The occasional baiting of Michael Fish over his famous weather forecasting error in 1987 has happened again.
Andy Haldane, the Chief Economist of the Bank of England, recently issued a bit of a ‘mea culpa’ when he said that the failure to predict the financial crisis and an overly pessimistic Brexit forecast was a ‘Michael Fish Moment’ for economists.
He also talked about economics learning from the changes the MET office made to its forecasting methods and the use of big data.
I can’t argue with that last point. But there is an extraordinary historical oversight being made here.
Economics had already experienced plenty of its own ‘moments’ to learn from between the time of Mr Fish’s forecast and the time of the financial crisis. And you only needed to wait about 48 hours for the first.
Back in 1987, immediately following the weekend that Mr Fish made his erroneous forecast, there was another event that was a bit more global in scale. A financial crash known as Black Monday when stock markets collapsed around the globe.
In the States the Standard & Poors index of the largest 500 stocks fell by a quarter and it’s futures contract by 29%. I know this first hand as I was attempting to trade them over the day on behalf of clients.
As an event it completely challenged the prevailing economic thinking about ‘modern portfolio theory’, ‘the efficient market hypothesis’ and ‘random walks’. A one day move in equity markets of that size was supposed to be virtually impossible.
When I say ‘virtually’, we can actually put a number on it. Given certain assumptions, it’s one chance in ten to the power of minus 160. Or as it was put in one academic paper in 1995 –
‘Even if one were to have lived through the 20 billion year life of the universe and experience this 20 billion times…..that such a decline could have happened even once in this period is virtually impossible.’
In modern parlance this was an ‘epic fail’.
Two years later I had a front row seat in Tokyo at the bursting of the Japanese asset bubble, the ripples from which are still being felt today. And then there was Black Wednesday as the UK left the ERM, the Savings and Loan debacle in the States and the Asian Financial Crisis.
Then we had the LTCM bust where two Nobel prize winning economists from 1997 lost $4.8 billion in 1998.
The turn of the century saw the dotcom bubble burst and then, finally we get back to Mr Haldane’s Michael Fish moment of the financial crisis.
It seems appropriate to use a weather analogy. And what those events demonstrated was, not only did financial floods occur more often than economic models predicted, they could be larger and arrive in batches. The financial dams built against them were also too low, poorly made or might not even exist. We even had examples where risk analysis was so poor in some markets that financial institutions were effectively buying flood insurance from other institutions lower down the hill.
But the most damning accusation is the likelyhood that, due to over confidence in their models, Central Bank policy increased the risk and scale of the last big flood in 2008.
If the dismal science of economics really does attempt to learn from other disciplines such as meteorology in order to counter that over confidence that would clearly help – but a mirror and a larger dose of humility would probably help more.
The quote is from ‘Recovering Probability Distributions from Contemporaneous Security Prices’, Jackwerth and Rubinstein (1995) and is referred to in the book ‘Black-Scholes and Beyond’ by Neil A. Chriss
- past performance is no guide or guarantee of future returns;
- the value of stock market investments can rise and fall over time, so it is quite possible to get back less than what you put in, depending upon timing;
- this blog does not constitute financial advice and is provided for general information purposes only.