A fair chance for the European Central Bank

In a couple of tweets I published on the occasion of the last press conference of the ECB president on June 5th, I gave my first assessment of the package of measures decided by the ECB: not overwhelming. Overall the market seemed to share this assessment as the exchange rate hardly moved while short term rates (Euribor, OIS and Bubills) only came down in a limited way (Chart1). 

Chart 1: Change in money market curves since the 6th of June
Source: Bloomberg
My “not overwhelming” assessment was based on 6 factors:
  1. While the ECB could have reduced MRO and deposit rate by more than 10 bp, say by 15 or 20, it decided to stop at the minimum possible reduction,
  2. The Targeted Longer Term Refinancing Operation (TLTRO) is conditional and fairly complicated, which will make it necessarily less enticing,
  3. The rate reduction does not, in practice, apply to the TLTRO given the spread applied to it (10 bp),
  4. There was no attempt to fight stigma, which is negatively affecting the demand of banks for ECB funding [1],
  5. The decision not to sterilize any more the liquidity created by the Security Market Program purchases will only have a transient effect on excess liquidity, and therefore on short term rates, unless there is an incentive for banks to demand more liquidity on a persistent basis from the ECB,
  6. The announcement that work on ABS is being intensified, while welcome, remained short of a concrete content [2].
Of course I recognized 3 favourable factors in the package:
  1. A potential maturity of 4 years, which I identified as “enticing” in a previous post [3],
  2. The factor of 3 in the relationship between the net increase in lending and what can be drawn in TLTROs to be allotted in 2015 and 2016,
  3. The clear statement that the ECB could do more if the announced package would not be enough to avoid too long a period of too low inflation.
After the teleconference I spent a great week in Andalucìa and, while admiring that wonderful juxtaposition of Arab and European civilization that are Malaga, Seville, Cordoba and Granada, I thought further about the package to come to a more pondered assessment.
A first critical aspect in this more pondered assessment is an evaluation of the prospects of  banks demand for the TLTRO. To come to this evaluation it is useful to examine separately the first two operations, to be allotted in September and December of this year, from the subsequent 6 operations, to be allotted in 2015 and 2016.

The two operations of 2014 can be best be seen as an unconditional, but limited in size, rollover for 2 years of the two 3 years VLTROs (Very Long Term Refinancing Operations), which will expire at the end of the year. In fact banks can borrow up to an amount of 7% of their lending to the private sector (excluding mortgages) on April 30 2014, but will have to return the money in September 2016 if they will not have respected the lending conditions.
One first consideration to estimate what could be the demand for the first two operations is that this should mostly come from peripheral banks. It is in fact doubtful that core banks will find it interesting to borrow large amounts from the TLTRO, even if this is in practice unconditional for the first two years. Thus the total of 400 billion mentioned by the ECB President looks more like an upper limit than an unbiased estimate. My estimate starts from table 1. My point if departure is 7% of the lending of peripheral banks to the private sector, excluding mortgages, [4], which gives 155 billion, assuming that these banks will borrow as much as possible from the TLTRO To add the demand for TLTRO from banks in core jurisdictions, an estimate of the amount of their borrowing in the still outstanding 2 years VLTROs is added, assuming that they would want to roll it over. Adding the two components, my upper estimate for the demand for the first two TLTROs will be, at 289 billion, quite lower than 400 billion mentioned by Draghi, as it should predominantly come  from peripheral banks, which only have about 40% of the total lending to the private sector excluding mortgages.

Table 1: Upper estimate of TLTRO demand
Banks from:
Loans to Private sector X-mortgages (1)
TLTRO allowance (2)=7%* (1)
Remaining VLTROs drawings
Upper Estimate of TLTRO demand
(4)=(2) for non-core banks + (3) for core banks
Source: ECB, Author’s calculations; Private SectorX-mortgages is Private sector excluding mortgages.

A second factor to guess which demand there will be for the first two TLTROs is how convenient in cost terms they will be. It is useful to compare the 25 bp fixed cost of these two operations to the situation that was prevailing before the ECB move as well as that which is prevailing now, after the move.
Before the ECB move, banks could borrow under the Longer Term Refinancing Operations  (LTRO) at the same 25 bp cost now available for the TLTRO and roll it over for two years. The difference between this kind of borrowing and borrowing under the TLTRO would have been:

