What Actually Causes Financial Crises
Every financial crisis feels like a surprise. Every single one. And then, six months later, everyone says “it was obvious.” The 2008 meltdown, the 2020 COVID crash, the 2023 banking scare – each time, the post-mortems reveal the same ingredients that have been causing financial disasters since the Dutch tulip bubble. The recipe has not changed in 400 years. What changes is the packaging. So let us unpack the recipe, because understanding it is the single most useful thing you can do for your portfolio before the next crisis arrives. And it will arrive.
Why Does the Same Thing Keep Happening?
Financial crises are not random meteors falling from the sky. They are the predictable result of human behavior interacting with financial systems that amplify it. The core ingredients are always the same, just wearing different costumes.
Leverage – the universal accelerant. Every major crisis has excessive leverage at its heart. In 2008, investment banks were leveraged 30:1 or higher. In 2023, Silicon Valley Bank had concentrated its balance sheet in long-duration bonds that collapsed in value when rates rose – a form of hidden leverage through duration mismatch. Leverage is what turns a manageable 5% loss into an existential 100% loss. It is the difference between a kitchen fire and the building burning down.
The insidious thing about leverage is that it is invisible until it matters. Derivatives create leverage without requiring upfront capital. Complex financial instruments obscure how much borrowed money is really at play. A hedge fund can have modest-looking positions that carry enormous hidden leverage through options, swaps, and futures. When things go wrong, everyone discovers the true exposure at the same time, and that discovery itself accelerates the panic.
Complexity – the fog machine. If leverage is the accelerant, complexity is what prevents anyone from seeing the fire until it is too late. Before 2008, mortgage-backed securities were sliced and diced into tranches so complex that the people selling them did not fully understand them, the rating agencies could not properly assess them, and the regulators could not track them. The complexity was not accidental. It was profitable. Each layer of complexity generated fees, and each layer made it harder to see the underlying risk.
In 2025, the complexity has migrated. Quantitative trading strategies, algorithmic market making, decentralized finance protocols with nested smart contracts, and exotic structured products create similar fog. The specific instruments change. The fog does not.
Complacency – the memory eraser. This is the human element. After a long period of stability and rising prices, people stop worrying. Risk premiums shrink. Lending standards drop. Everyone starts to believe that the current conditions are permanent. From 2003 to 2006, the U.S. housing market had not declined nationally in decades, so lenders built their models on the assumption it never would. By 2007, people were making mortgage payments they could barely afford, betting that rising home prices would bail them out. When that assumption broke, the entire edifice collapsed.
The economist Hyman Minsky formalized this pattern decades ago. His insight was elegant and slightly depressing: stability itself creates instability. Here is how it works. During good times, borrowers take on more debt because they feel confident. Lenders approve riskier loans because defaults are low. Investors accept lower returns for higher risk because nothing bad has happened recently. This gradual shift from cautious lending to aggressive speculation happens naturally, almost inevitably, during any prolonged period of prosperity. Eventually the system reaches a tipping point – now called the “Minsky moment” – where even a small shock reveals that the accumulated risks are unsustainable, and the whole thing unravels.
You can see this pattern in every crisis. By 2019, corporate debt was at record levels, covenant-lite loans were everywhere, and credit spreads were thin. Then COVID hit in 2020, and within weeks, the corporate bond market froze. In 2022-2023, after a decade of near-zero interest rates, many institutions and startups had built their entire business models around cheap money. When rates rose aggressively, the weakest links snapped: crypto lenders, regional banks, commercial real estate funds.
Misaligned incentives – the invisible hand pushing in the wrong direction. Compensation consultants, fund managers, and loan officers all face the same structural problem: they get paid based on short-term performance, but the risks they take manifest over years. A mortgage broker in 2006 got paid for originating loans, regardless of whether those loans were ever repaid. A hedge fund manager collecting 2-and-20 has enormous incentive to take risks that look brilliant 95% of the time and catastrophic 5% of the time – because the fees collected during the good years are never clawed back.
The accounting profession plays a role too. When companies can mark illiquid assets “to model” rather than “to market,” they can report profits that do not exist. Bonuses get paid on phantom earnings. By the time reality catches up with the accounting, the people who created the mess have already cashed their checks. This is not cynicism. This is the documented history of Long-Term Capital Management in 1998, Enron in 2001, Bear Stearns in 2007, and countless smaller blowups in between.
What Actually Triggers the Crash?
