Case study by Schroder Investment Management

  • Signatory type: Investment Manager
  • Region of operation: Global
  • Assets under management: £470.5bn

Why ‘environmental disruption’ could impact long term investment returns

Over the last few years, we have been considering the economic and disruptive forces we think will shape the future investment landscape. One of these forces is ‘environmental disruption’.

The incorporation of climate change into long term return estimates is important for two reasons.

First, clients such as pension funds, endowments and insurance companies often need to think about asset allocation over a long-term horizon to be consistent with the liabilities for which they are responsible. As a result, they often ask our opinion about the return and risk assumptions that they are planning to use in this process.

Second, our portfolio construction process involves some modelling and consideration of possible optimised portfolios before we overlay our shorter investment views to arrive at suitable portfolios for a wide variety of client outcomes. This modelling process incorporates long term return and risk estimates from our Economics team which we strive to make as realistic and rigorous as possible.

We are often asked by our clients how we became comfortable with making this change in our assumptions. Our Climate Progress Dashboard had already shown that a significant amount of work needs to be done across government policy, innovation, the existing resource base, and finance to limit global warming to two degrees. That is the level which we felt would naturally start to influence investment returns in the longer term. Our Economics team have evaluated the academic research and considered which were the most relevant in helping them estimate the likely impact on returns.

How our ‘three step’ process can measure the impact of climate change on financial returns

Ultimately, the potential ways through which climate change could impact growth and financial returns are too numerous, and indeed often unknown, for us to hope to model every moving part, particularly considering data constraints in poorer economies.

Instead, we chose to adopt, after collaboration between our Economics and Sustainability teams, a three step framework for incorporating climate change into our 30-year return assumptions.

The first step focuses on what happens to output as temperatures rise, which we refer to as the ‘physical cost’ of climate change.

The second considers the economic impact of steps taken to mitigate those temperature increases, or the ‘transition cost’. This is slightly more complicated, in that there are a range of possible transition scenarios.

Finally, we adjust for the effects of stranded assets. This is where we take account of the losses incurred where oil and other carbon-based forms of energy must be written off. This is when it is no longer possible to make use of the assets and they are left in the ground.

As can be seen from Figure 1, asset class returns (particularly equity returns as shown) are likely to substantially change in a negative way using this approach for a number of countries but a handful are likely to benefit from climate change. This will have a direct impact for the modelling conducted on behalf of our clients over this time horizon and also for the results from our optimiser.

Figure 1 - The impact for equity investors: a big reshuffling of relative returns, and large absolute reductions

Figure 1

Source: IMF, IEA, World Bank, US Census Bureau, Schroders

In fact, much of the climate change that has already occurred is baked into this 30-year time horizon and efforts to limit warming will only yield benefits after 2050.

We do not publish return estimates over longer time periods (as most clients do not have longer time horizons than this) but if we did, we would expect some of the mitigation measures to start to have a positive impact on some of the returns that have been significantly impacted. That is against a world in which no mitigation is attempted.

Example – Using our analysis to manage portfolios

As can be expected, there have been many conversations about the analysis and its impact on expected returns. We dynamically manage portfolios for our clients so in reality we might overweight a market where we have environmental, social and governance (ESG) concerns if we think that the valuation, for example, still accounts for the impact of these, or where we do not consider that the ESG issues will impact over the time frame of our trade. Indeed, we often take active shorter-term currency positions in our portfolios and so far there seems to be little evidence of a link between ESG and currency performance over this time horizon.

However, as can be seen in Figure 2, an unconstrained optimiser using the 30-year assumptions with and without the climate adjustments results in different efficient frontiers. A longer-term horizon investor using these new estimates may assign higher allocations to, say, the UK equity market and less to the Australian market to achieve the same risk/return profile. The efficient frontier for the climate change adjusted returns cannot achieve the same level of returns (the green line is shorter than the grey line) because the returns of the riskiest assets are reduced when climate change is considered.

Figure 2 – Climate adjusted returns result in a new efficient frontier

Figure 2

Risk is measured as expected volatility of the portfolio returns. An efficient frontier is the portfolios that offer the higher return for a given amount of risk (or the lowest risk for a given amount of return). Efficient frontiers for a USD-based portfolio, built with a universe of regional equities and bonds, optimised to achieve a return of 5%, with maximum 80% equities, maximum 20% credit, maximum 30% government bonds, and cash between 5 and 10%. For illustration only.

As can be seen, at lower levels of risk, the new assumptions result in the ability to find portfolios that have slightly higher returns - this is because the optimiser chooses to switch out of bonds that are expected to experience lower returns due to climate (such as Australia) into countries with significantly better returns (such as Canada).

However at higher levels of risk, portfolios are likely to have slightly lower returns because some equities that were expected to deliver higher returns such as Pacific Basin excluding Japan and Emerging markets are likely to be impacted by climate more than some lower returning regions. At the highest level of risk, the climate adjusted return portfolio is unlikely to be able to reach the upper level of returns possible before the climate adjustment.

So far we have focused on climate in our work on future returns but we have also developed tools that help our bottom-up stock selectors to better understand ESG risk and for these we consider a much wider range of factors. Our longer-term aim is to incorporate more of these factors in our top-down asset class analysis where we believe their inclusion is relevant.

The forecasts included should not be relied upon, are not guaranteed and are provided only as at the date of issue. Our forecasts are based on our own assumptions which may change. We accept no responsibility for any errors of fact or opinion and assume no obligation to provide with any changes to our assumptions or forecasts. Forecasts and assumptions may be affected by external economic or other factors.