There are a bunch of things in the ECB post-mortem note just released by SocGen's Michel Martinez, reproduced below, but here are the punchlines.
First, on the impact of ECB QE on the economy: "we argue ECB QE could be five times less efficient than in the US. In December, press reports suggested that the ECB had run studies suggesting that a €1000bn QE programme would only boost price levels by 0.2-0.8 after two years, five to nine times less efficient than the studies for the US or the UK. The impact on GDP is not provided, but it would be reasonable to assume the same impact as on inflation on a cumulated basis."
In other words, it will be an outright failure as it "triest" to boost inflation expectations and the European economy in its current format. That, as a reminder, is its stated purpose.
So what does SocGen suggest? Simple: the same thing every Keynesian says when justifying why a piece of occult economic voodoo fails to work: it wasn't big enough. To wit:
"The potential amount of QE needed is €2-3 trillion! Hence for inflation to reach close to a 2.0% threshold medium term, the potential amount of asset purchases needed is €2-3tn, not a mere €1tn."
Or as Charles Dickens would put it:
And since there is nowhere near enough bond supply in Europe, the ECB will have to proceed with monetizing, drumroll, stocks.
Should the ECB target such an expansion of its balance sheet, it would have to ease some conditions on its bond purchases (liquidity rule, quality...) or contemplate other asset classes- equity stocks, Real Estate Investment Trust-(REIT), Exchange-traded fund (ETF)...- as the BoJ, previously.
Because what tens of millions of unemployed Europeans really need to help their lot in life, and to boost their confidence, is for the central bank to buy the stocks sold by the richest 0.001%.
* * *
Full note from SocGen:
Large QE with symbolic risk sharing
The ECB announcement today was well in line with our call that the sovereign QE programme could be large scale but not pari passu. The share of mutualisation is symbolic (1/11%). Yet, they point to higher volume than expected overall (€1140n in total) and reinforces the probability that the ECB would reach (or go over) its stated intention of 1trn increase in balance sheet.
A) The size of QE programme is €60bn per month, €1140bn in total
The main measure is an expandable asset purchase programme that includes European agencies and sovereign and complement the current programmes (Covered bond and ABS purchase programmes). Those new programmes will start in March 2015 and run until end September-16 or until the inflation outlook converges to 2.0% medium-term, which means that it could be bigger. The combined purchases will amount to €60bn per month. Apparently, the expanded programme will not include corporate bonds.
There was a large majority to today’s announcement while the Governing Council was unanimous on the idea that QE is a legal monetary policy tool.
Hence, the ECB will purchase €1140bn (60*19) from March 15 to September 2016. Today, the pace of covered bond and ABS purchases is close to €13bn per month, so additional purchases represent €47bn per month. The ECB stated that “purchases of securities of European institutions will be 12% of the additional asset purchases”. A quick rule of thumb suggests €230bn in ABS and covered bonds, €110bn in European agencies and €800bn of sovereign bonds.
The ECB also decided to cut the spread on the TLTRO rate, that would now be equal to the refinancing rate (0.05% instead of 0.15%)
B) Criteria to be specified in March
We know that the new programmes are running until September-2016 at least and that purchased bonds will be hold up to maturity. Obviously, purchases will be made on the secondary market for European issues and government bonds.
In terms of rating, it is likely that the conditions will be the same as for the ABSPP and CBPP3. First, the ECB will purchase investment grade bonds.
Secondly, for Greece and Cyprus (which are not investment grade), the condition would be that those countries “have an ongoing programme with the EU/IMF”. This would suggest that any failure to extend the current Greek programme that expires at the end of February would exclude Greece from any asset purchase programme.
Can the ECB buy at negative yield? Yes, said Draghi during the press conference
Which maturity? Details will have to be specified in March but Draghi suggested that maturities could be of 2-30 years.
Interestingly, Draghi said that it will have two limits on its purchases: 30% of the issuer outstanding and 25% of the issue. This latter limit will prevent the ECB of having a blocking minority in CACs, with the aim “to ensure that the ECB is pari passu”. As we argue below, this is not convincing, given the low degree of risk sharing.
C) Minimum symbolic risk sharing (8% risk sharing only on government bonds
As we expected, purchases of government bonds will be done according to capital key. The point is the degree of mutualisation is minimum(1/11). For most purchases, the National Central Banks (NCB) will purchase their national government bonds and bear the credit risk. The silver lining is that the less the credit risk is mutualised, the larger a QE program could become in volume terms.
The main piece of information in the ECB communiqué is here: “With regard to the sharing of hypothetical losses, the Governing Council decided that purchases of securities of European institutions (which will be 12% of the additional asset purchases, and which will be purchased by NCBs) will be subject to loss sharing. The rest of the NCBs’ additional asset purchases will not be subject to loss sharing. The ECB will hold 8% of the additional asset purchases. This implies that 20% of the additional asset purchases will be subject to a regime of risk sharing”.
So the ECB indicates that the degree of risk –sharing is 20%, which seems a good compromises. However, regarding government bonds, the decree of risk sharing seem symbolic(8/100-12=1/11). To put is simply, the ECB will purchase €70bn on a risk sharing basis while the NCB will purchase the remaining €730bn. Noteworthy, the ECB will likely give objectives to the NCBs (purchases according to the capital keys).
This approach is consistent with our long held view that the ECB QE could not be both large and pari passu. Legal and political hurdles remain large because of the two articles of the European Treaty: Article 123 on prohibition of monetary financing and Article 125 (no bailout clause or no mutualisation clause). The ECB might well be pari-passu ex ante as Draghi argues. Yet, in the case of a debt restructuring, either the CB would avoid the debt restructuring (remind that the Eurosystem avoided the Greek PSI in 2012) or the (bankrupt) national government would probably be obliged to recapitalize its NCB. In both cases, the bigger are the purchases, the larger is the expected loss given default of the private sector. Investors risk seeking such way of proceeding as a lack of confidence in the euro area. Hence the final outcome on sovereign bond spreads might be uncertain debt sustainability concerns increase in the future. The flow of purchases will be positive but lower liquidity and higher expected loss given default will play out negatively.
Will it work?
In a joint paper with rates strategists (What kind of ECB sovereign and what impact?, we argue ECB QE could be five times less efficient than in the US. In December, press reports suggested that the ECB had run studies suggesting that a €1000bn QE programme would only boost price levels by 0.2-0.8 after two years, five to nine times less efficient than the studies for the US or the UK. The impact on GDP is not provided, but it would be reasonable to assume the same impact as on inflation on a cumulated basis. These figures are consistent with our own estimates.
The potential amount of QE needed is €2-3 trillion! Hence for inflation to reach close to a 2.0% threshold medium term, the potential amount of asset purchases needed is €2-3tn, not a mere €1tn. Should the ECB target such an expansion of its balance sheet, it would have to ease some conditions on its bond purchases (liquidity rule, quality...) or contemplate other asset classes- equity stocks, Real Estate Investment Trust-(REIT), Exchange-traded fund (ETF)...- as the BoJ, previously.
So the onus will remain on delivery of better-designed fiscal policy and structural reform. But it is difficult to be hopeful on these fronts.
* * *
Of course, this means that the time to frontrun the expansion of ECB QE 1 has begun. The only problem is that for Draghi to act, stocks will have to crash first, and they can't crash if they are frontrunning the event the follows from their crash.
Good luck figuring that one out.
