The Short Debt Cycle: Five to Eight Years, Every Time

You've lived through it and you probably didn't name it. A stretch of years where everyone's hiring, the apartment gets more expensive every lease, your customers pay on time and ask for more. Then, without a war or a crash you can point to, it turns. Budgets freeze. A client "pauses." A friend gets let go. Two years later it's fine again, and nobody can quite say why.

That's not weather. It has a shape, a length, and a throttle, and the same hand has been on the throttle for seventy years.

Lesson 2.1 showed you the chain: spending is income, and most spending is credit. This lesson shows you what the chain does over time. The question: why does the economy speed up, overheat, stall, and restart on a schedule nobody votes on?

Wednesday evening, October 19, 1955. The Waldorf Astoria in New York. The room is full of investment bankers, and the man at the podium is the one person in America they'd rather not hear from. William McChesney Martin Jr. runs the Federal Reserve. He's forty eight, a former stock exchange president, and he has just raised the discount rate again into an economy that's doing very well.

He knows what they think of him. He says so. "The Federal Reserve, as one writer put it, after the recent increase in the discount rate, is in the position of the chaperone who has ordered the punch bowl removed just when the party was really warming up."

Read the sentence he put right before it, because that's the one with the mechanism in it. Precautionary action to prevent inflationary excesses, he says, "is bound to have some onerous effects, if it did not it would be ineffective and futile."

That's the whole job in one line. The point of raising rates is to hurt. Not to punish, to cool. The central bank's tool is the price of credit, and the only way to slow a credit-driven economy is to make credit cost more until people stop borrowing.

Martin ran the Fed for nineteen years, through four recessions. He didn't cause the cycle. He inherited it, and he's the first chairman who said out loud that his job was to lean into it. Ask the tycoon's question. Who made the rule? A committee that sets a number. Who gains when the bowl comes out? Everyone who borrowed before it did. Who pays when it's removed? Whoever needed to borrow next.

Ray Dalio calls it the short-term debt cycle, and his description is the cleanest one in print. Here it is, gear by gear, built on what you learned in 2.1.

Gear one is cheap credit. When the central bank's rate is low, borrowing is easy. Banks make loans, loans create deposits, deposits get spent, and every dollar spent is a dollar of someone's income. Incomes rise. Because the extra money came from credit, not from producing more, spending grows faster than the stuff there's to buy.

Gear two is the heat. When spending outruns output, sellers raise prices. That's inflation, which means the same goods costing more dollars because more dollars are chasing them. Workers ask for raises to keep up. Asset prices run ahead of both, because cheap money looks for a home and houses and stocks are where it lands.

Gear three is the punch bowl. The central bank sees prices rising and raises its policy rate, the overnight rate banks pay each other, which sets the floor under every other interest rate in the country. Mortgages, car loans, business lines, credit cards: all of them reprice. A loan that made sense at four percent doesn't at eight. New borrowing slows, then stops.

Why does a higher rate bite so hard? Because most debt isn't paid off, it's rolled. A company with a five-year loan refinances it, and if the rate has doubled, the same loan now eats twice the cash every month. That cash was going to be somebody's income. Multiply that by every loan in the country coming due, and you have the brake.

It also bites late. A rate change today reaches the economy over the next year to eighteen months, as loans roll and budgets reset. That lag is why the committee overshoots in both directions: it keeps pressing until the damage shows, and by then it has pressed too long.

Gear four is the stall. Now run 2.1 backwards. Less credit means less spending. Less spending means less income for whoever was being paid. He cuts his spending, and the next person's income falls. Companies see orders drop and cut staff. A recession is what the chain looks like when credit is pulled, and it's always credit that gets pulled first.

Gear five is the refill. With the economy cold and prices calm, the central bank cuts. Existing debt gets refinanced at the new lower rate, which frees up cash. New borrowing restarts. The next turn begins.

Here's what makes it a cycle rather than a circle. Each turn ends with more debt outstanding than it started with, because the refill never takes the pile back to zero. It just makes the pile cheaper to carry. Hold that thought. It's the whole of the next lesson.

Why five to eight years? Because that's roughly how long it takes for cheap credit to build enough heat to force the bowl out, and then for the stall to cool things enough to bring it back. It isn't a law of nature. It's the reaction time of a committee plus the reaction time of a hundred million borrowers.

The National Bureau of Economic Research is the official scorekeeper. A committee of economists dates every peak and trough, usually a year after the fact.

Twelve complete cycles since 1945. The average, peak to peak, is seventy five months. Call it six years and change, which is why the title of this lesson says five to eight. The shortest expansion lasted a year.

The longest, the one that ended in February 2020, lasted a hundred and twenty eight months, nearly eleven years, and ended in a two-month recession that had a virus for a trigger and a credit stop for a mechanism.

Now lay the throttle under it. The Fed's rate was under one percent when the series begins in 1954. It reached nineteen percent in the summer of 1981, the hardest punch bowl removal in the record, and the recession that followed ran sixteen months.

