I was just reading a blog that reported the typical pre-retiree had $42,000 in a 401(k) account as of 2010. Yet, I happen to know the actual figure is more like $100,000. These two numbers reveal how much confusion can arise over a simple statistic—how much people have saved in their 401(k)s.

What’s the discrepancy? In 2010, 60% of pre-retiree households (age 55–64) owned a tax-deferred retirement account, that had a median value of $100,000.* The figure includes money in any of the alphabet soup of retirement accounts—401(k), 403(b), 457, profit-sharing, money purchase, etc. The money may be in a current or prior employer plan; it may also include funds rolled over (or contributed to) an IRA.

What about the other 40% of pre-retiree households? They have no retirement accounts—the value is zero. That’s where the $42,000 figure comes from. $42,000 is the retirement account balance of all pre-retiree households—the 60% with a median value of $100,000 and the 40% with zero. As we say in the research world, the $100,000 value “excludes the zeros.” The $42,000 value “includes the zeros.”**

For the six in ten older households who have retirement accounts, a median balance of $100,000 will be a useful supplement to Social Security (and any other income sources such as pensions). It might generate $4,000-5,000 per year if used as a regular income source. That’s hardly generous, but these balances are expected to continue to grow in the future and become an even larger source of retirement income.

But why do four out of ten older households not have a retirement account at all? There are two main reasons.

First, many smaller companies don’t offer a retirement plan. This is sometimes called the “coverage” problem—as in, which workers are “covered” (or not) by a plan at work. The second reason is that these households once had an account—but cashed it out earlier in life. This is the so-called “leakage” problem—as in money “leaking” out of a retirement account and being used for other purposes.

U.S. 401(k)s are distinctive in that plan participants gain access to 100% of their account balance upon job change or loss. To discourage use of the funds, the tax code imposes income taxes plus a 10% penalty tax on any amount withdrawn. And indeed, many dollars saved are preserved for retirement. Account owners also have access to funds while they are working, through loans or various types of hardship and other withdrawals, although these are a much smaller source of “leakage.”

However, some research suggests that, in a given year, one dollar leaks out of the system for every seven dollars contributed to retirement accounts (through leakage in all its forms).*** That’s a high level of friction. It’s probably worse in tough economic times, when some who have lost their jobs use retirement accounts as a form of private unemployment insurance.

As we think about maximizing retirement savings, it seems like there are two distinct issues. One is figuring out ways to expand tax-deferred savings to more companies. There are a variety of proposals floating around in Washington, too complex to itemize here, that attempt to do that. And the other question is whether to impose further restrictions on the flexible access that account holders enjoy today.

Meanwhile, these facts highlight the strategy for individuals seeking to help maximize balances in a tax-deferred retirement account. Work for an employer who offers a plan. Save regularly in it. And avoid the temptation to tap savings, particularly when changing jobs.

* Bricker, Jesse, Arthur B. Kennickell, Kevin B. Moore and John Sabelhaus, “Changes in U.S. Family Finances from 2007 to 2010: Evidence from the Survey of Consumer Finances,” Federal Reserve Bulletin, 98 (2), June 2012, Table 6.
** Proving that even such a simple comparison isn’t so simple – technically the $42,000 comes from sorting all pre-retiree households by aggregate net worth, looking at the middle 20% of households, and then arriving at the median holding in that group. It’s not exactly the same calculation I describe, but it’s close enough.
*** Bryant, Victoria L., Sarah Holden and John Sabelhaus, “Qualified Retirement Plans: Analysis of Distribution and Rollover Activity,” Wharton Pension Research Council Working Paper 2011-01, Figure 1.

Note: All investing is subject to risk, including the possible loss of principal.