The Fed raised rates last week. Here is what they missed. The unemployment rate looked strong last month. Look closer at how it is built and there is some alarming data that is not being reported. Here is what I found.
Mike Turner | Sep 21, 2026
If you have a home equity line, your rate went up last Wednesday.
If you have been shopping for a house, your rate went up last Thursday.
Those are two different rates. They are set in two different places by two different groups of people. And last week they both moved the same direction, which is the thing I want to walk through, because it says a lot about where we are and where this might be headed.
Let me start with what actually happened.
What the Fed did
Last Wednesday the Federal Reserve raised its short-term interest rate by a quarter of a point, to a range of 3.75 to 4 percent. Every member voted for it. It is the first increase since July of 2023.
The chairman, Kevin Warsh, gave the reasons plainly. Prices have been rising too fast for too long. The job market is strong. Business investment is booming. In his words, he would be "hard pressed to describe broad financial conditions as restrictive." So they removed, as he put it, "a dose of accommodation."
One other thing came out of that meeting that matters more than the quarter point.
The committee published its own forecast, and the middle of that forecast has the rate a quarter point higher by the end of this year and no cuts at all through the end of 2027.
What is striking is that three months ago the same people expected to be cutting by now.
So Wednesday was not a one-time move. It was a significant change of direction, in writing.
I have concerns about this decision, from the data I am seeing. I will come back to that. But first I want to show you what the Fed is focused on, and what they may be missing.
The Fed says the job market is strong. Here is what that number leaves out.
The Fed's case rests on the job market. Unemployment is 4.1 percent. The chairman called that "consistent with full employment."
If they believe we are at full employment, then their actions make sense. They believe the economy can handle a higher rate.
So I want to explain how that unemployment number is built, because most people have never had it explained, and it is not what it sounds like.
Picture a restaurant that counts how busy it is by counting the people standing in line. That is a reasonable thing to count.
But it misses everyone who drove up, saw the long line, and went home. They wanted to eat, but they are not in the count.
The unemployment rate works the same way. To be counted as unemployed, you have to have looked for work in the last four weeks.
If you looked for six months, found nothing, and stopped looking, you are not unemployed anymore. You are "not in the labor force." You disappear from the 4.1 percent.
Nobody is trying to hide that fact. That is just how the number was designed, decades ago, and it has always worked that way.
It means the headline unemployment number improves when people give up looking.
So here is what the headline unemployment number leaves out, from the same government report, released earlier this month.
There are 7 million people counted as unemployed. There are another 5.7 million people who say they want a job and are not counted, because they have stopped looking. That second group is nearly as big as the first, but they are not being counted.
Count everyone who wants work or wants more of it, including people working part time who cannot get full-time hours, and the government's own broad measure comes to about 13 million people. Nearly double the headline.
Fewer adults are working or looking for work than in January. The share is down half a percentage point, which sounds small until you do the math. It is more than a million people who stepped out of the count in eight months.
And here is the number that stopped me. Only about two thirds of adult men are working or looking for work. Outside the pandemic months, that is the lowest share in records going back to 1948. In 1949 it was 87 percent.
I went looking for why, and most of it is age. The count includes every man over 16, and a lot of them are now past 65. Economists at the St. Louis Fed found that aging alone explains more than the entire drop in participation over the past seven years.
So one man in three not working is mostly a country getting older. That is not a scandal. But it is not full employment either. It is fewer workers carrying more retirees, and one more reason the headline looks better than the ground under it.
The report that came out earlier this month showed 162,000 new jobs in August, which sounds healthy. The average over the twelve months before that was 31,000 a month, which is not strong.
One good month does not erase eleven weak ones, and the month before August was barely above zero.
Now look at the whole year instead of the one month, because the Fed says it watches trends. Over the past twelve months the economy added about 31,000 jobs a month. Health care alone added about 32,000 a month.
Take health care out and the rest of the economy added no jobs at all in a year.
Hospitals and clinics hire because the country is aging and Medicare pays the bill. They hire in recessions too. That is not a strong economy. That is one sector holding the number up.
