The post-hoc (correlation implies causation) reasons for why the initial LTRO spurred bond buying are many-fold but as Nomura points out in a recent note (confirming our thoughts from last week) investors (especially bank stock and bondholders) should be very nervous at the size of the next LTRO. Whether it was anticipation of carry trades becoming self-reinforcing, bank liquidity shock buffering, or pre-funding private debt market needs, financials and sovereigns have rallied handsomely, squeezing new liquidity realities into a still-insolvent (and no-growth / austerity-driven) region. Concerns about the durability of the rally are already appearing as Greek PSI shocks, Portugal contagion, mark-to-market risks impacting repo and margin call event risk, increased dispersion among European (core and peripheral) curves, and the dramatic rise in ECB Deposits (or negative carry and entirely unproductive liquidity use) show all is not Utopian. However, the largest concern, specifically for bondholders of the now sacrosanct European financials, is if LTRO 2.0 sees heavy demand (EUR200-300bn expected, EUR500bn would be an approximate trigger for 'outsize' concerns) since, as we pointed out previously, this ECB-provided liquidity is effectively senior to all other unsecured claims on the banks' balance sheets and so implicitly subordinates all existing unsecured senior and subordinated debt holders dramatically (and could potentially reduce any future willingness of private investors to take up demand from capital markets issuance - another unintended consequence). We have long suggested that with the stigma gone and markets remaining mostly closed, banks will see this as their all-in moment and grab any and every ounce of LTRO they can muster (which again will implicitly reduce all the collateral that was supporting the rest of their balance sheets even more). Perhaps the hope of ECB implicit QE in the trillions is not the medicine that so many money-printing-addicts will crave and a well-placed hedge (Senior-Sub decompression or 3s5s10s butterfly on financials) or simple underweight to the equity most exposed to the capital structure (and collateral constrained) impact of LTRO will prove fruitful.
The spread between senior and subordinated financial risk has compressed off crisis wides but remains elevated at prior crisis peaks. While not cheap explicitly, it would be a most direct way to hedge an oversize LTRO's impact on bank's balance sheets and obviously is a quite contrarian play.
Notably, even with spreads rallying today, this spread started to leak back wider as perhaps smart traders are seeing this opportunity.
LTROs are increasingly punitive and lead to subordination of senior unsecured bank debt.
The haircut structure and increased asset usage effectively means that further ECB liquidity is increasingly punitive, utilising ever more balance sheet. The more the facility is used the greater the degree of subordination to senior creditors, which previously would have partially relied on the assets, now pledged to the ECB, as security against senior debt. This problem is particularly pertinent given that banks have already been using the covered bond markets to raise funds, which require over-collateralisation in order to achieve higher ratings and to meet the criteria laid down by the ECB in order to be deemed eligible collateral for operations.
Nomura: How big will the next LTRO be?
The next 36-month operation is likely to be big; the question is how big? It should be bigger than the shorter LTROs as the long-term operations have a major advantage with regards to the timing of the payment of interest on borrowings. Besides the haircut taken on the collateral, the interest cost, from a cash perspective, is only settled at the end of the term of the repo meaning a reduction in interim funding cash flows. This is a major advantage over the 1w/1m/3m funding roll to cash-strapped banks.
Upside reasons to consider:
- The inclusion of a broader spectrum of loans should lead to a significant amount of additional balance sheet available for repo through the ECB operations. The downside to this is the significant haircuts on these assets.
- With the reduction in credit criteria on loans eligible. There is perhaps potential for funds drawn through the ELAs in Greece and Ireland to transfer to the mainstream ECB LTROs.
- Deposit levels in peripheral countries, and the degree to which they have been fluctuating, may be instructive as to the amount of funding that may be taken down. The higher the volatility of these deposits the more attractive the ECB insurance option should be.
Factors that could weigh on the overall size of the next operation:
- Banks have taken a significant amount of ECB funding already, with a marked increase from Spanish and Italian banks. According to national central bank data, an incremental €26bn and €57bn was taken by banks domiciled in those countries respectively. The Spanish take at the end of December 2011 was €132bn against the €210bn borrowed by Italian banks. We think the funding taken will cover 2012 bank redemptions to a significant degree.
- Decreased volume of liquidity rolling from old LTROs and MROs :-> For the December operation €50bn rolled for the October 12-month operation as well as from the 3-month.
- The ECB dropping its reserve ratio from 2% to 1% means liberation of funds, though this should largely affect MROs on the margin, but funds that were not provided through liquidity operations will now be available to banks increasing available liquidity.
- Bigger picture, the level of deleveraging in the banking sector as well as the private sector could lead to reduced funding requirements.
In aggregate we think that the total level of funds taken down through the ECB operation will be less than the previous round. In our estimation this is likely to be in the €200-300bn range.
If the size is bigger than this, perhaps in the range of €500bn or greater, the effect on bank balance sheets in Europe will be distinctly negative in our view, and would make future wholesale and term funding from private sector sources significantly more difficult.