Many of you may have read the recent Wall Street Journal article titled "Warren Buffett Comes Down The Buyback Chimney" that brought the news that Berkshire Hathaway (BRK.B) paid $1.2 billion to repurchase 9,200 Class A shares (BRK.A) at a price of $131,000 per pop. What is notable about this story is not only that it is remarkable that someone managed to amass a $1.2 billion fortune in Berkshire Hathaway over the course of several decades, but also the fact that it is an example of a share buyback done right: at Berkshire Hathaway, there is a policy that no share will be repurchased at a price above 120% of book value (when Berkshire announced the buyback program last year, the cutoff was 110% of book value).
That's how it should be. A company should determine some kind of objective criterion that will set the threshold for a stock buyback program to ensure that value is created by stock repurchases, as opposed to a blanket, "We're going to buy back $1 billion worth of stock over the next twelve months, whatever the price may be." The problem is that the evidence over the past decade indicates that the experience at Berkshire Hathaway is far from normative in corporate America.
Here's an excerpt from CNN Money that came out in February 2009 that discussed the difference in stock buyback programs between 2007 and 2009:
"Companies repurchased huge chunks of their own stock when prices were peaking. Buybacks for S&P 500 companies hit a high of $150 billion in the fourth quarter of 2007, according to Thomson Reuters. The market crash put an end to this: Companies in the index should buy back around just $10 billion in the current quarter. This is a shame."
This is why stock buybacks don't work. If you're doing it right, you're going to stop buying back stock when stock prices are high (with the company likely flush with cash because things are going well). Also, you should be accelerating your stock buybacks when the price of the stock has been beaten down, but as the numbers from CNN indicate, corporate America bought back about 15x as much stock during the fourth quarter highs in 2007 as it did during the market lows of 2009. This is why dividend investors cringe when they hear about money that could be funding dividend growth getting diverted to stock buyback programs.
For a more updated perspective, here is what the New York Times had to say at the end of 2011:
"After diving in the wake of the financial crisis, buybacks have made a remarkable comeback in recent years, with $445 billion authorized this year, the most since 2007, when repurchases peaked at $914 billion."
See a pattern here? When the stock market recovers, companies then have the cash to engage in buybacks. This is problematic because the stocks are trading at higher prices. Buybacks collapsed during the financial crisis because it takes real cajones to say, "Yes, the world is falling apart. Instead of surviving, we're going to try and buy back $5 billion worth of stock." That's a hard sell during times of extreme pessimism. Dividends, on the other hand, are often a sign of strength and serve as a reassuring signal-yeah, the economy is tough, but at least Coca-Cola (KO) is still growing its earnings and dividends, as the thinking might go.
Sure, there have been some companies, notably Exxon-Mobil (XOM), IBM (IBM), and Wal-Mart (WMT) that bought back stock before and throughout the financial crisis and managed to deliver a net benefit to shareholders. Exxon bought back 2.5 billion shares over the past eight years, earning approximately 4.9% on its buybacks over that time frame. Adjusting for executive compensation, IBM has reduced the share count by 500 million over the past eight years, earning a 15.3% annual return in the process. And Wal-Mart has bought back just shy of a billion shares since 2004 (not adjusting for dilutive executive compensation), earning a return of 8.9% annually on its share repurchases. So yes, some companies can and do create value with their stock buyback programs.
But comprehensively, this has been proven to be the exception rather than the rule. When you look at the 2006-2010 business cycle, the fact that companies spent 15x as much money on stock buybacks during the stock market peak in 2007 compared to the trough of Q1 2009 ought to indicate that maybe corporate America doesn't have a good track record of making money for American shareholders when they use stock buybacks as a device. We can talk all day long about how stock buybacks make sense in theory, but as long as most American companies continue to confuse the moment when stock buybacks should be accelerating with the moment when the buyback programs should be eliminated, we should stop getting excited anytime a company announces a new buyback program. At the very least, we should ask this follow-up question anytime a new buyback program is announced: What are the price terms for conducting the buyback? If there's nothing specific, investors should either brace themselves for some destruction of shareholder value, or look for investments elsewhere.