Now that a few people (including some actuaries) have read “What’s Your Future Worth?” I am starting to get questions on one of the more counterintuitive concepts I propose. Specifically, in the book I have taken the position that there is no “correct” discount rate to use when comparing the future to the present and that to make good decisions about the future, each person should introspect deeply and determine their own “personal rate of discount”. No one seems to doubt that if left to their own devices most people will choose not to delay gratification and opt to receive benefits (monetary or otherwise) sooner than they “should” according to traditional economic theory. Many retirement planners and others in the financial world believe that this is simply a mistake and that their job is to help their clients overcome their “emotional barriers” and think objectively about the future.
I disagree and think that what many financial planners are perceiving as emotional errors are often just manifestations of not only our own individual and completely valid “time preferences” but that the traditional “foregone investment” approach to setting a discount rate is in fact based on a specific (and not necessarily correct) assumption as to the nature of time itself.
How time flows and the Time Value of Time
“I just read with great delight your article on ‘the time value of time’…It was a real eye-opener”
John Haley FSA and CEO of Towers Watson
“I wonder …whether you (in your article) meant to be ‘tongue in cheek’”
David Wesley M.D. and Vice President of Cologne Life Reinsurance Company
My ruminations on the nature of Time and how it flows differently for different people began decades ago and in 1996 I was inspired to put some of those thoughts together in my first, and (thus far) only, attempt to add something new to the field of actuarial science. In essence I started from the observation that for most of us, time feels like it is accelerating as we get older, and that while the months and years seemed to drag on forever when we were young, now they seem to fly by in the blink of an eye. I then defined a concept called “perceived time” based on the (hopefully) reasonable assumption that time is perceived to pass at a rate that is proportional to age. In other words a year will “seem” to last twice as long when you are 20 as it does when you are 40. From there I went on to restate a number of actuarial elements (discount rates, life expectancies, annuity values, etc) using “perceived time” instead of the normal definition of time in developing the formulas. The results were interesting, surprising and in my view explained a lot of the “irrational” choices that people make when faced with economic decisions that require them to choose between getting something now versus later (eg whether to take a lump sum payment or a lifetime annuity from their retirement plan)
At the time I was very satisfied with what I’d come up with. My article was published in the Jan/Feb 1997 issue of “Contingencies” a fairly well respected actuarial publication
( http://www.peterneuwirth.com/?page_id=16 ) . The response was reasonably positive and other than Dr. Wesley’s skepticism, most readers took my paper seriously and found its premise intriguing, if not obviously true. Over time, however, I have come to realize that while there may be a kernel of truth in my “theory”, the nature of our relationship to time, is far more complex than I had imagined.
It turns out that there is a pretty extensive amount of academic study devoted to just this question, though it is framed in a slightly different way. Specifically the term “time preference” is defined by economists as the way people value the present as compared with the future and over the last 40 years quite a few economists have conducted a myriad of often ingenious experiments to determine what the average individual’s personal rate of discount rate is and how it varies by circumstance. The results of these experiments show, first, that personal rates of discount are extremely high (14%-179% according to a 2002 survey of all the studies on the subject), but even more confounding is that the rate will vary by all sorts of factors that are “irrational”. For example:
People show dramatically different discount rates when the choices are presented at different points of time (eg today vs one year from now as opposed to one year from now vs two years from now).
People treat decisions affecting expenses (eg whether to pay $100 now vs pay $110 a year from now) very differently than they treat choices about income (eg whether to receive $100 now vs receive $110 a year from now).
The magnitude of the dollars involved has a big impact on the discount rate used to value future payments ($2000 payable a year from now is worth more today (relatively speaking) than $20 payable a year from now)
When multiple time periods and cash payments are considered (eg payments and/or expenses at 3 or 4 different points in time), choices are inconsistent and do not reflect a stable discount rate.
When risk is combined with choices between the present and the future are given, even when the probabilities of outcomes are specified (eg 50% probability of $1000 one year from now vs $200 for certain today), the answers received vary dramatically and unsystematically both when the probabilities are varied as well as the time periods.
So what exactly is going on? Well, one obvious conclusion is that more work is required to understand the way we, as humans, actually process the passage of time and view the future as compared to the present. On the other hand, as scattered and chaotic as the findings above are, there do seem to be two consistent patterns that suggest my original concept of the “Time Value of Time” plays a part. First, the fact that under almost all circumstances there is a positive discount rate suggests that people prefer to receive something now vs. later. There are some limited circumstances where people’s discount rate is actually negative and individuals prefer to delay gratification, (eg the “Christmas Club” savings account that Banks used to set up where people paid a small fee for the opportunity to not access to their money), but these are the exceptions to the rule. In addition, there seems to be a consistent finding that people’s discount rate is lower the longer into the future one looks. For example Leonard Green and others showed that people would choose to receive $100 today rather than $110 a month from now, but would prefer $110 payable 13 months from now vs $100 payable 12 months from today. The phenomenon becomes even more pronounced when the choice is extended out several years in the future. This is most dramatically illustrated by Richard Thaler who showed that over periods up to one year people manifested personal rates of discount of more than 100%, but over a ten year period average discount rates are under 20%. This finding is called “hyperbolic discounting” and one of the consequences of the concept of “perceived time” as I constructed it is that people’s discount rates will be hyperbolic and tend to decrease as both an individual ages and as an individual looks further into the future (all of these findings can be found in “Time Discounting and Time Preferences: A Critical Review” Journal of Economic Literature 40:pp351-401).
The point here is not to determine exactly how we humans perceive and evaluate the passage of time. That will take many more years of research to figure out. But what we can say, and this is the crux of my argument, is that people do seem to have their own “natural” way of perceiving the flow of time and it seems to have very little to do with how much they could earn on money that is set aside. To see this consider that the studies referenced above took place over a 40 year period during which prevailing interest rates and expected investment returns in general varied widely. As far as I could see there was no meaningful correlation between people’s “personal rate of discount” and the prevailing expectations of investment returns at the time of the study.
The financial planners out there may say that all of the above may be valid but people still should factor in their expected rate of returns when determining a discount rate, because by the time the future arrives they will be happy they did so. Maybe so, and there is no doubt that people should be aware of how much they might earn on money that is set aside, but my belief is that we live in the present and not the future and therefore it is important for us to make decisions in the here and now that are most consistent with who we are as human beings and that if we are true to our own internal values (including our time preferences) we will ultimately be making the right decisions for ourselves and be happier for it.