Warren Buffett is America's most admired businessman ever. He's built a conglomerate worth hundreds of billions of dollars from almost nothing, by dint of incredibly astute investments spanning decades.
The modesty of his lifestyle shames the usual CEO excesses, while the decision to give away most of his fortune demonstrates a generosity befitting his legend.
Buffett has also taken a courageous and controversial stand on the issue of taxation, advocating policies that would cost him money in the name of what he sees as greater fairness. Because I happen to agree with his arguments, they add to my esteem of the man. (And I always wonder if Buffett's business critics have a political axe to grind. That's definitely not the case here.)
In light of the above, criticizing Buffett's business decisions feels a bit like arguing that a Mozart symphony could use a little something more right there. None of us could possibly have enough credentials.
Still, despite the cult that's grown around the "Oracle of Omaha," he remains a fallible man, as demonstrated by occasional missteps:
Buffett has a history of promoting favored stocks and companies via privileged investments that his followers can't possibly duplicate, as evidenced by his purchases of preferred stakes in Goldman Sachs, General Electric and recently Bank of America.
He railed against derivatives, then sold derivatives that promptly lost him money.
He's had accounting disagreements with the Securities and Exchange Commission.
He also showed poor judgment in failing to take timely and decisive action against a top subordinate who front-ran Berkshire's buyout of Lubrizol (LZ).
It's too early to know whether this week's announcement of the first share buyback in Berkshire's history will prove another Buffett home run or one of his rare strikeouts. But it's a striking departure from his long-held policy of reinvesting profits and leaving it to markets to reward shareholders.
Despite the subsequent spike in Berkshire's shares, the buyback doesn't make them a better buy. If anything, it warrants extra caution.
As Buffett acknowledges, it will only prove advantageous if Berkshire's shares rise over time, and many companies have proven to have mistimed their purchases. Think of the billions that Hewlett-Packard and Cisco Systems have poured down this particular drain in recent years. Historical evidence suggests that dividends have been a better way to reward shareholders.
Buying more of Berkshire's ample cash flow seems a far better bet than keeping its $40-billion or so of spare cash in minimally yielding Treasuries. On the other hand, what does that say about Buffett's estimation of his other portfolio investments and potential buyout opportunities?
There is an extra hitch with buybacks when it comes to conglomerates. These are assembled on the premise that the market has underestimated the value of a particular company being added to the fold.
So the empire builder must possess a self-confidence bordering on arrogance in disregarding the consensus valuation. Sometimes the process introduces us to a Buffett, and other times to a Dennis Kozlowski.
When a conglomerate repurchases its own shares, it's using its resources to manipulate the market gauge of the success of all its prior purchases. Buffett may be right that Berkshire is a bargain at a 25 per cent discount since February and a 34 per cent haircut from the record high set in late 2007.
Or maybe the market was right in evaluating the decline of Berkshire's insurance earnings, its increased cyclical exposure, and the risk posed by the rising market volatility and low interest rates. The point is that conglomerates are inherently hard to value, and the buybacks make that task even harder by goosing the share price in the short run.
But Berkshire is no ordinary conglomerate. It's one of the most decentralized ones ever, relying on Buffett's frequently excellent judgment in identifying the best managers and fostering a corporate culture that lets them do their thing with a minimum of interference.
So far, that's worked out well. But as auditing expert Francine McKenna points out, Berkshire's auditing practices are far from exemplary, while its decentralization and reliance on managers' adherence to a "culture" (propagated, it's worth mentioning, by the 81-year-old founder) are red flags based on historic precedents.
What will happen when Buffett fades from the scene? Will Berkshire's purchases of its own shares still look good then?
I don't know, and neither does Buffett. But this is one instance where no one should be especially tempted to front-run the great man.
Igor Greenwald is senior editor of Moneyshow.com