5 Key Points
- Nominal Safety vs. Real Safety: A 5% U.S. Treasury guarantee protects your principal dollars, but leaves your future purchasing power completely exposed to inflation.
- The 30% Equity Engine: Adding just a 30% equity allocation to an all-bond portfolio boosted sustainable retirement spending by over 50% in historical models.
- Software Blind Spots: Most financial planning tools fix inflation at a static rate (e.g., 2.5%) across all Monte Carlo trials, masking real-world stagflation risks.
- Bonds Can Silently Lose Half Their Value: From 1940 to 1980, long Treasuries delivered a -1.74% annual real return, cutting purchasing power in half despite zero defaults.
- Multi-Regime Stress-Testing: Off-the-shelf probability scores fall short for extreme allocations; true planning requires testing against rate shocks, sequence risk, and persistent inflation.
A client challenged me recently, and it sent me down a rabbit hole I’m still thinking about.
The question was straightforward: why not lock in an entire portfolio with 30-year Treasuries yielding a bit over 5% and live off the income? Paraphrasing a lot, he asked: "Sean, why would I take any equity risk at all? I can lock in 5% for thirty years, guaranteed by the U.S. government. Run it through the planning software. I bet it works."
So I did. And the software agreed with him: high probability of success, smooth projections, and plenty of money left over at the end. Yet I felt unsettled by the results and told him I needed to dig deeper into the projections before recommending a final spending target.
The result bothered me all weekend because I knew something was off. I’ve done previous analyses of 100% Treasury portfolios over historical periods, and they failed miserably against real purchasing power (here is one comparing the cost of a 6 pack to Treasury bond yields over time). Yet here it was, mapped out in my financial planning software, looking pristine.
I decided to rebuild the analysis from scratch using primary data: actual Treasury returns, actual CPI inflation, and actual equity performance against inflation-adjusted spending. What I found is worth sharing, because the "lock in 5%" instinct is one of the most seductive, and expensive, mistakes an investor can make.
The instinct isn't crazy – inflation is just tricky
Let's start by being fair to the idea. A 30-year Treasury really does pay you a fixed coupon, and the federal government really will hand you back your principal at maturity. In nominal terms (actual dollars, there's essentially no default risk. That part is true.
The problem is that "dollars" and "purchasing power" are two very different things, and only one of them buys groceries.
Here's the trap in one sentence: a 5% coupon locked in for 30 years is only a good deal if inflation behaves consistently predictable for the entire 30 years. You've frozen your income. You have not frozen your cost of living. And when you fix one side of that equation while the other floats, you've quietly taken on one of the largest risks in all of finance and yet somehow managed to relabel it as "safe."
What actually happened to a "safe" bond portfolio: 1940-1980
I don't want to argue or try to explain using theory, so let's use history. Consider an investor who put money into a rolling portfolio of long (10-year) U.S. Treasuries at the start of 1940 and reinvested every coupon for the next 41 years. This isn't a cherry-picked crash; it's four straight decades of actual history.
This data was compiled YOY but summarized here as a starting and ending point to spare you all the detail.
Read that bottom row again. For 41 years, a portfolio that never missed a coupon and never defaulted still lost about half its purchasing power. The nominal balance nearly tripled, and yet the investor got quietly, steadily poorer the entire time.
And it wasn't one bad patch dragging down an otherwise fine stretch. Every non-overlapping decade told the same story:
Three of four decades were clearly negative in real terms; the fourth was a rounding error above zero. Both 20-year sub-periods (1940-1959 and 1960-1979) were negative too.
Now, the obvious objection: "That's ancient history." Fair. So let's look at the recent stuff.
- 2014-2023: The 10-year Treasury returned roughly 1.5%/yr nominal while inflation ran about 2.8%, a (1.3%) real return for a full decade, within the "modern" era everyone considers normal.
- 2022 alone: Long Treasuries had their worst year in modern record, down ~10.6% nominal against 8% inflation, roughly an (18.6%) real loss in a single year. Long-dated zero-coupon Treasuries fell around 39%.
The uncomfortable result: 1940 began from historically low starting yields, which is the same setup we've had in the 2020s. Buying a long bond today locks in today's yield regardless of what inflation does for the next thirty years.
