Bottom line: Real interest rates don’t reset on bonds you’ve bought; inflation adjustments reset, even on already purchased bonds, every 6 months.
At pickleball a week ago, a number of us got talking about U.S. Treasury I-Bonds. My wife and I have each bought our quota of $10,000 annually for 2021 and 2022. A pickleball friend was wondering whether or how the I-Bond interest rate gets reset once you’ve bought the bonds.
Although the federal government often explains things badly, in this case the Treasury does a good job. It’s here.
But I’ll go through it anyway.
The I-Bond interest rate is made of 2 components: a real rate that stays constant once you’ve bought the bond and a rate to adjust for inflation. The bonds that my wife and I bought had a real rate of 0.00%. The latest ones, which are selling between November 1, 2022 and April 30, 2023, have a real rate of 0.40%. (Actually, there’s a third small component, as you’ll see below.)
The semi-annual inflation rate for the bonds we bought earlier this year was 4.81%. The semi-annual inflation rate for bonds bought currently is 3.24%.
So on the bonds we bought earlier this year, we earned 0.00% (real) + 2*4.81% (inflation adjustment), which gives 9.62%.
The formula is: Fixed rate + (2 x semiannual inflation rate) + (fixed rate x semiannual inflation rate).
Notice that I’ve left out the third term in the calculation just preceding this formula. It’s for the inflation adjustment on the real interest rate, as any good Fisherian economist (named after Irving Fisher) can tell you. But in the case of the bonds we bought with a 0.0% real interest rate, that term goes to 0. Zero times anything is still zero.
Notice something interesting. As far as I can tell, the inflation rate the Treasury uses is backward looking rather than forward looking. So because inflation was particularly high in the 6 months preceding May 1, 2022, my wife and I got an unusually high interest rate. What we investors care about in choosing investments is the future inflation rate. So, looking at buying an I-Bond today, if you expect the future inflation rate to be lower than the annualized rate for the 6 months preceding November 1 (3.24% * 2), you should be more likely than otherwise to purchase an I-Bond.
Incidentally, I didn’t even know about the existence of I-Bonds until sometime in 2021. But it turns out that they’ve been around since 1998, as the Treasury link shows.
The picture above is of Irving Fisher.
READER COMMENTS
vince
Nov 25 2022 at 2:22pm
When they came in the late 90s, they were paying inflation plus about 3.5%. I regret not investing in them then. The $10,000 limit should be lifted–there’s no good reason to impose it on citizens who want the inflation protection. BTW, a couple can add another $5,000 each annually from their income tax refund.
A couple important features:
1. You can’t withdraw for one year.
2. The penalty for withdrawing within 5 years is 3 months interest.
3. The interest accruals are tax deferred (by default).
David Henderson
Nov 25 2022 at 5:23pm
Thanks, Vince.
Mark Barbieri
Nov 25 2022 at 6:19pm
Unless I’m mistaken, the $5,000 from the tax return has to be claimed in the form of a physical bond…a piece of paper you have to keep track of rather than just having it as part of your account. For that reason, I have never bothered with it. The limit is too small and the gain too little for that extra hassle.
And while I’m complaining, their website has the worst password entry I’ve ever used. They don’t support two-factor authentication and their password entry is by clicking on a virtual keyboard. It’s atrocious an encourages the use of shorter, less secure passwords.
Finally, I haven’t seen a way to connect my account to online accounting services like Mint, Quicken, etc. Not sure why they can’t contract this stuff out to a competent company like Fidelity or Schwab to manage.
David Henderson
Nov 26 2022 at 11:15am
The web site you go on to buy the bonds is a disaster. I learned pretty quickly not to click on any explanations because when I did so, I had to start over.
John C Goodman
Nov 25 2022 at 8:51pm
David:
Good piece on I Bonds.
You probably know that Larry Kotlikoff think the government should do more. Let people convert their pension income and private annuities into inflation indexed annuities. Also, we should index the income tax on the inflation adjustment itself. (In fact, we should really index the entire tax code.)
Normally, I want government to do less. But since inflation is government caused and the private sector seems unable to insure against changes inflationary policies, we should have institutions that limit the ability of government to gain from the inflation it causes.
Thomas Lee Hutcheson
Nov 26 2022 at 6:28am
At least capital gains should be indexed. The preferential rate, rebasing for inheritors, and complicated rules of rollovers, etc. can be though of as ways of not taxing inflationary gains from capital sales.
