But the tax treatment of 401(k)s isn't as advantageous as it first appears. For that reason, contributing the maximum amount may not be the best idea for all workers. In fact, economists Laurence J. Kotlikoff of Boston University and Jagadeesh Gokhale of the Federal Reserve Bank of Cleveland argue that many workers would do better by switching contributions from their 401(k)s to other accounts.
The problem is that while postponing taxable income gives you a break now, it raises your future taxes. First, payouts from 401(k)s during retirement are taxed as regular income. In contrast, long-held investments in ordinary, unsheltered accounts are taxed at the (lower) capital-gains rate when sold. A fat 401(k) could generate so much fully taxable income that it pushes you into a higher bracket.
It will also subject more of your Social Security benefits to income taxes. Even the most-praised feature of the 401(k)--that it reduces your current taxable income--has a small drawback. If it pushes you into a lower tax bracket, it will reduce the value of your mortgage deductions while you're young and still paying lots of mortgage interest.
Kotlikoff and Gokhale conclude that most Americans should stop regarding the 401(k) account as their primary vehicle for retirement savings. They find that most people should contribute only 4% to 6% of their pay to their 401(k)s. (The national average for plan participants is just under 7%, and many people contribute far more.) Meanwhile, they say, people should save much more than they do now in ordinary, unsheltered investment accounts--about 17% of pretax income for household incomes over $100,000.
Kotlikoff and Gokhale did a customized version of their study for BusinessWeek Investor using ESPlanner, a detailed financial-planning program published by Economic Security Planning, of which Kotlikoff is president. (The $65 program, which requires a fair amount of data entry, is available through www.esplanner.com.) For starters, they took the case of a hypothetical married couple, age 25 and earning equal salaries, who plan to have two children by age 30 and send them both to college. They assume that the couple's income will rise 1% a year after inflation until retirement. They also assume a 3% employer matching contribution, and they figure that the couple will earn 6% after inflation on their investments, although the results are similar when lower or higher rates of return are plugged in.
Their conclusion: Contribute at least up to the level at which your employer makes matching contributions. But beyond that, watch out. Kotlikoff and Gokhale say that if their 25-year-old couple has a household income of $50,000 to $100,000, each spouse should put in just 4% of pretax pay. From $150,000 to $250,000 in family income, the right level is in the range of 5% to 6%. It's not until family income reaches $300,000 that each spouse should max out at $11,000 each, which for them is a 9% contribution rate. (The program assumes that their companies' plans allow contributions to go that high.)
That's for a couple just starting out. Let's say the same couple is 45, earning no more than $250,000 combined, and both spouses have been contributing 13.5% a year since age 25. ESPlanner concludes that they should stop making contributions altogether.
Instead of overstuffing the 401(k), ESPlanner recommends that families save more outside it. For example, Kotlikoff says that a Roth IRA is a better deal than a 401(k) for people whose incomes allow them to qualify. (The current income ceiling is $95,000 for individuals and $150,000 for couples.) While Roth contributions are made with aftertax income, distributions in retirement are tax-free. Kotlikoff says workers should ask their employers to make contributions to a Roth IRA in place of the company 401(k) match. "If we're inducing people to save in 401(k)s on the basis of tax savings that aren't there, that's wrong," he says.
Let's say that you're not convinced by Kotlikoff's argument and you still want to max out your 401(k). In that case, at least be sure to do it in a way that gets you the maximum possible company match. If you set the percentage of your pay too high, you could hit your ceiling before the end of the tax year. After that, you won't be able to make any more contributions. And because you're not making contributions, the company will stop making matching contributions. You'll get less free money from the company over the year than if you had contributed a lower percentage of your salary and hit the annual limit at the end of December.
Take an example that assumes you can contribute up to the IRS ceiling. Your pretax income is $10,000 a month. You contribute 11%, or $1,100, to your 401(k). At that rate, you reach the IRS limit of $11,000 at the end of October. The company matches the first 3% of your contributions, amounting to $300 a month. But in November and December, you make zero contributions, so you get zero match. If you had contributed 9% a month instead of 11%, you still would have nearly maxed out your contribution by yearend: $10,800. Plus, you would have gotten two more months of matching--a free $600.
If you think you might be overcontributing, it's not too late to fix things (table). Most 401(k) plans allow employees to change their contribution rates at least once a month. Doing a little arithmetic could earn you hundreds or even thousands of dollars.
Again, though, that's assuming you want to contribute the maximum to your 401(k). Kotlikoff and Gokhale make a case that a 4% to 6% contribution rate is the right level for most two-income families. There are other considerations, too. For instance, do you have a cash cushion equal to at least three to six months of living expenses? If not, save up before you raise your 401(k) contributions.
We've all been bombarded with advice about where to invest our 401(k)s. It's time to think more about how much to put into them in the first place. By Peter Coy