Economies do not grow in a straight line — they run in cycles, driven mostly by credit. This interactive explorer walks through the mechanism in twelve short lessons, then lets you step through the four phases of a cycle, compare yield-curve shapes, see what rising and falling interest rates typically do, and test what stuck with a thirteen-question quiz. It is an educational model of how the machine works, not a forecast and not advice.
What an economic cycle actually is
An economy is the sum of its transactions, and a transaction is one buyer handing money or credit to one seller. That second word is the whole story. Credit can be created quickly out of a promise to repay, it spends exactly like money, and it lets an economy consume more than it produces for a while — which means it must consume less than it produces later. That push and pull is what makes activity cyclical rather than steady.
The loop is self-reinforcing in both directions, because one person’s spending is another person’s income. More spending means more income, which supports more borrowing, which funds more spending. Run it in reverse and the same mechanism produces a contraction. Underneath both sits productivity growth — the slow, comparatively steady rise in what can be produced from the same effort, which is the only durable source of higher living standards.
The two cycles running at the same time
The framework in this tool separates two cycles that are usually discussed as one, because confusing them leads people to expect the wrong remedy.
- The short-term debt cycle — roughly five to eight years. Credit expands, spending outruns production, prices rise, the central bank raises rates, borrowing and spending fall, activity contracts, the central bank cuts, and the cycle restarts. The policy rate is the throttle.
- The long-term debt cycle — decades. Across many short cycles, debt tends to rise faster than income, because people generally prefer to borrow and spend rather than pay down. Eventually debt service outgrows income and the loop reverses at a point where rates are already near zero — so the usual remedy is unavailable. That is a deleveraging, and it is a different problem from a recession.
There are only four ways a debt burden comes down: austerity, restructuring or default, redistribution, and money creation. The first three are deflationary and the fourth is inflationary, which is why the balance between them — not any one of them — determines whether the process is orderly. The lessons in the tool work through each.
The four phases, briefly
- Expansion — credit is available, output and employment grow, confidence builds. Economies spend most of their time here.
- Peak — maximum output, stretched capacity, the highest price pressure, and a central bank that is usually tightening. Only ever obvious in hindsight.
- Contraction — credit tightens, spending falls, unemployment rises. Historically much shorter than expansions, and responsible for most of the visible damage.
- Trough — the low point and the point of maximum pessimism, which is exactly when the conditions for the next expansion are assembling. It never feels like a bottom from inside.
Select any phase in the explorer above and you get what is happening underneath, the indicator readings that typically accompany it, the market tendencies observed across past cycles, the prevailing psychology, and the mistake people most reliably make in that phase.
What the indicators can and cannot tell you
The single most useful habit when reading economic data is sorting indicators by when they move rather than by how much attention they get.
- Leading — yield curve, PMI, credit spreads, building permits. Noisy, frequently wrong, but they move first.
- Coincident — industrial production, real income, employment. They describe the present, which is genuinely useful and genuinely late.
- Lagging — the unemployment rate, CPI. Reliable descriptions of what already happened. Waiting for these to look healthy has historically meant acting well after the turn.
The uncomfortable state — leading indicators deteriorating while coincident ones still look fine — is the one that precedes most turns, and it is precisely the state in which confident predictions are worth least. No single reading identifies a phase; several read together, with their lags in mind, is the honest method.
Every U.S. recession since 1929
Cycle frameworks are easier to trust when you can see the actual record. Peaks and troughs below are the NBER Business Cycle Dating Committee’s official dates; peak unemployment is the highest monthly rate recorded around each episode (BLS); the equity column is the approximate peak-to-trough decline of the bear market associated with each recession.
| Recession | NBER peak → trough | Months | Peak unemployment | Associated equity decline | What set it off |
|---|---|---|---|---|---|
| Great Depression | Aug 1929 → Mar 1933 | 43 | 24.9% | −86% | Crash, bank failures, monetary contraction |
| Roosevelt Recession | May 1937 → Jun 1938 | 13 | 19.0% | −54% | Premature fiscal and monetary tightening |
| Post-war demobilisation | Feb 1945 → Oct 1945 | 8 | 3.9% | −27% (1946) | Wartime spending collapsed |
| 1948–49 recession | Nov 1948 → Oct 1949 | 11 | 7.9% | −21% | Post-war inventory adjustment |
| 1953–54 recession | Jul 1953 → May 1954 | 10 | 6.1% | −15% | Korean War wind-down, Fed tightening |
| Eisenhower recession | Aug 1957 → Apr 1958 | 8 | 7.5% | −21% | Fed tightening against inflation |
| 1960–61 recession | Apr 1960 → Feb 1961 | 10 | 7.1% | −14% | Fed tightening, fiscal restraint |
| 1969–70 recession | Dec 1969 → Nov 1970 | 11 | 6.1% | −36% | Vietnam spending pullback, tightening |
| Oil-shock recession | Nov 1973 → Mar 1975 | 16 | 9.0% | −48% | OPEC embargo; stagflation |
| 1980 recession | Jan 1980 → Jul 1980 | 6 | 7.8% | −17% | Volcker’s first tightening round |
| Volcker recession | Jul 1981 → Nov 1982 | 16 | 10.8% | −27% | Deliberate disinflation at a 20% policy rate |
| 1990–91 recession | Jul 1990 → Mar 1991 | 8 | 7.8% | −20% | S&L crisis, Gulf War oil spike, tightening |
| Dot-com bust | Mar 2001 → Nov 2001 | 8 | 6.3% | −49% | Tech bubble unwind; 9/11 |
| Great Recession | Dec 2007 → Jun 2009 | 18 | 10.0% | −57% | Subprime crisis, Lehman, global contagion |
| COVID recession | Feb 2020 → Apr 2020 | 2 | 14.7% | −34% | Pandemic lockdowns |
The bear-market windows do not line up with the NBER dates — markets have usually turned down before the recession was dated and turned up before it ended, which is the point the trough phase makes. Pre-1957 equity figures use the S&P Composite and vary between sources; treat one-point precision as false precision.
