age is more than just a number

Yields on 10-year U.S. Treasuries hit an all-time low yesterday. Before you spin a story using recent events: remember long rates have been trending down for thirty odd years. And that's true in most advanced economies. So think bigger than jobs day or Brexit or liftoff. And while I've got you thinking in decades not data releases ... also consider that the share high-growth young firms, aggregate productivity growth, and general satisfaction have all been trending down since early 2000s. And again not unique to the United States.

No single factor has a chance at explaining all these trends ... Still I think a common thread of population aging and reduced risk taking is worth exploring. The idea that aging can change individual behavior is nothing new but sometimes the gradual and the familiar are easy to discount. Also, and a bit more provocatively, I want to argue that effects of population aging go well beyond the behavior and views of older individuals.

Let me start with a few general points on population aging ...

This chart from a recent UN report shows how widespread population aging has been across much of the world and is expected to be in the coming decades.


And looking more specifically at the United States in 2015, we see from a Pew report that median age of the population also varies quite by race and ethnicity.


So how might population aging lead to less risk taking?

As we age, we are less willing to take risks.

In numerous studies, including my job market paper, older individuals are less willing to take risks than younger ones. In fact, I was able to see how much risk tolerance changed with age (and other factors) in decade-long panel study of older adults. Aging by a decade led to a 17 percent decline in risk tolerance. For comparison, women were 14 percent less risk tolerant than men, on average, even after taking into account several other observables including age. My main takeaway from this work was that persistent differences across individuals create more variation in the willingness to take risks than the changes within person over time. However, of the factors that seem to cause risk preferences to change, aging was by far the most robust in my data and shows up in other studies, including those with younger adults.

The Great Recession likely "aged" our young adults.

Aging is more than adding up of years. It is about gaining life experiences. Malmendier and Nagel have an excellent "Depression Babies" papers on how the financial events that individuals actually live through shape their willingness to take risk. Extending their findings to today, keep in mind that Millennials have surpassed Boomers as the largest living cohort in the United States. You might think that this large, younger cohort would help balance out the effects on aggregate risk tolerance of Boomers aging. And yet, Millennials came of age in the Great Recession and thus their life experience is heavily influenced by those negative financial shocks. There is some good news on this front though, as Millennials age and the Great Recession will not loom as large in their total life experiences. Plus the researchers found that the most recent events experienced have more effect on the willingness to risks than earlier events. Still at least, anecdotally, many have suggested that the Great Depression had some lasting effects on those who grew up during during it.

Older peer networks may cause spillovers in risk taking.

We can move beyond the individual and also think about social impacts of population aging. A new working paper on consumption network effects by Di Giorgi, Frederiksen, and Pistaferri got me thinking about the effects of having older, more risk averse peers. I have only started to dig into this very cool empirical study with detailed Danish data that shows substantial co-worker peer effects on spending. In terms of mechanisms, they find some evidence of "keeping up with the Jones" but not status goods or risk sharing. Now that still leaves open the theoretical question of why the level of peer spending affects our own. The researchers hypothesize that peers may be a source of information, particularly on experience goods. In any case, such network effects suggest that the preferences of our peers could end up affecting us too. For what it's worth, working closely with a bunch of energetic, ever-working twenty-somethings at CEA for the past year definitely reduced my consumption of leisure. And being back at the Board, I can feel a some shift in the change-loving vibe around me. None of this says we sync up our behavior perfectly, but we are social and, on aggregate, having more people in our networks who are less willing to take risks probably creates some spillovers.

Working at older firms makes us act older too.

