America’s Brewing Debt Crisis: What Dodd-Frank Didn’t Fix

Robert Litan. Foreign Affairs. Volume 95, Issue 5. Sep/Oct 2016.

Almost as soon as the financial crisis struck in late 2007, policymakers began working to prevent another one. The roots of the crisis, they contended, lay in reckless lending and excess debt. Banks had made massive loans to “subprime” borrowers, who had little ability to repay them, and the banks funded these investments with borrowed money. When the U.S. housing bubble burst, millions of Americans defaulted on their mortgages, and the overleveraged banks collapsed. The government had to bail them out, and U.S. taxpayers picked up the bill.

In July 2010, U.S. President Barack Obama signed the Dodd-Frank Wall Street Reform and Consumer Protection Act. Reformers hoped that the act-known as Dodd-Frank, after its Democratic co-sponsors, Senator Chris Dodd of Connecticut and Congressman Barney Frank of Massachusetts-would make another financial crisis less likely. And to some extent, Dodd-Frank has succeeded. During the crisis, too many financial institutions lacked enough capital to withstand losses on their loans. But now, thanks in part to the act, banks have to fund themselves with more capital and less debt, which equips them to absorb more losses in a future downturn. And banks have largely stopped making subprime loans, since Dodd-Frank rules require those who give loans and securitize them to bear some losses in the event they sour.

Nevertheless, during the 2016 presidential campaign, Dodd-Frank has come under attack from both sides of the aisle. In the Democratic primary, Senator Bernie Sanders of Vermont argued that its reforms did not go far enough. He called for the government to break up the largest U.S. banks and reinstate Glass-Steagall, the 1933 act that separated commercial from investment banking, until Congress repealed it in 1999. Republicans, meanwhile, including the presidential candidate Donald Trump, believe Dodd-Frank went too far, and Republican legislators have sought to repeal it at every opportunity, arguing that its regulations are crippling U.S. banks and stifling growth.

Both criticisms distract from the real problem with the act, which is that it left some key problems unaddressed. In dealing with reckless lending and excess leverage, it misses one of the most important causes of the crisis: “runnable liabilities,” or short-term debt that the government does not insure. The U.S. financial sector holds trillions of dollars of such debt, including uninsured bank deposits and the short-term liabilities of other financial institutions, such as overnight loans. What makes this kind of debt so dangerous is that during a crisis, short-term lenders, unlike long-term ones, can demand their money back immediately, leaving borrowers unable to pay all their creditors quickly. The financial sector stops lending money, credit dries up for consumers and businesses, and the economy grinds to a halt. This is what happened in 2007 and 2008, when massive runs on short-term debt spread panic throughout the financial sector and helped trigger the Great Recession.

Although short-term debt poses one of the greatest threats to the financial stability of the United States, Dodd-Frank has done little to mitigate it. Fortunately, several experts have proposed ambitious ways of dealing with the problem, including expanding federal insurance of bank deposits, allowing the Federal Reserve to lend money to more firms in the case of a panic, and banning unregulated financial institutions from issuing runnable liabilities. These are good ideas, and if Congress passed any of them into law, the odds of a future financial crisis would be significantly lowered.

After Dodd-Frank

The Dodd-Frank Act set out to solve one of the central problems with the U.S. financial system: that some banks, such as Citigroup and J.P. Morgan, were “too big to fail.” When those banks were faced with collapse, the government had to come to the rescue, or else risk allowing the whole economy to go down with them.

Dodd-Frank was supposed to solve this problem in two ways. First, the act raised the minimum capital requirements for all banks and imposed especially strict requirements for those with $50 billion or more in assets—the “systemically important financial institutions” (SIFIS). Dodd-Frank also requires banks to hold more liquid assets, money they can use to pay back depositors during a sudden panic. And the act gives a new body, the Financial Stability Oversight Council (FSOC), the authority to designate certain large financial institutions as SIFIS, which the Fed can then regulate more stringently.

Second, Dodd-Frank gave Washington new powers to preemptively shut down large, complex banks and other financial institutions, making bailouts unnecessary. The act gave the Federal Deposit Insurance Corporation (FDIC) powers to close SIFIS without taxpayers’ bearing the cost; instead, shareholders, creditors, and managers would lose out without causing wider damage to the financial system. Banks must now prepare “living wills,” plans that detail how regulators can shut them down in case of emergency.

