- Swedish pension fund AP4 has switched from using low-carbon equity indices to an in-house strategy with a traditional benchmark.
- The fund’s in-house strategy appears to be paying off, as it recorded a 19.2% overall return last year, with a further 23% cut to emissions in its global equity portfolio.
- AP4 accepts that tracking error will rise as it excludes more and more high-carbon stocks from the benchmark.
Use of low-carbon equity indices is growing fast as asset owners increasingly make net-zero emission pledges and look for ways to help achieve them. Accordingly, such benchmarks – which provide exposure to companies with lower overall carbon emissions than traditional market capitalisation indices – are proliferating. In the past month alone Dutch fund house Robeco and Singaporean bank OCBC (in partnership with Singapore Exchange) have unveiled new offerings.
Yet some investors – notably Norway’s $1.3trn sovereign wealth fund – have chosen not to use these types of benchmark, while others feel they have outgrown them, as Capital Monitor reported recently.
Sweden’s Fourth National Pension Fund (AP4) is a case in point. The institution, with SKr527.6bn ($56.1bn) in assets, was an early adopter of low-carbon indices, in 2012. But it has since developed a proprietary strategy for its SKr209.7bn global listed equity allocation (which excludes its SKr88.7bn of Swedish equities) and switched back to using a traditional benchmark, the MSCI All World. The in-house approach combines active and passive investing.
The fund made the move to achieve more speed and flexibility and reduce costs, says Pontus Lidbrink, who manages the global equity portfolio. AP4 can now incorporate new climate data much more quickly to keep the portfolio aligned with a pathway limiting the global temperature rise to 2ºC and the fund’s overall emission targets, he tells Capital Monitor.
“We are learning so much more by doing it ourselves,” Lidbrink adds.
What’s more, the approach appears to be paying off. In 2021, AP4’s return was 19.2% after costs, and the fund cut its carbon emissions for the listed equity portfolio by an additional 23%.
AP4 aims to reach net-zero portfolio emissions by 2040 and has an interim target of halving its emissions by 2030 from a 2020 baseline (see chart below). It had already cut the carbon footprint of its listed equity allocation by 15% in 2020 to less than half of the level it is for a broad global equity index.
The fund measures the carbon intensity of its portfolio companies emissions as well as its absolute emissions level. The latter has steadily fallen, to 12.8 tonnes of CO2 equivalent per million Swedish krona as of 2020 from 22.2 tonnes in 2015.
AP4’s in-house strategy combines a quantitative, environmentally focused approach with fundamental stock selection in the most carbon-intensive sectors. One example of a major allocation decision is that it has excluded every US energy name, “as we don’t think they’re participating in the transition”.
AP4 dubs its in-house approach “quantamental”, with high tracking error – the divergence between the price behaviour of a position or portfolio and that of a benchmark – in the fundamental parts and more moderate tracking error in the quantitative parts of the portfolio, Lidbrink says.
Unlike off-the-shelf climate indices, the fund’s low-carbon strategy incorporates both forward-looking environmental data and fundamental stock selection, Lidbrink says. It has been better able to do this since building up a team for fundamental thematic global equity management in 2020.
AP4 had started in 2012 by allocating €1bn to track MSCI’s Low Carbon Leaders Index. It gradually raised that to around one-third of the global equity portfolio using a combination of internal and externally managed low-carbon benchmark portfolios.
“By 2017 we moved everything in-house and changed from tracking low-carbon indices to our own model,” Lidbrink says. “By 2020, 100% of the global equity portfolio had a low-carbon strategy attached.”
AP4 also started incorporating Scope 3 emissions into its global equity strategy from 2020. Scope 3 usually refers to emissions not produced by the company but by its suppliers and its customers; AP4 is so far only including “upstream” emissions – that is, those generated by its suppliers.
Tracking error challenge
AP4 will, however, face the challenge of a significant increase in tracking error from 2030, Lidbrink says, when it aims to cut portfolio emissions by half again by excluding more high-carbon stocks, thereby shifting even further away from the equity benchmark.
Tracking error has long been an issue for investors looking to cut portfolio emissions. The more an allocation tilts to low-carbon stocks, the more it deviates from standard market benchmarks, such as the S&P 500 or MSCI All World.
Hence index providers and some other industry experts suggest that investors should attribute less importance to tracking error if they want to achieve net-zero portfolio emissions.
How should one address the tracking error issue? AP4 sees a certain amount of tracking error – up to, say, 75 basis points (bp) – as acceptable, but beyond that it gets more challenging.
The fund, and other investors, argue that the growing use of forward-looking data is key here.
Hence the incorporation of such data by the Church of England Pension Board when it constructed the TPI (Transition Pathway Initiative) Climate Transition Index Series with benchmark provider FTSE Russell, which launched in January 2020.
Data providers had said it was too difficult to embed forward-looking data into passive funds, says Adam Matthews, chief responsible investment officer for the Church of England Pension Board. “But once you started looking into it, it’s quite possible.”
It was a question of developing a tool for extracting forward-looking data from company disclosures and benchmarking it against the goals of the Paris Agreement, he says. “TPI had developed that approach and was in a position where we were able to model that into a passive fund.”
Keeping tracking error low for such benchmarks was challenging, however, concedes FTSE Russell’s Aled Jones, the company’s head of sustainable investment for Europe. That is especially true for the EU’s climate benchmarks, which are based on a sophisticated methodology “with lots of different inputs and quite punchy outcomes required” in order to achieve net zero, he adds.
The FTSE Russell index for the EU’s Paris-aligned Benchmark – the more ambitious of the two on emissions reduction – has a tracking error of 165bp.
“You are not going to do much for less than 100 basis points,” says Jones, “and I think clients increasingly understand that and are starting to relax the constraints around tracking error.”
It seems they will have to if they really want to move the needle on cutting emissions.