Treasury Inflation-Protected Securities

‘Inflation Guy’ explains seasonal adjustment (or lack thereof)

Posted on September 10, 2023 by Tipswatch

By David Enna, Tipswatch.com

I often caution readers about “non-seasonally adjusted inflation” and how it can create confusion and even distress as it sometimes skews away from headline inflation, the number you see reported each month.

Non-seasonally adjusted inflation is important for investors in Treasury Inflation-Protected Securities (where it is used to adjust principal balances each month) and U.S. Series I Savings Bonds (where it is used to set future variable interest rates).

Each month, the Bureau of Labor Statistics includes a paragraph deep down in its CPI report (this one is for July inflation):

Not seasonally adjusted CPI measures

The Consumer Price Index for All Urban Consumers (CPI-U) increased 3.2 percent over the last 12 months to an index level of 305.691 (1982-84=100). For the month, the index increased 0.2 percent prior to seasonal adjustment.

Once the inflation report is issued, I plug that number into the mega-chart on my Inflation and I Bonds page:

This chart shows you two things: 1) non-seasonally adjusted inflation ran at 0.19% in July (not the rounded 0.2% reported by the BLS) and that means principal balances for all TIPS will increase 0.19% in September. And 2) Four months into the I Bond’s rate-setting period, inflation has run at 1.28%, with two months remaining. The I Bond’s new variable rate — to be reset November 1 — appears to be heading toward a range of 3.2% to 3.4%.

What is this NSA inflation?

A new podcast from Michael Ashton, known around the financial world as the “Inflation Guy,” takes a crack at explaining non-seasonally adjusted inflation (NSA) in plain English. Although this is an esoteric, inside-baseball topic — and Ashton actually uses baseball as an analogy — it’s important for investors in TIPS and I Bonds to understand the basics of NSA. Take a listen:

https://t.co/cujQPbItDh
— Michael Ashton (@inflation_guy) August 31, 2023

In his intro, Ashton notes that the inflation derivatives market is suggesting that U.S. inflation will decline from August to December 2023. He notes:

Does this mean that deflation is upon us? No, and the reason why is something called ‘seasonal adjustment.’ It turns out that seasonal adjustment is critically important for inflation markets, and the market’s pricing is not nearly as outlandish as it looks.

Ashton clarifies something that I have noticed and written about over the years: that NSA tends to run higher than seasonally-adjusted ‘headline’ inflation from January to June, and then lower from July to December. At the end of 12 months, both NSA and seasonally-adjusted inflation will match. So if you run higher in the first half of the year, you’ll run lower in the second half.

He notes that the average investor has no interest in NSA versus official inflation, but investors in TIPS and I Bonds do:

Inflation people definitely care, because what gets traded in the market, what TIPS are based on, is non-seasonally adjusted inflation, whereas the number that gets reported, the headline number, is seasonally adjusted. ….

We know that prices have a strong tendency to fall in December, so we can adjust that number and then when see the official CPI report the December number is no more likely to be negative than any other CPI number.

The reality is that prices did decline in that particular December, so bonds tied to inflation ought to reflect the reality, not the seasonally adjusted reality.

This isn’t just a “theoretical” premise — in fact, non-seasonally adjusted inflation consistently runs lower in December than the official seasonally adjusted number.

So, what does this mean for investors of inflation-protected products? The key thing is to recognize that your inflation expectations could take a temporary dive toward the end of the year.

For TIPS investors, the December number sets inflation index for the month of February. As Ashton points out, that means that a TIPS maturing in January (with inflation accruals set by inflation through November) could logically have a lower real yield to maturity than a TIPS maturing in April (with inflation accruals including the month of December). He says:

If you look at a whole list of TIPS, the whole TIPS yield curve, you’ll notice that TIPS that mature in April very often have slightly higher yields than similar maturity bonds that mature in January or July. … You’ll get these little bumps every April.

Given a choice between those two bonds, I’d always rather have the January maturity that doesn’t have to experience that print that I am pretty sure is going to be negative. So investors have to be compensated to hold the April maturity.

