Too Big to Fail: Too Big to Exist

The phrase "too big to fail" is usually read as reassurance. If the institution were failing, the government would save it. The institution is too important to let collapse.

This is the wrong reading. The phrase should be read as a warning. If an institution is too big to fail, it is too big to exist.

What the Label Really Means

An institution described as "too big to fail" is one whose failure would have system-wide consequences. Its counterparties would fail. Its clients would lose access. Its role in critical infrastructure would break. The costs of its failure would be borne not by the institution or its executives but by the entire system—which means by society, by taxpayers, by everyone else.

The label is not a compliment. It is a description of systemic risk.

The Incentive Problem

Once an institution is understood to be too big to fail, the incentive structure breaks. The institution knows that:

This produces a moral hazard. The institution takes risks it would not take if failure were possible. It bets big, using cheap borrowed money, knowing that the losses—if they materialize—will be socialized.

The 2008 financial crisis demonstrated this perfectly. Banks took enormous risks using leverage because leverage was cheap and because the banks understood (correctly) that their failure would be unacceptable to the government. When the risks materialized, the banks were bailed out. The executives kept their prior-year bonuses. The public absorbed the losses.

The 2008 Bonuses from Bailout Funds

In the aftermath of the 2008 financial crisis, banks received trillions of dollars in emergency support from taxpayers. In many cases, the same banks continued paying executive bonuses out of the bailout funds.

The asymmetry was perfect: upside was privatized (executives kept their bonuses). Downside was socialized (the public paid for the bailout).

This is not capitalism. Capitalism requires that those who make the decisions also bear the consequences. This was capitalism for profits, socialism for losses.

The violation produced political fury and deep loss of trust in financial institutions that has never fully recovered. Every society that tolerates this asymmetry is storing political instability that will be paid later.

The Policy Response That Works

The correct policy response is not to make institutions "unfailable"—to guarantee they will never fail. This is impossible and creates moral hazard.

The correct policy response is to make institutions small enough that their failure is survivable.

If a bank is so large that its failure would break the financial system, the bank is too large. The solution is not to protect the bank. The solution is to break it into pieces small enough that the failure of any piece is survivable.

This is not theoretical. The Glass-Steagall Act (repealed in 1999) had required that commercial and investment banking be separated. This separation meant that a failure in investment banking wouldn't break the commercial banking system and vice versa. The separation created redundancy. It made the system more robust.

When Glass-Steagall was repealed, the regulatory justification was that unified banks could be more efficient. The real effect was to concentrate risk. Banks grew larger. Dependencies multiplied. The 2008 crisis was the result.

Why Banks Keep Growing

After 2008, regulators imposed capital requirements and stress tests meant to reduce the risk of large banks. But banks have continued to grow. They are larger now than they were before 2008.

Why? Because of the moral hazard. A large bank knows it is too big to fail. The capital requirements are real but not sufficient to change behavior. The bank still takes risks it would not take if failure were possible.

The regulations are meant to reduce the risk from large size. But they don't eliminate the fundamental problem: the bank is large, and its failure is unacceptable to the system.

The only real solution is size reduction.

Why Size Reduction Doesn't Happen

Size reduction doesn't happen because:

  1. Banks don't want to shrink. Large size is profitable and provides pricing power.
  2. Executives don't want to shrink. Large banks pay larger salaries to executives.
  3. Regulators are captured. The largest banks have the most lobbying power and can influence regulation in their favor.
  4. Policymakers fear disruption. Breaking up a large bank involves regulatory complexity and the risk of short-term disruption.

So nothing happens. Banks continue to grow. Risk concentrates. The next crisis approaches.

The 2008–2023 Pattern

The 2008–2023 period featured multiple "too big to fail" failures and emergency interventions:

Each required emergency intervention at massive public cost. The lesson has been taught repeatedly and not learned institutionally.

The banks are larger than in 2008. The concentration of risk is greater. The next crisis will be more expensive.

The Historical Precedent

The Code of Hammurabi, roughly 3,700 years old, required that builders of collapsed houses die within the rubble. The responsibility was direct and inescapable: if your structure fails and kills people, you are held accountable.

This rule—that those who make the decisions also bear the consequences—is ancient and nearly universal across societies that have learned to be stable.

Modern capitalism has largely abandoned this rule for large institutions. The executives who made the decisions don't lose everything when the institution fails. The institution is bailed out. The executives move on.

A system that does not align the decision-maker's interest with the decision's consequences is a system that will accumulate catastrophic errors.

What Would Actually Work

Three policies would actually reduce the fragility from too-big-to-fail:

  1. Hard cap on bank size. No bank should be larger than, say, 2% of GDP. Once you reach that size, you cannot grow. Period.

  2. No bailouts. If a bank fails, let it fail. Accept the disruption. It will hurt, but it will also create incentives for size reduction.

  3. Skin in the game. Executives of large institutions should have their personal wealth tied to the institution's solvency. Bonuses should be in long-term equity, not cash. If the institution fails, the executive fails with it.

These policies would eliminate the "too big to fail" problem by ensuring that institutions are either small enough that failure is survivable, or that decision-makers bear the consequences of failure.

Neither policy is implemented because both threaten the largest banks and their executives.

So we wait for the next crisis, knowing it will be larger than the last one, and knowing we won't be prepared for it.