Are ESG Alpha and Beta Benefits in Corporate Bonds a Mirage?

In this new paper, members of the quantitative research and portfolio management teams at Coolabah Capital Investments (CCI) present new evidence on whether the claimed alpha and beta benefits of ESG filters in both the global equities and corporate bond markets exist. You can download a full copy of the 10 page paper here or here, or review the executive summary enclosed below.

Executive Summary

A recent trend amongst retail and institutional investors has been the desire to construct portfolios that are compliant on an environmental, social and governance (ESG) basis. But this begs the question as to whether ESG factors have any relationship with future risk and return. Put differently, can ESG drive alpha and/or reduce beta? While there is a growing literature on the value of ESG factors (see Bektic (2017)), this paper presents alternative out-of-sample methods for isolating these relationships and considers both equities and corporate bonds in a global context.

Leveraging off one of the world’s leading ESG score providers available via Bloomberg, we initially find that ESG factors are not statistically significant in linearly predicting future equity alpha after controlling for market and industry betas on an out-of-sample basis for companies based in Australia, the US, Europe and Japan.

Our research is subsequently expanded to investigate non-linear relationships between ESG factors and future performance by examining quintile-based portfolios using the Capital Asset Pricing Model (CAPM) framework. Adopting this approach, we find that the best and worst ranked quintile portfolios on their ESG scores were statistically significant in generating alpha and reducing beta, out-of-sample, for companies in Australia and the US, but not for those located in Europe and Japan. Within the ESG scores, we found that the “governance” factor is by far the best predictor of future returns and risk, in Australia, the US and Europe.

In Section 3 we study the relationship between ESG factors and future corporate bond performance, focussing on the US market where the best available TRACE data exists. Employing a similar quintile-based portfolio and CAPM method, we find no statistically significant effects on alpha (in the CAPM sense), but ostensibly statistically significant beta reduction benefits.

Importantly, however, we demonstrate that better (worse) ESG scores are associated with superior (inferior) credit ratings, which is known to decrease (increase) beta.

In order to investigate the marginal effect of ESG scores, we construct quintile-based portfolios with similar characteristics along the three major dimensions that affect a bond’s risk and return profile, namely its credit rating, the issuer’s industry and the time to maturity (or call). After equalising the industry, tenor and rating composition of the five portfolios, we find that ESG factors do not significantly contribute to positive alpha in the bond market. In fact, the best social quintile portfolio has statistically significant negative alpha; though given the non-monotonicity of the results and the multiple applications of confidence interval tests without multiplicity adjustments, we cannot conclude in favour or detriment to ESG effects on alpha.

More importantly, however, once we adjust for credit rating and industry factors, we find that the previously observed beta-reduction benefits of ESG factors are no longer significant. In fact, the best environment and social quintiles have significantly higher beta, which is undesirable. In addition, better total ESG quintiles have betas that are nearly monotonically increasing and almost significant, which is again, undesirable.

We conclude that there is value in analysing ESG factors when considering individual investments in equities but less so in corporate bonds. There is a case for ESG alpha and beta benefits in the equity markets in Australia and the US, with governance by far the most important factor. We could not identify any objective ESG alpha or beta benefits in the corporate bond market, with ESG insights apparently already captured in companies’ individual credit ratings.

While ESG factor analysis is an important part of any investment process, participants need to understand the strengths and weaknesses of ESG scores.


RMBS default rates trending higher as house prices plunge

Coolabah developed the world’s first compositionally-adjusted, or hedonic, regression-based index of RMBS default rates, covering all prime deals that Bloomberg reports on (you can download the paper here). We update this index monthly, and the results are enclosed below.

Contrary to S&P’s SPIN Index, we find that Aussie RMBS arrears (nb: legally there is no difference under home loan contracts between being in “default” and in “arrears”) have been steadily climbing since 2014. This is consistent with the RBA’s latest data on home loan default rates, which Governor Phil Lowe published during the week (see chart below). Observe that according to both the RBA and Coolabah’s indices, defaults have been rising for many years. In fact, Aussie RMBS arrears are approaching their GFC peaks.