The act also took aim at financial derivatives, which many politicians blamed for the crisis. Derivatives are contracts whose payout depends on the performance of another asset, such as oil or a foreign currency. One particular type of derivative, credit default swaps, which allow buyers to insure against the failure of a company to pay back its loans, has been especially controversial. The insurance giant AIG issued far too many of these contracts without insisting on enough collateral, and when the mortgage market collapsed, AIG collapsed as well, prompting a massive rescue by the Federal Reserve. What’s more, before the crisis, the market for derivatives was opaque: instead of trading derivatives on formal, transparent exchanges, individual firms bought and sold them privately with little oversight. Dodd-Frank requires many derivatives to be settled through central clearing-houses, where regulators can more easily monitor them.

Born To Run

These measures to rein in subprime loans and excessive leverage have no doubt strengthened the U.S. financial system. But the problem is that these factors, although they contributed to the Great Recession, did not lie at the heart of the financial panic; runs on short-term debt did.

In 1933, after roughly 9,000 banks collapsed as savers rushed to withdraw their money during the Great Depression, Congress created the FDIC to insure deposits up to a certain amount (initially $2,500, but by 2007, the number had reached $100,000). The move helped prevent bank runs, since people no longer worried that they might lose all their savings if their bank collapsed.

Then, in 2008, as the financial crisis spread panic throughout the economy, Congress raised the amount of the deposits that the FDIC would insure from $100,000 to $250,000, covering roughly half of the $12 trillion that the country now holds in bank deposits. Bank runs have thus become even less likely, although not impossible.

But for other financial institutions, for which the government has not stepped in to provide insurance, the risk of runs remains high. “Shadow banks” are financial institutions that are similar to banks, in that they also issue very short-term liabilities, but are not regulated as such. These include investment banks, moneymarket mutual funds (a low-risk, lowyield investment option), and various financial firms. These shadow banks and other issuers of short-term debt collectively account for roughly $16 trillion in short-term debt (dwarfing the $6 trillion of insured bank deposits), and no equivalent of the FDIC exists to prevent the holders of these instruments from running on the institutions that carry this debt. Meanwhile, even in banks, deposits above $250,000 are still at risk of a run, as are Eurodollar deposits (dollar-denominated accounts in foreign banks), which the FDIC does not protect.

In the years leading up to the financial crisis, shadow banks relied increasingly on runnable debt. Until the mid-1990s, such debt was equivalent to around 40 percent of U.S. GDP, but by 2008, the figure had reached 80 percent. This debt carried lower interest rates than longer-term debt and was thus a cheaper source of funding. It took a number of forms, including commercial paper, a kind of short-term debt issued by corporations; money-market mutual funds; and repurchase agreements, or repos, a type of short-term loan that allows a borrower to sell a bond and promise to buy it back within a few days. In a crisis, lenders could run on all these financial instruments.

In 2008, they did. The investment banks Bear Stearns and Lehman Brothers experienced runs on their short-term debt. Investors also began to flee money-market mutual funds, which started to collapse; the Treasury Department had to step in and issue an unprecedented blanket guarantee of all of them. And federal regulators, afraid that there would be a run on bank deposits above $250,000, merged failing banks with stronger ones and temporarily guaranteed all accounts. All of this happened in just six months, and mostly in September 2008.

Problem Solved?

Today, runnable debt remains a major problem. Dodd-Frank focused on reforming the banks, but shadow banks remain out of the regulators’ reach. The act did create the FSOC to eliminate debt bubbles before they burst. But the FSOC will not spot every emerging bubble. After all, almost every economist missed the signs of the last financial crisis.

The FSOC also barely mitigates the risk of a run on uninsured deposits. Its power to designate certain banks as “systemically important” may indirectly address the problem, by signaling that the government would be more likely to bail out these institutions than others. So might a Dodd-Frank provision that enables the FDIC to borrow from the Treasury to pay short-term creditors who might otherwise pull their money out of a failing financial institution. But a fair amount of uncertainty remains; it isn’t clear how willing a future government will be to take such action, given the backlash against the bailouts of “too big to fail” banks and the forced mergers during the last crisis.

