
The sovereign bond market currently stands at about $63trn, according to estimates – a figure that dwarfs the $41trn corporate bond market. But despite its size, even leading responsible investors shy away from addressing sustainability in their government debt portfolios.
Canadian pension fund CDPQ broke ground in 2017 when it pledged to “treat climate change in the same manner as we treat risk: as fundamental in our decision-making process”. The only major asset class to be exempt from the commitment? Sovereign bonds. In 2019, Niklas Ekvall, the CEO of pioneering Swedish pension fund AP4, told RI that government debt was a particularly hard topic to address in ESG terms, joking that then-US President Donald Trump probably wouldn’t have been interested in AP4’s opinion on his climate policy. Last week, European policymakers tabled a new ‘gold standard’ for green bonds that gives sovereign issuance special dispensations.
Sometimes govvies are ruled out of investment universes on ethical grounds – earlier this year, €21bn Danish fund Industriens Pension sold out of bonds from Belarus and Myanmar on human rights grounds, joining other pension investors like PGGM and APG – but little has been done on climate grounds. Two years ago, the Swedish central bank exited sub-sovereign holdings in Queensland, Western Australia and Alberta over their high carbon footprints, in a move that climate bond expert Ulf Erlandsson said at the time had “opened the floodgates” for similar announcements elsewhere; but there were none.
Victoria Barron describes sovereign bonds “the elephant in the room” for sustainable investment.
“Especially for pension schemes, because we can’t get away from them and we can’t pick and choose which bonds we’re invested in,” she says, pointing to limitations around currency exposure, and the need to hold domestic government debt for inflation reasons.
Barron is the head of responsible investment at the £57.5bn BT Pension Scheme (BTPS) – the largest in the UK’s private sector. BTPS has committed to becoming Net Zero by 2035, and a big part of that decarbonisation will come from fixed income. This is because the scheme is closed, and therefore moving into low-risk debt instruments to reduce volatility for beneficiaries ready to draw their pensions.
This shift has put Barron, like many other investors in the same situation, face to face with the elephant in the room.
“It’s an asset class that investors can’t really measure at the moment, and we need to do something about that,” she says. And that’s how ASCOR was created. Short for ‘Assessing Sovereign Climate-related Opportunities and Risk’, ASCOR is a new collaboration between BTPS, the Church of England Pensions Board and big groups like the Institutional Investor Group on Climate Change, the Net-Zero Asset Owner Alliance, Ceres, the PRI and the Transition Pathway Initiative.
“We decided we needed to do something collaborative, independent and academically supported,” explains Barron. “We need to come up with a framework to assess sovereign debt, and help us decide how we engage with sovereign debt offices and policymakers. We’re a really big pension scheme, with a good number of fund managers, but we still don’t have time to speak to all of those countries individually, so imagine what it’s like for other funds.”
Even if BTPS did have those resources, Barron points out, there is no means of assessing performance or progress, so effective engagement would be difficult.
“We can’t compare countries, because all of this national information is expressed differently,” she says. “If you’ve ever had a look at this stuff you’ll know that there are some countries whose National Statistics Office holds the data, and others will have it in the NDCs [the national climate commitments made as part of the Paris Agreement] and others won’t have it at all”.
This is a problem for data providers, too.
Elnaz Roshan leads the development of climate risk modelling for German specialist Right.Based On Science, and has spent months developing ‘temperature alignment’ metrics for around 160 sovereign issuers. The project was commissioned by BVV, the €31bn pension scheme for Germany’s finance industry, which already uses similar corporate metrics from Right.Based On Science.
Roshan describes data as the “biggest challenge” in the project.
“When it comes to doing this kind of measurement for sovereigns, the whole economy matters, so we consider national emissions,” she explains. “But there aren’t enough data sets or projections for country-level emissions, so we take the broader emissions data – for OECD countries, for example – and we allocate the right proportion of that to a country within the OECD.”
‘There might even be a version of Climate Action 100+ that brings investors together to engage with countries’ – Victoria Barron, BTPS
Dividing the emissions by GDP would favour developed countries, so Roshan splits them based on population, coming up with a ‘emissions per capita’ figure. After that, she calculates how much the world’s climate would warm up if all countries had that emissions-per-capita number, and that gives the sovereign ‘business-as-usual’ and ‘target’ temperatures – the latter being the national decarbonisation pathway required to help keep the world below a certain temperature (i.e. 2 degrees).
The UK, for example, has a business-as-usual temperature of 2.1°C under the assessment, and a target temperature of 1.7°C if it wants to align with the Paris Agreement, meaning it’s currently overshooting by 0.4°C.
Christian Wolf, BVV’s Head of Risk, explains that it has committed to reaching Net Zero by 2050, and has begun efforts to achieve this across its entire portfolio, “asset class by asset class”.
“Sovereign bonds are a relevant part of the investment portfolio, which means we need an instrument that’s able, or potentially able, to cover all types of asset classes,” he says.
Like Barron, he notes that integrating climate considerations into sovereign bond portfolios is more challenging than other asset classes, citing the risks of double counting and the need to be able to compare the results with existing assessments and ratings for corporate securities.
“A broader set of risks are relevant, and kind of different, too,” he says. “Like physical and transition risks, which both have a socioeconomic component for sovereigns”.
The ASCOR project also intends to consider both physical and transition climate risk, says Barron, “as well as policy, economic and governance factors that can influence a country's performance”.
But she insists: “The output of all that will be an assessment, not a rating, because we aren’t interested in bashing countries over the head based on a number”. She argues that it is “impossible” to factor all relevant political and economic context into such a figure.
Both BVV and BTPS plan to use their research as an engagement tool – something Wolf describes as “the most powerful and effective instrument here”.
That’s not easy with governments, though, says Barron. “I think every investor agrees that sovereigns are quite hard to engage with. With corporate investments, if you identify a risk, you could use your in-house models to reduce exposure, or – if you can’t do that – you call up management, or you attend the capital markets day, and ask them how they’re dealing with that risk. But there’s no way of doing that with a country.”
She wants ASCOR to create a way to assess investment risks and opportunities stemming from how governments deal with climate change, and their progress on NDCs; but she also thinks it will enable a new generation of collaborative engagement.
“We can go to sovereigns and show them our assessment, and they can access the same, open-source information and give us better data where we need it,” she says. “In the future, there might even be a version of Climate Action 100+ that brings investors together to engage with countries,” she goes on, referring to the influential shareholder engagement network focused on corporate decarbonisation.
For now, BVV is beginning to analyse its sovereign bond holdings based on the new temperature alignment scores, and will be working with Right.Based on Science and other, unnamed market participants to test the model on other tricky asset classes.
And ASCOR is recruiting members from Africa, Asia and Australasia that can contribute to technical know-how and funding for the project.
“We're hoping to be able to discuss the project in more detail for COP26,” Barron says. “But it’s a lot of work, and we won’t jeopardise it just to meet a deadline, because this is such an important topic for investors.”