The “4% rule” with annuities (and/or bond ladders)

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
StillGoing
Posts: 360
Joined: Mon Nov 04, 2019 3:43 am
Location: U.K.

The “4% rule” with annuities (and/or bond ladders)

Post by StillGoing »

There may be some here who are interested in a working study (i.e., not peer reviewed) of how the inclusion of annuities affects the portfolio (or income) longevity using the so called “4% rule”

Stocker, Alan, Using Lifetime Annuities to Increase Retirement Portfolio Survivability (November 30, 2022). Available at SSRN: https://ssrn.com/abstract=4289339 or http://dx.doi.org/10.2139/ssrn.4289339

For those who have neither time nor inclination to read through the paper, here are the headlines:

The number of years, Y it took for retirement income derived from the combination of an investment portfolio and a lifetime annuity (level annuities and those with 3% escalation and inflation-linked were tested) to fall below a constant, inflation-linked, target income was investigated using historical asset returns for retirees in the UK and US.

There was a threshold value RT in the annuity payout rate, AR. When AR was greater than RT the purchase of the annuity increased Y compared to the case where no annuity was purchased, and vice versa.

For inflation linked annuities (while these may no longer be available in the US, they remain available in the UK), the threshold was the same as the target income. For example, for the 4% target used for the US, the annuity payout rate needed to be greater than 4% and for the UK where a target just over 3% was used, the payout rate needed to exceed 3% to increase Y.

For nominal annuities the value of RT was different for the two countries and depended on which annuity options were used (i.e., level or 3% escalation). For example, for US retirees with a target income of 4% of the initial portfolio value, RT was approximately 7.3% for level and 5.6% for 3% escalation annuities purchased at the start of retirement.

RT was higher when the purchase of the annuity was delayed for D years after the start of retirement since, in the worst historical retirements, the reduction in real portfolio value was greater than the increase in annuity payout rate with age.

Two approaches were tried to reduce RT following a delay in the purchase of the annuity.
  • Inflation linked bonds (ILB), e.g. TIPS, with a maturity of D purchased at the start of retirement were then used to purchase the annuity.
  • In addition to using an ILB with maturity D to purchase the annuity, a ladder of ILB was constructed to provide the target income for the delay period D.
For example, for US retirees with a target income of 4%, purchasing a level annuity after a delay, D of 10 years, and assuming an ILB yield to maturity of 1%, RT was 13.3%, 9.6%, and 9.0% for the portfolio, ILB held to maturity, and ILB ladder approaches, respectively.


For those who would prefer to consider an inflation linked bond ladder rather than an annuity, the following study might be of interest

Stocker, Alan, Using Inflation Linked Bond Ladders to Increase Retirement Portfolio Survivability (October 24, 2022). Available at SSRN: https://ssrn.com/abstract=4256534 or http://dx.doi.org/10.2139/ssrn.4256534

which found that portfolio (income) longevity depended on bond yields (no surprise there), ladder duration (ladders of between 5 and 30 years were tested), and the proportion of the initial portfolio spent on constructing the ladder.

cheers
StillGoing
Post Reply