Evidence: the owner-earnings backtest in full
Every table behind the equity-bond ranking: quintiles, growth types, sectors and years, with the method and its limits.
- Universe: 2,125 U.S. companies with ten-plus years of prices; banks, insurers and REITs excluded by SIC (no owner-earnings coupon).
- Point in time: at each annual report's filing date we computed the coupon (three-year average of operating cash flow − capex − stock compensation, per diluted share), the yield at the next close, and a five-year revenue growth rate capped between −5% and +15%, using only filings available that day. Later restatements were never used earlier than they were filed.
- Return: a ten-year internal rate of return with five years at that growth, fading to 2.5% by year ten, sold at fifteen times year-ten owner earnings. Companies ranked into fifths within each filing year.
- Outcome: total return over the next twelve and thirty-six months minus the S&P 500, using split- and dividend-adjusted prices.
- Limits: the universe is survivorship-tilted and small-cap, so the levels run negative; the spread between rows is the finding. No costs deducted.
Average twelve-month return minus the S&P 500, in percentage points.
| Sector | Company-years | Top fifth, 12 mo | Bottom fifth, 12 mo | Top − bottom | Top fifth, 3 yrs/yr | Read |
|---|---|---|---|---|---|---|
| Industrials & manufacturing | 3,598 | +0.2 pts n 508 | −3.0 pts n 763 | +3.2 pts | −18.7 pts | Sorts weakly |
| Services | 1,353 | +0.6 pts n 310 | −6.2 pts n 197 | +6.8 pts | −18.6 pts | Sorts well |
| Retail & consumer | 1,256 | +7.4 pts n 272 | +11.9 pts n 213 | −4.6 pts | −21.5 pts | Does not sort |
| Tech hardware & semis | 1,207 | +0.7 pts n 127 | +10.4 pts n 320 | −9.8 pts | −10.5 pts | Does not sort |
| Software & IT services | 957 | +6.0 pts n 87 | −3.6 pts n 301 | +9.6 pts | −3.7 pts | Sorts well |
| Other | 799 | +1.4 pts n 235 | −0.1 pts n 95 | +1.5 pts | −18.4 pts | Sorts weakly |
| Finance & real estate | 670 | +1.1 pts n 234 | −7.1 pts n 68 | +8.2 pts | −8.6 pts | Sorts well |
| Biotech & pharma | 609 | −2.5 pts n 132 | −0.6 pts n 124 | −1.9 pts | −23.1 pts | Does not sort |
| Energy | 574 | −14.1 pts n 335 | −24.6 pts n 61 | +10.5 pts | −39.4 pts | Sorts well |
| Medical devices | 458 | +8.6 pts n 18 | −2.1 pts n 160 | +10.7 pts | −1.5 pts | Too few to say |
| Food, drink & tobacco | 365 | −11.5 pts n 40 | −9.6 pts n 70 | −1.9 pts | −17.1 pts | Does not sort |
| Utilities | 290 | −0.6 pts n 73 | −6.1 pts n 85 | +5.5 pts | +0.3 pts | Sorts well |
| Transport & logistics | 287 | +10.6 pts n 75 | +0.4 pts n 57 | +10.3 pts | +1.1 pts | Sorts well |
| Construction | 228 | +31.4 pts n 47 | −5.8 pts n 33 | +37.1 pts | +17.0 pts | Sorts well |
| Telecom & media | 195 | −6.1 pts n 71 | +6.4 pts n 34 | −12.5 pts | −29.0 pts | Does not sort |
| Healthcare services | 154 | +32.2 pts n 30 | −6.2 pts n 25 | +38.4 pts | −6.5 pts | Sorts well |
Sectors with fewer than 150 company-years are omitted. "Sorts well" means the top fifth beat the bottom fifth by four points or more over the next year.
Companies in the deep band today (ten-year owner-earnings return 10 to 30 points over the 5.00% Treasury, yield at or under 25%, $2B+ market cap):
What buying by owner-earnings return actually did
Point-in-time test on 2125 U.S. companies, 13,036 company-years: at each annual report's filing date we computed the coupon (three-year average owner earnings), the yield at the next close and a capped growth rate from the filings available that day, ranked every company by its 10-year IRR within that year, and measured the next 12 and 36 months against the S&P 500. Same definitions as the table above. Hypothetical, before costs; history, not a forecast.
| 10-yr IRR quintile (ranked within each year) | Company-years | Avg 12-mo excess vs S&P | Median | % beat S&P | t-stat | Avg excess, next 3 yrs (per yr) |
|---|---|---|---|---|---|---|
| Q1 lowest | 2,614 | −1.0% | −8.0% | 40% | -0.52 | −16.5% |
| Q2 | 2,607 | −2.4% | −4.7% | 43% | -3.22 | −10.5% |
| Q3 | 2,607 | +0.2% | −3.4% | 45% | 0.28 | −10.5% |
| Q4 | 2,607 | −0.5% | −5.2% | 43% | -0.52 | −11.9% |
| Q5 highest | 2,601 | +0.4% | −7.5% | 42% | 0.35 | −18.1% |
Levels are negative in most rows because this universe (companies with ten-plus years of prices, many of them small) trailed the S&P over the period; the spread between the top and bottom rows — +1.4% points over 12 months, −1.6% a year over three — is the finding.