This last one is the most speculative and underscores the mess of causality. But before opining, let me stress that declines in various measures of entrepreneurial activity are a topic of great interest, regardless of the source. I am not necessarily arguing that population aging and a smaller pool of risk takers are reducing firm dynamism, though it might fit the profile of a global shock. Here I want to pose another spillover: The idea that firm aging may also be making us individually less willing to take risk. How institutions affect us is a massive topic, so I will stick to an imperfect anecdote. Questions about the macro-economy at CEA and the Board have a lot of overlap, as do the economic tools, outside research, data, etc. available to answer the questions. Still work flows a bit differently in a place that is almost entirely "reborn" with each Administration and than in one with more staff continuity. And I am not arguing relative merits. Older organizations offer benefits: sometimes that memo from twenty years ago is exactly what you need and reinventing the wheel all the time is inefficient. But older organizations, which also tend to expand in size with age, generally set a higher bar for change and create frictions that reduce risk taking.

In summary, age is more than just a number ... it can tell us something about a mindset, a set of life experiences, and even some social cues. As with any common thread, it is a messy, incomplete story of how risk taking may be changing, but it still seems to me that population aging, in its many forms, deserve more attention when we try to make sense of macro trends.

PS: I didn't quite fit in my post, but Kocherlakota's piece this week on the power of the old was the nudge that got me to write up my musing on this topic.

PPS: I linked to un-gated versions of all the papers. Some of them have been published (including mine, yeah) but those tend to be gated links. Also the reference section of each paper will give A LOT more research to read.

**Opinions here are mine and should not to be attributed to anyone with whom I work.**

claudiasahm

economist - my views here are my own

Also on this blog

SHARE:  Email · Facebook · Google · Twitter · Tumblr · Kindle
SUBSCRIBE:  Receive an email on new posts from claudiasahm

Comments

Really nice piece! Another dimension in which aging might be showing its effects is in the decline of migration (Molloy et al.'s work). Also, I recently saw some interesting research from the biology community about how effective age differs across people who are the same calendar age. This might also explain risk tolerance heterogeneity across individuals.
2016-07-10, Tyler Ransom

There may be similar correlation to things like industry concentration over the same time period.
2016-07-10, john

Age impacts our supply and demand for money. The Fed can choose to counteract, over-compensate or under-compensate for those impacts. I wish they'd read your post and then cut IOR. They paid 0% IOR for 95 years and only started paying the banks IOR in 2008.
2016-07-10, bill

Millennials without education debt (not just college) are effectively shut out of the middle class while those with education debt are very much in a middle age situation. Worse still, they tend to be shut out of the marriage market. Education debt is effectively inescapable unlike other forms of debt so these people are in a kind of house debtors prison. Anyone thinking that the Millennials offer a solution to our economic problems needs to consider these issues. Good to see you back online!
2016-07-10, Richard Knapp

To follow up on Tyler's comment above. Molloy et al (2016) http://www.brookings.edu/~/media/projects/bpea/spring-2016/molloyetal_decliningfluiditylabormarket_conferencedraft.pdf examine the relationship between aging and declining labor market fluidity. They find some role for demographic shifts (age and gender) but stress that these can explain only a portion of the decline in labor market fluidity. Also a nice discussion of aging spillovers on p 35-36 ... it's complicated. And I agree there is more to think about with firm characteristics.
2016-07-10, claudiasahm

If you actually tie the various points here together, you see that it isn't age, in and of itself, that reduces the tolerance for risk, it is experience. That fits the general downward trend with age, but also accounts for lower risk tolerance among those who came of age during downturns, that this, those who have taken risks and lost or watched others of their cohort take risks and lose. It also suggests a reason that risk taking has been declining over the past four decades. One is much less likely to get a positive payoff for doing so, and this has little to do with demographics, but a lot to do with the increasing power of large corporations and the extremely wealthy.
In fact, it is actually possible for risk tolerance to increase with age. In the 1970s, a lot of women, having raised their children to an age of independence, struck out in the workforce or in small businesses. They had experienced a long period in which risk was rewarded, though less so for those with young children to care for. As soon as that latter factor vanished, they became risk takers. In other words, restoring the rewards of risk taking would increase the tolerance for risk.
2016-07-10, Kaleberg


  • Notify me upon new comments

☺ Got it