This article was written by Kartik Ghia, Systematic & Index Solutions Researcher, Bloomberg.
Assessing the role of commodities
Over the past decade, whether viewed through the lens of climate risk or sustainable economic development, sustainable investing has become an increasing prominent theme within investors’ discussions. While commonplace in equity and fixed income portfolios, sustainable investing has yet to be fully embraced by commodity investors, not least because of questions surrounding the objective of such commodity portfolios and the lack of any consensus about what this means with respect to portfolio construction. Furthermore, there is a fundamental misunderstanding about sustainability principles applied to broad-based benchmarks versus more concentrated thematic baskets—which often focus on specific sectors.
The distinction between the two investment approaches is directly related to the role of commodities within investors’ portfolios. Asset owners tend to view commodities as having three main uses: to provide diversification to equity and fixed income holdings, hedge inflation risk and deliver exposure to the economic cycle. Given differences in sector return profiles, production levels and economic roles, meeting the three objectives requires exposures to a broad-basket of commodities—for example the Bloomberg Commodity index which contains 24 US dollar denominated commodities. At the forefront of investors’ minds is whether introducing a sustainable objective maintains the asset allocation properties of their existing commodity portfolios. Significant differences would likely necessitate a re-evaluation of the role of commodities within the policy portfolio. In contrast to equity and fixed income markets, commodity portfolios contain far fewer securities and correlations between those securities tend to be relatively low. This can make the task of re-weighting commodity exposures tricky given the need to balance sustainability objectives and maintaining the existing returns profile.
Why investors should care about sustainable development
Sustainable development can broadly be defined as a plan for economic development that is mindful of the depletion and/or degradation of natural resources. Measures such as greenhouse gas (GhG) emissions, water usage and biodiversity are among the more popular measures cited by practitioners as indicative of natural resource depletion. This, together with GhG emissions being one of the key inputs into the assessment of the consequences of climate change, makes GhG emissions a natural starting point from which to evaluate portfolios.
Unlike for equity and fixed income portfolios, commodity investors tend to get exposure via exchange traded derivatives. Given futures holdings do not directly affect physical demand and supply and that the derivatives volume, for a given commodity, is multiples of the physical volume, why do financial investors need to care about the underlying GhG emissions?
There are two important reasons. First, the derivatives’ returns reflect movements in the price of the physical underlying. As it stands, without the commodity, there is no commodity futures return. To gain exposure to the return profile offered by a broad benchmark, the physical commodity must be produced. This, in turn, creates GhG emissions from the production processes. Second, the price sensitivity of commodity producers and consumers differ, with producers generally more price sensitive (this is referred to as the ‘elasticity of supply and demand’ respectively). The result is that output tends to decline more during periods of falling prices than demand falling during periods of equivalent price rises.
Academic research on corporate investment under uncertainty (e.g., Investment under Uncertainty, Dixit and Pindyck), which can be interpreted as either increasing variability in profitability or declining profitability, highlights a negative correlation between capital expenditure and increasing uncertainty. Since capital expenditure typically entails purchasing new equipment, this suggests increasing expenditure results in more efficient production processes—often both in terms of output per unit and GhG emissions. Long-only investors provide price support to producers seeking to hedge future output (funding support), which can be viewed as facilitating productivity improvements.
The framing of long-only investors providing funding support together with the use of derivatives-based instruments has important implications. In the case of equity portfolios, a commonly used measure is portfolio emissions—meant to be representative of the GhG emissions ‘owned’ by the investor. Investors in commodity futures cannot lay claim directly to the underlying physical asset yet support prices resulting in a benefit to producers hedging output. A metric reflecting commodity investors’ role is an adaptation of the portfolio emissions measure. For each US dollar invested in a commodity portfolio, funding support can be defined as a weighted sum of the GhG emissions associated with each of the commodities. This measure is a macroeconomic analogy to portfolio emissions; representing the weighted amount of emissions supported by the futures investment. Translated to the notion of a sustainable portfolio, the measure of interest is the difference in funding support between the traditional benchmark and the sustainable portfolio—with the difference driven solely by the re-allocation of commodity weights.
How is the data collected?
Typically, GhG emissions data is collected via companies’ annual and sustainability reports. However, in this case, as each commodity is produced by multiple companies, many of which are private (and hence non-reporting), this is not an option. Furthermore, whereas the data collected by companies tend to be defined by scope 1,2 & 3 boundaries, it is more meaningful to define the emissions cycle of commodities according to the production processes that are required to produce those commodities. In the case of commodity futures, this can reference the futures specifications of the underlying commodity, which can be viewed as the equivalent of scope 1 and 2 emissions as reported in companies’ sustainability reports.
