Why Retirees Aren't Running Out Of Money

There’s a lot of concern that Americans aren’t saving enough for retirement. I can give you a bunch of reasons to doubt that conclusion: More Americans are participating in retirement plans today than ever before, total retirement plan contributions are rising, employers are contributing more to retirement plans, Millennials are starting to save earlier than Gen-Xers or Baby Boomers did, and so on.

But here’s one way to think about it: A good sign that people aren’t saving enough is if they run out of money once they’re retired. But that’s not happening. In fact, for most retiree households, wealth increases over the course of retirement. But why does this happen and what does it mean for retirees?

I thought about this based on an article in which financial writer Michael Kitces examines the safe withdrawal rate for the drawdown of retirement wealth. Kitces shows that a retiree who holds a 60-40 stock-bond portfolio and follows the conventional “4% rule” for spending down retirement wealth would be more likely to quintuple his retirement wealth than he would be to run out of money within 30 years.

I’m a little skeptical about these kinds of drawdown rules and the “Monte Carlo” simulations that are used to explore them. I’m not questioning Kitces’ numbers. But we don’t know whether investment returns will be the same going forward as they were in the past, and (to my mind) there’s good reason to believe they won’t be. But more importantly, just looking at the probabilities of different events doesn’t tell you the values that retirees place on them. Assuming you retired with a decent income, quintupling your wealth won’t improve your quality of life nearly as much as running out of money would hurt it.

But Kitces’s article made me think of recent research looking at how retirees’ wealth changes over the course of their retirement. In theory, wealth should decline overtime. And, if they’ve designed things perfectly it should run out just after they’ve died, maybe leaving a few unpaid bills behind for good measure. But that’s not what’s happening.

Starting with the Federal Reserve’s Survey of Consumer Finances, let’s look at retirees who were ages 65 to 67 in the year 1989. The median household in that age group had a net worth of $122,318 in 2013 dollars. By 1998, when this group of retirees was aged 74 to 76, their median net worth had risen to $219,146. By ages 83 to 85 in 2007 their median net worth had increased to $247,532 and even after the housing crash the median net worth in 2013, for households then aged 89 to 91, had declined only to $202,400. It’s hard to see these figures as showing the typical retiree household running out of money. And note that these figures are medians, not means, so they’re not influenced by huge wealth holdings of the mega-rich.

Now, one explanation for rising wealth is that poorer retirees tend to die younger. In other words, wealth could fall for all retirees over time but when poorer retirees die that leaves only the richer ones behind. Doubtless this is true. But it doesn’t fully explain the increase in retirement wealth. Economists David A. Love of Williams College and Michael G. Palumbo and Paul A. Smith of the Federal Reserve used the Health and Retirement Study to track the assets of individual households over the course of retirement, thereby eliminating the effects of differential mortality by household wealth. Instead of looking only at wealth, they looked at the income that retirees’ wealth-holdings could provide them over the years. Love, Palumbo and Smith found that household wealth declines more slowly than remaining life expectancies, which means that the annual incomes retirees can provide for themselves increases rather than falls as they age. While not every household’s wealth rises, retirees were more than twice as likely to experience a large increase in their affordable annual income than a large decline.

In a recent study published in the Journal of Financial Planning, Chris Browning, Tao Guo, Yuanshan Cheng and Michael S. Finke looked at retirees of different wealth levels. The authors analyzed income, spending and wealth for a group of 65 to 70 year olds who retired in the year 2000. Using data from the Health and Retirement Study, they followed 704 retirees from 2000 through 2010, tracking how much they received in income each year, how much they spent, and how their financial assets changed over that 10-year period.

The authors divided retirees into five groups, ordered by their wealth. Across all five wealth quintiles, average financial assets increased from 2000 through 2010. That is to say, after ten years of retirement most retirees had more financial assets than when they started. Moreover, in the top three wealth quintiles, average annual spending didn’t even equal average annual incomes. In other words, most retirees weren’t even spending what they collected from Social Security, employer pensions and other common income sources, much less spending down their savings. The authors found that a retiree household with typical wealth could increase its annual spending by 8% over a 30-year retirement while still holding 40% of their financial assets in reserve against a health emergency or an extremely long lifespan.

There are different reasons why retirement wealth would rise rather than fall. Say, investment returns might be high or home prices might increase. But one simple reason is that retirees simply don’t spend that much. Yes, some retirees splurge on luxury cruises and other retirees are faced with high healthcare costs. But in the Survey of Consumer Finances, slightly over half of households aged 65 and over state that they don’t even spend the income they receive. About one-third of older households state that their spending is about equal to their incomes, while less than 15% state that their spending exceeds their incomes. If you’re not spending your whole income, there’s a good chance your wealth is going to rise.

Some of this lack-of-spending is precautionary–that is, you spend less today to protect against, say, a costly health problem later in life. And part of the decline in spending is due to retirees having more free time. For instance, retirees spend a lot less on food than working-age households, but neither the quality nor the quantity of their food seems to be any lower. The reason: Retirees shop around and pay lower prices even for identical goods. They also eat less at fast food restaurants and spend more time preparing food at home. But they don’t eat less at quality restaurants, showing that lower food spending isn’t because they don’t have the money to spend.

But personally I don’t think that’s all of it. The life cycle model in economics states that households will maximize their lifetime “utility”–or enjoyment–if they smooth the "marginal utility" of their spending from one year to the next. If we assume that households have the same capacity to enjoy their spending from one year to the next, as well as assuming that their own personal time preference for consumption is the same as the interest rate at which they can borrow or save, then they’ll tend to spend the  same amount every year. (I’m leaving out a lot here that’s not directly relevant to the point, so indulge me.)

Here’s my gut theory: The capacity to derive utility from spending declines as individuals grow older. For instance, a “young” retiree who is in good health would enjoy taking an around-the-world cruise with his spouse for which they didn’t have time while they were working. As the retiree ages, however, everyday aches and pains might make travel less enjoyable. His spouse may pass away, leaving him no one with whom to undertake expensive activities. Or, on the bright side, he might simply derive more pleasure from spending time with his grandchildren than he does from buying a fancy new car. Older retirees are more easily satisfied by the money they do spend, so the marginal utility they derive from spending–that is, the personal reward from spending an extra dollar–equals the marginal utility of early-retirement spending even if they don’t spend as much in total. I’m not sure how well this idea holds up in technical terms: After all, if you’re age 85 and have lots of money, you might have been better off spending more of that money at age 65, but you can’t go back in time to change it. I just think that older retirees derive relatively less pleasure from the things that money can buy. And so long as they have other things that they do derive pleasure from, be it family, friends, hobbies or social causes, that’s perfectly fine.

This makes retirement planning harder, since it's tough to anticipate what your future self will want to spend money on. The upside, however, is that the vast majority of retirees report being able to live comfortably on the money they have and few seem to be running out of money. That's something we can be grateful for.