Inflation Is 3.7%. Bitcoin Is Down 38%. So Much for the Hedge?

On October 14, the government said consumer prices rose 3.7% in a year, with gas up 34%. Bitcoin, the asset that was supposed to love this, is down 38% from its high. So where’s the hedge? The 3.7% undercounts what your dollars are losing, and bitcoin’s boring stretches have a habit of ending in the opposite.

The backdrop is the ugliest mix the Fed has faced in decades. Oil has traded above $100 a barrel since October 7, with the war with Iran choking tanker traffic through the Strait of Hormuz. September's CPI jumped from 3.4% to 3.7%. The economy added just 29,000 jobs in September, and unemployment rose to 4.2%.

The Fed raised rates a quarter point in September, its first hike since 2023. On October 28 it held at 3.75% to 4%, but Governor Lisa Cook and Cleveland Fed President Beth Hammack dissented in favor of another hike. Chair Kevin Warsh summed up the oil problem this way: "I am not going to be the guy who says that oil prices haven't gone up."

Prices are rising, hiring has stalled, and rates are going up, not down. Meanwhile bitcoin trades near $77,800, down from about $126,200 last October.

It's a fair question. It deserves a better answer than "zoom out," so we'll take it in two parts: how much your money is actually losing, and why bitcoin hasn't moved the way people expected.

Is 3.7% even the real number?

The Consumer Price Index is a model, built from a basket of goods, survey data, and a long list of judgment calls about what to count and how. Those calls have changed over the years, and the changes have been designed to push the number down.

The big one came in 1983. Before then, the CPI tracked what it actually cost to buy a home, including the mortgage interest. After 1983, the Bureau of Labor Statistics swapped that for "owners' equivalent rent," an estimate of what homeowners would pay to rent their own house. Borrowing costs dropped out of the index entirely. Then in 1996 the Boskin Commission told Congress the CPI was overstating inflation, which led to more adjustments: substitution (if beef gets expensive, the index assumes you switch to chicken) and quality adjustments (if your new laptop is faster, part of the price increase gets counted as "you're getting more," not as inflation).

Each change has a defensible rationale on paper. Taken together, they turned the CPI from a cost-of-living number into something closer to a cost-of-staying-roughly-as-comfortable number, with the government deciding what "comfortable" means.

It's also worth asking who benefits from a lower number, because the government told us. In March 1995, Fed Chair Alan Greenspan testified to the Senate Finance Committee that about 30% of federal spending, including Social Security, was tied to the CPI, along with the income tax brackets. If those adjustments ran one percentage point lower, he estimated, "the annual level of the deficit would be lower by about $55 billion after five years." The Boskin Commission's final report made the same case in bigger numbers: an overstated CPI would add about $1.07 trillion to the national debt by 2008, making the bias alone, by the commission's reckoning, the fourth-largest federal program after Social Security, health care and defense.

Put plainly, the biggest payer of inflation-linked benefits and the biggest debtor in the country is also the one that measures inflation. A smaller number shrinks retirees' raises (the 3.6% Social Security increase for 2027 was announced the same morning as the September CPI), lets bracket creep raise taxes without a vote, and makes the debt look more manageable. When the GAO asked in 2000 how much of the bias had been fixed, the Bureau of Labor Statistics replied that it "does not believe it is currently possible to produce reliable estimates of bias in the CPI."

There's an irony in Greenspan leading that push. Almost three decades earlier, in a 1966 essay called "Gold and Economic Freedom," he wrote that "in the absence of the gold standard, there is no way to protect savings from confiscation through inflation."

So what happens if you measure it differently? A few people have tried.

Larry Summers and three co-authors put borrowing costs back in. In a 2024 NBER working paper, Marijn Bolhuis, Judd Cramer, Karl Oskar Schulz and Summers built inflation measures that include the cost of money, the way the CPI did before 1983. By their method, the inflation that started in 2021 would have peaked around 18%, roughly double the official 9.1%, and was still running about 8% in early 2024 while the official number was near 3%.

ShadowStats, run by economist John Williams, publishes an "alternate CPI" that estimates inflation as if it were still calculated the way it was in 1980. Its most recent widely cited reading put inflation around 7.6% when the official rate was 3.0% this past spring.

The Chapwood Index skips the model and tracks the prices of 150 things people actually buy (no substitution, no quality adjustments) in the 50 largest U.S. metro areas. Its 2025 report puts the national average around 11% to 12% a year, with the Bay Area above 14%. Chapwood doesn't publish all its underlying data, so it's hard to audit, but it lines up with what a lot of families report at the checkout.

These three measures use different methods and land on different numbers. Every one of them comes in well above 3.7%.

Skip the basket and count the money

There's a simpler way to check, and it doesn't depend on anyone's basket.

Henry Hazlitt opened his 1960 book What You Should Know About Inflation with a definition most economists have since forgotten: "Inflation, always and everywhere, is primarily caused by an increase in the supply of money and credit. In fact, inflation is the increase in the supply of money and credit." Rising prices are the symptom. The cause is more dollars. And creating dollars is easy: in a 2002 speech, future Fed Chair Ben Bernanke reminded his audience that the government has "a technology, called a printing press (or, today, its electronic equivalent)" that lets it make as many dollars as it likes "at essentially no cost."

So count the dollars. In February 2020, the U.S. M2 money supply was about $15.5 trillion. In September 2026 it was $23.4 trillion. That's about 51% more dollars in six and a half years. Over the same stretch, the official CPI rose about 30%.

The economy didn't grow anywhere near fast enough to soak up a 51% increase in dollars. The rest showed up in prices the CPI doesn't count, like stocks, land and the purchase price of a house, or in things it counts lightly. The extra money didn't vanish. A lot of it went into assets.