Not as inflation hedges, but to better track inflation expectations, Treasury should issue more TIPS of intermediate tenor, 1, 2, 3, and 7 years to complement the 5 and 10 year securities.
A nominal GDP security would be nice for those that think THAT should be a Fed target.
vince
Nov 26 2022 at 3:12pm
Why should capital gains have any special privilege over other forms of income?
Jim Glass
Dec 2 2022 at 1:07pm
Why should capital gains have any special privilege over other forms of income?
The value of a capital asset is all the income it will produce into the future discounted to current value. Simplest example: an interest only perpetual bond like a British consol. The bond’s price today is the current value of all that future interest.
So if the interest the bond pays is taxable when paid out, and the value of the bond’s future interest payments is also currently taxable (as a capital gain), double taxation of the same income will result.
The Supreme Court repeatedly held during the early days of the income tax that capital gain is not income. Then Congress defined it to be taxable income, and as Congress can legally define a fish to be a bird (seriously, look at environmental legislation) that held.
Note that in the USA’s official national accounts prepared by the Bureau of Economic Analysis there are lines for “income”, but there is no “capital gain income”.
John C Goodman
Nov 25 2022 at 8:56pm
Here is the link to our proposals: https://www.goodmaninstitute.org/wp-content/uploads/2022/06/The-Hill-Editorial-Kotlikoff-Goodman-Final-web-copy.pdf
Vivian Darkbloom
Nov 26 2022 at 4:10am
…At a somewhat higher level of income, the Internal Revenue Service
(IRS) begins taxing 85 percent of benefits. At that point, seniors are facing a marginal tax rate that is 85 percent higher than young people with the same income.
It to me a while to understand that you meant by that. We can argue forever the concept of “income” (“income”, like “reality” should always be in quotation marks). I believe that the initial rationale for excluding (part of) social security benefits was to account for the fact that recipents have a partial cost basis in those benefits somewhat like the basis in a private annuity where contributions are ratably deducted until they are exhausted. Current employee contributions to social security are after-tax and employer contributions are not taxed. Each case will be different depending on lifetime contributions and benefits; however, at least in part, the Code sacrifices accuracy here for some simplicity (that is, relative to what it would be if actual basis were considered).
Nevertheless, if one considers “income” here to be a current accretion to wealth available for consumption, the inverse would be true up to the taxable social security thresholds. That is, for every (additional) dollar of social security my “marginal tax rate” is 100 percent lower than young people with the same “income”. Also, quite arguably (again, basis aside) my *actual effective tax rate* is always lower than that younger counterpart.
One further point: The “marginal tax rate” here (in the economist’s sense) is somewhat like the old donut hole. I suspect that, at the margin, when most readers of this blog receive their social security, 85 percent of all benefits will be taxed so that the marginal rate reverts again at some point to the same (or lower) as that “younger person”.
I often hear from economists that lawyers are responsible for our complicated tax code, but this proposal strikes me as yet a further complication. In the interest of simplification, this lawyer would like to make a modest counter-proposal: If we are going to have a social security system as we know it, let’s start taxing *all* social security benefits at normal graduated income tax rates. That would help shore up social security for the younger generation, eliminate a modest amount of complexity and avoid other inevitable increases in tax that would distort marginal incentives even more (the incentive to work and save is much less important to a senior than a younger person). It would also more closely align current “income” available for consumption and “ability to pay” to the tax rate. There is always a tension in the tax code between “complete equity” and simplicity. This is one compromise this senior would be willing to make in favor of the latter.
Thomas Lee Hutcheson
Nov 26 2022 at 6:39am
I agree. It’s better to just think of Social Security as another part of the welfare system like health insurance and unemployment insurance, just one that is, unfortunately, (under)financed by a tax on wages. Let’s move to financing them — health insurance and unemployment insurance, too — with a VAT. The VAT had not been invented in 1935, but Congress in 2021 does not have that excuse 🙂
Thomas Lee Hutcheson
Nov 26 2022 at 6:46am
On “income,” we should really be taxed on employer “contributions” to SS, Medicare and health insurance purchased for employees. The latter would lead to better decisions about whether to take the coverage offered by the employer of purchase an individual policy. It would be a way to start moving away from the employer as agent of the welfare state.
vince
Nov 26 2022 at 3:25pm
“On “income,” we should really be taxed on employer “contributions” to SS, Medicare and health insurance purchased for employees.”
