One way to evaluate retirement savings is to look at a retirement savings to income multiple. Taking a household’s total retirement savings and dividing by typical household income is a simple measure of retirement readiness. The retirement savings to income multiple combined with age is, like BMI, a good summary statistic for looking at a household’s preparedness for retirement.
What Retirement Savings to Income Multiple Should You Have by Age?
Fidelity released a Viewpoint editorial which attempted to encapsulate a number of factors into a multiple recommendation. For age 30, 35, 40, 45, 50, 55, 60, and 67 they came up with a target multiple for retirement savings:
| Your Age | Fidelity Recommended Retirement Savings to Income Multiple |
| 30 | 1x |
| 35 | 2x |
| 40 | 3x |
| 45 | 4x |
| 50 | 6x |
| 55 | 7x |
| 60 | 8x |
| 67 | 10x |
Basically, the guidance states that you should have your typical salary stashed away by age 30, scaling to ten times your typical salary by age 67.
Calculating your own retirement savings to income multiple? Importantly, the multiple is based on savings that you solely earmark for your own retirement. Money you save for one-off spending, donations to charity, or college spending, for example, shouldn’t be included.
Fidelity notes there are a number of caveats that either lower or increase the targets. See their tool for more.
How Many Households Hit These Retirement Savings to Income Multiples?
We were curious as to the number of households in the United States which currently match the Fidelity benchmarks. Using 2016 SCF data with our home-built measure of retirement savings (that we call ‘expansive’) we found that households currently hitting the benchmarks are few and far between. Just 5.21% of 35 year old householders have 2x income saved in retirement accounts. 60-year old householders were most prepared – but only 7.59% had 8x typical income saved.
We use a more expansive definition of retirement savings than most sites. Notably, we include savings accounts, taxable brokerage accounts, and other saving vehicles. The data doesn’t include defined benefit pensions with periodic payments (such as Social Security) and we’ve left out real estate holdings.
Some households will downsize or otherwise tap primary home equity in retirement. Other households hold individual properties as an investment and will use recurring income or capital gains from those properties in retirement. Caveats in mind, if you’d like to expand the definition even further, we have a net worth by age calculator which includes all forms of assets.
What Multiples are Households Achieving?
As we looked at in the recent article on retirement savings by age, most households either have retirement accounts or some form of savings they can tap in retirement. After we ran the numbers, we were curious to what multiples households in America had actually achieved.
The following table lists our estimates of the multiple for median and 75th percentile households:
| Age | Fidelity Multiple | Median Household Multiple | 75th Household Multiple |
| 30 | 1 | 0.090 | 0.398 |
| 35 | 2 | 0.130 | 0.524 |
| 40 | 3 | 0.219 | 0.942 |
| 45 | 4 | 0.332 | 1.397 |
| 50 | 6 | 0.395 | 1.704 |
| 55 | 7 | 0.456 | 2.001 |
| 60 | 8 | 0.726 | 3.153 |
| 67 | 10 | 0.459 | 2.647 |
And here’s the gap visually:
Are the Retirement Savings to Income Multiple Suggestions Realistic?
If you are even a casual user of Twitter (follow me here!), you know that the Fidelity multiples went viral due to their “unrealistic expectations”. The 35-year old guidance of ‘2x salary’ drew particular ire on Twitter, with a multitude of humorous responses suggesting it is wildly out of reach. Seemingly, 35-year olds on Twitter have a shortage of retirement savings but no shortage of snark.
Here’s my personal favorite from the genre.
Snark aside, the numbers are very much in reach if you target them from your early career. 10% savings into >= 60% stock from age 23-35 would have covered most of today’s 35 year olds. (As an aside, a typical younger household should actually consider 80-100% of their allocation in risky assets.)
Sadly, that savings rate is more aspirational than descriptive for much of America… which is currently mired in the low-single digits. While the US savings rate covers all households, it’s more important to save while you’re young because of the time value of your investments.
If you aren’t yet saving 10%, you should make the sacrifices to get there. For more information, our personal finance basics article covers most of the framework.
Methodology on Retirement Savings to Income Multiples
The multiples themselves come from Fidelity, while estimates of retirement savings come from the 2016 SCF. Our extensive savings estimate methodology is detailed in our post on retirement savings in America.
There are too few samples to use a single age in the numbers. In each age group we added householders one year younger to one year older to the set. For example, ’30 year old householders’ includes 31 and 29 year old householders as well.
For the multiple, we used our tally of retirement savings versus the household’s reported typical annual income. Typical income is what a household expects to make in a typical year. This would eliminate (some) variation due to large capital gains from house sales, business sales, and the like.
It makes no sense to compare pre-retirement income to savings when considering readiness for retirement. It only makes sense to compare *expenses* to savings. If you’re going to lose your income and live off of your savings, what matters is the rate at which you are going to spend it once you retire and how long you’re going to need it to last.
I agree on ‘multiple of savings’ – but I put this measure in the same category as BMI. It’s useful on a population level, and it’s not too bad a guideline for a typical person earning on a W-2 with typical spending, saving, and tax-paying habits.
