Making sense of carbon price forecasting

Oct 21, 2022

In Brief

Over the past few years, carbon markets have captured increasing attention. This has been accompanied by a number of investable products emerging. Even so investors remain confused as to what lies ahead. We exmaine price fundamentals in this paper, which complements our carbon research published earlier this year
There are already around 100 countries identifying carbon pricing schemes as a central pillar of their net zero strategy. As a result, understanding the pathway for regional and global carbon pricing has become more important than ever
In addition, as an increasing number of corporates are also stress-testing an internal carbon price, and in some instances using this carbon price in business decision-making, providing forward guidance to carbon pricing is becoming increasingly warranted
When it comes to price forecasting, we find that those commodities with higher levels of volatility, such as European carbon prices, have significantly higher associated forecasting errors. According to our calculations, if the consensus analyst forecasting error over the past decade persists into next year, it implies an average EU carbon price of €97/tonne, compared to today’s current spot price of €68/tonne
When it comes to the track record of commodity analysts versus the respective forward curve as a predictor of the coming 12-month price, our work finds commodity forwards typically outperform the consensus analyst in the accuracy of their predictions, although mercifully for the analyst community by only a small margin
We are long-term bullish for European carbon prices reflecting fuel switching trends in the short-term and strong supply-side fundamentals over the long-term. However, a more dramatic downturn in industrial activity in Europe and an accompanying spike in risk aversion naturally pose downside risks
We also believe that the sell-off in global bond yields may have an impact as carbon prices are the net present value of future environmental liabilities discounted in today’s value. However, rising inflation ought to have a neutralising effect as it affects both sides of the equation (numerator and denominator), unlike a nominal bond.

Macro
ESG

Michael Lewis

Head of Research, ESG

1 / Analyst forecasting errors

It’s all about the vol

Not all commodities are equal when it comes to price forecasting. Figure 1 compares commodity volatility on the one hand with the average absolute analyst forecasting error for various commodities over the past decade. To calculate the forecasting error, we take the consensus forecast for each commodity at the start of the year and compare it with the final price outturn. The chart shows a clear positive correction: namely, the higher the volatility, the larger the analyst forecasting error for that commodity.

This work suggests that you can have a relatively high degree of confidence when it comes to the year ahead price forecast for gold and copper since price forecasts typically only diverge from the final outturn by no more than 10%. But when it comes to energy commodities, you need to take price forecasts with a heavy pinch of salt. Our analysis shows that commodities within the energy complex such as Brent crude oil and US natural gas have higher levels of volatility which are associated with large forecasting errors particularly compared to base and precious metals. A large part of this volatility divergence can be explained by the variation of consumption-to-inventory ratios across the commodity spectrum, which are typically lower for metals and higher for energy commodities.

When it comes to commodities within the energy complex, one commodity stands out, EU carbon prices. These have historically been the most challenging with an average absolute forecasting errors of 35%, or more than three times that of gold. Also noticeable is how the size of the forecasting errors over the past decade has been rising significantly and specifically from 2018 onwards as carbon prices have found a possible new and higher price plane.

Assuming the consensus price forecast for next year does not change between now and year end, and that the historical forecasting error between 2010 and 2021 persists into next year, it would imply an average 2023 EU carbon price forecast of €97/tonne (€71.6 x 1.35) or c.45% above the current spot price.

Click here to download the full article.

Making sense of carbon price forecasting

1 / Analyst forecasting errors

It’s all about the vol

Long View Q2: The risk of the ‘new normal’

The Global Market Portfolio

The corporate journey to carbon neutrality

CIO View