October 19, 2026 – Welcome back to a new installment of the Safe Withdrawal Rate Series. Please check the Series Landing Page. Today’s topic is an update on a post I published in 2024, Part 61, Safety First, where I examined whether a significant shift into (relatively) safe financial instruments and asset classes, single-premium immediate annuities (SPIAs) and inflation-protected Treasury bonds (TIPS), can improve my retirement safety. The verdict back then was mixed: For early retirees with long horizons, TIPS didn’t offer much security at the real rates prevailing in 2024, and annuities carry significant inflation risk. We’ve come a long way since 2024; TIPS real rates have risen significantly, and one of the most requested topics recently has been whether today’s impressive TIPS real rates, north of 3% for the 20- and 30-year maturities, warrant a second look.
Let’s get started…
What are TIPS?
Before we get started, let’s first make sure we’re all on the same page. TIPS stands for “Treasury Inflation-Protected Securities.” They are bonds issued by the US federal government that pay a guaranteed real interest rate, in contrast to regular Treasury bonds that pay nominal interest only. As a result, TIPS pay coupon payments, and the bond’s notional value rises with the Consumer Price Index (CPI). Future real interest payments also rise as the bond’s notional value increases, and at maturity you receive the (inflation-adjusted) notional value, i.e., the initial value adjusted by the US CPI.
Ex ante, with mostly efficient markets and intelligent market participants, there shouldn’t be any arbitrage opportunity. In other words, the yield difference between nominal bonds and TIPS with otherwise equal characteristics (maturity, creditworthiness) must be roughly the inflation rate financial markets expect over the lifetime of the TIPS. In fact, the rate differential between nominal bonds and TIPS with the same maturity is commonly used as a gauge for market-implied CPI inflation expectations. We call that the “break-even inflation rate,” i.e., at this inflation rate, the three variables, nominal rate i, real rate r, and inflation rate pi satisfy the Fisher Equation:
… for small enough nominal and real interest rates. And if we look at the time series chart of this break-even inflation rate, it’s been hovering at around 2.25% for the last few years. That seems like a reasonable inflation forecast for the next 10 and 30 years if you believe my former employer, the Federal Reserve, does its job. Of course, if you believe that inflation will be materially higher than that, you’ll definitely prefer the TIPS route over nominal bonds. But most rational market participants seem to believe there is no arbitrage between TIPS and nominal bonds and CPI will be slightly above 2% going forward. Also note that, historically, the breakeven inflation rate has fluctuated wildly: during recessions, it normally drops precipitously. This raises a market-timing idea: during a recession, ride the duration effect in nominal bonds, and when the breakeven seems unrealistically low, say below 1%, shift into TIPS, but that goes beyond the scope of today’s post.
How to implement TIPS: ETFs vs. TIPS bonds vs. TIPS ladders
There are different routes to get exposure to TIPS. I list them below, together with the pros and cons:
1: TIPS ETF
Just like for any other asset class/sub-class/style, we have a decent selection of relatively low-cost ETFs. From the major providers, we have VTIP (Vanguard), SCHP (Schwab), TIP (iShares), and STIP (also iShares, but shorter-maturity TIPS, only 0-5 years).
Pros:
- Easy to trade in and out, even in small quantities.
- Likely zero or low commissions and very low bid-ask spreads.
Cons:
- The ETF prices are subject to fluctuations. Specifically, if real rates change, you’re still subject to a duration effect in TIPS, just as nominal bonds fluctuate when nominal yields change. More on that below.
- Some ETFs have sizable expense ratios: iShares TIP: 0.18%. Vanguard’s VTIP and Schwab’s SCHP have the most competitive cost at 0.03% p.a.
- The ETF returns were not materially different from nominal bond ETFs since 2004. More on that later.
TIP’s current distribution yield is 4.71% as of 10/16/2026.
2: Individual TIPS Bonds
Pros:
- No expense ratio.
- You get real cash flow, guaranteed over the bond’s lifetime.
Cons:
- The cash flow is very uneven: relatively modest interest payments (smaller than those from a nominal bond of comparable maturity), but then one large payment to return the principal and accumulated CPI adjustments upon maturity.
- In the secondary market, TIPS bonds could be cumbersome and expensive to trade. New issues are available through TreasuryDirect, which lowers the cost of purchasing TIPS, but the TreasuryDirect website has its own issues. Large brokers may offer zero commissions, but bid-ask spreads may be large.
To manage uneven cash flow, retirees might want to buy not just one TIPS bond but a ladder of bonds to create a CPI-adjusted paycheck, which brings me to the next option…
3: TIPS Ladder
Building a full TIPS ladder is more complex because we need to choose which individual bonds to buy to match cash flow needs over time. Conceptually, that’s relatively easy; I’ll illustrate with a numerical example; see the chart below. You purchase a bond expiring in the last year. The notional would be the cash flow need divided by (1+r), so that the notional repayment and interest exactly match your need. Next, you go to the second-to-last year. Your bond from the first step generates interest in that year, so you need to cover slightly less additional cash flow. You buy bonds worth that remaining flow divided by (1+r) for that year, and so on. In the end, you’ll have a payment stream that precisely matches your cash flow needs. Sweet!
Pros:
- We can generate any cash flow pattern we like, including a (mostly) flat cash flow without big balloon payments in the end.
- A wonderful free website, TIPSladder.com, helps you build a ladder from scratch. You enter your cash flow needs, the horizon, etc., and the site spits out a recommended buy list (i.e., which CUSIP to buy and in what quantity), plus other useful stats, such as the duration and real IRR of the TIPS portfolio. Another great resource for monitoring TIPS yields and breakeven rates across maturities is StreetStats.
Cons:
- For some retirees, TIPS bonds might not be “granular” enough. If you buy them in the secondary market (i.e., in your brokerage account), you can only buy in $1,000 increments. And sometimes not even that, which brings me to the next point…
- Tracking secondary market bond prices and getting the sizing right is cumbersome. Just like for any other bond, TIPS prices change with real interest rates (so TIPS can trade above or below par just like nominal bonds), but the definition of par also changes with CPI adjustments. So, a 2036 TIPS bond issued in 2006 may trade at par, but par isn’t $1,000. Instead, it is $1,000 in 2006 dollars, so we need to add 20 years of CPI adjustments, probably around $1,600. If you want to buy $100,000 worth of TIPS bonds, don’t blindly buy 100 bonds, or you might get the sizing wrong! To be sure, the TIPSladder.com site will do the calculations for you.
- Gaps exist in the TIPS maturity structure: we have no maturing bonds in 2037, 2038, or 2039. Workarounds exist, and the TIPSladder.com site I mentioned above can help; for example, buy bonds that expire in the earliest year after the gap and liquidate them early a few years later, with modest uncertainty about the exact price you will get.
What can possibly go wrong with TIPS?
One reason TIPS can disappoint is that there is no free lunch. I looked at return data for the TIP ETF (iShares TIPS) and IEF (7-10 year nominal Treasury bonds) since 2004, and the two ETFs are significantly correlated. The TIP fund slightly outperformed the nominal bond fund, but not by much. Another major disadvantage of the TIP ETF: during periods of economic stress, such as the global financial crisis in 2008/9 and the pandemic shock in 2020, only nominal bonds rallied, while TIPS sank in early 2009 and lagged nominal bonds during the 2020 recession. Of course, TIP eventually caught up with IEF again, but using TIP instead of IEF in a 75/25 portfolio would have provided less diversification.
Thus, the TIP ETF was less useful for diversifying equity risk. If you put the bond funds into a diversified 75/25 portfolio, performance was almost identical whether you used 25% TIPS or 25% nominal bonds. So, historically, adding TIPS to our retirement portfolio through the ETF route likely won’t move the needle much in SWR simulations; more on that below. Also notice that over this current decade, TIP had a measly 1.87% annualized nominal return, or -2.04% real. That’s better than its nominal counterpart (-0.64% nominal and -4.46% real p.a.), but it’s still not exactly a great inflation hedge either. The best inflation hedge in the post-pandemic inflation scare was the S&P 500! (Side note: OK, maybe the S&P was only second-best, with the Nasdaq taking the trophy, but that’s a story for another time!)
Taxation could be a hassle in (at least) two ways. First, most retirement savers I know will have an equity-heavy portfolio while accumulating. How do you reshuffle this portfolio into TIPS upon retirement? In taxable accounts, it will likely not be worthwhile to incur massive taxable capital gains to shift into TIPS. Most retirees will probably want to shift only assets in tax-deferred accounts (i.e., IRAs, Roth IRAs, etc.).
Second, even if you could easily shift assets into TIPS, they pose another headache in your taxable accounts. That’s because the real interest payments are taxable every year, and so are the CPI adjustments to the principal. So even though you don’t receive the notional increase until the bonds mature, you still have to pay tax on those gains along the way. How unfair! With my equity investments, I don’t pay tax on unrealized gains. However, with TIPS, because the government guarantees the notional adjustments, they are taxable income every year. This quirk can create really unappetizing tax bills. Imagine a retiree shifted his/her entire $5,000,000 in a taxable account into TIPS and gets a 3% real return. Nice, that’s $150,000 a year and a pretty rich retirement budget. Even after paying taxes. But wait, what if you encounter a nasty inflation shock that year, with a 10% increase in the CPI? You will get a tax bill for $650,000, i.e., the $150k in real interest income plus the $500k nominal adjustment. According to NerdWallet’s tax estimator (married filing jointly, using the standard deduction), you’ll be in the 35% tax bracket and will pay $ 153k in federal income taxes, more than the annual interest payment. Even if CPI is more normal, say 2.5%, you pay income tax on $275k of income ($150k of interest plus $125k of CPI adjustment), so about $43k. Your 3% annual real yield drops to about 2% after tax. But note that you don’t pay state income taxes on TIPS income. That’s irrelevant to retirees in states without an income tax (FL, TN, TX, AK, WA, NV, WY, SD), but retirees in Oregon, California, and other states with potentially high marginal tax rates might consider this feature.
Finally, one concern is government meddling with the CPI calculation. True, this is probably tin-foil-hat conspiracy-theory territory, but with growing debt burdens and tight government budgets, I wouldn’t completely rule it out. The government wouldn’t save much money from TIPS bonds by underreporting CPI by 0.5% or 1% a year, because the TIPS market is relatively small. The far bigger incentive would come from both Social Security inflation adjustments (COLA) and the CPI-adjusted parameters in the progressive tax system, where underreporting CPI would push more folks into higher tax brackets. I hope that this wouldn’t happen here. But precedents of governments fudging inflation data are plentiful: Argentina, Chile, Turkey, Venezuela, and Zimbabwe, to my knowledge.
Simulations with TIPS
As usual, I like to simulate two scenarios: 1) the typical traditional retiree with a 30-year horizon and asset depletion, i.e., zero final assets as the success criterion, and 2) a FIRE scenario with a 50-year horizon and only partial asset depletion, i.e., the retirees like to keep an inflation-adjusted 25% value of their current portfolio as their final portfolio value to serve as a cushion and/or bequest target. Version 2 is roughly what I target in my personal situation.
