On January 9, 2018, Kodak ($KODK) announced that it was launching a “major blockchain initiative and cryptocurrency.” Their stock jumped from $3.10 per share on January 8th, to a high of $13.28 (closing at $10.70) January 10th, inflating the market cap from $132M to $455M.
On that day (1/10/18), over 100 million shares of Kodak stock were traded. Average daily volume for $KODK is around 250 thousand, so this would be 400 times the usual trade volume. Given that the company has around 40 million shares outstanding, it would seem that entire company value was traded twice over, and then some, that day.
Was every share really bought and sold twice that day? Or, was it just two high frequency traders buying and selling the single share between themselves with fractional cost differences? Did someone really invest an additional $320M in Kodak? Or, was the net additional investment only $7.60 from driving up the price of a single share?
Share price, trade volume, and market cap do not answer these questions. The same set of statistics could support either explanation, and everything in between.
Approaching this as a data analysis problem, I wondered what metric would help one distinguish high frequency noise from real changes in value?
Cumulative Cost Basis
Market capitalization is a hypothetical — how much would someone need to spend to buy every outstanding share at today’s market price.
But I would also be interested in reality — how much did everyone spend to buy the shares they own today.
We could get this value by adding up the cost basis of every outstanding share. Let’s call this the Cumulative Cost Basis (CCB). The CCB does not tell you how much the corporation may be worth. But it does tell you how much money people have been willing to collectively invest into a given stock — how much skin they have in the game.
Looking at changes in CCB over time, you could tell whether people are putting money into or taking money out of the stock, and at what rate.
High frequency trading back and forth of a few shares at fractional price difference would have negligible effect on the CCB, even if the share price changes drastically. On the other hand, large scale transfer of ownership would be reflected in the CCB, commensurate with the gains or losses of the trading parties.
PS The problem is, I am not sure if there is any single entity that knows the CCB of publicly traded companies.
PPS The level of my financial knowledge is moderate, at best. It is quite possible that this metric, or a better one, already exists and I’m simply unaware of it.