Choose your country :
Choose your country ?
The ECB has already started “tapering” less corporate purchases than other programmes since April 2017, the month which saw the reduction from €80bn to €60bn of monthly purchases. Recently, President Draghi was not the only member of the ECB Governing Council noting that in 2018 the “ECB will continue buying sizeable quantities of corporate bonds in the programme”. “Sizeable” gives no indication as to a specific number for 2018 but surely points to an increase in CSPP purchases in percentage terms on the overall programme and relative to the public sector programme. In this paper, we address the ECB CSPP from three different angles: its structural impact on the EUR corporate bond market, the state of the art of the programme and developments in the programme in 2018 and beyond, together with the rationale behind our expected scenario.
CSPP has structurally impacted the EUR corporate bond market and will continue to do so. Let’s see how
Market size and composition: strong support for non-financial BBBs and BBs
The supply of non-financial IG corporate bonds was quite low before the CSPP announcement, despite ongoing QE in other asset classes at that time. Volumes were led by the domestic sector (companies domiciled in the Eurozone), especially after March 2016 ECB’s CSPP announcement. CSPP’s positive effect on supply was both quantitative and qualitative, in that periphery names were back and average duration of new debt trended higher. The EUR HY primary market has been busier, as well since March 2016, after seven months of very poor new issuance levels. Volumes of new debt were back thanks to CSPP, but they did not accelerate on the same trajectory as IG market activity. However, also in the case of HY new debt, the return of investor risk appetite has been driven by the increase in average duration of new debt. Among nonfinancial IG names, BBBs’ market size is up by 42% since the CSPP announcement. It had grown at half this pace in previous years following the “Believe me!” speech. In the chart below, the vertical light blue line depicts the CSPP announcement, the vertical dark blue line, Mario Draghi’s “Believe me!” speech. Among speculative grade issuers, since the CSPP announcement, BBs’ market weight has moved from 63% to 74% of the overall EUR HY market. The trend on the EUR HY market has been remarkable over the few years, not only in terms of overall size but also average quality.
The demand side: stable investment inflows into IG bonds
The CSPP triggered strong demand from private investors into EUR corporates: inflows into the IG asset class were strong and steady, with few exceptions (Trump’s election, for example), while products dedicated to speculative grade were more volatile, feeling for example the negative impact from Brexit. Inflows into IG were concentrated on short-term instruments, thanks to the stronger search for yield in this segment.
The impact on relative performance within fixed-income asset classes
The two charts below show excess returns over Bunds delivered by crossovers and periphery/Italian bonds before and after the CSPP announcement. The first chart ranges from the famous Draghi speech (“Believe me!”) and the CSPP announcement, while the second covers the following period. The two charts show quite a different relative pattern: this is not only true in terms of cumulated outperformance but also when adjusting it by volatility.
The impact on sensitivity to equity implied volatility and political risk
Empirical evidence from the last 18 months seem to point to less sensitivity of the asset class to temporary spikes in equity implied volatility. The Brexit shock in particular constituted the first real market “stress test”, but also, in the wake of political and geopolitical risks, recent history seems to confirm a reduced correlation between spreads and V2X with respect to pre-CSPP “regime”.
CDS-bond basis has moved to steady positive levels. The “tapering” announcement failed to spur a compression, underlying market perception of quite a remarkable CSPP effect in 2018, too. As the chart shows, the announcement and the start of CSPP (respectively indicated in the chart with the light and dark blue vertical lines) led to a new regime of the basis that is still persisting in full.
The other side of the coin: the (undesired) effect of tighter valuations
EUR HY spreads are below their fair values. The gap peaked at 170 bp of the opposite sign at the beginning of 2016. Since then, thanks mainly to the ECB’s CSPP and stronger macro and micro momentum, spreads have steadily moved towards fair values, finally reached in December 2016. Since then, HY spreads have been consistently below fair values. Together with market spreads, however, fair value has also trended down, on the back of improved fundamentals, tighter sovereign spreads and persistently low volatility.
Where CSPP stands 18 months after launch
Lots of leeway and still a long way to go…
After 18 months of CSPP activity, the ECB’s cumulative purchases of corporate bonds reached EUR 130bn in value by the end of November. Current holdings make up a bit more than 15% of the almost EUR 840bn in total eligible debt. The primary market’s contribution remains limited, with its share on overall holdings steady at 14% of CSPP operational activity in 2017. This means that the ECB continued to play a fairly important role in secondary market transactions.