Building jobs were up, which sounds like the part of the economy that borrows to grow is doing fine. But when I looked inside that number, the crews that build houses lost about 21,000 jobs this year. The crews building data centers, chip plants, and the power lines to feed them gained roughly 110,000.
The building boom the Fed sees is real, however it is narrow, and it runs on corporate cash, not on the rate that went up Wednesday.
The government's own spending report says the same thing: home building is down about 7 percent from a year ago, and outside data centers, commercial construction is at its lowest level in three years.
Meanwhile the white-collar side shrank all year. Software and media lost more than 100,000 jobs. Finance lost about 100,000. Government lost about 200,000. Manufacturing, for all the talk, came in about flat.
One more number, and it is the one that needs to be highlighted.
Wages rose 3.1 percent over the past year. Prices rose 3.4 percent. The average paycheck buys less than it did a year ago. And when it says "prices rose 3.4 percent," that is an average of everything a household buys, so if you drive a diesel pickup fifty miles a day, your price jump this year is way worse than 3.4 percent.
The craziest part to me is that none of those numbers are hidden. They are in the same report as the 4.1 percent, so the Fed board must see them.
Yet the Fed reads the headline and calls it full employment. Why? The report is clearly telling a different story.
Can the unemployment numbers be trusted?
There is one more thing about that 4.1 percent unemployment stat that I did not know until I dug into this, and I think most people do not.
It does not come from unemployment claims. It comes from a survey.
Every month the Census Bureau interviews about 60,000 households and asks each adult what they did last week. The sample has been the same size since 1981, while the country grew by 60 percent.
And people have stopped answering. Fifteen years ago about nine households in ten responded. Today it is about two in three.
The government's own report to Congress this spring says so, and economists at the Atlanta Fed wrote in April that if this continues, "confidence in monthly national employment statistics will diminish."
Here is the part that matters. The research they cite found that the households that stop answering are more likely to be the ones without work.
In their words, "the unemployment rate could be biased downward." The people the number is built to catch are the ones most likely to have quit picking up the phone.
You can see the wobble in the revisions. July's job number was first reported as a loss of 23,000. A month later it was revised to a gain of 21,000. Did they make more calls and get different answers? This comes from surveys, not reported financial data.
I know I have never responded to a survey like this. And if I did, I am not sure how I would answer. Do you trust the person asking? Maybe, or maybe not.
I am not saying the number is fake. I am saying it is a survey of a shrinking, self-selected group of people, and the direction of the error runs toward things looking better than they really are.
What people are actually saying about the economy.
We do not have to guess how people feel about this economy. They are asked, every month.
The University of Michigan has been asking Americans how they feel about their finances since 1952. This month's reading is the second-lowest in the history of the survey. The lowest was four months ago.
Think about what that means. People feel worse about their finances today than they did at the bottom of the 2008 housing crash. Worse than after the dot-com bust. Worse than in the spring of 2020 when the country shut down. Worse than at any other point in seventy-four years of asking, and it has happened twice in the same year.
That is not a bad month. That is a massive swing in consumer confidence in the wrong direction.
The New York Fed, part of the same Federal Reserve that raised rates on Wednesday, asks people whether they expect unemployment to be higher a year from now. This month, 44 percent said yes. In the thirteen years that survey has run, the only readings higher were the two months in the spring of 2020 when the country shut down.
And Walmart, the largest retailer in the country, reported a month ago that its sales growth was the slowest in six years. Its chief financial officer said out loud what the surveys show. When gas gets above $4, he said, "consumers are making trade-offs."
So last Wednesday we had the chairman of the Federal Reserve saying the economy is strong and the moment calls for optimism, and we had the people in that economy reporting the most concern about their finances in 70 years of asking. Both of those things are true on the same day. That gap is the whole story.
The real reason the Fed raised rates
I want to give the Fed its due here, because this is not a letter about the Fed being foolish. The people in that room likely know every number I just gave you. They raised rates anyway, and they have a reason.
