Just a hunch now, but Italian, Spanish, and related bond yields began falling coincident with the first ECB LTRO. The question is why, as I saw no operative channel of consequence from ECB liquidity provision of 3 year funds on a floating rate basis to the term structure of rates.
So it seemed to me that also coincident to the LTRO was some entity giving the nod to its banks to buy those bonds, or some reason sellers of those bonds backed off.
I’m now thinking it may have been the BOJ giving the nod to its member banks to buy euro member debt denominated in euro and keep the fx risk on their books, with the assurance govt policy would keep the yen weak and guarantee the banks an fx profit.
We learned after the fact that Japan had been selling yen well before they announced their new weak yen stance. And having their banks buy euro member euro denominated debt directly weakens the yen vs the euro.
The timing of the events- the LTRO/yen sell off/yen policy change- is close enough to get my attention.
So Japan managed to weaken the yen and firm euro member debt prices all under the cover of the ECB LTRO operation which they gladly allowed to take the credit.
In any case, I don’t expect any more from this next LTRO than I expected from the last, but I am keeping a close eye on the yen.
Warren – have you commented before on the similarity between the ECB’s LTRO and how you would see the banking system working under your “Proposals for the Banking System”?
The LTRO effectively removes market discipline from [a part of] the liability side of the balance sheet.
If Greece does leave the EZ, this could cause deposit flight to German banks from all across the EZ. The only way to stop this, either ahead of Greece leaving or once it has already left, would be to ensure that lost deposits would be replaced by ECB funding.
So maybe Europe ends up with some of your proposals by the back door??
yes, already happening with ecb funding as you state.
except the ecb is trying to exit being the funder…
ECB runs its interest rate policy via overdrafts against collateral. As banks draw on overdrafts and post collateral, they need to replenish their unencumbered collateral / liquidity buffers. It does not mean they will buy govies as collateral rules have been extended but there is strong feeling that they actually do.
sarko trade? for a eurozone bank to buy its own sovereign is like a free option. upside: the carry trade (maturity disparities and variable rate notwithstanding it’s pretty hard to cock up) and political brownie points (nationalisation risk if they don’t do it) vs. downside: none, as they’re toast anyway if the sovereign fails, whether or not they do this, so zero incremental risk.
remember the EZ bond market is 25 people (banks outside the top 25 are too small to matter) and they were all in the same room with the ECB guys just before they decided the LTRO.
true
@CiG,
I read somebody’s quote today (bloomberg article) that in effect ECB is “subcontracting” banks to buy government debt.
That could be a possibility – that they just agreed to give cheap loans to banks in exchange to them buying government bonds. That way ECB can adhere to its ideology of non-intervention, and yields on bonds can still be down.
I wonder – if that is the case – then how long can it last.
http://mobile.bloomberg.com/news/2012-02-27/draghi-may-have-shelved-ecb-s-bond-purchasing-program-as-loans-fuel-rally
the banks are still at risk of loss, so it only goes so far
@WARREN MOSLER,
Exactly. And they were always able to do it…the MRO operates on full allotment for a long time. Since the new collateral requirements were not in effect in the first LTRO, the only difference is duration. But that doesn’t seem too much of a risk, since if financial instability continues, the ECB will keep operating on full allotment.
yes, agreed
A bank can finance a bond purchase through an MRO or a 1m/3m/1Y/3Y LTRO. Functionally they are quite the same. They all pay variable rate (set by the ECB through its MRO’s, the 3Y-LTRO will not pay 1% for 3 years but the rates of the MRO’s during its term) and they all have the same risk management (if an Italian bond drops from 95 to 85, the bank must post more collateral). The only real differences are:
1) Interest payment happens at maturity. Therefore a 1Y/3Y LTRO makes carry trade easier since the bank will earn the interest rate on the bond it purchased before having to pay ECB.
2) There’s probably no collateral substitution (although the ECB documentation is not clear), meaning that you have to post collateral with more than 3 years maturity on the LTRO (Warren do you know anything more?).
I was equally puzzled by the drop in periphery bond prices after the LTRO. My assumption was that it was due to balance sheet effects. Periphery banks had a huge rollover problem for their liabilities (bank bonds) maturing in 2012. The ECB removed that by essentially providing a 3 year term loan (hence the relaxed collateral rules) making the banks ‘solvent’ again.
I always thought that you actually need a buyer of last resort for bond prices to fall (like Fed was for MBS’s in QE1). Warren’s explanation seems quite reasonable. Warren please do post any data whenever you get a hold of them.
@Kostas Kalevras,
Hadn’t seen your reply before I posted a response…I totally agree with your analysis.
Maybe its not appreciated that China is behind the Japanese funding moves following China agreeing to buy Japanese debt. This is a standard move under Chinese strategy to acquire global interests.
There has been talk in Japan for at least two months from some reliable sources that the yen would weaken and the Japan interest rate would go to 1%. I have noticed the yen/euro/usd connection since last August, and they seemed to be in a deadly embrace. I knew at some point that somebody would have to give. Looks like the BOJ has found a way out. For now…
Warren (in case you read comments of old posts) there ‘re some interesting numbers in the latest ECB monetary statistics:
http://www.ecb.int/stats/money/aggregates/bsheets/html/outstanding_amounts_A30.A.U2.2100.en.html
http://www.ecb.europa.eu/press/pdf/md/md1201.pdf
MFI Assets – General Government securities increased from 1381,8 b€ on Nov to 1448,1 b€ on Jan, a change of 66 b€. For Spain the number was 51,7 Italy 32,6 and Ireland 4,3. Portugal was unchanged.
But in the monetary developments statistics, new loans to government (in the form of securities) were 58 and 35b€ for Dec and Jan, for a total of 93b€.
Unless i am not reading the numbers correctly, this means that European MFI’s bought 93b€ of new European government debt in Dec-Jan but only kept 66 billion on their balance sheets. That does not look like full scale carry trade government bond buying from banks, nor can it explain the fall in yields. Just my 0.02€.
thanks, as you suggest, we’re still in the dark as to what’s actually going on