On balance, Morgan Stanley feels that broad-based QE, (i.e. large-scale purchases of government bonds) is further away for the ECB than the market currently believes. Presently they only assign a subjective 40% probability to such a step being taken; whereas the euro rates market is already pricing in the ECB resorting to a broad-based purchase programme with a very high probability of 80-100%. Goldman agrees warning specifically that "Sovereign QE is not imminent... and indeed may never happen." It appears no matter what, disappointment is guaranteed for the market.
As Morgan Stanley points out, ahead of the November ECB meeting, the newswires seem to be picking up again on the possibility of the ECB taking additional policy measures before year-end. Our interest rate strategy team believes that the bond market currently assigns a very high probability of 80-100% or higher to the ECB starting to buy government bonds. In our view, such expectations are likely to be disappointed.
QE is not a panacea: In our view, the negative long-term side-effects of QE on the financial system would undermine the ECB’s efforts to repair the bank lending channel through the AQR, the TLTROs and the two purchase programmes. Furthermore, we think QE would be inconsistent with the ECB’s recent decision to cut the deposit rate further into negative territory.
In our view, the ECB would face some serious political and legal risks if were to move into the sovereign space. Given the explicit ban on monetisation of government debt in the EU Treaty, it might be easier for the ECB to buy private sector debt instead of public sector debt – especially with the German Constitutional Court already taking a critical view on the compatibility of OMT with the German Constitution as well as the EU Treaty. This legal dispute also would put the Bundesbank in a very difficult position should the Governing Council decide – contrary to our expectation – to push ahead with government bond purchases.
For QE not just to foster ‘Japanification’, the ECB would need to make sure that governments press on with structural and, if possible also, institutional reforms. One way to keep the pressures on governments would be to link the parameters of the purchases, notably the countries whose bonds are bought, to fulfilling the requirements of the Stability and Growth Pact (SGP). This link would reinforce the governance rules EMU has been built on, notably rule-based fiscal discipline. Given that the decision as to whether a country is making enough of an effort to bring down its deficit/debt is taken by the Ecofin, finance ministers would effectively need to sign off on the list of sovereigns the ECB buys. This would free the Governing Council from taking politically charged decisions on countries repeatedly breaching their SGP targets.
Beyond these problems associated with sovereign QE, we also believe that the ECB’s surprise decision to cut the deposit rate again is inconsistent with a central bank that is close to embarking on a major QE programme. After all, the negative deposit rate is effectively a tax on the excess reserves that the banks hold. Having recently increased this tax from 10bp to 20bp is to us at odds with the direct consequence of a major QE programme that is to pump excess reserves into the banking sector. For the two policies to be consistent with each other, the ECB would need to raise the deposit rate (and possibly also the refi rate) before expanding its balance sheet through the excess reserves of the banking system. In our view, the Executive Board would be aware of this inconsistency and therefore would not have pushed for another rate cut in September if it was in the final stages of preparing a QE programme.
Because a negative deposit rate incentivizes banks to reduce their excess reserves at the ECB as much as possible, the negative deposit rate also limits the ECB’s ability to increase the size of its balance sheet materially. Instead of piling any payment received from the ECB in selling assets into a QE programme into excess reserves at the Bank, like their counterparts in the US and in the UK do, euro area banks will likely try to reduce their borrowing under the various ECB refinancing operations instead. Hence investors, who are convinced that the ECB needs to do QE, might also want to consider the prospect of an interest rate increase that would need to coincide with it to make it effective.
Furthermore, we would highlight that broad-based asset purchases could be counterproductive to the health of the financial sector. This is because after the initial relief rally, historically they have caused lower returns on a variety of assets, notably government bonds, over the longer run. The dominance of bank finance in the euro area explains why the bank lending channel is the main focus of the ECB’s policies, ranging from the AQR to the TLTROs and the asset purchase programmes at the moment. At the present juncture, the ECB will hope to leverage synergy effects between the AQR, the TLTROs, the ABSPP and the CBPP3 in order to bring down bank lending rates, reduce financial fragmentation, slow the pace of deleveraging and eventually boost new lending volumes.
