“We found that an active, ESG-tilted bond portfolio strategy generates superior outcomes compared to a relevant benchmark that does not explicitly take ESG scores into account,” said Bhupinder Bahra, co-head of quantitative research group, global fixed income, currency & commodities at JPMAM.
“There is a typical assumption among investors that their portfolio’s yield will be lower if their process incorporates ESG factors. However, their major focus should be on their fund’s returns and risks. Total return and downside protection is all-important for credit investors,” he told FSA in an interview.
The report, “Systematically Investing in ESG in Global Bond Markets”, was based on a quantitative approach that overlaid MSCI ESG scores on investment grade, high yield and emerging market fixed income, as represented by their respective ICE Bank of America Merrill Lynch global corporate bond indices.
All unscored bonds were removed from the raw market indices in order to create an ESG benchmark, and the research found that the resulting indices with only ESG-rated bonds generally had higher excess returns, lower volatility, higher return-to-risk ratios, better drawdown characteristics, lower yields and lower spreads than their respective market indices.
“The contingent liabilities related to ESG are not necessarily factored into the credit agencies’ assigned ratings, so ESG scores can be additive to traditional credit ratings,” said Bahra.
“Moreover, and perhaps surprisingly, cross-sectional correlations showed that MSCI’s E, S and G scores are not related to one another or to credit ratings. That might suggest that the MSCI itself is differentiating more exactly between the three factors.”
Building an ESG bond portfolio
Next, active ESG-tilted portfolios were constructed for investment grade, high yield and emerging market bonds, derived from their ESG benchmarks. They were immunised against the three main systematic market factors that influence the prices and volatilities of corporate credit instruments, namely: market sector, rating and duration.
They bought only the top 20% of ESG-scored bonds within so-called “rick cubes”, which are the intersections of sector and rating categories and duration buckets.
“In this way, we could ensure that any active outperformance was driven purely by our idiosyncratic ESG-based selection methodology rather than by some other, inadvertent market-based influence, or beta,” said Bahra.
The gross returns of all their ESG-tilted strategies were either comparable to or higher than those of their relevant benchmarks. This was the case for the investment grade bond portfolio after transaction costs, but net returns of the high yield and emerging market portfolios was less than their ESG benchmarks.
However, in all cases post-transaction cost volatility and drawdown was less.
“Systematic back-testing of corporate bond portfolios proves that an ESG overlay can enhance portfolio outcomes via lower drawdowns, reduced portfolio volatility and, in some cases, marginally increased risk-adjusted returns,” said Bahra.
He believes that the next step is to create a model that incorporates the back-tested findings into real-time portfolio construction.
JPMAM already integrates ESF factors into the $5oobn of assets managed by its GFICC team. They use MSCI scores, but are not tied down by them.
“MSCI’s 37 ESG pillars” provide wide coverage of the factors that can be used for an ESG-compliant portfolio, but JPMAM’s analysts often disagree with its weightings in its overall score, according to Bahra.
“For instance, we might weight privacy and security higher for banks because of cyber attack potential, or weight environmental measures higher for industrial companies in anticipation of future regulatory actions”.
Nevertheless, a JPMAM flagship fund – the Global Corporate Bond Fund, has underperformed its conventional benchmark over three years, with a cumulative return of 11.38% compared with 13.62%, according to FE Analytics (chart below).
Its annual volatility is lower (2.71% versus 2.73%), but its drawdown is higher (-3.54% versus -3.25%).
Bahra believes that the new methodology has the potential to clearly determine how ESG factors actually attribute to the performance of a fixed income fund.
“By immunising market – sector, credit rating and duration – factors, the attribution of ESG features to performance can be identified,” he said.
ESG-tilted strategies back-tested results relative to ESG benchmarks
(1) Pre-transaction costs: mainly higher excess returns and return-to-risk ratios
(2) Post-transaction costs: uniformly lower volatility and lower maximum drawdowns
Source: ICE BoAML, MSCI ESG Research, JPMAM GFICC Quantitative Research Group (31 March 2019)
JP Morgan Global Corporate Bond Fund vs sector average and benchmark index