Could the failed German bond sale be good news for the Euro?

File this one under "desperately looking for a silver lining in the unfolding disaster".

Suppose, just suppose, that all 17 Eurozone countries were identical. Suppose they were all at the Eurozone average. Suppose they all faced the same default risk and moderately high interest rates on their government bonds. What would the European Central Bank do then?

Is there any chance, if all the "why should country X pay to bail out country Y?" arguments were eliminated by supposing all 17 countries were identical, the ECB would finally step in and act as lender of last resort to all 17 governments? Not just by buying bonds when it felt it really needed to. But by making a very clear commitment to buy as many bonds as were needed to push bond yields down to some specified level?

I don't really understand how a bond auction can "fail". Does it mean there simply weren't enough bids to sell all the bonds? Or does it mean there weren't enough bids that met the reserve price? Or does it mean the Bundesbank felt the prices offered weren't high enough, and decided to buy a large part of the bonds itself? In any case, the people who seem to understand what it means are saying it was a very bad thing.

But could it be a good thing? If it means Germany is moving towards Greece, even if ever so slightly, then it might be a good thing. Because anything that puts all 17 countries in the same boat might just weaken one major argument against the ECB doing what it needs to do to try to save the Euro. What's wrong with a bailout, if even the Germans need bailing out too?

If the ECB does decide to act as lender of last resort to Eurozone governments, what interest rate should it charge? Here's one way to think about that question:

Would it make sense for the ECB to lend to banks at a lower rate of interest than it lends to Eurozone governments? If the banks were safer than the governments, it might make sense. But that assumption seems to me to be false. Eurozone banks hold a lot of government bonds. Eurozone banks are highly leveraged. The ECB loans to the banks are "protected" by government bonds held as collateral. If the bonds held as collateral go bad, the ECB loans to the banks go bad. And the banks will very probably go bad too. The part of the repo agreement where the banks promise to buy back the government bonds they "sold" to the ECB could never be fulfilled if the bonds go bad.

Put it another way. If the ECB thinks it makes sense to lend to a Eurozone bank at (say) 1% interest and a (say) 10% haircut on the government bonds used as collateral, why wouldn't it make sense for the ECB to offer exactly the same deal to the Eurozone governments directly? The government gets to borrow at 1%, and the 10% haircut means the ECB is protected from the first 10% of any default.

Does that make sense? (I'm not sure.)

Comments

[Max]

"But could it be a good thing? If it means Germany is moving towards Greece, even if ever so slightly, then it might be a good thing."

Or a bad thing, if it reinforces the austerity-everywhere mindset. It's like the S&P downgrade of the US - insignificant in itself, but becomes a political talking point.

I don't think there's any real danger of Germany losing its special position, which is not based on fundamentals, but on the fact that Germany can't go down without taking the entire Euro zone with it. It's too big to fail, hence safe.

[Doctor Why]

Markets are worried about three main risks:
1) Debt restructuring for GIIPS
2) The Eurozone break-up
3) And inflationary solution to the debt crisis

The ECB’s intervention will reduce the first two risks and increase the third, so the net effect on long-term nominal rates will be different for each country and may be positive or negative.

The problem here is that there are many possible inflationary scenario – real rates may go up or go down, and nominal rates may stay at their present level of go up by hundreds bp – and we don’t know which of them is more likely. For example, if the Eurozone is in a very bad liquidity trap (if the equilibrium long-term real rate is strongly negative), then long-term nominal rates will stay where they are despite higher inflation, and buying government debt will have a net positive effect. On the other hand, if the equilibrium real rate is close to zero, then the market is already very close to equilibrium, and a higher inflation target will mean higher long-term nominal rates. In this scenario the net macroeconomic effect of bond purchases may well be negative, at least in the short-term.

So, market expectation severely limit the ECB’s options, and if it is not prepared to directly target the entire yield curve, it may actually be better off doing nothing.

However, even if the ECB were willing to target the entire yield curve, I think it would still be a mistake to make it responsible for managing country spreads. Periphery countries do need reforms and below-market rates are a good way to reward good governance; and, from this perspective, it makes more sense to convert EFSF into a bank so that it can borrow directly from the ECB and buy government debt subject to explicit approval from Germany and other Eurozone countries. Moreover, a solution involving EFSF is less likely to provoke fears of an inflationary scenario, and may therefore work even within the current monetary policy framework (i.e. without targeting LT nominal rates).