This doesn’t always hold up in reality, because TIPS have a lot of factors affecting real yield to maturity. A TIPS with a very high inflation accrual and high coupon rate will generally draw a slightly higher real yield than one with a low inflation accrual and discounted price. Plus, today’s TIPS market is skewed toward higher yields on shorter maturities.

In all these issues through 2028, the January TIPS (highlighted in green) has a higher inflation accrual and generally a higher coupon rate than the TIPS maturing in April and July of the same year. And that tends to push the real yield higher. So at this point, do the TIPS maturing in January look relatively more attractive? I’d say yes, but that’s a quick observation.

But if you take a careful look at this chart, you can see that the TIPS maturing in April of each year consistently have a higher real yield to maturity than the TIPS maturing in July, even with the variance in coupon rates and accrued inflation. That is evidence of the “NSA effect” because the TIPS maturing in July gets the benefit of generally higher NSA from January to May.

Is NSA deflation concerning?

Ashton points out that inflation derivative markets are pricing in slight deflation through the end of the year.

The answer is it’s not terribly unusual to see NSA deflation especially when inflation is kind of low in those last four months of the year. … What the SWAP market is telling you is yes, inflation is expected to be unchanged or slightly lower. … But once you seasonally adjust those numbers you get something like a 2.2% annualized rate.

I Bond investors are less affected by these seasonal skews because the rate-setting periods for I Bonds — April to September and then October to March — cross through the high months and low months and more or less balance out.

Last year, for example, non-seasonal inflation ran at -0.10% in November and -0.31% in December and then jumped to 0.80% in January and 0.56% in February 2023.

The main take-away is to realize that inflation-protected investments are are tied to non-seasonally adjusted inflation and not the headline number you see reported each month. And that will mean some ups and downs across the year, but it all balances out after 12 months.

* * *

Feel free to post comments or questions below. If it is your first-ever comment, it will have to wait for moderation. After that, your comments will automatically appear. Please stay on topic and avoid political tirades.

David Enna is a financial journalist, not a financial adviser. He is not selling or profiting from any investment discussed. I Bonds and TIPS are not “get rich” investments; they are best used for capital preservation and inflation protection. They can be purchased through the Treasury or other providers without fees, commissions or carrying charges. Please do your own research before investing.

Posted in Investing in TIPS | 20 Comments

Question: What is a ‘normal’ real yield?

Posted on September 7, 2023 by Tipswatch

By David Enna, Tipswatch.com

One of the very first Treasury Inflation-Protected Securities I ever bought — CUSIP 912810FH6 — was auctioned on April 7, 1999, with a real yield to maturity of 3.899% and a coupon rate of 3.875%. I still own that little gem, which has racked up an inflation index of 1.85668 on its way to a maturity date of April 15, 2029.

As great as that investment was, it spoiled me. Real yields didn’t hang near 4.0% for long. Soon we were seeing 2.7%, then 2.5% then 2.0%. Those were still great real yields, but not my highly coveted 3.899%. So I lost interest in TIPS for awhile.

Now, after more than a decade of very low and even negative real yields on TIPS, we are again back to a 2.0% real yield across nearly the entire TIPS maturity spectrum. I’ve been investing aggressively at these yields, but there is always the question: Are we arriving at a “new normal”? Here is where we stand today:

Does this look normal? I’d say “not at all,” because this yield curve is extremely flat. In normal times, you’d expect longer-term yields to be higher than shorter-term yields. To get to normal, either the longer-end needs to rise or the shorter end needs to fall.

But then, what is “normal”? I don’t know the answer; I wish I did. Can recent history tell us? I looked back through real yield information compiled by the Federal Reserve Bank of St. Louis, with daily real yield estimates dating back to 2003 for the 5- and 10-year TIPS; 2004 for the 20-year; and 2010 for the 30-year.

Can charts tell the story?

In each of the following charts you will note the spike higher in real yields during the financial crisis of late 2008 and then again in the pandemic distress of March 2020. In both those cases, panicked investors were selling everything. Yields returned to more typical levels relatively quickly.

But after each of those market panics, the Federal Reserve reacted with aggressive quantitative easing (bond buying that pushed yields down) and slashes to short-term interest rates, resulting in a near 0.0% return on safe investments.