TIPS ETF replication
Let’s start with the easiest TIPS implementation and compare two portfolios, both with 75% large-cap stocks, one portfolio with 25% nominal bonds and the other with TIPS. For the TIPS returns, I use a spliced data series: the actual TIP ETF returns since 2004, and an estimated factor model before 2004. I used a similar technique in the previous post (SWR Series Part 65), where I had to fill in the early return data for Bill Bengen’s new retirement portfolio. I look at the factor model output (you can find these data in the Google Sheet, tab “ETF Factor Exposures” for over 230 ETFs, mutual funds, and stocks). First, note that even the IEF isn’t exactly a 100% 10-year benchmark bond. It captures slightly negative exposure to the 30-year bond, a symptom of its shorter duration, since the IEF uses bonds with 7-10 years to maturity. The TIP ETF has even less exposure to 10Y bonds and also captures other factors: about 11% of combined equity beta (US and international), plus commodities and gold, which clearly captures the inflation hedge. These slope parameter estimates are quite intuitive!
1) 75/25 portfolios: how much will a TIPS ETF help?
Let’s check how my toolkit simulations compare if we replace the 10-year bonds with a TIPS ETF. To this end, I copied the TIPS returns (simulated before 2004, actual TIP returns thereafter) into the Custom Series in my Google Sheet. The TIPS series already includes the expense ratio, so I set its expense ratio parameter in my Google Sheet to zero.
Before we look at the safe withdrawal rate simulations, let’s review the return stats first; see the table below. The return stats are almost identical. A 75/25 portfolio with nominal vs. real bonds will differ by only a few basis points (0.01%). Worse, the TIPS ETF will also raise portfolio risk and both equity correlations and betas across all subperiods. So, let’s not get our hopes up: don’t be too disappointed if the TIP ETF doesn’t materially change your safe withdrawal rates!
Let’s compare safe withdrawal rates by decade for the four scenarios: two retirement horizons and two portfolios. See the table below! Including the TIPS ETF has no material effect on the historical withdrawal rates. If anything, retirement performance slightly deteriorates. For the traditional retiree, the 1960s ($38,180) were the fail-safe, but very closely followed by the 1920s ($38,367). Including TIPS makes the 1960s slightly better but the 1920s slightly worse, for a net loss in the fail-safe withdrawal amount. That’s intuitive: TIPS did well during the inflationary 1970s, but nominal bonds would have outperformed TIPS during the deflationary 1920s and 1930s.
Likewise, in the FIRE retiree (50 years with a modest bequest target), you significantly reduce 1920s performance for the same obvious reason: TIPS didn’t help you during the deflationary Great Depression! Nominal bonds did much better.
Well, that wasn’t too impressive. But wait, there is hope! The TIPS ETF simulations miss that TIPS rates are currently much more attractive than they likely would have been historically. So, why do I care about how the TIPS ETF would have performed in 1937 (subject to a lot of modeling uncertainty, by the way), when we have today’s TIPS rates to work with? That leads me to the next section…
2) TIPS bonds and ladders
Let’s now model both individual TIPS bonds and TIPS ladders using today’s rates. For the TIPS bond, I assume our retiree considers a 30-year bond with a 3.17% real interest rate, according to Bloomberg, as of October 16, 2026. For the TIPS ladder, I assume the retiree has access to a range of TIPS bonds available as of October 16, 2026. I took a screenshot from the TIPSladder.com site and found that to secure a $100,000 (average) annual real income, you’d have to spend just under $2,000,000 today to purchase the entire 30-year bond ladder. You can choose a few parameters, and they all generated costs between $1.969m and $1.980m for the TIPS ladder today. So, in my simulations, I assume that for every $1,000 in guaranteed TIPS ladder income for 30 years, I’d need to convert $19,800 into TIPS today. That’s amazing, because that’s a 1/19.8 = 5.05% guaranteed withdrawal rate over 30 years.
Of course, this 5.05% withdrawal rate has one major drawback: you have zero upside! If you live 31 years, you’d be destitute, while traditional Bengen/Trinity-style simulations hedge the worst case but leave substantial excess funds after 30 years, sometimes more than the starting portfolio value!
To study the trade-offs between hedging the worst case and keeping some upside potential, I use the following six scenarios for our traditional retiree with a 30-year horizon. In each scenario 2-6, I optimize the failsafe withdrawal rate, though subject to different constraints. Specifically, I optimize the TIPS allocation and the allocation across the non-TIPS portfolio (Stocks, 10y bonds, 30y bonds, TIPS ETF, and Cash/Money Market):
- The baseline 75/25 portfolio with nominal bonds. No TIPS ETFs, bonds, or ladders. The initial portfolio is $1,000,000, and the horizon is 30 years.
- TIPS, all-in: Well, to model this in the SWR toolkit, I can’t set the initial portfolio value to $0, because this would trigger a division-by-zero error. So, I constrain the TIPS investment to be at most $999,000 and the remaining financial portfolio to be at least $1,000.
- Obviously, the allocation in portfolio 2 would leave very little in the final portfolio. Thus, I assume our retiree is still OK with depleting the portfolio in the worst case but wants to retain 50% of the initial portfolio at the median of the final portfolio value distribution, conditional on a CAPE above 20 in the initial month.
- Same as Scenario 3, but now let’s target $1,000,000 for the median of the final value distribution.
- Same as Scenario 4, but let’s assume our retiree has $500,000 in highly appreciated stocks in a taxable account. So, we have the additional constraint of maintaining that equity allocation in the non-TIPS portfolio.
- Same as Scenario 5, but I assume the TIPS rates are much lower: the standalone 30-year TIPS has a real rate of only 2.25%, and the TIPS ladder is more expensive and only affords you a 4.5% safe withdrawal rate (equivalent to a 2.12% IRR of the TIPS ladder, about a whole percentage point less than today’s rates. I want to check how much is left of the TIPS route if rates ever “normalize” again (then again, maybe TIPS rates are normal today)
Let’s look at the simulation results:
- This is the typical Bengen/Trinity/ERN-style result: You get a slightly below 4% failsafe. The roughly 3.82% rate occurs twice, once in 1929 and once in 1968. Even though the worst-case scenario leaves you with nothing after 30 years, there is massive upside potential: even conditioning on the CAPE above 20, you’d still have 1.6x your portfolio after 30 years at the median of the final distribution.
- All into a TIPS ladder (well, at least 99.9% in, leaving a small financial portfolio cushion) gets you a roughly 5.05% safe withdrawal rate. It’s as flat as it gets throughout the decades because you have essentially no financial portfolio. You also wipe out your portfolio with precision; the median final value is just under $1,500. This all affords you a 32.25% richer retirement budget!
- To guarantee a comfortable $500k cushion in the end, you need to maintain only about $145k in the non-TIPS portfolio, invested almost entirely in equities (98.1%). You’d still do a massive shift into the TIPS ladder: over $855k. The failsafe didn’t deteriorate much: It’s still 47,443, so 4.74%. This is 24.26% higher than in the benchmark.
- If you want the median final value to be 100% of your initial portfolio, you would still transform more than two-thirds of your portfolio into the TIPS ladder. You’d again keep an almost all-equity portfolio. The failsafe is now down to $44,352, or about 4.44%. This is 16.16% better than the benchmark
- Requiring the $500k minimum equity portfolio severely hampers your TIPS conversion options. Sure, you could still transfer $500k into a TIPS ladder, but that would leave you with a $500k all-equity portfolio, which apparently would be too risky. So you’d instead keep a diversified portfolio (72.7% stocks, rest bonds and TIPS) and transfer only $312k into TIPS. You get an improvement of about 11% in your retirement budget, all while maintaining significant upside to your portfolio in the end.
- Same as scenario 5, but with lower TIPS yields. The TIPS ladder is now $263k. You use a 68% equity portfolio; the rest is mostly in TIPS (23.5%) and nominal bonds (8.4% combined). The failsafe is now $40,743, or 4.07%. That’s 6.71% more than in the benchmark.
Notice that none of the scenarios called for any investment in the single 30-year TIPS bond. The optimal allocation to maximize the failsafe would only seek the TIPS ladder! This makes sense: a large payoff in year 30 is less useful if you strive to maximize the failsafe with zero final value.
What I find amazing is that with today’s TIPS rates, even in the most cautious setup, you can increase your retirement budget by over 10%. That’s impressive, and it’s a bigger boost than a reverse glidepath, where you can squeeze out about 7% extra.
For the 50-year retirement with a 25% final value target, I go through the following progression of scenarios:
- The baseline 75/25 portfolio with nominal bonds. No TIPS ETFs, bonds, or ladders. The initial portfolio is $1,000,000, with a 50-year horizon.
- Add the single 30-year TIPS bond to the mix (but not yet the TIPS ladder).
- Add the full 30-year TIPS ladder to the mix.
- Require a minimum $250,000 financial portfolio.
- Require a minimum $500,000 financial portfolio.
- Require a minimum $750,000 financial portfolio and a $ 500,000 equity portfolio.
- Same as scenario 6, but with lower TIPS rates.
Let’s look at the simulation results:
- You get a 3.23% failsafe. This time, the 1900s, 1920s, and 1960s are the close competitors for the worst-case scenario. Even though the worst-case scenario leaves you with nothing after 30 years, there is massive upside potential: even conditioning on the CAPE above 20, you’d still have 3.86x your portfolio after 30 years, based on the median of the final distribution.
- Adding just one TIPS bond won’t do much: Even though I set the upper limit of TIPS exposure at $999,000, it’s optimal to buy only about $20k. The portfolio shifts down to 63% stocks, 13% 10y nominal bonds, 9% TIPS, and 15% cash. The failsafe withdrawal amounts don’t change much, only about 1.18% higher overall and 0.52% higher post-1926. Not much to see here!
- Adding the TIPS ladder option has a massive impact on your retirement finances. The retiree would shift almost $882k into the TIPS ladder, while the 30-year TIPS bond remains unutilized. That’s intriguing: the $999,000 upper limit remains unused. The remaining $112k of portfolio value is all in an equity portfolio. With this move, the retiree would have created a massive +25.94% in the retirement budget since 1900 and an even more impressive +34.17% in the post-1926 era. Specifically, the safe withdrawal rates are 4.07% overall and 4.36% post-1926. Also noteworthy: you don’t give up much of the upside. The median final value would have been almost 2x the inflation-adjusted initial amount. What’s intriguing is that the TIPS ladder in this optimized solution is overfunded, i.e., the cash flow from that ladder exceeds the failsafe withdrawal amount. So you create a new dollar-cost-averaging machine: use the $112k initial stock portfolio and add more every month until year 31 of your retirement. Sequence Risk solved!
- The next scenario restricts the equity portfolio to above $250k, say, because highly appreciated stocks are in a taxable account. Again, the optimal portfolio asks for 100% equities in the non-TIPS financial portfolio. The failsafe withdrawal rate is still 3.985% for all years and 4.09% for post-1920 cohorts, translating into a quarter more
- When you need at least $500k in your equity portfolio, your TIPS ladder will be $411,680, with the remaining portfolio 85% stocks and 15% 30-year nominal bonds. This move gives you roughly an 11- 12% higher annual retirement budget.
- If we further constrain the total non-TIPS portfolio to $750k, we’re now looking at a $250k TIPS ladder, $750,000 in the remaining portfolio, allocated to 72.3% stocks. This move affords you a 7.28-7.98% higher retirement budget.
- Finally, with lower TIPS rates, some of the advantage disappears: you expand your retirement budget by only 4.33% in the 1900s and 3.65% in the post-1920 cohorts.
Again, these results are very impressive. At today’s rates, even with a conservative assumption that we move only a quarter of the portfolio into a TIPS ladder, we can pull off a 7% boost to the retirement budget. About the same ballpark as the reverse glidepath.
Also, in all scenarios, you would have maintained pretty strong upside potential for the final portfolio value. That’s because targeting $250,000 in the worst case means you get sizable final portfolio values in the not-so-bad cases, thanks to 50 years of compounding! So, over the longer horizon, only the initial portfolio size and unrealized equity gains in taxable accounts pose any true constraint on the TIPS ladder conversion!
