2013 Conference

Financial Services

Financial Services

Previous research on planned retirement ages has excluded workers who say they will never retire. This issue is important both in financial planning and in assessments of retirement adequacy.  Using the 1995, 1998, 2001, 2004, 2007, and 2010 Survey of Consumer Finances datasets, we present descriptive analyses, a logistic regression on the head's plan to never retire, and a multinominal logistic regression comparing the head's plan to never retire to each of three categories of planned retirement age. (In this draft, the multivariate analyses include only the 1995-2007 datasets, but 2010 will be added for the conference presentation.) For the 1995 to 2007 SCF datasets, 14% to 16% of full-time employed household heads age 35 to 70 stated they would never retire, but in 2010 this proportion increased to 19%.  Based on the logistic regression results for “never retire” versus any specific planned retirement age, the likelihood of planning to never retire decreases with current age, with education, and with net worth, and did not vary substantially between 1995 and 2007.  The multinominal logit results are similar but more complex because of multiple comparisons, but the main difference between the “never retire” and specific retirement age over 70 workers are on life expectancy and current age, with both results indicating higher levels related to being more likely to state specific age over 70 than to state “never retire.” For comparisons between the” never retire” response and specific ages 70 and younger (before 62 and 62 to 70), higher net worth, especially at a younger current age, was related to being less likely to state “never retire,” and a more positive assessment of retirement adequacy was related to being less likely to state “never retire.”  These results have implications for retirement planning and for assessments of retirement adequacy, as the usual assumption is that planned retirement age is exogenous.