Global banking regulators envisage a low-risk world where banks fund themselves through long-term deposits and bond issuance. However, the reality on the ground in a Europe ravaged by a sovereign-debt crisis could not be more different. With both euro and dollar term markets closed for the foreseeable future and US money market managers turning their backs on the sector, European banks are becoming ever more reliant on the shortest-dated funding sources and central bank liquidity programmes.
There is little sign that conditions are likely to change. With markets still waiting for further details from Decembers European Council summit, and banks on both sides of the Atlantic suspecting each other of hiding more toxic sovereign assets than previously thought, the availability of market-based funding in general, and dollars in particular, is extremely limited.
The root of the problem at the moment is sovereign risk, says a European banking analyst in London. Banks dont trust other banks assets. No matter how much capital gets put in, its not possible to get comfortable on the liability side of things.
The failure of John Corzines MF Global, and subsequent revelations about its substantial yet hidden exposure to European peripheral sovereign debt, have renewed suspicion among banks regarding their actual positions in Greek, Italian, Irish, Portuguese and Spanish government bonds, official denials to the contrary notwithstanding.
It is always a concern when money flows to the front end of the curve, mused one global DCM head during a recent meeting with Euromoney. And that it is most certainly doing. Money rates in both euros and dollars show an acceleration of fear in the interbank market. For example, the spiking Euribor-OIS spread, which measures the difference between the floating rate at which banks lend to each other and the European Central Banks overnight rate, jumped by more than 20 basis points during September. Although it fell slightly in early October, the classic bank credit-risk indicator spiked up again around the time of the MF Global bankruptcy filing on October 31, spending most of December around the 100bp mark, a level last breached in the first half of 2009.
The collapse of John Corzines
Furthermore, the euro/dollar basis swap, which measures the cost at which European borrowers can fund themselves in dollars, has touched levels not seen since mid-2008. With the three-month basis, the most liquid part of the market, dropping lower than minus 150bp, the fall in appetite among US money to extend dollar funds to European banks is driving an acute shortage of dollar liquidity in Europe. Moreover, the launch of the coordinated dollar liquidity swap by the ECB and Fed in early December only temporarily relieved euro/dollar basis stress, implying that the collapse of dollar supply is likely to remain a serious problem in 2012.
Concerns over the potential impact of sovereign credit issues on eurozone banks continue unabated, note Madeleine Gish and Dan Tronstead, prime fund managers with Wells Fargo Advantage Money Market Funds.
US money markets have also bifurcated between the haves and the have-nots, with the US subsidiaries of foreign banks generally on the wrong side of the fence. The latest statistics on US money market funds show that capital flight from European banks, which broke out in the summer, continued in the fourth quarter; US managers cut exposure by a further 9%, according to Fitch Ratings.
As a result, the US subsidiaries of European banks, which had been a valuable source of dollar funding for international financial conglomerates, are now net borrowers from their own headquarters, placing further strains on European dollar liquidity, says Ray Stone, president of Stone & McCarthy Research, a Princeton-based company providing analysis to institutional investors. US banks, meanwhile, which relied heavily on foreign subsidies for term funding after the Lehman bankruptcy, are now net lenders to foreign subsidiaries. For foreign, read European.
The swings in net positions vis-à-vis foreign-affiliated offices appear to be related to shifts in alternative funding options, says Stone. Presumably, foreign banks operating in the US face a less receptive market for their dollar-denominated certificates of deposit. US offices of these branches have witnessed a plunge in large time deposits as US money market funds have a sharply diminished appetite for this paper.
Between June and mid-November 2011, CD issuance by US branches of foreign banks fell to $782 billion from $1 trillion, while foreign banks borrowed $265 billion from their head offices in mid-November, up from $84 billion. At the same time, foreign banks onshore US cash balances fell by $367 billion, while US banks core deposits increased by $379 billion, mostly as a result of outflows from money market funds.
Data from the Federal Reserve show a similar contraction in the supply of US commercial paper, which shrank by around $100 billion over the course of 2011. Again, US subsidiaries of foreign banks suffered the worst reductions in demand. However, the daily supply of CP from the financial sector in general also fell by $10 billion over the year as investors shifted allocations away from financial sector credits. The tenor of funding available from US CP investors fell from an average of 44 days in July to an average of 41 days by December.
Shut out of unsecured interbank and capital markets in the eurozone and in the US, European banks, dealers say, have increased their reliance on short-term secured markets, principally repo. Despite a lack of aggregate European repo data and the fact that banks use it to finance trading activities as well as to fund balance sheets, it is possible to piece together an indicative picture from data gathered in surveys conducted by the ECB and the International Capital Markets Association (Icma).
In its November monthly bulletin, the ECB noted that the annual growth rate of M3, a monetary aggregate that includes interbank repurchase transactions, had increased to 3.1% in September. The marketable instruments component of M3 increased to 7.3% from 5.3% in August.
|Deposit flows of foreign banks' US subsidiaries|
|To Q3 2011|
Relatively large numbers of repurchase transactions continue to be conducted via central counterparties, the ECB said, adding that this highlights distrust among banks and points to a preference for collateralized interbank lending via electronic platforms rather than direct unsecured interbank lending.