But the market is fully pricing it in already...
We estimate that the euro rates market is already pricing in the ECB resorting to a broad-based purchase programme with a very high probability, i.e., 80-100%. This figure is based on work we previously did, which suggests that a €1 trillion government bond purchase programme would depress 10y Bund yields by 50-70bp, the assumption that QE affects bond yields primarily through depressing the interest rate term premium, and from assigning the majority of the 60bp decline in 10y Bund term premium year-to-date to increased expectations of QE, as the other factors that typically drive term premia (i.e., interest rate and inflation vol) cannot explain the scale of the decline.
However, it would be consistent with the experience of QE in the US and UK, where the announcement of the subsequent purchase programmes did not lead to lower government bond yields.
* * *
Paradoxically, the extent to which the market is expecting QE reduces the need for the ECB to do QE, to the extent that the ECB would be looking to engineer euro sovereign yields lower.
* * *
Goldman agrees, warning that ECB sovereign QE: Not so imminent...
The unique institutional setting of the Euro area increases the (political) fixed cost of adopting sovereign QE relative to other jurisdictions. In an uncertain macroeconomic environment, higher fixed costs increase the option value of waiting before engaging in sovereign purchases. This reasoning both: (1) explains why the ECB has been a laggard relative to its peers in making sovereign purchases in the past; and (2) suggests that the ECB will continue to prove more reluctant to engage in sovereign QE in the future than the conventional market narrative assumes.
ECB sovereign QE: Challenges to the conventional narrative
Market participants currently focus on the prospects for ‘sovereign QE’ by the ECB.
A typical narrative is as follows. In the Euro area, inflation is, for all practical purposes, at zero. Likewise for growth and policy interest rates. To head off protracted, Japan-like deflation and stagnation, something needs to be done. Conventional policy easing has been exhausted: unconventional measures are required. Sovereign QE – large-scale central bank purchases of government debt – has worked in the US and UK (even if we are not quite sure why). Ultimately, the ECB will have no option but to follow the Federal Reserve and Bank of England along this path. And given ongoing declines in inflation and the recent weakening of area-wide growth, there is no time to waste in moving forward.
This narrative contains important truths. On our reading, the leadership of the ECB recognises rising threats to price and macroeconomic stability in the Euro area. This prompted the ECB to ease policy in June and September. And there is a preparedness to do more. We expect further measures from the ECB over the coming quarters.
Yet sovereign QE by the ECB is not part of our base case for 2015. The Euro area is not the US or the UK: rather, it is a unique – and uniquely fragile – construct of 18 (soon to be 19) separate countries, which share a common currency but have their own fiscal policies and political institutions. And the autumn of 2014 is not the spring of 2009: the starting point for implementation of sovereign QE is quite different today.
On this basis, we identify three (inter-related) reasons why sovereign QE in the Euro area may be less likely than the conventional narrative presumes.
- Political constraints. The distributional impact of sovereign QE is likely to be more politically contentious in a multi-country monetary union (such as the Euro area) (especially if it has a cross-border character) than in a unitary state (like the US, UK or Japan). Political considerations play an important role in constraining ECB actions.
- Economic effectiveness. With the risk-free yield curve in the Euro area already both low and flat, and credit spreads compressed, the scope for sovereign QE to ease domestic financial conditions further from here is limited.
- Financial structure. The Euro area has a bank-centred financial system. Direct market interventions such as sovereign QE may be less effective in this environment, should the banking system fail to transmit any easing impulse into the real economy.
Such considerations help to explain why we have not seen sovereign QE from the ECB as yet. They also weigh against the announcement of sovereign QE in the coming months.
Sovereign QE in the Euro area is more distant than the conventional narrative implies (and, indeed, may never happen). That said, it is clearly not impossible. As reflected in our ECB preview published earlier this week, we see a 1-in-3 chance of sovereign QE by the ECB through the middle of next year. There is (and always has been) a point at which the macro data are bad enough to trigger sovereign QE, notwithstanding the factors listed above that weigh against it. As the Euro area macro data deteriorate (and particularly as longer-term inflation expectations drift downwards (Exhibit 3)), this trigger point gets closer.