Friday, September 2, 2011

Dan Ariely on Spending in Retirement

Noted behavioral economist Dan Ariely has a new blogpost making a few controversial points.  Here I would like to focus on his criticism of the traditional replacement rate rules of thumb (which are somewhere in the neighborhood of 70%).  Those rules of thumb are indeed just rules of thumb and everyone will have a unique situation, but these rules were derived by looking at household survey data and investigating how much people spend on average before and after retiring. He said that in conducting surveys about what people would like to spend after they retire, he adds up the costs of these activities and finds that their ideal replacement rate is more in the neighborhood of 135%.

I’m confused, as this seems to be the end of the story.  I could imagine if he asked people about what they would like to be spending currently, he might also find that they would like to be spending much more as well.

Economics is all about “maximizing happiness” as he rightly notes, but this must be done in light of one’s constraints.  And it usually involves “consumption smoothing,” which means that people can get the most happiness if they are able to spend about the same amount from year to year rather than letting this fluctuate (as his approach would do, someone is going to have to save an awful lot to have a 135% post-retirement replacement rate, which means they will have a much lower standard of living prior to retirement).

So we can’t just suggest that financial planners are giving bad advice because his surveys show that people would like to spend much more after retirement.

I have a new paper that will be in the upcoming October 2011 Journal of Financial Planning, “Getting on Track for a Sustainable Retirement: A Reality Check on Savings and Work.” That link is to a rough draft version.

Using the approach outlined in my paper, we can see what still awaits someone hoping for a 135% replacement rate when they retire. This approach is about how much one needs to replace from their personal savings; Social Security and other income sources would be added on top. So let me use “115%” as the target replacement rate from savings, which would provide somewhere in the neighborhood of 135% with Social Security too.

And let’s consider the case of a 40-year old. This 40-year old already has accumulated some wealth, and the following table shows her feasible retirement age for different scenarios regarding her current wealth accumulation, future savings rates, future stock allocations, and desired replacement rates.  More about specific assumptions are provided in the paper.

What we can see from this table is that unless she has already accumulated quite a lot of wealth (as represented by multiples of her final salary), she is going to be rather old before reaching the point where she can feel comfortable retiring with such a high replacement rate.  And this assumes she can keep the same job and same salary for so long. Life is indeed about tradeoffs, such as in the bottom part of the table where we see that if she currently has saved 4 times her salary, she can potentially retiree at age 69 with a 115% replacement rate, but she can retire at age 58 with a 50% replacement rate. Now she needs to decide whether to delay retirement in order to spend more or not.  

I think this is a real tradeoff that Prof. Ariely must consider, and it seems to me that a good financial planner should be able to help someone evaluate these tradeoffs to come to the best happiness-maximizing compromise solution.  



  1. Anything less than 100% is inadequate and even 100% will not allow dealing with inflation. Any reduction in work related expenses is more than offset by other expenses in retirement. The idea that no saving is required in retirement is inaccurate. One must still have an emergency fund and replenish it otherwise the funds set aside for retirement income will be at risk of unplanned depletion.

  2. Dick, thanks for the comment. All my calculations do include inflation adjustments. And I'm not really in the business of saying what an appropriate replacement rate is. This research is just saying how much more one needs to save and work in order to have been able to finance the replacement rate of their choice through age 100 in the worst-case scenario from history.