Financial Gurkha | New York City | August 12, 2026

Disclaimer: This article is for educational purposes only and is not investment advice. Trading and investing are subject to volatility and market risk, including partial or entire loss of capital. Please consult your financial advisor before making investment decisions. Financial Gurkha and its writers are not liable for your losses, nor do we take credit for your gains. CPI figures are from the U.S. Bureau of Labor Statistics Consumer Price Index Summary for July 2026 (USDL-26-1378, released August 12, 2026). Energy outlook figures are from the U.S. Energy Information Administration Short-Term Energy Outlook, August 2026.

July 2026 CPI: The Headline and the Momentum Disagree#

The 60-Second Version#↑ Contents

The seven things that matter:

1. Annualise the last three months and headline inflation is 0.8%, core 1.6%. Both below the Fed's 2% target. The 3.4% year-over-year figure is dominated by one month — March — that has not yet dropped out of the base.
2. Energy fell 1.5% in July and is still up 14.7% over the year. Those two facts are not in conflict; they are the same fact seen from different distances.
3. Gasoline is down 2.9% on the month and up 24.6% on the year. Fuel oil is up 39.1% over twelve months — the single largest annual increase in the report.
4. The shelter number is quieter than it looks. Rent and owners' equivalent rent both rose 0.3%. Headline shelter came in at 0.1% only because lodging away from home fell 2.8%. Rents are not cooling.
5. Medical care services rose 0.6% — the hottest services line in the report. Hospital services +0.5%, physicians' services +0.2%.
6. Airline fares rose 2.2% on the month and 25.5% over the year, while motor vehicle insurance fell for a second month. Transportation services is two opposite stories in one line.
7. Food is unremarkable at 3.0%, and that is the point — the component that caused the most political pain in this cycle has normalised.

The one-line read: the disinflation is real and it is broad, but it is disguised by a base effect that will not clear until the spring. Anyone reading only the 3.4% headline is looking at March, not July.

What's in this report#

• July 2026 CPI: The Headline and the Momentum Disagree
  • The 60-Second Version
    • What's in this report
  • The Number Everyone Will Quote, and the One That Matters
  • Energy: The Whole Story in One Component
  • Where the Barrel Goes From Here
  • Shelter: The Number Is Quieter Than the Reality
  • Medical Care: Services Hot, Goods Deflating
  • Transportation Services: Two Opposite Stories
  • Food: Normal, Finally
  • Record Output, Flat Emissions: Healthy or Not?
    • Why production and emissions have decoupled
    • Now the year that was actually unhealthy
    • The verdict
  • What This Means for the Fed
  • Financial Gurkha Takeaways
  • Frequently Asked Questions
  • About This Analysis

The Number Everyone Will Quote, and the One That Matters#↑ Contents

The Bureau of Labor Statistics reported that consumer prices rose 0.1% in July, after falling 0.4% in June. Over twelve months the all-items index is up 3.4%, down from 3.5% in June. Core inflation — all items less food and energy — rose 0.2% on the month and 2.5% over the year, down from 2.6%.

Here is the monthly path for 2026 as published:

Now compound the recent months and annualise them iAnnualised rateTaking a short run of monthly changes and projecting the same pace forward for a year. Three months of 0.2%, 0.0% and 0.2% compounds to 0.4% over the quarter, which is roughly 1.6% if repeated for four quarters. It is noisier than a twelve-month figure but it responds to turning points much faster — which is exactly what you want when a one-off shock is sitting in the base.:

Both current-momentum figures sit below the Federal Reserve's 2% target. The gap between the 3-month and 12-month readings is not a rounding artefact. It is one month doing enormous work.

Look again at March: +0.9% headline, with energy up 10.9%. Energy carries a weight of roughly 6–7% in the index, so that single component contributed somewhere around 0.65 to 0.76 percentage points of that 0.9% print. March was almost entirely an energy shock, and it will sit in the year-over-year comparison until the spring of 2027.

That is the central fact of this report. The disinflation is real, it is visible in the monthly data, and it is being obscured by arithmetic.

Energy: The Whole Story in One Component#↑ Contents

Energy fell 1.5% in July after falling 5.7% in June — and is still up 14.7% over twelve months. Here is the detail.

