On a share market of most liquidity and least mispricing

The main idea: require all listed firms to act as their own shares’ clearinghouses, considerably narrowing the gap between market cap and book value. 

0. Posts on this blog are ranked in decreasing order of likeability to myself. This entry was originally posted on 18.06.2022, and the current version may have been updated several times from its original form.

 

1 All equity is overvalued

1.1 There’s a fundamental bias built into equity (and most other capital) markets: its far easier to cause a stock’s price to go up than down.

1.2 If you think stock X is underpriced, just go and buy it. What can you do if you think stock Y is overpriced?

1.3 You can sell, but this means that only those who hold the stock in the first place are entitled to vote on whether its overpriced or not, compared to how everyone is entitled to vote on whether its underpriced.

1.4 You can sell short. Which mechanism is based on leverage, hence very risky to a degree buying a stock that may lose value isn’t. The difference between losing money you have and losing money you don’t even have.

1.5 Moreover, shortselling is a complex technical endeavour, open to a select few professionals, again compared to anybody who can go in and buy a stock they dreamt about.

1.6 But finally, shortselling may be an inherently short-term action (see 1.61 though), and you cannot hold such a position for very long. On the other hand, you can HODL forever, giving ample time to the market to come to its senses in regards to your pick. When you sell short you really allow the market to stay irrational for longer than you can stay solvent.

1.61 As a counter to 1.6, commenter Jonathan Ray at Astral Codex Ten tells me that it is indeed feasible and rather quite cheap to hold a short position over a decade, especially one taken over a major firm. Whilst the other issues mentioned here continue to hold (I think) - and I could add others I suspect also hold such as counterparty risk or the chance of the position being closed on you by the underlying holder at the worst possible time - the feasibility of shorting over such long time horizons takes much of the edge from my theory. With the severity of the issue of overvaluation being less than originally thought, this post is downgraded a few notches. 

1.7 So, whatever its merits, shortselling is a limited fix that only takes care of the very worst and very obvious cases of over-pricing, and even that when its not being demonised or made illegal. Other alternatives such as options and whatnot all share the very same issues: leverage and complexity in execution.

1.8 Hence, the forces that prevent over-pricing are inherently inferior to those preventing underpricing. All equity is overpriced, as a rule.

1.81 The fundamental issue with shares is that there isn't such a thing as an anti-share (which doesn't prevent Scott Alexander from having a bit of fun though) which can be traded. Imagine a prediction market where all contracts are Yes contracts, and no No contracts can be traded or are enabled at all. You can buy, sell, short or take options of the Yes contracts, and will rely of these tools alone to gauge the probability of the issue being analyzed. Would the probabilities produced by such a market be as unbiased as those produced by a standard design where both Yes and No contracts can be traded?

1.82 A final point before moving on. The above applies to nearly every market, with a precious few exceptions like prediction markets. Is then everything overvalued? Can everything be overvalued at once?

1.83 Yes, but not really. There’s a huge difference between a good or service whose value is linked to its ultimate consumption, and money or financial markets. There’s no “true price” of the former, and a loaf of bread is truly worth what buyers will bid for it. Even if you prevent the resale of bread (or xBoxes) on pain of death, the prices of these goods won’t inflate at all. You’ll just end up with a much more variable price in time, as speculators are not allowed to smooth it. Eventually, all a buyer needs to do is estimate his own enjoyment of the good. 

1.84 Financial markets are, on the other hand, derivative markets in the broadest sense: we all play for money, not utility. Given one’s risk appetite and time preference (and these two you can disentangle nearly at will), there is indeed one true “fair price” for Tesla stock. The only challenge is that this price must be estimated in the face of an uncertain world. In this kind of market, various knowledgeable parties have various models of the true price, and limiting buying and selling to some parties will indeed skew the price. 

1.85 So, is the financial sector as a whole overvalued? Yeah maybe, but back to shares now.

1.9 I’ll confess this to be a weird way to start this post, given that what I’ll be proposing below does not solve this almost at all (perhaps to a very minor degree only). But it has to be said.

1.10 What also has to be pointed out is the curious practice of denouncing insider trading whilst allowing the greatest insider of them all - the firm itself - to trade at ease by way of issues and buybacks. Of course there's nothing wrong with insider trading, but that the gain in market efficiency allowed by having the issuing firm trade is such as to survive the onslaught against insider trading is relevant to the below.

