Good index, bad index

This piece started with the idea of whether we could use the good bank, bad bank construct as an analogy within investment. Further thinking and some initial discussions suggests there might be something in it.

Formalising the bad bank analogy

Wikipedia defines a bad bank as a corporate structure which isolates illiquid and high-risk assets held by a bank. The goal of segregating the "good" assets from the "bad" assets is to allow investors to assess the bank's financial health with greater certainty. A bad bank structure also permits specialised management to deal with the problem of bad debts.

If we port this idea to investment, what would it look like if we split the market cap index into a good index and a bad index? Would that give us greater clarity on appropriate valuations? Would it more effectively starve bad index constituents of capital? Would, or could, the bad index portfolio be managed differently?

A failure to think ahead

A little further thought shows that the investment world has already started to do this. BlackRock’s new focus on sustainability includes the decision to remove “from our discretionary active investment portfolios the public securities (both debt and equity) of companies that generate more than 25% of their revenues from thermal coal production[1]”. So having more than 25% of your revenue come from thermal coal makes you a bad business; 25% or less a good one.

Or there is FTSE and the Church of England creating the FTSE TPI Climate Transition Index. Here the definition of good is having public targets aligned to the Paris agreement, so Shell and Repsol are in the index while ExxonMobil, Chevron and BP are not[2].

However, both these case studies are informal, or partial, applications of the bad bank analogy. After separating the good and bad, only the good will be held, monitored and managed. But the bad assets still exist. And a formal application of the analogy would see them subjected to specialised management.

All things are possible…

The DSM case study can offer us hope. DSM was a coal miner (Dutch State Mines) which transitioned to petrochemicals, and then transitioned again into its current form as a health, nutrition and materials business. The first transition occurred under state-ownership and may have been a necessity (the last mine closed in 1973). The second transition occurred under private ownership and, given the timing of acquisitions and divestments, may have been a deliberate strategy operating under foresight. So, leopards may not be able to change their spots, but coal miners can become health companies.

…but beware the fallacy of hasty generalisation

However, just because one coal miner transformed itself into a health company we cannot conclude that all fossil fuel companies can. This could also be an example of survivorship bias. We do not know how many coal miners attempted to transform, and therefore we do not know the failure rate.

This seems like a good point to return to our bad bank. The assets in the bad bank are bad assets. They are loans that should not have been made (the benefit of hind sight). Once, they were deemed to be worth 100 cents per dollar of loan. They now reside within the bad bank because no-one now expects them to pay back in full. The management game is now to minimise the losses. We have left behind the era of maximising returns. I imagine the management now gets labour intensive – conversations with individual borrowers about how they might change their behaviour, and in so doing protect the lender’s position as far as possible.

All analogies break down at some point, but I think ours holds for the moment. There is a set of bad business models[3], but at the time they were funded they were not considered to be bad. It is only with the benefit of hind sight that we can state the capital should not have been allocated. So far, so good. However, what makes these business models bad? Is it that we no longer expect them to return 100 cents per dollar of capital provided (the analogy is precise)? Or are they bad because we now judge their impact to be unacceptable, despite them continuing, for now, to provide an acceptable financial return (the analogy is much looser)?

What does continue to hold, in my opinion, is that the management game changes. It becomes very labour intensive, involving deep conversations with individual company managements about managing the bad business model down to zero. The managements may have individual incentives to resist decline – or even grow the bad business model.

But what about fiduciary duty?

We are currently in thought experiment territory, rather than dealing with reality. Also, I persist in differentiating between business models and companies. So, I am assuming that in most cases a company has a mix of business models, particularly large companies.

Further, I am taking it as given that we would prefer to bequeath a +1.5C world to our grandchildren rather than a +3C world, provided it doesn’t cost us too much. And if it were financially beneficial to us, then we would absolutely prefer to hand on the cooler planet. Here, I am assuming that a business model that sells fossil fuels for burning will be shut down sometime before 2050.

Combine these two assumptions and the conversations with an oil major would be about managing the mix of their business models through time – presumably growing the renewables business model (but could be a new health division!) while shrinking the sales of fossil fuels for burning.

The fiduciary duty question then becomes whether this strategy better preserves, or enhances, financial value relative to an alternative path. The most obvious alternative is business as usual, in which company managements decide how fast they can sell fossil fuels for burning, and for how long. The dangers here include the already-mentioned incentives, and cliff-edge legal risk.

A potential global defining moment

On 27 February 2020 the UK’s high court found that the ministers’ failure to take into account the UK’s climate change commitments made the Airports National Policy Statement unlawful. In plainer speak, Heathrow will not now be able to build its third runway. Adding capacity for more flights is not compatible with achieving climate change commitments. The significance of this ruling is enhanced by the UK’s hosting of CoP26 (delayed to 2021). It could be seen in time to have been a defining moment.

Applying this to our current argument, it seems feasible, if not likely, that the selling of fossil fuels for burning could, at some point, be deemed unlawful. And if that were to ever happen, it would not only be future revenues that would suffer. There would be compensation cases brought looking to claw back previously booked financial returns.

Back to the indices

We are now ready to start thinking about constructing our good index and bad index. There will be technical aspects to address, some of which we will not have anticipated at this stage. There will also be the really hard aspect – defining what is good and what is bad. Here we give one thought on each area. Clearly much more thought will be needed if this idea is to be taken forward.

Regarding the technical, there is choice between threshold and proportional representation. The threshold approach would assess each company against some metric and then allocate all of that company’s securities to the good index or the bad index. The proportional representation approach seems to sit better with the concept of a mix of business models, as it would allocate a proportion of securities to the good index, and the remainder to the bad index depending on the business model mix.

As for defining good and bad, the simplest approach would be to have a single issue, such as climate change. Even here very hard decisions will need to be made. Does any level of positive emissions automatically imply ‘bad’? And does the basis of assessment include scope 3 emissions or not? This simple approach would be equivalent to dropping ‘SG’ from ESG. It could be a means to an end, but would probably not be acceptable to those who, rightly, care about S and G issues. However, adding in multiple issues will force us into further uncomfortable decisions – how much child labour does it take to move a negative emissions security from the good to the bad index?


[2] Accessed 13 Feb 2020

[3] I am deliberately avoiding the term ‘bad companies’. I am assuming the employees and, maybe, the management are well meaning and acting in good faith. It is not necessarily their fault that the business model is now judged to be flawed.