Big Spring Capital

Dave Ramsey’s 12%, illustrated

For decades, Dave Ramsey has told his audience to expect 12% a year from the stock market and to plan their retirement around it.

The number is not invented – it comes from a real average of the S&P 500’s real yearly returns. But an average return and a consistent return are very different things…and the difference between them can compound into millions of dollars.

Where the 12% comes from

Ramsey Solutions publishes the number, and its source, in an article they actively update on their website:

“The historical average annual return of the S&P 500 is 10–12%. … The historical average annual return from 1928 through 2025 is 11.86%.”

Ramsey Solutions, Can You Really Get a 12% Return on Your Investments?

That 11.86% is an arithmetic average: take each of the 98 yearly returns from 1928 through 2025, add them up, divide by 98. The number is real and accurate.

The trouble is what it gets used for.

An average describes the past. Typed into a retirement calculator as a growth rate, it becomes a promise that your money will grow that much every year – and money does not grow at the average of its returns.

It grows at the compound rate, and over those same 98 years the S&P 500’s compound rate was 10.02%.

A difference of 1.84 percentage points might seem small. But compounded across a lifetime of investing, it creates two totally different universes.

A simple example of the problem with averages

Take a two-year investment of $1,000. Year one it gains 50%, year two it loses 50%.

YearReturn$1,000 becomes
Year 1+50%$1,500
Year 2−50%$750
Average return implies0%$1,000
Actual result−13.4% a year$750

The average says you broke even. But based on the real dollars in your account, you’re actually down 25%.

The sequence is irrelevant – reverse the order and you still end at $750. The damage comes from the simple math of losing and recovering.

Down 50% needs +100% to get back to even, not +50%.

The same math, across 98 real years

Since 1928, the S&P 500 has had 26 losing years.

Each one opened this same kind of gap between the average return and the actual account balance…and across the full 98 years, the gaps compound into something enormous.

If $1,000 invested in the S&P 500 at the start of 1928 had actually grown 11.86% every year, it would have become $58.9 million by the end of 2025.

But growing at the returns the index actually posted (including reinvested dividends), the same $1,000 became $11.6 million.

At a consistent 11.86% every single year$58.9Mwhat the average implies
versus
At the actual yearly returns$11.6Mwhat the index delivered

The average implies 5.09× more money than the market actually produced.

Growing at a consistent 11.86%Growing at the actual returns
For the first several decades the two lines look almost identical – the gap does nearly all of its damage late. The shaded area is money the average implied and the market never produced. Hover any year for both values.

Check it yourself

To see that this gap is not merely an artifact of a super-long date range, pick any starting investment amount and any span of years between 1928 and 2025.

The S&P 500’s average return and its compound return (which determines the actual account balance) recompute, and the chart above updates.

Any time you have a down year (a loss), the average will overstate the actual returns.

$
1928 or later
2025 or earlier
Projected at the average$58,902,24811.86% every year
Actually delivered$11,570,011the real year-by-year path
Average return11.86%arithmetic mean
Compound return10.02%what it actually grew at
Overstatement5.09×98 years, 19282025

The last thirty years

Ninety-eight years makes the math (and the chart) vivid, but nobody invests that long.

So take just the last thirty, 1996 through 2025. The S&P 500’s average over that stretch was 11.80%– almost exactly the number Ramsey Solutions advertises.

$1,000 invested at the start of 1996 and projected to grow consistently at 12% would have become $29,960. The actual returns produced $18,708.

At a consistent 12%$29,960what the projection promised
At the actual yearly returns$18,708what the index delivered

More than a third of the money predicted by the “average” never materialized.

And not because this was a time period where the market was way off the average – its average really was 11.80%. But because six of the thirty years lost money (2000, 2001, 2002, 2008, 2018, 2022), and every loss had to be recovered before growth could resume.

That is the whole difference between an average return and a consistent one.

If your plan assumes 12%

None of this is to say that the stock market is an empirically bad investment.

If you can stomach the volatility, the liquidity tax, and the “number go up” nature of pure equity investing, it’s an okay place to put your money.

But don’t run your retirement math on a straight-line 12% return – because getting that assumes a version of the stock market has never existed (and never will).

This is one of the major reasons we love cash-flowing real assets, like real estate and oil & gas. When returns come primarily from recurring cash flow rather than paper-price movement, consistency becomes possible – and a consistent return that never breaks compounding doesn’t require years of recovery just to get back to even.

Questions

This topic always invites plenty of good questions. Here are the most common ones I get:

Does anyone actually plan around the average?

Ramsey Solutions’ article states that “the historical average annual return from 1928 through 2025 is 11.86%.” That 11.86% is the arithmetic mean of the period, not its compound rate.

A professional building a projection uses a compound figure. The number Ramsey Solutions hands its audience is the average, and it is the one that ends up typed into retirement calculators as a growth rate.

Does this include dividends?

Yes. Every figure here is S&P 500 total return with dividends reinvested, from the NYU Stern dataset maintained by Aswath Damodaran – the same source and the same 1928–2025 window behind the 11.86% figure.

Price return without dividends would make the actual path materially worse, not better.

Isn’t this just sequence-of-returns risk?

No. Sequence of returns risk is about the order returns arrive in, and it mostly bites once you are withdrawing.

The gap between an arithmetic mean and a compound rate exists regardless of order, and regardless of whether you ever withdraw a dollar. Shuffle the 98 years into any sequence you like and the ending value is identical. The +50% / −50% example above works in either order.

If only the starting and ending values matter, isn’t the compound rate the only number anyone needed?

Yes. $1,000 to $11.6M across 98 years is 10.02% a year, and that is the complete description of what the money did.

