Yesterday, FSOC wisely abandoned firm-specific SIFI designations for asset managers and adopted the activity-based framework FedFin forecast this fall. That I advocated it in a paper prepared for the Federal Reserve is, I trust, not why we forecast it – I try very hard to keep my views from affecting our forecast. I’ll just chalk this up to great minds…. With this happy precedent, I’ll turn now to what I think FSOC should do to address asset-management systemic risk.
To date, many large asset-management companies have resolutely stood by a wholly implausible claim about their own resolvability: that is, that all they do is handle other people’s money and, thus, an asset manager can fail without damaging either investors or the broader market. A quick look at the largest asset managers and their funds shows clearly why this just isn’t the case in practice.
It’s nominally true that all the assets in a fund like Pimco’s huge Total Return one are not Pimco’s. But, to boost yield, Pimco plays fast and loose with these funds. As the New York Times laid out yesterday, funds use all sorts of structuring to boost return. Between both these capital-market ploys and their direct holdings, they control far bigger shares of vulnerable sovereign markets than all but the biggest “national champion” banks.
Asset managers are also direct participants in central counterparties (CCPs) and the array of financial-market utilities in which risk is increasingly concentrated. CCPs have about zero capital ahead of those clearing through their systems to ensure resilience under market or operational stress. Instead, risk cascades through a “waterfall” in which clearing banks and other major counterparties take a series of increasingly hard falls. End-users – that is, retail investors, endowments, and pension funds – may ultimately share or even bear these costs. For example, transactions can be subject to “tear-up” of the payments that a counterparty thought it was due. Banks are required to hold increasingly large amounts of capital to protect both themselves and end-users from CCP risk – asset managers, not so much.
Most of these risks may well prove to be liquidity – not solvency – ones both to asset managers and their investors. But, in sharp contrast to big banks, asset managers have no central-bank backstop – or, at least no official one. The FRB has increasingly been talking up a possible role as “market-maker of last resort.” The more it talks, the more the market thinks it will walk down this aisle in extremis. However, to do so or even to get the market now to expect the Fed to stand behind asset managers is to give the industry a terrific taxpayer guarantee for which none now pays. One can argue about whether big banks pay enough for central-bank liquidity – and, whatever the cost, it’s going up because of all the new rules that bar big banks from counting on the Fed. However, to the extent banks go to the FRB, they absorb a lot of up-front regulatory cost and a lot of collateral requirements for any actual funding. No such rules apply to asset managers and it’s far from clear how they could post collateral since most of their assets — once they find them in the re-hypothecation or hedging chain — aren’t their own.
Asset managers are also subject to significant amounts of operational risk. This is not just because of their big CCP role and, thus, exposure. There’s also zero reason to assume that they are any more immune to cyber-attack than big banks and, in some ways, the risk here is greater again because the funds at risk mostly aren’t theirs. When a big bank is cyber-assaulted, it uses capital and reserves to buffer the impact to itself and to customers. An asset manager would almost surely simply transmit the cost to investors and, thus, pose potential systemic risk.
Even without coming under cyber fire, asset managers are – like banks – at risk from rogue traders, systems failures, and natural disasters. Banks have to hold operational risk-based capital against all these hazards, as well as engage in very extensive and, now, binding recontingent-recovery plans. Nothing like this applies to any asset manager.
In short, asset managers may be especially vulnerable to operational risk and they aren’t resolvable as currently constituted or regulated. Solve for this, and much systemic risk is similarly solved. Not all systemic risk would then disappear – like big banks, very big asset managers may still be able to externalize some of their risks – but most systemic risk is meaningfully addressed if asset managers – like big banks – are really forced to have fun only with their own dollars and reputations, not everyone else’s.