Tim Cook cares, a little. Photographer: Noah Berger/Bloomberg
Tim Cook cares, a little. Photographer: Noah Berger/Bloomberg

Apple's stock is trading at around $565 today. Apple thinks that number is too high, and so is going to cut it by six-sevenths, to some number around $80, by giving everyone six extra shares for each share they currently hold. This is not the sort of story where I explain to you that changing the nominal price of a stock doesn't change the economic value of the company; I hope we are all grown-ups here.

So, if the split doesn't have any economic impact, why bother? I've seen two explanations. One is that doing this will probably get Apple included in the Dow Jones Industrial Average, because they're not grown-ups at the Dow, and they weight their index by nominal stock prices, which is ridiculous.1 Now, if this is your explanation you need to dig a bit deeper, because who cares if Apple is included in the Dow? Getting added to the S&P 500 index is important, because it brings in a new, large, stable investor base of index funds, increasing value for shareholders. But not very much index money is invested in the Dow, because the Dow is ridiculous, and index investing is all about avoiding ridiculousness. So doing this is unlikely to attract many new investors, though you can't entirely rule out non-index but nonetheless Dow-focused investors.2

The more straightforward explanation that I've seen is that the split is, in Tim Cook's words, intended "to make Apple stock more accessible to a larger number of investors." Now, look. If you can afford to invest $80 in Apple stock, but you cannot afford to invest $565 in Apple stock, you should not be investing $80 in Apple stock. I can point you to some very nice index funds, or just, like, I mean, open a checking account.

But of course that's not what Cook means. He's probably referring to people who can legitimately afford to invest $8,000 in Apple stock, but not $56,500. Those people have a problem, because they want to buy stock in units of 100 shares, because you're supposed to buy round lots:3

"I think other companies will follow suit because it's difficult to buy round lots when a stock price gets too high," said Kim Forrest, senior equity research analyst, Fort Pitt Capital Group in Pittsburgh. "Higher prices per stock makes them difficult to buy for $500,000 portfolios."

Here, again, though, you need to go a bit deeper. Who cares about round lots? Historically, "trading commissions for odd lots are generally higher on a percentage basis than those for standard lots," and I suppose someone somewhere is still being charged extra to trade odd lots. But in the real world, it costs $9.99 to trade one or six or 100 or 10,000 shares of any stock; that is the price to trade stock, it has nothing to do with lots.4 Your cost to buy 100 shares of New Apple for $8,000 is the same as your cost to buy 14 shares of Old Apple for $8,000: $8,009.99.

But round lots do matter, for market structure reasons. Odd lots -- orders of fewer than 100 shares -- are treated differently under the Securities and Exchange Commission's national market rules. They are not part of the national best bid or offer, and are not "protected orders" under Regulation NMS.5 So pretend that some investor has a limit order to buy 100 Apple shares at $565, another has a limit offer to sell 100 shares at $565.25, and a third has an order to sell 20 Apple shares at $565.10:

You come along to your broker and ask to buy 20 shares at the market price. The displayed national best bid and offer are $565.00 bid, $565.25 offered: The $565.10 offer isn't part of the national best bid or offer, because it is for an odd lot. Your $9.99-a-trade broker sells your order flow to some internalizing market maker, the algorithmic trading firm that actually trades with you. The market maker sees the $565.00/$565.25 national best bid and offer, and improves on it to sell you the stock at $565.24 -- a better price than the "national best offer," but a worse price than the actual national best offer for the number of shares you want to buy. Then the market maker, if it feels like it, goes and lifts the $565.10 offer, leaving it with no position and making it 14 cents a share.6

That's just a $2.80 profit, but it's automatic and risk-free.7 Do a lot of that and you can make a lot of money trading without ever having a losing day.

But it's also frustrating for investors, as you may have heard. It looks like a classic case of socially useless high-frequency market making. Here, a "real" buyer and "real" seller who would have been happy to trade with each other -- at $565.10 or $565.24 or anywhere in between -- instead trade with a high-frequency trader who takes the 14 cents that the "real" traders would otherwise have kept. The more odd-lot trades there are, and the wider the spread between the best round-lot bid and the best round-lot offer, the more often this will happen, and the more money will bleed from "real" investors to high-frequency market makers providing a socially dubious service.

The higher the dollar price of a stock, the more odd-lot trades there will be, and the wider the spread will be.8 So high dollar prices pretty directly take money from investors and give it to high-frequency traders.

Now, like so many of the concerns that people have with high-frequency trading, this is a pure issue of market structure. The SEC could solve it in various ways, perhaps by getting rid of Regulation NMS entirely,9 or more simply by "protecting" odd lots. This recent BlackRock white paper calls for the latter solution:10

According to the SEC MIDAS platform, odd lot executions for equities account for 18% to 24% of transactions, which is a meaningful proportion of daily trading activity. Yet, although executions are now publicly reported, we still have a two-tiered market as it pertains to odd lots. The consolidated quote feeds still do not include odd lot bids and offers. In addition, odd lot quotes are not protected from being traded-through while round lots are protected per the Order Protection Rule in Reg NMS. ...