At the same time, house prices are experiencing their biggest falls in 40 years (see Sydney chart below), which is particularly negative for RMBS because the only thing protecting the bond is the value of the home. This in turn means that leverage or the loan-to-value (LVR) ratios underpinning recently issued RMBS are likewise rising fast. When we model deals on a bond-by-bond basis we see big spikes in the share of borrowers with LVRs over 80% and 90% in deals issued since 2017.

Another significant negative for Aussie RMBS is that prepayment speeds are dropping like a stone (see our numbers below), which blows out the expected life of the bond and therefore reduces the expected credit spread an investor is earning relative to the spread they assumed when they bought the bond.

The deflation of the great Aussie housing bubble in an orderly fashion while the unemployment rate has fallen from 6.4% to 5.0%, GDP growth is demonstrably positive and higher than most OECD countries, the government’s budget is almost in balance, and the sovereign’s credit rating has been recently upgraded to AAA “stable”, is exceptionally positive from a financial stability perspective and credit positive for the banks on a medium term basis. This is because the banks have government-guaranteed funding via deposits, access to emergency RBA liquidity, and the prospect of government-guarantees of their senior bonds during crises. They are also constantly refinancing their home loan books with better quality assets written according to tighter lending standards. Finally, the Aussie banks in particular have slashed risk-weighted leverage in half since 2014 with the biggest build-up in equity capital in modern banking history as a result of APRA’s new “unquestionably strong” capital framework.

CCI1CCI2CCI3sp-gov-2019-03-06-graph14 (1)sydneyhpi

Hybrids vs High Yield

Notwithstanding the franking debate last year, ASX hybrids were the second-best performing asset-class behind long-dated government bonds in 2018 with an attractive 4.9% gross return, smashing the AusBond Floating-Rate Note Index (2.3%) and the Aussie equities market including dividends (down -3.5%).

Major bank hybrids are rated by S&P at BB+, which is technically in the so-called “high yield” or sub-investment grade category. Up until May 2017 they were, in fact, rated BBB-, which placed them in the “investment-grade” band, but a downgrade of Australia’s economic risk score by S&P saw both Tier 2 bonds and Additional Tier 1 (AT1) capital hybrids notched down to BBB and BB+, respectively.

We are of the view that S&P will likely upgrade these securities to their pre-May 2017 levels because it has recently put Australia’s economic risk score on trend for an improvement back to its April 2017 mark as our housing imbalances unwind and the federal budget moves back into balance.

Nevertheless, since hybrids are technically rated as high yield, this begs the question: how have they have performed compared to the global high yield market? To answer this we use the benchmark Bloomberg Barclays Global High Yield Total Return Index in unhedged form.

It is important to examine returns on an unhedged basis because currency hedging can drive dramatically different results depending on the currency you use. Historically hedging US dollar assets into Aussie dollars has, for example, artificially increased your returns by about 4% annually over the last couple of decades (and more than 2% per annum over the past 7 years) according to CBA’s research. This is because Australia has over this period had a much higher cash rate than the US. The average RBA cash rate since 1999 has been 4.1% compared to a 1.8% average Fed funds rate. Since, however, the Fed funds rate, which is currently 2.5%, has risen above the RBA cash rate (1.5%), hedging US assets into Aussie dollars has reduced, not increased, returns.

Put differently, you could mislead investors if you presented US high yield returns hedged into Aussie dollars as the basis for future return expectations if you believe that short-term cash rates in Australia will continue to remain well below equivalent US rates, which is certainly the current market consensus.

Given this, I simply focus on the unhedged returns of each asset-class to enable a true like-for-like comparison of how hybrids and high yield have historically performed before considering exogeneous factors like currencies.