On top of that, the money printing has started again. M2 grew 5.71% over the past year, the fastest pace since May 2022, even though the Fed was hiking during part of that time.

Wartime makes this worse, and it always has. Writing about how governments paid for the First World War, John Maynard Keynes (no hard-money crank) warned in 1919 that "by a continuing process of inflation, governments can confiscate, secretly and unobserved, an important part of the wealth of their citizens." Oil above $100 because of a war gets reported as an energy story. Keynes would have recognized it as a money story too.

You can also check it against your own life. Gasoline is up 34% from a year ago. Airline fares are up 22%. Anyone who renewed home or car insurance in the last few years knows that line item didn't go up 3%. If your personal inflation rate is closer to 7% than 3.7%, a savings account paying 4% is losing you about 3% a year, and you're still paying tax on the interest.

We made the same argument from the paycheck side in Why $100K Feels Broke in 2026. The losses add up a few percent at a time, mostly where the official number doesn't look.

So why is bitcoin down?

If dollars keep losing value, why did bitcoin fall from $126,000 to under $78,000?

Bitcoin doesn't work like an inflation thermometer that ticks up 0.3% whenever a CPI print comes in hot. It's a new monetary asset still in the middle of being adopted, and that process is anything but smooth.

Nick Szabo explained why in his 2002 essay Shelling Out: The Origins of Money. Long before coins, people stored value in collectibles like shells, beads and gold. Szabo argued that "the primary and ultimate evolutionary function of collectibles was as a medium for storing and transferring wealth," and that what made a collectible trustworthy was "unforgeable costliness," a cost the recipient could verify and nobody could fake. Shells became money because they were hard to forge and could be checked, not because a government said so.

Bitcoin is the first collectible built for the internet with exactly those two properties. But no money starts out as the thing you price groceries in. Vijay Boyapati built on Szabo in The Bullish Case for Bitcoin (2018) and described money developing in stages: first a collectible, then a store of value, then a medium of exchange, and finally the unit everything is priced in. While an asset is moving between those stages, its price moves in big waves. Each wave brings in a new group of holders, overshoots, and then falls back to a higher base than the wave before.

So in the short run, bitcoin trades like what it is right now: a scarce asset people are still learning to value, sensitive to interest rates, leverage and headlines. When the Fed hikes and the cost of borrowing goes up, leveraged buyers get squeezed, and bitcoin drops. When an exchange gets hacked for $6.3 billion, as Binance was on October 29, it drops some more.

Over the longer run, bitcoin has protected savers better than anything else. Bitcoin closed February 2020 at about $8,600. It's around $77,800 now, roughly nine times higher, while the money supply grew 51% and gold gained about 146%. Over 13 months, bitcoin hasn't been a hedge. Over six and a half years, nothing has done the job better. Which of those timeframes matters depends on how long you plan to hold.

Boring usually comes before exciting

Right now bitcoin is boring, and we mean that literally.

On October 21, K33 Research reported that bitcoin's 28-day realized volatility had fallen to about 1.2%, near its lowest level of 2026, with the price stuck in a narrow range since September 21. Over the past year, bitcoin's realized volatility has run about 43%, compared with a five-year average of 81%. Even the Binance hack only knocked it a few thousand dollars lower, into a $76,000 to $78,000 drift. Trading volume is thin, and people have mostly lost interest.

K33's head of research, Vetle Lunde, put it plainly: "Expansions typically follow such compressions, and we expect BTC volatility to pick up shortly."

He has history on his side. Bitcoin has gone through several of these quiet stretches:

• Summer 2020. Bitcoin spent weeks stuck around $9,000 to $9,500, with volatility at the bottom of its range. It closed July at about $11,300, December at about $29,000, and passed $60,000 by the following spring.
• Summer 2023. Bitcoin sat near $26,000 to $30,000 for months. In August, its 90-day volatility hit the lowest level since 2016, and Glassnode described the mood as "apathy and exhaustion." By March 2024 bitcoin was above $73,000.
• 2016. A long, flat grind between halvings, followed by the 2017 run to nearly $20,000.

There's another pattern that has held in every cycle so far. Each bear market has bottomed higher than the last: about $152 in 2015, about $3,200 in 2018, about $15,500 in 2022, and roughly $58,000 to $60,000 in June 2026. The lows keep stepping up because, cycle after cycle, more people decide this collectible is worth holding.

Where we land

The hedge isn't broken. It's working on a longer clock than the news cycle, and the dollar is losing value faster than the official number admits. Somewhere between the CPI's 3.7% and Chapwood's 11% is the rate your savings are actually losing, and the money supply is growing again.

We'll also put our own opinion on the record, as an opinion and not a promise: we think this stretch, with bitcoin under $100,000, fear in the air and nobody paying attention, is going to look like a gift in hindsight. Once bitcoin gets back above six figures, we don't expect to see it below $100,000 again. History has rewarded people who bought during the quiet stretches, and today's setup of stalled hiring, $100 oil, a Fed that can't print barrels, and a money supply that's started growing again isn't a world where hard money gets less useful.

We could be wrong about the timing. Nobody can call a bottom, and this isn't financial advice. So if you act on any of this, do it in a way that doesn't require you to be right about next month:

• Buy on a schedule, not on a feeling. Small recurring purchases take the guesswork out of quiet markets.
• Skip the leverage. Every one of those big waves shook out the people who borrowed to ride it.
• Take custody. Whatever you buy during the boring part only helps you in the exciting part if it's still yours when you get there. Keep it in a wallet where you hold the keys.

Bitcoin Well sends every purchase straight to a wallet you control, so you can buy on a schedule while things are quiet.

Originally published at bitcoinwell.com/blog