It seems incongruous that this segment of the market is not afforded the same protections that are extended to other orders. BlackRock supports removing all distinctions for odd lots in order to extend Reg NMS protection to every order in the market. This would improve transparency, reduce complexity, and eliminate the two-tiered market which exists today largely for historical reasons.

But that requires the SEC to do something. You can't really count on the SEC to do anything about the market structure it created; it's fond of that structure. But this particular problem can also be solved, or at least reduced, without SEC involvement, by just making stock prices lower and round lots more frequent. So why wouldn't Apple do that for its shareholders?

1 I mean, it's objectively ridiculous. That said, I've half-defended it as "a nostalgic Jesse Livermore throwback to when common stocks had a $100 par value and were supposed to trade roughly at par," and I sort of stand by that. You could build a whimsical historical market psychology in which (1) the right price of a stock is $100 and so (2) everyone wants his stocks to trade at $50-$80 so they feel cheap. I don't know. Also don't even get me started on the word "Industrial"; the Dow is ridiculous.

2 Can you? "I only invest in blue-chip companies like the ones in the Dow, not fly-by-night upstarts like these Apple characters." I don't know. If there is much demand like this, it's probably retail, and crotchety. I guess there's Dogs of the Dow or whatever, too, and now Apple does have a dividend.

Also, I suppose there are non-shareholder-value-based reasons for management to want to be included in the Dow -- look at us, we're successful, we run a Dow Jones Industrial Average company -- but, come on, they run Apple, surely they don't need validation from the Dow.

3 Also there is the straightforward money-illusion issue here -- $565 a share sounds expensive, but $80 sounds more reasonable, so you'd rather invest your $8,000 in 100 $80 shares than in 14 $565 shares. This is too dumb to discuss, though that is not to say that it is too dumb to exist. It surely exists.

4 Or maybe it's $8.95? Also, I quoted Investopedia for the wrong answer so it's only fair to point out that they actually have the right one:

While trading commissions for odd lots may still be higher than for standard lots on a percentage basis, the popularity of online trading platforms and the consequent plunge in brokerage commissions means that it is no longer as difficult or expensive for investors to dispose of odd lots as it used to be in the past.

5 Here's Reg NMS. Relevant language is mostly in the definitions: the "National best bid and national best offer" means what it sounds like; the "Best bid and best offer mean the highest priced bid and the lowest priced offer"; but:

Bid or offer means the bid price or the offer price communicated by a member of a national securities exchange or member of a national securities association to any broker or dealer, or to any customer, at which it is willing to buy or sell one or more round lots of an NMS security, as either principal or agent, but shall not include indications of interest.

Rule 611 is the Order Protection Rule of Reg NMS, and provides that trading centers must try to "prevent trade-throughs on that trading center of protected quotations in NMS stocks." A "trade-through," going back to the definitions, is a trade at a price lower than a protected bid or higher than a protected offer; those in turn are basically the (round-lot) national best bid or offer.

6 These numbers are hypothetical, but one trader sent me a snapshot of the Apple order book at 11:00:01 a.m. yesterday. Here's the top:

So if you went to your broker and asked to buy five shares of Apple at the market price, the broker's internalizer could sell to you at $566.24 -- the national best offer for 100 shares -- and then immediately go buy those shares for $566.20, the national best offer for nine shares.

Now, the Reg NMS rules are not identical to the broker's duty of "best execution," and brokers and internalizers may avoid doing this for best-execution reasons. "Flash Boys," and the BlackRock paper I quote further down, suggest that they at least don't always avoid doing it.

7 And comes out of your pocket, etc., though of course some of it goes back into your pocket in the fact that the payment for order flow is what makes your commission only $9.99. So don't get too worked up about it.

8 A good thing to read about nominal stock prices is this 2012 Felix Salmon post, which includes a fascinating chart suggesting that stock bid/offer spreads tend not to go below about 2 basis points. So a $50 stock would tend to have a spread of about a penny or so, but a $565 stock would tend to have a spread of about 12 cents or so. And in fact Apple's national-best-bid-or-offer spread runs around 16 cents, or 3 basis points. After a seven-for-one split, that spread should go to about 2 cents, creating less opportunity for high-frequency traders to pick off "real" investors, because there's just less room to jump ahead of them on the price. Improving by a penny on a 16-cent spread can make you 15 cents; improving by a penny seven times on a 2-cent spread only makes you 7 cents.

9 A non-obvious solution. The benefit of that would probably be to reduce fragmentation: If there's no national market system, then people might gravitate toward one or two exchanges. And individual exchanges protect odd-lot orders; the problem is that the national market system doesn't, and most retail orders now don't go to exchanges.

10 It also notes that institutional orders often end up becoming odd lots after being partially filled. So it's not just your dentist who has this problem; it's BlackRock too. And it's worse for high-priced stocks:

In certain stocks, the frequency of odd lot transactions is very pronounced; for example, nearly 60% of Google trades are odd lots and this represents over 25% of share volume in Google.

To contact the writer of this article: Matt Levine at mlevine51@bloomberg.net.

To contact the editor responsible for this article: Tobin Harshaw at tharshaw@bloomberg.net.