Knowledge is power when it comes to retirement planning, Stanford researcher shows
A Stanford economics scholar found that people increase their savings when they use retirement income projections and have access to detailed information about the benefits of saving now for use later. When such information is absent, people save less.
A SIEPR study on retirement savings suggests people benefit from more information when deciding how much money to set aside.
People save more for retirement when they use detailed retirement income projections and information, according to new Stanford research.
"The findings suggest that people are not perfectly informed about the link between saving today and income in retirement," said economist Gopi Shah Goda, a senior research scholar at the Stanford Institute for Economic Policy Research. "Also, many overestimate the level of retirement income that current saving generates."
The research comes at a time when Congress is considering requiring retirement plan sponsors to provide participants of defined contribution plans with information about projected annual income at retirement. The study was summarized in a SIEPR policy brief.
"Many Americans must make choices about their retirement saving while faced with uncertainty about future rates of investment return, inflation, health status, income and other factors, and about how current contributions will translate into retirement income," Goda said.
Using a large-scale field experiment, the research involved randomizing university employees into different groups and providing one group with brochures that contained customized information about how contributions into retirement saving accounts translate into income in retirement.
Goda found that providing retirement income projections along with general retirement planning information induces individuals to increase their saving for retirement.
"We found that this low-cost intervention had a modest but statistically significant effect on retirement saving, increasing contributions by approximately 0.15 percent of income on average," Goda said.
People who were given the partial or full treatments of retirement projects were more likely to make changes in their contribution rates than those in the control group.
"This suggests that general information about retirement planning influenced behavior on this dimension," the study stated.
On the other hand, when information about the link between saving and retirement is absent, people save less.
The research also found that the assumptions used in generating the income projections – such as the expected retirement age and rate of return – are significant. "Those who received larger projections responded more to the brochures than those who received smaller projections," she said.
Overestimating and biases
Why do some people overestimate the level of retirement income needed?
"Linearized exponential growth bias," Goda explained, is a phenomenon that leads people to project future savings as though little or no interest were earned on future earnings. A growing body of literature has shown that this type of bias is pervasive and that biased individuals have lower levels of wealth and net worth, she added.
However, retirement projections contain a lot of numbers and charts – a presentation that not everyone is inclined to understand, Goda said.
"Many studies have shown that financial literacy is low and that many people struggle with basic financial concepts. Projecting retirement income requires complex calculations and several assumptions that can be challenging for even the financially sophisticated," she said.
There is also the issue of "present bias," otherwise known as procrastination. "The findings showed those who displayed procrastination tendencies were less likely to respond to our interventions," she said.
Goda's research colleagues include Colleen Manchester and Aaron Sojourner, both from the University of Minnesota.
Gopi Shah Goda, Stanford Institute for Economic Policy Research: (650) 736-0480, firstname.lastname@example.org
Clifton B. Parker, Stanford News Service: (650) 725-0224, email@example.com