Icmas most recent survey, which reports the outstanding repo contracts of 55 European financial institutions at the close of business on June 8, reveals that the overall European repo market was around 6.1 trillion, up from 5.9 trillion in December 2010.
Icma said: The repo books of the 49 institutions that participated in all of the last three surveys grew by 3.6% over the six months from the December 2010 survey, while the surveyed business of the 53 institutions that also participated in the June 2010 survey showed year-on-year growth of 10.2%. The growth in the total value of transactions over the last six months therefore continued the upward trend re-established in the first half of 2009.
The available time series, such as it is, stops in September, at about the same time that risk-off reasserted its grip on market psychology. It is likely, therefore, that Icmas December 2011 survey will indicate a slowdown or, more probably, a contraction.
Don Smith, an analyst with Icap, which owns the market-leading electronic repo trading platform Brokertec, reports a pull-back in private repo markets that he says is being partly caused by a shortage of high-quality collateral in the underlying bond markets as well as the increased role of the ECB as repo counterparty to many European banks. At 175 billion a day, daily repo volumes traded on Brokertec were around 35 billion lower in mid-December than they were at the end of the third quarter, Smith notes.
The result has been a reduction in position-taking in underlying government bond markets that typically involve short-term repo activity as well as an increase in demand to hold government bonds and not release these as collateral. This has been the case especially for German government debt, where the shortage of available collateral is increasing. Market repo volumes have fallen back since the summer, but this appears to largely reflect the risk-off environment rather than a reduction in bank reliance on this form of financing, Smith says.
The flip-side to German general collateral (GC) being quoted through Eonia, the eurozones so-called risk-free rate, is ballooning rates for Spanish and Italian collateral, and other peripheral eurozone sovereign bonds where participants still accept them as collateral. For example, in mid-December three-month Spanish GC was quoted around 80bp above Eonia, while three-month Italian GC was around 60bp wider. Local banks that are natural holders of domestic sovereign debt therefore face higher risk premiums when seeking to deploy peripheral collateral in private repo markets, while high nominal yields and rating downgrades make them more expensive to use as collateral with central counterparty clearing platforms and exchanges.
The increasingly harsh penalties associated with peripheral risk have pushed holders to source more desirable collateral, further reinforcing its scarcity value.
|Who holds what|
|Fitch's US money market fund survey of holdings|
|Click here for a bigger version of the graph|
While the repo rate on three-month German government GC has tightened close to zero from its summer wide of around 1.2%, and it remains around 40bp tighter than Eonia, the lack of general position-taking in the underlying bonds has prevented the emergence of widespread negative yields. In other words, there are very few instances of particular bonds standing out as trading with a scarcity premium relative to other bonds. This is fairly unusual and supports the view that there is little active position-taking going on in the underlying German government bond market or repo market, says Smith.
Against increased demand for German bonds, repo lenders are now entertaining requests to fund against long-dated paper as the lack of offers in sub-10-year collateral is forcing buyers to relax maturity restrictions and accept up to 30-year paper, according to Icap. Notwithstanding the widening of eligibility criteria, the risk-off mentality and hoarding of liquid assets has resulted in the German repo market winding down into the year-end, with activity on Brokertec falling from around 90 billion a day in July to less than 70 billion in December.
It is not just supply of German collateral in European repo markets that is limited but also high-quality collateral in general. In the US, more than $2.3 trillion of Treasury and agency mortgage-backed securities purchases under the quantitate easing I and II programmes and the roll-off of the Treasurys $200 billion supplementary financing programme exacerbated the effects of counterparty risk aversion and the flight to quality to make repo collateral and Treasury bills even more scarce. Like their European counterparts, US money markets continue to hoard collateral to meet potential liquidity demands. Almost all other private holders of high-quality securities are doing the same thing.
Although there is a cyclical tendency for financial intermediaries to flatten off repo positions and hold on to collateral over the year-end to flatter balance sheets, the scarcity of collateral is a trend that has been gathering steam since the Lehman shock, according to senior IMF economist and former emerging markets debt syndicate banker Manmohan Singh. He says that the velocity of collateral, or the rate of its use and re-use by financial intermediaries in the shadow banking system, is a vital yet largely misunderstood mechanism for the transmission of credit to the real economy. The chains between counterparties that pledge and receive the same underlying collateral lubricate the cogs of monetary policy (see page 74).
Singh estimates that the net reduction in underlying collateral, pledged and re-pledged (or re-hypothecated) through the financial system is approximately $5 trillion since 2008. This has attracted much attention recently, mostly from commentators using parts of his analysis to argue that collateral chains pose an unregulated risk to financial stability by re-using collateral to increase off-balance-sheet leverage.