Putting real dollars on it: the million dollar question
Abstract percentages don't land. So let's make it concrete with a round number, a $10 million portfolio, and ask the question that actually matters to anyone living off their money: how much can I safely spend per year? This is the “die with zero” number.
I built a model that applies actual historical bond returns and actual CPI year by year, subtracts taxes and spending, and solves for the maximum sustainable first-year spending that still leaves roughly $3 million (in future dollars) at the end – that’s as close to “die with zero” and I can ever recommend. I ran it two ways: a) 100% long Treasuries vs. b) a 70%/30% Treasury/S&P 500 mix with annual rebalancing and looked at it across two eras.
First, adding a modest 30% equity sleeve raised sustainable annual spending by roughly $167,000, about 55% more income, using the recent, friendly 40-year window. Same starting money, same ending target, one allocation change.
Secondly (and somewhat interestingly), the conclusion barely changes even in one of the worst bond stretches in history. In the 1940-1980 run, the 70/30 mix still out-spent the all-Treasury portfolio by about $169,000 a year. I thought there would have been an even worse result that only $2K/year lower spending.
There's another lesson: the all-Treasury investor's sustainable spending dropped 15% (from $302K to $258K) just by changing which 40 years they lived through. That swing is the inflation risk, a the true inflation risk that supposedly wasn't there.
(Fair-warning caveat: this model deliberately simplifies. It holds bond principal flat rather than modeling price swings from rate moves, and it uses a flat tax rate and a flat 2% dividend yield. It's built to show the directional impact of allocation and the real-world volatility of inflation — not to be a to-the-penny forecast. The gaps between the columns are the story, not the exact figures.)
So why did the planning software say the bond plan was fine?
If history is this clear, why did a professional-grade planning platform give the 100% Treasury plan a cheerful thumbs-up? The answer lies in the underlying methodology used to evaluate success.
RightCapital (RC for short) tests the risk of plans by running the plan through 1,000 trial iterations (a Monty Carlo simulation), factoring varying market returns and volatility into each trial, and reports the percentage of those trials that don't run out of money. A thousand various futures! It seems like it must capture every possible outcome.
You already know there is a catch so let’s start pealing back the layers of the onion. The key is what actually varies across those futures and, more importantly, what doesn't.
Issue #1: Inflation doesn't vary at all.
In RC, inflation is a single deterministic number (2.5% by default but it is constantly updated and can be updated manually if desired) applied identically across all 1,000 trials. This is a big problem. Market returns get randomized; inflation does not. A 1970s-style stagflation regime (the exact scenario that destroyed real bond returns for 40 years), literally cannot appear in any trial. The single most dangerous risk for a bond-heavy portfolio is switched off before the simulation starts.
Note: While I’m naming RightCaptial here because it’s the software I am most familiar with, this critique seems to apply to most of the major professional planning software used by advisors as documented by the advisor support organization Kitches.com and fellow planner’s reviews online. I also personally know several advisors that don’t use any planning software and only rely on personally constructed excel documents which could be even more susceptible this and who know what other methodology issues.
Issue #2: The assumptions are built on the recent past.
By the platform's own documentation, the inflation rate used and asset return assumptions were derived from roughly the past 40 years of data, one of the great disinflationary bull markets for bonds in history. The 1970s aren't in the training data. You're shown a future that resembles the friendliest stretch bonds and inflation have ever had.
Issue #3: The distribution has thin tails.
The default engine uses a "Standard" model leaning on a log-normal distribution, more simply put: what we think of a standard bell curve. But real markets have "fat tails," or a distribution where extreme, rare events happen more often than they would under a normal distribution The 2008 crisis was a 4-5 standard-deviation event the model would assign a near-zero probability. 2022's bond rout was similarly "impossible" by that math, yet it happened.
Issue #4: Every year is drawn independently.
These engines assume stationary parameters over 30-year horizons; each year is rolled fresh like independent dice. But inflationary periods (or regimes) are persistent. The 1970s were one continuous condition, not ten unlucky independent years that took a significant structural change to fix (The fix was President Carter sacrificing his political career and entire legacy by appointing Paul Volcker to fed chair and let him make independent fed decisions). Continuous conditions cannot be replicated by randomized independent draws, so they understate the odds of a long, bad run.