The emissions data is estimated using an approach called Life Cycle Assessment (LCA). This relies on a specified model of the production processes along with data sets which determine input values and parameter settings. Typically, datasets collected by industry and academic researchers are fed into modelling software to construct emissions estimates per commodity. Naturally, the results are dependent on both the dataset and modelling parameters. Accordingly, selecting representative datasets (process and geography wise) are important, as is selecting parameters according to some generally accepted standard (for example the IPCC 2021 Report). The estimates can be refreshed on a regular frequency to account for new datasets, model updates and shifts in reporting standards. Given the potential complexity of timely management of model updates and collecting and assessing the quality of new data sets, it can be operationally intensive to manage the estimation process in-house. An alternative is to use consulting firms that specialize in the field.
For illustration purposes, Figure 1 displays the result of the LCA-based estimates calculated in the first half of 2023. Grouped by sector and whether they are primary or derived in nature (of which we discuss more a little later), the heatmap represents a rescaling of the raw estimates.
Implications for portfolio construction
The notion of sustainable development implicitly assumes the pattern of economic development will remain approximately intact, while the adverse impacts to the environment will be lessened. A corollary to this is that the re-allocation of funds needs to be restricted to commodities that are at least weak substitutes. This is further supported by the nature of the energy sector—which is an input to all other sectors (agriculture, industrial metals, precious metals, and livestock). A straightforward way to address these two points is restricting the re-allocation of weights to commodities belonging to the same sector. The final consideration is to try and minimise double counting emissions by distinguishing between primary and derived commodities.
Large differences in liquidity between commodities (measured for example by average US dollar volume traded), even within the same sector, suggests tilts need to be a function of both the benchmark allocation and a weight assigned by the GhG emissions—the latter using an inverse measure of emissions, similar in form to that used when constructing equal volatility contribution portfolios.
Another implication of the sustainable development concept is the requirement of the GhG measure to be based on production units rather than on monetary value. Production processes evolve slowly as do energy-mixes. In contrast, commodity prices have an annualized volatility of between 15% and 45% and are uncorrelated to changes in emissions per unit of production. A good example is the large swings in prices over the past two years such as nickel and natural gas, with little to no change in emissions per output.
Commodity allocations are typically used for three purposes: providing diversification for equity and fixed income holdings, inflation hedging and exposure to the economic cycle. Asset owners seeking to adopt a portfolio orientated to sustainable themes are cognizant of the trade-off between maintaining the asset allocation and liquidity features of the portfolio, versus tilting away from commodities with relatively high GhG emissions per unit of production. Accounting for low inter-sector correlations, the significant differences in average liquidity by sector and the impact on economic development, controlling sector deviations relative to the benchmark provides a means to control for the asset allocation objectives above.
Two important metrics are sensitivity to inflation and correlations with equity and fixed income markets. There are two inflation measures—the first is realized inflation (quarterly changes in US CPI) and changes in inflation expectations (quarterly changes in the US 10-year breakeven rate). These are commonly referred to as ‘inflation betas’ and provide investors’ with an indication about the inflation hedging capabilities of the investible asset. Using monthly data since 2012, the inflation beta for the BCOM benchmark was 2.1 and 17.9 for realized inflation and changes in inflation expectations. Correlations with US equity and fixed income were 0.5 and 0 respectively.
Access to a carbon tilted benchmark
The Bloomberg Commodity Carbon Tilted Index (Bloomberg ticker BCOMCA Index <GO>) uses a GhG emissions per unit of production metric to reweight the BCOM benchmark. The carbon-tilted index comprises of five sector-based GhG emissions tilted portfolios, combined as per BCOM sector weights. Portfolio rebalancing in carried out annually. The equivalent inflation betas for the BCOM Carbon Tilted benchmark was 2.3 and 18.8 respectively. Correlations with US equity and fixed income are virtually unchanged (0.5 and -0.1 respectively).
The inflation betas together with the asset class correlations suggest the BCOM Carbon Tilted benchmark can be used in a similar manner to the BCOM benchmark for asset allocation purposes—for example within a policy portfolio. It has both displayed diversification potential along with a positive correlation to both types of inflation, which recognizing the emissions emanating from the assets from which returns are derived. Since 2012, the outcome of the tilting of commodity weights has been a reduction of approximately 20% in the associated reduction in emissions per unit compared to the BCOM benchmark. Via the tilting procedure, the benchmark can be customized to account for the trade-off between the reduction in associated emissions versus annualized tracking error against the BCOM benchmark.
The data and other information included in this publication is for illustrative purposes only, available “as is”, non-binding and constitutes the provision of factual information, rather than financial product advice. BLOOMBERG and BLOOMBERG INDICES (the “Indices”) are trademarks or service marks of Bloomberg Finance L.P. (“BFLP”). BFLP and its affiliates, including BISL, the administrator of the Indices, or their licensors own all proprietary rights in the Indices. Bloomberg L.P. (“BLP”) or one of its subsidiaries provides BFLP, BISL and its subsidiaries with global marketing and operational support and service.