Now the engine room. The business cycle described so far is real, but its amplitude, the difference between a mild inventory wobble and a 1929 or a 2008, is set overwhelmingly by one thing: credit. And the deepest theory of why credit cycles turn violent, and why they are endogenous (generated by the system itself, not just by outside shocks), belongs to a heterodox American economist who was largely ignored in his lifetime and then, in September 2008, suddenly explained everything: Hyman Minsky.
What is Minsky's financial instability hypothesis?
The Financial Instability Hypothesis (FIH), in one devastating sentence: stability is destabilizing. Minsky's insight inverts the intuition that a calm, prosperous economy is a safe one. His claim is that the calm itself plants the crisis, through a change in behaviour that a long expansion makes almost inevitable:
- Start from a period of stability: the economy is growing, cash flows are strong, asset prices are rising, and defaults are rare.
- Because defaults are rare, lenders and borrowers alike become more confident. Loans made two years ago were repaid; the borrower who looked risky turned out fine; the loan officer who demanded a big down-payment lost the deal to a rival who didn't. Success rewards boldness and punishes caution, so both sides rationally take more risk: lenders accept thinner margins of safety (higher loan-to-value, less collateral, weaker covenants, lower spreads), and borrowers take on more leverage.
- Rising asset prices validate the extra leverage retroactively. The borrower who over-paid for the building is bailed out by the building's rising value, which "proves" that leverage was smart and encourages more of it. Leverage begets rising prices begets more leverage, a self-reinforcing loop.
- Over the expansion, the composition of the economy's financing therefore migrates, steadily and invisibly, from safe structures toward fragile ones, until the system is so leveraged and so dependent on continued asset-price rises and easy refinancing that a small disturbance can bring it down.
Advertisement
What are hedge, speculative and Ponzi finance?
The three financing structures (define these precisely; they are the most examined idea here). Minsky classified every borrowing unit (a household, a firm, a bank, a country) by a single cash-flow test: can its expected income cover its debt obligations, and which ones?
- Hedge finance, the safe structure. Expected operating cash flow covers both the interest and the principal repayment in every period, from income alone. The unit does not depend on refinancing or on selling assets; it can service and retire its debt from what it earns. Example: a profitable FMCG company with a term loan whose EMIs (interest + principal) are comfortably covered by operating cash flow several times over; a homeowner on a normal amortizing mortgage whose salary easily covers the payment. Hedge units are robust to rising rates and frozen credit markets — they don't need the markets.
- Speculative finance, the fragile-but-solvent structure. Expected cash flow covers the interest, but not the principal. The unit is solvent (it earns enough to pay interest) but illiquid-dependent: it must continually roll over (refinance) the principal as each tranche matures, because it cannot repay it from income. It is betting that credit markets stay open and rates stay manageable. Example: almost every bank is structurally speculative, funding long-term loans with short-term deposits it must continually roll; so is a company that funds a long-lived asset with commercial paper it rolls every 90 days. Speculative units survive as long as they can refinance — and die the moment they can't.
- Ponzi finance, the doomed structure. Expected cash flow covers neither principal nor even all the interest. The unit can only meet its obligations by borrowing still more, or by selling assets, so it is betting entirely on rising asset prices (sell the appreciated asset to pay the debt) or on ever-easier credit (borrow more to pay the interest). Named, of course, for Charles Ponzi, because it has the structure of a Ponzi scheme: old obligations paid only with new money. Example: a subprime borrower in 2006 on an interest-only or negative-amortization "teaser" mortgage whose income never covered the real payment, who could only "afford" the house by refinancing against its rising price: fine while prices rose, insolvent the instant they stopped. Or a real-estate developer paying the interest on old project loans with the proceeds of new project loans. Or IL&FS (see India's business cycle). Ponzi units are alive only while the asset keeps rising and credit keeps flowing.
What happens at a Minsky moment?
The Minsky moment. As the expansion matures, the share of speculative and Ponzi finance in the total rises, quietly, because everything is being repaid on schedule as long as refinancing is available and prices rise. Then something changes: the central bank raises rates (making refinancing costlier), or an asset price stops rising, or one prominent borrower fails and lenders get nervous. At that point the Ponzi units, who needed rising prices or easy credit, can no longer refinance. They are forced to sell assets to meet obligations. But forced selling lowers the asset's price, which pushes the next leveraged holder underwater, forcing their sale, lowering the price further: a self-reinforcing cascade. The Minsky moment (a term coined by PIMCO's Paul McCulley in 1998, during the Russia/LTCM crisis) is precisely this instant: the point at which over-leveraged units are forced into distress selling and the credit structure begins to collapse under its own weight. The calm didn't prevent the crisis; the calm built it.
