Will Robson (MSCI): Measuring climate risk of real estate portfolios | European ESG Forum

Thanks everyone for having me. So as I accosted, I'm going to be talking about climate risk, measuring climate risk. So let me just share my screen. Here we go. So in this presentation, I'm going to answer two questions. The first is what is climate risk, and then the second is why shall I care. So on the first question, I'm going to look at the difference between exposure to hazards versus a measure of risk, and then within climate risk, make the distinction between physical risk and transition risk. And then emphasize the fact that talking about climate risk is really about the financial implications of all these kind of changing hazards as a result of climate change. Then in the second half, I'm going to talk about why should I care. This is really about what you can do with this analysis. And so it's about measuring and monitoring risk in a portfolio and then feeding those measures into portfolio management and investment decision making. It's about aligning your portfolio with the climate goals you may have set for yourself. Lots of governments and financial institutions these days are setting net zero targets, so trying to measure how closely aligned to those targets you are. The third element is around due diligence decision making, so whether that's new investments or refurbishment plans, understanding the climate impact of that. And then lastly, regulation and reporting, which is becoming much more abundant and much more demanding in terms of climate risk measurement.

So to the first question, what is climate risk? The first thing to understand about climate risk and the way that we measure it at MSCI is the climate value at risk is a forward-looking measure and it's return-based. So it's trying to take views about what may happen to the climate in the future versus now, what the costs associated with those things are, and how it might impact the portfolio. That's in contrast to carbon footprinting, so just looking at the emissions from a portfolio as of today is kind of backward-looking and it's not really talking about the future risk to a portfolio. So that's the key distinction we want to make in this measurement. Within climate risk, it's always all forward-looking, but it is made up of two elements. The first is transition risk and transition and policy risks. So this is looking at different views of the future in terms of how much warming is allowed to happen, so whether we limit ourselves to three degrees warming, two degrees, or one and a half degrees, which is aligned with net zero and is kind of widely accepted as the target that we need to hit to avoid the worst impact of climate change. It's looking at how much carbon needs to be taken out of the economy to achieve those targets and then what the financial implications of reducing that much carbon might be. On the physical side of things, it's looking at how the climate is today, how we expect it to evolve over the future, and what that might mean for the incidence, the frequency, the intensity of things such as extreme heat, extreme cold, coastal flooding events, river flooding events, cyclones, and wildfires. And so again, it's looking at how these hazards might play out over a portfolio over time, what the costs associated with those are, and how those compare to today's asset values. So combining all these things together, we're trying to express what might happen as a result of climate change and how it's going to affect the assets in the portfolio and what it might do to values.

So just a little bit more about that translation of hazards to financial risk. It all starts with assets. So on the transition side, we're looking at each individual asset in a portfolio. We're looking at the current state of energy emissions that that portfolio asset and the portfolio is producing. So that'll be a function of how much energy it's using, but also the mix of energy use types, so whether it's kind of renewable energy, coal-fired, nuclear power. That results in the current emissions. Then we look at how much we expect to need to reduce those emissions by under these different scenarios and under different regulatory regimes. And then using carbon pricing, understand how much it's going to cost to make that amount of carbon reduction. And then taking those costs, hit the portfolio out into the future, discounting those back to today's value, and comparing with the current asset value. It's trying to take these kind of abstract climate risks and measure them in financial terms, in return terms, in today's values. We do exactly the same for physical risks, but we look first at what physical risks, how we think they're going to change over time, where they're going to hit in terms of the geography around the world, the intensity of those physical risks, how much the sea level is going to rise, how much more we expect in terms of tropical cyclones for example. And then we overlay a portfolio's assets using geolocation and then again using various modeling techniques estimate the damage caused by these physical events and then discount those back to today's value and compare with the current asset value. So we're trying to estimate the potential impact on the portfolio's value today as a result of changes to the climate in the future.