Three things worth pulling out.

Fuel oil at +39.1% is the largest annual increase anywhere in the report. It carries a small weight — it heats a minority of homes, concentrated in the Northeast — but it is a clean read on distillate iDistillateThe refined product family that includes diesel and heating oil. Distillate is the barrel's industrial cut: it moves freight, runs farm and construction equipment, and heats homes in the Northeast. Because it is the fuel of goods movement rather than commuting, distillate prices tend to lead broader goods inflation — a rise feeds into the delivered cost of almost everything before it reaches a shelf. markets, which have been tighter than crude for most of this cycle. Diesel and heating oil come off the same cut of the barrel, and a distillate squeeze passes into freight costs before it reaches any consumer good.

The split between energy commodities and energy services is the whole disinflation story. Commodities — the stuff you buy at a pump — are down 2.9% on the month. Services — electricity and piped gas, which you buy on a regulated tariff — are still rising, at 0.3% monthly and roughly 4.2–4.3% annually. Utility rates reset slowly, on regulatory schedules, and they are still catching up to fuel costs from earlier in the year. Pump prices fall in days; a utility bill falls in quarters.

Now the seasonal read, which is the part worth thinking about. July is peak driving season. Heat across Texas, Oklahoma and the mid-country has been severe, and East Coast travel volumes have been heavy. Gasoline demand should be at its annual high — and gasoline prices still fell 2.9% on a seasonally adjusted basis and 2.1% before seasonal adjustment.

Prices falling into the strongest demand month of the year is a supply signal, not a demand signal. It says the barrel is comfortable. WTI is trading around $82 as of today, well below the roughly $102 peak reached earlier this year, and the EIA's futures-implied curve slopes downward from here.

Where the Barrel Goes From Here#↑ Contents

Cross-referencing the BLS data against the EIA's August Short-Term Energy Outlook:

The EIA has Brent averaging $87 in 2026 and falling to $69 in 2027 — a 21% decline — with retail gasoline following from $3.78 to $3.29. If that path holds, the energy component flips from a 14.7% year-over-year contributor to a drag on headline CPI somewhere in the first half of 2027.

The mechanism is supply. US crude production goes to 14.2 million barrels per day by 2027, a fresh record and up 7.6% from 2024. The NYMEX futures curve in the EIA's chart and the agency's own forecast both slope down through 2027, with the STEO running slightly below the futures strip. When the world's largest producer is adding barrels into a market where the forward curve is already in mild backwardation, the price of the marginal barrel falls.

What this means for a reader watching inflation: the single largest source of upward pressure in the current twelve-month figure is forecast to reverse. Nothing about that is guaranteed — a supply disruption resets it overnight, as March demonstrated — but the base case in the official outlook is disinflationary from here.

Shelter: The Number Is Quieter Than the Reality#↑ Contents

Shelter rose 0.1% in July and accounted for roughly two-thirds of the entire monthly all-items increase. That sounds benign. Read one level down and it is not.

Rent and owners' equivalent rent iOwners' equivalent rent (OER)The BLS does not treat buying a house as consumption — a house is an asset, and its purchase price would swamp the index. Instead it estimates what homeowners would pay to rent their own home, and uses that. OER is the single largest component of CPI at roughly a quarter of the index, which is why a 0.1 percentage point move in it matters more than a large move almost anywhere else. both rose 0.3% on the month. That is roughly a 3.7% annualised pace — above target, and unchanged in character from recent months. The headline shelter figure was pulled down to 0.1% almost entirely by a 2.8% fall in lodging away from home, which is hotel room rates.

Hotel rates are volatile, discretionary, and a small weight. Rent and OER are enormous and sticky. So the correct reading is: the cost of housing Americans is still rising at about 3.7% annualised, and the cost of hotel rooms fell in July. Only one of those tells you anything about the trend.

Shelter is up 3.2% over twelve months and remains the largest single obstacle between current inflation and a 2% handle.

Medical Care: Services Hot, Goods Deflating#↑ Contents

The medical care index rose 0.4% in July after falling 0.1% in June. Underneath, two components are moving in opposite directions.

