The 2009 stress tests published every big bank's hole — and private money filled it

By early 2009 the question hanging over American banking was simple and unanswerable from outside: which of the giant banks were secretly insolvent? Deposits and wholesale funding drifted toward whoever seemed safest, bank stocks priced in nationalization, and every official denial was read as confirmation. Treasury's rescue capacity was already strained, and the nineteen largest bank holding companies — holding two-thirds of US banking assets and more than half its loans — could not be recapitalized quietly, one rumor at a time.

From late February, more than 150 examiners from the Federal Reserve, the OCC and the FDIC sat inside all nineteen firms at once, running every balance sheet through the same 'more adverse' downturn scenario. Bernanke framed it carefully: not a solvency test but a buffer test — how much extra capital does each bank need to keep lending even if the economy does worse than expected? On May 7 the answer went public in one release: ten firms needed another $74.6 billion between them; Bank of America alone $33.9 billion.

The design turned a leak-prone guessing game into a priced event. Everyone learned every grade at the same moment, and Treasury pre-committed the Capital Assistance Program behind any bank that could not fill its hole privately within six months. That converted the worst case from 'unbounded losses, maybe nationalization' into something bounded, public and bridged — a risk investors know how to price. As Bernanke put it, the results 'should provide considerable comfort to investors and the public.'

Publish each bank's hole, backstop the tail, market pays

The panic ran on asymmetric information: each investor knew the banks knew more, so every bid assumed the worst. Simultaneous public grades collapsed the asymmetry instead of subsidizing around it.

The Treasury backstop was pre-committed and bounded, so investors could value each bank's worst case exactly. Uncertainty, not expected loss, was the binding constraint.

One scenario for all nineteen firms made the grades comparable and stopped the numbers from being negotiated bank by bank.

Framing it as a buffer test rather than a solvency test avoided both self-fulfilling runs and the legal optics of declaring large banks insolvent or sound by fiat.

The six-month window to raise private capital first meant government money served mostly as a guarantee, not a disbursal.

When panic is fed by uncertainty, not insolvency, credible simultaneous disclosure beats secretive support: publish the exposure, pre-commit the backstop, and private money re-prices and returns.

Ten of the nineteen firms were told to add a combined $74.6 billion of common equity; Bank of America's share alone was $33.9 billion. Each had six months to raise it privately, with Treasury standing behind any that could not. The one-off exercise became permanent: the Federal Reserve now runs supervisory stress tests every year and folds the results directly into capital requirements through the stress capital buffer. SCAP is remembered as the hinge of 2009 — the day the official numbers replaced the market's nightmare.