SustainaWeekly - ESG Bonds – Green bond investors are reclaiming losses in the primary

We actively track how much issuers end up paying above or below what the market would have commanded if it was unaware of a new bond deal being marketed (where we obviously look at secondary bond markets for guidance). Greeniums in the primary market, i.e. negative new issue concessions, are rare. Yet during the 30 month time horizon since we have been tracking new issue concessions on green bonds, the picture against regular bonds is quite mixed, especially since 2021. Green bonds price at both a lower and higher concession vs regular bonds during different periods. The chart displays the average concession over all deals for a particular month.

The difference could normally be explained by the variation in issuers coming to the market. For example, in March 2022 the chart on the previous page shows that higher concessions were paid on green bonds but this made sense since not only was the average composite credit rating on the 8 green bond deals a BBB1, these bonds were also offered at a longer maturity of 7.8 years on average. The 31 regular bonds priced during the same period had a composite rating of A3 and a 6.7 years of maturity on average. Paying higher concessions on weaker quality and higher duration risk makes perfectly sense to us, especially in the risk-off market at the time.

We are seeing a similar 10bp of higher new issue concession being paid on green bonds offered in October as we did in March. In terms of maturity being offered there is, again, a 1y difference in favour of the regular offering. But surprisingly, the green bond and regular bond offering quality has the same BBB1 composite credit rating. One could perhaps argue that perhaps the TenneT (a Dutch utility company) 20y green bond would pushed up the average new concession on green bond deals. But then we saw a higher concession being paid by Bouygues (a French conglomerate company) on a regular deal for a 10 year maturity, while Bouygues and Tennet have the same A3 composite credit rating. A slightly longer average maturity on green bond deals being offered this month was not the defining factor. But green bond investors have incurred bigger losses than broad based corporate bond investors because of their higher allocation towards real estate credit. Perhaps covering this through the primary is a way to cover a part of these losses incurred in the secondary. Indeed, there was nearly 60bp been paid in new issue concession on the Tennet 20y bond deal and after nearly two weeks spreads are 2bp tighter vs re-offer, while spreads on all other very long dated issuance, such as the Bouygues 20y, went wider reflecting ongoing weakness in longer dated credit risk.

A reversal in real estate credit could imply lower new issuance concessions

The EUR IG corporate green bond index came into existence at the end of 2018, perhaps at a same time when green bond issuance was surging. This green bond index has lower composite quality and higher duration than the broad EUR corporate IG index, yet still it managed to trade at tighter or similar spreads, until recently. Secondary green bond spreads today trade a record 20bp wider than the broad index.

The charts below show the difference in sector weights between the Corporate green bond index (Bloomberg ticker ERGN) and the Corporate all bond (regular and green) index (Bloomberg ticker ER00). Banks are included in these corporate indices. Real estate clearly has a bigger weight in the green bond index and the right hand chart shows that real estate has been a clear underperformer in terms of spread widening this year and the green real estate bonds were no exception. We note for example that 8y and 9y remaining maturity green bonds issued by less familiar real estate issuers such as Dutch CTP and Belgian VGP are trading close to 50% of face value and a corresponding yield of 9 to 11%!

Performance in the overall green bond corporate index therefore depends on a reversal of fortunes in the real estate credit space. A couple of weeks ago we showed in our dedicated credit publication called the ABN AMRO Monday Credit Crisp (see here) that the pricing of real estate credit had become so extreme, that for example the difference in credit spread between real estate and the broad corporate bond market, as expressed in Z-scores, had breached values which normally either flag oversold conditions or severe distress on issuer level. The chart on the left shows that this Z-score moves in tandem with interest rate volatility (proxied by the MOVE index). But interest rate volatility in the US has been strong as well. Actually the MOVE index captures US interest rate volatility, but since rates have also aggressively moved up in Europe we feel comfortable to use the MOVE as a gauge of European volatility as well. Still, the right hand chart shows that US real estate credit, in contrast to European real estate credit, has been more resilient as the Z-score remains under 2.

The most likely pivot in the fortune of European real estate credit will come when interest rate volatility starts to moderate. The good news is that we are likely approaching the peak in the Eurozone rate hike cycle and we should also see long term yields come down next year. Perhaps the higher leverage used by European real estate issuers vs their US peers could explain the more extreme reaction on this side of the pond to European real estate credit as a lower starting equity implies less cushion for bond holders when valuations drop. However, the imminent recession, where we expect 0.9% contraction in 2023 as a whole in the Eurozone, is unlikely to cause a heavy decline in property values due to a spike in vacancies or massive downward pressure on rents upon renewals.

Therefore, if spreads in real estate credit come down faster and stronger than the broad corporate space, perhaps green bond investors will see outperformance and then also do not need to be greedy in demanding new issue concessions.