So the case can be made that the entire period from 2008 to 2022 was not normal, subject to aggressive market manipulation by the Federal Reserve. That era ended in March 2022, when the Fed abandoned quantitative easing and began raising short-term interest rates, first slowly and then aggressively.

Here are the charts:

During the 2022 to 2023 cycle of Federal Reserve tightening, the 5-year real yield has been the quickest to respond to short-term rate increases and has been lingering around 2.0% since early July 2023. But as you can see in the chart, the 5-year has rarely broken through the 2.0% real-yield barrier at other times back to 2003.

My guess: The “normal” real yield of a 5-year TIPS appears to be in the range of 1.25% to 2.0%.

Surprisingly, 10-year real yields have been slow to creep up to the 2.0% level, but on Wednesday the most recent issue of this term was trading at 2.0% on the secondary market. Note that a real yield around 2.0% was common for the 10-year before the financial crisis of 2008.

My guess: A “normal” 10-year real yield appears to be in a range of 1.5% to 2.2%.

FRED’s tracking of the 20-year real yield dates back to 2004. Note that in the four years before the 2008 financial crisis, the 20-year real yield ran very close to — or well above — the 2.0% barrier.

My guess: It’s possible that 1.8% to 2.2% could be a baseline “normal” real yield for a 20-year TIPS.

FRED’s records for 30-year TIPS only go back to 2010 — two years after the turmoil of the financial crisis — so it’s hard to draw any conclusions. The Treasury did not issue 30-year TIPS from October 2001 tot February 2010. I consider the early years of these long-term TIPS, from 1998 to 2001, to be unreliable as a gauge for a “normal” 30-year real yield. Over that time, auctioned real yields ranged from 3.4% to 4.1%. I don’t think we are returning to those “glory days.”

My guess: I’d think that 2.0% to 2.25% could a baseline normal yield for a 30-year TIPS.

Can the U.S. economy handle this?

It has been surprising to see the U.S. economy continue to hum along even as short- and long-term interest rates have increased 300 to 500 basis points. No one seemed to expect this; the recession was supposed to start months ago. But because it hasn’t, the Federal Reserve has kept up the fight against inflation.

Bloomberg this week posted an article theorizing that the surge in U.S. interest rates could hamper the U.S. economy for at least a decade. From the article:

Central bank interest-rate increases reduce potential economic output for at least 12 years, in contrast to traditional theories of national economies that assume policy is neutral in the long run, Federal Reserve Bank of San Francisco research found. … For example, in response to a 1% interest-rate increase, real gross domestic product would be about 5% lower after 12 years than it would otherwise be, the researchers found.

In other words, the U.S. economy will feel pain as interest rates rise after years of near-zero yields on safe investments. But those yields were artificially created through Fed manipulation. The yields we are seeing in September 2023 are rising to historically valid levels, with almost no Fed manipulation on mid- to longer-term Treasurys.

The long view

The St. Louis Fed does track nominal 10-year yields all the way back to 1963, so you can get a better idea of the massive fluctuations in yields over the last 60 years, a period that included an intense battle against raging inflation that lasted into the early 1980s. Here is the chart:

I highlighted the 6.74% yield in January 2000, which marked the last period the nominal 10-year was yielding more than 6%. On Jan. 12, 2000, a 10-year TIPS auctioned with a real yield of 4.338%. Oh, and by the way, the U.S. inflation rate in January 2000 was 2.7%.

I note this because I’ve seen recent commentary (such as this from Bloomberg) suggesting that the the 10-year note could be heading to 6%. For example:

That’s an eye-popping figure. And Kaminski acknowledged that we’re coming off a “decade of really low interest rates.” But we’ve seen 6% rates before, she said, “so it’s not strange to think that we may have higher rates if we have surprises to the upside in inflation later in the fall.”

Katy Kaminski of AlphaSimplex says a 6% 10-Year is not crazy https://t.co/yudw7B1mXQ pic.twitter.com/IaBzdqRaW5
— BSurveillance (@bsurveillance) August 22, 2023

Final thoughts

The point is: It’s a new, uncertain era.

