Professor Riccardo Rebonato: “A significant risk re-pricing may be overdue.”

How well do you think institutional investors are approaching climate change and the risks associated with it?

There are many encouraging efforts to come to terms with the financial implications of climate change. Understanding what the climate future might look like is an essential first step in being prepared. Several international organisations have provided climate scenarios, which are invaluable.

However, so far these scenarios have been devoid of any assessment of their likelihood – relative or absolute – and this makes them difficult to use. Faced with a garden-variety market scenario, financial planners routinely build probabilities ‘in their own heads’ and qualitatively assess whether the scenario is worth losing sleep over or not.

But this is only possible because of a century-long experience of market crashes, credit crises, asset bubbles, interest rate hikes and the like. This ‘institutional memory’ is absent in the case of climate scenarios because we have not yet encountered this situation in the history of Western civilization, let alone of financial markets.

Any portfolio manager worth her salt can express an informed opinion about whether a market scenario such as ‘yields move up by 100 basis points in a month’ is reasonable or not – and she does not need to run a formal model to arrive at her conclusion.

So how can investors assess whether breaching the 1.5-degrees target in 20 years’ time is likely or not?

This is why investors and financial planners need science-based models to assess what they should worry about and what belongs to the category of ‘meteorite risk’. This lack of any probability assessment is a big gap in what is being provided to investors.

One should also keep in mind that standardised scenarios are great for comparability and reporting but can easily generate tunnel vision and encourage group think. The ‘wisdom of crowds’ is good indeed when it comes to estimating averages but fails badly when it tries to assess the tails of distributions.

So, my recommendation to investors is not to think that the ‘canned scenarios’ available cover all that can happen. Instead try to embed climate scenarios in the wider macro-financial picture. For instance, if subsidies prove more politically palatable than carbon taxes, and if subsidies – as it happening in the US and in Europe – acquire a progressively protectionist focus, what will the consequences be for trade agreements, globalisation, etc?

Or if the 150 million people living in the already extremely dry and agriculturally ‘marginal’ Sahel area were forced to migrate because of a modest temperature increase, what might the economic and political repercussions be for European countries?

Nobody can know with certainty how severe climate change in itself will be, but the nature of the problem is that it is deeply pervasive and has ramifications in every aspect of the economy.

How do you see the debate surrounding climate change, net zero and investors? Is it going in the right direction or taking the wrong course?

There is no doubt that emission abatement must play a key, and increasingly important, role in controlling climate change. Investors can play a significant part in this respect.

However, every scientist and the Intergovernmental Panel on Climate Change agree that all paths to a manageable level of warming by the end of the century require substantial carbon removal.

Unfortunately, we have very few practical carbon removal options, such as afforestation and reforestation, that can be deployed in scale now. Even the ones that we do have are no panacea, for instance, because of competition for land from afforestation.

Other removal technologies are expensive and require a lot of energy that must be provided by renewables unless we want to use up our carbon budget.

Unfortunately, talking about non-abatement routes to climate control is unpopular because of the perceived risk of moral hazard. However, if we fail to devote re- sources to direct carbon removal, the temperature outcome by the end of the century will be well outside the Paris targets.

So, we must indeed think of reaching net- zero soon – the sooner, the better – but we must start to think seriously about net-negative as well. All ‘experts’ agree on this point, but the importance of substantial carbon removal has rarely been on the radar screen of politicians, and, arguably, of investors.

The same investors should also realise that if the transformations of the economy associated with large carbon removal do not take place, then we should brace ourselves for much higher temperature outcomes.

Big transformational changes are afoot whether we act decisively or we don’t. The net-zero target via abatement, useful as it is, can create complacency: it is a necessary first step but not the be-all-and-end-all of climate control.

You have studied the climate risk premium in detail: what it is and why should investors care?

All risk premia depend on whether the security in question pays well or badly when we feel rich or poor. Equities attract a positive risk premium because an equity portfolio pays badly when the whole economy is in the doldrums. Investors do not like these ‘fair-weather friends’ and, therefore, pay less for them – lower price, higher expected return.

Conversely, US treasuries and bonds attracted a negative risk premium up to the Covid crisis because they were perceived as providing a hedge to equity wobbles: the ‘Greenspan put’ – that is, to act as insurance by performing well when the rest of the portfolio was doing poorly. So, the same expected cashflows can be valued differently if they materialise in good or bad states of the economy.

Investors should care a lot about this because the risk premium can be a substantial part of the expected return from an asset. Indeed, part of the current high treasury yields in the US and the UK are due not just to inflation expectations but also to the fact that the negative risk premium has evaporated.

