The widespread feeling after the November ECB meeting was that we should prepare for the beginning of the central bank’s exit strategy possibly, at least in terms of announcements, already at the December Governing Council (GC) meeting. The rationale for this is that on November 13 we will have the final confirmation that in 3Q 2009 the euro area emerged from the recession and we keep witnessing signals of stabilization in financial markets, with figures on 3Q earnings by major European banks being the most visible sign.
Although the ECB is under no pressure to hike the refi rate – the growth/inflation outlook and our Taylor Rule suggest that aggregate conditions do NOT call for a classic tightening anytime soon – the central bank is increasingly feeling the need to regain control of market interest rates via a progressive mopping up of excess liquidity. What we have seen so far is a swift and bold ECB reaction to deteriorating conditions in the financial and macroeconomic landscapes since October 2008 that triggered a significant reduction in money market interest rates. In addition, and this is something that is underestimated in comments on recent ECB policy, the unlimited cheap financing to banks has caused a massive rebalancing in banks’ balance sheets in favor of government bonds, helped also by European Commission statements that no single country would have been left to its own destiny. As a result, ECB figures confirm that banks’ net asset purchases of government bonds more than tripled and now stand at EUR 150bn: banks enjoyed a positive carry which was much needed for their impaired bottom lines, and governments were able to finance the heaviest supply in history which allowed them to undertake a significant (though not dramatic) fiscal easing. Something for everyone. Faced with a stabilizing macro and financial picture, the question is: what’s next?
What to expect in December and afterwards
The starting point for the ECB when they meet in December is a total liquidity outstanding of about EUR 686bn, with an excess surplus of about EUR 100bn. This surplus has been constantly declining since last June’s 12-month LTRO, but it is likely to increase again after the December operation. Let’s be clear from the beginning: the low September allotment (circa EUR 72bn) shouldn’t fool anybody. With the ECB already announcing that this is going to be the last auction at that maturity and with all the liquidity issues and balance sheet dressing needs which arise at the turn of the year, we expect a number definitely higher than two months ago. An allotment in the EUR 150-200bn neighborhood seems to us pretty reasonable. In this respect, it is important to note that at this week’s long-term refinancing operations the demand of liquidity was quite low with only EUR 0.8bn of bid at the 6M LTRO (vs. EUR 21bn of expiring liquidity), EUR 11bn of bid at the 3M LTRO (vs. EUR 13bn expiring), and EUR 2.5bn of bid at the 1M LTRO (vs. EUR 7.7bn expiring). In our opinion, this is a sign that institutions are shifting their interest to the December 12M LTRO.
This view is underpinned by the recent increase in the 12M OIS rate, which is now at 0.70% (30bp below the refi rate). The spread vs. refi represents the cost of getting 12M liquidity from the ECB instead of funding on the secondary market. The lower the spread, the higher the incentive to go to the ECB. It is interesting to note that since the June auction this spread has moved in the 20-40bp range. As of today, the spread 12M OIS vs. refi is in the middle of this range, and, importantly, lower than at the September auction (40bp). Should it move below the lower end of the range, it is likely to be a further sign of strong demand in December.
Clearly, the ECB would be pleased with a lower figure and, in order to achieve that, may even decide to apply a spread on the allotment rate. But we keep thinking that this is not going to be the case for a couple of reasons: 1) it’s an undesired arbitrary element in a procedure that so far has worked perfectly; 2) it would send a misleading signal to markets: is the ECB applying a spread only for the pure need of draining liquidity and regaining control of market rates, or is it a signal of a higher refi earlier than expected (this motive heavily clashing with the present cautious central bank’s reading of the cycle juncture)?
We are inclined to think that a spread may be applied only if Eurosystem’s staff GDP forecasts in December are way higher than in September, with an inflation projection for 2011 higher than 2%, implying a central bank willing to act on the refi in 1H 2010. However, recent remarks by Mersch and Nowotny imply that the upward revision to growth will be relatively contained. Faced with a strong EUR and a still-wide output gap, the ECB cannot see upside risks to inflation in the medium term.