Medical care services at +0.6% was the hottest major services line in the report. Hospital services rose 0.5% and physicians' services 0.2% — this is the cost of care providers, the labour-intensive part of healthcare, and it is not cooling. Health systems are still working through wage settlements from the staffing shortages of recent years, and those flow into billed rates with a long lag.

Meanwhile medical care commodities are in outright deflation at −2.7% over twelve months, with prescription drugs down 0.8% in July alone. Drug prices are falling; the people who administer them are getting more expensive.

A clarification worth making, because it is a common conflation: the CPI measures medical care services — what consumers pay for hospitals, physicians, dental and eye care, and health insurance. It does not measure Medicare, which is a federal insurance programme. Medicare reimbursement rates influence what providers charge everyone else, so the two are related, but a CPI print tells you about consumer prices for care, not about programme costs. If you want the Medicare picture you need CMS data, not BLS.

For anyone budgeting around care costs, the practical read is that the trend for services remains upward at roughly 2.7% a year and accelerating month to month, while the goods side is providing genuine relief.

Transportation Services: Two Opposite Stories#↑ Contents

Transportation services rose 0.3% on the month and 2.9% over twelve months. That composite hides an unusually wide split.

Airline fares up 25.5% over twelve months is the second-largest annual increase in the entire report, behind only fuel oil. Jet fuel comes off the same distillate cut that pushed fuel oil to +39.1%, and carriers have been disciplined about capacity. Add a heavy summer travel season and fares have very little reason to fall. July alone added 2.2%.

Motor vehicle insurance fell 0.3%, after falling 2.0% in June. This is genuinely good news and easy to miss. Auto insurance was one of the most punishing line items of the entire post-pandemic period, running at double-digit annual rates for a stretch, as insurers repriced for higher repair costs and higher used vehicle values. Two consecutive monthly declines suggest that repricing cycle has completed and is now unwinding.

For households, these two roughly offset in the index but not in the wallet. Insurance is a fixed monthly cost every driver pays. Airfare is discretionary and concentrated among people who fly. The composite of +2.9% understates relief for non-travellers and understates pain for frequent flyers.

Food: Normal, Finally#↑ Contents

Food rose 0.1% in July and 3.0% over twelve months. Food at home actually fell 0.1% on the month and is up 2.7% annually; food away from home rose 0.3% and is up 3.4%.

Three of the six major grocery groups declined in July. Meats, poultry, fish and eggs fell 0.7% as pork dropped 1.5%. Fruits and vegetables fell 0.1% — lettuce alone was down 16.4%, a weather-driven swing rather than a signal. Dairy is down 0.5% over the year.

This is the least interesting section of the report, and that is precisely why it matters. Grocery inflation was the most politically salient economic number of the last several years. At 2.7% for food at home, it has returned to something close to its long-run average. The remaining pressure is in restaurants at 3.4%, which is a wage story rather than a commodity story.

Record Output, Flat Emissions: Healthy or Not?#↑ Contents

The EIA outlook lets us ask a question the CPI cannot: the US is producing record volumes of oil and gas — what is that doing to emissions?

Between 2024 and 2027, crude production rises 7.6% and LNG exports rise 56.3%, while CO₂ emissions rise 1.1%. Output is climbing far faster than emissions.

Why production and emissions have decoupled#

The main reason is definitional and it is important: US CO₂ emissions measure fuel burned inside the United States, not fuel produced there. When American gas is liquefied and shipped to Europe or Asia, the emissions from burning it are counted by the importing country. A 56% increase in LNG exports therefore adds essentially nothing to the domestic total — it moves the combustion offshore.

That is not the same as reducing global emissions. If exported US gas displaces coal in the importing country, global emissions fall. If it displaces renewables or nuclear, they rise. The honest position is that the US figure looks better than the global reality warrants, and the direction of the global effect depends on what the gas replaces.

Now the year that was actually unhealthy#

Set the export effect aside and look at emissions against economic output:

2025 is the unhealthy year in this dataset. Emissions grew 2.4% while the economy grew 2.1% — the only year where the US produced more carbon per unit of output than the year before.