Conclusion
People have come up with all sorts of wacky solutions to Sequence of Return Risk. Like this one I found on Reddit the other day:
Thus, as always, I started this project with a great deal of skepticism. I looked at the TIP vs. IEF performance and thought that was it; there was no more work to do. What’s more, after the last few blog posts, shooting down the Bill Bengen small-cap stock approach in Part 65 and taking the even more offensive and idiotic Risk Parody, uhm, Risk Parity approach to the woodchipper in Part 64, I wasn’t really in the mood to write another post about a “negative result.” It would give more fodder to the trolls and haters who often falsely accuse me of just running a “hit job factory” on my blog. But that’s not true: I’ve written positively about gold, reverse glidepaths, trend-following/momentum strategies, and options trading for supplemental income in retirement.
Today I can add the TIPS ladder to the list of ERN-approved retirement safety enhancements, at least in today’s TIPS rate environment. Specifically, I was pleasantly surprised by how well a carefully modeled TIPS ladder works. It’s crucial to never treat the TIPS route as a zero vs. one decision (i.e., no TIPS at all vs. all all-in TIPS). We should simulate the TIPS ladder as part of your overall financial asset portfolio, something that, to my knowledge, only my SWR Toolkit can do properly. Of course, even with my toolkit, it took some hacking, i.e., downloading the Google Sheet into Excel and running the solver/optimizer there.
The strength of a TIPS ladder in today’s environment comes from its ability to support a 5+% safe withdrawal rate over 30 years, at least with current real rates hovering slightly above 3% for 20- and 30-year TIPS. To my knowledge, no other approach can guarantee that kind of (inflation-adjusted) initial withdrawal rate. Of course, some charlatans will push artificially boosted SWRs through hindsight bias, e.g., tying retirement to a rebound in the long-gone small-cap value-stock alpha premium from the 1930s and 1970s. But that’s wishful thinking by some ditzy internet trolls. In contrast, TIPS enjoy an actual government guarantee.
Sure, the TIPS ladder has the disadvantage of zero upside, while the Bengen- or Trinity-style failsafe rate will have lower failsafe amounts but far better outcomes in less-than-worst-case scenarios. Retirees should consider only partially converting to a TIPS ladder to keep enough of the upside while maintaining a safe withdrawal rate well above 4%. The partial TIPS ladder conversion may be the preferred route anyway because of the tax planning constraints mentioned above.
But the 5% SWR over 30 years, even with depletion, also helps you in longer retirements! I was surprised by this feature. I previously often dismissed the TIPS approach because TIPS only go out to 30 years, so you’d face reinvestment risk after the 30-year mark, and I suspected we shouldn’t use TIPS in that case. But today’s post confirms that 30 years of TIPS can still help you with a 50-year retirement. It’s almost like you create a “fresh shuffle” in the Sequence of Return Risk deck by shifting a large portion of your portfolio into a TIPS ladder and keeping the rest in equities. Today is not a bad time to do so, given the nosebleed-high CAPE ratio: 39.58 with Shiller’s methodology and 35.44 with mine, as of 10/16/2026, with an S&P 500 close of 7,659.44. You’d effectively create a fresh 30-year buy-and-hold portfolio with 100% equities (or potentially even contribute to that portfolio if your TIPS ladder income is higher than the annual budget), and thus completely hedge away the risk from today’s elevated equity valuations.
But alas, not everyone can pull off that portfolio switcheroo from significantly appreciating equities into TIPS. Most people I know have most of their portfolio in taxable accounts. TIPS create too much of a tax headache, both when shifting funds from equities to TIPS upon retirement and an ongoing tax drag during retirement. Most retirees will benefit from a TIPS ladder in a tax-advantaged account only. Finally, there is no guarantee that real rates will stay high. A TIPS ladder would be less attractive if rates declined again into the low-2% range or below. But TIPS certainly look like an attractive opportunity at today’s rates.
Please leave your comments and suggestions below! Additionally, be sure to explore the other parts of the series; see here for a guide to the different parts so far!
Title Picture Credit: WordPress AI + ERN edits
Technical Appendix
The Google Sheet link is here. It has the TIPS ETF return series.
I downloaded the Sheet as an Excel File and used the MS Excel Solver function in the tab “Cash Flow Assist.” You can check out the file here. All simulation results are in the tab Sim Results. Notice that modeling the TIPS investment in the Cash Flow Tab requires some additional fine-tuning of the formulas. For example, if we have a final value target, you need to adjust the target in the Excel Sheet. If you divert $250k from the portfolio and need $250k at the end, that’s now 1/3, not 25% of the non-TIPS portfolio.
Thank you for revisiting this, with the rates seen on the second week of october. Changing your 2024 verdict when the numbers changed. I hope people who get an oppurtinity to enhance their portfolios do so with an open mind regarding the tax headache. “a fresh shuffle in the Sequence of Return Risk deck” Love it.
Starting at today’s CAPE: the ladder hedges the fail-safe of the 1929- and 1960s-type cohorts. If you reran the same optimization conditional on a starting CAPE of 20 or below, I’d guess it keeps far less in the ladder, because the stock remainder’s expected return is then well above 3% real. If so, does that argue for a rule that moves ladder money into stocks after a deep equity drawdown, say 40–50% in real terms, rather than holding every rung to maturity? Or should the ladder be untouchable once bought, with only the surplus-into-stocks drip (DCA) the only way back in. Id think such a drawdown rule switch back into assets allow for even higher safe withdrawal rates.
Oh wait you kind of answered this type of question. Then a follow-up to that Part 61 reply, where you’d roll a heavy ladder back into stocks with an active glidepath (shift only when the S&P 500 is down) and doubted a valuation signal. Here the ladder is held to maturity and the only way back in is the surplus drip. With your solver: if the start were a CAPE of 20 or below rather than 35, how much of the portfolio would it still put in the ladder? If the answer is much less, which is the better way to get there, the drip, the active glidepath, or moving a block of ladder money into stocks after a deep drawdown of 40–50%?
The ladder optimization was performed unconditionally, so it would capture historical worst-case scenarios during periods of high CAPE. I didn’t bother looking at the CAPE-below-20 condition because we haven’t seen that scenario in a while.
You’ll be surprised that even slightly below 20, a few bad failsafes occurred, so maximizing the failsafes condition on CAPE under 20 wouldn’t change that much. The bigger issue would be that, in that scenario, real interest rates would be very low, so a TIPS ladder at low rates looks unattractive compared with an equity market primed for a rebound. So, a TIPS ladder may not be that attractive in that case.
It’s hard for me not to start wondering if making a fairly straightforward decision tree based input based on just CAEP and CPI could help with this even more – even without a pure solver, just establishing a rule to nudge equity allocations within a set range (with 60% to 85% as the guardrails to bounce within) and doing the TIPS ladder rebalance within that smaller bond/treasuries allocation can make some real movement happen on your 50yr 25% default FIRE type representation. I don’t want to reintroduce a ton of complexity into a model you’re probably having to throw Excel middleware at to optimize solvers, but just adding these three levers and some Crystal Ball abuse in excel, intuitively I feel like you could find a little bit more SWR without it being the typical post facto hackjob, but something a bit more like an equity/bond/TIPS ladder dynamic allocation strategy (optionally with momentum) that can be fine tuned for risk tolerance.
Yes, the holy grail would be a dynamic model that shifts across asset classes. I haven’t looked into this enough to check whether this works. Ideally, one would use features from both valuation and momentum. Valuation can take painfully long to revert to normal.
Truly great stuff Karsten! I love that you keep an open mind and changed your view now that things are different. Thanks for the breakdown, this was really helpful 👍
Thanks! Glad you liked today’s post!
Thanks for the analysis BigERN! Just to check if I understand correctly, am I right in thinking that in scenario 5 of the 50 year simulation, if I put $411,680 of the $1mm portfolio into the TIPS ladder and the rest in a 85%/15% equities/nominal bonds portfolio, I would be able to spend $36,091 in today’s dollars for 30 years, at which point I would have nothing left in the TIPS ladder (assuming I spent all $36,091), and then my spending for the subsequent 20 years of the 50 year period would be funded out of the equity portfolio which has been growing untouched for 30 years? And the amount I could spend for those 20 years would be (portfolio size in year 31 * 3.61%)?
If that understanding is correct, are the results multiplicative i.e. if you have $2mm total portfolio at the start, is the number $36,091*2=$72,182 real spending per year for the first 30 years?
The TIPS ladder will be depleted. You will not be able to fund the spending out of the TIPS ladder completely. $411k will only generate about 20k in annual revenue, so you still need to withdraw a bit from the non-TIPS portfolio, but the rate is so low that the portfolio will rise enough to sustain the retirement in years 31-50.
No. You’d still spend 36901. Not tied to the portfolio value.
Yes, the results are scalable: If you got $5m you multiply everything by 5.
This analysis is unsatisfactory. Market returns are non-ergodic, as are the investment and return paths of individuals. Academic research shows that investors in nominal assets are not compensated for residuary/unexpected inflation, thus should heavily bias their fixed income sleeves towards inflation protected bonds. This is especially important for retired investors whose future labor capital is presumptively zero. The point of TIPS investing is to protect against severe left tail events, not to maximize theoretical expected returns. It is pointless to chase rapidly declining marginal utility on the right side of the returns curve in exchange for sacrificing safety on the left side. Run your simulations for 20 – 30 years starting in 1929, 1966, and March 2000 and see if it alters your conclusions.
I showed that TIPS (through a ladder) helped, which seems to align with your general idea. Why be combative? I know that return patterns are non-ergodic. That’s the whole point of my SWR series, i.e., Sequence Risk. Have you read my series?
Your analysis of my analysis is unsatisfactory. Maybe you should have read my post to the end: I did run 30-year simulations starting in 1929, 1966, and 2000. TIPS ETFs did not help in 1929 and 2000, but they did help in 1966 and 1968.
The nominal ergodicity of markets as a whole is statistically buried on the noise that is consistent reversion to mean, especially when so much of equity valuation is tied to broad human perception of capital ROI returns.
We’ll keep reinventing the airport arrival paradox of showing up hours early because 1st percentile results are still disproportionately costly – the fact that TIPS meaningfully expunged the mid/late 60’s as the SWR allocation driver in longer simulations by partially ameliorating SORR during sustained inflationary periods is actually a relevant finding. In a stratospheric CAPE environment with decent-looking TIPS returns that can be laddered out without egregious shoe leather cost (particularly if it’s boomers or FIRE GenX/GeriatricMillenials recharacterizing RothIRA holdings), it’s something to actually look into, particularly if somebody is looking to pull the trigger and start working less but is justifyable wary of a $10T AI/Tech CAPEX boom not having exactly the second-order ROI effects we might want in a world that still can’t fix most supply chains.
My conclusion: holding TIPS in the TIPS funds/ETFs, such as TIP or VTIP, won’t have significant impact on the SWR because they only hold short to intermediate TIPS, and so cannot get full advantage of the current high rates. For that one has to build a 30 year TIPS ladder.
A comment: many people have no access to TIPS ladders (or individual bonds) in their 401(k), where most of their tax-protected money reside. However, there may be access to defined-maturity TIPS ETFs, such as IBII from iShares, via a brokerage window. Unfortunately, there is no way to build a 30 year ladder of these defined-maturity TIPS ETFs. It currently extends for 10 years only with new rung added every year.
