ECB bond buying flattens long-end yields
Listening to Draghi’s press conference at the beginning of June, the execution of the ECB’s Quantitative Easing (QE) program is proceeding rather smoothly. We tend to disagree.
Our calculations show that for some of the core countries, such as Germany and the Netherlands, the purchases of government bonds will move to the long end of the curve.
In addition, the ECB will need to increase the issue limit (the maximum percentage that may be purchased per outstanding bond), or the purchases of German and Dutch government bonds will come to a standstill. This could already happen quickly if interest rates remain at the same low level or decrease even further. The Brexit vote resulted in a rally in core-country interest rates, which magnifies the effect.
The ECB has set the following rules for the daily purchases in bond markets:
1. The maturity rule permits the ECB to buy bonds with remaining maturities longer than 2 years and shorter than 31 years.
2. The deposit rate rule only allows the ECB to buy bonds with a yield to maturity higher than the deposit rate, i.e. minus 0,4%.
3. The last relevant rule is the maximum percentage that may be purchased per outstanding bond. This limit, also known as the issue limit, is set at 33%.
Since the start of the QE program in March 2015, interest rates have declined significantly. The 10-year Bund, for example, is currently trading close to 0%. With declining interest rates, more and more bonds trade below the deposit rate, making them ineligible for the ECB to purchase. In other words, the pool of purchasable bonds shrinks with decreasing interest rates.
In practice this means that since the start of June, the minimum maturity that can be bought moved from 5 to 8 years in Germany. This is illustrated in figure 1. This also has a second-round effect: with a decreasing pool of purchasable bonds more needs to be bought per bond to get to the required monthly target of EUR 80bn. As a result, issue limits fill up much quicker.
Data from the ECB shows that during 2015 the bulk of ECB purchases in the German and Dutch government bond market were bonds with a relative short maturity. However, since the start of 2016 this concentration in the short end and middle part of the curve has moved to other parts of the curve. Figure 2 shows that since the end of last year the weighted average maturity of purchases has gradually increased for Germany and the Netherlands, while for many of the other countries that average has been relatively stable.
Reasons for buying longer maturities are the fact that short maturity bonds trade below the deposit rate and that many bonds in the belly of the curve are likely close to the maximum of 33%. This combination forces the ECB to buy core-country government bonds with longer maturities.
Moving the focus of the purchases to the longer end of the curve will not buy the ECB that much time. In May, the ECB bought almost EUR 20bn in Germany alone. With the average outstanding size per German bond of EUR 17bn, the 33% limit will fill up rather quickly. Without changes in the parameters of the purchase programs, we think that the ECB will struggle to buy government bonds in both Germany and the Netherlands around year end, while the current purchase programs runs until March 2017 at least, and probably a lot longer.
The ECB has options to increase the issue limit for bonds issued before 2013 to e.g. 50% or more. This will give the ECB leeway to continue buying for one or two more quarters. And since these older bonds are mostly concentrated in the long end of the curve, the ECB will again focus buying in that part of the curve.
We expect the ECB to announce an increase in the issue limit for these old government bonds from 33% to 50%. This will likely occur in September. Increasing this limit should be a fairly easy step, since this has recently been done for supranational bonds. This should provide some additional breathing space.
The Robeco Global Total Return Bond Fund, Robeco Euro Government Bonds and Robeco All Strategy Euro Bonds are positioned for flatter long-end curves. We expect curves to become much flatter because of the scarcity mentioned above. In addition, it follows also from the experience in other countries like Japan, where inflation is low and central banks are in expansionary mode. Since the beginning of June, the scarcity theme really got traction, with a significant and rapid flattening as a result, which is shown in figure 3. We remain positioned for a flatter curve.
Kommer van Trigt, head of Global Fixed Income and Marck Bulter, portfolio manager Global Fixed Income at Robeco