It is clear that the only ways that the ECB has to regain control of market rates is: 1) to wait until excess liquidity declines significantly (evidence suggests that this cannot happen with a liquidity surplus higher than EUR 80-90bn), and 2) that the “full allotment” procedure is no longer in place.
With respect to the first point, several aspects are worth stressing. First, all the liquidity that will be allotted at the December 12M LTRO will remain in the system until the end of 2010. This holds particularly true if the ECB decides to cancel the other extraordinary operations, as we expect (1M and 6M). Second, autonomous factors are an important component of liquidity demand: they are an important liquidity-absorbing factor but out of the ECB’s control. Namely, they correspond to the sum of: 1) banknotes in circulation; 2) governments’ deposits at Eurosystem central banks; 3) the investment portfolio of euro area central banks (due to an accounting reclassification enforced at the end of 2008). It is worth noting that in October 2008 autonomous factor jumped sharply, as a consequence of: 1) higher demand for banknotes, especially the high-denomination ones, related to a lack of confidence in the banking system at that time. According to ECB statistics, the level of currency in circulation has remained above its long-term average, despite government interventions in support of banks; 2) increase in the size of the deposits by euro area governments at the Eurosystem, mostly due to the various government programs introduced in response to the crisis; 3) increase in the euro area central banks’ holdings of government bonds.
In the medium term, autonomous factors could decline, returning to the level prevailing before October 2008. As stressed before, the increase in such a component is mainly due to crisis-related factors. Hence, it is reasonable to expect a decrease in such sub-components when the crisis is over. Ceteris paribus, the decline in the autonomous factor would mean a corresponding increase in the excess liquidity, thus putting downward pressure on overnight rates.
To this extent, recent remarks by Weber and Trichet suggest that the ECB is fine with an EONIA “fairly” lower than the refi rate for some more time, and that the full allotment procedure – at least at the weekly MROs – can be kept in place for longer. In order for the first condition to be met, especially if we are right on the allotment at the December 12-month LTRO, the ECB has to wait until July 1 when the massive June’s 2009 EUR 442bn will expire.
Given that the GC meeting in December will take place before the 12-month LTRO, the ECB would probably decide to err on the side of caution and leave some important decisions to be taken in January or throughout 1Q 2010, once the excess-liquidity picture will be clearer. Hence, at the December meeting we expect the ECB to announce:
- The end of 12-month LTROs;
- No spread on the December 12-month LTROs;
- Return to the normal “fixed amount/ variable rate” procedure for all the other LTROs from January (or a bit later in 1Q), with a high chance that 1M and 6M operations are canceled;
- Extension of the “fixed rate/ full allotment procedure” for the weekly MROs at least until the end of 1Q or sometime in early 2Q;
- A more detailed communication set (calendar, etc.) in January with a clearer excess liquidity information.
The trickier question is to define a reasonable EONIA path. However, under the working assumption of a resurging liquidity surplus in December and assuming that our bets are correct, we expect the EONIA rate to stay around current levels until July 1 2010 when the huge impact on excess liquidity of the first 12-month LTRO fades away. If the bidding at the December 12-month operation is not very sizable, then the situation will dramatically change after July 1, and the EONIA will jump toward the refi again, reversing the drop of last June. In the light of recent rhetoric, aimed at stressing the willingness to keep the EONIA lower, that would probably be the perfect outcome for the ECB. Indeed, under this scenario, the ECB would have ensured the beginning of the exit strategy without triggering any abrupt market reaction: as a matter of fact, investors would have plenty of time to prepare for the rise in EONIA of next summer.Two things may alter this unfolding of events: 1) as already argued, a heavy allotment in December. In this case, the exit strategy would start as well but it may last longer than the ECB desires, because the EONIA would stay lower for longer. Under these circumstances, the ECB may be forced to return to competitive auctions for the weekly MROs earlier (already in 1Q 2010?), or decide to lift the deposit rate, thus narrowing the corridor. 2) A stronger-than-expected recovery, coupled with upside risks on inflation that would significantly alter the ECB stance on the “traditional” refi rate policy. It’s difficult to gauge on point 1), whereas we feel more relieved on point 2). In our 2010 Outlook to be published at the beginning of December we’ll largely argue why.