A 401k with a brokerage window often can buy TIPS ladders. I do so in my Vanguard 401k which has a Schwab PCRA (a type of SDBA, self directed brokerage account). Schwab permits buying TIPS on the secondary market (and re-auctions I think, but I prefer secondary for anytime convenience).
Also purchasing TIPS in Vanguard RothIRA. And most people who have left their employer can rollover their 401k into an IRA, where they will have access to TIPS purchases at any decent brokerage.
Yes, good point. Once you leave the job, some people roll over to a Fidelity/Vanguard/Schwab IRA for better asset allcaotion options.
That’s another quirk: Tax-advantaged is not enough; you may have to transfer your 401k to a rollover IRA at, say, Fidelity, where you can buy the TIPS.
Anyone can build a proxy ladder by holding a long TIPS ETF and a short TIPS ETF and then every year rebalance to match the average duration to the average duration of a ladder
The long plus short TIPS ETFs approximates the TIPS ladder, but they are not equivalent. If the real yields rise abruptly and significantly just before you plan to withdraw, the value of both ETFs drops, and you will have to sell at a loss. With a ladder the real return of each rung at maturity is determined at the point of sale.
Yeah, you could duration-match that way. There is still a small potential mismatch if there’s a lot of curvature in the term structure, though.
Thank you for another excellent post, Big ERN! I was just about to buy a four-year TIPS ladder to bridge until I start Social Security at age 70. But now I’m thinking maybe I should buy some more for ages 70-90, too…
Thanks! Yes, it’s not a bad time to do that, at least for a portion of the portfolio.
You have a typo in the sentence “This time, the 1900s, 1960s, and 1960s are the close competitors for the worst-case scenario.” I respect your usual rigorous adherence to correct math and data, so I’m thinking you should probably fix this typo to keep this article consistent with your overall approach to reporting on these issues.
Thanks! Fixed the typo!
Thanks for fixing the typo. The article was one of the best data-driven takes I’ve read on TIPS. Fortunately, you reinforced my decision to build a 20-year ladder a couple of months ago. I of course could have squeezed some more yield if I had held off a bit, but I’m not going to cry over that, as I made a conscious decision that rates I saw in July were simply good enough to make my 20-year ladder a rational move, even if it wasn’t optimal.
Thanks! Yes, it’s hard to time this perfectly. But at the rates a few months ago this was still a deal.
Thanks for fixing the typo. The article was one of the best data-driven takes I’ve read on TIPS. Fortunately, you reinforced my decision to build a 20-year ladder a couple of months ago. I of course could have squeezed some more yield if I had held off a bit, but I’m not going to cry over that, as I made a conscious decision that rates I saw in July were simply good enough to make my 20-year ladder a rational move, even if it wasn’t optimal.
Great article, and great timing (the ~3% real yields have had me looking closely at TIPS lately too).
The thing that quickly became clear to me is the entry point timing for buying TIPS seems to be very important. There’s no point in buying TIPS when real yields are 1% or less. That’s why TIP ETFs don’t look good over the long-range – they’re buying TIPS at bad times.
Your bar chart for TIPS Ladder cash flow is the best illustration of a TIPS ladder I’ve ever seen.
I recommend also https://streetstats.finance/rates/tips which is how I track slope or inflection points on the yield curve, and how TIPS rates are moving week-to-week.
Thanks for the kind words. Yes, exactly: attractive TIPS entry points are rare. 2026 seems to be one of them!
Thanks for the link! I will include that in the post.
I do worry about the CPI being understated / under-estimated. It already is, pretty clearly from some sources I’ve read (ex, on OER – owners equivalent rent – particularly). But even with a slightly off CPI the TIPS real yields are still the best you can do on real yields at that risk level.
But if it drifts more, then potentially the TIPS ladder loses ground on your personal inflation rate – which of course can vary significantly from national CPI, particularly in high COLA coastal areas.
As much as it’s a valid concern, too many other things rely on CPI being at least vaguely tied to cost of living, so to me it’s less of a case of looking a U1/U6 unemployment to find real truth, and just one where reversion to mean means my odds are probably fairly close to a coin flip for ‘treasury error in my favor’ as a Monopoly chance card… as long as I’m not get absolutely ran over by tax implications of carrying a TIPS ladder, this feels like a reasonable addition to the strategy bin.
I’m just lazy, and kinda want Karsten to build an optimization decision tree between bond ladders, TIPS ladders, standard equity allocations based on CAEP, and see if Ritzholtz or somebody will grit their teeth and make a reasonable-load momentum fund to add at minimal effort, then I can put a fair bit of this into a ‘less is more’ strategy that hits my nominal lazy math target of a 3.65% SWR for my family (since I have a few fixed and inflation-adjusted income sources that let me cheat here)
Yes, the CPI is just a single national average. Housing runs hotter on average, by about a 1%point pace. Especially in HCOL areas, you’d often feel insulted by CPI adjustments alone. And that’s not even the Venezuela-style data fudging yet.
Nice post, and very timely! Questions/comments:
1) You wrote: “First, note that even the IEF isn’t exactly a 100% 10-year benchmark bond. It captures slightly negative exposure to the 30-year bond, a symptom of its shorter duration, since the IEF uses bonds with 7-10 years to maturity.”
Can you explain in a little lower-level detail what is meant here? They can’t be negatively correlated, can they? I’m not all that familiar with factor modeling, if that’s what you meant.
2) Today’s breakeven rates in the 2.25% to 2.5% range seem oddly optimistic (i.e. it’s lower than the actual experience of inflation for 5+ years!). Is it possible investors in TIPS have a higher inflation estimate in their minds, but are accepting a lower real yield as a way of paying for the option value of the TIPS? I.e. if TIPS are the only way to survive a scenario like “10% inflation for each year of a decade, starting years from now”, then perhaps TIPS have value beyond what nominal bonds offer. TIPS seem to offer today’s yield plus today’s inflation plus a call option on CPI going to the moon in future years. Could it be that only the first two of these sources of value are captured in the breakeven calculations? If so, the breakeven numbers don’t function as an accurate poll of investors’ expectations. Do you think TIPS include an option value?
3) It seems there is a spectrum of real rates, and how much it makes sense to allocate to TIPS. At one end of the spectrum, there’s no justification to ever buy TIPS (e.g. mid 2021, -0.85% real yield) and at the other end you need to back up the truck and go all-in with a 100% TIPS ladder allocation (e.g. a 4% real yield on TIPS). I think there’s probably more to uncover here. With the spreadsheet infrastructure you built to write this article, could you build a curve of optimal TIPS allocations at various real rates between 2% and 4%? The advice around TIPS allocation probably varies based on the rates prevalent at any given time.
4) One reason to NOT buy a heavy TIPS allocation is that inflation can go to zero or below during severe recessions. In those scenarios, as I understand it, you can not only lose previous years’ inflation adjustments all the way down to your original principal, but at the same time your yield/income can be slashed at a time when it would be hard to find a job. I’m sure your model covered this risk by including the Great Depression and 2008, but I’m pointing it out because it wasn’t mentioned. There might be possible deflationary scenarios that have not yet happened in which a TIPS-heavy portfolio with withdraws at the historical SWR would fail.
5) ImmediateAnnuities .com is today offering me (age: late 40s) a 7.22% lifetime stream of income, which seems to beat the ~5% WR for 30 years that a similarly asset-depleting portfolio of TIPS can support. However this would be a flat, nominal series of payments instead of an inflation-adjusted series, so the real purchasing power would decline over the years. The annuity at the higher rate would come out ahead if inflation was below a certain range (maybe less than the breakeven rate plus 2.22%?), and the TIPS would come out ahead if it was unexpectedly higher than that. If I was to go the annuity route or the TIPS ladder route as a sort of bond tent on steroids, I would want to have a buy-and-hold equity allocation growing untouched in the meantime, to make up for inflation and to provide upside. Rates have changed so quickly that I don’t imagine some of these scenarios have ever been modeled. Do you think I’m correctly considering the tradeoffs here, not including taxes?
6) Just a comment about the tax pitfalls: I don’t know how most people in the U.S. have so much of their assets in taxable accounts, and have to pay taxes to adjust their AAs. We have >85% of our assets in traditional and Roth IRAs and managed in several years of our careers to max them out. Having most of our assets tax-protected means we can immediately pivot to take advantage of market conditions like TIPS yielding 3.2% and incur no tax liabilities. A word to the wise: If you have taxable assets you need to be maxing out your Roth account contributions over a period of many years. If you are piling up savings in taxable accounts, you need to increase your 401k/403b deduction. If you can max out both Roth and traditional IRAs, and still have savings piling up in taxable, then you’re either heading for retirement within a few years regardless of what you do, or you have a severe spending problem.
The 7% fixed annuity is kind of mind blowing (I’m also in my 40’s). It seems to me that an annuity with that kind of payout could more than eliminate the sequence risk of, for example, my 2.8% fixed rate mortgage with 25 years until payoff… (I’d still have to consider the tax implications though!)
But it would expose you to significant inflation risk. If we have another 10 year inflation run like the 70s that 7% would be effectively cut in half. I’m guessing the TIPs ladder would be better. You could use Ern’s spreadsheet to compare.
On the flipside, if we get another disinflationary / deflationary recession like 2008, one’s TIPS income dries up, whereas the annuity or nominal bonds would not.
Nope. With deflation, TIPS will hold their value 100% when held to maturity and they will continue to throw off interest over their entire term as well.
100% of the initial nominal value, not 100% of what you paid if you bought it a few years after issuance
It’s important to be clear about what happens when you buy TIPS on the secondary market. You will absolutely get your purchasing power back (inflation + real yield). You aren’t getting shortchanged back down to a flat unadjusted $100 face value; you get the fully inflation-indexed principal based on the index ratio at maturity. I think you’re just clarifying that the baseline multiplier is the bond’s official par structure tracking historic inflation from issuance, rather than a refund of your specific transaction cost if you bought at a premium or discount, yes?
You are conflating a one year deflationary recession with multiyear deflation.
The former might happen.
But in the “Helicopter Ben” era, the chances of sustained deflation are vanishingly close to zero,
Very true. But I liked the idea of the annuity as a counterweight to the 2.8% mortgage.
That one has inflation risk, though! But I like the idea of the hedge against the 2.8% mortgage!
You have to be careful when looking at annuity payout rates. Those rates are not the equivalent of compounding interest from an investment. The ground lost to inflation over many years is substantial. I’m saying this even though I own an FIA with income rider whose payout rate is about 8%. I’m still satisfied with the purchase though. FIAs with income riders typically pay back principal mostly in the first decade, so there are no taxes to pay on that paid back amount. That lifts your net income during the crucial first few years of retirement and I find that the consequent relief of pressure from my growth engine in VT will help that growth engine do its job more effectively. But I also have a TIPS ladder, so at least part of my portfolio is immune to both market swings and inflation.
All good points:
1: If I regress the 7-10y IEF on 5y and 10y benchmark bonds, then you’d get the expected result: probably 0.30 exposure to 5y and 0.70 to 10y bonds. Without the shorter-maturity bonds, and with only 10y and 30y benchmark bonds, the statistical model picks up a negative weight on the 30y bond.
An analogy: You have Beck’s beer (decent German lager) and Ayinger Celebrator (heavy-duty, high alcohol beer) as your references. Now you try to “describe” Coors Light in that spectrum. Well, you can’t because it’s to the left of Beck’s in alcohol and taste. So, statistically, it’s a 0.9x Beck’s and a minus 0.3x Ayinger.
2: TIPS include somewhat of an option value. If CPI is negative, the notional adjustment is still zero, never negative. So, it’s a call option on CPI inflation with a strike at zero. That value might be worth something and could taint the price somewhat.
