Shares buybacks (repurchases) are often consider as tax-efficient alternative to dividends. Many academic studies have compared dividends and buybacks, here are my non-scientific notes.
To me it is a mystery how a company management can convince company owners (shareholders) that the best usage of cash is buyback.
Let's say a company has cash and can spend it to buy a business. Why do you think that the best decision is to buy the own business? The only correct answer in my mind (as a value investor) is " our own business is really good but strongly undervalued now, so it is good time to buy it." But as we know from scholar investigations most companies do buyback in overvalued times. Also many CEO/CFO" are biased and too bullish than consider their company (as were also documented in several academic studies).
For sake of simplicity let consider only 2 categories of companies - industry leaders and non-leaders.
Value metrics of leaders are usually higher than the same metrics for non-leaders, so argument that the leader is undervalued should not fly.
It should be even harder to sell buyback proposal for non-leader because obviously there are better companies on the market in the same and different industries.
To me it is a mystery how a company management can convince legal bodies like SEC that the management does NOT use insider information during buybacks.
Philip Carret in "The art of speculation" published in 1930 wrote
"Management has only so much cash as its disposal. If it pays dividends, I see it as a confession of failure. The company is saying 'We don't know where to invest the money so we'll give it to you,stockholders'".
I'd agree with this statement only with huge provision that management indeed has bullet-proof projects to invest MY (as company co-owner) money to grow the company and had demonstrated such ability. Shares buybacks rarely leads to company grow. Even debt repayment seems me better. Again many companies had terrible timing for shares repurchases I think because their management is biased.
I think at least a Board should propose buyback policy, shareholders should approve it (vote) and only then company can execute the buyback policy.
I think at most that the shares buyback should be again considered illegal activity similar to insider information trading.
Because of mistake in US tax system (double taxation) shares buyback can be presented as better cash distribution than dividends. This mistake in US taxes must be fixed. While IRS could declare a series of regular repurchases a constructive dividend and tax it as such, I do not know a case in which a share repurchase of a large, publicly traded company has been deemed a constructive dividend by the IRS. In fact, in a number of instances the company actually states in the repurchase announcement that they are repurchasing shares in
lieu of paying dividends (I wonder why the company CFO allows such declaration). Even in these instances the IRS has not declared the repurchase a constructive dividend.
IMO it means that IRS does not do proper work in this case and ordinary people pay more taxes (assuming that total government revenue is constant and if A pays less some has to pay more).
The following graphs shows how ineffective are companies in buybacks timing.
Michael J. Mauboussin recently published a good report "Share Repurchase from All Angles" there he used the same graph as I. He pointed (with Alfred Rappaport) that "Executives should follow the golden rule of share buybacks: A company should repurchase its shares only when its stock is trading below its expected value and when no better investment opportunities are available." I actually promoted the same idea but noted that probability of such event is close to zero (company MUST be the BEST investment opportunity).
1) Jeff Paul posted good buybacks analysis -see seekingalpha.com/article/724601-modeling...
2) Imagine that you are CFO or another top manager of IBM that suppose to propose BoD that to do with cash IBM has. In quarter 1 (Q1) you proposed buy PG shares because PG was strongly undervalued and it was rejected, in Q2 you proposed buy XOM shares because XOM was strongly undervalued and it was rejected, in Q2 you proposed buy MO shares because MO was strongly undervalued and it was rejected, .....
When CEO will fire you? 8-).
ONEOK Partners (NYSE:OKS) issued 11/13/2012 press release (finance.yahoo.com/news/oneok-partners-en...) where they wrote "ONEOK Partners, L.P. ... may, from time to time, issue common units... The partnership intends to use the net proceeds from sales under the program for general partnership purposes, which may include, among other things ... repurchases of securities."
Great! A company issues new shares with the goal to buyback shares !!! I do not understand something or this is total absurd!
I guess and hope that nobody just proofread the press release, meantime I sold my OKS common units...
I learn from seekingalpha.com/article/1078931-fiscal-... that Factset prepares good reports on dividends and buybacks:
I took Factset buyback data and compared them with S&P prices:
I think this picture tells as again that timing for buybacks is terrible for most of companies. IMO the quasi-fundamental reason is the following - a company has cash after period of good performance and good performance reflects in stock prices so buyback occurs at high P/E. On another side when P/E are low because of whole economy or industry or company issues the company is more reluctant to spend cash and stops buybacks.