Compelling large banks to hold more capital reduces their risk of failure, but as the last financial crisis demonstrated, during a widespread panic, investors and lenders lose all faith in the values banks have assigned to their assets, and many mistrust banks that claim to have enough capital. Depositors with more than $250,000 may still run on their bank at the first hint of trouble.

Regulators have also forced banks and other financial institutions to hold more liquid assets, which they can use to pay back depositors who want their money back immediately. But even this measure may not do enough to meet the demands of creditors in a full-scale panic, since no bank can have all its assets in liquid form and still turn a profit.

As inadequate as the existing measures are, however, the popular ideas for bolder reform would do little more to reduce the risk of a run by uninsured depositors or short-term debt holders. Consider Sanders’ proposal to break up the “too big to fail” banks. Turning one $2-trillionasset bank into four or five smaller banks would not make uninsured depositors any less likely to withdraw their money if one of the smaller banks faced difficulties, since their large deposits would still be uninsured. Such depositors would rationally conclude that if one of the smaller banks was in trouble, theirs might also be, potentially triggering a run. Nor would reinstating Glass-Steagall prevent runs if panic caught on, because separating commercial from investment banking would do nothing to stop uninsured depositors from running on commercial banks or repo lenders from refusing to roll over their loans to the investment banks.

Republican proposals, meanwhile, could exacerbate the risks posed by short-term debt. Conservative academics at Stanford University’s Hoover Institution, for example, have suggested designating a special district court to expedite bankruptcy cases. But making it easier for a financial firm to declare bankruptcy could make it more likely that lenders would lose their money if the firm collapsed, which might make them quicker to pull their money out. The special bankruptcy court could lessen this risk if it asked the Federal Reserve to act as the lender of last resort for short-term creditors to prevent them from panicking, but this would offer little improvement over the current system.

Fixing Finance

Yet the problem of runnable debt has solutions. One idea comes from Morgan Ricks, a former official in the Obama administration’s Treasury Department. In his new book, The Money Problem, Ricks argues that the government should drop the pretense that its insurance extends only to $250,000 worth of deposits. In fact, the government implicitly insures more than that, but only if depositors place their money in the big banks that governments have a strong incentive to protect for the sake of financial stability. Indeed, since the crisis, Americans have concentrated their assets in the largest banks. If the FDIC formally abolished the insurance ceiling and thus promised to insure all accounts, regardless of size, it would eliminate the risk of runs on banks. It would also put smaller banks on a level playing field with bigger ones, since people would no longer eschew the former for fear that the government would bail out only the latter. The move would make the financial sector less concentrated, which could introduce more competition.

Critics of this proposal argue that it would create a moral hazard. If the government insured everyone’s deposits, the logic goes, banks might feel emboldened to take greater risks, for example, by lending to riskier borrowers at higher interest rates. But this problem already exists today: customers with large accounts have moved their deposits to the biggest banks, gambling on future government protection in the event of a crisis, protection that the banks themselves are gambling on. The best way to limit this moral hazard is through the stiff capital requirements that regulators have imposed on large banks, measures that provide cushions for the banks in case of losses from bad decisions. But regulators need to enforce these standards more effectively than they have in the past.

Ricks has an even more controversial solution for the risks that shadow banks pose. He proposes banning any financial institution that isn’t a bank from issuing runnable liabilities-in other words, he calls for the end of shadow banking. Under his plan, the government would essentially outlaw money-market mutual funds, repos, shortterm commercial debt, and Eurodollar deposits. To cushion the blow to financial institutions, which would find it costlier to raise money, Ricks also suggests eliminating Dodd-Frank-something many banks have been advocating since Congress first passed the act. To put it mildly, this would be a big deal.

Hal Scott, a professor of international finance at Harvard Law School, has put forward a more traditional approach to mitigating the risks of short-term debt. In his new book, Connectedness and Contagion, he argues that the government should expand the Fed’s authority to step in as the lender of last resort. By setting clear ground rules for emergency lending in advance, rather than acting in an ad hoc fashion in the heat of a crisis, the Fed would remove uncertainty about when institutions are eligible to receive liquidity support. This, Scott claims, would leave creditors with no reason to run on the debt. It would put an end to the concept of “too big to fail” and give regulators time to reorganize and close failing financial firms, wiping out shareholders’ equity in the process but preserving the stability of the system as a whole.