The cause is visible in the same tables. Natural gas spot prices jumped from $2.19 to $3.53 between 2024 and 2025, a 61% increase. When gas gets expensive, power generators switch to coal — and coal's share of US electricity generation rose from 16% to 17% in exactly that year. Coal emits roughly twice the CO₂ per unit of electricity that gas does, so even a one-point share shift moves the national total.

2026 and 2027 reverse it. Gas prices ease to $3.44 and $3.31, coal's share falls back to 16% and then 15%, and solar climbs from 7% to 8% to 9% of generation. Emissions fall 1.8% in 2026 against 2.0% GDP growth, then rise only 0.5% against 2.4% growth in 2027. Both years show genuine decoupling.

The verdict#

Healthy, with two honest caveats.

It is healthy in the sense that matters most: from 2026 the US is forecast to grow its economy while holding emissions roughly flat, and to do it while producing record volumes of hydrocarbons. That combination — more output, more exports, flat domestic emissions, falling carbon intensity — is what a functioning energy transition looks like in a country that has not stopped drilling.

The first caveat is the accounting one above. Exported molecules are somebody else's emissions on paper, and the global ledger depends on what they displace.

The second is that the improvement is fuel-price-driven, not structurally locked in. The 2025 deterioration happened because gas got expensive and coal came back. Nothing in the 2026–27 forecast prevents that from recurring; it simply assumes gas stays cheap. The durable part of the improvement is the solar build — 5% of generation in 2024 to 9% in 2027, close to a doubling — because that capacity does not switch back when fuel prices move. The rest is a commodity price bet dressed as a trend.

What This Means for the Fed#↑ Contents

We covered the July FOMC decision two weeks ago, where three members dissented in favour of a hike and there was no dove in the room. This report complicates that stance.

The case for patience: headline is 3.4%, shelter is 3.2% and still running at 0.3% monthly in its rent components, and medical care services accelerated to 0.6%. Services inflation has not been solved.

The case for cuts: core at 1.6% annualised over three months and headline at 0.8% are both below target. The energy base effect that produces the 3.4% headline is arithmetic, not current pressure, and it mechanically unwinds by spring. The EIA has Brent falling 21% into 2027.

A committee that hiked into this data would be tightening against a shock that has already passed. A committee that reads the momentum sees an economy at or below target. Our read is that the 3-month figures are the more honest description of current conditions — but the shelter and medical services lines are the reason reasonable people on that committee still disagree.

Financial Gurkha Takeaways#↑ Contents

1. The 3.4% headline is March, not July. Three-month annualised headline is 0.8% and core is 1.6%. The year-over-year figure will not reflect current conditions until the spring energy shock rolls out of the base.

2. Gasoline fell during peak driving season. Prices declining into the strongest demand month of the year is a supply signal. WTI near $82 against a spring peak around $102 says the barrel is comfortable.

3. Shelter's 0.1% is a hotel story. Rent and OER both rose 0.3% — roughly 3.7% annualised. Housing costs are not cooling; hotel rooms got cheaper for a month.

4. Care providers are the stubborn line. Medical care services +0.6% monthly, +2.7% annually, with hospitals at +0.5%. Meanwhile prescription drugs are deflating. Labour is the cost that will not come down.

5. Insurance relief is real and underreported. Two straight monthly declines in motor vehicle insurance after years of double-digit increases. That is a fixed cost falling for every driver.

6. Record output, flat emissions — mostly healthy, partly accounting. Production up 7.6% and LNG exports up 56% against 1.1% emissions growth. But exported gas is counted abroad, and 2025's deterioration shows how fast coal returns when gas gets expensive.

Frequently Asked Questions#↑ Contents

What was the July 2026 CPI inflation rate? The Consumer Price Index rose 0.1% in July 2026 on a seasonally adjusted basis and 3.4% over the previous twelve months, down from 3.5% in June. Core CPI, which excludes food and energy, rose 0.2% on the month and 2.5% over the year.

Why is annual inflation 3.4% if monthly inflation is only 0.1%? Because of a base effect from March 2026, when the all-items index rose 0.9% on the back of a 10.9% jump in energy. That single month contributed an estimated 0.65 to 0.76 percentage points and remains in the twelve-month comparison until spring 2027. Annualising the most recent three months gives headline inflation of roughly 0.8% and core of 1.6%.