Here is the uncomfortable truth: the trigger almost does not matter. It could be an archduke getting shot. It could be a virus. It could be a mid-size bank making bad bets on long-term bonds. The trigger is just the match. The gasoline – leverage, complexity, complacency, misaligned incentives – was already pooled on the floor.
In 1998, a Russian debt default and the collapse of a single hedge fund nearly froze global credit markets. In 2008, it was subprime mortgages. In 2020, a pandemic. In 2023, a few regional banks with concentrated bond portfolios. None of these triggers were individually large enough to cause a global crisis. What made them dangerous was the system’s vulnerability at the moment they struck.
This is why predicting the next crisis is both impossible and unnecessary. You will never guess the trigger. But you can observe the preconditions: Are credit spreads unusually tight? Is leverage rising? Are lending standards loosening? Is everyone confident that a particular asset class “only goes up”? When these conditions align, a crisis is not a question of if, but when. The specific catalyst is almost irrelevant.
The concept of the “crowded trade” makes this worse. When thousands of fund managers are using similar strategies, similar models, and similar risk management rules, they will all respond to the same signal at the same time. Algorithmic trading amplifies this – when market conditions hit certain thresholds, automated systems start selling simultaneously, which pushes prices down further, which triggers more automated selling. This feedback loop is what turned Black Monday in 1987 from a bad day into a 23% single-day crash. Portfolio insurance – essentially automated stop-loss orders – was marketed as risk reduction but created systemic risk because everyone had the same plan for the exit. When they all headed for the door at once, the door was not wide enough.
In 2025, the same dynamic exists in different form. Quantitative funds managing trillions of dollars use correlated strategies. Passive index funds amplify momentum in both directions. 0DTE options create massive gamma exposure that market makers must hedge in real time. The plumbing has changed. The physics has not.
How Do You Prepare Before the Next One Hits?
If crises are inevitable, the question is not how to avoid them but how to survive them – and ideally, how to profit from the aftermath. Here is a practical framework.
Accept that crises are normal, not exceptional. Since 1900, the U.S. stock market has experienced a decline of 20% or more roughly once every 7-10 years, and a decline of 10% or more roughly every 2-3 years. Treating these as shocking, unprecedented events is a failure of imagination. Build your portfolio expecting them.
Manage leverage like your portfolio’s life depends on it – because it does. The single most common cause of permanent capital loss is not picking the wrong stock. It is leverage. A portfolio that drops 50% needs to gain 100% to recover. A leveraged portfolio that gets a margin call at the bottom has no chance to recover at all. The difference between temporary loss and permanent destruction is almost always leverage. No matter how smart your analysis, anything multiplied by zero is zero.
Keep liquidity reserves. Cash is not a drag on performance during a crisis. It is the weapon. The investors who did best after 2008 and 2020 were the ones who had cash available to buy when everyone else was forced to sell. Having 10-20% of your portfolio in cash or short-term bonds during frothy markets is not cowardice. It is strategy.
Diversify across truly uncorrelated assets. During a crisis, correlations spike – stocks, corporate bonds, real estate, and commodities can all fall together. But some things hold up better than others. Government bonds (especially short-duration), cash, and certain alternative strategies can provide ballast. The key is diversifying before the storm, not during it.
Watch for the preconditions, not the trigger. When you see excessive leverage, loose lending standards, compressed risk premiums, and widespread investor complacency, reduce risk. You do not need to know what will cause the next crisis. You just need to recognize when the system is fragile.
Key Takeaways
Financial crises are caused by the same four ingredients every time: excessive leverage, complexity that hides risk, complacency born from prolonged stability, and incentive structures that reward short-term risk-taking.
The Minsky insight is critical: stability breeds instability. Long periods of calm encourage exactly the behavior that makes the next crisis worse.
The specific trigger – a pandemic, a bank failure, a debt default – is almost irrelevant. What matters is how much hidden fragility has accumulated in the system.
You cannot predict crises, but you can prepare: avoid leverage, maintain cash reserves, diversify genuinely, and pay attention when risk premiums get unusually low and confidence gets unusually high.
Crises are not bugs in the financial system. They are features. They have happened roughly every decade for centuries, and they will keep happening. The investors who thrive are not the ones who avoid crises – nobody can do that – but the ones who are prepared when they arrive.
The best time to fix your roof is when the sun is shining. In investing terms, the best time to build crisis resilience is when everything looks fine. Because by the time it does not look fine, it is already too late.