Do We Really Need to Worry about Retirement Saving?

I personally believe, and more importantly all the work done at the Center for Retirement Research at Boston College suggests, that households are going to face real challenges in retirement.  Our National Retirement Risk Index projects that 53 percent of today’s working households are not going to be able to maintain their standard of living once they stop working.  Yet really smart economists who compare optimal savings with that reported in the Health and Retirement Study – a nationally representative longitudinal survey of older Americans – conclude that only a small fraction of households are behind in their retirement saving.  

I wouldn’t mind being wrong.  How lovely it would be if we really did not have to worry about how much money people will have in retirement.  But think about the numbers.  The Social Security replacement rate – benefits as a percent of pre-retirement earnings – is declining for the average age-65 worker from 42 percent to 36 percent as the full retirement age moves from 65 to 67.  Moreover, in the 2030s when the Trust Fund is exhausted, the program will be able to pay only three-quarters of that amount or 27 percent of pre-retirement earnings – a level not seen since the 1950s.  Congress will certainly put some new money in the program, but the average beneficiary will still see a reduced benefit, will have Medicare premiums deducted from this reduced amount, and will pay personal income taxes on the gross amount.

The only supplement to Social Security in the private sector will be balances in 401(k) plans.  The most recent national data (2010) show that combined 401(k)/IRA balances amounted to $120,000 for households that are approaching retirement.  These balances will be somewhat higher with the run-up of the stock market, but are unlikely to produce more than about $600 per month.  And these data only apply to households with 401(k)s; a significant percentage of households lack such coverage.  Hence I’m worried.

But John Karl Scholz and co-authors have published a number of rigorous studies that assess whether people are saving ‘optimally.’ ‘Optimally’ is defined as smoothing the expected marginal utility of consumption over time.  The authors define saving to include accumulation through pensions, Social Security, and housing equity as well as direct saving.  The results show that only a few households had less wealth than their ‘optimal’ targets.  These results appear to contradict the NRRI finding that 53 percent of households are ‘at risk.’  

A small part of the difference between the two approaches is the timeframe.  The initial optimal savings studies looked at cohorts with higher Social Security and more defined benefit pensions.  More recent studies, however, also conclude that most Early Baby Boomers are well positioned for retirement.  The more fundamental issue is the assumptions in the model.  Three stand out as important – how children affect replacement rate targets; what households consume in retirement; and how housing is treated.  

Our analysis assumes that households maintain relatively constant consumption over time, which implies that parents increase their per capita consumption when their children become financially independent.   In contrast, the optimal savings literature assumes that parents optimally chose higher household consumption when the kids are at home and lower consumption when they leave.  This pattern produces lower target replacement rates and fewer households falling short.  The problem is that the evidence does not appear to support this assumption.  We find that parents kick up their heels when the kids leave and increase their spending, raising their target replacement rate.  

The second issue is how much households spend in retirement.  Our analysis generally has people purchasing an annuity so that they spend a steady inflation-adjusted amount throughout their retirement.  For example, they spend $100,000 in constant dollars each year for as long as they live.  The ‘optimal saving’ approach has people drawing down their wealth.  If they were certain of being alive in each year, they would draw down constant amounts.  But households face the possibility of not being alive, so their consumption declines over time reflecting the declining probability of being alive.  To ensure steady consumption over the retirement period would require more money.  

The final issue is the treatment of the house in the optimal savings approach.  As just discussed, the model assumes that total consumption declines during the course of retirement, reflecting decreasing probabilities of being alive at older ages.  To avoid implausibly rapid declines in non-housing consumption, households would have to move to a smaller and smaller house.  Clearly people do not do this in the real world, so the decline in total consumption produces severe pressure on non-housing consumption.

So, the results from the optimal savings literature don’t make me feel better.  I will continue to worry.