Courtesy of the European economics team at UBS…
We want to meet the respondent who said €3 trillion. Were they just larking around or where they going for a wild outlier on the book in the hope they might strike lucky and look like a genius.
All we know is that it was someone inside UBS…
Either way, the average forecast figure comes in at €668bn, with a median figure of €629bn.
With reference to the methodology, UBS explain:
Let us start with the amount, we believe, banks really need. To estimate this amount, we used the work of our banks team. They reached the conclusion that the amount taken by banks will likely be €300bn. Our colleagues wrote: “There is much discussion of the likely size of the end-February LTRO take-up, with the high end of market expectations close to €2 trillion and many commentators expecting €1 trillion. We expect a much smaller figure – around €300 billion – for two key reasons.” To this number, we must add the amount of other LTROs maturing. Remember that LTRO stands for “long-term refinancing operation” and also comprises other long-term tenders, such as the one month, three month, six month and one year.
Adding that ELA sums could now also move over into LTRO because of expanded collateral rules:
The ELA amount is not explicitly mentioned in the balance sheets of the national central banks, but our research shows that the amounts in the case of Greece and Ireland total close to €45bn. We believe that the ELA programme is small in the case of Portugal – probably less than 10bn. Finally, we have to add an amount for probable ELA for the troubled banks in Benelux. In total, the ELA amount currently outstanding is likely to be around €100bn.
ELAs are an exceptional procedure, with the repo being two weeks’ maturity at a rate 200bp above the lending facility, which means that banks are currently paying 3.75% on their ELAs. The extension of the collateral rules will allow some banks to use LTROs as an ELA replacement on much more advantageous conditions: three years at 1%. However, the question is whether the banks using the ELA will have enough collateral to go to the LTRO. Again, according to our banks team, the answer is probably yes in the case of the Irish banks, but likely no to a large extent for the Greek banks. Thus, we would not expect the total ELA amount to migrate to the LTRO; in our opinion, two-thirds seems to be a reasonable estimate.
In other words, even with the expanded collateral rules, collateral constraints may still hinder some banks from tapping what they really want.