This has happened because investors are no longer willing to pay an ‘insurance premium’ because the insurance policy doesn’t seem to work anymore.

When it comes to hedging climate risk, when is it possible and when should investors do it?

If an investor has identified a robust hedging instrument, and wants to be insulated with respect to that risk, the hedge should be put in place as soon as the risk is identified. In some cases, deploying the insurance strategy continuously is too expensive: as in the case of out-of-the-money equity puts.

However, it is better to buy more out-of-the-money protection than to try to time the entry and exit points for the hedging strategy.

Having said this, recognising that a portfolio is exposed to a risk factor, such as climate, doesn’t automatically mean that the risk should be hedged away – it all depends on how handsomely the risk is rewarded and on the ‘staying power,’ such as internal or limit constraints, of the institution.

If an institution decides that it wants to ‘ride the risk’ – and extract the risk premium – then it should make sure that its risk-budget, for example, value at risk utilisation in ‘normal times’, is well below the limit. If not, the institution will see itself forced to liquidate the risky positions at the first sign of turmoil.

Therefore, are green assets hedging against risk or adding to it?

We have few empirical answers for this ‘trillion-dollar question’ and the empirical studies conducted so far have given contradictory answers. This is why state-of-the-art theoretical models can give investors some help.

Currently, a robust finding of these models is that the largest climate damages materialise if the global economy is ring on all cylinders: because of the link from economic expansion to emissions to concentrations to temperature increase to damages.

So, an asset that paid well in states of high climate damages, let’s call it ‘green’, would pay well when equities pay well and would, therefore, attract a positive risk premium.

One important observation: investors must distinguish between risk premia ex-ante and ex-post. If a security is perceived to perform badly in poor states of the world, its lower price already reflects this information, and the investor, therefore, enjoys the positive risk premium today.

However, if investors realise tomorrow that the same security pays badly when everybody feels poor, then the downward price adjustment will only occur tomorrow, and today’s holders will post a loss. There are reasons to believe that current valuations reflect climate risk partially at best: investors beware.

Can you explain the structure of the climate risk premium? Are long or short-dated assets more strongly affected?

As far as physical risk is concerned, the assets that could attract the highest risk premium – positive or negative – are long-dated, as it is long-dated cashflows that are more likely to be affected by physical climate risk.

Somewhat surprisingly, extremely long-dated assets – there are some treasury bonds with 100-year maturity – are not affected as much because, sooner or later, we expect the climate problem to be brought under control.

So, the sweet spot for physical climate risk premia is long, but not extremely long, dated assets.

So how does the climate risk premium depend on what you describe as future abatement policies?

The climate risk premium depends crucially on future abatement policies. If we abate little, then climate damages are going to be much larger, and the climate sensitivity of cashflows – the ‘climate beta’ – to climate outcomes will also be correspondingly larger.

An estimate of the magnitude of the climate risk premium is, therefore, a joint estimate of whether the largest climate damages will materialise when the economy is strong or weak and of the aggressiveness of our climate policies.

What I would add here is that the likelihood of abating too little is much, much higher than the likelihood of abating too much – so the risk premium has a similarly skewed distribution.

How robust then are the results to climate uncertainties and model limitations?

There is huge model uncertainty, and all projections should be associated with large error bars, which are too frequently forgotten.

Having said this, we do have valuable information, and the defeatist view that the problem is so complex that models are of no use is not constructive. The key trick is to use all the information we have while keeping in mind what we do not know. We should remember that knowing what we do not know is useful in itself.

Having said this, one of the most robust findings of climate/economy models is that we can expect the largest climate damages in strong states of the global economy, especially if robust growth occurs in yet-to-develop countries.

All models concur that the joint effects of demographic and economic growth of poor countries will have a profound effect on climate outcomes. What the models cannot tell us is whether this growth – if it happens – will be fuelled by renewables or fossil fuels.

You have mentioned that the market may be asleep at the wheel on climate change: what do you mean by that?

If we do little to tackle climate change and keep on kicking the climate ball into the high grass, temperature increases can take us to levels never seen by Homo Sapiens. Just 3-degrees would be unchartered territory. If, instead, we get our act together and act decisively, the whole economy will have to be rewired – profoundly and in a short time.

Either outcome should have a marked effect on valuations, either in the aggregate or at the sectoral level. Yet, the signature left in asset prices by these events is barely detectable. This makes me think that a significant risk re-pricing may be overdue.