client ages, reducing stock holdings as retirement approaches. Of course, it is essential that these target-date funds have reasonable fees.

For evaluations of how corporate plans stack up, consider the ratings offered by services like BrightScope.com. If you aren’t happy with what you find, complain to your company’s management. And if you are part of management, get busy.

Richard H. Thaler is a professor of economics and behavioral science at the Booth School of Business at the University of Chicago. He has informally advised the Obama administration.

Economic View: For the Wealthy, a 28 Percent Tax Solution

In a nutshell, the fewer deductions and other “tax expenditures” we have, the lower the rates can be. That part is simple.

The problem is that cutting rates is more popular than closing loopholes, especially those like the mortgage-interest deduction that are used by millions of taxpayers. But there is a possible solution.

I call it the modified Reagan 28 plan, or just the “28 plan” for short. It is a simple framework for thinking about tax policy. Though I’ve named it in honor of Ronald Reagan, it is similar in many ways to the ideas proposed by the Bowles-Simpson commission, which is a good starting point for any serious discussion of tax reform.

My first premise is that we should devise tax policy from the top down. As everyone now knows, nearly half of American households do not pay any income tax, though they do pay Social Security taxes, sales taxes and so forth. And many households that earn enough to start paying some income tax are essentially on a flat-tax system, because few itemize their deductions. That means they mostly just pay a percentage of their income above the threshold where taxes kick in. So let’s think about how to tax the rich, then work our way down the income ladder.

For this discussion, let’s define a household as rich if its income exceeds $1 million a year. In fact, my plan applies only to the income such households earn above that threshold. And I can state my idea in just one sentence: All income above $1 million a year for a household will be taxed at 28 percent. There are no deductions, and all income, including capital gains and dividends, is included. President Reagan favored something like this approach. His 1986 tax plan also taxed dividends and capital gains at the same 28 percent rate, as would the Bowles-Simpson proposal.

While we’re at it, let’s make the corporate tax rate 28 percent, too, because our current rate is high by international standards. Oh, and the estate tax exemption? On amounts above $3.5 million for individuals, the rate would be, of course, 28 percent.

What would this plan accomplish? First, by establishing the same marginal rate on all income sources, the incentives disappear for shifting from one source of income to another. Although there is much discussion about how taxes affect people’s willingness  to work, most people don’t have much flexibility about how many hours they’re employed. For the rich, however, it can be relatively easy to switch income from a highly taxed category to one that is taxed at a lower rate. One reason that Mitt Romney’s taxes have been so low — he paid an effective federal income tax rate of only 14 percent in the two years for which he has disclosed his returns — is that venture capitalists have figured out a legal way for the incentive fees they receive to be treated as capital gains income, currently taxed at a rate of only 15 percent. If we tax all income at the same rate, these games will end.

But what about the argument that taxing capital gains and dividends at the same rate as ordinary income will discourage investment? I don’t find this claim convincing. People are willing to buy bonds issued by corporations even though the interest is taxed as ordinary income, and I don’t see why investors need a special tax break to induce them to put money in the stock market, which has historically paid high returns. One bonus from adopting this plan would be a short-term windfall to the Treasury from investors realizing their capital gains now, in order to pay the current, lower rate. (Also, I do think that capital gains taxes should be adjusted for inflation.)

Wouldn’t nonprofit organizations suffer if we eliminated the charitable deduction for the wealthy? There might be some effect, but notice that by reducing the marginal rate to 28 percent, the tax incentive to donate would already be quite a bit lower than at current rates of 35 percent or more. And if this political season has taught us anything, it’s that rich people don’t need a tax deduction to contribute to causes they believe in. (Political contributions are not deductible.)

We could pay for the reduction in the corporation income tax by broadening that base as well. There are many special deals — like subsidies to oil companies — that could be eliminated.

OF course, I haven’t said what would happen to the households in the middle, or what the taxes would be on the first $1 million for the rich, but the Bowles-Simpson commission offers a comprehensive plan that is in the spirit of my suggestion. One possibility is to scale back deductions smoothly, starting at household incomes above $250,000, and completely eliminate them for incomes above $1 million. That would leave the popular deductions fully in place for those earning less than $250,000. A more radical plan, curtailing deductions for this large group, is probably politically infeasible, at least for now.

And what if the resulting revenue falls a bit short of what we need to start trimming the deficit? I suggest that we get our gasoline tax more in line with those of the rest of the world. Gradually raising it to something like $1 a gallon would both bring in revenue and help reduce emissions. In the long term, we could set the rate as a percentage of the price at the pump. Maybe 28 percent?

Richard H. Thaler is a professor of economics and behavioral science at the Booth School of Business at the University of Chicago.