We are normalizing, but at this point we don’t know what the “new normal” is. How high can real yields go? Will we overshoot to the high side? And, can it last?

Do I know? Of course not. But it is clear that today’s real yields of 2.0% above inflation are historically attractive.

• Now is an ideal time to build a TIPS ladder

• Confused by TIPS? Read my Q&A on TIPS

• TIPS in depth: Understand the language

• TIPS on the secondary market: Things to consider

• Upcoming schedule of TIPS auctions

* * *

Posted in Federal Reserve, Investing in TIPS | 11 Comments

Experiment: Let’s try out a very short-term TIPS

Posted on September 3, 2023 by Tipswatch

By David Enna, Tipswatch.com

Last week I wrote about my recent buying spree of medium- to longer-term Treasury Inflation Protected Securities, combined with some shorter-term nominal Treasurys and bank CDs. My goal is to take advantage of current high real yields to fill my fixed-income ladder out to 2043.

In the feedback I got to that article, several readers asked about the attractiveness of very short-term TIPS, which appear to have above-inflation yields higher than 3.6%. For example, at Friday’s close:

In general, I have recently preferred to use T-bills, bank CDs, and short-term Treasury notes for investments of less than 5 years, and TIPS for investments of 5 years or more more. That’s mainly because short-term nominal rates have been so appealing. But I decided to take a look at CUSIP 912828B25, the TIPS maturing January 15, 2024.

The goal: To set aside $15,000 for a January 2024 purchase of a new 10-year TIPS, which will mature in 2034. At this point, there no TIPS that mature in 2034 and I want to add this one to my investment ladder. I’ll probably buy it at the originating auction on January 18, 2024.

The account: Traditional IRA at Vanguard. Money for this purchase is being withdrawn from my holdings in Vanguard’s Short-Term TIPS ETF (VTIP).

The question: How would investing in CUSIP 912828B25 — maturing Jan 15 2024 — compare with investing in a 17-week T-bill, maturing Jan 2 2024?

So last week, as an experiment, I bought $12,000 par value of CUSIP 912828B25 after doing a quick analysis of the likely results. Nothing is certain, of course. But let’s take a look.

Note that to come up with proceeds of $15,000+ in January, I needed to purchase $12,000 par value of this TIPS. That is because it carries a lofty inflation index of 1.30749, meaning my $12,000 par would actually purchase $15,689.88 of principal. The coupon rate is only 0.625%, so this TIPS sold at a discounted price of 98.73.

The end result was that I paid $15,490.62 for $15,689.88 of principal. That is an immediate gain of 1.29%, or an annualized return of about 3.92% if nothing else happened until maturity. But … two things will happen: 1) The principal balance will continue growing (or possibly falling) along with U.S. inflation from August to November, and 2) there will be a final coupon payment of 0.312% on Jan. 15 based on the ending inflation-adjusted principal.

Also, remember that inflation accruals for TIPS are based on inflation two months earlier. The September inflation accruals have already been set by the 0.19% non-seasonally adjusted inflation reported for July. We already know the inflation index for this TIPS on Sept. 30. What we don’t know are the indexes for October (based on August inflation), November (September), December (October) and half of January (November).

My scenarios look at how this TIPS would perform if non-seasonally adjusted inflation runs at 0.2%, 0.1% or 0.0% for the months of August to November.

Based on this analysis, CUSIP 912828B25 could create a nominal annualized return of nearly 7.5% if inflation runs at 0.2% a month from August to November. That number drops to about 6.4% if inflation runs at 0.1% and 5.1% if inflation remains flat throughout those months.

Compare that to the return of a 17-week Treasury bill sold at auction last week:

The TIPS is the clear winner if inflation runs at a monthly average of 0.1% or 0.2% through November. But the T-bill is the winner if inflation is flat or declines in those months.

Where could this go wrong?

I do think deflation risk is higher for a very short-term TIPS (especially one with a high inflation accrual) than for a longer-term TIPS. The long-term investment has time to make up for a few deflationary months. The short-term investment takes an immediate hit.