The reason is history. Twice, in the 1970s and again around 1980, the Fed raised rates to fight inflation, then cut them as soon as the economy weakened. Both times inflation came roaring back, and it took rates near 20 percent and a deep recession to finally stop it. Every Fed chairman since has learned that lesson: ease too soon, and you pay twice. And with prices running above the Fed's target for five years, and people now expecting 4.6 percent next year, that lesson is front of mind.
When a reporter asked Warsh how a quarter point rate hike reopens the Strait of Hormuz, he gave the honest answer: the Fed cannot affect the price of oil or groceries, but it can keep those price rises from spreading to everything else. He is admitting the hike does nothing about the prices that are actually rising. He is saying it is aimed at everything else.
Here is the part I keep coming back to. There are two kinds of rising prices. One comes from an economy that is running hot, too much money chasing too few goods, people bidding up everything because they can.
A rate hike works on that kind. It makes borrowing more expensive, spending slows, and prices cool.
The other kind comes from a shortage. Diesel is over $6 a gallon because a pipeline is shut, a strait is contested, and refineries in Russia are dark.
That is what I wrote about last week. That is what I wrote about last week. Interest rates will not impact fuel supplies.
A rate hike works on that kind of inflation in only one way: by making people poorer until they buy less.
The Fed is treating this like an overheated economy. Most of the evidence says the economy is struggling to hold on.
The interest rate the Fed sets is the one that impacts small businesses.
So if the hike cannot touch the price of diesel, what does it touch? This is the part almost nobody explains.
There are two interest rates that matter to ordinary people. I wrote about them a couple of letters ago, and I will catch you up in two sentences.
The Fed sets the short-term rate, the one banks charge each other overnight. The long-term rate, the one on a 30-year mortgage, is set by the bond market, where the whole world shows up to lend the government money and names its price.
Those are two separate rates. The Fed can change the short-term one. It does not set the long-term rate.
When the Fed raised its rate on Wednesday, the prime rate went up with it, to 7 percent. Prime is what banks use to price a home equity line (usually prime plus a point or two), a business line of credit, and most credit cards, which run well above prime.
If you have a line of credit against your house, your rate moved up on Wednesday, and it moved because of the Fed.
Your mortgage rate did not move because of the Fed. It moved because the bond market, which is an open market, decided it wanted a higher rate to keep lending to our government.
The 30-year mortgage hit 6.95 percent on Thursday, the highest since January of last year.
So both rates went up last week, for different reasons. That is a punch to the gut for anyone hanging on with borrowed money right now. And I want to be specific about who that is.
My concern with the interest rate hike last week is not about mortgage rates. It is about the person with a house and an idea.
I have lived this. A great deal of small business in America starts with someone pulling a line of credit against their home to try something. It is how I did it, more than once. That line is priced off prime.
At 4 or 5 percent, a line of credit against your house is a great tool.
At 9 percent, which is roughly where a home equity line sits today, it is a risk most people will not take. The loan does not happen. The hire does not happen. The lease does not get signed.
There is an important nuance here that took me a while to understand, and I think it is worth two paragraphs.
When you take out a line of credit on your house or business, the bank does not hand you somebody else's savings. It creates the money on the spot.
New money enters the economy that did not exist the day before, and it goes to a contractor, a supplier, a new employee. When you pay the line off, that money disappears again.
Which means a high interest rate does not just make a loan expensive. It stops the loan from happening, and the money that loan would have created never shows up anywhere. Not in a paycheck, not in a storefront, not in a tax return.
Every business loan and home equity line that does not get created is a business that never got started, or never hired its first and second employee, or never bought the new equipment.
That is what the Fed discouraged on Wednesday. The price of starting something.
The problem with the last time the Fed lowered interest rates
Some of you might be thinking, because I thought it too: the last time the Fed cut rates hard, we got the housing frenzy of 2020 and 2021. We do not need another one of those.
I live in Boise, Idaho, and home prices here went up by 50 percent in about two years, creating a massive affordability crisis. We still have not recovered from that, and we certainly do not want to cause another spike like it.