[Mandos]

1) Debt restructuring for GIIPS 2) The Eurozone break-up 3) And inflationary solution to the debt crisis

ie, The Markets are worried about the entire gamut of potential solutions that don't involve mass starvation, pretty much. I can't think of a better case for jailing The Markets.

[rjs]

nick: "I don't really understand how a bond auction can "fail". "

greg ip:
http://www.economist.com/blogs/freeexchange/2011/11/german-bunds?fsrc=rss

[Frances Woolley]

Nick "desperately looking for a silver lining in the unfolding disaster" How about: you're not likely to run out of things to blog about any time soon?

[Phil Koop]

It's not too clear to outsiders what actually happened in that auction, so it's difficult to draw conclusions. Was the auction failure a signal that the buy-side has changed its mind about the riskiness of German bonds? Or is it a signal that the dealer banks are so strapped for funding that they don't want to participate ... which would be a sign that the last liquid repo market in Europe is breaking down. Either interpretation is bad news, of course, but news of very different character.

See http://ftalphaville.ft.com/blog/2011/11/23/759801/the-bund-that-broke-the-bundesbank/.

[bill woolsey]

Does added default risk on government bonds make interest rate targeting effective?

If there is loss of confidence in government's willingness to repay, and government's respond by cutting spending and raising taxes, can the central bank prevent this from reducing nominal expenditure by maintaining or even only modestly increasing its target for nominal expenditure?

[M.G. in Progress]

In fact it does not make sense. The fact is that eurobonds already exists inside the ECB budget. Repo agreements can not go on forever and sterilisation of bonds bought by ECB will have soon natural and economic limits. Thus the need to go as quickly as possible for eurobonds with EU memeber States collateral and joint guarantees. More than years ago I sketched out a proposal now being presented as well by the European Commission. Never say ever again...

[JP Koning]

"If it means Germany is moving towards Greece, even if ever so slightly, then it might be a good thing."

There was a terrible US bond auction back in February 2010, and look where long US bond prices are now. I wouldn't read too much into one bad German bond auction, although it would be worrying if a couple auctions failed.

[David Pearson]

Banks don't run deficits; sovereigns do. Temporarily providing liquidity to the banking system (the actual LOLR role) while it delevers is very different from financing net fiscal spending. First, politically it would be next to impossible to withdraw that financing given the impact on the economy. Second, the central bank has no authority over the direction of net fiscal spending, so it cannot reasonably argue that such an action is "temporary".

Deficit monetization is an easy policy to enter; it it is exceedingly difficult to exit.

[Nick Rowe]

Max: "Or a bad thing, if it reinforces the austerity-everywhere mindset."

Hmmm. I *hope* you are wrong.

Dr Why: "Markets are worried about three main risks:
1) Debt restructuring for GIIPS
2) The Eurozone break-up
3) And inflationary solution to the debt crisis

The ECB’s intervention will reduce the first two risks and increase the third, so the net effect on long-term nominal rates will be different for each country and may be positive or negative."

Normally, with a sensible central bank, and where monetary policy is currently right, an increase in expected inflation will mean an increase in nominal (and real) interest rates, because the bank will allow/ensure interest rates to rise to keep inflation on target. But if the ECB's policy is currently heading for deflation, a bit of inflationary pressure is just what is needed, to cancel out the disinflation.

Mandos: "I can't think of a better case for jailing The Markets."

Behind the "market" reification is a bunch of people (and their agents) trying to figure out where to hold their savings. Given current ECB monetary policy, that's not an easy decision.

rjs and Phil: thanks for those links. But I still can't really say I understand it. So I'm gonna just run with the "people who say they understand it say it's bad news" meme.

Bill: that brings us straight back to the same debate that me, you Scott, Paul Krugman, Brad DeLong, had about the US, a few months back. We never really resolved it.

MG: I don't quite follow you there. Sure, the ECB has been buying bonds. But on a piecemeal basis. There is no commitment that it will do so in future to hold yields down. So it hasn't had as much effect. It hasn't anchored expectations.

JP: But we have some fair idea that the Fed will buy US bonds if ever yields start to rise (unless the economy recovers). We don't know what the ECB will do.

David: the Bank of Canada hands over its profits to the Canadian government every year. So it has always financed part of government spending. My maintained assumption is that if the ECB loosened monetary policy enough to restore normal NGDP growth, and also fixed the liquidity problem, the 17 governments could be solvent. (Maybe not Greece). Their debt/GDP ratios aren't that high, compared to other countries that have much lower real interest rates.