In the closing months of the year, non-seasonally adjusted inflation tends to run lower than the official seasonally-adjusted CPI number you see reported each month. Last year, for example, non-seasonal inflation came in at -0.10% in November and -0.31% in December. So it is possible we could see a deflationary month before the end of the year, which would likely make the 17-week T-bill the winner.

I’d say that deflation seems less likely in 2023, with the Cleveland Fed currently forecasting a rate of 0.79% for August. That’s probably an over-shoot, but it definitely indicates August wasn’t a deflationary month.

Conclusion

No one knows where inflation is heading, even a few months into the future. But my judgment was that CUSIP 912828B25 is likely to out-perform the 17-week T-bill, so it was worth an experimental purchase to set aside money in January 2024.

• Now is an ideal time to build a TIPS ladder

• Confused by TIPS? Read my Q&A on TIPS

• TIPS in depth: Understand the language

• TIPS on the secondary market: Things to consider

• Upcoming schedule of TIPS auctions

* * *

Posted in Inflation, Investing in TIPS, Treasury Bills | 22 Comments

Oh no, I’ve become a Treasury market day-trader!

Posted on August 27, 2023 by Tipswatch

Real yields at 2% are too good to pass up. Or am I wrong?

By David Enna, Tipswatch.com

OK, it’s a silly headline. I am not really day-trading in the Treasury and fixed-income markets; I’ve only been buying and not selling. But I have dramatically stepped up purchases in the last four months.

All this started in mid-2022, as the Federal Reserve began ratcheting up short-term interest rates. At that time, I worked out a strategy to space out purchases of 13- and 26-week Treasury bills and then roll the investments over, always allowing access to part of the money within 4 weeks. I wrote about that strategy in July 2022 and followed up in August 2022. Those investments are still rolling over, now yielding 5.5%+.

At the same time — mid 2022 — I began liquidating my holdings in Schwab’s U.S. TIPS ETF (SCHP) and very gradually began using that money to buy individual TIPS, both at auction and on the secondary market. Real yields were rising, finally, and by November 2022 had hit decade-plus highs. Unfortunately, that trend reversed in the early months of 2023, with the 10-year TIPS real yield falling from 1.73% on Nov. 3, 2022, to 1.06% on April 4, 2023. So my buying spree slowed down.

Fast forward one month to May 2023 and real yields again began rising, up more than 40 basis points by the end of that month. So I began buying TIPS again, and a bit more aggressively. As it turned out, a new strategy formed: to build a TIPS ladder through the year 2043, with near-equal amounts in each year. To do this, I began liquidating my holdings in Vanguard’s short-term TIPS ETF (VTIP).

I’ve been doing this in a traditional IRA brokerage account at Vanguard, so taxes are not an immediate concern. My strategy combined buying longer-term TIPS (5 years or longer) combined with some nominal investments in the 2- to 3-year range, to prepare for paying future RMDs from this account. Plus I will need investable money in 2024 through 2029 to buy 10-year TIPS in each of those years to complete the ladder. (There are currently no TIPS maturing from 2034 to 2039.)

Here is what I have been doing since May 2023:

My ladder-building strategy is nearly complete, but I am still looking to fill out the years 2040 to 2043 with amounts equal to the earlier years. I am hoping to nail down 2.0%+ real yields for each of those investments in coming weeks. Some days I can find them on the open market; most days I can’t.

On the nominal side, I have been looking for decent yields to provide cash in my early RMD years (2026 and 2027). This IRA account also has a 20% holding in Vanguard’s Total Bond ETF (BND) and 20% in Vanguard’s Wellington Admiral (VWENX). So I have some flexibility for those future RMD withdrawals.

And I should point out that about 10% of my TIPS are old-school purchases that remain in a taxable account at TreasuryDirect. All those, except one, will roll off by 2029.

What is the strategy?

Any investment can look dumb in a few months. I could have been in cash for the last 12 months and just bought all these TIPS at the top of the yield cycle last week! But here’s my advice: When you are putting your money into a very safe investment with a known decent return: Buy without regrets. When you see real yields that are historically attractive, go ahead and invest. As long as you have a plan.