And the expansions in between
Derived from the same NBER dates — each expansion runs from one trough to the next peak. Put beside the table above, it makes the asymmetry hard to miss: contractions are the short, loud part of the cycle, and expansion is the state an economy is normally in.
| Expansion | Months |
|---|---|
| Mar 1933 → May 1937 | 50 |
| Jun 1938 → Feb 1945 | 80 |
| Oct 1945 → Nov 1948 | 37 |
| Oct 1949 → Jul 1953 | 45 |
| May 1954 → Aug 1957 | 39 |
| Apr 1958 → Apr 1960 | 24 |
| Feb 1961 → Dec 1969 | 106 |
| Nov 1970 → Nov 1973 | 36 |
| Mar 1975 → Jan 1980 | 58 |
| Jul 1980 → Jul 1981 | 12 |
| Nov 1982 → Jul 1990 | 92 |
| Mar 1991 → Mar 2001 | 120 |
| Nov 2001 → Dec 2007 | 73 |
| Jun 2009 → Feb 2020 | 128 |
Across the whole table the median expansion runs 58 months against a median recession of 10 months; counting only the post-war episodes it is 58 months of expansion against 10 of contraction. The longest expansion on record — Jun 2009 → Feb 2020 — ran 128 months.
Peak unemployment in the table above is worth reading against this one. In 1990–91 the recession ended in March 1991 and unemployment did not peak until mid-1992; in 2007–09 it ended in June 2009 and unemployment peaked that October. That gap is the lagging indicator doing exactly what a lagging indicator does.
Fed policy-rate cycles, 1979–2023
The short-term debt cycle’s throttle, as actually used. Target for the federal funds rate; the modern target range replaced a point target in December 2008.
| Period | Policy rate | What it was |
|---|---|---|
| 1979–1981 | ≈11% → 20% peak | Volcker’s deliberate disinflation. Inflation fell from roughly 15% in 1980 to about 3% by 1983; unemployment reached 10.8% on the way. |
| 1984–1986 | 11.75% → 5.9% | Easing once inflation had broken. |
| 1994–1995 | 3.00% → 6.00% | Preemptive tightening with inflation still low. Triggered a severe bond sell-off, and was followed by a long expansion. |
| 1999–2000 | 4.75% → 6.50% | Late-cycle tightening into the dot-com peak. |
| 2001–2003 | 6.50% → 1.00% | Post-bubble easing, held low for an extended period. |
| 2004–2006 | 1.00% → 5.25% | Seventeen consecutive 25bp increases at a self-described ‘measured pace’. |
| 2007–2008 | 5.25% → 0–0.25% | The financial crisis. The rate reached its floor and the balance sheet became the policy instrument. |
| 2015–2018 | 0–0.25% → 2.25–2.50% | The slowest normalization on record, followed by three cuts in 2019. |
| 2020 | 1.50–1.75% → 0–0.25% | Two emergency cuts inside two weeks, plus large-scale asset purchases. |
| 2022–2023 | 0–0.25% → 5.25–5.50% | 525 basis points in about sixteen months — the fastest tightening since the early 1980s. |
Deliberately stops at 2023. A page like this cannot keep a current-rate figure accurate, and a stale number presented as current is worse than none — for where the rate is today, read the Fed’s own statements or the FRED series.
How this explainer is built
The lesson sequence follows the structure of Ray Dalio’s widely published explainer How the Economic Machine Works, written here in our own words: transactions, credit, productivity, the short- and long-term debt cycles, the four levers of a deleveraging, and his three closing rules of thumb. Around that skeleton, the phase descriptions, indicator material, yield-curve section and central-bank balance-sheet material are standard macroeconomics drawn from primary sources.
