There's no way of knowing how many of Minnesota's big-company CEOs know about Henry Singleton, but they could probably learn a few things from studying the career of a man once called "the modern American capitalist."

Singleton was a CEO with a reputation for shrewd investing back when no one outside of Omaha knew who Warren Buffett was. Singleton ran a company called Teledyne, and investors in his company in 1966 made 53 times their money over 25 years, vs. 6.7 times for the S&P 500.

Singleton thought it was his job to use Teledyne's checkbook for the best investment ideas. If Teledyne wasn't in a particular industry already, that was a detail, not an obstacle.

The role skillful capital allocation can play in corporate growth is top of mind this week because Buffett this year has once again made a good case that his conglomerate structure at Berkshire Hathaway is ideal for long-term capital growth.

That's not conventional wisdom, of course. There aren't many synergies to be gained by putting unrelated businesses under one roof, the game plan of Buffett and Singleton. Modern corporations instead try to stay focused.

That's hard to argue with. There are good reasons why conglomerates went out of style.

Yet telling your shareholders that your main job as CEO is to be really smart about spending the company's money has also, unfortunately, gone out of style.

Carefully allocating capital is as important to narrowly focused companies as it is to conglomerates built by the likes of Singleton. 3M Co., with its thousands of products, is probably closest to the conglomerate model on the list of Minnesota's biggest public companies. But even companies more or less in one industry, like Ecolab, have what I assume are hundreds of separate business units.

These companies are nothing like the high-flying conglomerates of the 1960s. Then a company called Litton Industries, a defense contractor, bought Stouffer's frozen foods. Gulf and Western owned zinc mines, the Miss Universe pageant and Paramount Pictures.

The leaders of the conglomerates took an interesting idea — coolly allocate capital between unrelated business opportunities — and drove it into the ground like a tent stake.

A brutal bear market in the 1970s ended the party for most of them. Investors and analysts distrusted them, never knowing where the earnings were coming from. And it was the portfolio manager's job to diversify, not a CEO's.

The CEOs of the 1980s got rid of "noncore" operations. This idea was reinforced by a popular article called "The Core Competence of the Corporation" that appeared in 1990 in the Harvard Business Review.

The authors showed how the corporate winners would be those that organized themselves around strengthening a few technologies or processes that made the whole company more competitive.

The interesting thing about this argument and the approach of executives like Singleton and Buffett is that they shared the same observation about the weakness in how big companies organize themselves.

The champions of "core competence" thinking saw problems with separate business units, where the business unit manager essentially just cared about delivering results that met the budget, not cooperating to strengthen important competencies.

Buffett has talked about the same thing, although in financial terms. Capitalists are supposed to be great at allocating the money to the most productive things, but Buffett thinks it rarely works that way in corporate America.

When a business unit's prospects have soured, the rational thing to do is look for somewhere else to invest. That idea won't sit well with the business unit manager, who will balk at losing part of the expense budget.

What's worse is if the business unit that is starting to slide has been powering the whole company's earnings. Investing in something else now looks like an admission of defeat. So that day is put off. And capital gets wasted.

In Buffett's latest shareholder letter, he wrote about how he had stubbornly tried to grow See's Candies, a boxed chocolate manufacturer and retailer Berkshire bought in the early 1970s. It only later occurred to him to use See's cash flow to buy businesses with better growth prospects.

One of the ways Buffett learned to do this kind of thing, of course, was by watching Singleton. He's on record calling Singleton a master at running a business that way.

Singleton really was a unique CEO. When Teledyne's stock was highly valued, Singleton used shares trading at 25 times earnings to buy companies selling for five times earnings. When his own company's stock looked cheap, he used cash flow and borrowed money to buy back shares.

In the 1970s, he had his insurance subsidiaries buy common stocks with the excess cash that came from insurance float, calling stocks the "safe" investment. Even Wall Street veterans shook their head at this kind of Alice in Wonderland thinking, in the midst of a crushing bear market for stocks. Singleton was proved right, of course, and as other conglomerates came apart his continued to prosper.

After he stepped down, Teledyne was beset by scandals, and then the dismantling began. Even by the end of his tenure in the late 1980s, though, his reputation for brilliance had faded.

Singleton wasn't much for courting the stock analysts and didn't put much detail in news releases. It's one reason he was charged with the corporate crime of not having a business plan. It didn't help that he told Business Week in 1982 that "my only plan is to keep coming to work every day."

Even with Singleton's record, it would be silly to suggest that a top CEO like Doug Baker of St. Paul-based Ecolab needs to start acting just like him. It seems like a business plan is a useful thing for a CEO to have.

On the other hand, I hope Baker keeps his calendar completely free of meetings at least a few days a year. That way he could come into work, like Singleton did, eager to find out what opportunities the market has brought him.