I use the Solactive ASX Hybrids Index, which is the only independently produced index that exists. It starts in early 2012, which gives us a decent seven year timeframe with which to compare the risk and return profiles of these alternatives. Importantly, this period covers some big shocks for both hybrids and high yield, including 2012, 2015, early 2016, and again 2018.

A few things stand out. First, the correlation between ASX hybrids and global high yield is reasonably strong at 0.52 over the last seven years. Second, ASX hybrids have clearly much lower risk, with annual return volatility of 2.8% compared to high yield’s 5.7%. That is, hybrids have about half the risk of high yield. This is not surprising given that most of these hybrids are rated BB+ and very close to being investment grade whereas the global high yield index credit rating is on average three notches lower at B+. This was also true during the GFC when peak-to-trough losses on across the ASX hybrids market were lower than the global high yield index.

The final observation relates to the returns. The ASX hybrids market has returned 6.5% annually gross over the last 7 years compared to the global high yield market’s 5.8%. If we divide the hybrid market’s return by its volatility, the risk-adjusted return ratio of 2.35 times is also much higher than the 1 times ratio for high yield.

The much higher risk in high yield is also being driven by the fact that these assets are typically fixed-rate bonds rather than floating-rate. In Australia, the fixed-rate AusBond Composite Bond Index has historically had about 10 times higher volatility than the floating-rate AusBond FRN Index purely as a function of interest rate risk. Almost all ASX hybrids are, by way of contrast to high yield, floating-rate.

While hedging into Aussie dollars has historically made a massive difference, lifting the high yield index returns by 2.1% annually, this is not likely to be true going forward. Currently, hedging US dollar assets into Aussie dollars is reducing, rather than increasing, returns because the Fed funds rate has lifted so far above the RBA’s cash rate.

Some folks have been trying to encourage investors to shift out of hybrids into high yield because of concerns about non-taxpayers’ ability to claim cash refunds from the ATO on franking credits. North of 90% of all current investors who do pay tax will be able to continue to claim franking credits under Labor. And if Labor wins, it is plausible that the Senate will seriously dilute or block the policy entirely. So if you are thinking of switching, you need to consider several important risks.

The first is that ASX hybrids typically have much better credit ratings than global high yield and have historically displayed markedly lower probabilities of loss. As one example, the BB- average rating on the NB Global Corporate Income Trust (ASX: NBI) is several notches below major bank hybrids.

A second consideration is that retail investors are generally familiar with the Australian banks and insurers that dominate the ASX hybrids market. They will not be nearly as conversant with the hundreds of smaller companies that typically populate a high yield portfolio and which generally carry much higher probabilities of default and loss.

A third very important factor is that if you are evaluating historical US high yield returns hedged into Aussie dollars, the track-record has been boosted by about 2% to 4% annually simply through the entirely coincidental benefits of currency hedging. Today this historical tail-wind has completely disappeared and transformed into a return headwind. Accordingly, one should first evaluate the unhedged returns on the assets in question, and then consider whether hedging will be an opportunity or cost.


August RMBS Default Index Results

You can download the full report here..

Key Take-Aways: 

  • Contrary to S&P data, Coolabah’s globally unique hedonic index of compositionally-adjusted Australian RMBS default rates shows defaults trending higher after controlling for the date of the RMBS issue, the average life of the loans, the average LVRs and geographic biases
  • Coolabah has also developed another global first, which is a hedonically-adjusted mortgage prepayment index, which shows Australian prepayments (or CPRs) are slumping sharply lower to the detriment of RMBS investors given this blows out the weighted average life of RMBS bonds
  • Since April 2017 Coolabah has argued that Australian house prices would fall by 10%, a forecast recently embraced by ANZ, UBS and PIMCO amongst others. House price falls of 10% or more combined with higher default rates will very seriously threaten the credit ratings on junior RMBS tranches, from the AAA rated ABs and lower. Coolabah believes RMBS spreads are heading wider as this process unfolds over the next 1-2 years
  • On the other hand, S&P has recently put Australia’s economic risks core on positive trend for an upgrade because house prices are falling and credit imbalances are unwinding. Coolabah believes this will result in the major banks’ RAC ratios rising above 10%, earning them SACP upgrades from “a-“ to “a”, which would in turn upgrade their T2 and AT1 bond ratings to BBB+ and BBB- respectively while reducing by one notch the assumed government support underpinning their senior bonds’ AA- ratings (ie, this is also positive for major bank senior)