It touched near zero for most of the 2010s. It sat above five percent through 2023 and has been cut since; the August reading is 3.63 percent.

The turn we're in is worth a paragraph because it's the open case. The rate went from near zero to above five in about a year and a half, the fastest removal since 1981, with unemployment at 3.7 percent when it peaked. The textbook said a stall had to follow. The scorekeeper hasn't dated one.

Unemployment drifted up to a bit over four, the committee started cutting, and the question the desk asks every morning is whether gear four was skipped or only delayed. Nobody knows yet. That's what makes it a cycle and not a formula.

Look at the shape. Every peak in the rate is followed, within a year or two, by a shaded recession. Every valley is followed by an expansion. The line is the schedule. The cycle doesn't have a fixed length because the throttle is set by people, but it has never once been skipped.

The incentive map. Who made the rule? Congress, when it gave a committee the power to set the price of credit for the whole country. Who gains on the way up? Borrowers, asset holders, anyone whose income is a percentage of spending.

Who pays when the bowl comes out? The last people to borrow, the first people to be laid off, and anyone holding a long loan at a low rate who now needs to sell.

Here's the reveal. You've been taught the economy runs on confidence, on animal spirits, on the mood of the consumer. Martin knew better in 1955 and said it to the people least willing to hear it. The mood follows the rate. The committee moves the price of credit, credit moves spending, spending moves income, and income moves the mood, in that order, every time. Confidence is the last gear, not the first.

A digital tycoon sells into this cycle and invests through it. The office reads the lesson three ways.

First, know which gear you're in. The desk watches the policy rate, the unemployment rate, and the gap between long and short Treasury yields, which turns negative when the market expects the bowl to come out and turns positive when it expects the refill.

This week the rate is a bit above three and a half percent after a run of cuts, unemployment is near four, and the curve is positive. Read that as: the refill is under way, the heat hasn't built yet. That's a reading, not a forecast.

Second, borrow like the cycle is real. Fixed rates before the bowl comes out; short or none when it's out. A business that needs to borrow in gear four is a business that gets repriced at the worst moment. Runway is the office's answer, and the lesson on cash and runway will give you the number.

Third, buy the stall, not the party. Every valley in the rate chart is where assets were cheapest. The office's rules say when to move; the cycle says the opportunity arrives on a schedule and the crowd calls it a catastrophe every time.

Our paper fund reads it by its rules: real prices, a written reason on every move, no leverage, so the bowl can never be pulled from under us.

• NBER business cycle reference dates since 1945: peaks and troughs Nov 1948 to Oct 1949 (11-month contraction, 37-month prior expansion); Jul 1953 to May 1954 (10, 45); Aug 1957 to Apr 1958 (8, 39); Apr 1960 to Feb 1961 (10, 24); Dec 1969 to Nov 1970 (11, 106); Nov 1973 to Mar 1975 (16, 36); Jan 1980 to Jul 1980 (6, 58); Jul 1981 to Nov 1982 (16, 12); Jul 1990 to Mar 1991 (8, 92); Mar 2001 to Nov 2001 (8, 120); Dec 2007 to Jun 2009 (18, 73); Feb 2020 to Apr 2020 (2, 128). NBER.
• Average cycle length 1945-2020: 75.0 months peak to peak, 74.5 months trough to trough. All cycles 1854-2020: 59.2 months peak to peak. NBER.
• William McChesney Martin Jr. was Fed chairman from April 2, 1951 to January 30, 1970. Federal Reserve History.
• The speech was delivered October 19, 1955, released 7:30 pm Eastern, before the New York Group of the Investment Bankers Association of America at the Waldorf Astoria. The quoted passages: "In the field of monetary and credit policy, precautionary action to prevent inflationary excesses is bound to have some onerous effects--if it did not it would be ineffective and futile." and "The Federal Reserve, as one writer put it, after the recent increase in the discount rate, is in the position of the chaperone who has ordered the punch bowl removed just when the party was really warming up." FRASER, the original release, page 12.
• Federal funds effective rate (FEDFUNDS): 0.80 percent in July 1954; 19.10 percent in June 1981; 9.81 in May 1989; 6.54 in July 2000; 5.26 in July 2007; 2.40 in July 2019; 5.33 in August 2023; 3.63 percent in May, June, July and August 2026. FRED.
• Unemployment rate (UNRATE): 3.7 percent in August 2023; 4.3 in May 2026; 4.1 percent in July and August 2026. FRED.
• 10-year minus 3-month Treasury spread (T10Y3M): +0.89 percentage points on September 11, 2026. FRED.
• "The majority of money in the modern economy is created by commercial banks making loans." Bank of England Quarterly Bulletin, 2014 Q1.
• The short-term debt cycle as five to eight years of credit expansion and contraction driven by the central bank's rate: Ray Dalio, How the Economic Machine Works, 2013.