But it is worth being precise about what actually caused the spike, because it was not the short-term rate alone.
In 2020 the Fed cut the short-term rate to zero, and it also printed money to buy mortgage bonds directly, roughly $40 billion a month, month after month, which artificially drove mortgage rates down.
That is what pushed mortgage rates under 3 percent. The Fed essentially bought down the mortgage rate, aggressively.
During Covid there were a lot more buyers than sellers, so buyers fought over what few homes were available, with access to cheap loans, and it drove up prices quickly.
Today two of those three ingredients are gone. Prices are flat. The Fed is no longer buying mortgage bonds. And the mortgage rate is set at auction, at 6.95 percent, by a bond market that is not in a generous mood to go any lower.
If the Fed cut the short-term rate today, it would not touch the mortgage rate. The people sitting on 3 percent mortgages would keep sitting, which is rational, and which is also why housing supply has been so tight for three years.
Tight supply is what keeps prices up. It is also what makes a 2008-style crash very hard to produce, though I will never tell you it is impossible. Supply is the gauge, and I watch it every month.
There is a lot of equity sitting in American homes. About $18 trillion, and nearly $12 trillion of it could be borrowed against tomorrow. That is a lot of stimulus sitting untouched, gated by the high cost of borrowing against it.
I am not a fan of over-leveraging your home. But borrowing against my home's equity has helped me start multiple businesses. That is noteworthy.
What I am saying is, the Fed can encourage job growth without touching the mortgage rate, and that is the distinction I think Wednesday's decision missed.
If it were my call
I do not have all the answers here, and the people on the Fed board likely have information that I do not. But if it were my decision, here is what I would have done, and the pros and cons of it.
I would have lowered the short-term rate, the one the Fed actually sets, meaningfully.
Not to zero. But enough to bring the equity line and the business line of credit down to where someone with a house and an idea can act on it.
And I would have left the mortgage rate exactly where the open market has it, as in, not have the Fed buy mortgage bonds like it did in 2020 and 2021.
That keeps a lid on home prices while the short-term rate reaches the people who build things and hire people, to further stimulate jobs, innovation, and growth.
Here is what I would be accepting in exchange, because there is no action that does not have negative consequences.
Savers would earn less. Anyone with cash in a money market fund or a savings account would see their rate drop while prices keep rising. I wrote a whole letter about that cost, and it is real.
However, savers have options. They could, for example, move money into longer-term bonds, which today pay well over 5 percent, and accept that it is locked up longer. So the trade-off, in my mind, is worth it.
Another potential downside is that if you lower the short-term rate below the inflation rate, you risk weakening the dollar, which makes importing oil and other goods more expensive.
But if the lower rate actually does what it is meant to do, where people are more confident to borrow to grow their businesses, which creates more jobs and boosts growth, that can strengthen the dollar over time.
Raising rates, as the Fed has chosen to do, is acting out of fear of inflation running away in an overheated economy.
I am proposing that the economy is on far shakier ground than the Fed appears to see, and therefore I believe it should lean into stimulating small business growth.
An economy where the paycheck buys less every year, where 5.7 million people have stopped looking, where the mood is the worst in seventy years of asking, and where diesel is over $6 and gasoline is well over $4, is not an economy running hot. It is an economy hanging on. Raising the price of borrowing on top of that does not cool it down. It tips it.
Two rates went up last week. Only one of them was a choice.
That is how I see it.
Why I wrote this
I am curious about people, history, and how the world works, and I write these to sort through the noise. Like a puzzle I am trying to solve. I am not a financial advisor. I am not selling anything in these letters, and I do not take on clients.
What you get here is research, insight, and my opinion. Nothing here is a recommendation.
If you like what you learn from these letters, awesome, I would love to hear it. If you think I am missing an important piece of the puzzle, please share.
That's all for today.
Stay curious, cheers,
Mike T.
###
Forwarded by a friend? Get the Letter Manage Which Emails You Get... https://preview.convertkit-mail4.com/preferences
|