3: Correct: 3% is likely already a sweet sot where you would move large amount into TIPS. 4% would be a totla no-brainer. The question is how far can TIPS fall before you’d pass again. That’s something you’d heve to personalize. Depends on your preferences.
4: Not true. see the point above. There is a call option feature on CPI.
5: Hard to compare: if you subtract the inflation estimate, you’re right back at the implicit TIPS SWR again.
6: Yes, you could. I’m surprised not more people do that now. Or maybe they do becaue TIPS yield have fallen since the peak.
I only have feedback on item #6 – despite having a sizeable (99.X% of the human population is understandably envious) amount of money in tax-deferred and Roth accounts from maximizing contributions for well over a decade, most of my earnings are rear-loaded and almost windfall-like other than being recurring. I suspect I’m not alone. So, while similarly situated people are in the bin of ‘en route to short term retirement capability’, any amount of tax optimization is probably going to be quite welcome, and nudging SWR a tenth of a percent is something nontrivial, or at least relatively tangible in terms of some additional number of weeks worked or not as far as immediate spending horizon.
The tax system is meant to have high skilled high earners shoulder a disproportionate load anyway (because the generationally wealthy don’t want the unwashed masses with graduate degrees to be eligible, and will lobby accordingly since that’s aligned with their interests anyway), and I’ll probably hit my milestone of paying 7 figures in taxes by the time the decade is out as my badge of accomplishment for my troubles… it just is what it is.
Yea, I guess somehow we ended up earning modest incomes (keeping tax rates low) and also piling large percentages of that income into 401ks and Roths from our 20’s to 30’s (keeping taxes even lower, in the case of the 30-40% 401k deductions).
Over the past 5 years, we’ve never paid more than 5% of our AGI as “total tax” to the federal government or our state (married filing jointly, one kid, no special tricks other than the 401k). Our total state + federal income tax burden is around 8% per year, which is stupid low.
I guess people living in more expensive places or with more expensive lifestyles cannot afford such a high savings rate. Therefore they save less in their tax-advantaged accounts, and then pay a higher tax rate, and then have even less available to save. I saw the virtuous cycle earlier than most, stayed in a LCOL, maxed out my IRAs, and got rich despite being an income underachiever.
Great post, especially on the TIPS ladder, and how it resembles the reverse glidepath benefits, because it kind of is.
If I don’t have access to a TIPS ladder, but can buy a TIPS ETF, and “manually” do the glidepath/ladder by selling the TIPS ETF over 30 years, would the results be much different?
An evergreen TIPS fund can’t be used like a TIPS ladder because it can’t guarantee return of principle. Build a TIPS ladder and you’ll know that every single year you’ll receive return of principle, plus inflation adjustment, plus yield. A TIPS fund can’t reliably do that. Also I question the statement that you don’t have access to TIPS. TIPS can be purchased directly or they can be bought on the secondary market, even inside an IRA, with brokerages like Fidelity. Also, you can build a 10 year ladder using TIPS maturity ETFs that currently exist through 2036. If you can’t access TIPS because you don’t have access to the US market, realize that other countries often have similar vehicles, like index-linked gilts in England.
Two TIPS funds, one short and one long, rebalanced annually to match the average duration of a TIPS ladder provides equivalent returns to a TIPS ladder
The ETF option doesn’t hedge the payment stream as well as the actual TIPS ladder, though.
As others said, a “regular” TIPS funds can’t, but I would think that the iShares term TIPS funds like IBIF could. They do pay out the inflation part every year though, so I think you’d want to automatically reinvest dividends, and then manually do the ladder by selling just the “interest” part every year for each term ETF.
Yes, the target-year fund does this correctly. But if you plan to hold to maturity, why not implement the TIPS by hand? Saves the expense ratio.
Thanks! This is really helpful and relevant in today’s environment. Can you recommend an equivalent for European/EUR investors?
Most European investors should have the same pallette of TIPS bonds. They are often called different (Gilts in UK, “Teilindexierte Anleihen” in Germany, OATi and OAT€i in France, etc.) Check your local listings!
Great analysis, Karsten.
I will read it several more times, to make sure I ‘actually get it’
One followup question:
If my primary objective were to protect the first 10 years of retirement spending against SORR, would a ladder of iShares target-maturity TIPS ETFs (IBIG, IBIH, etc.) be a reasonable, lower-complexity alternative to building a ladder of individual TIPS?
Since these ETFs have defined termination years, they seem to offer many of the advantages of an individual TIPS ladder without the complexity of purchasing and managing individual bonds.
Would you expect a 10-year ETF ladder to provide comparable SORR protection, or are there meaningful drawbacks I’m overlooking?
I actually own the iShares TIPS defined maturity ETFs through 2036. I bought them a while back when I thought building a ladder with individual bonds was complicated. I later learned that it’s not complicated at all if you use tipsladder.com, and I then built a 12 year ladder from 2037 to 2048 with individual bonds. It was very easy to do. The ETFs may be a little more convenient, but they do have an ER, so there’s a cost to the convenience.
That would work too. I would check the expense ratios, though. You might be able to perform the same task with much lower fees when buying the actual TIPS for a buy-and-hold portfolio.
iShares iBonds target-maturity ETFs have an expense ratio of 0.10%
Thanks, Aubrey! For small ladders up to 10 years old, that’s not a bad option. For large account values, I’d still go with actual TIPS bonds, especially if you know they’ll stay until maturity.
I bought 30 year Tip bond to withdraw the interest while allowing the principal to be used as a reset at maturity. I am allowing some stocks to glide up without rebalancing because both ends march upwards over time. I can’t sell ass to generate income, so I use deferred income annuities to trigger income for specific periods ant different intervals. The principal payment to buy these are much smaller than what I would have needed for a stocks heavy portfolio. As such the combined income from everything is enough for me and there is reasonable increases from dividends and Tips adjustment plus add one of annuities to make it all work out for a significant lower principal than the normal SWR system
Nice! You’re all set for a worry-free retirement!
I too have been very tempted to use a deferred annuity to offset SORR in the specific timeframe where that might be an issue, beyond the timeframe of a bond tent.
For example, if you are a 40 year old early retiree, you could bond tent from ages 40-50, but then you have 12 long years to go until the Social Security lifeline at 62, during which time a lot could be going wrong if a SORR event occurred or had just occurred, and the portfolio is on a depletion path. If you can manage to avoid spending your principle in that timeframe of vulnerability when many simulated retirements go into terminal decline, it seems you’d be much more immune to SORR.
As a specific example, I can buy a period certain annuity today that starts payments in 10 years, and pays for the following 10 years. The payout rate I was just quoted is 23.26%. So if I wanted to guarantee $40k income during those future years, starting a decade from now, I would need to buy ((1/0.2326)*40k=) $172k worth of this annuity. That might not buy much after a decade of inflation, and I haven’t calculated the IRR as opposed to something like a ladder of zero coupon bonds, but it’s enough to greatly reduce withdraws from the rest of one’s portfolio during that timeframe. A planned windfall like this might just be the relief an otherwise-failing portfolio needs to survive another 40 years.
But the TIPS ladder at today’s ~3% rates offers yet another alternative that can be extended out nearly 30 years as part of a glidepath strategy. This seems more straightforward than the annuity idea, but it didn’t make so much sense until rates rose. You could even kinda overbuild your bond tent – essentially the exercises in ERN’s post – to supply lots of overlapping income from the TIPS so that the pressure to withdraw money from selling equities is reduced until much later in retirement.
Does the TIPS ladder eliminating SOR for the first 30 years (of a 50 year retirement) make a future Social Security payment (say in 20 years) more impactful to the overall withdrawal rate than you have discussed previously? Thanks for all the great work you do!
Not sure if there’s a connection between the impact of future cash flows. You’d certainly adjust your TIPS ladder if, for example, you expect your Social Security to start 25 years into FIRE, so you bridge only 25 years with a TIPS ladder and then do nothing or reduce the ladder for the years when SocSec starts.
Age 70. New York State resident. During the past 18 months, I had previously set up a ~24 year TIPs ladder for the period 2028-2052, using tipsladder.com to produce around $103K inflation-adjusted income per year by using approx. $1.8M from IRAs to construct, in six dollar cost averaging purchasing waves, during the set-up dollar-cost averaging purchasing period April 2025 – August 2026. I will hold all these individual TIPs until maturity. I got decent real yields during this period, but didn’t get the real yields spike this past month. Oh well. Can’t time the market. Still satisfied. My $3.3M non-TIPs portfolio is in 60% equities / 40% nominal short-term treasuries. I maxed out my IRAs into the TIPS purchases, but I’m wondering if I should put an additional small allocation from my nominal treasuries that are in my taxable brokerage account into more TIPs ladder holdings at the latest real yields, even considering the income tax drag on the inflation increases in the TIPs, due to that they would be in the taxable account. I wouldn’t need to sell any equities at a capital gain to purchase these additional TIPs. I would just sell some of the nominal short-term treasuries that are already in my taxable account. I don’t like the idea of the tax drag on the inflation increases on any new TIPs that might be purchased in my taxable account, but I like the idea of really good current real TIPs yields and having even more secure income. I’m also concerned about the Federal debt level, and future administrations changing the inflation adjustments or the terms / timing of paying back the Federal debt. So I’m on the fence about it. Would be interested in your insights. Are there any instances where it’s actually worth it to purchase / or add to a TIPs ladder in your taxable brokerage account in this context?
I would avoid TIPS in taxable accounts due to tax drag. It’s similar to the dividend tax drag in taxable accounts, especially in HCOL areas with high state taxes (CA, NY, IL, etc.). It’s clearly not a deal-breaker, but it’s painful. So I suggest keeping the IPTS in tax-deferred accounts.
Oh, good, I’m not the only one excited with those 2.5% real returns on my recent TIPS purchases ! Great write-up and advice, as always!
Oh, and when can we expect front loaded Prison Terms variables in the SWR sheet? 🤣🤣
Thanks! Haha, I will leave that Club Fed modeling to the individual retiree. People should just make sure that they don’t end up in the ADX Florence supermax facility.
Great post. On the gap between TIP’s 4.71% distribution yield and its return: we measured the same thing for nominal Treasury funds. TLT paid 4.86% over the 12 months to Oct 16, 2026 and returned −8.28%. $100 in TLT from Oct 31, 2023, paying out 5% a year, was $91.85 after 35 months; over 255 rolling 36-month windows since 2002, a 5% payout ended under $100 in 118. Method and windows here, in case it is useful for the series: https://portfolio-terminal.com/blog/bond-fund-5-percent-payout-not-return
TIP has a shorter maturity and duration than TLT, so that’s not a fair comparison.
You’re right to point out that distribution yield matters. You can lose a lot more of the principal.
That said, I suspect yields will fall again, so some bond ETFs will pay a handsome yield and can also appreciate.
Great post on a topic that has been top of mind lately! Do you have any thoughts/intuitions on the pros and cons allocating a portion of one’s portfolio to a TIPS ladder while still in an accumulation phase (with retirement still ~5 years out)? Would it make sense to construct the ladder to start providing withdrawals from the TIPS portion of the portfolio starting, for example, 5 years out and building the ladder for the next 25 years to follow?
I think chasing single-digit returns from bond ladders is less advised during accumulation than chasing double-digit returns from stocks. However these are strange times. There are definitely scenarios where the TIPS ladder would outperform a stock-heavy portfolio over the next 5 years, for example. But we get into questions of market timing when we stray from historically optimized portfolios.
Great question!