[Sina Motamedi]

Makes sense to me.

[Ian Lippert]

Hasnt "ECB bailing out the european countries" always just been a euphemism for "Germany bailing out the european countries". How does this equation work once the only thing holding up the Euro tanks?

[Doctor Why]

Nick: "But if the ECB's policy is currently heading for deflation, a bit of inflationary pressure is just what is needed, to cancel out the disinflation."

With inflation at 3% and long term nominal (risk-free) rate at 2%, monetary policy seems to be about right to me, so an increase in expected inflation is likely to increase nominal rates. Therefore, buying more government debt may be counterproductive, as a decrease in country spreads may be more than offset by an increase in the risk-free rate.

[Patrick]

I'm no Euro expert, but if the ECB promised to buy unlimited quantities of member nations bonds to maintain yields at a specified target, how does that constitute Germany bailing out the GIIPS? Yes, the ECB might takes loses and Germany would be on the hook for some of those loses, in theory. But the risk is pretty low - if they step-in there's a fighting chance that everyone other than Greece can dig their way out. If they don't step-in, everyone is definitely hosed. Are Germans such a bunch of moralizing bigots that that they'd rather have a definite catastrophe rather than the slim prospect of the appearance of maybe perhaps picking-up part of the tab for those dirty lazy Latins? I thought they had evolved passed that once they gained readmitted to the human race.

BTW, the Germans didn't complain when the imbalances worked in their favor. They didn't moralize over the evils of exporting inflation, their banks making foolish loans, and generally pumping-up a gigantic bubble in the periphery. If they want to make it a morality play, then they ought to at least tell the whole story.

[Bob Smith]

Nick: "Their debt/GDP ratios aren't that high, compared to other countries that have much lower real interest rates."

Case in point: "http://business.financialpost.com/2011/11/24/uk-yields-fall-below-germany/"

Dr. Why.

The fact that bond markets are spooked by Euro debt, but not UK debt, suggests that the markets aren't as worried about inflationary solutions to the debt crisis, i.e., risk 3, as that's at least as likely, if not much more so, in the UK as it is for the Euro, so much as actual default (with Germany left to pick up the pieces), i.e., risks 1 and 2. I have a vague recollection of a recent piece in the Economist about the apparently paradoxical willingness of investors to have their investment inflated away rather than defaulted on.

[Nick Rowe]

I really should have made this two posts. Because the bit at the end of my post seems separate.

Let me repeat it, by asking 3 questions:

1. Which is more risky: Eurozone governments or Eurozone banks? I would say the banks are riskier. Either both fail together, or the governments let the banks fail. I can't see the governments failing and the banks staying solvent. Agreed?

2. Assuming 1 is true, is the ECB lending to banks at a lower rate of interest and easier terms than it is lending to governments? I think it is. Agreed?

3. Assuming 2 is true, what the ECB is doing just doesn't make sense to me, on any grounds. Agreed?

[Sina Motamedi]

^^
Nick, I completely agree with 1-3. I don't see how anyone could not.

[Nick Rowe]

Sina: thanks. Anyone else? I should post this, as a separate post. But I'm scared I'm missing something big and obvious. So I'm looking for reassurance that I might be on the right track.

[Doctor Why]

I think 2 is not true since loans to banks are collateralized by government debt and the ECB applies a valuation haircut.

[TMDB (123)]

#1 is probably false in Greece
#2 is false in Germany ( 3 month German yield is 0.25%, ECB refi rate is 1.25% )
#3 makes sense on the grounds of public choice, as ECB imposes haircuts on the market value of collateral

[Sina Motamedi]

@TMDB
Yes but the ECB doesn't provide 0.25'% to Germany, Germany gets that from somewhere else. The ECB does provide 1.25% lending to banks, but nothing to sovereigns.

@Doctor Why
ECB may haircut the collateral, but it's still providing a lower rate to what it lends. I think that's the key.

[Nick Rowe]

Doctor Why and TMDB: OK, suppose I'm a Eurozone government. I write "IOU E100" on a bit of paper, and the ECB lends me E80 (20% haircut) at 1.25%, on the understanding that it will give me back my bit of paper when I repay the loan. Unless I'm Germany, I would jump at that chance. But that's the same deal the banks get, when they repo government bonds.

[Bob Smith]

Dr. Why: "#1 is probably false in Greece"

What, the Greeks can't tax the bejesus out of their own banks? (On further thought, given the (in)efficiency of Greek government, you may be right).

