Clearinghouses as Liquidity Partitioning

The Dodd-Frank Act established that certain swap contracts which previously were traded bilaterally (directly between buyers and sellers) must be traded through clearinghouses instead. Critics of this clearing mandate have mounted two main objections: a clearinghouse shifts risk instead of reducing it; and a clearinghouse could fail, requiring a bailout. In my article Clearinghouses as Liquidity Partitioning, recently published in the Cornell Law Review, I counter both objections by showing that clearinghouses engage in a socially valuable function that I term liquidity partitioning. Liquidity partitioning means that when one of its member firms becomes bankrupt, a clearinghouse keeps a portion of the firm’s most liquid assets, and a matching portion of its short‑term debt, out of the bankruptcy estate. The clearinghouse then applies the first toward immediate repayment of the second. Economic value is created because the surviving clearinghouse members are paid much more quickly than they would be in a bankruptcy proceeding. Meanwhile, the bankrupt member’s outside creditors are not paid any less quickly: they still are paid at the end of the bankruptcy proceeding, which the clearinghouse does nothing to prolong. These rapid cash payouts for clearinghouse members reduce illiquidity and uncertainty in the financial sector, the main causes of contagion in a crisis. And because the clearinghouse holds only liquid assets, it avoids the maturity mismatch between short‑term liabilities and long‑term assets that characterizes the balance sheets of many financial institutions. A clearinghouse therefore is much less likely than its members to fail during a crisis.

A clearinghouse achieves liquidity partitioning by engaging in netting. Thus, when a member fails, the clearinghouse uses short‑term debts owed to the member to immediately repay short‑term debts owed by the member. In this way, cash is intercepted on its way toward the bankruptcy estate and redirected toward other financial firms, who may be suffering their own liquidity shortages. The clearinghouse thereby shifts cash from lower-value to higher-value uses, decreasing liquidity pressure on the financial sector and thus the need during a crisis for a taxpayer-funded bailout.

Critics who doubt that the clearing mandate will reduce systemic risk emphasize how a clearinghouse’s impact on creditor recoveries is zero-sum: instead of reducing total losses to creditors when an insolvent firm fails, a clearinghouse merely shifts losses from some creditors to others. Although it is true that a clearinghouse’s impact on the simple sum of creditor payouts is zero-sum, its impact on payout speed is not. By engaging in liquidity partitioning, the clearinghouse accelerates cash payouts to some creditors without slowing down payouts to others, thereby decreasing total illiquidity risk in the financial sector.

Critics also argue that a clearinghouse could itself fail during a crisis, requiring a government bailout. But because a clearinghouse engages in liquidity partitioning, its balance sheet is symmetrical in terms of duration: it accepts responsibility for its members’ short-term debts only to the extent that the members can provide it with short-term, liquid assets. By contrast, many financial institutions—including many clearinghouse members—engage in maturity transformation: they raise funds through short-term loans (including demand deposits) and invest the funds in long-term, illiquid assets. As a consequence, these institutions tend to have low cash ratios—that is, small cash holdings relative to their short-term debts. During a crisis, the failure of one such institution can cause creditors to fear that others are exposed to similar risks and therefore are about to fail as well. Creditors may then “run” on financial institutions generally, withdrawing cash and refusing to renew short-term loans. Because such institutions typically have more short-term debt than cash on hand, they may have to seek bankruptcy protection (or, if banks, enter receivership) even though their underlying business activities remain profitable.

This sort of liquidity shortage is precisely the hazard that a clearinghouse protects against: it intercepts cash headed toward a bankruptcy estate and uses it to make immediate payouts to financial institutions. Indeed, faster payouts reduce uncertainty about the clearinghouse members’ financial stability, which decreases the likelihood that their creditors will run in the first place. At the same time, liquidity partitioning shields the clearinghouse from the risks associated with its members’ long-term, illiquid assets, making the clearinghouse much less likely than its members to become either illiquid or insolvent. Critics who argue that the clearing mandate will simply shift bailouts from swaps dealers to clearinghouses in the next crisis have not recognized that clearinghouses are less likely to require a bailout because they avoid the maturity mismatch between assets and liabilities that characterizes many other financial institutions.

Whether the clearing mandate will in fact realize the systemic benefits of liquidity partitioning will depend on how it is implemented by rulemakers. These systemic benefits derive from netting, which is what keeps cash out of a bankruptcy estate and in circulation in the financial sector. Clearinghouses make it possible for parties to engage in a particular type of netting called multiparty netting, which occurs when the bankrupt firm has a claim against one counterparty and a debt with another, and the former is used to pay the latter. A second type, called multicategory netting, occurs when a bankrupt firm has debits and credits across different contract types. Parties do not need a clearinghouse to engage in multicategory netting, and indeed a clearinghouse can undermine multicategory netting opportunities if it fails to accept a variety of contract types for clearing. For the mandate to be effective in reducing systemic risk, it must be structured so that it increases opportunities for multiparty netting more than it reduces opportunities for multicategory netting.

The full article is available for download from SSRN here.