A Pension Promise to Oneself

Abstract

Saving for retirement is not hopeless. Well-run DB pension plans provided retirement income for generations. When plans failed, it is because they broke the rules. The same applies to individuals. By understanding a set of rules on how much to save and how to invest and then sticking to those rules—that is, by making a pension promise to oneself—retirement income goals can be met. Fulfilling this promise requires more saving than most people are accustomed to.

With defined benefit (DB) pension plans on the ropes, defined contribution (DC) savings plans are the principal means by which Americans now accumulate assets for retirement. However, most DC plan participants do not save enough to retire. We show how they can.

We note that if the economic resources that enable a DB plan to make its lifelong payouts are instead given to a DC plan, the latter should have enough resources to make the same payouts. The trick is getting money into a DC plan when contributions are voluntary. By showing workers that they can retire on DB-like levels of income if they save a large percentage of their incomes, invest conservatively, and use longevity pooling, we can motivate them to make and keep a pension promise to themselves.

By assuming that investors can earn only the real riskless rate, which is currently zero, we can put an upper bound on the required savings level. We develop illustrations for three worker prototypes: a Columbus, Ohio, teacher, representing middle-income earners; a San Diego sanitation worker, representing lower-middle earners; and an Austin, Texas, software developer who later becomes an executive, representing upper-middle earners. We assume that in retirement, each worker needs 70% of final pay, inflating with the US Consumer Price Index.

We begin by subtracting each worker’s expected Social Security benefit from the 70% replacement income target. We multiply the resulting income requirement by a retirement multiple (RM), currently 21.47, to arrive at the savings target. We derive the RM as follows: At the current real riskless rate of return, which is zero, the desired consumption level from age 65 to age 85 is guaranteed by saving 20 years’ postretirement income (net of Social Security); consumption after age 85 is provided by a deferred income annuity (DIA). At age 65, the annuity is priced at 1.47 years’ income. The RM is the sum of the two numbers 20 and 1.47.

Because DIAs for which we were able to obtain price quotes provide a nominal, not real, benefit, inflation after age 85 is not hedged. If we could have found a real DIA, we would have “bought” it, increasing the RM slightly. A work-around is to buy a little more than the required amount of the DIA and if one lives beyond age 85, save some of the income benefit for future rather than present consumption.

We assume that the three workers start their 40-year careers at age 25 and save 10%, 14%, and 15% of income, respectively. We estimate their real salary growth using published data for the teacher and sanitation worker and our own estimates for the software developer. The teacher needs to save 28.4% of income in his 20th year of work and 32% in his 40th and final year of work. The sanitation worker needs to save 21.2% in the 20th year and 27.3% in the 40th year; the developer needs to save 21.2% in the 20th year and 19% in the 40th year.

Such high savings rates are completely consistent with the size of required DB plan contributions using realistic market return assumptions. To get employees up to such high rates, the Save More Tomorrow (a registered trademark of Shlomo Benartzi and Richard Thaler) program is a template. Saving 52%, 54%, and 25% of future real pay raises, respectively, achieves the desired overall savings. The savings rates needed to achieve an adequate payout are reached today in peak earnings years for the middle- and upper-income cohorts.

At higher (2% real) rates of return, the required savings rates are substantially lower but not trivial and still require much more saving than most DC plan participants are accustomed to.

Although not everyone will achieve these savings rates, almost everyone will retire. They will mostly do so by engaging in personal fiscal adjustments, adjustments to either production or consumption. One can work harder (e.g., by getting a second job), work smarter (by getting additional education or training), consume less before retirement, or consume less after retirement (e.g. by moving in with relatives). Pushing these levers is more effective than trying to increase one’s investment return because the latter often backfires and makes the investor worse off.