MMFs, deposit insurance, and regulation in the age of shadow bank runs | FT Alphaville

MMFs, deposit insurance, and regulation in the age of shadow bank runs

Deposit insurance on non-interest bearing accounts — it was in October 2008 that the FDIC started it, through the Transaction Account Guarantee, or TAG.

Until we looked a bit more closely, we hadn’t guessed that the issue could offer much insight into the complexities of shadow banking regulation.

But that was until we understood its importance to money market fund investors in the years since the crisis…

First a step back

US money market funds, remember, have had their margins squeezed by the low-rate environment and by having to comply with an amended Rule 2a-7, which limits investments in lower-tier commercial paper and requires shorter-maturity and more-liquid holdings. And they still face an uncertain regulatory future while the SEC thinks about tougher capital standards and forcing a floating NAV.

Making life marginally more difficult has been quantitative easing and, more generally, elevated demand for low-risk assets. For all of QE’s other benefits, its removal of a large share of US Treasuries and Agencies from the market and onto the Fed’s balance sheet has meant a smaller total of such assets for MMFs to buy, correspondingly lower yields on the ones they do buy, and less appropriately “safe” collateral for their potential repo counterparties. (Offsetting this trend somewhat has been a few years of higher fiscal deficits, requiring the issuance of more government securities.)

So money market funds remain in a tense situation. Their total assets under management remain well below their peak, more on which in a second, and as the gang at Fitch tells us, their struggle for incremental yield conflicts with their ongoing risk aversion. We say ongoing, but really it’s part of their business model.

Perhaps nothing exemplifies this difficulty more than their recent return to European financial institutions. Having recognised the renewed, LTRO-boosted stability of French and German banks, MMFs finally began lending to them again in February while taking some money out of the safe havens to which they’d flocked…

… but according to Credit Suisse, they’ve returned to these banks at much shorter maturities than before they starting exiting last summer:

Seems like fickle money we’re talking about.

In the case of the largest money market funds, some have resorted to entering into reverse repos on risker assets in a reach for higher yield; the NY Fed argues that this doesn’t apply more broadly to the whole tri-party repo market, while still acknowledging that even the limited trend among the bigger funds should be carefully watched.

So, now a quick recap of flows, the FDIC, and Dodd-Frank:

Now, money market fund assets stood somewhere north of $3.7 trillion at the start of 2009. By the end of last year they had slightly less than $2.7 trillion, according to the Fed flow of funds report, though their balances were only down about $70bn since December 2010.

All after FDIC began TAG. (Deposit insurance on other deposit accounts was, and is, capped at $250,000.)

The point then was to prevent a run on bank deposits. But the program was also meant to expire at the end of 2010, and instead Dodd-Frank extended it through the end of this year. The goal this time was to give smaller banks a chance to capture some of these deposits from the perceived “safer” Too-Big-To-Fail institutions.

Instead, three things happened:

1. Cash in non-interest bearing accounts, which has nearly doubled since the crisis, especially accelerated after the extention in Dodd-Frank and now stands at $1.4 trillion (chart via Barclays Capital):

2. The deposits that accumulated between the crisis and Dodd-Frank’s passing mostly stayed at the big banks. Ie Dodd-Frank’s stated goal wasn’t met.

3. Assets that left MMFs and would have returned by now if the insurance had expired… haven’t. But it’s tough to pinpoint the exact amount:

It is difficult to say for certain how much the FDIC’s provision of unlimited deposit insurance has pulled out of money funds. After all, some portion of the aggregate amount of balances over the normal insurance cap would be on deposit at the “too-big-to-fail” money center banks regardless of the presence of unlimited deposit insurance given their perceived government support. As a result only a portion of the $1.4trn in non-interest bearing checking accounts over the $250,000 insurance cap would otherwise be held at money funds.

Based on the growth in these balances since the Dodd-Frank extension went into affect, we reckon that between $500bn and $600bn of this amount would be, in the absence of unlimited coverage, held at money market funds.

That’s an awful lot of money, and the logical follow-up question is just how much we can expect to return to money market funds if the insurance isn’t extended. That too is difficult to estimate. Back to Barcap:

Assuming that the 19 largest banks, which hold a preponderance of the insured deposits, decide not to participate in 2013, it is not entirely clear that the $500-600bn in balances that left money funds since the end of 2010 would return.