So that's a very brief overview of kind of what we consider to be climate risk and how we go about measuring it. So I'll give you a few examples of the kind of analysis that can be done at portfolio and asset level and how you might use these metrics in portfolio analysis, in portfolio management, and reporting to clients or regulators. So the first chart here is around measuring the risk. And by measuring the risk, we can start to kind of identify the drivers of those risks and think about how to determine action to mitigate those risks. So this is looking at data from our UK quarterly property index and we're essentially taking every single asset within that index and calculating the climate value at risk both from transition risk and physical risk and then aggregating up to the index level. And here we're kind of reporting the numbers across different property use types of the index. So you can see on the left-hand side, the industrial sector presents the highest climate value at risk and that's mainly driven by the physical risk side of things. From the map on the right-hand side, all the blue dots are industrial assets and you can see those are very kind of broadly spread across the country because they're kind of distribution centers, logistics sheds, spread around the country along the sides of motorways, around the edges of towns, but more geographically diverse. If you compare that to the office sector though, which is just to the left, it's got the lowest risk and those are represented by the red dots on the map. You can see they're much more concentrated in major urban areas, particularly London in the southeast of the UK. Those urban areas tend to be more protected from physical risks like flooding, which is the flooding and coastal flooding is the most dominant risk in the UK. And so things like the Thames Barrier help to protect assets from those kind of risks and so the physical risk is much lower for offices than it is for industrial, for example. When it comes to the transition risk side of things, for this analysis we're just using broad market proxies rather than individual asset emissions data that we're not yet collecting from every single client. But using that, you can understand that different sectors have different levels of emissions on average, but also what's important is comparing the costs associated with that carbon emission reduction against today's asset values. And generally speaking, industrial assets are cheaper on a per square foot basis. So even if you have two assets with exactly the same carbon emission reduction requirements, the industrial assets would tend to have a higher risk because you're expressing those reduction costs relative to today's value per square foot. And so that's another example of different elements that feed into the calculations of these risk levels. So for some assets, cheaper assets, the carbon reduction costs are more significant from an economic basis because of the cheaper asset value, whereas central London offices and high-value assets can withstand bigger carbon reduction costs, all else being equal.

The next slide is from a blog that we wrote just a couple of weeks ago. You can find that on our website. It's around this kind of road to net zero that we're seeing in the run-up to COP26 in Glasgow this year. Lots of governments around the world are starting to make renewed commitments to reducing carbon and aiming for net zero by 2050 usually or earlier. And so what this chart is doing, there's a lot of information here, but the horizontal axis is really showing the percentage reduction in carbon emissions required between a two-degree scenario and a one-and-a-half-degree scenario. So the two-degree scenario, basically at the Paris Agreement back in 2015, all governments got together and aimed to reduce emissions so that we could keep global warming below a two-degree limit. Now as we head up to COP26, I think there's a broad recognition that we need to keep warming below one and a half degrees. And so everyone's stepping up to kind of make bolder carbon reduction requirements. So on the right-hand side of the chart, you have countries where the Paris Agreement commitments were already quite ambitious and so the shift from a two-degree to a one-and-a-half-degree scenario is not that great. On the left-hand side of the chart, you've got countries that had very minimal commitments to carbon reduction from the Paris Agreement, namely South Africa and Japan. So if they move towards a one-and-a-half-degree commitment, it would be a big shift in policy and requiring a huge amount extra carbon to be taken out of the economy to make that commitment. In the middle there, I've highlighted Poland, Hungary, and Czech Republic as countries that have kind of middle-of-the-range shifts in terms of policy that would be required to kind of maintain warming at one and a half degrees. On the vertical axis, you have the current carbon intensity of the real estate stock of those countries as well. So the black line across the middle is the global average. So countries like South Africa, Japan, and Italy are generally more carbon intense than the global average, whereas countries like Sweden and Norway and Switzerland have a lot more renewable energy in their power mix and are much less carbon intensive at the moment. Again, Poland and Czech Republic are kind of broadly in line with global averages. Poland is very similar to the US in terms of its carbon intensity of the real estate stock. And the size of the bubbles here is multiplying those two axes together, so it's kind of the total amount of carbon per square foot that needs to be taken out of the real estate stock if we were to move from a two-degree alignment to a one-and-a-half-degree alignment. And so you can see on a square foot basis, Poland is broadly in line with the US in terms of the extra amount of carbon per square foot that would need to be taken out.

This chart takes that same analysis, so two-degree scenario on the left-hand side and a one-and-a-half-degree scenario on the right-hand side, and looks at how those changes in carbon reduction requirements would shift into climate value at risk. So I've highlighted the CEE countries here in red. But if we focus on a comparison of Poland versus the US, because given they were so similar in the previous chart, Poland increases in the risk level from about five percent to ten percent moving from two degrees to one and a half degrees, but it only shifts up slightly in terms of the rankings. Now the United States shifts from about two and a half percent to five percent. So a smaller shift in Poland even though it has a similar amount of carbon reduction to kind of take out additionally on a per square foot basis. And this comes back to that point around capital value per square foot. So the Polish real estate stock is in general a bit cheaper, so the same kind of costs in terms of carbon reduction is going to be more significant from a financial perspective for Poland than it is for the US. So that translates into the relative valuation impact from climate change.

So beyond the risk analysis itself, you can also use metrics from our modeling to assess how closely aligned you are with your own climate goals. So I talked before about the different warming scenarios. So the three degrees is what was agreed at Paris. Everybody aimed for two-degree warming, but actually the commitments that were made, subsequent work from the UN determined that would only get you to three degrees warming. So the UN gap report issued analysis to tell us what would be needed to maintain warming at two degrees, and so that's represented in the two-degree scenario. And then the one-and-a-half-degree scenario is aligned with carbon neutrality by 2050. So for each country and property type, we can think about a reduction pathway for each of these scenarios and then kind of plot them in the chart on the right-hand side here. So if you take the red dots on here as being the average carbon intensity today, and then you need a certain amount of carbon reduced to maintain alignment with three degrees, two degrees, and one and a half degrees, plotting this line means that for any carbon intensity, you can get kind of what effectively the alignment with one of those scenarios. So in this example, take country A and property type B. Carbon intensity of 100 on average, this asset is quite efficient and so it's got a carbon intensity of only 40, and then that translates into a warming potential of nearly 2.5 degrees. So while it's much more efficient than the average asset in that country in the sector, there's still a ways to go to get to alignment with one and a half degrees. But being able to compute this number on an asset-by-asset basis and then aggregate to a portfolio level gives you some metrics to understand how closely you are aligned with net zero if that's a target.