[TMDB (123)]

Nick,

Yes this is what happens. But the end result is 7.3% yield on 2 year Italian IOUs.

[Jacques René Giguère]

Bob Smith: being inflatesd away is gradual and can be shared by selling your bond to someone else willing . Default is absolute and can't be shared.

[Doctor Why]

To clarify:

From a purely technical perspective, lending to a bank is safer simply because the bank provides extra credit support. True, this extra credit support may not be very useful in some extreme cases, but it does protect the ECB against some events.

From a more general perspective, monetary policy works by setting interest rates for certain classes of financial assets and you basically argue that government securities should be on this list. Maybe they should, but that's not how the ECB was designed, and fortunately there are other ways to reduce the average yield on the Eurozone debt (eurozone bonds, EFSF, IMF etc.) which may be easier to implement.

[Prem]

NR: "Put it another way. If the ECB thinks it makes sense to lend to a Eurozone bank at (say) 1% interest and a (say) 10% haircut on the government bonds used as collateral, why wouldn't it make sense for the ECB to offer exactly the same deal to the Eurozone governments directly? The government gets to borrow at 1%, and the 10% haircut means the ECB is protected from the first 10% of any default.

Does that make sense? (I'm not sure.)"

It does not make sense unless you are implying that the 11% extra collateral available to the ECB in a repo arrangement is essentially worthless in a sovereign debt default. The ECB's risk of loss is the intersection of (1) the probability that the sovereign will default with zero recovery and (2) that the bank will also simultaneously default with zero recovery. This is not the same as lending to the sovereign at the same with the same collateral hair cut.

In any case, how exactly can the ECB lend to the sovereign at a 1% interest rate but with a 10% haircut? if it buys the bond from the government at a 10% discount, it simply means that the yield-to-maturity of the bond is greater than the stated coupon, and there is no loss cushion made available in the form of a third pary bank's equity.

[Bob Smith]

Jacques: "being inflatesd away is gradual and can be shared by selling your bond to someone else willing . Default is absolute and can't be shared"

Unexpected inflation can be shifted to someone willing to sell, but I wouldn't expect that current holders of debt have better knowlege about future inflation that current would-be buyers of debt. If higher inflation is expected, that would presumably be incorporated into pricing (just as a default would be incorporated into pricing). I'm not sure what the difference is between being told (i) that you'll get 50 cents on the dollar at maturity and (ii) that you'll get a dollar that's worth 50 cents in current dollars at maturity (subject to the obvious point that there is a difference to the extent there is a coupon on the debt). In either case, I'd expect you'd take a similar haircut if you tried to sell the debt to a third party.

That said, I suppose that, with inflation, there may be buyers with more optimistic assessments of future inflation (whereas default, as you note, is fixed), and so it may be easier to offload debt subject to inflation than debts subject to default (sort of a "greater sucker" theory of bond markets). The other possibility is that default and inflation are not exclusive and that while the UK may whittle away its debt with inflation, once Greece (for example) partially defaults on its debts, it might whittle away the remaining balance with inflation (either as a result of inflationary monetary policy at the Euro level, or from by going back to the Drachma). Certainly it wouldn't be hard to believe that a country that can't manage it's fiscal policy might also not be able to manage its monetary policy.

[Robillard]

Free exchange had a description of the bund sales process from a bund trader:
http://www.economist.com/blogs/freeexchange/2011/11/german-bunds

[Nick Rowe]

Prem; "The ECB's risk of loss is the intersection of (1) the probability that the sovereign will default with zero recovery and (2) that the bank will also simultaneously default with zero recovery. This is not the same as lending to the sovereign at the same with the same collateral hair cut."

Good point. If I guarantee my daughter's loan, that increases its safety unless my solvency is perfectly correlated with my daughter's. If I guarantee my own loan, my own solvency is perfectly correlated with itself, of course, so there's no point in doing that.

But the haircut still has some effect, I think. If I offer one of my own bonds as collateral, and the lender takes a 50% haircut, then if I default but still pay 50 cents on the dollar (or more) the lender comes out OK. It's like being ahead in the queue of creditors, sort of, I think.

[Jacques René Giguère]

Bob Smith: totally agree on your third paragraph.
As for the "greater sucker" theory, it is the basis for all markets. You buy because you think it's worth more than what I think when I sell.. In most physical markets, it is obvious that the carrots are worth less to the farmer than to the cook and thwe trade makes sense.
In many finacial transactions, both sides think of the other as the sucker.

