Here is this week’s advice on how to get a $1 million retirement: an explanation of “15% or more” for retirement savings.
You should be saving 15-20% of your gross income for retirement. Why?
Simply put, 15% is roughly the savings rate needed to retire with a similar income after a 40 year career. 20% is even better because life happens. You may have trouble saving some years, the market may perform poorly for an extended period of time, and who knows what will happen with Social Security.
To illustrate this, I took median personal income data based on Census Bureau data, extrapolated it out over a 40-year career and took a look at what saving 10%, 15%, and 20% would provide in retirement income on top of the median Social Security benefit.
This model still works for radically different income levels because everything is based on percentages, but I wanted real data because people tend to earn much less when they are younger and that affects how much you’ll have when you retire.
|Age||Personal income||Savings at 10%||Savings at 15%||Savings at 20%|
All dollars are 2015 dollars.
What does retirement look like for those people?
It looks pretty good, but I wouldn’t want to be the person who only saved 10%. And yes, the 15% saver got to a $1M nest egg after 40 years of saving with only a median income.
Let’s look at a 4% safe withdrawal rate from retirement investments plus media.
|Median Social Security benefit||$16,020||$16,020||$16,020|
|4% retirement withdrawals||$26,744||$40,116||$53,488|
|Total retirement income||$42,764||$56,136||$69,508|
What can we conclude?
- 10% is just enough if Social Security benefits don’t go down, nothing seriously interrupts your retirement savings during your working years, and the market does pretty well. That is a lot of “ifs”.
- 15% is good for a solid retirement that would be sufficient even if Social Security benefits are significantly reduced. You can also survive a few bad years along the way.
- 20% is much safer. Not only could you survive without Social Security, but if the market does poorly over the coming decades, you aren’t totally screwed. If the market grows just 1% slower, the 20% model looks more like the 15% model. It might also let you retire better or earlier. Early retirement may not even be a choice. The median retirement age in the US is 62 and many of those retirements are due to health issues or inability to find work.
Understanding these numbers
Note that all dollars are 2015 dollars so you don’t need to think about “how much will $X be worth in 10, 20, 30, or 40 years?”.
This means that the nominal dollar amounts shown at age 65 here are likely much lower than they will be actually be in 40 years. If the inflation rate stays at about 2%, the actual value of the 15% portfolio would be about $2.2M, but since $2.2M would only have the value of $1M in 2015 dollars, it’s easier to just think about everything in 2015 dollars.
That’s also why this post uses a growth rate that includes the value-reducing effect of inflation (6% rather than 8% or something higher).
Is this pessimistic enough?
I tried to generate a “middle of the road” look at the future based on today’s numbers, but we have no way of knowing what the future growth of the markets is going to be. My point here isn’t that 15% or 20% is enough no matter what, but that a 10% savings rate is not really where you want to be.
Also bear in mind that while the 4% safe withdrawal rate historically works in the US, it is definitely optimistic. If applied on historical data from other developed countries, it ends up being much too high (you run out of money early). A more pessimistic model might use 3% or 3.5% instead.
- 6% post-inflation growth is assumed. The long-term historical average for the US stock market is about 7%. We use a lower number because you can’t expect a 7% return. Bonds return less than stocks and we have no way of knowing what the future performance of the stock market will be.
To be more specific, the 6% number is the median post-inflation CAGR across all 40 year periods on cFIREsim with 85% stocks, 15% bonds, 0.1% expenses, and annual rebalancing. Note that cFIREsim only uses large-cap US stocks for stocks and US Treasuries for bonds (a more diversified portfolio is usually recommended here). There is a spreadsheet link below if you want to try different rates of return.
- The income data is the average of the incomes for men and women roughly interpolated out to get numbers for every single year. This includes data from non-primary earners in two income households (e.g., parents who mostly stay at home) which lowers the numbers somewhat. Financial Samurai has a nice article on the data.
- Here’s my spreadsheet if anyone wants to look at the numbers or change any of the assumptions (e.g., rate of return or safe withdrawal rate).