Value investors understand better than many that differences between ‘should-be’ and ‘are’ stock prices can be large and persist for very long periods, Keynes noting famously in the 1930s ‘The market can stay irrational longer than you can stay solvent”. The solve-for-y of these discrepancies is flows.
With this in mind researchers Xavier Gabaix and Ralph S. J. Koijen from the universities of Harvard and Chicago respectively set out to see if they could quantify what they refer to as the ‘dark matter’ (flows) in asset prices.
What they found astonished them and was shown, by a poll of peers who came nowhere close to guessing the impact, to be beyond the present realm of theorists (finance theory says flow-effect should be zero, because for every buyer there’s a seller).
As the research pair put it “Our findings are not only at odds with traditional theoretical models, but also with the prevailing common wisdom in our profession at the time of writing..” (their bold).
So what are the observed, real world, effects of flows? Let me try and summarize the main points:
- A dollar of new investment in a stock market results in a five-dollar rise in stock prices. This is because there is not a seller, in fact, for every buyer.
- Government stock purchases and company buy-backs therefore can and will have a meaningful impact on prices [See P.36 where Hong Kong’s 1998 intervention gets a shout-out. That resulted in a 4:1 stock price recovery, very close to the theoretical 5:1 highlighted in the paper].
- Stock price elasticity comes out around 0.2 which is to say if stock prices rise 5% the demand will fall by only 1%. Rational behavior models predict much higher elasticity.
- The price impact of flows is, theoretically at least, perpetually long lasting i.e. there is no mechanism for correction.
- The researchers are calling their work ‘The Inelastic Market Hypothesis’ and stress, in so doing, the work is merely a hypothesis at this stage.
- 1/3 of stock market fluctuation can be described by flows. By quantifying the impact of this ‘dark matter’ on valuation it’ll make it easier for future researchers to drill down into the causes of flow-variance.
- Hedge funds, who in the past were assumed to provide a flow balancing mechanism, don’t act counter cyclically. Moreover, representing only around 5% of markets, they’re just too small to offset the bigger flow picture.
- The biggest flow-effect producers are households and mutual funds (including ETFs) who are responsible for 54% and 30% degrees of influence respectively.
I wrote cheekily in an introduction to last week’s paper new ideas can often be summed up by old wisdom. In this case a snarky observation could be the paper reveals nothing more than stock prices are often driven by ‘more buyers than sellers’ (and vice versa); but that would be a fatuous summary here.
What’s new is not that flows move prices. The revelation is the magnitude of the effect and the mechanism behind it which has its roots in a supply-inelasticity conventional theory holds just isn’t there.
This paper shows not only is supply-inelisticity an important explanatory variable in asset price description (and manifestly present), but also, it’s effect is huge.
You can access the paper in full via the following link Why Stock Prices Move.
Happy Sunday.