21-for-20 stock split ... why?

I happened to be looking at ADM (Archer Daniels Midland), and noticed its strange split history:

3-Jun-86 [3:2]
5-Dec-89 [3:2]
17-Aug-93 [21:20]
6-Dec-94 [3:2]
15-Aug-96 [21:20]
14-Aug-97 [21:20]
20-Aug-98 [21:20]
19-Aug-99 [105:100]
24-Aug-00 [105:100]
30-Aug-01 [105:100]

A few 3 for 2 splits, OK. But why on Earth do you split 21 for 20 (or 105 for 100)? It seems to me all that accomplishes is to cause a lot of people to have odd numbers of shares, and maybe to have to make a lot of petty payments for cash in lieu of fractional shares. 10 splits, and they’ve increased the number of shares by a factor of less than 5.

Just a WAG…

This appears to be a way of giving the shareholders a 5% annual dividend tax free. If you notice, the amount has been pretty steady at 5% and always occurs in mid-late August.

The other splits (non 5% ones) are the “real” splits of the stock.

What ** igloorex** said. It is a way of giving a tax-free dividend, while encouraging people to hold on to the stock; it takes a while to accumulate enough shares to be worth selling.

I’m used to seeing it with smaller companies anxious to increase the number of shares available (to lower volatility). It is a little surprising to see it happens with such a large company, but I’m by no means an expert in the field.

Slight hijack . . . How is this a tax free dividend? I am under the impression that the value of the stock that you own will still remain the same.

Of course I am probably missing something . . .

The value does indeed remain the same, but if you get 5% more stock, any increase in price gives you 5% more money.

At least it’s not a reverse split.

My wife owns some stock in Annie’s Homegrown, whore stock has been tanking beautifully recently. To the point where they are talking about a 1 for 2000 reverse split.

This doesn’t make sense to me.

Say you have 20 shares @ $100 each.

At the time of the split, the share price is diluted to give the same total value:

20 shares @ $100 each = $2,000
21 shares @ $95.24 each = $2,000

a 5% increase in either the pre-split or post split stock still gives the same total value:

20 shares @ $105 each = $2,100
21 shares @ $100 each = $2,100

I understand what you are saying in terms of math but the stock market is based on human irrationality and emotion at least a much as a hard analysis of what a stock is worth.

Some stocks tend to hover around a set price (say $25 a share) based on these factors. If a company splits a stock in a small enough way to cause the value to drop to $24 dollars immediately, you may still find that shareholders bid it right back up to $25 dollars a share in a very short time through shear habit.

I have no idea if this actually works consistently but apparently this company believes it works for them so they keep doing it.

Shagnasty ,

I understand what you are saying, but then why wouldn’t all companies effect a stock split when their price reached a certain (relatively low) point?

IANASB (I am not a stock broker or a son of a bitch), but I would imagaine that there would be a premium to a high value/low float security.


The “tax free dividend” statement does seem to be the idea. I went and looked in on the ADM board on the Motley Fool, and, sure enough, found a thread started by somebody carping about the ridiculous fractional splits.

Some respondents brought up the same “tax free dividend” statements. Of course, as has been pointed out, both here and there, the split does nothing theoretically to increase the value of your stock - when it splits 21-for-20, the share price drops to 20/21 of the previous day’s level and you are holding slightly more shares of slightly diluted stock.

However, a shareholder on the fool boards mentioned that every year when they do this, the price always comes right back up in a few days. Apparently, the mechanism is that there are enough naive investors out there watching nothing but the share price, who don’t notice the split, and just say something like “Hey, ADM just dropped a buck. I think I’ll buy some.”

It probably only works with large, stable, well-known companies that can foster that sort of thinking.

Now I have a different question. If this kind of stunt to slightly inflate your market multiple on a regular basis actually works, how come every “widows and orphans” blue chip on Wall Street isn’t pulling it?

Splits are supposed to be simply an accounting measure to keep share prices in range, and, in the case of normal splits, to increase liquidity by making more shares available for trade (reverse splits are still to keep the share price in range - normally, to keep it high enough to avoid delisting).

But the psychology that goes along with them can be wierd.

I would think just the opposite (about the stock price not your son of a bitchedness). A low float security would be harder to sell when the time came because the stock is not as liquid as some.

And, of course, while I was posting that, some similar observations floated in.

Don’t look at the percent; it’s misleading. A 5% increase is always a 5% increase. But look at the absolute number:

Say the stock price rises $1

20 shares @ $101 each = $2020
21 shares at $96.24 each = $2021.04

Say the price rises $5

20 shares @ $105 each = $2100
21 shares at $100.24 each = $2105.04

So each rise in price increases the value of the stock more quickly.

OK, I agree. I was thinking about securities like Berkshire-Hathaway (spelling?). However, the market forces at work on a unique security such as B-H probably do not belong in this discussion.

RealityChuck , wouldn’t there have to be significantly more demand for a stock with a greater amount of outstanding shares to increase in price? What I mean is that shouldn’t we be dealing in percentage increases instead of dollar amounts?

I am probably the one being naive, but I can’t reconcile the idea that unknowledgable investors and day traders would pounce on a stock because a price went down solely because of an X for X+1 split and have such a dramatic affect upon price.

I think the idea is to pay avoid paying a taxable dividend. If you have a $20 per share stock that it going to pay out a $1 dollar per share cash dividend, the day after the dividend is legally “paid,” (which is the day on which whoever owns the stock becomes entitled to receive the dividend aka the “ex-dividend date” – actually, this isn’t quite correct, the ex-dividned date is actually the last trading date, but I digress) the stock should open trading at $19 per share because the company now has assets of $1.00 per share less than it did before.

By not paying a cash dividend and splitting the stock, the company avoids this price drop and, hence, confers a benefit on the stockholders, at least in theory. Of course, the stock should also fall almost 5% when the float increases 5%, but at least it’s not taxable!

NB, some companies actually declare stock dividends. This is the worst of both worlds. The ADM stock split scheme makes a lot more sense than a true stock dividend and creates the same effect, at least when there’s only one class of stock.

IANA Accountant

I shouldn’t have said “tax free”…it’s really more “tax deferred.” If the company were to pay out a regular dividend, it would be taxed as income in the year it was distributed. Done this way, the tax wouldn’t come due until the holder actually sold the stock, and then quite possibly at a lower long-term capital gains rate.