If you plan to retire in the next few years, I wouldn’t blame you for locking in the TIPS ladder now. Say you plan to retire in 2031; you could map out 2031-2056 with a ladder today (but avoid the TIPS for 2026-2030). The TIPS ladder site lets you choose any payment pattern you like.
I had been waiting for this post, but got restless. The logic and my calculations/projections won out. I purchased a 30-yr TIPS ladder as a liability-matching portfolio to cover baseline expenses, ex-mortgage. For the mortgage, which is too cheap to pay off, I purchased a STRIPS ladder since those payments are fixed nominally. So that’s done. My TIPS ladder still has steps: steps down at 65 a bit due to cheaper health insurance, steps down at age 70 due to the start of SS income. For years 31+ and niceties above the necessities, I’m relying on equity returns, except for the first 8 years, which are in a combo of MM, Tbils, TIPS, bonds. I locked in the first years only because of the market’s ATH and stratospheric CAPE. I don’t have a FIRE approach in terms of a magic number of 25x, so I had some flexibility (being significantly above 25x… due to my own life experience of lifestyle creep and market drawdowns) so I’m able to trade off some potential returns for lower SRR. I imagine that given this current bull market, many readers have similar flexibility.
Oh, TIPS in my taxable are a PITA, but c’est la vie. Year 1 tax return would have been hard to navigate without AI, years since has been follow the prior year. The “phantom income” isn’t a problem in my perspective; I enjoy real, after-tax income.
Thanks for sharing! Good plan! It was hard to time the TIPS yield spike perfectly, so it’s expected you didn’t grab the peak. But the plan looks solid: Customize it to align with your other cash flows (Social Security, etc.).
Good luck!
Thanks for digging into the data on this — great post! I was surprised not to see a mention of thebestthird.com, which has a tool designed specifically to create a portfolio comprised of a TIPS ladder plus equities. The website is run by Stefan Sharkansky, who has also published a paper supporting this strategy: https://www.tandfonline.com/doi/full/10.1080/0015198X.2025.2541567
I tried it out. I don’t like that the tool recommends using TIPS in taxable accounts, though.
Great analysis, Karsten! Thanks for all your efforts.
I’ve been building a TIPS ladder since retiring and grown it in the last year that it is now ~40% of our portfolio – mostly covering our spending needs over the next 10 years. My own analysis using the Toolbox with the particulars of our financial picture prior to this post suggests we should go with an even higher allocation. This post’s analysis confirms that. However, I’m really cautious about putting a majority of our assets in a narrow asset class, even if it’s government bonds.
Is that a valid concern? Or am I over thinking it?
I think people in the FIRE community have a natural tendency to fear allocations where the money will be gone forever: annuities or TIPS ladders that deplete the principal. It must be against our nature as radical accumulators.
Once we get over the emotional hump, a decent TIPS allocation makes sense, though. TIPS are quite safe.
The next level is to diversify currency risk. Ray Dalio and others have been warning of a USD devaluation as the only plausible outcome from a debt/GDP trajectory that seems out of control and with no will to control it.
The problem is that devaluation =/= inflation. The two often go together, but do not necessarily have to. If the USD lost 40% of its value against an international basket of currencies or commodities, that does not mean wages rise or that margins rise. If wages don’t rise, and instead we all just get poorer and work for less purchasing power, then prices for services don’t have to rise. Prices for commodities or imports might rise, but these comprise a minority of CPI.
In this scenario, TIPS would not adjust enough to offset the loss in purchasing power from being in US dollar assets. The people providing you services as a retiree might be working for Mexico-level wages, but the price of goods or energy might be very high. If you ever wanted to move to a more functional country, you’d lack the financial resources.
The devaluation scenario discussed here is what happened with the British Pound in the quarter-century after they went deeply into debt from the World Wars.
What countries will appreciate vs. the USD? Europe? Have you looked at what’s going on in France right now? Will they beat the US economy?
Japan, with its demographic and debt problems?
China with its demographic problems and communism not really working?
Another great analysis/post.
You may want to consider a short regularly updated post (or section of the SWR series intro post) to list a simple scorecard or table of strategies that are worth considering to improve on base equity index/bond SWR/SCR rates and which ones are not. They could link to the blog post supporting each conclusion, but having a short summary table / post with just the bottom line of which approaches work and which ones don’t would be very useful for people who trust your expertise but don’t want to dive into all the detail of each blog post. It would also help counter the naysayers who think you are just being overly conservative and reject every alternate approach.
And maybe describing which ones can be combines and strategies to do so. For example, this was already done with gold in the momentum strategy. But what about previous posts about leverage or how can gold or momentum be incorporated with a TIP ladder? Just make them part of the non-TIPS allocation? Glidepath?
Good point. TIPS ladder with depletion of that position is a glidepath of sorts already.
Haha, I thought the SWR landing page would do that. But who reads a 6000-word post these days? I will consider that suggestion. 🙂
Exactly. The TL; DR version. Minimal to no discussion, just the scorecard in a very simple and concise format, potentially with links to the detailed analysis on each.
If adding all those links is too much trouble, just a statement pointing to the landing page detailed index.
Could be the beginning of the landing page or a standalone post prominently linked from the top of the landing page. Just an easy way for people who already trust your rigorous analysis to bottom line it quickly.
Sounds like a plan. Will work on this!
The issue is that the “rules” change over time as interest rates, valuations, accounting standards, and inflation move around over time.
This post is an example: A heavy TIPS ladder allocation makes sense today whereas it did not make sense 5 years ago.
Similarly, if stocks lost half their value in this quarter, a very high stock allocation with a high WR might start to make sense because CAPE would no longer be elevated.
Also note how nobody’s talking about gold anymore now that juicy bond yields are available. Things change.
Karsten would need to put these tradeoffs into some kind of flexible model to create a single piece of advice that would be applicable years later. Or, he could create a series that says, basically, given today’s conditions this is the AA and WR that makes the most sense right now. The latter is a simpler approach.
Good point about the need for conditionality for a timeless table/summary. Karsten will of course decide what makes most sense to him, but if it were me, I’d rather create a single timeless summary that only needs to be updated with new research, rather than every time market conditions change in a way that materially changes the effectiveness of one of the possibly effective approaches. Simply because the ongoing effort to maintain the summary would become tedious over time and is at greater risk of being out of date. Maintaining the summary would unnecessarily take time away from his ability to publish new research.
The use case I see is for someone who already trusts Karsten’s expertise and rigor, maybe used his SWR toolkit for a tailored SCR, and wants to know what (s)he could consider to increase SCR beyond that number. Just eliminating the myriad things that don’t work, and pruning the list down to the few that might, dramatically simplifies this task. But it only works if the summary remains accurate over .
Karsten could make it timeless by simply saying here is the list that work under some conditions. Read the detailed post for that strategy to see if it applies under current conditions or not. In cases when the condition is easy to describe, he could include it in the summary.
OK, so here’s our situation: we are early retired, have about a $4M total portfolio, $1.5M of which is in a taxable brokerage. We use the CAPE-Based SCR from the toolbox to determine our monthly withdrawals. We have a 50 year horizon, are currently 50% S&P500, 30% Non-US stocks, and 20% 10y US treasury. $1.5M of the $4M is in a taxable brokerage, the rest is in a mix of 401k, tIRA and Roth IRAs between my spouse and me.
Let’s say I want to construct a 30y TIPS ladder using half of our portfolio (all within the 401k/tIRA/Roth accounts); per TIPSLadder.com today that would yield an average of $130,924 p.a.
Would the correct way to model this in the Toolbox be to set the Parameters & Main Results tab as 50% S&P500, 20% 10y treasury, 30% Non-US Stocks, horizon 600 months, FV target to, say, 40% (if we want to leave $800K at death). Then on the cash flow assist tab, set Portf. at Retirement to $2M, and drop in supplemental cash flows of $130,924 annually for 30 years starting Jan 2027?
Assuming this is the correct way to enter the data into the toolbox for a ~50% TIPS Ladder, could we then continue to use the CAPE based rule to know our target withdrawal?
Lastly, my understanding is that money is fungible between our taxable and tax-advantaged accounts, so we’d actually be selling equities from our taxable brokerage and reinvesting the income being generated from the TIPS in our tax-advantaged accounts. What should we reinvest that income into within those accounts?
Good thinking. Did you try this? If yes, does it roughly agree with the SWR gains ERN’s simulation shows?
With 50% reallocated to the TIPS ladder, would you still maintain the same AA (50/20/30) for the other 50% as you currently have?
I have Gemini Pro and I think i got it to do a simulation! Its amazing. It did what you suggested by adding to the supplemental cash flow and creating a new portfolio entry as “Custom: TIPS”. I will need to do a lot more cross-checking, but this looks really promising.
Nice. Thanks for sharing.
Yes, that’s how you’d model the TIPS ladder!
Apologies, I made a mistake on TIPSLadder.com and a 30y ladder @$2MM would yield around $104K annually.
Yes, I am planning to keep that same Asset Allocation for the portion not invested into TIPS.
Ah, yes. That seems more realistic!
Sounds correct: you need to adjust the FV target % to reflect the lower initial portfolio of only $2m.
$130,000 in annual revenue out of a $2m TIPS portfolio seems off. That’s a 6.5% WR. With a 3% yield and a 30-year horizon with depletion, you should get only about 5% or $100k out of a $2m TIPS ladder.
Yes, you can still use the CAPE-based tab. It will take into account 1) your current portfolio, 2) your discounted cash flow of TIPS, 3) bequest target.
Karsten, this post is IMO one of your best ever! Thank you so much.
One small correction, perhaps – both the simulation tables are labeled as “Simulation results: 50 years”. The first one should be 30 years?
I use your SWR toolkit, but my technical understanding of financial calculations is limited at best. How can I do an A/B comparison of my current model in your sheet (I have a 70/30 portfolio with very little in TIPS ETF) with a hypothetical TIPS ladder portfolio. Would appreciate any guidance here.
Thank you!
Thanks! Fixed the typo.
You already found the answer to the other issue. Gemini has figured this one out! 🙂
Even with a modest 18% allocation of total portfolio to TIPS ladder, my SCR for worst case with 0% failure probability jumped from 3.4% to 4%. This is for a 40 year retirement horizon with a 10% Final Target Value.
Does this seem realistic? Seems too good to be true and I am questioning my results.
If this is indeed correct, ERN, do you realize what you have done? This could accelerate ER timelines for so many people if they can pull this off. This is groundbreaking.
Correction – went from ~3.4% to ~3.7%. Not that huge, but still a ~9% jump in SWR!
Yeah, that’s more realistic. Not that groundbreaking. But it’s a good boost in the SWR!
I replied to the other comment: probably check your math. This seems a bit too big of an impact from an 18% TIPS allocation.
I start with the disclosure I am not a fan of TIPS ladders, and prefer owning essentially 100% stocks in retirement (and have been retired for 14 years).
Thanks. I agree with #8, thanks. I disagree with all other points. I can’t addreas them all, but here are a few pointers:
1: The opportunity cost of less long-term growth is small. In fact, the 50y example shows that the median is way overaccumulating, so you should take some chips off the table.
2: False: The strategy hasn’t worked for 20y due to low real rates. Now the real rate is high again, so it should work. In fact, a strategy with 100% stocks has worked for so long, and I have benefitted from it, but that makes a lot of people worried that the party will eventually stop. Like it did in 1929 or 2001.
3: Life expectancy is just the expected value. You may live much longer than the life expectancy. That’s a simple fact from statistics. So, I suggest people target a life horizon much longer than their life expectancy to cover longevity risk. I call this my life horizon. Most FIRE folks can expect to live well past the 30-year horizon of a TIPS.
