Initial bankruptcy filings among National Football League (NFL) players begin very soon after retirement and continue at a substantial rate through at least the first 12 years of retirement, according to a study published by the National Bureau for Economic Research (NBER).

Researchers from the California Institute of Technology, University of Washington and George Washington University set out to test how the life-cycle hypothesis applies in a group of people whose income profile does not just gradually rise then fall, as it does for most workers, but rather has a very large spike lasting comparatively few years. One of the central predictions of the life-cycle hypothesis is that individuals smooth consumption over their economic life cycle; thus, they save when income is high to provide for when income is likely to be low, such as after retirement.

A perfect test sample for this income spike is professional athletes, so the researchers used data for NFL players drafted by NFL teams from 1996 to 2003. They note that a career lasting six years (the median length) will provide an NFL player with more earnings than an average college graduate will get in an entire lifetime, plus a modest pension. Reduced income in retirement and the uncertainty of their career length are presumably known to players, so the researchers assume that to maintain a smooth level of consumption after the predictable post-NFL income drop, a rational, patient player should save a large portion of his NFL earnings and enter retirement with a high net worth.

However, the study found 1.9% of nearly 900 players in the sample filed bankruptcy by their second year of retirement, and 15.7% filed bankruptcy by their twelfth year. Using regression analysis, a very pessimistic prediction gives a bankruptcy rate of around 40% by 25 years after retirement.

The median level of earnings across all players is about $3.2 million (in year-2000 dollars). The study found players with longer careers have much greater earnings and opportunity to save for retirement, yet their bankruptcy rate during retirement is no lower than those with shorter careers and lower earnings. In addition, the bankruptcy rate of retired football players is in line with that of the general population of that age group (from early 20s to age 34).

Documents on the NFL Players Association website indicate NFL players are offered an annuity program, with payments beginning at the age of 35, or five years after his last credited season in the league (whichever is later), a savings plan in which players can defer income and teams may contribute a 2-for-1 match, and the Bert Bell/Pete Rozelle NFL Player Retirement Plan, which offered a monthly benefit of $470 per credited season in years 1998 through 2011 and $560 per credited season in years 2012 through 2014.

The researchers say future work will focus on what types of players have higher bankruptcy risk, perhaps indicating behavioral biases, correlates of predictably poor financial decision making, and social variables.

“This evidence should also inform helpful interventions, both for these athletes and for other workers with unusual income profiles,” they conclude.

The study report is available for purchase or a free download here.