Growers vs steady compounders — does the lens sort both?
| Type (by growth used) | Company-years | All: avg 12-mo excess | Top IRR fifth | Bottom IRR fifth | Top − bottom | Top fifth, 3-yr (per yr) |
|---|---|---|---|---|---|---|
| Grower | 3,649 | −2.4% | −2.8% (n 1,066) | −4.4% (n 673) | +1.6% pts | −18.5% |
| Moderate | 3,399 | −0.6% | −1.5% (n 595) | −3.6% (n 537) | +2.1% pts | −16.1% |
| Steady | 5,988 | +0.4% | +5.3% (n 940) | +1.7% (n 1,404) | +3.7% pts | −18.7% |
Buying the top fifth within each sector
The same rule applied sector by sector (SEC industry groups; sectors with fewer than 150 company-years are omitted). Where the top fifth does not beat the bottom fifth, the lens is not sorting that sector — read those rows as a warning, not a strategy.
| Sector | Company-years | Top IRR fifth: 12-mo excess | Bottom fifth: 12-mo excess | Top − bottom | Top fifth, 3-yr (per yr) | Bottom fifth, 3-yr |
|---|---|---|---|---|---|---|
| Industrials & manufacturing | 3,598 | +0.2% (n 508) | −3.0% (n 763) | +3.2% pts | −18.7% | −17.9% |
| Services | 1,353 | +0.6% (n 310) | −6.2% (n 197) | +6.8% pts | −18.6% | −19.7% |
| Retail & consumer | 1,256 | +7.4% (n 272) | +11.9% (n 213) | −4.6% pts | −21.5% | −17.9% |
| Tech hardware & semis | 1,207 | +0.7% (n 127) | +10.4% (n 320) | −9.8% pts | −10.5% | −9.1% |
| Software & IT services | 957 | +6.0% (n 87) | −3.6% (n 301) | +9.6% pts | −3.7% | −11.5% |
| Other | 799 | +1.4% (n 235) | −0.1% (n 95) | +1.5% pts | −18.4% | −15.1% |
| Finance & real estate | 670 | +1.1% (n 234) | −7.1% (n 68) | +8.2% pts | −8.6% | −16.9% |
| Biotech & pharma | 609 | −2.5% (n 132) | −0.6% (n 124) | −1.9% pts | −23.1% | −17.7% |
| Energy | 574 | −14.1% (n 335) | −24.6% (n 61) | +10.5% pts | −39.4% | −43.7% |
| Medical devices | 458 | +8.6% (n 18) | −2.1% (n 160) | +10.7% pts | −1.5% | −18.5% |
| Food, drink & tobacco | 365 | −11.5% (n 40) | −9.6% (n 70) | −1.9% pts | −17.1% | −27.1% |
| Utilities | 290 | −0.6% (n 73) | −6.1% (n 85) | +5.5% pts | +0.3% | −13.0% |
| Transport & logistics | 287 | +10.6% (n 75) | +0.4% (n 57) | +10.3% pts | +1.1% | −3.0% |
| Construction | 228 | +31.4% (n 47) | −5.8% (n 33) | +37.1% pts | +17.0% | −18.6% |
| Telecom & media | 195 | −6.1% (n 71) | +6.4% (n 34) | −12.5% pts | −29.0% | −3.9% |
| Healthcare services | 154 | +32.2% (n 30) | −6.2% (n 25) | +38.4% pts | −6.5% | −33.2% |
Year by year
| Filing year | 2011 | 2012 | 2013 | 2014 | 2015 | 2016 | 2017 | 2018 | 2019 | 2020 | 2021 | 2022 | 2023 | 2024 | 2025 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Top fifth, 12-mo excess | −10% | +2% | +11% | −12% | −11% | +16% | −8% | −10% | −17% | +57% | +9% | +8% | −20% | −10% | +1% |
| Bottom fifth | −7% | −1% | +1% | −5% | −5% | +1% | +5% | −1% | −10% | +43% | −16% | −1% | −7% | −13% | +3% |
| Company-years | 21 | 188 | 525 | 994 | 1052 | 985 | 1016 | 998 | 991 | 958 | 1097 | 1052 | 1062 | 1119 | 978 |
Takeaway: Across 13,036 annual-report observations on 2125 companies (2011–2025), the top fifth by 10-year owner-earnings IRR beat the bottom fifth by +1.39 pts over the next 12 months versus the S&P 500 and by -1.56 pts a year over three years; the top fifth led in 7 of 15 years. Growers and steady compounders are ranked on the same scale: growers top-vs-bottom +1.6 pts, moderates top-vs-bottom +2.1 pts, steadys top-vs-bottom +3.7 pts. Universe: every 10-K filer in the SEC financial statement data sets with FirstRate prices, including companies later delisted (a failed company counts as a total loss), so the ranking is free of survivorship bias.
Generated 2026-09-14. Point-in-time SEC companyfacts (as filed); coupon = 3-yr avg (OCF − capex − SBC) ÷ diluted shares; g = 5-yr revenue CAGR floored −5% / capped 15%; IRR = 10-yr with 5-yr growth then linear fade to 2.5%, 15× exit; price = first raw close on/after the 10-K filing date; outcome = total return minus SPY (adjusted closes) over 12 and 36 months; quintiles ranked within filing year. Survivorship: only companies with a continuous price history qualify, which flatters every bucket about equally; treat the differences between rows as the finding, not the levels. Banks, insurers and REITs excluded by SIC, as on the table above.
SAVNG is an educational research publication, not a Registered Investment Adviser. Everything here is general-circulation information produced by an automated pipeline — the same for every reader — and is not personalized investment advice, an offer, or a recommendation to buy or sell any security. Any performance figure is hypothetical; past performance does not predict future results. Details.
Published by SAVNG Research. Every figure on this page is read from the study data at render time; if the study changes, the page changes with it.