HKMex is cause for concern, not congratulation
The regulatory enthusiasm for central counterparty clearing houses (CCPs) is unbounded. Having encouraged their spread from commodities markets to futures and option markets to cash equity markets to repo markets to swaps markets, regulators are now looking at injecting them into stock loan markets as well. Perhaps the withdrawal this week by the Hong Kong Securities and Future Commission (SFC) of the licence to operate an ATS of the Hong Kong Mercantile Exchange (HKMEx) will prompt an overdue pause for thought.
What took place at HKMex was a live experiment in winding down open positions at a failed exchange. This is an entirely different proposition from winding up a failed clearing broker, as happened in the aftermath of the failure of Lehman Brothers in September 2008. In that case, the successful unwinding by LCH.Clearnet of a default with a notional value of $10 trillion became the main inspiration behind the Group of Twenty (G20) decision to make the clearing of OTC derivatives mandatory.
As it happens, LCH.Clearnet was also clearing house to HKMex, and the closing of open positions was completed in an equally orderly fashion. Collateral is now being returned to clearing members (HKMex itself held no client assets). The fact that the winding up process has gone so smoothly can be regarded as a vindication of the increased safety afforded by shifting ever more responsibilities on to market infrastructures, and it is already being interpreted by infrastructural enthusiasts in exactly that fashion.
But the necessity of actually withdrawing from an infrastructural entity its right to operate ought to be the occasion not for another bout of self-congratulation, but alarm. The greater the responsibilities the market infrastructure is asked to bear, the greater the risk of buckling under the strain, and the greater the scale of the catastrophe when it does finally occur. CCPs in particular - several of which have failed in the past - are being asked to bear enormous liabilities on the basis of nugatory amounts of equity capital.
True, maybe CCPs do not need that much capital. After all, they run completely matched books: what is owed on one side is matched by what is owned on the other. But they are still vulnerable to a major broker default and, worryingly, it is the clearing brokers that dominate their affairs. The best measure of this marsupial relationship is how the default fund contributions paid by clearing brokers remain low relative to the initial margin payments contributed by their buy-side clients.
It also explains why, when they are asked by regulators to find additional capital, CCPs are apt to reply that it is only their capital fragility which keeps them safe. Too much capital, they argue, would only encourage their clearing members to take greater risks. These entities have, to put it plainly, no money. Which means that all that stands between buy-side users of a CCP and material losses in an event of a broker default is the ability of a VaR-based, CCP-designed algorithm to ensure that margin coverage is always adequate. That is not a reassuring prospect. It is rightly said that, far from centralising and mitigating counterparty risk, CCPs merely re-package it as liquidity risk in the markets for collateral.