4: You obviously don’t understand bond math. Bonds trade at/below/above par all the time. That’s not a valid concern.
5: TIPS funds pay back real interest, and a portion of the maturing bond’s principal (i.e., the accumulated CPI adjustments). Just like a TIPS ladder. This concern is invalid.
6: Not a valid concern.
7: Not a valid point. Depends on your life horizon and personal parameters.
Thanks for responding. We will agree to disagree.
Thank you for letting me share my thoughts.
One quick response: You’re misunderstanding — as you have before in Reddit — the issue of asset location for TIPS ladders. You argue that somehow buying a TIPS ladder inside a tax-deferred account creates an additional tax burden for the retiree. That’s incorrect. When you allocate IRA funds to a TIPS ladder, you’re just changing the wrapper that holds the tax-deferred funds. Part of that money has belonged to the US government since the day you earned it. You or your heirs are going to pay ordinary taxes on that money regardless of whether you invest that money in the S&P or a TIPS ladder.
Choosing to hold a TIPS ladder inside the IRA is a neutral move in terms of taxation, so there is no taxation downside when compared to holding TIPS in a taxable account. Taxation is just different for taxable account money, and you could certainly decide that a ladder is worthwhile for you in a taxable account despite the phantom income, but as Big Ern points out, the tax-deferred account is a smart place to hold it.
I don’t misunderstand anything. I challenge the conventional wisdom that, all things being equal, it’s best to hold a TIPS ladder in a tax deferred account.
https://docs.google.com/spreadsheets/d/1uFI5VX2wuTTqrF1gUhf_VDdB92E6nojhd8ZlyfAgfJI/edit?gid=0#gid=0
In the early years, it is marginally more tax efficient to hold in a tax deferred account. In the later years, the holder of the TIPS ladder has converted the return of capital into taxable income, and substantially more over time.
People contemplate the tax implications of RMDs versus Roth Conversions over time. Placing a ladder in a tax deferred vehicle, based on running the numbers over the 30 years, substantially increases taxable income for the holder. That is definitely a valid concern that cannot be hand waved away.
Nope. The ladder by no means “substantially increases taxable income for the holder.” You can’t on the one hand claim that TIPS are a lame investment because they don’t return an amount equal to equities and then in the same breath argue that by year 30 the income generated is somehow creating a higher tax bill than the investor would have otherwise experienced.
To repeat, a TIPS ladder inside an IRA doesn’t generate taxes that are somehow proportionally higher than they would have been without the ladder. Sure, if you parked all that TIPS money in cash instead of TIPS, there would likely be growth that doesn’t keep up with inflation and that would mean fewer taxes, but there would also be much less income generated.
The money in the IRA is going to be taxed as ordinary income whenever you withdraw it, full stop. You can invest the IRA money in whatever you want to. If your entire investment philosophy is total return, then sure, just put the whole thing in equities and pay ordinary taxes on the principal and proceeds whenever you withdraw. Or invest part of it in TIPS and, yes, pay ordinary taxes whenever you withdraw. Or hold cash or cash-like assets in that IRA and pay ordinary taxes when you withdraw. There’s nothing special about TIPS income that creates an outsized tax bill. Everything that comes out of that IRA is subject to ordinary taxation.
All of that said, what happens if we just give you the benefit of the doubt and do an evidence-based evaluation of your comment and the accompanying graph? It doesn’t work out well. You’re committing a textbook accounting fallacy by isolating one asset class in a vacuum and confusing gross withdrawals with net economic tax cost.
If that same pre-tax space had been occupied by broad equities (which historically grow at a higher expected nominal rate), the ending account balance and subsequent Required Minimum Distributions (RMDs) would be dramatically larger, resulting in a far higher annual tax burden in later years than a modest, self-liquidating TIPS ladder would ever produce.
A maturing TIPS ladder is designed for liability-matching decumulation. Once the final rung matures, that pre-tax space is exhausted—it doesn’t leave behind a bloated, RMD-generating monster.
Touting a line graph showing nominal IRA distributions trending upward over 30 years without adjusting for inflation, phantom tax drag in taxable accounts, or the alternative RMD impact of higher-growth assets doesn’t prove holding TIPS in an IRA is inefficient—it just misattributes the fundamental mechanics of a traditional IRA to the asset sitting inside it. Also, the actual economic value and cash flows delivered to the retiree are identical in the taxable and tax-deferred versions.
Your comparison is fundamentally misleading. While your blue line shows taxable income shrinking as the ladder ends, in the first 15–20 years, the taxable ladder creates a serious cash-flow headache. You owe ordinary income tax each year on inflation adjustments you haven’t received in cash yet. In an IRA, those phantom taxes are sheltered.
The money sitting in the traditional IRA went in pre-tax decades ago (or grew tax-deferred). The principal inside an IRA was already owed to the IRS. The TIPS ladder didn’t convert principal into income; the IRA wrapper itself dictates that all distributions are taxable, regardless of whether you hold TIPS, cash, Apple stock, or gold.
To build a $1,000,000 TIPS ladder in a taxable account, you had to earn roughly $1,300,000 to $1,400,000 upfront and pay income taxes on it before buying the bonds. In an IRA, you built that ladder with pre-tax dollars. Celebrating that a taxable ladder has lower taxes on withdrawal completely ignores the massive tax bill paid upfront.
For 3 you make a good point. A 30 year TIPS ladder at the age of 65 may indeed be too long, and a 20-25 year ladder is more likely to be spent before death. Longer ladders can be useful for early retirees, say, the joint life expectancy of a male/female couple at 55 is over 30 years.
As Mr. Unhappy doesn’t think I have the right to express my opinions, here is exactly what I wrote over on Reddit about male and female mortality and the back ended value of TIPS ladders.
Here are the mortality percentages for male and female, compared to age 65.
Investing in long term assets where the majority of the benefit–the escalation of the inflation adjustments PLUS getting the substantial discount at purchase–at maturity, when maturity is past your life expectancy, makes zero sense to me.
Just out of curiosity: where’s that offensive Reddit post? Please share the link; I’m curious now, haha!
There were three.
https://www.reddit.com/r/Bogleheads/s/tmzVzW9hqI
https://www.reddit.com/r/Bogleheads/s/g0N43GVV8s
https://www.reddit.com/r/Bogleheads/s/snCKvhN0UH
Happy to share, but did not feel it was fair to you for me to spam your comment section with what I wrote.
Yeah, it’s one of those Brandolini’s law examples. It would take me way too long to debunk all those. I will leave at that.
Karsten took the smart approach and decided your long missives were so misguided that they didn’t warrant engagement. I probably would have saved myself a lot of time if I had invoked Brandolini’s law as well.
Haha!
Your responses show zero concern for sequence of risk. If you can fund your spending with 2.5% or less of your portfolio then you can afford to take this attitude and you are not really the audience of this post (or most of the blog in general).
I think SORR is vastly overrated. Your number of 2.5% is also wrong.
Quoting a source who has run the numbers:
The above is from Part 1 of this series.
https://earlyretirementnow.com/2016/12/07/the-ultimate-guide-to-safe-withdrawal-rates-part-1-intro/
Consistent with Part 1, written in December 2016, I have been virtually 100% stocks over that time. That has proven to be a far better choice than virtually all of the alternatives, such as being 75/25 stocks/bonds, as bond performance over that period has been below inflation.
The 2.5% Marc was referring to is likely from Part 2 of my series, the large table: Failsafe of 60y retirement with 100% stocks: 2.58%
You’re right, there are instances where the 4% rule’s success probability is highest, but that’s a different objective function. For an unrealistic goal, it usually makes sense to roll the dice. But the failures would be more catastrophic. And the failsafe WR is usually maximized with around 75/25.
Please don’t quote me on that. That was when yields were really low. Today’s bond yields are much more attractive, especially compared to earnings yields.
Its ironic that you are attempting to use this blog to prove a point that SORR is vastly overrated. It looks like you retired near the beginning of a very long bull run. No wonder you are pro 100% stocks. Its possible that the CAPE was low enough at that time that you could make a good argument for 100% stocks. But the situation is different now with the elevated CAPE. As somebody who has been 100% stocks during accumulation, the numbers are clear that I would significantly increase my safe withdrawal rate with some form of glidepath/tips ladder.
Hi Marc, when you say “glidepath/tips ladder”, do you just mean that you cover your full expenses in the early years using a tips ladder and leave your equity portfolio to grow and so once the tips ladder is depleted you will naturally have increased your exposure to equities because you spent down all the bond exposure that bought the tips ladder?
That is correct, although I will probably end up doing a 10 year cash glidepath instead of a tips ladder for a couple of reasons (the biggest being that I am not quite ready to pull the trigger and the rates are already going down).
Thanks for your reply!
1: I didn’t mean it that way. Overaccumulation is a good problem to have. I simply stated that you can raise your failsafe withdrawal rate without giving up much upside.
2: Again, you don’t understand bond math. The low yields of the past and the losses in 10y bonds (even worse in 30y bonds) are water under the bridge. You have 20/20 hindsight bias. Knowing that in 2018, I would have stayed 100% equities. It’s my duty to inform my readers
3: Here are survivorship probabilities for an extremely healthy 65yo couple:
Prob of being alive at the end of the Simulation Horizon (30y, age 95 for both):
Spouse 1 22.62% (male)
Spouse 2 32.18% (female)
Only Spouse 1 15.34%
Only Spouse 2 24.90%
Both 7.28%
Either (joint survivor) 47.52%
5: The TIPS ETF currently pays about 5% in taxable distributions. That’s about the same as a TIPS ladder when you account for the real rate and CPI adjustments. Pro tip: Check out the 2022 distributions: you will see a massive taxable income flow. Essentially all from CPI adjustments. Real yields were zero at that point.
6: Others have weighed in on the tax issue. I agree with them disagree with you.
7: Fair point. Maybe a 10y TIPS ladder is all you need to hedge out the worst part of SoRR. But I know a lot of retirees who like to hedge longer. The last time we had a CAPE ratio this high, in 2000, the equity market would still be down quite badly in 2010, especially in real terms.
I will try to keep my replies brief.
Sageillini, your third point is a prime example of how you are not to be trusted with math and statistics.
You’re confusing conditional probability with unconditional probability. You took the cohort survival count at age 95 and divided it directly by the cohort survival count at age 65 out of the original birth cohort of 100,000.
Here is why this math is voodoo and what the SSA Table 4C.6 actually says:
Social Security Administration (SSA) period life tables track a hypothetical cohort of 100,000 live births. Every number in the “Number Alive at Age x” column represents how many people out of those original 100,000 babies are still alive at age x.
Out of 100,000 babies born, 79,094 make it to their 65th birthday. (Roughly 21% of the original babies died before age 65).
Out of those same 100,000 original babies, 6,333 make it to their 95th birthday.
You calculated 6,333/79,094 which comes to 8%. By dividing 6,333 by 79,094, you are including all the males who already died between birth and age 65 in the denominator. You are essentially asking: “Out of 100,000 newborn babies, what percentage will make it to 95 AND were part of the group that survived to 65?” That answers a completely different question.
According to the SSA data, a 65-year-old male has an average remaining life expectancy of roughly 17 to 18 years (taking him to age 82 or 83).
You think you caught a flaw in the math, but you just proved you don’t know the difference between a static one-year population snapshot and actual cohort longevity.