All but the
Indeed, a common criticism levelled at the Feds quantitative easing programmes has been their failure to increase the flow of credit to the real economy and stimulate growth. Banks in receipt of the Feds freshly printed dollars have tended to deposit money back with the Fed, rather than lend it on to the real economy. Singh maintains that collateral chains connect owners of collateral (typically asset managers) with those who need it. He says that shorter, less elastic chains have an even greater potential to disrupt the credit channel than the higher collateral haircuts that have accompanied the latest round of deleveraging. This is because they reduce the velocity of collateral, and its availability to a party that needs it elsewhere in the economy.
A shortage of acceptable collateral would have a negative cascading impact on lending similar to the impact on the money supply of a reduction in the monetary base, he says. Thus the first-round impact on the real economy would be from the reduction in the primary source collateral pools in the asset management complex (hedge funds, pension and insurers), due to averseness from counterparty risk. The second-round impact is from shorter chains from constraining the collateral moves, and higher cost of capital resulting from decrease in global financial lubrication.
Notwithstanding Singhs efforts to promote the broader economic benefits of rehypothecation, his work does encourage regulators to improve their understanding of banks increased reliance on rehypothecation as a source of funding. In terms of policy, regulators will need to consider the re-use of pledged collateral when defining bank leverage ratios, Singh argues. Also, given asset managers demand for non-M2 types of money, monitoring the shadow banking system will warrant closer attention well beyond the regulatory perimeter.
While some banks promote the happy idea that they can avoid both capital market dysfunction and central bank humiliation with an elegant ensemble of emergency deleveraging and private placement funding, the reality is that all but the super-heavyweight universal conglomerates will likely lean heavily on the new ECB liquidity programmes for the foreseeable future.
Despite a relatively low take-up until recently, the improvement in terms offered by the ECB has encouraged increased participation across its range of facilities. For example, the dollar swap line was relatively expensive compared with market rates until a coordinated 50bp cut and a reduction in collateral haircuts from 20% to 12% made the facility more attractive in the fourth quarter of 2011. The improvement in economic terms spurred an immediate response, with European banks taking up $50 billion in the latest three-month dollar operation, and $5 billion in the one-week tender. However, the relatively small number of bidders (34) could mean that many remain fearful of the reputational risks, or simply couldnt source the collateral to exchange.
Meanwhile, the introduction of the 1% three-year Long Term Refinancing Operation (LTRO), which replaced the 13-month programme, has surprised some banks with the generosity of its terms. The LTRO is a lot more positive than wed originally expected. We had been prepared for a two- or three-year facility, but the flexibility of what was announced is very appealing, and allows banks to roll other operations into this operation, says Guy Mandy, fixed-income strategist at Nomura in London.
Although the facility allows banks to fund assets with the ECB for up to three years, it provides the flexibility to roll longer-dated operations into shorter-dated ones after one year. If there is an improvement in market-based funding during that time, banks can exit the programme and not have assets encumbered for the full three years, he says. However, the possibility that stronger banks will deposit their ECB bills back with the bank, rather than make them available as collateral, remains an issue.
This brings us back to the collateral question bothering Manmohan Singh. How can policymakers defrost the collateral freeze, extend the rehypothecation chain and encourage financial intermediaries to start lending to each other again? In this respect the mis-sequenced meetings of European central bankers followed by policymakers represents another missed opportunity to address the root cause.
Clearly, the European planners got their last meetings of 2011 in the wrong order. If sovereign credit risk is driving the collateral squeeze in private repo markets, then shouldnt the policymakers have met before the ECB to hammer out the fiscal compact before the bankers sat down to work on financial remedies to the bank funding crisis? In any event, LTRO might yet prove to be quantitative easing by another name. As the runaway success of the Spanish government bond auction immediately after its introduction suggests, with banks taking down 6 billion of new paper at 5.2% versus expectations of just 3 billion, the new programme provides a timely incentive for European banks to pile back into sovereign debt to offload it to the ECB, although the extent to which this is actually happening remains unclear.
However, with Italy facing debt redemptions of 115 billion in the first four months of the new year alone, and the European banks potentially needing to refinance around 260 billion of term funding in the first quarter, the convergence of short-term credit risks in the European banking system almost wholly reliant on short-dated funding will surely force the ECB to act in 2012. It is important to provide as many incentives to purchase sovereign debt as possible, says one banker. Taking assets off balance sheets and replacing them with central bank bills when people clearly need collateral helps de-leveraging, and decreases counterparty credit risk. Therefore we need a big balance sheet to step in.
The IMF is potentially too small, only the ECB has the capacity to do it, either through quantitative easing or an extended securities market programme. There will come a time, in the near future, when there will be a realization that the ECB needs to do something, and we will probably need another risk event to force that. However we should be trying to avoid that kind of situation. But despite Draghis acknowledgement in mid-December that small and medium-sized banks are now suffering repo collateral shortages, it might yet take a failed core sovereign auction or another bank bailout to force him to bite the bullet.