A new hedonic approach to measuring mortgage default risk

Measuring and monitoring the true level of mortgage delinquencies across an economy is essential for asset pricing and financial system stability. Yet public measures of mortgage default risk almost always use simple averages across pools of individual assets, including balance-sheet loans or indices tracking default risk across portfolios of residential mortgage-backed securities (RMBS). These approaches are, like median house price indices, afflicted by compositional biases that can lead to spurious inferences regarding the direction of default rates. Sources of bias include artificial changes in default rates attributable to: increases in the volume of new loans being written or securitised RMBS added to indices; changes in the proportion of transactions with higher loan-to-value ratios (LVRs); the introduction of less seasoned RMBS transactions with a lower weighted-average loan age; and/or borrower characteristics that have higher probabilities of default (e.g. tilts towards investment borrowers). To address this problem, we have developed the first known hedonic regression-based indices of mortgage default risk that explicitly control for compositional biases through the models’ characteristic-based independent variables. Whereas simple average measures of default rates across securitised loan portfolios have declined in recent years, which suggests that the risk of loss has been declining, our hedonic mortgage default index implies exactly the opposite: that is, compositionally-adjusted default rates have, in fact, been increasing sharply in recent times. Read the full paper here.


Yield Report Interview

In our series of fund manager profiles, we speak to Coolabah Capital Investments which manages Smarter Money Investments funds. Smarter Money Investments provides specialist active management in the in the liquid cash plus & short-term fixed interest asset-classes.

The Coolabah/Smarter Money Investments’ portfolio management team is led by experienced fixed-interest, economics research, and managed investment specialists Christopher Joye and Darren Harvey, who are both Executive Directors and Co-Chief Investment Officers.

Click here for more

Fixed-Rate Bonds Crater

By Portfolio Manager Christopher Joye

Australia’s main fixed-income index—against which almost all superannuation funds benchmark—delivered its fourth worst return in modern history over the 12 months to 30 June 2017. In particular, the exclusively “fixed-rate” (rather than “floating-rate”) AusBond Composite Index returned just 0.25 per cent over the last year before you deduct fees.

After accounting for Australia’s 2.1 per cent inflation rate, investors in these fixed-rate bonds actually suffered a “real” return loss of minus 1.85 per cent from what is supposed to be a bullet-proof asset-class dominated by super-safe bonds.

Indeed, the AusBond Treasury Index, which is made up of twenty-plus AAA rated Australian government bonds, suffered an outright loss with a terrible return of minus 1 per cent over the year to 30 June (or minus 3.1 per cent after inflation).

Continue reading “Fixed-Rate Bonds Crater”

Active vs Passive

One of our portfolio managers recently published a column in The Australian Financial Review on the contentious “active vs passive” investment debate, which you can download in full here. The column opened as follows:

“Buying a “passive” or “indexed” fund is lobotomised investing, predicated on the beliefs that the market is smarter than you are, investors are systematically rational, assets are always properly valued as prices move in an unpredictable “random walk” and, finally, one has no ability to identify those “active” managers that do consistently beat benchmarks. (Yes, they exist.) The problem is that every single one of these assumptions has at various times and across different sectors proven to be wildly incorrect, save for the question on your relative intellectual quotient and/or financial markets expertise. For those who never want to engage in analysis and/or are convinced they possess fundamentally inferior faculties, passive strategies may certainly be preferable to an “active” approach. And let’s be frank, many folks fall into this category. Yet let’s also bust the utterly misguided myth that passive investing is universally superior to active.”