This is what the author of this blog wrote. I do not know what “at the end of the Simulation Horizon” means. The author can clarify, which is why I wrote the response to HIM.
https://www.ssa.gov/oact/STATS/table4c6.html
That table shows the life expectancy of a 65 y/o male is 18.12 years, which means 83.
https://www.ssa.gov/cgi-bin/longevity.cgi
That SSA calculator shows someone who turns 65 today, born 10/29/1961, has a 84.3 year life expectancy (which is the average).
So even with simple math, which you might even be able to understand, if a 65 y/o male buys a TIPS longer than a 2045 issue (here is the math, 1961 + 84 = 2045, do you agree with that?????) there is more than a 50% chance they will not be alive to see the maturity of that TIPS.
And if the simulation horizon is 30 years–and again I do not know, but this blog projects out 30 years–the 22.62% chance the male is alive at 95 is inconsistent with the social security life expectancy tables and with effing common sense, because one will be lucky to know a handful of 95 year old males, much less 22%.
And the math is exactly correct. If 79,094 are alive at 65, and 6,333 are alive at 95, the typical male has an 8% chance of making it from 65 to 95. But you don’t even have the gonads to admit I’m right.
30 years, what else?
Regarding 4, please help me understand. Let’s say you have 2M in taxable all in stocks and 2M in tax-deferred in the same stocks. You are trying to decide in which account to buy a 1M ten-year TIPS ladder to take advantage of the current high yields. That way you move from 100% stocks to 75/25. Let’s say you buy TIPS via target-maturity ETFs, so that the inflation adjustments are paid every year, and there is no phantom tax issue. If you buy in taxable, you will have to sell 1M of stocks realizing all the gains, and over the next ten years you will also be paying taxes on the ladder income. If you buy TIPS ladder in tax-deferred, the only possible tax consequence is that the TIPS ladder will have a different return from the stock portfolio, and you have to eventually pay the tax on the difference, when you withdraw from tax-deferred. If the ladder and stocks have the same return over the next decade, which seems possible due to high CAPE and high TIPS yields, then there is no tax consequence. The conclusion is that you will probably pay less tax if TIPS are in tax-deferred. What am I missing in your opinion?
You’re attempting to reason with an unreasonable person. Sageillini somehow amassed a lot of wealth, retired into a historic secular bear market and therefore easily did well with a 100% equity stake between then and now. He takes that n of 1 and extrapolates that to the world at large, arguing that everybody should be 100% in equities 100% of the time. A single academic paper by Cederburg supports his position, but tons of academic work and lived experience of investors suggest that he’s fortunate to have benefited from circumstances that may or may not make his approach a winner in the future. The alternative narratives in support of strategies like a TIPS ladders upset his worldview and threaten his conviction that his narrow path is the only true path. That bias taints his ability to anchor his critique of TIPS in actual verifiable math and statistics, and he spends a lot of time torturing numbers, trying to cram them into his small view of the world. It’s all pretty transparent and it falls apart when you subject his arguments to a basic level of evidence-based scrutiny. His distorted math gives him false confidence that he’s cracked a code that the entire financial services industry and reasonably well-informed lay people are apparently unwilling or unable to acknowledge.
It’s almost impossible to answer this question without more specifics. For example, how old are the individuals (the standard deduction changes at 65). What is the basis in that $1 MM of stock sales? 25%? 50%? Are they taking social security (or not).
For example, lots of people want to do Roth conversions. The advice is to use funds outside the IRA you want to convert to pay the tax to maximize the benefit. But like your example, the investor may incur a significant amount of capital gains tax to get the cash to pay the upfront tax on the conversion–and that may nix the deal.
Or what if the investor is currently 75/25–as this blog advocates–with the 25% already in bonds in the taxable account? Those bonds might be held at little gain or (very possibly) a loss?
In your example, it makes little sense to sell $1 MM in stock all at once in a taxable account. For one, you get in higher brackets and incur the net investment income tax. That has nothing to do with whether one wants to invest in TIPS. But selling stocks to fund annual retirement cash flows is more tax efficient than withdrawing a similar amount from a tax deferred account, because the return of the basis is tax free and capital gains have preferential rates.
And I’m not sure what you mean by ladder income? Are you talking about the total cash flows or just the (approximately) 23% of additional income from coupons, inflation adjustments, and recognizing the discount?
But again, all things being equal, between choosing to have the TIPS (actual TIPS) in a tax-deferred account converts the return of capital–77% in my example in a 10 year ladder–to taxable income. But everyone’s situation is different. But an investor with significant investments in their taxable account that is reasonably tax efficient to access, they will be better off buying the ladder in a taxable account.
You continuing to double-down on not understanding what the ‘advantaged’ part of tax-advantaged accounts means is a very impressive lesson in stubborn wrongheadedness that only the internet can provide.
3: You are confusing expected lifetime (which means you might outlive that in roughly 50% of the cases, which is not a good way to plan your retirement) and potential/likely survivorship of a couple.
You also ignore that not all Americans have the same survivor probabilities. A healthy retired white-collar worker has a massively longer life expectancy than in the
And finally, you’re confused about single survivorship vs. joint survivorship.
4: You are too confused about the tax treatment; I just give up. Just let it be.
5: No, you’re wrong. I’ve done simulation with both 10y and 30y TIPS ladders. For most examples I came up with the 30y ladder was better.
Not thrilled to see you outside of Reddit spreading TIPS hatred with your usual stale and false arguments. However, I’m very happy to see that Big Ern is shooting you down on multiple levels.
You need to one day realize that your ultra-early retirement and your successful use of 100% equities throughout that retirement are both outliers. Very few people can retire as early as you did. And only well-overfunded retirees have the luxury of dodging sequence of returns risk whole staying in 100% equities.
As usual, your mistake here is that you like to treat your n of 1 as though it tells a statistically valid story for everyone else. Unlike you, I recognize that your path can work for some people some of the time. You take the opposite position, acting like the rest of us who see the value of TIPS are making irrational decisions and you fail to acknowledge the very basic valid answers to the questions you pose.
You’re older and set in your ways, but you may want to consider opening your mind and allowing for the possibility that you can be successful with your plan and we can be successful with ours.
Honestly, your obsession with repeated failed efforts to poke holes in the TIPS narrative is a bit strange. It’s also unique. I’m honestly wondering if there may be a screw loose that’s driving this campaign of yours.
If he changed his 3.5% claim of SWR given 100% equities above to 2.5%, he’d basically be correct.
And the rest of us would merely be pointing out that he could safely have a higher withdrawal rate with better portfolio allocation if he wished.
Of course, that’s moot if he just wants to maximize expected value of bequest.
But his claim of 3.5% withdrawal rate being safe for 100% equities is simply false.
As anyone who has read this site or can use ERN’s sheet already knows.
I tested out glidepath strategies using the TIPS ETF return data and it provided higher SWR’s at every bequest amount that i tried and for longer retirement lengths.
Have you looked at glidepaths for this instead of using ladders?
I haven’t added this to a glidepath model. The TIPS ladder already acts like a GP. In fact, I’d almost assume that it’s the other way around to get even better results: Use a TIPS ladder to hedge your expenses for the first 10-30 years. During that time, keep 100% equities. But when the TIPS ladder runs out, do a 75/25 again to hedge the SoRR at that new retirement date.
Thanks for doing this Karsten! I, and others, asked in the comments of the Bengen post about doing a TIPS blog post and you seemed to question if there was enough to make it worth doing. Incredible that you put in the time and effort (that I and others aren’t capable of doing) and learned something yourself and then shared it all with us in a very humble way. Crazy that when I posted my comment it was October 8th and the 30 yr real yield was 3.38%, you used 3.17% in the article (based on the 10/16 data) and today on 10/26 it is 3.07%.
Quick question, I’ve seen some people suggest that with the US debt where it is we could see the US fed cap interest rates to manage the debt (financial repression). I don’t really believe this will happen but if it did would it make a ladder of longer term TIPS (or nominals) more attractive since these bonds aren’t callable? Could you see the fed actually capping interest rates to manage the ballooning debt? Here’s an article about it in the 40s – https://www.federalreservehistory.org/essays/wwii-to-the-treasury-fed-accord
And a paper that discusses it happening again – https://www.imf.org/-/media/files/publications/wp/2026/english/wpiea2026160-source-pdf.pdf
I hope we don’t get there. Europe would be the first to try this, though.
Thanks! Yes, I realized I needed to push this out before it got too late, haha!
Whoa, financial repression?! Hard to believe we’ll see that here.
Just recently played around with the updated calculator sheet. Really looking forward to your updated post about non-US stocks soon. I remember commenting earlier about the South Korean investors having a hot degen summer propping up Samsung and SK Hynix via leveraged ETFs in https://earlyretirementnow.com/2017/08/23/how-useful-is-international-diversification/
Just like the bell curve meme in that earlier post, I’m more of a VT and chill guy as the dumbass on the left end. Apparently putting 64% in S&P 500 and 36% in non-US stocks (basically VT weighting as of this writing) gets me a SCR of ~3.5% with the same failure rate as rolling a nat 1 on a d20 (5%). I like those odds lol
Yeah, I updated the international return series. For most investors, it wouldn’t make too much of a difference.
Not going to lie, I actually love the D20 reference for SWR long tail odds management – for my longer horizon with a bunch of COLA adjusted consistent payments I can get a number closer to 3.65% with slightly higher equities allocation, a TIPS ladder, bonds, and some precision metal/countercyclical mixes. I’m trying to push it to where I would need successive Nat 1’s on a D20, which does put me closer to 3.3% unless I start playing flexibility games (because I know my mandatory costs won’t change, but I’m probably willing to make CAEP-based adjustments on foreign travel or my spending line categorized as ‘bougie restaurant experiences’).
Not sure what Nat 1 and D20 refer to, but I’m glad you found this post useful.
LOL, its a D&D (Dungeons and Dragons) reference. A D20 is a twenty sided dice used commonly in that game and a Nat 1 means a natural (unmodified) 1 being rolled. Its basically means the worst 5% outcome (and a double Nat 1 would be .25%).
Ahh, that makes sense now! Thanks for clarifying!
great post! good to see a seal of approval on today’s real rates. I figure a very human approach might be to ladder some comfortable amount of undeniable annual spending into a ladder such as 15 or 25k/annum knowing that you’re boosting your SWR to some degree
Also, good to see that a moderate approach is best bc the only way I can ladder TIPS in a retirement account is via my solok so going big isn’t even an option.
Yes, that’s a good way to implement the TIPS ladder. Note that I just did a very generic case, but you will do well customizing this to your situation.
Thank you for another great article!
I’m getting close to retirement and only discovered TIPS recently. Since then, I’ve been reading every book and article I can find on the subject, including your earlier article, Part 61.
When I read the old article, Part 61 a month back, with TIPS offering real yields close to 3%, I wasn’t fully convinced by some of the conclusions, especially since the analysis was based on the much lower yields available in 2024.
After a lot of research, I’ve become a big fan of TIPS and concluded that a TIPS ladder is the best approach for my retirement needs. I was comfortable enough with the strategy to move 20% of my total portfolio into a 30-year TIPS ladder during the second week of October, locking in those attractive real yields.
It’s great to see your latest article, with backtesting using current TIPS yields, supporting some of the same conclusions I had reached. My retirement portfolio is now structured around a 60/20/20 allocation to stocks, nominal bonds, and a TIPS ladder.
Thanks again for the excellent research and for sharing your insights!
Thanks! Yes, there is definitely a sweet spot now. A while ago, with real yields of about 2.2%, the TIPS ladder would not really contribute that much. But it would have huge opportunity costs (lost upside). But with the current yields, it’s hard to ignore.