Retirement income: the value of deferring Social Security
In this article we are going to focus on one specific issue under current Social Security rules: the relative utility of deferring the commencement of a participant’s Social Security benefit to age 70. (To be clear: we are talking about the deferral of the commencement of Social Security, not the deferral of retirement.) We discuss this issue because it has implications for the design of DC/401(k) payout options and for participant retirement planning.
The strategy of deferring the commencement of retirement has gotten some attention recently. Leading commentators in this area include economists John B. Shoven (Stanford University) and Sita N. Slavov (American Enterprise Institute), who have published a number of papers on the issue (see, for instance, their booklet Efficient Retirement Design, March 2013).
Shoven and Slavov discuss a number of techniques for increasing retirement income based on elections under current Social Security rules. We are going to focus just on the simple strategy of deferring commencement to age 70.
We begin with a brief review of Social Security rules for how a deferred benefit is calculated. We then discuss the relative value of deferring or not deferring and the key role of longevity and interest rates in that calculation. We conclude with a discussion of the significance of this analysis for plan design.
Before we begin our comparison, let’s briefly review Social Security’s rules on deferral. Generally, a covered individual may begin receiving (reduced) Social Security benefits as early as age 62. ‘Full’ benefits are available at the individual’s Social Security retirement age (SSRA). An individual’s SSRA depends on her year of birth, as follows:
|Year of Birth||SSRA|
|1937 or earlier||65|
|1938||65 and 2 months|
|1939||65 and 4 months|
|1940||65 and 6 months|
|1941||65 and 8 months|
|1942||65 and 10 months|
|1943 — 1954||66|
|1955||66 and 2 months|
|1956||66 and 4 months|
|1957||66 and 6 months|
|1958||66 and 8 months|
|1959||66 and 10 months|
|1960 and later||67|
In the rest of this article, to keep the discussion as simple as possible, we will use an individual born in 1952 contemplating retirement at age 62 in 2014. And we will only be looking at the situation of a single individual – while the analysis still generally holds for couples, spouse and survivor benefits complicate it significantly. Our example is assumed to earn $50,000 in 2013 and, based on a full earnings history, has a ‘Primary Insurance Amount’ (PIA) under Social Security – i.e. the ‘full retirement’ benefit – of $20,515.
Credit for delay
If our example individual commences Social Security prior to ‘full’ retirement age (age 66 in our example), the benefit is reduced, and if commencement is deferred beyond age 66, the benefit is increased up to age 70. The table below summarizes the ‘early/late retirement’ adjustment factors applicable to our sample participant:
|% of PIA
Our example individual can (a) commence Social Security today and receive $15,386 per year ($20,515 x 75%, equal to 31% of final pay), or (b) defer until age 70 and receive a benefit of $27,080 per year ($20,515 x 132%, equal to 54% of final pay), or (c) commence at an intermediate age with an intermediate benefit. Note that the benefit at age 70 is 76% greater than at age 62 (note: in all cases, benefits are indexed for inflation after age 62.)
The value of deferring
When an individual delays receipt of Social Security, payments now are, in effect, exchanged for larger payments in the future. Whether this is a ‘good deal’ for any given individual depends on two variables: how long the individual lives and interest rates.
The significance of longevity here is relatively intuitive. For instance, if the participant dies at age 69, under the age 62 strategy she will get 7 years of payments; under the age 70 strategy she will get nothing. On the other hand, if the individual lives to, e.g., age 90, the age 70 strategy, with its higher payments, is more valuable. So, the strategy we are discussing here, and annuitization generally, are unattractive for those with a short life expectancy.
The significance of interest rates is a little less intuitive. Consider: the lower interest rates are, the more expensive annuities are. That is, at a 1% interest rate $100,000 buys a smaller annuity (e.g., from an insurance company) than it would at a 3% interest rate. Individuals who delay Social Security commencement to age 70 are, in effect, buying a bigger annuity with the payments they are foregoing during the period age 62-70. Interest rates have gone down a lot in recent years, but the Social Security ‘early/late retirement’ adjustment factors are fixed – the age 70 benefit is still 76% higher than the age 62 annuity. The lower interest rates are, the more valuable deferring Social Security is. When interest rates are low enough, deferring to age 70 begins to look like arbitrage – it buys an increased annuity at a rate much more favorable than the market.
Thus, Shoven and Slavlov argue that:
Quantifying the value of deferring
As we mentioned, the strategy of deferring Social Security may not make sense for those with a short life expectancy. In addition, confident investors who are optimistic about asset returns may prefer not to defer. For most of us, however, deferring looks like a compelling value.
Insurer prices give us some insight into this value. Here’s the math1 – in today’s environment, an age 62 annuity, indexed to CPI, in the amount of $15,386 per year, costs about $325,000 for a male, while a deferred to 70 annuity of $27,080, also indexed to CPI, costs about $345,000. In other words, for a male with a standard life expectancy, deferring Social Security to age 70, versus age 62, is worth $20,000 today.
For women, the value of deferring Social Security is more dramatic, due to greater life expectancy. The age 62 indexed annuity costs about $350,000 while the deferred to 70 annuity costs about $400,000, so the value to a 62 year-old woman with a standard life expectancy of deferring Social Security until age 70 is $50,000 today.
In our example, it’s hard to overstate how meaningful an additional $50,000 in retirement wealth is. If interest rates keep moving up, this value will decline, but in the current environment, this is a deal that is impossible for commercial annuities to beat.
Which is to say: for those contemplating retirement and interested in annuitizing a portion of their retirement wealth, deferring Social Security may be the beginning and end of the story.
The value of an additional annuity
Our second article in this series discussed how annuities can be an effective tool that may be useful to 401(k) plan participants to manage risk in a way that will increase their retirement income. Most obviously, annuities are a hedge against longevity risk. Thus, some participants will want to use their 401(k) account to buy annuities.
At current interest rates, however, annuities don’t provide a lot of ‘bang for the buck.’ Rather than annuitizing retirement wealth, participants can get a much better deal by spending down retirement assets and deferring Social Security.
Why does this matter to sponsors?
We’ve gone through this extended comparison of the relative value of an age 70 (vs. an age 62) strategy to a plan participant because it has implications for the design of payout options under a 401(k) plan. Admittedly, these are difficult issues to understand, but there is an effort underway to educate 401(k) plan participants about the utility of delaying the start of Social Security. That effort dovetails with the more general effort to encourage participants to consider buying an annuity to provide for at least some piece of their retirement income.
As a result of these efforts, some participants may want to pursue a payout strategy of delaying the start of Social Security, taking larger 401(k) payments during the period age 62-70 and smaller payments after age 70. For participants who want to ‘self-design’ such a strategy, sponsors may want to make available tools that will aid in that project. Moreover – given the advantages that delay provides – sponsors may want to make available information about the relative benefits of deferring.
We would also note that, to the extent that ‘delay’ catches on as a good strategy, sponsors that do not help participants figure out a way to bridge the years between retirement and a (delayed) age-70 Social Security commencement may find some participants reluctant to retire.
As we stated, our analysis is simplified; probably most significantly, we only analyzed the relative value of deferral for a single individual. The Shoven and Slavlov analysis considers a number of other issues that come into play when Social Security is deferred, especially the rules applicable to two- and one-earner couples and the role spouse and survivor benefits may play in a deferral strategy. Those interested can find their booklet here.
1Current annuity pricing information comes from http://incomesolutions.com/AnnuityCalculator.aspx and http://www.immediateannuities.com/deferred-income-annuities.