First, officials at the SEC are discussing some significant changes to the structure of money funds that include floating NAVs or establishing capital buffers with redemption gates. Faced with either option and given the traditional money fund investor community’s vehement opposition to floating NAVs and anything less than 100% daily liquidity, it is difficult to see how eagerly they might return to prime and government-only money funds.

Second, as noted above, there is a fair amount of uninsured balances held at the largest money fund banks that may assume the balances have de facto unlimited deposit insurance by being at “too-big-to-fail institutions.

As a result, while some of the $500-$600bn will likely return to money funds, we suspect a sizeable portion could remain at the largest money center banks. The proportion that stays will be higher if, the SEC moves money funds to floating NAVs (which we consider to be unlikely) or if, the SEC imposes daily redemption limits (far, more likely).

What’s the point?

What the trends of the last few years show is that the minuscule yield certain investors would get by putting cash in money market funds, was deemed less valuable than the government guarantee that backs non-interest bearing accounts.

Absolute safety over meagre yield.

And the investors who made the switch from MMFs to these insured accounts are, we suspect, similar those who represent the constant demand for safe assets that Gary Gorton’s recent preliminary paper described.

In other words, these are investors looking for short-term savings vehicles only if those vehicles are either buying safe assets directly or lending against safe assets as collateral. This is money that would rather be in the shadow banking system.

And from a certain point of view, it still is.

Deposit insurance wasn’t just a lender-of-last-resort protection for traditional deposits. An imperfect but useful way to think about this insurance is that it also represents an increase in the collateral available for shadow banking money. And some of that money decided that the new collateral on offer was better than the collateral then (and now) available to money market funds.

Money market funds did receive some explicit help during the crisis after the Reserve Primary Fund broke the buck. But the deposit insurance turned out to be an implicit backstop for the shadow banking system, just dressed in the guise of explicit help for the traditional banking system. Since this post is in need of a namedrop-bomb, this ain’t your classic Diamond & Dybvig, though the principle is the same.

Despite their shrunken assets, money market funds are still competing for a smaller pool of safe assets and face a foggy future, as noted at the top. (By the way, we’ll have more on this in a future post.) So rather than park their money with MMFs, some investors are now guaranteed to get back their money back 100 cents on the dollar with the safest backing on earth as collateral. And they didn’t even have to pay for it.

Which brings us to the question of whether deposit insurance should be extended or not. And if so, for how long?

Sorry, but there’s no easy answer to this question.

In an ideal universe, our preferred solution would probably be for the insurance to remain in place until the economy has recovered sufficiently to lift the values of non-public sector collateral that’s appropriate and available to the shadow banking system.

But after that, things get trickier. This deposit insurance functioned as a kind of bailout, with the government assuming the risk for MMF investors to at least have a place to park their money where its value will be guaranteed.

Unless the government will happily stand ready to do the same in the future event of another run on MMFs — an unpopular idea, surely, and which also brings with it the obvious moral hazard dilemma — then some other regulations will be necessary.

One idea is for fee-based insurance, where the banks pay the government for the insurance and then pass along the cost to investors. This seems like a reasonable idea to us, with the caveats that the pricing will be tricky to get right and that many banks simply won’t want to participate. The latter is an especially important worry during a time when the largest banks are still implicitly Too-Big-To-Fail.

In other words, such a policy might have to be forced on the banks.

Another option, of course, is to bring added transparency to repo and the other short-term lending markets in which MMFs and other similar players operate. This is fodder for a much longer discussion, but added transparency is also difficult to get right, as it too risks pushing shadow banking money into even murkier corners.

So let the insurance expire too early and you risk putting further strains on repo and short-term lending markets if the money retreats back to the MMFs. Extend it without new regulations and you’re simply continuing a bailout that’s cost the bailed-out absolutely nothing.  Extend it with regulations and you must face some thorny questions about just what those regulations should look like. And all this has to be done by a Congress that’s, shall we say, not always moved to the optimal solution.

None of these options is appealing, really. But then, the shadow banking system existed unregulated for the better part of three decades — nobody said getting it right would be easy.