Piggy bank in jaws of vice (Credit: Cathy Yeulet/Thinkstock) (Image: Cathy Yeulet/Thinkstock)

There’s a decent amount of research out there that suggests retirees underspend, an effect that’s been dubbed the “retirement consumption gap.” 

The idea that retiree households under-consume has important policy implications, as well as for retirement savings in general: Why save all that money for retirement if you’re not going to spend it?  

Most research exploring retirement spending has focused primarily on retirement asset levels (i.e., whether assets are being depleted and the extent the depletion would be considered optimal) or used surveys of retiree households, but has not also considered pre-retirement spending (i.e., the retirement liability) when noting the effect. 

To better understand retiree spending, I recently teamed up with Warren Cormier, who heads up the Retirement Research Center at the Defined Contribution Institutional Investment Association, on some research that was just released. 

In our research we not only focus on retiree spending, but we also consider pre-retirement spending along with assets available to fund retirement, such as savings and pensions, using data from the Health and Retirement Study. Considering assets, along with pre-retirement spending, should provide a richer context for the understanding of spending than a review of  retiree spending alone.

Research Results

We found a few interesting things in our analysis. 

First, most households are not on track to replace pre-retirement income. We estimate only 18% of retirees have enough retirement wealth to maintain pre-retirement spending levels using a relatively conservative replacement metric. 

This finding is not that surprising given media coverage of retirement readiness (retirement crisis!); however, it’s important context when understanding the spending choices of retirees.  

Second, most retirees deplete financial assets, but at different rates. Median real financial assets were 35% lower 10 years following retirement and 65% of households had fewer assets 10 years after retiring, in today’s dollars. 

Households with the lowest initial funded ratios tended to spend down financial assets the most, relatively speaking, and had the fewest assets initially (in absolute terms). While retirees might not like spending down their savings, they definitely appear to be doing so.

Third, spending declined in real terms over the first 10 years of retirement for 75% of households, with a median total reduction of 23% (i.e., approximately 2% per year). This has important implications when forecasting retirement spending in a financial plan. 

While it’s common to assume spending increases annually with inflation during retirement, reality is quite different. I’ve dubbed this effect the “retirement spending smile” in past research, if you want to learn more. I typically suggest running at least a second plan where the retirement need is assumed to grow at a rate that is 1% lower than inflation.  

This can free up some cash earlier in retirement when the retirees might be able to better enjoy it (e.g., to take that cruise … if anyone is actually taking cruises anymore).

Finally, the percentage of households that can fund their retirement consumption (i.e., have a funded ratio of 1 or greater) increases dramatically during the first 10 years of retirement, from 18% to 48%.  

This shift is largely a result of reductions in spending, especially among households with the lowest initial funded ratios, and suggests that households, at least in early retirement, attempt to “right-size” their spending to better align with available resources. So, the reason households appear to be spending less is that they need to do so in order to get their financial house in order.

Spending Before Retirement

Perhaps the most interesting group of households we analyzed were those who had more than enough wealth to cover pre-retirement spending (i.e., a funded ratio greater than 1). This group is especially relevant to financial planners, since they tend to have accumulated the most financial wealth and therefore are more likely to need help managing these assets and are more desirable as clients given there are more assets to manage.  

While these well-funded households are the minority (18% of the total), 29% had more wealth, in today’s dollars, 10 years after retirement, and 54% have funded ratios that increase (at least slightly) over the period of study. 

These well-funded households could increase consumption, but appear to choose not to do so. While this underconsumption would appear to be irrational (at least according to economists), there are likely other factors driving this behavior that cannot easily be observed in survey data, such as the desire to leave a bequest, uncertain medical expenses (especially late in retirement), uncertain life expectancy, etc.  

Understanding the financial behaviors of this group will be an important strand of literature going forward.

The Study’s Implications

Overall, what the analysis suggests to me is that any type of retirement consumption gap observed early in retirement can likely be explained by households attempting to match resources to available spending (i.e., “right-sizing” their consumption). 

I’m not sure to what extent this gap persists further into retirement (e.g., by year 15 or 20). It could be that households that are forced to “right-size” during the first 10 years of retirement continue to reduce spending even though they may not need to (i.e., the belt continues to tighten) and evidence of reduced spending by well-funded households suggests this may be the case (and it may explain other findings in the space). 

Regardless, the relatively poor funded status of most households suggests that pre-retirees should focus on saving more for retirement, although traditional approaches to estimating required retirement savings may be overstating the actual amount needed based on various observed behaviors (i.e., spending declines in real terms, and overfunded households tend to under-consume).


David Blanchett is head of retirement research for Morningstar Investment Management LLC. Views expressed are his own and do not necessarily reflect the views of Morningstar Investment Management LLC. This blog is provided for informational purposes only and should not be construed by any person as a solicitation to effect, or attempt to effect transactions in securities or the rendering of investment advice.