Tariffs, Energy, and a $10,000 Decision: Where 2026 Inflation Actually Landed, Costed Over 5 and 10 Years
Updated September 28, 2026. Every rate and index change below was read on that day from the primary sources listed at the end — the Bureau of Labor Statistics, Freddie Mac, and the U.S. Treasury. No asset price forecast is made here, and no retailer price or product rating is quoted.

The question that gets asked about tariffs and crypto is usually "what will this do to the price of X." That question has no answerable form. A different question does: which line items in a household budget actually moved this year, and what is the cost of leaving money idle while deciding? Both have published numbers behind them, and the second one is a straight comparison between options that most households never write down.
What the August 2026 CPI release shows, by category
The Bureau of Labor Statistics released August 2026 CPI on September 11, 2026. Reading the categories rather than the headline changes the story:
| Measure | Change | Period |
|---|---|---|
| All items, seasonally adjusted | +0.4% | month over month, August |
| All items, not seasonally adjusted | +3.4% | 12 months ending August |
| All items less food and energy (core) | +2.4% | 12 months |
| Energy | +16.3% | 12 months |
| Food at home | +2.2% | 12 months |
| Apparel | +3.6% | 12 months |
| New vehicles | +0.6% | 12 months |
If import duties were the dominant force in a household's cost of living right now, the categories most exposed to imported goods should be leading. They are not. Apparel is up 3.6% and new vehicles are up 0.6%, both at or below the all-items figure, while energy is up 16.3% — more than four times the headline. That does not mean the tariff channel is imaginary; it means that for the average basket in the year to August 2026, it was not the biggest thing happening. Anyone budgeting around trade headlines while ignoring the utility bill has the weights backwards.
The same percentages, on one household's monthly bill
Percentages are hard to act on. Applying those 12-month rates to a stated set of monthly spending assumptions — groceries $900, household energy and motor fuel $400, apparel $150, and an amortized new-vehicle cost of $500 — gives this:
| Category | Assumed monthly spend | 12-month CPI change | Added cost per year | Added cost per month |
|---|---|---|---|---|
| Groceries | $900 | +2.2% | $238 | $20 |
| Household energy + fuel | $400 | +16.3% | $782 | $65 |
| Apparel | $150 | +3.6% | $65 | $5 |
| New vehicle (amortized) | $500 | +0.6% | $36 | $3 |
| Total | $1,950 | blended 4.8% | $1,121 | $93 |
Two readings fall out. The blended increase across this basket is 4.8%, well above the 3.4% all-items figure, purely because the basket is weighted toward energy. And 70% of the added $1,121 comes from one line — energy — while apparel, the most tariff-exposed category in the list, contributes $65 a year, about $5 a month. These are assumptions applied to published rates, not measured household data; change the spending mix and the total changes with it.
The $10,000 that hasn't been committed yet
Here is the comparison the trade headlines never make. Suppose a household has $10,000 that is not needed for expenses. Three places it can go, priced with rates published in the last week of September 2026:
| Option | Published rate | Value after 5 years | Gain, 5 years | Value after 10 years | Gain, 10 years |
|---|---|---|---|---|---|
| A. Pay down 30-year mortgage principal | 7.03% (Freddie Mac PMMS, week ending 9/24/2026) | $14,045 | $4,045 | $19,727 | $9,727 |
| A2. Pay down a 15-year mortgage | 6.42% | $13,649 | $3,649 | $18,631 | $8,631 |
| B. 5-year Treasury, held to maturity | 4.98% (Treasury par yield, 9/25/2026) | $12,751 | $2,751 | $16,258 | $6,258 |
| B2. 10-year Treasury | 5.17% | $12,866 | $2,866 | $16,555 | $6,555 |
| B3. 1-year Treasury, rolled | 4.50% | $12,462 | $2,462 | $15,530 | $5,530 |
| C. A volatile asset such as bitcoin | no published forward rate exists | Cannot be computed. See below for the break-even instead. | |||
Option A wins against every Treasury option in this rate environment, by $1,295 over five years and $3,469 over ten against the 5-year note. The reason is mechanical: mortgage interest avoided at 7.03% is a certain, untaxed return, while Treasury interest at 4.98% is certain and taxable. At a 24% marginal federal rate the Treasury option's after-tax equivalent falls to roughly 3.78%, which widens A's lead to about $2,000 over five years.
Option C is where the original framing of this article lived, and it is the one that cannot be tabled honestly. There is no published forward rate for bitcoin, so no five- or ten-year figure can be computed. What can be stated is the break-even: to beat option A, a volatile holding has to clear 7.03% a year, every year, after tax and after any fees — and it has to do so on a schedule the household cannot control. The U.S. Securities and Exchange Commission's investor alert on bitcoin describes the loss and fraud risks that sit alongside that requirement. A 7% hurdle is not exotic; the problem is that A delivers it with certainty and C does not deliver it at all in any given year.
