Between 2000 and 2020, the real interest rate on 20-year US government bonds fell from 4% to 0%. During that period, a balanced retirement portfolio roughly doubled in nominal value. By every conventional measure — account balance, annual return, Sharpe ratio — investors did well. By the one measure that matters — can this portfolio buy a real retirement income? — many investors went backwards. The cost of funding a 25-year inflation-adjusted spending stream nearly doubled. Portfolios grew. Funded ratios shrank. This is what interest rates actually do to your retirement, and almost nobody tracks it.
The Rate Nobody Watches
The FIRE community is obsessed with returns. Annual return, compound return, inflation-adjusted return. Every portfolio update on Reddit includes a chart showing the growth of $10,000 over time. The number goes up. This feels like progress.
But there is another number — one that moves in the opposite direction — that has more power over your retirement outcome than any return you will ever earn. It is the real interest rate: the rate at which you can lock in a guaranteed inflation-adjusted income stream.
When the real rate is 4%, locking in $40,000 per year of real retirement income for 25 years costs approximately $640,000 (the present value of a 25-year real annuity at 4%). When the real rate is 0%, the same $40,000 per year costs approximately $1,000,000. The income is the same. The price doubled.
This is not a theoretical construct. It is the market price of retirement. TIPS — Treasury Inflation-Protected Securities — deliver this rate directly. The yield on a 20-year TIPS is the real rate at which you can buy a known real cash flow at a known future date. When that yield falls, the cost of every future dollar of real spending rises.
The Two Decades That Broke Retirement Math
From January 2000 to January 2020:
| Metric | 2000 | 2020 | Change |
|---|---|---|---|
| 20-year TIPS real yield | ~4.0% | ~0.3% | −3.7% |
| Cost of $40k/yr real income (25 years) | ~$640k | ~$980k | +53% |
| S&P 500 (nominal, total return) | 1,469 | 3,258 | +122% |
| 60/40 portfolio (nominal) | — | — | ~+140% |
| CPI (cumulative) | — | — | +50% |
An investor who started 2000 with exactly enough to fund a $40,000/yr real retirement income — a funded ratio of 100% — and held a 60/40 portfolio would have seen their account balance more than double. In nominal terms, they got richer.
But the cost of their liability rose by 53%. After adjusting for both the portfolio's growth and the liability's growth, their funded ratio went from 100% to approximately 90–95% — despite strong positive returns.
If they held a more aggressive portfolio (say 80/20), returns were higher but so was volatility. Through the 2000–2002 and 2008–2009 drawdowns, withdrawals during negative return years created a sequence-of-returns drag that further eroded the funded ratio.
The investor who felt wealthiest in 2020 — "my portfolio doubled!" — may have been less able to fund their retirement than they were in 2000. The number went up. The purchasing power over lifetime income went down. The mechanism was interest rates.
The Real Interest Rate Beta (βᵣ)
In the Campbell-Viceira framework, the sensitivity of an asset's real return to changes in the real interest rate is called the real-rate beta (βᵣ). It measures how the asset behaves when real rates move.
Your retirement spending liability has a large positive real-rate beta. When real rates fall, the cost of your liability rises. This is the same as saying your liability has long duration — it extends 15–25 years into the future, and lower discount rates increase its present value.
To hedge this, you need assets that also rise when real rates fall — assets with positive βᵣ. Specifically, you need assets whose βᵣ approximately matches the βᵣ of your liability.
| Asset Class | Real-Rate Beta (βᵣ) | Implication |
|---|---|---|
| Long-duration TIPS (10–25 yr) | +0.8 to +1.0 | Closely matches liability. When rates fall, TIPS prices rise nearly as much as the liability grows. This is the hedge. |
| Long nominal bonds (TLT) | +0.7 to +0.9 | Rises with falling rates, but also carries inflation risk (βπ negative). A partial hedge at best. |
| Intermediate bonds (BND) | +0.3 to +0.5 | Some protection, but duration too short. Matches maybe 30% of the liability's rate sensitivity. |
| Broad equities | −0.1 to +0.1 | Near zero. Equities respond to growth expectations more than real rates. Essentially unhedged against rate moves. |
| Cash / money market | −0.8 to −1.0 | Opposite direction. When real rates fall, your reinvestment rate falls too. Cash increases the gap between assets and liability. Maximum reinvestment risk. |
| REITs | −0.2 to −0.4 | Net negative. Operating income benefits from low rates but leverage refinancing risk and cap rate sensitivity dominate. |
| Short-term TIPS (VTIP) | +0.1 to +0.2 | Barely positive. Duration too short to match liability. Better than cash, far worse than long TIPS. |
The pattern should now be unmistakable. Your liability has a real-rate beta somewhere around +0.8 to +1.0 (long duration, real spending). The typical FIRE portfolio — heavy equities, some intermediate bonds, maybe cash — has a portfolio-weighted βᵣ near zero.
The gap is the entire liability. When real rates move, your portfolio barely responds while the cost of your retirement swings dramatically.
Cash: The Wolf in Sheep's Clothing
Post 5 identified cash as dangerous because of its zero duration. Now the mechanism is precise.
Cash has a real-rate beta of approximately −0.9. This means: when real rates fall by 1%, the real return you earn on cash falls by 0.9%. Meanwhile, the cost of your liability rises by approximately 15% (for a 15-year duration liability experiencing a 1% rate decline).
The investor sitting in cash "waiting for a better entry point" is not being conservative. They are making the largest possible bet that real rates will rise. If rates stay flat, they earn nothing and the liability drifts higher. If rates fall further, they lose on both sides: lower reinvestment income and higher liability cost.