…trough a widespread portfolio
Although holdings are still limited with respect to eligible debt, the ECB spread its purchases over a remarkably large number of issuances. The number of bonds in the CSPP portfolio has, in fact risen to 1046 as of this writing, thanks to weekly additions. With just 69 issues of the eligible universe not yet included in the ECB portfolio, this means that around 94% of estimated eligible issues have been “affected” by the ECB, or about 92% of eligible debt. This broadening of the portfolio is even more remarkable in light of the accelerating activity recorded in the primary market since the launch of the programme.
A “benchmark-neutral” approach…
As we know, the ECB doesn’t communicate the amount of debt purchased in each bond. However, last June (the first anniversary of the programme’s start) the ECB released interesting details on its holdings at that time. Detailed numbers regarded not only country, sector and rating classification, but, more interestingly, also the year of bond issuance. In a nutshell, and as widely expected, ECB stated that “the composition of CSPP holdings generally mirrors that of the CSPP-eligible bond universe”, and that “the breakdown of CSPP holdings by country follows that of the CSPP-eligible bond universe very closely. Nor are there major deviations … in terms of sectors of economic activity or rating group”. In October, the ECB resumed disclosing CSPP purchases on the breakdown of holdings by ratings, sector and country. This “benchmark-neutral” approach is quite understandable, in our view, given the aim to avoid unwanted market distortions or discriminations. In this respect and based on end-October data, the ECB’s portfolio, like the benchmark, is tilted towards BBB-rated debt (49% of holdings, vs. 45% of the universe), consumers and utility sectors (making a combined 54%), and core countries’ corporates (54% of holdings have been issued by French and German companies). Italian and Spanish domiciled companies both make up 11%, weightings very similar to benchmark ones.
…but a longer duration, likely to make reinvestments more persistent over the next years (2019-2021, in particular)
What about reinvestments over the coming years? The most interesting numbers released by the ECB in June regarded the breakdown of holdings by year of issuance. The ECB underlined the fact that “CSPP holdings tend to be skewed towards bonds issued more recently”. This is due to both participation in primary markets and the better liquidity of newly issued bonds. In June 2017, therefore, more than half of holdings were concentrated in bonds issued in 2016 and H1-2017, while these bonds accounted for only around one third of the eligible universe. In a nutshell, the ECB “underweights” older issues and “overweights” bonds issued over the most recent quarters. We have run some analysis on the current CSPP portfolio and because of strong activity in primary markets following data publications, this overweight is now likely to be even higher than in June. Since then, in fact, the weight of CSPP bonds issued in 2016 and 2017 has increased from one third to 40%, suggesting that corresponding holdings are now probably higher than the “more than half” figure released almost six months ago. This is an important indication in assessing reinvestments’ volumes of corporate bonds, as they will obviously depend on the portfolio’s average maturity. We therefore first analysed the distribution of the around 1,000 bonds held by the ECB by year of issuance. Apart from the 40% issued in the past two years, another 23% was issued in 2014 and 2015 and another 22% in the two years before that. This leaves us with a maturity schedule that looks longer than the benchmark’s. That’s why the apparently disappointing volume of corporate bonds maturing over the next 12 months doesn’t come as a surprise, based on the above analysis. The distribution in maturities of bonds issued in the last 24 months, in fact, gradually increases to “peak years” not before 2023, with very few bonds maturing in 2018 and 2019. The CSPP portfolio, therefore, likely differs from benchmarks almost only in terms of average duration, which, based on the numbers released and primary market activity, should be longer.
Why and how the ECB will taper CSPP less than other QE programs in 2018
ECB has been “tapering” corporate purchases less than other programs since April 2017…
Draghi was not the only ECB Governing Council member to point out that in 2018 the “ECB will continue buying sizeable quantities of corporate bonds in the programme”. Novotny, as well, after Draghi remarks following the October announcements, said that “corporate bond-buying share may rise from January”. These comments were consistent with the October press release, where “sizeable” was mentioned, too, referring to private sector purchase programme. Actually, according to the ECB’s numbers, the step down from EUR 80 to EUR 60bn in monthly purchases since April 2017 has already been “corporate friendly” on a relative basis, as CSPP’s weight in overall purchases increased. As the following chart shows, in fact, average weekly purchases fell by 24% for public sector bonds, but by a lower 14% in the case of CSPP, raising the weight of CSPP volumes relative to PSPP from 12% to 14%. “Sizeable” doesn’t give clue on a specific number for 2018 but surely points to an acceleration of CSPP purchases in percentage on overall programme and relative to the public sector. If monthly purchases fall, let’s assume, to EUR 5bn from the current EUR 6.5bn, this would account for 16% of overall future QE purchases (up from the current 11%) and 22% of PSPP purchases (vs. the current 14%). Relative to current ECB portfolios, the increase would be “sizeable” for corporates. An additional amount of around EUR 45bn would mean an increase of 33% over year end CSPP EUR 130 bn, while adding reinvestment to new purchases, the PSPP portfolio would increase by just 11%.