- Ray Dalio, How the Economic Machine Works — economicprinciples.org
- NBER Business Cycle Dating Committee — recession dates and how they are determined
- Federal Reserve — H.4.1 balance-sheet release, and the FRED database for rate and yield series
- Federal Reserve Bank of New York — yield-curve recession-probability model
- U.S. Bureau of Labor Statistics (CPI), ISM (PMI), and Cboe (VIX) for the indicator definitions
Two deliberate omissions. The tool carries no live data and no current-conditions readout: a static page quoting today’s inflation print or Fed balance sheet is wrong within weeks, and a stale number presented as current is worse than no number. And it makes no call on which phase we are in, for the reason given in the FAQ below. Where historical market behavior is described, it is described as a tendency measured across past cycles after the fact — never as a prediction and never as a recommendation.
Common mistakes with cycle frameworks
The first is treating a framework as a timing device. Knowing that risk is accumulating tells you nothing about when it will be realized, and the gap between those two things has ruined far more portfolios than ignorance of cycles ever did.
The second is single-indicator thinking — usually the yield curve, occasionally the unemployment rate. Every indicator here has failed at least once, and the ones with the best records have the widest lead times, which makes them nearly useless for acting on.
The third is forgetting that you are inside the psychology the model describes. The framework says confidence peaks at the top and despair at the bottom; that applies to the person reading it. Recognizing the pattern in the abstract is easy, and acting against your own sentiment in real time is not.
Frequently asked questions
How long is an economic cycle?
There is no fixed length. The short-term debt cycle — the business cycle most people have lived through — has typically run something like five to eight years in post-war economies, but the range is wide: some expansions have lasted a decade or more, and contractions have often been under a year. The long-term debt cycle plays out across decades. Anyone quoting a precise cycle length is describing an average, not a schedule.
Can this tool tell me what phase we are in right now?
No, and that is deliberate. It teaches you what each phase looks like and which indicators lead, coincide with, or lag the turn — but it carries no live data and makes no call on the present. Phases are identified confidently only in hindsight; in the United States the NBER’s dating committee routinely confirms a turning point a year or more after it happened. A tool that told you today’s phase would be selling you false precision.
Is the yield curve really a reliable recession predictor?
It is the strongest single signal in modern U.S. data — an inverted term spread has preceded every U.S. recession since the 1960s, which is why the New York Fed publishes a recession-probability model built on it. It is also routinely oversold. The lead time has varied from roughly six months to two years, there has been at least one widely cited false signal, different maturity pairs invert at different times, and the whole track record rests on a handful of recessions. Respect it as a warning; do not treat it as a countdown.
Does understanding cycles help with investing?
It helps with expectations and with sizing risk — knowing that contractions are normal, that unemployment peaks after the trough, and that markets have historically turned before the data did makes it easier to hold a plan through a bad year. What it does not do is give you timing. The gap between ‘this risk is building’ and ‘this is when it breaks’ is where most confident predictions die, and no framework closes it. Nothing here is financial advice.
Whose framework is this based on?
The lesson sequence follows the structure of Ray Dalio’s widely published explainer How the Economic Machine Works — transactions, credit, productivity, the two debt cycles, and the four ways a debt burden comes down. The explanations here are written in our own words and mixed with standard macroeconomic material on indicators, the yield curve and central-bank balance sheets. It is an independent educational implementation, not affiliated with or endorsed by Ray Dalio or Bridgewater Associates.
Does the tool store anything about me?
No. Your lesson progress and quiz answers live in the page’s memory only, the tool makes no network requests, and nothing is written to storage or sent anywhere. Reload the page and it starts fresh.
Keep reading
The articles this tool was built alongside, in the order they make sense:
- The Economic Machine Explained: How Credit, Debt, and Cycles Actually Work — the long-form companion to this tool
- Short-Term vs Long-Term Debt Cycles — why the distinction changes what a downturn means
- Why Interest Rates Move Markets — the policy rate, and what it actually transmits to
- The Yield Curve Inversion, Explained Without the Jargon — the signal, and its real track record
- Recession Indicators: Which Ones Actually Predict Downturns? — which gauges have held up, and which have not
- What Is Inflation, Really? — the mechanism behind the number
- How Banks Actually Create Money — where credit comes from in the first place
- Stagflation 101 — what happens when the usual trade-off breaks
Disclaimer: This tool and page are educational content, not financial advice. I am not a licensed financial advisor, and nothing here is a recommendation to buy or sell any security or asset. The cycle model described is a simplification of a complex system; it can be wrong, late, or misleading, and historical patterns do not guarantee future results. Investing and trading involve risk, including the possible loss of the money you invest. Do your own research and consider consulting a licensed financial professional before making investment decisions. Read the full Disclaimer.
Attribution: The framework taught here follows Ray Dalio’s publicly available explainer How the Economic Machine Works. This is an independent educational implementation in our own words; it is not affiliated with, authorized by, endorsed by, or reviewed by Ray Dalio or Bridgewater Associates, and it reproduces none of the original’s text or artwork.