You could attempt to adjust for these issues by putting in overly conservative buffers, like bumping the inflation assumption to 3.5-4%, lowering the bond return input, or looking at the stress-test module but each of those are blunt tools. If we make each of these numbers overly conservative, we could just end up with a plan that says you can never retire.
The best way to minimize the issues identified above, in planning or in the real world, is to have a higher equity allocation in the portfolio. Equity returns have proven time and time again to beat inflation over any material duration. The equity portion of portfolio will provide inflation protection but will only mask the bond portfolio issue. There are cases where we want or need a very high level of bond allocation so we need a way to look at that risk properly.
Stress-testing the way it should be done
Once I stopped trusting a single rosy projection, I rebuilt the spending analysis through eight lenses, each relaxing an unrealistic assumption (thanks to a pretty powerful AI planning platform, Hazel). Instead of one number, you get ranges of numbers which then you can measure against risk tolerance.
The value in any single row isn’t the “correct” answer but looking at the pattern is how we can figure out what is. Rows 2-3 fix how returns are shaped and whether bad years clump. Rows 4-5 isolate sequence risk. Rows 6-7 switch on the inflation and rate risks the standard engine switches off, precisely the two risks a long-bond investor is most exposed to.
When you can say "we ran your plan through eight models, including the ones the software can't produce, and here's the number that survives all of them," you're having a far more honest conversation than "the probability of success is 94%."
Okay, no bonds, no planning software right?
Nothing I said above should be taken as an argument against bonds or using planning software. They both have flaws but we should also not ignore why they are also valuable.
On Bonds
Bonds are essential for near-term spending (and mid-term in early retirement), dampening volatility, and protecting capital you cannot afford to see fluctuate. Caution just needs to be used when considering extreme concentration and ensure your decisions are not mistaking nominal yield for purchasing power.
- Nominal safety and real safety are entirely different. For a treasury bond, the government guarantee protects your nominal principal, not your purchasing power. Over a 40-year retirement, purchasing power is all that matters.
- Treasury Inflation Protected Securities (TIPS) could be another potential solution but they are better utilized to match a known spending number, not solving for a max spending question. The longest length is also 30 years so a longer retirement would introduce our inflation risk question back into the plan.
- A little equity diversification does heavy lifting. Moving from 0% to 30% equities lifted sustainable spending by roughly a third in both favorable and hostile inflation eras.
- Match your bonds to their purpose. Use fixed income for short-term liquidity and stability, not as a 30-year purchasing-power growth engine.
The 5% coupon feels like certainty. But certainty about the number of dollars you'll receive is not the same as certainty about what those dollars will buy.
On Planning Software
RightCapital or any other planning software is not the bad guy either. Financial software provides very important structure for planning, ensuring that nothing is missed and we don’t build one-off spreadsheets looking for confirmation biases. Further, most planning software will provide solid numbers for most households. However, we need to be cognizant of two items:
- Treat any single probability-of-success number with suspicion, especially for plans that are outside the ordinary (such as extremely bond-heavy, or extremely stock heavy plans).
- If it’s an outlier plan, think about what the biggest risk is. Is it accounted for in the model? We have discussed here the risk inflation variability poses to an extreme bond portfolio, but an equally extreme equity allocation may be more susceptible to things like Sequence of Return Risk (SORR) than the planning software is able to measure.
- Stress-test explicitly for identified risks in outlier plans.
- Once the risk is identified, find a method to quantify the risk. For example, if the plan is extremely equity concentrated, we may want to more heavily weigh the regimes like “Lost Decade First” in formulating the headline number.
At Purpose Built, our job as your personal CFO isn't to hand you a glossy 50-page software report with a 99% probability of success and send you on your way. It’s to look under the hood, identify the blind spots, and stress-test your family's wealth against the real-world conditions that standard software defaults turn off.
True financial security doesn't come from hiding in a 5% nominal yield and hoping inflation stays quiet, it comes from building a resilient, multi-asset portfolio engineered to protect your purchasing power through every economic regime. Being educated about possible risks ensure you are prepared for them.
If you're ready for a second opinion for possible blind spots in your current plan, contact Purpose Built today to schedule a strategy session and if we are a good fit. Let's make sure your wealth is built for reality, not a simulation.
Frequently Asked Questions (FAQ)
Q: Why did my financial planning software give a 100% Treasury portfolio a 90%+ probability of success?
