Sam Palmisano, CEO of IBM (NYSE:IBM), announced in October 2011 that its board of directors had approved a $7 billion increase in its stock repurchase program to $12.2 billion. Palmisano boasted: ” IBM’s higher value, higher-margin business strategy has enabled the return of over $109 billion since 2003 to our shareholders through share repurchases and dividends.”
While impressive, I question the effectiveness of the share-buyback strategy. Read on, and I’ll explain why.
Andrew Hallam, author of the Millionaire Teacher:The Nine Rules of Wealth You Should Have Learned in School, suggests that rational share prices follow profits. If IBM’s profits increase 100% over a decade, so too should its stock price. Although the spread between the two rises and falls over time, generally they run together. Since the end of 2003, IBM’s cumulative total return is 134%. Hallam believes that earnings over those next eight years should have increased by a similar amount.
According to Morningstar, however, earnings per share grew from $4.32 a share in 2003 to $13.06 in 2011, an increase of 202%. What gives? Share repurchases, that’s what. Fewer shares outstanding naturally increased EPS.
If you use net income (which isn’t affected by the buybacks) rather than EPS, the increase is actually 106%. Add another 11% in dividend yield for payouts over the eight-year period, and your total return ends up being 117%, some 17% shy of IBM’s actual total return of 134%.
Extending this one step further, if you accept 117% as the hypothetical total return using net income and dividends, you can make the argument that the total return without share repurchases was 117% compared to 134% with share repurchases, meaning the difference of 17% is attributable to share repurchase. That’s an extraordinary waste, when you consider it spent $90.1 billion on buybacks over eight years to generate an additional 17%.
Imagine what IBM could have done with the $90 billion. It paid out $18.4 billion in dividends over eight years, just 11.5% of the $109 billion Palmisano refers to above. What if the amounts were reversed, and IBM paid out $90.1 billion in dividends and $18.4 billion in share repurchases?
The result would have been a much higher share count, resulting in an EPS increase nowhere near the actual 202%. However, net income would still have increased 106%, and shareholders would have an extra $90.1 billion in dividends. Ultimately, I believe, the shareholders would have been better served with dividends instead of share repurchases. Here’s why.
Using its share count from the end of 2003, which was 1.76 billion shares, I’ve added 260 million shares that IBM issued between 2004 and 2011 for employee purchase and incentive plans. That gives us an estimated share count of 2.02 billion. For simplicity sake, we’ll use this as the outstanding number of shares for all eight years
If you take the $90.1 billion and divide by eight (the number of years) and then divide the answer ($11.26 billion) by the share count of 2.02 billion, you get an annual dividend of $5.58. Taking the high and low stock prices between 2004-2011, you get an average price of $132.20 and an average annual yield of 4.2%. Multiply yield 4.2% by eight (the number of years) and add that to 106%, which is the net income increase from the previous paragraph, and you get a hypothetical total return of 140%, slightly higher than the actual at 134%.
Furthermore, had IBM paid a dividend of this size, its stock likely would have attracted significant interest from institutional and individual investors alike, driving the stock price up even higher.
From 2004 to 2011, IBM paid an average price of $109.55 a share to repurchase its stock. As share repurchases go, it didn’t too badly. Having said that, shareholders still would have been better off with more dividends and fewer buybacks. The stats don’t lie. Share repurchases do far less to positively affect stock prices than dividends do.
It seems even successful companies will mess up capital allocation from time to time.
As of this writing, Will Ashworth did not own a position in any of the stocks named here.