How much did energy prices change in July 2026? The energy index fell 1.5% in July after falling 5.7% in June, but is still up 14.7% over twelve months. Gasoline fell 2.9% on the month and is up 24.6% annually. Fuel oil fell 1.7% on the month and is up 39.1% over the year — the largest annual increase in the report. Electricity rose 0.1% and utility gas rose 0.7%.

Why did gasoline prices fall during peak summer driving season? Falling prices during the strongest demand month of the year indicate ample supply rather than weak demand. US crude production is at record levels and forecast to reach 14.2 million barrels per day by 2027, and WTI is trading near $82 against a peak around $102 earlier in 2026.

Is rent inflation actually slowing? Not materially. Headline shelter rose only 0.1% in July, but rent of primary residence and owners' equivalent rent each rose 0.3% — roughly 3.7% annualised. The low headline figure came from a 2.8% fall in lodging away from home, which is hotel rates. Shelter is up 3.2% over twelve months.

What is happening with healthcare costs? Medical care services rose 0.6% in July, the fastest of any major services category, with hospital services up 0.5% and physicians' services up 0.2%. Over twelve months medical care services are up 2.7%. Medical care commodities moved the other way, falling 0.6% on the month and 2.7% over the year, with prescription drugs down 0.8% in July.

Does the CPI measure Medicare costs? No. The CPI measures medical care services — what consumers pay for hospitals, physicians, dental and eye care and health insurance. Medicare is a federal insurance programme, and its costs are reported by CMS rather than BLS. The two are related, because Medicare reimbursement rates influence what providers charge other payers, but a CPI print does not measure Medicare spending.

Why are airline fares up 25.5%? Jet fuel is refined from the same distillate cut as heating oil, which is up 39.1% over twelve months, and carriers have kept capacity growth restrained while summer travel demand has been strong. Airline fares rose 2.2% in July alone.

Is US car insurance still rising? No. Motor vehicle insurance fell 0.3% in July after falling 2.0% in June — two consecutive monthly declines following a period of double-digit annual increases driven by higher repair costs and vehicle values.

How fast is food inflation in 2026? Food rose 0.1% in July and 3.0% over twelve months. Food at home fell 0.1% on the month and is up 2.7% annually; food away from home rose 0.3% and is up 3.4%. Grocery inflation has largely normalised, with the remaining pressure concentrated in restaurants.

Are US CO₂ emissions rising with record oil and gas production? Only marginally. Between 2024 and 2027 the EIA forecasts crude production rising 7.6% and LNG exports rising 56.3%, while CO₂ emissions rise 1.1%. The main reason is that US emissions count fuel burned domestically, not fuel produced or exported — combustion of exported LNG is attributed to the importing country.

Was the 2025 increase in US emissions healthy or unhealthy? Unhealthy. Emissions rose 2.4% in 2025 while GDP grew 2.1%, meaning carbon intensity worsened — the only year in the forecast period where that happened. The cause was a 61% rise in natural gas spot prices, which pushed power generators back toward coal, lifting coal's share of electricity generation from 16% to 17%. In 2026 and 2027 gas prices ease, coal's share falls to 16% then 15%, solar rises from 7% to 9%, and emissions intensity improves in both years.

What does this CPI report mean for Federal Reserve policy? It cuts both ways. Three-month annualised headline (0.8%) and core (1.6%) are below the 2% target, and the energy base effect driving the 3.4% headline unwinds mechanically by spring. But shelter is still rising 0.3% monthly in its rent components and medical care services accelerated to 0.6%, so services inflation remains unresolved.

About This Analysis#↑ Contents

Financial Gurkha is an independent markets research desk run by Kanchan Sharma from New York City. Every figure above is sourced to the BLS Consumer Price Index Summary for July 2026 or the EIA Short-Term Energy Outlook for August 2026, and where we have annualised, derived or estimated a figure we have said so and shown the arithmetic.

Need a full intrinsic valuation, or macro research built to order? Schedule a consultation with the Financial Gurkha desk.

Reported from New York City. Sources: BLS Consumer Price Index Summary, July 2026 (USDL-26-1378) and EIA Short-Term Energy Outlook, August 2026.