GS: Five questions on the ECB's sovereign purchases, ten days on
Focus: Five questions on the ECB's sovereign purchases, ten days on
Bottom line: On March 5, the ECB presented technical details for its sovereign debt purchases under the Public Sector Purchase Programme (PSPP), following January's announcement of the programme by the Governing Council. This programme is expected to be maintained at least until end-September 2016, unless the ECB "see[s] a sustained adjustment in the path of inflation which is consistent with [the ECB's] aim of achieving inflation rates below, but close to, 2% over the medium term". After the first 10 days of this new programme, we address five questions related to its implementation, impact and future practicalities. The first week of the programme appears to have evolved in line with the ECB's announcement. We believe there is little risk that PSPP purchases will stop before end-September 2016, but expect the ECB to exploit the flexibility it has allowed itself with respect to the composition and maturity of purchases to the full.
At its press conference on March 5, 2015, the ECB presented technical details of its sovereign debt purchases under the Public Sector Purchase Programme (PSPP). According to ECB President Mario Draghi, this programme is expected to be maintained at least until end-September 2016 and "in any case until [the ECB] see[s] a sustained adjustment in the path of inflation which is consistent with [the ECB's] aim of achieving inflation rates below, but close to, 2% over the medium term".*
After one week of implementation of the PSPP, we address five questions that have arisen on the purchases.
1. Are the PSPP purchases in the first week consistent with the EUR60bn per month target?
Yes, but on the tight side.
The ECB's EUR60bn per month target implies a purchase rate or 'run rate' of around EUR3bn per working day and EUR14bn per week (although purchases of sovereign debt under the PSPP are only one component of the intended run rate; purchases of ABS and covered bonds also need to be allowed for).
The official figures published show that the ECB purchased EUR3-4bn in the first day (i.e., slightly above the required run rate) and EUR9.8bn in the first week (slightly below the required rate). While the amount for the first week as a whole appears on the tight side relative to the target, we expect an increased rate of intervention by the end of this month.
2. Have the purchases had an impact on financial markets?
Yes, to a significant degree across a relatively large spectrum of assets.
One of the motivations behind the PSPP was to sustain the Euro depreciation and consolidate further the improvement in financial conditions. The liquidity injection through the PSPP could thus encourage consumers and entrepreneurs to bring forward their spending plans and thereby have a positive impact on growth (via the impact of Euro depreciation on exports) and the outlook for price developments (possibly reversing the decline in inflation expectations seen from the middle of last year).
Recent developments in various financial asset prices suggest that the impact of purchases under the PSPP is in line with these ambitions:
Equity prices have maintained their rising trend, with the Eurstoxx index breaching the level of 370 since the start of the PSPP purchases (Exhibit 1);
Bond yields have decreased significantly since the start of the PSPP across (core and peripheral) countries and maturities. Today, the 10-year yield currently stands at around 0.2% for the German Bund, 0.4% for the French OAT, 1.3% for both Italian BTP and Spanish bonds. Similar developments have been observed for public issuers in smaller Euro area countries (e.g., 0.4% for the 10-year Belgian bonds and 1.7% for Portuguese bonds);
Sovereign spreads narrowed further (Exhibit 2), while the EUR, at 1.05, reached its lowest level in a decade against the USD (see our
recent analysis on FX developments);
Increased upward pressures on excess liquidity (Exhibit 3) have increased further downward pressures on the EONIA in the near future. All in all, excess liquidity currently stands at EUR156bn and it could rise by another EUR700bn by end-September 2016. Such levels would maintain the EONIA close to the rate on the deposit facility with low volatility, as experienced at the peak of the ECB's balance sheet in June 2012. These expectations seem to explain the downward movement of the EONIA forward curve since January (Exhibit 4).
3. Could the ECB stop purchases under the PSPP before end-September 2016?
We think it is unlikely, given the relatively high bar that the Governing Council has set for it to cease purchases.
