Forced Retirement Risk and Portfolio Choice
Guodong Chen, Minjoon Lee, Tong-yob Nam
- Older Americans and Canadians face a significant risk of being forced to retire before their planned retirement ages. On average, every year, 4 percent of older American employees are forced to retire; 10 percent of older Canadian employees experience involuntary job separations.
- The chance of forced retirement is higher after stock market crashes, as witnessed during the Great Recession in 2008-2009 and once again during the current coronavirus crisis.
- The optimal portfolio adjustment in late life, under the forced retirement risk, is to have a relatively conservative portfolio before retirement and increase the share of risky assets after retirement. This is the opposite of conventional wisdom that suggests reducing the share of risky assets as individuals get older and enter retirement.
RESEARCH QUESTION AND CONTEXT
Conventional wisdom on late-in-life portfolio management is to reduce the share of risky assets as one gets older and enters retirement. Whether this is indeed an optimal reallocation strategy or not depends on the size and the nature of the labor market risk older workers face.
In this project, we focus on a specific form of late-in-life labor market risk that the literature has overlooked—the risk of being forced into retirement. Then we examine the implication of this risk on the optimal portfolio allocation strategy in late life.
OVERVIEW OF THE CURRENT STATE OF THE PROJECT
This project is composed of two parts: Part 1 based on the U.S. data and Part 2 based on the Canadian data. A complete working paper from Part 1 is revised and resubmitted to the Journal of Empirical Finance (impact factor: 1.244). We have recently started working on Part 2. In this executive summary and the industry report, we focus on the main findings from Part 1 and their implications. We will also describe some preliminary findings from Part 2.
1. Empirical Findings from the U.S. Data
Using the self-reported reason for retirement from the Health and Retirement Study (HRS) data, we find that older Americans face a significant risk of being forced to retire before their planned retirement ages. Workers in the age range 60-64 who want to continue work face on average 3-4 percent chance of being forced to retire every year.
We also find that the chance of being forced to retire goes up after the stock market crashes: For example, after the burst of the dotcom bubble in 2002 and after the Great Recession. The negative correlation between the forced retirement risk and the stock returns amplifies the implication of this risk on portfolio choice.
2. Findings from the life-cycle model
Using a structural life-cycle portfolio choice model, we find that the optimal strategy, under the estimated forced retirement risk, is the opposite of what conventional wisdom suggests. When faced with the forced retirement risk, older workers should increase the share of risky asset in their financial portfolio as they approach and enter retirement. Intuitively, while they are still working, they face a risk of being forced into retirement when the market goes down — recall what happened during the Great Recession and what is happening during the coronavirus crisis. When they retire, on the other hand, they are no longer exposed to the labor market risk, which gives them more room to invest in risky assets in their financial portfolio.
3. Implications of the Key Findings
The main findings from this project reveal that the common advice of financial advisors regarding late-in-life portfolio adjustment, as well as the design of target-dated funds, are based on invalid assumptions. A key premise behind conventional wisdom is that older workers do not face much labor income risk that is correlated with stock returns. Our findings suggest that that is not true, once we take into account the forced retirement risk. Financial advice, and target-dated funds, should acknowledge significant labor market risk faced by older workers and hence suggest a relatively more conservative portfolio before retirement than after retirement.
4. Preliminary Results from the Canadian Data
We use the indicator of involuntary job separations from the Labor Force Survey (LFS) data to examine the late-inlife labor market risk among Canadians. Our preliminary results from the LFS suggest that older Canadians face a significant risk of involuntary job separation. On average, about 10 percent of the workforce in the age group 60-64 experience an involuntary job loss per year. As expected, the risk of involuntary job separation increases after a large negative return on the S&P/TSX index, though the correlation is weaker than in the U.S. data.
We will update our empirical analysis using the LFS as data for the first half of 2020 becomes available. We anticipate that the pandemic had disproportionally impacted older workers when we also had a substantial loss in the stock market. We will then incorporate the estimated risk of involuntary job separations into the life-cycle portfolio choice model we used for the U.S. analysis. We want to examine whether the main message from the U.S. analysis — the optimal portfolio choice around retirement is to increase the stock share as one nears and enters retirement — holds for older Canadians.