By HAL S. SCOTT
Originally published by the Financial Times on March 1, 2018
Ten years after the start of the financial crisis, US policymakers are finally starting to deal with the stiff regulatory reaction that it spawned. The US Senate will soon take up the first bipartisan effort to rein in the impact of the Dodd-Frank financial reform bill on smaller banks within weeks. But that should be just one step in reform.
The bill would exempt banks with less than $100bn in assets from the Federal Reserve’s annual stress tests and the Federal Deposit Insurance Corporation’s annual living wills requirement that each bank explain how it could be safely wound down in a crisis.
However, the legislation would not eliminate the burdens those requirements create for banks with more than $100bn in assets. These 37 banks — 10 of which are headquartered overseas — account for more than 75 per cent of US banking assets. What they can or cannot do is critical to the economy.
Bank regulators and supervisors do not need to wait for legislation — there are a few simple steps they could take to to free these big banks up to provide more credit to the economy.
The Fed’s stress tests are effectively the binding constraint on bank capital and thus lending. They require banks to prove they could survive extreme adverse scenarios while still complying with global capital requirements. The process has two major deficiencies.
First, the Fed’s adverse scenarios are extreme to the point of incredulity. For example, the latest stress test assumes an increase of the unemployment rate from 4.1 to 10 per cent over seven quarters. That has not happened in the 70 years since today’s measure for unemployment was adopted. There is no open consultation with experts, industry or the general public as to whether these scenarios make sense.
Do we really want banks to hold enough capital to survive events that have no US historical precedent? If such an extreme economic event did occur, would any amount of capital be enough to withstand the panic it could trigger?
The second problem is that the Fed’s stress tests depend on secret government financial models to predict bank losses. Since these models are never held up to public scrutiny, there is no way to know whether they are accurate or well-justified.
By contrast, the Bank of England allows lenders to model their own losses and then compares the models with each other and with its own version. If differences are reasonable they are permitted. This helps “model monoculture” in which every bank adapts its holdings in order to pass the tests and they all end up holding assets the government model favours. A diversity of bank strategies is far preferable given that risks are hard to predict.
The living wills process requires banks with more than $50bn in assets to hold minimum amounts of “safe” assets; currently this stockpile totals more than $4tn in government debt, significantly more than called for by the global rules. Individual bank requirements are secret and confidential, so it is impossible to tell whether some banks are being hit with excessive requirements. If unchecked, this allows completely arbitrary government action.
There is empirical evidence that higher bank capital requirements cut lending and economic growth. A recent Fed paper concludes that a 1 percentage point increase in capital ratios could reduce the level of long-run GDP growth by 7.4 basis points. Without knowing how the secret models affect capital requirements, it is impossible to say whether and by how much they restrict our economy. High liquidity standards have an even more direct impact on lending. For every “safe” asset, for example a US Treasury, that a bank holds, there is one less loan it can make.
A more transparent process for setting liquidity and capital standards is critical. The Treasury Department has proposed opening up the stress tests and living wills process to public comment. The Fed, FDIC and other bank regulators should act on those recommendations. Meaningful banking reform depends on strong leadership at the helm of each of the bank regulators. They will no doubt be subject to criticism but most worthwhile change usually is.
Mr. Scott is Nomura professor of international financial systems at Harvard Law School and director of the Committee on Capital Markets Regulation.