[Determinant]

Here is a good piece from the New York Fed about how its bond auctions work: http://newyorkfed.org/aboutthefed/fedpoint/fed41.html

The government goes into a debt auction with a financing requirement. Those bonds are going into the Consolidated Revenue Fund to pay for spending.

When a bond auction fails, the amount of money bid by dealers is less than the government's financing requirement. The government didn't get all the money it needs out of this auction.

In Germany's case where the Debt Agency often keeps a percentage of bonds for itself and sells them later on the secondary market, this can be made up quickly. But the fact is that right now the Government of Germany has less money than it budgeted for in its coffers. This may change in two days or two weeks, but that is what happened.

Note that in the NY Fed's case competitive bond auctions are accepted by discount rate, lowest rate first. Changing the discount rate is not going to elicit new bids and more money immediately, it's not just a bookkeeping change.

Bond auctions fail every few years in advanced markets. It often has as much to do with the state of market players as it does with the government. If the next auction fails or if a Dutch or Austrian auction fails, then it's time to really worry.

[K]

Nick,

If you default with 50% recovery, the lender keeps the collateral and then claims the 50% loss in the bankruptcy court. They'll recover 50% of that claim so you end up with effective 75% recovery. In general a loan collateralized by your own credit loses L^2 on default where L is the fractional loss on default. This all assumes that the bankruptcy court will recognize such an arrangement as a proper collateralized loan. I've never seen this done, so I don't know if there's precedent, and it would depend on the jurisdiction, etc. It's almost explicitly fishy, so you might be taking your chances in court.

[Nick Rowe]

Determinant, K: Aha! Thanks.

[K]

I was ignoring the haircut. (And when I say haircut I mean in the usual sense of the amount of overcollateralization of the loan.) If the $100 loan is backed by $200 of bond then there would be zero loss.

Effectively what you are doing is introducing senior debt. Corporate entities do this sometimes if they are in trouble and their bond indentures don't explicitly prevent it. If the ECB wants to provide senior financing to countries I don't know if there's any law specifically preventing it, so it might not need to involve collateral. It *will* cause bond spreads to blow up though as they would be getting in back of the line behind the ECB.

[Andrew F]

"OK, suppose I'm a Eurozone government. I write "IOU E100" on a bit of paper, and the ECB lends me E80 (20% haircut) at 1.25%, on the understanding that it will give me back my bit of paper when I repay the loan. Unless I'm Germany, I would jump at that chance. But that's the same deal the banks get, when they repo government bonds."

So why don't these countries do that? Can't a country put some money in a new bank, lever that up buying their own debt, and repo it back to the ECB? Why can't a country just create banks to take advantage of powers that banks have and they do not?

[K]

Oh, yeah. One more thing. "Insolvent trading" is if an entity enters into an arrangement that depletes its assets in order to put off default when that entity knows full well that it's going to default. This robs current creditors and its *illegal*. So if I borrow $100 by providing $200 of collateral when I'm about to go bust, a court may reverse the transaction (and throw me in jail). If you are going to subordinate current creditors you'd better be sure you are allowed to do it.

[Determinant]

Andrew F:

The initial yield, the real rate the government is paying, is not 1.25% then. The rate is adjusted with 80 euros and the principal, not 100.

In your case the yield diverges from the coupon rate. Normally this divergence is small but these are not normal times.

Ben Graham's classic "The Intelligent Investor" is a wonderful introduction to basic bond trading. The man loved bonds and it shows in his writing. He started his career in the railroad paper and bond market. Until the 1940's that was the majority of the US corporate debt market, actually. US Railroads went bankrupt with regularity and Graham was a master at the credit evaluation of railroads.

Graham shares a few of his rules in his book which is all the more entertaining when he turns his eyes on Penn Central, the combination of the Pennsylvania Railroad and New York Central which went bankrupt in 1970. That bankruptcy led to many changes in the industry and created the healthy North American rail sector we have today.

[K]

Andrew F: "Why can't a country just create banks to take advantage of powers that banks have and they do not?"

They could. But then bond holders would see that they are about to get stuffed and yields would blow up. And then the offending country will not borrow another penny in the bond markets. And then the gig is up.

[vjk]

JP:

"There was a terrible US bond auction back in February 2010"

What do you mean terrible ? The February 2010 auction was not as spectacular as usual, but still as far as I recall, bids exceeded acceptance about 2 times -- the US Treasury sold all it wanted at the action. I would not call that a failure.