Yields could keep going higher, obviously. Maybe even much higher if you take a dire view of the U.S. borrowing needs in the next few years. Site reader Amit this week provided a link to this video, laying out a very bearish view of the U.S. bond market:

I have no idea who these guys are or anything about their “All-In” podcast, but the speaker, David Friedberg, makes valid points about the potential extreme level of borrowing needed to fund U.S. deficits in the near future. That could lead to higher nominal yields, which would also drag real yields higher, most likely.

So there is always the possibility that buying at today’s decade-high real yields could look bad into the future. But because these investments are Treasury Inflation-Protected Securities, if held to maturity they will deliver a return of about 1.8% to 2.0% above inflation. That is a certainty.

These investments are part of an asset allocation in a single account — a traditional IRA — that just needs to keep pace or exceed U.S. inflation for the investments to provide what I need: A stable flow of cash until I reach age 90. So I am buying now to lock in that real return, an opportunity we didn’t have in the decade-long stretch of ultra-low yields.

I know a lot of readers are also adopting this “safe-withdrawal” strategy, first promoted by financial author Allan Roth in his October 2022 article, “The 4% Rule Just Became a Whole Lot Easier.” This strategy, as Roth notes, is only possible when real yields are above 1.7% — the range we are seeing today.

So with that in mind, I have been acting to build this ladder of very safe TIPS investments through 2043. It’s not day-trading, but it has become an obsession, I’ll admit.

• Now is an ideal time to build a TIPS ladder

• Confused by TIPS? Read my Q&A on TIPS

• TIPS in depth: Understand the language

• TIPS on the secondary market: Things to consider

• Upcoming schedule of TIPS auctions

* * *

Posted in Bank CDs, Inflation, Investing in TIPS, Retirement, Treasury Bills | 54 Comments

30-year TIPS reopening gets real yield of 1.970%, highest in 12 years

Posted on August 24, 2023 by Tipswatch

By David Enna, Tipswatch.com

The Treasury’s $8 billion reopening auction of CUSIP 912810TP3 — creating a 29-year, 6-month Treasury Inflation-Protected Security — went off pretty much as expected Thursday, generating a real yield to maturity of 1.970%.

While that result fell just short of a milestone 2.0% real yield, it was higher than where this TIPS was trading most of the morning, at about 1.93% to 1.94%. So investors should be pleased. The real yield of 1.970% was the highest for any 29- to 30-year TIPS auction since February 2011.

This TIPS had its originating auction on Feb. 16, 2023, which generated a real yield of 1.550% and set the coupon rate at 1.50%. So, six months later, real yields for this term have increased 42 basis points. And that means this TIPS sold today at a discount, because of the wide gap between the auctioned real yield (1.970%) and the coupon rate (1.50%).

Investors got an unadjusted price of 89.533532 for $100 of value. The inflation index on the settlement date of Aug. 31 will be 1.02632, pushing the adjusted price to 91.890055, still well below par value.

In essence, investors paid about $91.89 for about $102.63 of principal, and will now collect accruals to principal matching U.S. inflation, along with 1.5% annual interest on the inflation-adjusted principal. Here is how the pricing worked out:

Inflation breakeven rate

With a nominal 30-year Treasury bond trading today at 4.29%, this TIPS gets an inflation breakeven rate of 2.32%, which is in line with the February result but higher than other auctions of this term over the last decade. Investors in this TIPS are betting that inflation will average higher than 2.32% over the next 29 years, 6 months.

As this chart shows, 30-year inflation expectations have been rising since 2020, but have been in a range of 2.2% to 2.4% for the last several months:

Reaction to the auction

The bid-to-cover ratio was 2.42, which indicates reasonable demand. The yield came in fairly close to secondary market trading, so by all indications this auction went off without a hitch.

After strong increases in real yields over the last two weeks, the trend turned slightly downward on Wednesday, with the Treasury’s 30-year real yield estimate falling from 2.07% on Tuesday to close at 1.93% on Wednesday. So today’s end result of 1.97% looks like a positive auction for investors.

Here is the trend in recent TIPS auctions of this term: