Too Big to Fail: How Large Size Creates Systemic Fragility
"Too big to fail" sounds like a measure of importance. It's actually a measure of moral hazard.
When a bank becomes large enough that its failure would cascade through the entire financial system, governments face a choice: let it fail (catastrophic systemic damage) or bail it out (transfer the loss to taxpayers). The government chooses bailout. The bank survives. The taxpayer loses.
This isn't an accident or a one-time mistake. It's the structural outcome of allowing banks to grow past a critical size.
The Asymmetry Problem
A bank that is "too big to fail" has an asymmetric payoff structure: - Upside: Private. Profits, bonuses, executive compensation accrue to shareholders and officers. - Downside: Public. Losses from failed bets are socialized to taxpayers.
This is the definition of a free option. The bank makes money when things go well. The public loses money when things go poorly.
Given this structure, the rational strategy is to take large risks. Every bet is asymmetric: you keep the gains, the public bears the losses. Even if most bets fail, the occasional big win exceeds the accumulated losses of failures.
This is exactly what happened before 2008. Large banks, protected by the implicit government guarantee, took increasingly large risks. The risks accumulated silently until they couldn't be contained.
When the collapse came, the bank was bailed out. The taxpayers lost.
The Size Feedback Loop
The moral hazard problem created a size feedback loop:
- A bank becomes large
- Size creates "too big to fail" status
- The implicit government guarantee makes the bank's debt cheaper to borrow
- Cheap debt allows further growth
- Growth increases systemic importance
- Systemic importance reinforces the implicit guarantee
The bank becomes locked in growth. Staying the same size means losing the competitive advantage of cheap funding. The only option is to keep growing.
The tragedy: the banks that are most protected from failure are the ones most incentivized to take risks. The banks that are most fragile are the ones with the strongest government backing.
What Actually Happened
In 2008, financial institutions made massive bets on the assumption that housing prices would never decline nationally. This assumption was never tested — housing had risen for decades. The models were built on data from this one historical period.
When housing prices declined — for the first time in the data — the assumption collapsed. Institutions that were leveraged 30:1, 40:1, 50:1 on this assumption faced ruin.
The government bailed them out.
AIG, Citibank, Bank of America — all received billions in bailouts. The losses were socialized.
The irony: the banks that had the strongest balance sheets and best risk management before the crisis were the ones that benefited most from the bailout. They survived. The weaker, riskier institutions also survived because the government couldn't let them fail.
The Policy Failure
The failure wasn't in the bailout itself — faced with systemic collapse, bailout was necessary. The failure was in allowing banks to grow that large in the first place.
Taleb's solution is radical and simple: prevent banks from becoming too big to fail by limiting their size.
Not through increased regulation. Through simple limits: a bank cannot hold more than X% of national deposits. No bank can be larger than Y as a percentage of GDP. If banks want to grow past this, they must split.
This removes the systemic importance. Remove systemic importance, and the implicit government guarantee disappears. Remove the guarantee, and the incentive to take excessive risk disappears.
The alternative — regulating a "too big to fail" bank — is a game of regulatory whack-a-mole. Every regulation designed to prevent one type of risk creates incentive to take a different type of risk.
The Broader Problem
The "too big to fail" problem isn't unique to banking. Any institution large enough to damage the entire system if it fails transfers fragility to the public.
Companies become dependent on government bailouts (automotive, airlines). Pension funds become critical to the financial system (if they fail, millions lose retirement income). Infrastructure projects become "systemically important" (failure would cascade widely).
The principle is universal: size creates fragility that governments cannot allow to fail, which transfers fragility to the public.
The solution at each scale is the same: prevent the institution from becoming large enough that its failure matters systemically.
This isn't punishment of success. It's recognition that success that requires public support to survive is success purchased at public expense.