As the leading edge of Baby Boomers enter retirement, the savings of this generation of 50-to-70 year olds is very much on the minds of economists and policymakers. When economists study consumption patterns, they frequently use a simple life cycle model where people save during their working years and then spend down during their retirement. While this seems reasonable on face, people in general do not spend down their assets as fast as these models would predict.

Researchers have suggested the slow spend-down rate is because people want to leave money for their offspring, for fear of high unexpected expenses, or due to uncertainty in remaining life. New research by economists John Ameriks of the Vanguard Group, Joseph Briggs and Andrew Caplin of New York University, Minjoon Lee and Matthew Shapiro of the University of Michigan, and Christopher Tonetti of Stanford University looks at spending in retirement using a very interesting new data set and highlights the important role that health has in retirement spending.

These papers are important because they are using a detailed new data set to study identify what influences spending in retirement. In the first of the group’s two papers, the authors describe the Vanguard Research Initiative, an exciting new data set and survey of a sample of 9,000 Vanguard clients over the age of 55. Beginning in 2013, these data include detailed information about demographics, health, income, and assets as well as questions about decision making.

The second paper from the group looks at how people in or near retirement spend their money. Using the new Vanguard data, the six economists find that much of the consumption trends among people in retirement are strongly associated with the perceived likelihood for the need of long-term care. Healthy retirees tend to say that they want to leave money for family members. However, as people decline in health, they spend less in anticipation of high long-term care costs. Thus, the authors’ research implies that access to a public long-term care option such as Medicaid reduces the precautionary savings among low-income retirees. Intriguingly, though, the survey component of the research reveals that respondents tend not to have a favorable opinion of public care options compared to private ones.

These findings indicate that savings decisions for retirement are more complex than the simple models that economists frequently use and also that public policy can have an impact. Because inherited wealth is a major factor associated with growing inequality and there are concerns about who on the income ladder has the means to save adequately for retirement, the data and analysis produced by this research team are an important contribution to our understanding of wealth dynamics. This work further highlights the importance of good data in economic thinking.