This is what happened to the generation of investors who held cash from 2009 to 2020 waiting for "rates to normalise." Real rates went from 2% to 0%. The cost of a real retirement income rose by roughly 30%. Cash returned approximately 0% real. Their funded ratio declined steadily for a decade — and they never saw a drawdown. Their accounts had the same balance. They just could afford less retirement.
In the funded-ratio framework, this is clearly visible. In the conventional framework — where success is measured by account balance — it is invisible.
Brandon Ganch — the Mad Fientist — built an influential FIRE career around tax optimisation: Roth conversion ladders, HSA strategies, tax-loss harvesting. His tax analysis is excellent and I have no quarrel with any of it. But here is the uncomfortable arithmetic: the best tax strategy in the FIRE toolkit saves perhaps 0.3–0.5% per year in effective tax drag. A 1% move in real rates — which happens routinely — changes your funded ratio by 12+ percentage points on a typical unhedged portfolio. The Mad Fientist's readers are optimising a 0.3% variable while ignoring a 12% variable. They are sharpening the knife while the house is on fire. Tax optimisation is valuable. But doing it on a portfolio that has a 15-year unhedged duration gap is optimising the wrong thing — or, more precisely, the right thing at the wrong level of the hierarchy.
The 2022 Reversal: A Natural Experiment
In 2022, the Federal Reserve raised rates aggressively. The 20-year TIPS yield went from roughly 0% to approximately 2% in under a year. Stock and bond portfolios dropped by 15–30%.
By every traditional measure, 2022 was a catastrophe. "Worst year for bonds in decades." "60/40 is dead."
By the funded-ratio measure, 2022 was the best year for unfunded retirees in two decades.
The cost of a $40,000/yr real income stream fell from ~$1,000,000 to approximately $770,000 — a 23% decline. If your portfolio fell 20% but your liability fell 23%, your funded ratio improved. You got poorer but you could afford more retirement.
The entire financial media told retirees to panic. The funded ratio said: this is the buying opportunity you've been waiting for — real income just got 23% cheaper.
This is not contrarianism. It is arithmetic. The funded ratio measures what matters — the ratio of assets to the cost of the income you need. When rates rise, both sides fall, but the liability falls more (because it has longer duration). The ratio improves.
This is precisely what liability-driven pension funds experienced in 2022. Their assets fell. Their liabilities fell more. Their funded ratios improved. Some went from underfunded to fully funded for the first time in a decade. This was not luck — it was the mechanical consequence of having long-duration assets that moved with the liability.
How to Measure Your Rate Exposure
The diagnostic is the same as for inflation in the previous post. For each asset, multiply its weight by its real-rate beta. Sum the results.
Example — Typical FIRE portfolio:
| Holding | Weight | βᵣ | Contribution |
|---|---|---|---|
| VTI | 70% | 0.0 | 0.00 |
| VXUS | 20% | 0.0 | 0.00 |
| BND | 10% | +0.4 | +0.04 |
| Portfolio total | +0.04 | ||
| Liability | +0.9 | ||
| Gap | −0.86 |
Your portfolio's βᵣ is +0.04. Your liability's βᵣ is +0.9. The gap is −0.86 — meaning 96% of your rate risk is unhedged. A 1% fall in real rates decreases your funded ratio by approximately 13 percentage points.
Same portfolio with 30% long TIPS:
| Holding | Weight | βᵣ | Contribution |
|---|---|---|---|
| VTI | 50% | 0.0 | 0.00 |
| VXUS | 15% | 0.0 | 0.00 |
| TIPS (long, individual) | 30% | +0.9 | +0.27 |
| BND | 5% | +0.4 | +0.02 |
| Portfolio total | +0.29 | ||
| Liability | +0.9 | ||
| Gap | −0.61 |
The gap narrows from −0.86 to −0.61 — a 29% improvement. Still substantial, but the 30% TIPS allocation now absorbs nearly a third of the rate risk.
To fully close the gap, you would need a portfolio βᵣ of +0.9. That implies roughly 100% of the bond allocation in long-duration TIPS — which is exactly what liability-matching pension funds do. The trade-off: lower expected return (giving up the credit spread and equity premium on that portion) in exchange for certainty that rate moves won't destroy your funded ratio.
The Three-Beta Dashboard
With Posts 4 (bad beta), 6 (inflation beta), and this post (real-rate beta), you now have the three numbers that actually describe your portfolio's risk relative to your retirement:
| Factor | What It Measures | Your Goal |
|---|---|---|
| Bad beta | Exposure to permanent loss | Minimise (mostly broad index, minimal concentration) |
| Inflation beta (βπ) | Sensitivity to unexpected inflation | Close to zero (TIPS replace nominal bonds) |
| Real-rate beta (βᵣ) | Sensitivity to real interest rate moves | Match the liability (~+0.9 for long retirement) |
These three numbers — not "asset allocation," not "risk tolerance," not "years to retirement" — are what determine whether your portfolio can fund your consumption. Pension funds have tracked them for decades. The retirement industry has never shown them to an individual investor.
In the capstone post of this series, I'll show you how to calculate your funded ratio — and give you a tool that does it in 60 seconds. Subscribe so you don't miss the launch.
This is Post 7 of The Funded Ratio — a series on amifunded.com applying pension fund mathematics to individual retirement portfolios. Built on the Campbell-Viceira (2002) intertemporal asset allocation framework, the lifecycle consumption theory (Merton, 1969; 1971), and the real-rate dynamics analysed in Campbell, Sunderam & Viceira (2017). No financial advice is given. All methods use publicly available academic research and market data.