Hedging Armageddon

I’ve been watching the second season of Silo on Apple+, a series in which a character played by the brilliant Rebecca Ferguson struggles to survive in a dystopian world in which humanity has been forced underground by an environmental apocalypse. I’ve always loved dystopian fiction - the novels on which Silo are based, by Hugh Howey are excellent. Partly, it’s because dystopia acts as a warning against a naive belief in the inevitable march of human progress. Partly it’s that the shape our dystopias take says a great deal about the concerns of the present moment.

All this got me thinking about how (relatively) safely we stay anchored to historical experience when thinking about tail risk in financial markets. I remember early in my career working with a risk manager who spoke with grim relish about the increasing likelihood of “five-sigma events” (meaning five standard deviations away from the mean). Our thinking about hedging tail risk and tail dependencies has evolved a great deal since then, but we still remain wedded to models that are in some form of dialogue with the past.

In our Markets Meeting this week, my colleague Rupert Goodall presented the following slide as a bit of fun at the end of the session. It shows the performance of the Trend factor - a generally reliable hedge against crises - at different times on the famous Doomsday Clock, showing how close we are to the end of days.

The idea of hedging Armageddon has come up in a number of investor meetings recently. Our investors have generally very long-term time horizons - they are pension funds and Sovereign Wealth Funds who need to look forward multiple decades when thinking about their investment horizons. It’s striking to me that culturally we can’t get enough of bleak visions of the future, but that we in finance have remained relatively constrained in what paths we permit ourselves to imagine the world taking. As one very long-term investor put it to us in a meetings a few weeks ago: “If I believe that there is a small but not zero possibility that the world will suffer some sort of apocalypse, is it not my duty to reflect this in my allocations?”

So what if we moved beyond the ways we usually advise our investors to hedge against extreme negative market moves (if you want to read more about our attempts to crisis-proof portfolios, I strongly recommend The Best Strategies for the Worst of Times by Otto Van Hemert , Campbell Harvey , Henry Neville et al)? What assets might provide some sort of genuine hedge for portfolios against the long dark tail of Armageddon?

Firstly, not all apocalypses look the same. The performance of markets under different scenarios are likely to vary dramatically , while the geographical distribution of risk will change based upon which of the various apocalypses we model. Let us consider what feel like the three most likely catastrophes (at least if our novelists and filmmakers have got things right): nuclear war, environmental collapse and a global pandemic (one more durable and deadly than our recent experience of Covid).

Let’s start with nuclear war. The first call, I suppose, is on defense stocks. This is something Henry Neville spoke about at our Stockholm Conference last week – the idea that a multi-polar, more isolationist world will require military spending to rise back to the kinds of levels we saw in the height of the cold war seems reasonable to me. As such, we should expect a significant uptick in the valuation of the major defense contractors.

The outperformance of defence stocks ought to be visible both as a long-term secular move as countries increase military spending and in short-term spikes in response to moments of heightened crisis, as discussed in this excellent paper by Viksit Verma makes clear: The Impact of Wars on US Defense Stock Returns.

Historical precedent isn’t hugely useful here. If we look back to times in which the world has come dangerously close to the edge, it looks like the markets weren’t hugely bothered. As this paper, about the markets’ response to the Cuban Missile Crisis, makes clear, the “Kennedy Slide” was already well underway by the time the real tension kicked in in in mid-October 1962. Obviously there were geopolitical tensions around in the run-up to the crisis, but the market had appreciated by more than a quarter over the course of 1961, and, if anything, the crisis and its resolution gave the markets a new sense of direction.

In nuclear scenarios, geographical diversification would be paramount. There’s a reason that preppers and tech tycoons are buying property and land in Australia and New Zealand. Thinking of potential loci of conflict, the Southern Hemisphere immediately appears a safer bet, with no countries on that side of the globe possessing nuclear weapons. The rising value of land in remote, politically stable regions like New Zealand and Patagonia reflects their appeal as what are known as “lifeboat geographies”, with their isolation offering a hedge against the breakdown of more heavily-populated Northern Hemisphere states. Permutable.ai runs a nice piece of machine learning that scrapes huge amounts of data to create a real-time map of global geopolitical risk.

As far as climate risk goes, I wrote about this some time ago in my Forbes column. The important thing to recognize is that climate risk is a clear an inescapable threat, and while its mean impact will approach gradually, tail events are already with us – and they are creeping towards the mean. There will be clear beneficiaries from climate change: shipping firms able to exploit northern sea routes; miners in previously inaccessible polar regions; and, above all, Greenland (perhaps Trump’s attempt to buy it from Denmark a few years back wasn’t such a bad idea after all).

For pandemics, we have a recent playbook to turn to: you need to hedge the immediate drawdown risk, then invest in the pharmaceutical firms and technology companies who will eventually – because that’s what humans do – deploy extraordinary ingenuity to solve seemingly insoluble problems.

?One of the great uncertainties in any post-apocalyptic scenario is whether conventional market mechanisms, including valuations, would survive at all. Markets rely on trust, functioning infrastructure and enforceable contracts, all of which would likely break down. Assets like land may retain intrinsic value, but liquidity would likely vanish and the ability to convert holdings into other forms of wealth (or even use them) could be severely constrained. Currencies would likely be so volatile as to be valueless, while a collapse in property rights enforcement could lead to widespread confiscation or squatting.

While my crypto friends are currently celebrating their winnings, I’m not sure that digital currencies are the answer in this kind of scenario. It may be that having currencies that are outside the remit of national governments offers a degree of narrative appeal, at least. But their utility in a genuine Armageddon scenario depends on the survival of internet infrastructure and digital networks.

It leaves us with gold: a time-tested store of value in scenarios of societal collapse. It is portable, fungible, and has been rising dramatically in value even as real rates have fallen – breaking a longstanding historical relationship. Perhaps the markets know something we don’t…

Or perhaps, at least, people are thinking hard about how to hedge themselves against those parts of the tails that are not captured by traditional crisis alpha strategies. The gradual ticking up of central bank gold holdings suggests that this may already be under way.