An article published in December 2013 by the Centre for Economic Policy Research “The roots of shadow banking” by Enrico Perotti of the University of Amsterdam, ECB and CEPR grabbed our attention. The author blames repo for the bad lending in the mortgage markets during the financial crisis.
The paper looks like your basic shadow banking piece, suggesting that the run on shadow banking funding was no different than a loss of confidence leaning to a bank run. We have seen this before and agree with the premise….but there are some interesting bits.
One of the author’s primary assertions is that the ability to liquidate upon default (a/k/a the safe harbor provisions that both repo and derivatives benefit from) created an environment ripe for fire sales. The liquidation of Lehman was cited as an example. Oddly enough Perotti did note that no one lost money in those liquidations, but prices were forced down creating negative externalities. Reports that Lehman’s centrally cleared derivatives portfolio were unwound without losses have been around for a while. It begs the question: if the derivatives were unwound without a loss, this must mean that the margin was enough to absorb negative market movements. Were those margins so huge to protect the non-defaulting counterparties but still create negative externalities? We would suggest that the negative externalities were more about counterparty uncertainty and frozen markets, less about prices falling. The regulatory push to CCPs would support that. But we digress.
Back to repo.
“…A jump in market haircuts, and ultimately a refusal to roll over security loans or repos, is the shadow banking system’s equivalent to a classic bank run. As a security borrower cannot raise as much funding from its own illiquid assets, it is forced to deleverage fast or go bust. In both cases this triggers fire sales…”
OK. No major issues there.
“…The guaranteed ease of escape for repo lenders led to the final burst in the pace of mortgage lending and repackaging in 2004–2007, when credit standards fell through the floor…”
Wait a second. Did he just imply that the ability to liquidate quickly when counterparty defaults on a repo encouraged mortgage lenders to ease their credit rules? That seems a major stretch.
Perotti does have some ideas on how to attack the safe harbor “problem”. Most readers of this blog have read about them before. The author does have some new ones: registering claims subject to immediate liquidation upon default and to charge a fee for the right to liquidate upon default.
“…At the macroprudential level, once collateral held under safe harbour were registered, policymakers would be able to track its evolution, finally enabling the mapping of contingent liquidity risk. If the stock appears to grow too fast, various steps may be undertaken…”
This is a variant of trade repositories and serious thought should be given to adding a “tick the box” for safe harbor eligibility when collecting data. Decoding and understanding the chains of data will be another thing altogether.
The idea of a fee to pay for the privilege to liquidate is intriguing. It is a first cousin the concept expressed by academics and regulators about creating specific capital charges on securities financing businesses (which was in the context of preventing fire sales). Quick liquidation, as we have written before, means that haircuts can be lower. If you can assume it will take a couple days to liquidate versus being stuck in bankruptcy court for months (or longer), the expected end price will be closer to the price the trade was put on at (plus/minus variation margin). Lengthening the liquidation period will mean bigger haircuts in order to protect against the greater potential exposure.
All of these ideas are likely to evolve. But the seeds of change are out there in one shape or form…and the industry needs to be prepared.