Where the two sides of the comparison do not line up
Three differences make A and B less interchangeable than the table implies, and they cut in opposite directions.
- Liquidity. A dollar of mortgage principal is gone until the house is sold or refinanced. A Treasury can be sold, at whatever the market pays that day. If the $10,000 is also the emergency fund, option A is the wrong answer regardless of the arithmetic.
- Tax treatment. Mortgage interest avoided is not taxable income; Treasury interest is federally taxable. That gap is what pushes the effective comparison from 2.05 percentage points to roughly 3.25.
- Rate direction. The 30-year average moved from 6.95% to 7.03% in a single week in September 2026. If mortgage rates fall materially, prepaying at today's rate still locks in avoided interest at today's rate — which is the point — but a future refinance changes the arithmetic for everything not yet prepaid.
Which way does the cheaper error run? Prepaying money that turns out to be needed costs a forced refinance or an expensive short-term loan. Leaving money in a 1-year Treasury while deciding costs the difference between 4.50% and 7.03% — about $253 a year on $10,000, or roughly $21 a month. That is a small enough number that taking a month to decide is cheap, and a large enough number that taking three years to decide is not.
If the money is earmarked for spending inside 12 months
The table above assumes the $10,000 is not needed. If it is — a roof, a car, tuition — then the comparison changes shape entirely and the relevant number is the one in the CPI table. A household facing the $93-a-month increase computed above needs about $1,121 a year more than it needed last year just to hold the same basket. Against that, a 4.50% one-year Treasury on $10,000 yields about $450 before tax. In other words, for a household whose basket is energy-heavy, parking the money does not keep up with the household's own inflation rate, let alone the 3.4% headline. Reducing the energy line is the higher-return move in that case, and it is the only one of these options that does not depend on any rate staying where it is.
Tip: Since energy was the single largest contributor in the table above, the cheapest diagnostic is usually measuring rather than guessing which circuits are driving the bill. A plug-in home energy monitor reports actual draw per appliance, which is the input the utility bill never gives you. (These are Amazon Associate links — we may earn a small commission on qualifying purchases.)
Six figures worth checking against your own numbers
- +16.3% — energy, 12 months to August 2026. Compare it to your own utility and fuel spend.
- +3.4% vs +2.4% — all items versus core. The gap is food and energy.
- 7.03% — the rate your mortgage prepayment earns, risk-free and untaxed, if that is your rate.
- 4.98% — the 5-year Treasury par yield on 9/25/2026, before tax.
- About 3.78% — that Treasury yield after a 24% marginal federal rate.
- $253 a year — the cost on $10,000 of choosing the 1-year roll instead of prepaying, at these rates.
What this article could not confirm
Customs duty receipts for fiscal 2026 were not retrieved from the Monthly Treasury Statement in time for this update, so no figure for tariff revenue appears above. The share of any CPI category increase attributable to import duties is not published as such and is not estimated here. The household spending assumptions are stated inputs, not survey data, and the CPI category weights that produce the official all-items figure were not applied. State and local tax treatment of Treasury interest is not covered. No forecast of bitcoin, energy or mortgage rates is offered.
Related reading on this blog
Sources
- U.S. Bureau of Labor Statistics, Consumer Price Index — August 2026 (released September 11, 2026) — +0.4% month over month seasonally adjusted, +3.4% over 12 months unadjusted, core +2.4%, energy +16.3%, food at home +2.2%, apparel +3.6%, new vehicles +0.6%.
- Freddie Mac, Primary Mortgage Market Survey (week ending September 24, 2026) — 30-year fixed 7.03%, up from 6.95% the prior week; 15-year fixed 6.42%.
- U.S. Department of the Treasury, Daily Treasury Par Yield Curve Rates (September 25, 2026) — 1-year 4.50%, 2-year 4.81%, 5-year 4.98%, 10-year 5.17%.
- U.S. Securities and Exchange Commission, Investor Alert on bitcoin — risk disclosures referenced in the option C discussion.
This article summarizes published government and agency data and performs arithmetic on stated assumptions. It is not personalized financial, tax or investment advice, and no security or asset is recommended here. Jurisdiction is the United States; rates are as of the dates given and change weekly or daily. Compounding calculations assume the stated rate is earned or avoided annually with no fees, no rate changes and no state or local tax, which no real account matches exactly.
Start Here: every guide by topic · About · Privacy Policy
Comments
Post a Comment