What makes MIT economist James Poterba’s research so valuable is that his work focuses as much on public policy as on the private-sector financial behavior that is the more typical field of observation of retirement researchers.
So it is unsurprising that a jury of his distinguished peers has honored Poterba, who is also president of the National Bureau of Economic Research (NBER), with the Retirement Income Industry Association’s (RIIA) award for applied retirement research.
That award, sponsored by AdvisorOne and Research Magazine, will be presented at a ceremony in Boston at RIIA’s annual conference in October. But following the announcement of the award last week, Poterba has shared some of his award-winning perspective with AdvisorOne readers.
The practical, advisor-useful research for which Poterba has been recognized is fully displayed in the most recent study the MIT economist has conducted, together with Steven Venti of Dartmouth and David Wise of Harvard.
In a landmark study NBER published in February, Poterba and his fellow researchers showed that nearly half of all Americans die with financial assets of less than $10,000. The study further found that many of the families observed had no housing or private pension assets either, thus relying exclusively on Social Security.
The researchers eschewed the retirement readiness index and other measures that merely try to figure out where people stand at the beginning of retirement.
“That’s like asking: ‘Am I leaving enough time to get to the airport?’” Poterba says. Maybe not on a day when authorities have closed off a key road, he answers.
So Poterba and his colleagues examined the issue from the opposite perspective. Instead of looking at the beginning of retirement, Poterba says “you can look at a different time clock: the date of death and looking back from there”—using a regularly updated data set based on two decades of biannual surveys of the asset and health dynamics of elderly Americans.
Looked at this way, Poterba says one can see that many “households reached retirement with a reasonably substantial level of financial assets, but by the time we see them at the end of life, their assets have dropped.” And of course there are households that both begin and end retirement with a low level of assets. But the data show that there are many financial shocks that can emerge at the end of life—the cost of modifying a home, the cost of moving to be near caregivers, out-of-pocket medical costs, to name just a few. And these shocks produce a great deal of heterogeneity in financial experiences in retirement, variation that is difficult to capture in summary measures such as a retirement readiness index.
The study thus replaces a static image of investing whereby a household saves for retirement, and then dis-saves on a precise income distribution schedule, with a more dynamic understanding of retirement realities.
What does tend to get drawn down in a regular and predictable way are Social Security and defined-benefit pension assets, which leads to questions regarding the value of annuities.
While many a retirement researcher has struggled with the so-called annuity puzzle—the question as to why more people for whom annuities would seem to be sensible investments don’t purchase them—Poterba, who has conducted voluminous research on the topic over several decades, is notably unpuzzled.
“There are two elements to an annuity decision,” he said. “One is asking what it would cost to purchase an annuity. The other is thinking about the benefit calculations associated with annuities.”
Poterba says many households simply lack the financial wherewithal to purchase annuities; their financial assets may not be sufficient to meet minimum purchase requirements, for example. At the other end of the spectrum, he says, there are households with asset levels where they are not likely to draw down assets in their lifetimes.
“There’s a middle group where Social Security won’t protect all their living standard going forward, and they may have sufficient financial assets to make the cost-benefit analysis,” Poterba continues. But for this middle group, whether to purchase the annuity depends on that cost-benefit analysis of “household specific circumstances” and the market price of annuities. “There is a large degree of heterogeneity” among middle-market households, he says.
One more of the innumerable key issues upon which Poterba’s research has cast light is the feared “asset market meltdown” about which some demographic analyses have warned. In an important paper published in 2001, Poterba—prompted by widespread fears at the time that the aging of the baby boomers was a demographic time bomb for stocks—investigated the relationship between the age structure of the population and stock market returns.
Poterba projected that demand for equities in the years 2020 to 2050 would not decline sharply, arguing that age-wealth profiles would decelerate only gradually—in contrast to the robust surge in asset values on the way up, as the age and wealth of the baby boomers swelled in the 1980s and ’90s.
Market developments since the publication of his research has not altered in any way Poterba’s conclusions.
“I think it is very important not to take a narrow national perspective on this. We live in a global capital market. If there’s an aging population in Japan, they may be able to invest in assets supplied in other countries,” Poterba says.
He adds that income growth in places like China, India and Brazil will supply the buyers for assets of aging populations in the U.S. “As their populations grow, their demand for U.S. financial assets will presumably rise,” he says.
While this is true for assets like stocks that can be easily traded across borders, Poterba warns the case may be different for “fixed assets that are difficult to trade to foreign investors—owner-occupied housing, for example.”
On balance, Poterba expects only “relatively modest” headwinds facing stock markets in the developed world as a result of demographic shifts.