If you’ve got multiple streams of income, max retirement accounts, side businesses and real estate investments etc. it just doesn’t apply – but that demographic that will know their annual spending habits offhand. It’s much easier for someone to quote an income instead of calculating spending or even savings. Body fat percentage is similar; people who know it offhand are going to throw out BMI as worthless. (My BMI was 28.4 yesterday, ha – not worried)
Nice work. Interesting stuff.
If I understand it right, your “expansive” definition does not include business ownership or real estate does it?
I think both of those would tilt it some. ~10% of households have business equity and ~10% have real estate (other than primary home).
Also, social security is on top of all this right?
Thanks!
Yes, you nailed it – everything in NFIN in the flowchart is not included: http://sda.berkeley.edu/sdaweb/docs/scfcomb/NetworthFlowchart.pdf . It also doesn’t include vehicles (recreational or otherwise) and OTHNFIN ‘other’ assets. Papers I’ve read list things like racehorse shares, antiques, art, etc. in that category.
And yes, no Social Security. These calculations I’ve been doing lately really underscore how much Social Security is relied on by Americans. It’s a hugely important promise that many people will need to fill the gap.
I had seen the RAM idea tossed out there a few times, and while it works for higher incomes (say 150% of the median), at lower incomes, it does tend to be way more than you actually need to replace your income once transfer payments and OASDI are included. All the guesses in the world for how much you should have saved don’t really seem to show the complexity of the problem, because they inevitably, due to lack of data, cannot approximate the additional income from those sources.
When I look at these charts, I try to remember that 1, everyone isn’t screwed because social security, and 2, social security is hugely important to the standard of living of a good portion of this country. Could you imagine if social security were to have to make major cuts to benefits in the future? Because this chart really does a good job of showing what would happen without it around.
Absolutely agree (echoing my reply to Jim!). Doing this math really reveals the importance of Social Security to most folks’ retirement planning; even relatively well off households (e.g. the 75th percentile) has a liquid savings gap.
And +1 on the limitations of the data. I’m toying with switching over to the PSID and building a model for Social Security/periodic pension payments but I’ll probably spike on budget/investing/debts with the SCF next instead. Call that one ‘eventually’.
Holy moly, how did I not know this data set existed!?! Now I’m down a rabbit hole! The SCF is great for employed households, but this takes it to a whole new level!
Thank you!
It’s a good set and covers many of the blind spots from working with the SCF. Beware though the two aren’t directly comparable: https://psidonline.isr.umich.edu/publications/Papers/tsp/2014-03_%20Measuring_Wealth_and_Wealth_Inequality.pdf
It’ll probably be some time before I dig in (asset allocations, debts, and investment mixes are on my list right now) but if you see anything interesting that could use a treatment drop a comment or you can mail pk [at] dqydj [dot] com
Enough with the acronyms…how many people really talk this shop?
Haha, touché. I’ll fix that:
SCF – Survey of Consumer Finances, from the Federal Reserve: https://www.federalreserve.gov/econres/scfindex.htm
PSID – Panel Study of Income Dynamics, from the University of Michigan: https://psidonline.isr.umich.edu/
I\’ll throw mine on here too: RAM (Retirement Account Multiple) refers to the strategy of retirement planning described in the post. OASDI (Old Age, Survivors, and Disability Insurance) commonly referred to collectively as social security, but includes a ton of other important programs that fall under the same flat income tax.
Yes, this does underscore the urgent need for Social Security — AND Medicare! Republicans note: These are not entitlement programs. Working Americans have saved (directly through their paychecks) all their working lives for them!
“entitlement” means that you have a right to it. So by definition social security and medicare are entitlement programs. They aren’t welfare.
Social Security is definitely an entitlement program (albeit an important one) and not a savings program. The whole system is based on the premise that present working folks pay for present retirees (yes, there is the IOU “trust fund” that was created, but that’s not the core system).
Ida M. Fuller became the first person to receive an old-age monthly benefit check under the new Social Security law. She paid a total of $24.75 between 1937 and 1939 on an income of $2,484. Her first check, dated January 31, 1940 was for $22.54. She passed away in 1975 (100 years old) and had received a total of $22,888.92 in monthly payments from Social Security.
Things get tough if the population growth lags or people start living longer since you end up with more retirees being supported by the working population per capita.
Social security was absolutely set up for a growing population, and if the ratio of employed to retired workers shrinks, it puts the system in more of a bind. Currently the tax isn’t projected to be capable of keeping up with its obligations indefinitely. At some point, we will need to increase revenue to continue this level of payout. That could happen by either raising the (flat) tax, adding a second tier, or removing the “cap” in income above which the tax is no longer applied. If none of that happens, payout will have to be cut, and the hundred million people (give or take) that rely on the income will find themselves up a creek without a paddle, at which point, popular sentiment would almost certainly lean toward the confiscatory end of the spectrum to make up the shortfall.
I use the term “entitlement program” the way Republicans do: to imply welfare. We should have raised the cap on SS tax 20 years ago. (DQYDJ, could we see the figures on how much more this would add to SS coffers?) BTW, I love how people are quick to point out that my SS checks don’t actually come from the money I made, as though the government had been stuffing it under a mega mattress for years. I realize it’s paid for by young wage earners today. What folks always fail to point out is that I have already paid the SS benefits for our parents and grandparents! It is not fair now to deny it to my generation — or the next.