EBRI Issue Brief

Individual Retirement Accounts: How Balances of Older Account Owners Change Over Time

Mar 21, 2019 31  pages


  • Assets held in individual retirement accounts (IRAs) have been growing rapidly, rising from $1.47 trillion in 1996 to $9.20 trillion in 2017. As a result of this growth, these assets have become a major share of all retirement assets held by retirees. Thus, how they are used is of great interest to those focused on retirement adequacy due to this use having a tremendous effect on Americans’ retirement.
  • This Issue Brief examines changes in IRA balances over a three-year period from end-of-year 2012 to end-of-year 2015.While three years is a rather brief time, it allows for evaluating whether individuals owning IRAs, particularly those near or in retirement, are drawing down the assets very quickly or are preserving or even growing them. This has become a key policy issue, as it helps shed light on whether Americans use their IRA assets in a way that allows them to maintain a standard of living throughout retirement beyond what any Social Security benefits provide. This is also the next evaluation to be done after the assessment of whether workers are accumulating sufficient assets in the individual account plan retirement system.
  • This study tracks Traditional and Roth IRAs that were in the EBRI IRA Database in 2012. Each of these accounts will be followed to see if they are maintained or closed. If maintained, the IRA’s balance changes from 2012 to the end of the study in 2015 are examined. The focus is on those around retirement ages — 60 years old or older. In 2012, the EBRI IRA Database had 19.8 million Traditional and Roth IRAs. Traditional IRAs amounted to 14.4 million, while Roth IRAs numbered 5.4 million.
  • Of the IRAs owned by those ages 60 or older, slightly over 80 percent still had positive balances after three years. Not until the owners reached ages 85 or older did a significantly higher percentage of accounts not have a positive balance (were not closed or depleted) after three years. There were only small differences between IRA types (Traditional vs. Roth) in the likelihood that balances were still positive after the three years. However, accounts with larger balances were more likely to still have positive balances.
  • If the IRA still had a positive balance in 2015 at the end of the study, just under a third (32.0 percent) had a balance less than what it was in 2012. However, Traditional IRAs were far more likely to have a lower balance than Roth IRAs. Specifically, 36.5 percent of the Traditional IRAs had lower balances in 2015, compared with 13.2 percent of the Roth IRAs.
  • Accounts owned by older individuals were more likely to have a decrease in their balances. For IRAs owned by those ages 60–64, 20.8 percent had a balance decrease. This percentage continued upward with the age of the IRA owner, reaching 63.3 percent for those ages 85 or older.
  • Almost half (45.0 percent) of accounts either went to zero — were depleted or closed — or decreased in size since 2012. However, this proportion rose sharply with age, going from 35.0 percent of accounts owned by those ages 60–64 to 76.9 percent for those owned by individuals ages 85 or older. Accounts owned by males had a somewhat higher likelihood of a decline at 44.6 percent vs. 41.9 percent for accounts owned by females.
  • Of all IRAs owned by those ages 60 or older, the median balance change over the period was an increase of 11.9 percent. Consequently, the balances of many IRAs owned by those ages 60 or older experienced fairly substantial increases even during a time when the owners were near or at retirement and could be expected to start spending down their assets.
  • Some of the factors that lead to larger balance changes include the IRA type, age of the account owner, the account balance size, and the number of annual withdrawals taken.
    • Roth IRAs were much more likely to have had positive account balance changes than Traditional IRAs: the 25th percentile balance change for Roth IRAs was an increase of 9.9 percent, compared with a decrease of 8.5 percent for Traditional IRAs.
    • Older account owners experienced lower balance increases than younger account owners near or in retirement: IRAs owned by those ages 85 or older decreased 5.9 percent over the period studied at the median, compared with 21.1 percent growth for IRAs owned by those ages 60–64.
  • Withdrawal behavior shows that those younger than age 71 who made a withdrawal took out a larger share of their accounts than those who were just at the beginning of the required minimum distribution age, where they were potentially only making a withdrawal because they were required to do so.
  • A comparison of these IRA results with a study of overall nonhousing wealth changes shows that the changes in IRA balances are in line with the changes in wealth for retirees over the three-year period.
  • Seeing such a large percentage of IRA balances increasing in this period is encouraging from the perspective of IRA owners not outliving their retirement assets. However, some may be doing too well, as they may be in position to spend more. While these results show what happened in this short time period, a longer time period may have different results. Thus, a further examination is warranted of effective behaviors and strategies for retirees to achieve the balance of not outliving their assets while at the same time maintaining the appropriate standard of living given the size of their nest egg.