Doing buyback company management essentially say "We believe that our stock is THE BEST purchase in whole stock market". How often it can be true?
Added 13 Jan 2013:
a) Warren Buffet noted: "Many CEOs never stop believing their stock is cheap" according to "How to Make Money in Dividend Stocks: Everything You Need To Get Started in Income Investing (The Art & Science of Investing)" by Edward Croft, Ben Hobson and Dave Brickell.
b) Again: How often buybacks are good? Are they better than dividends?
Let's assume that 1250 firms practice buybacks only (set B) while other 1250 firms practice dividends only (set D). Assume that at given business day stock of only 1 firm is the best investment (yes, the best means 1 even in my pure English). Assume that these "bests" are random and unique. We have about 250 trading days in a calendar year. It means that a single firm from set B and set D is the best investment only once in 10 years.
Smartest management of B-kind of firm makes buyback only in this day (again once in 10 years). Do you know such firm? I'm not.
Investors of D-kind of firms can use cash to buy the best stock each time they have enough dividends, let's say for simplicity each quarter. Do you know such investor who makes the best purchases regulary? I'm not. But dividend investors have 40 time higher probability to make best investment than buyback firm managers. I believe that this 40X is enough for amateur investors to oversmart any B-type firm management.
Added 15 Aug 2013
Bin Jiang and Tim Koller compared dividend payments and buybacks in the paper "Paying back your shareholders" (see fig. below).
They noted that "Some investors, too, prefer repurchases because they can then choose whether or not to participate. Institutional investors, for example, can maintain their investment in a company without the transaction costs of reinvesting dividends. Individual investors, by not participating in a share repurchase, can defer taxes on the dividends and turn them into capital gains even years in the future" Well it makes sense IMO.
When the authors argued that "Does it matter whether distributions take the form of dividends or share repurchases? Empirically, the answer is no. Whichever method is used, earnings multiples are essentially the same for companies when compared with others that have similar total payouts (Exhibit 3). Total returns to shareholders (NASDAQ:TRS) are also the same regardless of the mix of dividends and share repurchases (Exhibit 4)."
Let look on these exhibits more carefully. I indeed do not see any pattern in the exhibit 3, but the exhibit4 tells us a story (see boxes and lines I added to it below). For companies with relatively small payout (below 65%) total return to shareholders increased with reduction of buybacks (green box and line). The companies with dangerous payout (above 95%) did not return to shareholders more than 14% and subset of these companies with dividends below 66% of total payout delivered only less than 7% return (red boxes). For companies with relatively big payout (66%-95%) total return to shareholders increased with portion of buybacks in the mix (blue line).
Well each of us see that (s)he wants to see. To simplify Bin Jiang and Tim Koller argued (IMO correctly) that companies should return to shareholders cash they cannot use for NPV investment. The firms with high but not dangerous payout should on the first glance prefer buybacks (if we can continue the blue line to no dividend case), but I'd rather argue that their management don't think deeply enough to find promising R&D opportunities. If my eye is correct, firms with low payouts should ignore buybacks.
Extended quote from my comment to good article seekingalpha.com/article/1647072-dividen...-22417192
"The problems with buybacks IMO are:
a) CEO/CFO as well as many retail investors buy stock when they have money and usually it is exactly time when company has very good financials and fully or over-priced.
b) US firms rather stops buybacks when dividends when their finances are not so good and stock price is low then because (as in a) of course) market recognized it.
So it might be the buy high/sell low case + too frequent trading.
c) Buybacks were illegal in US before ~ 1980 - so not proper for good statistical analysis historical data exist
d) Buybacks are still illegal in many countries (IMO correctly, because CEO/CFO might act on insider information), so diversification is limited."
e) Many CEO are perma-bulls on stock of their company and agree to do buybacks at any price. I've never heard that some simple investment ideas (as for example dollar average) were applied to share repurchasing.
f) Any buyback splits shareholders into 2 groups (former and still co-owners) and cannot be good for both these groups. If the price of the company stock next day after buyback is higher than at buyback former co-owners loose, if the price is lower existing co-owners loose. Do you like if somebody makes you a foolish looser? If yes - you must vote for buybacks.