The key insight of the latest Gorton-Lewellen-Metrick paper, charted:
The findings are preliminary, but the authors calculate that the safe asset share — the percentage of safe assets to total assets in the US economy — has been roughly the same since 1952, at about 33 per cent.
This paper hasn’t received much attention since it was presented at an AEA conference two weeks ago, but if its findings hold then we think it might be a bigger deal than has been recognised.
The stability of demand for safe assets has held during a time in which the assets of the financial sector as a percentage of all assets in the economy have climbed from 25 to 40 per cent, with most of the growth in total assets being financed through debt rather than equity.
Less surprising to those who have followed Gorton’s work is that the composition of the safe assets created by the financial sector has changed dramatically in the last three decades with the rise of the shadow banking system — though Gorton & co take a crack at measuring the changes in distribution with some precision:
(Note that “money-like debt” refers to “commercial paper, net repurchase agreements, federal funds, money market mutual fund assets, interbank transactions, broker-dealer payables, and broker-dealer security credits”.)
A (very) brief primer
To understand the importance of the finding, it helps to know some background on what Gorton & Co mean by “safe assets”. Those already familiar with Gorton’s previous work can skip to the next section.
Since the relevant topic is how to prevent a run in the shadow banking system, we’re primarily talking about debt here — specifically, the debt eligible to be used as collateral in repo and short-term secured lending markets.
Gorton defines debt as safe when it is information-insensitive, which means that the reliability of its being paid back is unquestionable and should remain through fluctuations in financial market conditions and the economy.
Investors, banks, hedge funds and other private market players who use the repo markets as a short-term savings vehicle therefore have no incentive to perform any diligence on the assets and find new information about them. Their resilience is assumed.
Gorton’s explanation of the US crisis that started in 2007 is that information-insensitive debt switched to information-sensitive when the Libor-OIS spread started spiking in 2007 and made clear the deteriorating conditions of interbank markets. And because not enough US Treasuries and other kinds of government-backed debt (agencies, munis) existed to meet the demand for this kind of collateral, the debt causing the problems was the senior tranches of asset-backed securities used as collateral in repo markets.
Why wasn’t the switch limited to those securities that were exposed to rapidly disintegrating fundamentals in the underlying assets?
Gorton and Metrick have previously argued that the panic wasn’t caused directly by the revelation that subprime-related ABS values were plummeting; this had already happened earlier than 2007. The problem was that the lack of transparency in repo markets meant that investors had no way of distinguishing between repo borrowers whose collateral was subprime-related and those whose collateral was relatively safer. So they started raising haircuts, from zero in most cases, indiscriminately across all repo counterparties. The run was on — and so was the credit crunch.
That’s a simplistic overview of a very complicated issue — for a better catch-up click here — but Gorton’s primary theme is that shadow banking does qualify as a real and necessary kind of banking. The run on shadow banking that started in 2007 therefore had similarities with the bank runs that were common before the onset of the Quiet Period in banking that started in 1934.
It is intuitive to see how the preliminary findings of this new paper complement Gorton’s previous writings.
If the demand for safe assets remains constant, and its presence is a necessary part of the credit channel provided by financial intermediaries, then a shortfall of supply is an endless and ineradicable threat to credit markets — and therefore an endless and ineradicable threat to amplifying the effects of an economic downturn.
There is something about Gorton’s work that has always been deeply unsettling.
He partly explains why in his book, Slapped by the Invisible Hand (emphasis ours):
Paradoxically perhaps, this is somewhat contrary to modern finance theory, which focuses on equity markets and extols the virtues of “market efficiency”, that is, the idea that equity prices contain lots of information. That may be fine for equity markets, but for much of the debt market there should be no reason for prices to reflect a lot of information. In fact, the needs of the economy are for precisely the opposite.
The core problem is that there is no such thing as a safe asset, as the world has so painfully learned. “Safe asset” is just a phrase that describes assets perceived to be safe enough. But we can never completely eliminate the possibility that an asset will go from safe enough to not safe enough.
To put this in Gorton-speak, there is no such thing as an information-insensitive asset that cannot one day switch to information-sensitive. “Insensitivity” has its limits.
Yet the demand for information-insensitive assets seems to hold across time and across varying economic conditions, and nobody knows why, as the authors of the paper write.
Had we been thinking about this during a benign period for global financial markets, we would instinctively have wished for this illusion of safety, of insensitivity, to be shattered. Sure, we might have recognised that even the simplest financial transactions are based on some kind of trust, but at this scale and complexity it probably would have seemed dangerous.
It turns out that the illusion not only persists, but has come to be a necessary part of the banking credit transmission channel — and it is the very shattering of the illusion, when it happens, that exacerbates a crisis.
The Gorton-Lewellen-Metrick paper’s findings about the persistence of safe asset demand seem to reinforce the disturbing notion that there is no going back.
We have to think about this a lot more, but we’ll probably be back later with a post on the confounding difficulty of finding any solution.