A study of actual interwar commodity trades by John Maynard Keynes finds that pricing was as efficient as modern derivative trading using the Nobel Prize-winning BSM model developed a half-century later.
One of the most important
innovations in modern finance is the Black-Scholes-Merton (BSM) model of option
pricing, developed in the early 1970s and recognised in 1997 by the Nobel Prize
in Economics. It is often thought that before the 1970s traders had little idea
how to accurately price options.
Yet a new Cambridge
Judge Business School study that looks at the actual commodity option trades of
the famed economist John Maynard Keynes between 1921 and 1931 suggests that
traders in that interwar era may have been able to “intuit” what
amounts to fair value and adjust their prices to the market environment – five
decades before the modern BSM model.
“We find that
interwar option prices were no more mispriced than in modern times and were as
sensitive to changes in volatility – the key valuation parameter in the BSM
model,” says the study just published in the journal Economic History
Review. “Traders of tin and copper options in the 1920s transacted
with Keynes via his broker at prices fairly close to their BSM theoretical
Such interwar commodity options were “modestly” overpriced – “as in modern markets” – and responded systematically to market changes as reflected in volatility even though traders had no knowledge of the BSM model – whether for puts, calls, or straddles at-the-money (selling both a put and call option at the same strike price and expiration date), the study found.
fascinating is that the study suggests – and we can only speculate – that for
the options traded by Keynes in copper and tin, market participants were
intuitively taking into account the parameters of the BSM model, including
volatility, without any knowledge of the model itself,” say the study’s
co-authors. “Perhaps this is because these traders were able to estimate
the relationships between the key elements of BSM and option prices, even
without the benefit of the model. But we simply do not know.”
The new study hand-collected
the actual prices paid to London Metal Exchange dealers by Cambridge-based
economist Keynes in the over-the-counter market for short-dated options on
metal futures from 1921 to 1931, based on Keynes’s detailed investment records –
and then analysed how “fair” those prices were as judged by the BSM
The conclusion was
that for Keynes’ trades, pricing errors (defined as the difference between
traded prices and BSM-theoretical prices) averaged 15 per cent, which is at a
similar level to those observed in modern options trading that actually uses
the BSM model. Previous historical studies, based on South African mining
stocks in the early 20th Century and US stock option pricing in the
1870s, found pricing errors at two to three times that level.
The study acknowledges
that not every trader is as smart as the well-connected Keynes, so less
sophisticated investors may not be the recipient of the same efficient pricing.
“We do not deny
this possibility but at the same time we believe that this does not invalidate
our test or weaken our results,” the study says. “What matters,
whether in the 1920s or today, is whether trades between informed participants
exhibit very little mispricing.”