 

2 Centralising market clearing

2.1 Here’s the idea: all listed firms would be required to publish a price for one share, at which they would buy and sell unlimited quantities from and to all comers. I'd furthermore disallow any bid-ask spread, but am less certain on this.

2.2 Once this is established, nearly all trading outside of each firm’s offering will cease, as no one has an incentive to buy at a higher price or sell at a lower price. The published price becomes the uniform price.

2.3 Nothing stopping the firm from updating the price as frequently as needed in light of the flows, but the price has to be publicly accessible and actionable, and a trade requested at a certain time must be fulfilled at the price published at that time.

2.31 Obviously, the great dynamism of shareholding under this model would make dividend impracticable, hence all distributions would be by way of buybacks (jack the price of the share for a bit). Shareholders would be entirely anonymous though, at least as far as their relationship with the firm is concerned. 

2.32 That very same dynamism of shareholding would make the standard AGM model of governance obsolete, requiring a rethink of sorts. At one extreme, one could put forward the idea of severing shareholding and ownership at all, with either Board members or CEOs nominating their own successors in a Yarvinesque and Moldbugian fashion respectively. Anyway.

2.4 Obviously the office adjusting the price such as to match inflows and outflows would be of key importance with every listed firm, as mistakes or sloppiness could lose billions in seconds. But these are, after all, publicly traded firms for whom the above'd be just another Tuesday.

2.5 Still, some safeguards must be in place, lest we create the weird prospect of a “run” on a firm where all holders decide to redeem at the same time. Some may even have trouble believing that setting prices which by changing by cents can cause billions of flows in milliseconds from algorithmic traders could even be possible if managed by humans.

2.6 To avoid all this, there would be some firebreaks, whereas market cap changes of (say) 5% up or down from the session start would automatically shut trading. At which point the issuer would decide if trading would resume before the regular session concludes or not. Regardless, trading would automatically restart with the next session. Since market caps are themselves affected by massive trades, the limits of this system would move with the firm and you’d never reach zero (hm, maybe some minimal cap standards are in order, otherwise the firm trading for 1000 bucks would stop trading every other minute).

2.7 All right, so no runs or takeover by algorithmic traders. Still, why make what is now largely automatic trading such as to require active and careful oversight from each firm?

 

3 The benefits

3.1 First and most obviously, this market is now nearly infinitely liquid, which is a huge difference compared to the vast majority of stock today outside of the main indexes.

3.2 When it comes to most countries outside the main ones, whole markets are very illiquid. Such a design may be the only way to have a functional equity market where market depth is otherwise insufficient to allow for one.

3.3 But liquidity is a secondary consideration only, the main idea is to prevent overpricing to a slightly greater degree than today.

3.4 If a stock is over-priced it will be sold back, without anyone having to shop around for a “greater fool” who thinks it’s actually underpriced. Not a huge help since those who make the call still must be holders in the first place, but their counterparties list has now exploded, so that’ll help a bit.

3.5 Helping a bit more would be the prospect of a firm going bankrupt because it does not have the cash to honour its promise of redemption, or has issued far too many liabilities against stock. A whole new type of bankruptcy.

3.6 But even more importantly, the combination of the great boost in liquidity and the little boost in fair pricing may produce far more anchored and honest prices that would be far less prone to cyclic behaviour. Of course, cycles and bubbles are monetary in the first instance, so this system would not eliminate them entirely. But the degree to which such a market could respond to monetary influxes would be checked a bit in magnitude and much in timespan.

3.7 One can go so far as to say that the way the market is run nowadays gives rise to entire periods during which prices seem completely made up, helping leftists of every stripe to sell their tired old ideas. Well, no more, now mispricing actually costs the firm something. I can be convinced that this is the only honest way to issue stock.

3.8 Finally, you know those firms with huge value to book ratios? Yeah. Yet another interesting result of such a system would be to make market caps much more tangible than they are now. Under such a system, a market cap is a potential liability (you may have to pay it out in cash), making firms whose book value is much smaller than their market cap sweat a bit. Perhaps high enough value-to-book ratios would themselves precipitate a run. Firms may need to hold more equity and be less leveraged, but its mostly the caps that would go down to more realistic levels.