Can You Predict a Financial Crisis?

Dr. Stephen Davies is, among other things, Distinguished Fellow in History at the John Locke Institute. He is the author of The Wealth Explosion: the Nature and Origins of Modernity. In 2006, Dr Davies predicted that within two years there would be a major financial crisis. The rest, as they say, is history.

Financial crises have been a major and recurring feature of the world economy for centuries. There are some eras when they are less common, such as the Thirty Glorious Years after 1945, and there are others when they happen with increasing frequency and severity, such as the later nineteenth century or the years since 1971. (The crisis of 2008 was only the last of a whole series going back to 1997 or even earlier). Right now, there is concern that there may be a crisis in the aftermath of the Covid-19 pandemic and there is good reason for this. However, pandemics, perhaps surprisingly, do not always lead to economic crises and collapse once they are over. Moreover, the worries about an impending crisis as big as or bigger than that of 2008 were rife before Covid-19 first appeared in the world’s news stories. The concerns were based on several factors but above all very high levels of private debt, particularly corporate debt but also household debt (with car purchase loans a particular concern).

"Pandemics don't always lead to

economic crises."

Were central bankers and finance ministers right to be anxious? Should they still be, regardless of what effect the pandemic may have? The more general question this raises is that of whether we can ever tell if a financial crisis is just around the corner. Are there things that, if we see them, should always make us nervous? Or is the very nature of financial crises that they are caused by things that are not only unexpected and unpredicted but fundamentally impossible to predict – black swans in Nassim Nicholas Taleb’s now well-known expression? If the latter, then we can never know when a panic or crisis is going to happen, only that one will happen sooner or later. If that is the case then we can only hedge against most such downturns by pursuing certain kinds of investment strategy and avoiding others but otherwise not worry about it. (Taleb offers advice on what to do in that regard.)

In one sense financial crises are unexpected almost by definition. Clearly the people caught up in them, particularly those who suffer financial losses as a result, did not expect them to happen – that is why they come as an unpleasant surprise, and it is the element of surprise that leads to the characteristic panic as confident assumptions are suddenly overturned. More generally we may say that if investors, financiers, the general public were genuinely expecting a crisis then they would behave differently, and in that case the crisis might well not happen – in other words it is precisely because enough people do not expect a crisis that there is enough of the kind of behaviour that leads to one. There is an important qualification to this however, which is the phenomenon of people expecting a crisis but thinking it is not as imminent as it actually proves to be – this is a rational but short-sighted response to certain incentives, as we shall see.

Moreover there are, of course, always people who are predicting financial doom and gloom – often market analysts. Paul Samuelson captured this in his witty but unkind quip in 1966, that the stock market had predicted nine of the last five recessions. In other words there is no reliable way of using stock market fluctuations to tell when a market is at a peak and about to undergo not just a significant correction but also something that will usher in a decline in economic activity. Many doom predictors are like the clock in Lewis Carroll’s famous logical paradox: a stopped clock is more accurate than one that loses a minute per day because it is correct twice every day. If you predict a crisis often enough and for long enough eventually you will be proved right, but that does not show that your analysis is correct or sound. The impressive cases are those of people like Robert Schiller who has correctly predicted both of the last major recessions, including the crisis of 2008.

In fact Schiller’s case shows us something: there are in fact phenomena which, while not clearly predicting a panic or crisis, should still make us very nervous and will indicate that a crisis has become more probable than before. There are a number of these and we can identify them by looking at the history of financial crises, as has been done by a number of economic historians such as Charles Kindleberger. The crucial point is that these signs are not part of the higher-level investment market, although they are connected to it. This matters because a central feature of financial crises is that in the run up to them financial markets of all kinds become in a sense deranged or detached from reality. The phenomena I am about to describe are signs of this.

"When everyone starts to tell us that nothing lies ahead but sunlit uplands, then we should really start to get nervous."

The first (and the one that Schiller used to make his accurate predictions) is when there is a market in a particular product or asset, which we are told, has no downside. In other words when there is a bubble in a particular sector or kind of product. That by itself is not enough; the bubble has to be in an area that has a significant part of the total speculative investment capital pool invested in it and which leaves major financial institutions open to exposure should it go south. The commonest place for this is real estate, whether land or buildings, and construction, but it can also be equities, as in 1928 and 1929 or the South Sea Bubble, or any one of a number of specific sectors or commodities - railway shares in the mid nineteenth century, for instance, internet company shares in the late 1990s, or tulips in seventeenth century Holland. The sure sign that a bubble is about to burst is when many well paid and apparently sober people tell you that, for structural reasons, this particular market can only keep on going up, and that you the punter must buy into it before it’s too late.

A bubble of this kind is damaging in itself when it bursts because it is a massive misallocation of investment: people are buying the asset in question, increasingly, not for use or to enjoy an income stream but simply in the expectation that the price rise will continue and they will be able to cash in a capital gain or leverage it. This reflects the second classic warning sign which is a mood among investors of what Alan Greenspan (!) once described as irrational exuberance, a kind of giddy optimism that feels there is no limit to what can be done or the profits that can be made. That is why even smaller scale bubbles in a peripheral asset that are not harmful to the wider economy in themselves (such as the current one in bitcoin or the dotcom bubble) are a warning sign because they reveal the extent to which a kind of speculative mania has come to grip previously cautious investors. The classic sign of this is when economists (who really should know better) tell us in portentous tones that we are now so smart that we have worked out how to end the business cycle. (Irving Fisher is the best-known case, in 1929, but Robert Lucas joined him in 2006.)

The third warning sign, and the one that is the most alarming, is when a story comes to be believed by many commentators that the old truths and principles of sound investment are for the birds. We are in a ‘new economy’ we are told, where those old rules do not apply: this time it’s different. This is classic hubris in almost every case and has the traditional outcome of hubris – nemesis. What all of these three indicators show is a state of affairs where the investment market has become detached from the underlying economic reality of physical assets, production and trade.

Unfortunately, the consequences of this are often very painful but modern governments and monetary authorities have a range of tools at their disposal that enable them to stave off these consequences, often for some time. The result, however, is to make those costs – caused by the misallocation of resources already mentioned, and people's misguided actions based on a belief that they are much richer than they actually are – even higher. Moreover this government action creates very dangerous incentives: many recognise the signs and realise that things cannot continue as they are but they also know that the longer you stay on the bus before it goes over the cliff, the more you will make, so nobody wants to be the first person to get off. Even if you know a cliff is coming, nobody is sure how far away it is, and – if you are very lucky – the greatest profits are made just before it is reached. If the government is going to step in, then even if you do go over the cliff you won’t lose so much, so the incentive to keep going is even greater.

Why this state of affairs arises is something to think about in a later column. But what of now? We certainly have the first of the three signs now, with simultaneous bubbles in equities, bonds, real estate, bitcoin and other crypto-currencies, and things such as old masters. However, we do not have either of the other two signs (yet), although there are alarming indications that some people are starting to get giddy again. It is when everyone starts to tell us that the bad days are behind us and nothing lies ahead but sunlit uplands that we should really start to get nervous.

The problem, of course, is that even if we know a crisis is likely, we cannot tell exactly when it will come: even if we are talking of a white swan rather than a black one, as we were around 2006, we still cannot be sure of when it will fly in.

There is also the question of what the economic impact of the pandemic will be, and whether it will trigger a panic regardless of underlying factors. But that is another topic.

616 views0 comments

Recent Posts

See All

Email:  .  Oxford: +44 (0)1865 566166  .  Princeton: +1 (609) 608-0543 

Would you like to join our mailing list?