On Monday, we highlighted the following question posed by David Einhorn to Mario Draghi: “Mario Draghi says he sees no sign of a bubble in the sovereign debt market, which raises the question: what does Mr. Draghi think a bubble in sovereign debt might look like that isn’t already evident?”
Late last week, we echoed this sentiment when we remarked that “just because you don’t yet own all of something doesn’t mean you are not making something scarce kind of like just because you can still get water out of the faucet in California doesn’t mean there’s not a drought, and if one key indicator of scarcity in credit markets is the degree to which certain issues trade special then Draghi is either ignoring the ‘evidence’ or he doesn’t know what evidence of scarcity looks like.” Then this morning, Bill Gross offered his two cents, calling 10-year Bunds “the short of a lifetime” (will Mario Draghi dare anyone to put the trade on?).
Given all of the skepticism, it seems logical to ask the following: “who is still buying Bunds at these levels besides the ECB?” Here’s Citi with more:
The most frequent client question is not based around economics or any market technicals but simply, ‘who is buying Bunds at these levels’. The question usually comes with a look of incredulity. Other than the ECB buying, Figure 1 shows the net buying by investor classification, based on Citi’s flows, which we think are representative of the market. The data shows that banks and asset managers have been the biggest buyers in the most weeks. Hedge funds had been short (perhaps erroneously positioned for a Fed QE style reaction) while pension and insurance industry flow has been more stable. Netting out all investors, German government bond flows are positive (more buying than selling), excluding central banks. Why? As we have remarked before, banks are buyers and likely to extend duration, while asset managers will be funding it costly to be short Bunds given the performance, even in global funds which tend to have more room to short core European bonds.
As Citi goes on to note, the expansion of PSPP-eligible assets did not include any new German issues, meaning that the Bundesbank’s job (i.e. hitting its monthly targets as per the capital key) isn’t about to get any easier and the scarcity that Mario Draghi says doesn’t exist isn’t likely to dissipate.
The headline here is that there were no changes to the existing German QE pool. The Bundesbank clearly decided against taking more risk in FMSWER purchases for example, which remain off the eligible PSPP security list. That continues the squeeze on Bunds.
And as we’ve noted on too many occasions to count, the very design of Q€ makes its full implementation questionable because there’s a certain degree to which it becomes a victim of its own success — that is, the more successful the ECB is at driving down yields, the less assets there are to buy because the program doesn’t allow for purchases below the depo rate. Put differently, the depo rate floor and the program’s goal aren’t entirely compatible...
The market is exercised by the risk of QE execution failure in some core market simply because of the ECB QE design flaws. Conceptually, there is a problem in setting a quantity as well as a yield floor (-20bp). In practical terms, what concerns the market is that as the market rallies, the ECB is forced to buy further out on the curve, as more and more paper trades below -20bp. Figure 5 shows that without a yield floor the current market value of government bonds available to buy would be over €902bn.
Citi sees the ECB having to go out to 7-years on the Bund curve in the space of just six weeks in order to stay above the depo rate assuming current trends hold...
Based on current yield trends, the buying will have to start at the 7y point in just 6-weeks, and will take the eligible pool of paper available down to €516bn on our estimates. With a 25% issue limit that means a maximum size of €129bn is possible, and that would make it a close race as to whether the ECB can actually buy the near €144bn total QE volume that we expect in German sovereigns.
In our view, the September meeting is likely to be key rendezvous point for discussing technical difficulties in some core countries on the available pool of paper to complete the programme. This would follow July where there is a huge negative supply of -€146bn, measured as issuance minus coupons, redemptions and ECB buying.
And the ECB will have to make “adjustments”, because if they do not, they risk a “forced” taper and the attendant “tantrum” and it would not look good for Mario Draghi — who has made it very clear that he intends to see PSPP implemented in full — to have to later admit that the ECB will in fact not be able to carry out the program as promised. As we’ve discussed previously, the central bank’s options for freeing up more purchasable assets include raising the issue limit on non-CAAC bonds to 50%, buying corporate bonds, lowering the depo rate, or removing the floor altogether. Here’s Citi again:
Opening up the limit on bonds would help on the technical issue of not having enough paper to buy to complete the programme and we have noted in the past that there is no obvious reason why the ECB limit on non-CAC paper can not be higher than 25%. For instance, if the ECB bought non-CAC paper to a 50% limit, this would increase the eligible paper in 2y to 30y that is not trading below -20bp, by €294bn in Germany, compared to €176bn eligible paper under current rules.
Buy corporate debt: The idea is to keep the sanctity of the €60bn monthly purchase number to get the balance sheet to right-size. – The problem is that corporate spreads are also tight and not representative of euro area funding, but nevertheless, this could be a fall back option.
Note also here that buying corporate debt may be inevitable anyway because, as we discussed last week, the further below zero the risk-free rate falls, the wider spreads to corporate credit become, leading to mispriced risk, something the ECB may need to step in and "correct" (and by "correct" we mean drive rates on corporate bonds below zero in a complete perversion of credit markets).
Cut the deposit rate: Draghi has ruled this out but one argument often heard is that cutting the deposit rate would open up room for more buying as the new yield floor would be at a lower deposit rate. – We see this argument as weak, primarily because we expect yields to collapse to the new depo rate, given that banks will continue to treat high quality paper as fungible with central bank reserves. Moreover, the larger tax on the banking system probably infers even more yield seeking behavior.
Remove the -20bp limit: Simply removing the -20bp limit would immediately address the issue. – This is very unlikely. The rationale for imposing the -20bp floor to purchases is the dark side of the risk segregation in that losses as well as profits are kept locally. The Bundesbank for instance would be guaranteed a greater loss on purchases assuming the -20bp deposit rate is not a permanent feature. That is, the Bundesbank’s share of profits from the negative rate to banks, would not offset portfolio losses.
To sum up the above, 30% of German debt trades at or below the depo rate and some 60% carries a negative yield. The way things are going now, central bank Bund purchases will have to be in maturities of 7 years or more within just 6 weeks, and of course that timeframe could accelerate meaningfully should things take a turn for the worst in Athens.
Ultimately, the math doesn't add up and it appears as though modifications to PSPP's structure will be necessary (perhaps at the ECB's September meeting) in order to prevent a forced taper. Program adjustments may include depo rate cuts, an elimination of the depo rate floor, or, in the most likely scenario, corporate bond purchases. Meanwhile, the Bund curve continues to flatten as yields converge on -0.20% and, as we noted earlier today, "far from being the short of a lifetime right now, Bunds are in fact quite the opposite ... because if the ECB will have no choice but to buy even more Bunds from the private market, then the sellers can demand any prices for these Bunds, up to and including the ECB's hard (for now) floor of -0.20%"