Screening out fossil fuel producers and related companies doesn’t mean you’ll have to accept lower returns on your investments, according to the Australia Institute’s calculations. It compared a portfolio of a specially selected group of ASX 200 companies whose business model doesn’t depend on fossil fuel with the full ASX 200 list over a projected 10-year period. The findings? Getting rid of fossil fuel-related investments had “no significant impact” on returns.
We have discussed this particular report before but it worth revisiting.
This is how the Australia Institute describe their methodology:
We used these classifications to make ‘fossil free’ portfolios by screening out Tiers from the ASX 200. First we eliminated Tier 1 and Tier 2 companies from the ASX 200. Using this screen, US analysts Aperio group constructed an ‘optimised’47 portfolio excluding these stocks and simulated performance based on historic data. The portfolio tracked the broad share market very closely, achieving very similar month to month returns to the ASX 200.
In plain language – they started off with the ASX 200 and deleted 21 firms and then re-optimized the weights of the remaining firms to mimic the portfolio return of the original 200 firms. They then claim:
These results suggest that screening out fossil fuel extraction and downstream industries can have negligible impact on risk-adjusted returns.
First a quibble – the Australia Institute do not claim “no significant impact” – that does have a distinct meaning. Rather it says “negligible impact”. The Australia Institute report indicates a tracking error of 0.88%. This is what I said last time:
Considered negligible by whom? I hunted around for some benchmarks for tracking error and found this – it’s a bit dated, but damning.
An annual study by investment dealer Morgan Stanley found that the average tracking error for U.S.-listed ETFs (excluding inverse and leveraged versions) was 0.52% in 2008. If this average is weighted by assets under administration, the average comes in lower, at 0.39%.
So the tracking error just from screening fossil fuels is 0.88% while the overall average tracking errors for passive funds is 0.39% – 0.52%?
The far greater problem is that the methodology involves what finance academics call look-back bias. They have optimised known returns, after the fact, to replicate the performance of an index. The secret to success, however, is to optimize the performance of a portfolio using unknown returns, before the fact, to replicate the future performance of an index. In other words, there is no reason to believe that actual investors could do as well as the simulation suggests they can given the actual information that they would have available to them when making decisions. In the case of perfect information excluding the fossil fuel firms would result is investors under-performing the ASX 200 and have a tracking error of 0.88%.
Not only have Choice misrepresented a dodgy study, they don’t understand just how dodgy the results are.