Not our words, but those of Richard Comotto of the European Repo Council.
In case readers are not familiar with Mr Comotto, he’s the author of the ICMA’s semi-annual survey of the European repo market — probably the best (if not the only) overview of the repurchase market in Europe.
Comotto’s job involves interviewing repo market participants throughout the year, in a bid to try and understand what’s really going in the repurchase world.
The reason his study is so valuable, meanwhile, is because nobody else really does the same thing, and statistics remain hugely scarce in the market.
As he tells FT Alphaville, the European market continues to be fragmented, compartmentalised and opaque. Getting data on how many deals are being done, meanwhile, is particularly hard. So is finding out the bid-ask rates being charged, plus when and how often securities are, or are not, going ‘special’ — that is, become temporarily over-priced due to scarcity or unique levels of demand.
Regarding the amount of information that’s in the public sphere, Comotto explains:
“For Europe, you have the ICMA survey and some data on some individual countries. There’s very little on Germany. Italy has electronic data. France and UK have good data. But there isn’t a Platts-equivalent in the repo market.
“Brokers only see a certain type of business going through. But a lot of business bypasses them, so they’re not always a good indication. But the repo market has better transparency than many others (eg commodities & derivatives).”
“The repo market is like an iceberg. You only see a tip.”
Nevertheless, there are some obvious trends emerging.
Among the most prominent, he says, is the continued push towards collateralised lending across the entire board — something that happens to tie in with what we have been hearing from other sources too.
What Comotto emphasises, however, is that while that trend may have begun as far back as the 1990s, the recent European crisis has now nearly completely vaporised what little unsecured interbank lending was left in the market. What’s more, the demand for tri-party transactions — where collateral is managed by a custodian rather than bilaterally — has almost doubled from less than 25 per cent before the Lehman crisis to almost 50 per cent since.
The lack of unsecured lending options is one reason why a number of new players have recently flocked into the European repo market. Spanish banks, for example, have most recently sought access to the cross-border market via central counterparty institutions — presumably as they prepare for ECB liquidity to be phased out completely.
As Comotto noted to us:
With the crisis, even if you have customers happy to put it into you unsecured you’re not so willing to lend it out on the same basis. So banks previously not into repo are now getting into it.
If the crisis hadn’t happened maybe they wouldn’t have.
The other detectable trend, meanwhile, is just how fussy banks continue to be about the type of collateral they accept for repo deals on an interbank level. According to Comotto, the Lehman crisis pushed the proportion of government securities being demanded in repo transactions to 83.6 per cent versus 81 per cent in June 2008.
That might not sound like a lot, but according to Comotto it is a substantial sum in absolute terms, especially when you consider that the number already started at a high base. What’s more, the figures reverse what was previously a downward trend.
The remainder of the collateral, meanwhile, is now exclusively focused on good quality alternatives like covered bonds, supranational bonds or equity. ABS-backed transactions, though, have almost completely vanished.
All of the above hence points to no real improvement in interbank borrowing costs and what’s worse a potential quality-collateral run in Europe, much like that seen in the US post the Lehman crisis. As Comotto states (our emphasis):
The concern now is there may not be good govts, to supply demand.
Although he further stresses:
Currently it’s a concern, rather than a fact. What’s worrying the repo market is that regulators want more use of high quality collateral and use of CCPs and yet already the demand for collateral is quite broad and already there are doubts about supply, linked to downgrades of sovereign issuers.
We might find ourselves in a situation there isn’t enough AAA to satisfy our needs. So there is a big effort now in Europe to use bank loans as collateral. Or we have to adjust our perspective and realise there is no such thing as perfect collateral.
But where’s the evidence of market disruption?
Government repo markets traditionally reflect potential supply issues via the ‘special’ securities market as well as settlement failures (as participants opt to fail on deliveries, rather than source overpriced securities).
In a crisis situation — both of these metrics would naturally be expected to rise, just as they did in the US after the Lehman crisis. Comotto explains the situation thus:
When you get a financial crisis, people want to do things pretty fast. They have to sell positions and provide collateral quickly, and if you have securities stuck in the pipeline you start to get fails. The problem then spirals. If I fail, others fail.
The interesting thing about the European sovereign crisis this year, however, was that even though the number of fails did go up — participants knew this was mainly down to the market’s fragmented and choppy structure in Europe.
Unlike the US — where fails zoomed higher in the aftermath of the Lehman crisis — the European repo market never completely seized up. (Nor did it seize up after the Lehman crisis either, for that matter.) As Comotto explains:
Repo kept functioning, whereas unsecured evaporated completely.
One big reason for this were the central bank facilities available in some major collateral markets in Europe, which ensure securities can never go special.
The Bank of England, for example, uses a standing repo facility to create phantom copies of those securities it believes are too scarce. The Bundesbank, meanwhile, holds back a certain portion of new issuance for the purposes of lending into the market as and when needed.
As Comotto states:
“These systems come into play in the most difficult circumstances. Scarcity is not a problem until people start failing. During the crisis, which was an exceptional event, fails did start to go up, but it wasn’t as bad as it could have been because of these procedures.”
But why then have Euribor rates been largely unfazed, if there was really such a collapse in unsecured term lending, we hear you ask?
Well, according to Comotto, it’s down to the way the Euribor rate is compiled.
As he explains:
The strange thing is when you look at Euribor fixings there was some movement but not a lot. Yet you would have had difficulty to borrow for 12 months.There’s no business, at those levels. None at all. So what have they been quoting? It’s their best judgment where the rate would be if it was used — according to their models. It’s based on rates implicit in other transactions. It’s a very very difficult thing to assess. And six months is now a very long time in unsecured lending.
Which means — especially in the longer-term contracts — Euribor may have become nothing better than a model-dependent banking guesstimate.
The Collateralized Lending Regime: An Under-reported Shift in Capital Structure – FT Long Room
Developments in repo markets during the financial turmoil – BIS
Frozen in the Greek repo markets – FT Alphaville