Germans, on the other hand, sold only half of what they wanted, unless something is lost in translation/reporting -- that's clearly a failure.

I am not aware of a single failed US bond auction in the last 10 years in the sense of the treasury having been unable to sell as much as wanted. Are you ?

[vjk]

Nick:

ECB cannot be the LLR to anyone, neither banks nor governments, by design under the current legal framework.

The design can be changed of course, given the actors desire to do so is strong enough, just as the FRA can be changed by the US Congress. After all, both are human institutions, not something like physical constants.

[vjk]

Nick:

"the Bank of Canada hands over its profits to the Canadian government every year. So it has always financed part of government spending. "

Unless I am mistaken re. Canadian arrangements, that is an odd kind of "profit". The profit is the interest that the government pays to the Central bank on the government issued security holdings that the CB remits back to the government in a manner similar to what the feds do with US treasury interest, minus the CB operating expenses. Thus, the "profit" is the interest the government would have to pay to a non-CB security holder, but now it gets part of that interest back due to the fact that the security holder is the CB.

[Nick Rowe]

K: "It *will* cause bond spreads to blow up though as they would be getting in back of the line behind the ECB."

Interesting point. Not sure if it's right or wrong. Will have to think about it.

I think it's *maybe* wrong. Even if the new lender is first in line ahead of me, my loan will be safer if the new loan fixes a liquidity problem, and also increases NGDP and fixes a solvency problem.

vjk: "Unless I am mistaken re. Canadian arrangements,..."

You are not mistaken. But it's exactly the same as the BoC financing part of government spending by printing money.

[vjk]

Nick:

"You are not mistaken. But it's exactly the same as the BoC financing part of government spending by printing money."

Don't know.

In normal circumstances that, as far as I know, still exist in Canada, the bond shuffling operations are conducted by the CB to maintain the overnight interbank interest rate. So, the original bond travels from the government to the public to the Central bank to the public, etc. Note, the CB never buys directly from the government. Only if it did so, would it be direct financing of the government as traditionally understood.

One can claim I guess that the fact that the government pays less interest during the period the bond is held by the CB amounts to financing the government, but I'd imagine this way of putting things unnecessarily obscures the simple fact that the government simply pays less interest on its debt.

Likewise, one could claim that the bank finances the mortgagor when the latter refinances the loan, not that the mortgagor simply pays less interest to the mortgagee.

[Nick Rowe]

vjk: Think about it this way. $100 bond, 5% interest.

Governemt sells bond to public, which sells it to the BoC. The BoC prints $100, which it gives to the public, which gives it to the government. Then every year after that the govt gives the BoC $5, which the BoC gives right back to the govt.

It's exactly the same as if the BoC prints $100 and gives it to the govt.

The only difference is that, if it wants to, the BoC can sell the bond to the public and reduce the money supply, without requiring the govt to increase taxes immediately by $100.

[the_iron_troll]

We live in terrifying times.

[vjk]

Nick:

"It's exactly the same as if the BoC prints $100 and gives it to the govt."

It is a very simplistic and therefore incorrect description. You present a scenario in which the public is a mere mechanical conduit of the CB base money to the government. In fact, as your own article shows, the public should be willing to buy the government debt and that's what gives that debt any value, by design, public willingness to buy it. After the public blessed the government pieces of paper, the act of swapping base money for bond is immaterial as it is an exchange of two assets of roughly the value as Bernanke and friends discover to their chagrin with their QE exercises (there may be some secondary effects in QE swaps influencing commodities and stocks).

Additionally, in normal times, even assuming your description is correct,, the volume of repo operations using government securities as general collateral for one or two weeks typically, is so insignificant that in no way can it be considered as "financing" the government.

[vjk]

"two assets of roughly the value" should read "two assets of roughly the same value"

[Sergei]

People seem to draw too many conclusions about this auction. Lets look at options that eurobanks face:

1. Buy German 10y bond at 1.98%
2. Buy ECB 1w deposits at 1.25%

After a politically imposed 50% haircut on Greek bonds, you have to be dead stupid to think that all your risks over the next 10 years are worth about 75bps.

Honestly, no surprise Germans were not able to find enough idiots to play this game.

There is no more to this story than this. Next month ECB cuts the rate and there will be more buyers for the risk in eurozone even at 1.98.

Monetary policy rate and government bonds are and will always be a simple arbitrage for banks.