  1. Rolling over a 3 month operation is not the same as having a fixed 2 year operation,
  2. There was a (very small) risk that the rate on the LTRO coud be increased,
  3. There was a (very small) risk that the Fixed Rate Full Allotment (FRFA) modality would not be prolonged after the date of July 2015.
Overall, only the first factor is likely to be important but, compared to the situation prevailing before the ECB decision, the offer of the TLTRO did indeed provide somewhat more favourable conditions, especially to peripheral banks.
After the ECB move, banks can either borrow for 2 years under the TLTRO at a cost of 25 bp or under the LTRO at 15 basis points and roll-over the borrowing for two years, with the assurance that the FRFA modality would be maintained until the end of 2016 and only the very small risk of an increase of the rate by the ECB. This of course reduces the relative attractiveness of the TLTRO with respect to rolling over borrowing from the LTRO. My sense is that this effect would be strong enough to reduce the demand of the first two operations of the TLTRO below the around 300 billion presented in table 1, even if I am not able to determine by how much.
In conclusion, my overall judgment is that the first two, unconditional, 2 year TLTROs will not even allow rolling over the maturing VLTROs and thus do not deserve much more than my “not overwhelming” initial assessment. 

On the subsequent 2015-2016 TLTROs I maintain the reservations that I expressed in a previous post [6]. However,  Andalucìa, made me somewhat warmer on their possible effect. Let me explain why.
  1. The enticement of the 4 year maturity can be significant: the 25 basis points cost for 4 year fixed maturity (at half the level of Euroswaps) is pretty convenient, as can be seen in chart 2, even after the fall in rates that was determined by the ECB package,
  2. Credit demand could increase in 2015 as the recovery takes hold and the banks could have better incentives to satisfy this demand, given the higher margins between cheap ECB funding and more remunerative lending,
  3. Balance sheet cleaning and bank re-capitalization due to the Comprehensive Balance Sheet Assessment conducted by the ECB, which will be over at the end of this year, should make banks more able to exploit profit opportunities from lending.
Chart 2:Money market curves at 13th of June

Source: Bloomberg


Of course, there is a negative impact on demand, deriving from the complications to which banks will have to subject themselves, including special reporting requirements, to borrow under the TLTRO. However, I regard this negative factor not sufficient to fully offset the three positive ones mentioned above. In particular, I think the third point is potentially very important: in a way the two hands of the ECB, the monetary policy one and the supervisory one, are working jointly and we could see a useful synergy effect.
This could be reinforced by the effect on short-term rates coming from higher excess liquidity. This will be buffeted by different events:

  1. The stopping of the sterilisation operations will inject around 160 billion, 
  2. At the turn of the year, instead, there will be the withdrawal of liquidity because of the maturity of the two 3 years LTROs, whose outstanding amount is still around 425 billion,
  3. If my guess is right, the two TLTROs at the end of this year should partially compensate the maturing VLTROs, up to the maximum amount of around 300 billion estimated in table 1,
  4. In 2015 liquidity could be supported by the demand for the successive TLTROs.
  5. All through the period, excess liquidity will be influenced by the demand of banks at “normal” MRO and LTROs as wells by so called autonomous factors, which are difficult to forecast.
While it is difficult to get to a bottom line from the above mentioned 5 events, my sense is that liquidity should be abundant and thus keep the overnight rate (EONIA) as well as the Overnight Index Swap (OIS) Rate close to the bottom of the new corridor, hovering just below or around 0.
Would this be enough to make inflationary expectations again firmly anchored, thus re-establishing the firm pursuit of the ECB objective to keep inflation below but close to 2.00 per cent? President Draghi evidently believes so, when he says: “Together, the measures will contribute to a return of inflation rates to levels closer to 2%.” Even if I find it contradictory that, in the same paragraph, he says that: “Inflation expectations for the euro area over the medium to long term continue to be firmly anchored in line with our aim of maintaining inflation rates below, but close to, 2%.” If that was really the case, why come out with such an articulate package?

I believe myself there is a fair chance that the joint effect of the monetary policy move decided in June and the conclusion of the Comprehensive Balance Sheet Assessment could bring back inflation, albeit only in too long a period to be really defined as medium, towards the desired level. Fair chance I think can be interpreted as 50/50, so leaving still a significant risk that a less favourable scenario would develop, in which the ECB could have to do more. This would become apparent, however, only a few months from now.


[***] Madalina Norocea provided research assistance

[1] Is the ECB really so generous?
[2] It is not even clear whether the possible purchase would apply to all ABS with private assets as underlying, so including Residential Mortgages Backed Securities (as in this quotation “the Eurosystem will consider purchasing simple and transparent asset-backed securities with underlying assets consisting of claims against the euro area non-financial private sector,” from the press conference, or only to ABS having loans to non financial firms as underlying, as one would derive from this other quotation ”On the ABS, we did not discuss the scope of the ABS, other than again reaffirming that it should be real economy-oriented, oriented towards non-financial companies of the private sector. ”)
[3] An enticing Very Long Term Refinancing Operation from the ECB
[4] Banks in Italy, Spain, Ireland, Greece, Portugal
[5] My guesses of where rates would be after the ECB package were published in a post on May 20th.
[6] An enticing Very Long Term Refinancing Operation from the ECB