…and there is considerable rationale for a further increase in the weight of CSPP in 2018
We guess that there are many reasons behind a QE strategy based on a pace of corporate purchases that should not slow down too much in 2018. Some of them lie more on the technical side, while others have more to do with effectiveness of monetary policy transmission and fundamental rationale.
On the technical side, the “scarcity issue” doesn’t affect the asset class at all, as the ECB currently owns just 15% of the overall eligible corporate bond universe, or 22% when the issuer limit is taken into account. This is in sharp contrast to the much higher numbers of the ECB’s cumulated purchases for example in covered bonds and German government bonds, the latter almost in the 90% region of capped Bunds. The issuer share limit is higher for corporates than for public bonds and CSPP, among QE operations, is the latest or most recent programme to be launched. It became operational in June 2016, while the PSPP was launched one year before and the third covered bond programme started even 18 months before the CSPP. Flexibility, a feature often cited by board members when they refer to QE, has probably to do with available room present in the CSPP-eligible universe usable to ease pressure on “crowded” segments and ease also sceptical views about the sustainability of future purchases. The CSPP programme, then, by its nature tends to raise fewer “political issues” than PSPP, as no capital key rule is applied and the share of core countries looks dominant with respect to other countries, coherently with the composition of market corporate benchmark.
CSPP accounts for just 5% of overall QE purchases. 84% of the ECB’s portfolio consists of public sector instruments and another 10% of covered bonds, while ABS make for the residual part of 1%. According to the numbers released by the central bank, out of around EUR 140bn of maturing bonds over the next 12 months corporate bonds would make up just EUR 4bn, obviously much less than PSPP (EUR 107bn) but also less than covered bonds (EUR 19bn) and even less than ABS (EUR 7bn). The ECB could therefore switch part of reinvestments from other private sector programmes into the corporate one.
Moving on to the rationale regarding the transmission of monetary policy, a recent working paper published by the Bank of Italy (Macroeconomic effects of non-standard monetary policy measures in the euro area: the role of corporate bond purchases, N. 1136, September 2017), together with other papers addressing the same topic, evaluated the macroeconomic effect of the CSPP. The findings look quite encouraging in terms of contribution to GDP growth, especially in the event of a prolonged programme. As we have already shown in previous issues of cross assets, CSPP has played a crucial role in easing corporates’ financing conditions in terms of access to lower borrowing costs and higher credit volumes available, ultimately supporting a recovery of investments and employment. A premature stop to the programme might mean an unwanted headwind to encouraging trends in both areas (investments and employment) currently driving the macro recovery, especially in this phase of the cycle and at a time when sovereign spreads and bank financing conditions already fully benefitted from CBPP and PSPP, which have been in place for a longer time. Therefore, in this respect, as CSPP was the last programme to be introduced by the ECB, it could also be the last programme to be ended. This means that the CSPP could even outlive the PSPP.
The average credit quality of the EUR HY markets has improved quite substantially since the CSPP announcement
The CSPP triggered strong demand from private investors into EUR corporates
Current holdings make up a bit more than 15% of the total eligible debt
EUR corporate bond valuations have stabilised at high levels. Significantly, this asset class received direct support from the ECB through its Corporate Sector Purchase Programme (CSPP). In this article, we look at the impact the tapering of this programme will have on the market.
Valentine AINOUZ, Sergio BERTONCINI
The CSPP became operational just a few months ago, but it has already produced dramatic effects on the EUR credit markets. The flexibility of the programme, together with its sustained purchase trajectory, which is spread widely on a very large number of bonds that cover all IG ratings, curve buckets and sectors with issuers from many countries, contributed to the effectiveness of the CSPP.
Head of Rates and FX Research