As stressed by Mr. Draghi, the PSPP is expected to last until there are clear signs of a "sustained" adjustment in the inflation path, bringing it into line with the ECB's price stability objectives. In theory, the ECB could thus stop the PSPP if such adjustments occur before end-September 2016. However, we believe that such a sustained adjustment is unlikely, for at least three reasons:
The (negative) impact of oil price declines on consumer price inflation in the Euro area is likely to be faster than the (positive) effects of Euro depreciation, even if these effects largely cancel out at a two-year horizon. This explains our baseline scenario of a subdued inflation outlook by end-2016;
The stickiness of inflation will likely limit a rapid upturn in inflation figures in the short run. Inflation persistence in the Euro area has risen of late: we estimate that inflation in the previous month explains up to 40% of the current month's inflation;
Consequently, a recovery in inflation expectations is likely to be slow, especially if recent past inflation figures increasingly affect expectations in the short run.
All in all, we therefore believe that a sustained upward adjustment in inflation in line with the ECB's definition of price stability (below but close to 2%) will take time, and significant improvements should not be expected before end-September 2016. Furthermore, by setting a high hurdle to suspending the PSPP, any shortfall in the data relative to the ECB's projections (assuming the PSPP is maintained until September 2016) would make a strong argument for not suspending the programme before end-September, possibly even suggesting that it be maintained for a longer period.
4. Could the ECB encounter difficulties in purchasing bonds in some segments, thereby endangering the credibility of the programme?
No, we do not anticipate problems with making purchases at the announced pace, at least in the first few months.
Concern has emerged in the market about the feasibility of purchases at the pace announced by the ECB on the grounds that some investors (especially institutional ones) may prove reluctant to sell their high-rated public bonds owing to regulation and lack of alternatives. In addition, the ECB has imposed a set of limits on the character of its sovereign purchases: i.e., the distribution of purchases is in line with the ECB's capital key; purchase limits of 25% and 33% by issue and issuer, respectively; and no purchases of bonds yielding less than the rate on the ECB's deposit facility (i.e., below - 20bp). As a result of these self-imposed constraints, an arithmetical problem could arise if the yield curve in some large countries (say Germany) were to flatten out below the -20bp threshold and the stock of purchasable bonds fell below the ECB's intended purchase volume.
After one week of the PSPP (and subject to uncertainties related to any extrapolation of future developments), these factors do not appear likely to constrain the purchase programme, at least in the near term. Today, although the limit set for the purchases of negative-yielding bonds (-20bp) does not seem to be a concern yet for the Euro area on average (around 94% of public bonds in the Euro area still have yields to maturity above -20bp), the country-based picture may be different (e.g., for Germany, where only 69% of bonds have yields above this limit, whereas all public French bonds have yields above the -20bp limit).**
In the event that the ECB's ability to buy becomes limited, possible further adjustments to the PSPP could be envisaged (such as adjustments to the 25% and 33% ratios, as explicitly mentioned by the ECB itself, or increased focus on substitute assets*** with a possible extension of this list in the future). By contrast, while possible in theory, we do not necessarily believe that a further downward move of the deposit rate is a real option, although it is too early to have a definite view on this issue at this stage.
5. Can we expect the ECB to be flexible in the quantity and composition of its purchases on a month-to-month basis?
Yes, we think this is likely.
Although the stipulated amount of EUR60bn can be seen as a monthly target, the ECB is likely to maintain a degree of flexibility in the composition of purchases (e.g., regarding the breakdown between mutualised (20%) and non-mutualised (80%) purchases, as well as between the national agencies and supranational debt instruments) over the coming 19 months to end-September 2016.
The ECB has decided to publish the aggregated purchase amounts on a weekly basis and additional information on average maturities on a monthly basis. In last Friday's announcement, the ECB did not provide the distribution of purchases by country (or by central bank) behind the aggregated amount of about EUR9.8bn. Partial evidence so far from the market suggests a diversity of purchases in terms of maturity, although with a slight bias towards the long end of the yield curve for many countries (e.g., 4/5/6/9/10/15/16/20/30-year maturity for Germany; 2/3/5/10/15/17/20-year+ maturity for France; 3/5/6/30-year maturity for Italy or 2/8/10/15/25/30-year maturity for Spain). Regarding debt issued by agencies, purchases have tended to focus on maturities in the 5- to 15-year segments. Despite some necessary caution given the limited window of observation, purchases in the first week thus tend to support our view that PSPP purchases will be made on a basis whereby maturity and composition vary quite flexibly, with little in the way of preannounced patterns.