Which is exactly why 11.86% is the wrong number to hand someone planning a retirement.

Doesn’t diversification solve this?

Partly. Combining genuinely uncorrelated assets lowers a portfolio’s volatility, and lower volatility narrows the gap between its average and its compound return. The gap cannot close completely while any volatility remains, but it shrinks.

Which is an argument for holding assets that do not move together – though plenty of what gets sold as diversification is not. Public REITs run a 0.68 correlation to the stock market, while private real estate runs 0.14.

Is Dave Ramsey wrong about everything, then?

No. His debt advice has gotten an enormous number of people out of a hole, and the behavioral discipline behind it is genuinely valuable.

Return assumptions are a different skill from debt-payoff discipline, and this is the specific place where the advice does real damage – because an inflated growth assumption does not feel like a mistake until you are thirty years into it.

So what do I do with this?

One thing: run your own plan at the compound number and see if it still gets you where you need to go. For a lot of people the answer is yes, and they can stop reading here.

If the answer is no, the useful question is which part of the assumption to change. I spend my time on assets whose return comes mostly from recurring cash flow rather than price movement, because a return you collect is not one you have to recover. Check out my newsletter and podcast (linked below) where I cover these aspects in more detail.

Sources

Show the full table, 1928–2025
YearReturn$1,000 actualAt 11.86%
1928+43.81%$1,438$1,119
19298.30%$1,319$1,251
193025.12%$987$1,400
193143.84%$555$1,566
19328.64%$507$1,751
1933+49.98%$760$1,959
19341.19%$751$2,191
1935+46.74%$1,102$2,451
1936+31.94%$1,454$2,742
193735.34%$940$3,067
1938+29.28%$1,215$3,431
19391.10%$1,202$3,838
194010.67%$1,074$4,293
194112.77%$936$4,802
1942+19.17%$1,116$5,372
1943+25.06%$1,396$6,009
1944+19.03%$1,661$6,722
1945+35.82%$2,256$7,519
19468.43%$2,066$8,410
1947+5.20%$2,174$9,408
1948+5.70%$2,297$10,524
1949+18.30%$2,718$11,772
1950+30.81%$3,555$13,168
1951+23.68%$4,397$14,730
1952+18.15%$5,195$16,477
19531.21%$5,132$18,431
1954+52.56%$7,830$20,617
1955+32.60%$10,382$23,062
1956+7.44%$11,155$25,797
195710.46%$9,988$28,857
1958+43.72%$14,355$32,279
1959+12.06%$16,086$36,107
1960+0.34%$16,141$40,390
1961+26.64%$20,441$45,180
19628.81%$18,640$50,538
1963+22.61%$22,854$56,532
1964+16.42%$26,607$63,237
1965+12.40%$29,906$70,736
19669.97%$26,925$79,126
1967+23.80%$33,333$88,510
1968+10.81%$36,936$99,007
19698.24%$33,892$110,750
1970+3.56%$35,099$123,885
1971+14.22%$40,090$138,577
1972+18.76%$47,611$155,012
197314.31%$40,798$173,397
197425.90%$30,231$193,962
1975+37.00%$41,417$216,966
1976+23.83%$51,286$242,698
19776.98%$47,706$271,482
1978+6.51%$50,812$303,680
1979+18.52%$60,223$339,696
1980+31.74%$79,337$379,984
19814.70%$75,608$425,050
1982+20.42%$91,048$475,461
1983+22.34%$111,388$531,851
1984+6.15%$118,238$594,928
1985+31.24%$155,176$665,487
1986+18.49%$183,868$744,413
1987+5.81%$194,550$832,701
1988+16.54%$226,729$931,459
1989+31.48%$298,103$1,041,930
19903.06%$288,981$1,165,503
1991+30.23%$376,340$1,303,732
1992+7.49%$404,528$1,458,354
1993+9.97%$444,859$1,631,315
1994+1.33%$450,776$1,824,789
1995+37.20%$618,465$2,041,209
1996+22.68%$758,733$2,283,296
1997+33.10%$1,009,873$2,554,095
1998+28.34%$1,296,071$2,857,011
1999+20.89%$1,566,820$3,195,852
20009.03%$1,425,337$3,574,880
200111.85%$1,256,434$3,998,861
200221.97%$980,396$4,473,126
2003+28.36%$1,258,436$5,003,639
2004+10.74%$1,393,592$5,597,070
2005+4.83%$1,460,902$6,260,883
2006+15.61%$1,688,949$7,003,423
2007+5.48%$1,781,504$7,834,029
200836.55%$1,130,364$8,763,145
2009+25.94%$1,423,580$9,802,454
2010+14.82%$1,634,555$10,965,025
2011+2.10%$1,668,881$12,265,477
2012+15.89%$1,934,066$13,720,163
2013+32.15%$2,555,868$15,347,374
2014+13.52%$2,901,421$17,167,573
2015+1.38%$2,941,461$19,203,647
2016+11.77%$3,287,671$21,481,200
2017+21.61%$3,998,137$24,028,870
20184.23%$3,829,015$26,878,694
2019+31.21%$5,024,051$30,066,507
2020+18.02%$5,929,385$33,632,395
2021+28.47%$7,617,481$37,621,197
202218.04%$6,243,288$42,083,071
2023+26.06%$7,870,288$47,074,123
2024+24.88%$9,828,416$52,657,114
2025+17.72%$11,570,011$58,902,248
Return data: Annual Returns on Stock, T.Bonds and T.Bills, 1928–Current, Aswath Damodaran, NYU Stern School of Business.The 12% claim: Can You Really Get a 12% Return on Your Investments?, Ramsey Solutions.

More like this

A newsletter and a podcast for people who would rather own things that pay them than watch numbers on a screen.