Fisher's debt-deflation, the mechanism of the crash. Irving Fisher, ruined personally in 1929, wrote the classic account of why the cascade is so vicious ("The Debt-Deflation Theory of Great Depressions," 1933). As over-indebted units sell assets to pay down debt, they push asset (and goods) prices down, but the debt is fixed in nominal terms, so falling prices raise the real burden of the debt. Fisher's chilling paradox: "the more the debtors pay, the more they owe" in real terms. Each round of deleveraging makes the remaining debt heavier, forcing more selling. That is a deflationary spiral which, unchecked, is exactly how a recession becomes a depression (see the 1929 crash), and it is why the central bank's lender-of-last-resort role (M7.02) and fiscal support (M7.04) exist: someone must break the spiral by supplying money and demand when the private sector is all selling at once.
How do you read the credit cycle in practice?
The credit cycle as the analyst sees it. In practice, the Minsky dynamic shows up as a credit cycle you can read: in the boom, credit growth accelerates, lending standards loosen (the SLOOS survey, M7.02), credit spreads compress to complacent lows, covenant-lite lending proliferates, and leverage ratios climb; at the peak, spreads are at their tightest and standards at their loosest exactly when risk is highest (the cruel irony: credit is cheapest and easiest to get at the top); then a trigger widens spreads, standards snap tight, credit growth collapses, defaults surge, and the contraction is on. Watching the credit cycle (spreads, standards, and credit growth) is watching the amplitude dial of the whole business cycle, and it is why credit spreads are a premier leading indicator (see leading, coincident and lagging indicators).
How do different schools explain downturns?
Everything above is one theory of the cycle, and it is a minority one in academic economics. An analyst meets the other four constantly, in central bank speeches, in newspaper columns and in the assumptions behind a forecast, and mistaking one school's prescription for a consensus is how you end up surprised by a policy you should have expected. Learn to place a claim before you argue with it.
| School | What causes a downturn | What policy should do | The strongest objection to it |
|---|---|---|---|
| Neoclassical and real business cycle | Real shocks to productivity and preferences. Fluctuations are the economy's optimal response to a changed environment, not a failure | Very little. Stabilisation policy mostly adds noise; the useful work is supply-side | It struggles to explain why an "optimal" response involves millions of people wanting to work at the going wage and failing to find work |
| Keynesian | A collapse in aggregate demand, driven by expectations, that wages and prices are too sticky to clear quickly | Fiscal expansion, plus monetary easing, to fill the demand gap until private spending returns | Timing and politics: stimulus arrives late, is hard to withdraw, and can raise debt without raising potential output |
| Monetarist | Central bank error. The Great Depression was a monetary contraction the Fed allowed rather than an inevitable collapse (Friedman and Schwartz, 1963) | Rules over discretion: keep money and nominal spending growing at a steady, announced rate | The empirical relationship between money aggregates and nominal spending broke down after the 1980s as financial innovation moved the target |
| Austrian | Credit expanded below the natural rate distorts the structure of production toward projects that were never viable; the bust is the necessary liquidation | Let the malinvestment liquidate; intervention prolongs the adjustment and stores up the next one | The prescription accepts an enormous, concentrated human cost, and debt-deflation shows the liquidation can overshoot into a spiral that destroys sound firms too |
| Minsky and the credit cycle | Endogenous financial fragility. Stability itself shifts the economy's financing mix toward speculative and Ponzi structures until a small shock forces a general repricing | Lean against credit growth in the boom with macroprudential tools; backstop the system in the bust to stop the fire sale | It is better at explaining crises after the fact than at dating them in advance, and its policy implication requires acting when everything looks fine |
Read across the "what policy should do" column and a live disagreement becomes legible. A monetarist and a Keynesian both want the authorities to act in 2008 and disagree about which authority. An Austrian and a Keynesian disagree about whether the action helps at all. And a real-business-cycle economist and a Minskyian disagree about whether finance belongs in the model.
The frame chosen here is the credit-cycle one, and the honest reason is worth stating rather than assumed. It was chosen because it is the most useful to an analyst, not because it is the most likely to be correct. Its unit of analysis is the balance sheet, which is what you can actually read: the hedge, speculative and Ponzi test from the Minsky section runs on a company's own filings, on a bank's loan book, on a country's external accounts. A real-business-cycle model needs a productivity shock you cannot observe until afterwards. A monetarist model needs a money aggregate whose link to spending is now unstable. A Keynesian model gives you the right question about demand and no company-level instrument. Minsky gives you a test you can apply on a Tuesday afternoon to a name you cover.
The cost of that choice, stated plainly: the credit-cycle lens over-predicts crises. A framework whose central claim is that calm is dangerous will call a top several times before one arrives, which is exactly the failure the section on timing warns about and why the module insists on reading the regime rather than calling the turn. Practice item P19 asks you to identify the school behind a stated prescription, and getting that right is the difference between disagreeing with a policy and not recognising it.