Similarly for a due diligence use case, you can think about running this analysis on a prospective asset that you're looking to buy, an existing asset that you already have but perhaps you're thinking about trying to run a capex project to reduce carbon emissions in the asset. So you can run this analysis pre and post the expected carbon reduction. So here we've got an asset that's got average carbon emissions for that country in the sector and has a climate value at risk of a little over one percent. And then running the same asset under the same scenarios but with half as much carbon emissions gives a much lower climate value at risk. And then you can also see the impact on the warming potential there as an example.

And then the final couple of slides, I just wanted to make a point around regulation and reporting. So TCFD is a Task Force on Climate-related Financial Disclosures. It's been around for a little while now, but it's gaining a lot of momentum in terms of providing a standard for voluntary disclosure around climate risk and how that's impacting the governance of a company, the strategy of it, what risks are present in the business model related to climate change. But this is kind of evolving as a standard that many regulators and governments are using to base their own regulation around. So increasingly the financial industry is gravitating around this through choice and volunteering, but also through regulation increasingly. And then just a quick note on SFDR. This is EU regulation around financial disclosure around sustainability. And there are three broad areas that we'd like to talk about in relation to real estate. And the first is around adverse impacts and bringing some standard reporting to highlighting the kind of potential adverse impacts to a portfolio. So this will be some mandatory reporting around fossil fuel exposure and efficiency of buildings. The next column is around ESG risks, more specifically around climate risk. So the type of measurement around transition risk and financial and physical risk, although it's not prescribed on exactly how you should measure that, you do need to be able to kind of talk to the risks facing your portfolio. And so the methodology as I've just described is useful in aligning and providing metrics around that to regulators. And then the final element is around ESG impact. So this is really designed around the versioning supply of impact funds and trying to put some standards around how we measure whether a fund is actually having impact or not. And so potentially you can use analysis such as climate value at risk to illustrate and track the climate risk exposure that you have in the portfolio and hopefully the changes to your portfolio over time reducing that climate risk would be a demonstration of impact in that sense.

So with that, I think I will bring it to a close. Just want to summarize the benefits of this kind of approach to climate risk measurement. It's kind of rigorous climate science behind, peer-reviewed climate science behind the modeling, but translating all that climate science into financial impacts through financial modeling. The important thing about our modeling is that the same kind of basic methodologies apply to real estate as well as equities and fixed income. So you're talking the same language as other asset classes. And if you're trying to gain exposure or raise capital from other asset classes, then that consistency is important. And then also potential for benchmarking against your investment peer groups. There's plenty of solutions out there to give you a sense of how high your climate risk is compared to a broader peer set, but if you want to align that peer set to the same peer set that you're benchmarking your investment performance against, again that's a benefit of our solution here. And for the uses, just to reiterate, there's a huge importance around regulation and reporting that's coming up, but actually embedding this into investment decision making, into risk measurement and portfolio management, individual deal analysis, due diligence, and then also understanding your alignment with your climate goals or your net zero commitments. So with that, I will hand back the floor to any questions that there might be.

Yeah, so all that analysis was done on a per square foot basis, so we're controlling for the overall size of the real estate markets there. So Poland and the US were about the same size in the bubble, but this is the US carbon reduction requirements on a per square foot basis. So in absolute terms, the US would be far bigger than Poland, but it's normalized for the size of the real estate markets.

Yeah, so it's a function of two main things. It's how carbon intensive the energy mix is at the moment for the real estate sector in that country, but also it's how ambitious the carbon reduction commitments of that country was from the Paris Agreement compared to net zero. So under a net zero scenario, we assume that everybody has to reduce everything down to zero by 2050. Under the Paris Agreement, different countries made different levels of commitments. So some countries maybe made commitments of sort of 10 or 20 percent reduction and others were much more ambitious like 80 percent. So that's kind of driving as well.

I think it's possible. Some companies are making very... well, it has to be possible, we have to find a way. I'm not sure every sector in the economy can completely remove carbon. There's going to be some offsetting that has to happen in some elements of some markets. And it won't all just be around reducing demand, it'll be through technology improvements as well. So yeah, it's very complicated. It's a very high hurdle to hit, but unless everybody is kind of focused on it, first of all measuring their current exposure is the first step to understand how to mitigate it.