A: Most professional planning software treats inflation as a fixed, static number across every simulated trial while randomizing investment returns. Because persistent stagflation regimes (like the 1970s) are mathematically excluded from the simulation, the model fails to capture the true purchasing-power erosion that a 100% bond portfolio faces.
Q: Isn't a 5% guaranteed Treasury yield safer than risking money in the stock market?
A: It depends on how you define "safety". A Treasury bond carries virtually zero default risk in nominal dollars, but it carries immense purchasing-power risk. If inflation averages higher than expected over 30 years, your fixed income stays the same while your grocery, healthcare, and living costs double or triple.
Q: Can’t I just use Treasury Inflation-Protected Securities (TIPS) instead of standard Treasuries?
A: TIPS are an excellent tool for matching known, fixed spending needs. However, the longest maturity available is 30 years, which reintroduces reinvestment and inflation risk for longer retirement horizons. Additionally, TIPS are designed to match inflation, not to maximize sustainable spending across a multi-decade retirement.
Q: How much stock exposure do I actually need to protect my spending against inflation?
A: You don't need an overly aggressive portfolio to see a massive impact. Historical data shows that introducing just a 30% equity sleeve (a 70/30 bond/stock mix) significantly improved portfolio survival and boosted sustainable income by over 50% compared to a 100% Treasury allocation.
Q: Does this mean I should avoid Treasury bonds altogether?
A: Not at all. Treasuries play a vital role in providing short-term liquidity, dampening portfolio volatility, and funding near-term spending needs. The danger lies in extreme concentration - using fixed income as a 30-year growth engine rather than a short-term stability anchor.
Final Thoughts
A 5% guaranteed coupon feels like the ultimate financial security blanket… until inflation quietly eats away at what those dollars can actually buy. True risk management isn't about eliminating volatility at all costs; it’s about choosing which risks are worth taking. Trading temporary market fluctuations for permanent loss of purchasing power is a trade no long-term investor should make. By combining short-term fixed income for immediate stability with a disciplined equity engine for growth, you build a retirement strategy engineered to survive real-world economic regimes - not just a friendly software simulation.
If you want to see how your portfolio and retirement plan stack up to risks you may not have thought about, reach out to Purpose Built today to see if we would be a good fit to help you reach your financial independence goals.
About the Author
Sean Lovison, CPA, CFP®, is a fee-only financial planner and founder of Purpose Built Financial Services. After spending 14 years as a corporate chief financial officer (CFO), receiving and designing compensation plans, he decided to help others navigate their plans.
Purpose Built Financial Services is a state registered (PA and NJ) advisor but legally able to virtually serve high-earning households across America from New Jersey to California to Alaska (yes, we have a client there!). While we specialize in the unique tax complexities of the NJ/NY/PA tri-state corridor, our 'Personal CFO' model is designed for tech leaders nationwide who require sophisticated equity and tax coordination.
All written content on this site is for information purposes only and is general in nature. Opinions expressed herein are solely those of Sean Lovison and Purpose Built Financial Services (PBFS), unless otherwise specifically cited. The material presented is believed to be from reliable sources, and no representations are made by our firm regarding other parties' informational accuracy or completeness. All information or ideas provided should be discussed in detail with an advisor, accountant, or legal counsel prior to implementation.
The information on this site is provided "AS IS" and without warranties of any kind, either express or implied. To the fullest extent permissible pursuant to applicable laws, PBFS disclaims all warranties, express or implied, including, but not limited to, implied warranties of merchantability, non-infringement, and suitability for a particular purpose. PBFS does not warrant that the information will be free from error. None of the information provided on this website is intended as investment, tax, accounting, or legal advice, as an offer or solicitation of an offer to buy or sell, or as an endorsement of any company, security, fund, or other securities or non-securities offering. The information should not be relied upon for purposes of transacting securities or other investments. Your use of the information is at your sole risk. Under no circumstances shall PB be liable for any direct, indirect, special, or consequential damages that result from the use of, or the inability to use, the materials on this site, even if PB or a PB-authorized representative has been advised of the possibility of such damages. In no event shall Purpose Built have any liability to you for damages, losses, and causes of action for accessing this site. Information on this website should not be considered a solicitation to buy, an offer to sell, or a recommendation of any security in any jurisdiction where such offer, solicitation, or recommendation would be unlawful or unauthorized.
.jpg)