As effective as Ricks’ and Scott’s proposals may prove, however, they would face major political problems. Ricks’ plan to outlaw shadow banking would surely invite fierce opposition from the firms in question. And his proposal to repeal Dodd-Frank-which he envisions as a bipartisan compromise in which Democrats agree to get rid of the legislation they support in exchange for measures that eliminate the possibility of future financial panics-seems unlikely to get very far in Congress. Scott’s idea to expand the Fed’s ability to lend to troubled firms runs counter to a rule the Fed adopted in 2016 to limit lending of last resort to specific institutions. It’s hard to imagine such a quick reversal of policy happening at a time when much of Congress is openly hostile toward the Fed.

Jeremy Stein, a former member of the Board of Governors of the Federal Reserve, and Robin Greenwood and Samuel Hanson, both professors at Harvard Business School, have offered a more practical, but also more limited, plan. They argue that the government should increase the supply of public, or government, short-term debt to accommodate investors’ demand for safe financial instruments and so that investors do not have to rely on privately issued short-term debt. They suggest that the Treasury could unilaterally and gradually replace longer-term Treasuries with these shorter-term government obligations, which are immune from runs because investors view them as safe assets.

Their solution is not as far-reaching as either Ricks’ or Scott’s, but at least it does not require congressional approval, since the tactic would amount merely to a change in Treasury Department policy. The problem, however, is that issuing short-term government debt would expose Washington to swings in short-term interest rates, introducing more volatility into the federal budget. In the end, then, the strategy might prove more expensive than other approaches. Long-term interest rates have fallen to their lowest levels since the early 1950s, and the government could save money by issuing longerterm, rather than shorter-term, debt.

But perhaps the boldest proposal comes from Mervyn King, who was governor of the Bank of England during the financial crisis. In his book The End of Alchemy, King argues for even higher capital and liquidity standards for both traditional banks and shadow banks. King would phase in these tougher requirements over a period as long as 20 years. And like Scott, King wants clearer policies from central banks on when they will act as a lender of last resort. He argues that central banks should lend to almost anyone with sufficient collateral, and not just to banks-a significant expansion of central banks’ lender of-last-resort role, but one that would help provide liquidity when it is most needed.

King’s idea is clear and logical. Yet it is likely to face as much hostility as Ricks’-if not more, since King is tougher on banks. And Congress would likely have the same reaction to King’s plan as it would to Scott’s, since it also expands the role of central banks.

One Step at a Time

Most of the reforms that politicians have advocated have neglected the problem of runnable debt, and the academics’ proposals are currently politically impractical. But there is a way forward: regulators should focus on more moderate reforms that reduce the role of short-term and other uninsured debt in the financial system. Some of these reforms would be possible under existing law, while others would probably require new legislation.

The Federal Reserve has already suggested one useful reform: mandating that issuers of repos back those instruments with extra collateral. The requirement should discourage investment banks from using repos for funding and, in a crisis, reassure those lending through repos that their loans are sound and will be repaid.

Regulators could also discourage investment banks from issuing short-term debt by requiring them to hold capital in an amount that increases in proportion to their short-term liabilities (an idea similar to the “risk fee” that the Democratic presidential candidate, Hillary Clinton, has proposed for banks and SIFIS that rely on short-term debt). U.S. regulators could take this step on their own now, although it would make the global banking system safer if they could persuade regulators in other developed economies to adopt a similar system.

Meanwhile, Congress could downsize money-market funds, stopping short of outlawing them altogether. To do so, it should eliminate the current $250,000 ceiling on insured bank deposits. The change would make money-market funds less attractive to large investors, since they could invest in banks with full protection without having to take on extra risk.

Among all the potential causes of the next financial crisis, the massive amount of runnable debt ranks as the most probable. Yet so far, policymakers have overlooked this problem, perhaps believing that all will be fine if they simply cut big banks down to size or promise never to protect any of their large depositors. This is wishful thinking. The financial system will never be immune from crises, but solutions exist that may go a long way toward reducing the risk.