Yes, There Is a Retirement Crisis

In an article in National Affairs, Andrew Biggs, of the American Enterprise Institute and Sylvester Schieber, an independent pension consultant and former chairman of the Social Security Advisory Board, question whether we are, in fact, facing a pension crisis. 

Among a long list of targets suggesting the nation might have a problem, they criticize two key assumptions underpinning the National Retirement Risk Index (NRRI), calculated by the Center for Retirement Research at Boston College.  The Index shows that 53 percent of working-age households will be unable to maintain their standard of living in retirement.

Underpinning the NRRI is a life-cycle savings model.  Households save during their working years and dis-save in retirement, their goal being to smooth consumption over their lifetimes.  When setting replacement rate and wealth accumulation targets, the NRRI makes two important assumptions.  The first is that the household’s consumption target does not drop when the kids leave home.  The second is that, during its working life, target consumption increases at the rate of economy-wide wage growth.   

If consumption really did decline and savings for retirement correspondingly increased once the kids left home, then the NRRI treatment would clearly be inappropriate.  Although the question is as yet unsettled, the existing empirical evidence suggests that total consumption does not decline at that time, and that parents accustom themselves to a higher standard of living.   

When constructing the NRRI targets, my colleagues and I made a conscious decision to assume that households had a preference for a standard of living that increased during their working lives at the rate of economy-wide wage growth.  This assumption reflected our belief that households care not only about their absolute standard of living, but also about their relative standard of living.  They want to be able to participate in societal gains, and the cost of participation increases over time.  Back in the 1960s, most people likely thought they were doing just fine.  But few people alive today would want to be stuck at a 1960s standard of living. 

Beyond the specific critique raised by Biggs and Schieber, it is important to note that the NRRI incorporates several assumptions that overstate available retirement income and potentially understate the share at risk: 1) households are assumed to retire at age 65; 2) they take out a reverse mortgage; and 3) they annuitize all of their financial assets, including the proceeds of the reverse mortgage.  In reality, the average household retires earlier than 65, does not take a reverse mortgage, and does not annuitize its assets.  Finally, households are only classified as “at risk” if they fall more than 10 percent short of a target replacement rate.

In short, we are unpersuaded by the Schieber/Biggs criticisms that the NRRI shows an overly pessimistic picture of the retirement challenge.  The evidence does not support the contention that households decrease consumption when the kids leave home, and we think wage indexing is consistent with households’ goal of maintaining their relative consumption over time.  Moreover, the NRRI includes several conservative assumptions that could understate risk levels.  

Finally, we also don’t agree with their conclusion that “public policy [should] aim for minimum levels of retirement income that are sufficient to avoid poverty and destitution, while allowing individuals and households – who know their needs and preferences better than a government planner – to decide how much to save on top of that minimum.”  The implication of that approach would be a dramatic reduction in Social Security – a terrible idea given the weaknesses in the rest of the retirement income system.