The other day I learned about the nominal share price puzzle. Like, since the Great Depression the price of pretty much everything has changed dramatically. Except stocks! Stocks cost exactly the same. The average share of stock cost $35 then and it costs $35 now.
And like, the stock market has gone up so much since then! If you bought a share of GE for $38 in 1935 it would be worth like $10,000 now. Except it wouldn’t, because GE has split their stock a ton of times so you would actually just own a whole lot of shares that are worth $30 each.
Why do they do this? It costs GE money in administrative costs. It costs shareholders money in trading costs. And it doesn’t have to be this way: Japan and the UK both have totally non-constant nominal share prices.
The authors of this paper suggest one reason could be to market to individual investors. But if that were the case why wouldn’t share prices at least keep up with inflation? And this hypothesis would also predict that as stocks have become mostly held by institutions rather than individuals, the effect would diminish, but it hasn’t.
Another fun theory is that when stock prices are low relative tick sizes are high, so companies keep their prices low to compensate market makers for providing liquidity in their stock. But that would predict that stock prices would change when tick size changed in 1997, and they didn’t. And do executives at GE really lie awake at night worrying that if their share price goes above $100 no one will provide liquidity in their stock anymore?
The authors end up concluding that there’s no good economic explanation and everyone does it because, uh, it’s what everyone does. It was kind of unsatisfying.
Update to this post: there is in fact a reasonable economic explanation!
Matt Levine:
If you bid $10 per share for 100 shares of a stock on a national stock exchange, then your bid is “protected”: Under the rules of the national market system, no one is supposed to sell shares to someone else at a lower price without selling to you at $10 first. But “odd lots” of fewer than 100 shares are not protected: If you bid $10 per share for 99 shares, and the highest bid for 100 shares is $9.99, then people can sell 100 shares – or 99 shares – at $9.99 and ignore your bid. (It seems like it would be irrational for them to do that, but in practice this means that a wholesaler can buy from a retail customer off the exchange at $9.99 instead of at your best-in-the-market $10 price.) Lower stock prices, the theory went, encourage more round lots, which means that more trades will take place at the “real” best bid or offer, and fewer will take place at worse prices that bypass odd-lot best bids or offers.
And this recent paper (link) shows that there are plenty of cases where people trade through odd lots. It doesn’t really explain the whole since-the-Great-Depression thing, since Reg NMS has only been around since 2005, but it does give a good reason you might consider splitting your stock today.