Better yet, get employers out of the health insurance business.
vince
Nov 26 2022 at 4:00pm
Social Security recipients are especially shortchanged by lack of inflation indexing. A single retiree today gets the same $16,000 exemption that one did in the mid 1980s. Today, the equivalent exemption would be $44,000.
The Social Security taxable income computation is especially perverse. As nonSS goes up, say to about $20,000, an additional $1 of income creates $1.85 of taxable income. How’s that for a marginal rate?
The linked article criticizes the SS earnings test for early retirees. Right or not, SS doesn’t incentivize early retirement.
Jim Glass
Dec 2 2022 at 1:22pm
Social Security recipients are especially shortchanged by lack of inflation indexing….
People forget that Social Security went broke for the first time (it will again) in 1982. The increase in the taxable portion of benefits to 85% and lack of indexing was part of a carefully disguised “means tested” reduction of benefits to help bail out the program, one that would grow perpetually into the future.
The “income tax” you pay on your Social Security benefit isn’t included in USA general revenue like all other income tax is. It is paid back to the Social Security Administration’s account.
It is a net reduction of your benefit, determined by the amount of income you have — disguised so that nobody knows it. And this is not an accident.
Vivian Darkbloom
Dec 2 2022 at 3:24pm
A small correction: Per the 1982 Act, the taxation of SS benefits starts to kick in with an income of $25K, not 16K.
Spencer
Nov 26 2022 at 10:52am
The Fisher Effect (“tendency for nominal interest rates to change to follow the inflation rate”) was lost because economists don’t know a credit from a debit. I-bonds aren’t necessary.
I.e., the economic engine is being run in reverse. Interest is the price of credit. The price of money is the reciprocal of the price level. To increase the real rate of interest, you tighten the money stock, and drive the banks out of the savings business (which doesn’t reduce the size of the payments system), increasing the supply of loanable funds, but freezing the money stock. But the removal of Reg. Q Ceilings did the opposite.
Japan is a good example. Japan’s “lost decade” is due to the impoundment and ensconcing of monetary savings in their banks. The BOJ has unlimited transaction deposit insurance, the Japanese save more, and keep more of their savings in their banks.
“Japanese households have 52% of their money in currency & deposits, vs 35% for people in the Eurozone and 14% for the US.”
It’s stock vs. flow. Secular stagnation is directly due to the deceleration in velocity (as predicted in 1961 by Dr. Leland James Pritchard, Ph.D., Economics, Chicago 1933, M.S. Statistics, Syracuse).
Professor emeritus Pritchard never minced his words, and in May 1980 pontificated that:
“The Depository Institutions Monetary Control Act will have a pronounced effect in reducing money velocity”.
The deceleration in N-gDp was axiomatic:
Dr. Philip George corroborates this theory with his: “The Riddle of Money Finally Solved”.
Spencer
Nov 26 2022 at 11:07am
It’s truly a conspiracy (“a secret plan by a group to do something unlawful or harmful”). Banks don’t lend deposits (from a system’s perspective, not an individual bank’s experience). Deposits are the result of lending. Ergo, all bank-held savings are frozen, lost to both consumption and investment.
The ABA paid the economists that debated time deposit banking $3,000 in 1961 not to discuss the issue again. I.e., “Disintermediation is Made in Washington”.
See: “Should Commercial banks accept savings deposits?” Conference on Savings and Residential Financing 1961 Proceedings, United States Savings and loan league, Chicago, 1961, 42, 43.
An increase in time deposits adds nothing to GDP. All monetary savings, income not spent and shifted into time deposits, originate within the payment’s system, not outside it. It’s just stock vs. flow. Only savings deposits owners can spend/invest their balances outside of the banks.
Mark Brophy
Nov 27 2022 at 8:28pm
I-Bonds are a poor investment because the principal is not indexed to inflation, only the interest is indexed.
David Henderson
Nov 27 2022 at 10:34pm
You wrote:
I’m not claiming they’re a good investment although they’re better than many. As I pointed out in the piece, the interest rate is based on past inflation, not future inflation. But let’s say hypothetically that the inflation rate stays constant at 3% every 6 months, or 6% for a year, and that the real rate is 0.40%. At the end of the year, your bond is worth $106.40. So yes, the principal is indexed.
Comments are closed.