There's a good piece in the New Yorker this week called "Board Stiff." The writer, James Surowiecki, makes the case that corporate boards still aren't doing a very good job minding the store for shareholders. Despite "reforms" like increasing the number of outside directors and increasing the ethnic diversity of corporate boards, he argues, the boards of publicly traded companies still aren't effective in anticipating problems or preventing business meltdowns. The main reason, he cites, is that board members still rely on their CEOs for information. There's no clear autonomy or ability to challenge the CEO's thinking.

One reason is that the CEOs of publicly traded companies still play the largest role in selecting directors, which results in a loyalty system that makes it difficult to rock the boat. Directors don't have enough power or time to really direct; instead, they typically see their most important job as selecting the CEO. It's not until there's a crisis of confidence in the CEO that the Board steps in, and by then it's too late.

I've worked extensively with corporate boards. I've also worked extensively with the boards of many other types of organizations: non-profits, public agencies, universities, and cooperatives. One thing stands out: the CEO typically doesn't serve on those boards. That confers some clear advantages:

• First, it's a lot easier to clarify the roles of the Board and the CEO when there's clear separation of powers.
• Second, it enables the Board to structure its work so that it truly understands the issues of the company and can set overall direction and policy.
• Third, it forces the Board to be held accountable. It can't fall back on the excuse that "we relied on the CEO."
  

That's a powerful case. But implementing a CEO-less board of directors runs up against a counter-veiling force: the ability of CEOs, under the current system, to control their boards and not be governed by them. That, fundamentally, is what stands in the way of fixing corporate boards.

A recent study measured successful CEOs for the Big 5 personality traits (openness, agreeableness, conscientiousness, extraversion, and stability) and found the closest correlation with conscientiousness. The authors of the study, Steven Kaplan, Mark Klebanov and Morten Sorensen (“Which C.E.O. Characteristics and Abilities Matter?”) relied on detailed personality assessments of 316 C.E.O.’s and measured their companies’ performances. So where do you think you are on the conscientiousness scale? Here are some sample questions:

• I am always prepared.
• I am exacting in my work.
• I follow a schedule.
• I get chores done right away.
• I like order.
• I pay attention to details.
• I leave my belongings around. (reversed)
• I make a mess of things. (reversed)
• I often forget to put things back in their proper place. (reversed)
• I shirk my duties. (reversed)

Regardless of whether you think you score high or low on this scale, you should not leap to the conclusion that these are the most important traits of successful CEOs. On the contrary, I can point to dozens of case studies in which passion, honorable behavior, and humility played a much greater role in defining successful leaders and successful companies. The real difference between the successful and not-so-successful CEO has nothing to do with personality. It has to do with traits that are learned, like persistence, efficiency, analytic thoroughness and the ability to work long hours. It would be easy to miss this point if you were inclined to put too much faith in nature, not nurture.

The waitress who came to our table didn't make eye contact. She took our order without smiling. As we got our food, I noticed she was treating other customers testily, too.

"I wonder why she waits on tables," my lunch companion said to me. "She clearly isn't enjoying it."

"I wonder about her manager," I said.

We started talking about techniques for converting lackluster employees into first-rate ones. He told me a story about a fast-food business he had managed when he was younger. The cashier in the drive-through window was a woman who never smiled. He sent her to customer service training, but he couldn't get her to change her attitude.

"What did you do?" I asked.

"I told every employee they had to collect data on how well they were doing their jobs. For each employee, I set up a measure that would improve their performance."

"What did you tell her?"

"In her case, I told her to count how many smiles she received from customers each day."

"That's brilliant!" I said. "Did it work?"

"I saw the improvement right away. Because she was measuring smiles, it made her more conscious of what I expected her to do. When she smiled, she found that people smiled back. In a week she was a completely different employee."

That's a great example, I said, of the Pygmalion effect. People perform better when they are expected to perform better.

My friend asked me if I had written about it.

It's a basic part of managing well, I said. Your attitude toward other people affects how they perform. By communicating to her the way you did, you raised her awareness and helped her perform better. Nice going.

That woman could charm the buns off a hot dog, he said with a smile.

I was talking about good governance with an elected official, a member of a city council. "What do you do to encourage good governance?" I asked him.

"I ask people what's going on. I ask employees at all different levels how they're doing. That's how I find out which departments are well managed, and which are not."

"Really?" I asked.  "Is that how you spend your time?"

"I think that's why the people elected me - to find out what's going on." He said the city charter permitted him to "get whatever information he wanted from whomever he wanted." He could call a parks employee, ask why the grass hadn't been watered, and expect the employee to tell him.

"Who would let the city manager know?" I asked.

"It's not my job to tell the city manager," said he said. "I assume the employee will tell him."

"But couldn't that lead to chaos?" I asked.

"I don't know."

"What about determining the overall direction of the city, and setting long-term policy, and setting performance goals for the city. Isn't that the city council's role?"

"We don't have performance measures," he replied.

"Well, that explains a lot." I smiled. "Listen, you have no formal feedback mechanisms. So you've had to go out and invent your own. I can understand that."

He nodded. "Do you know how I measure whether the streets are being cleaned? I put a newspaper in the gutter and then check a day later to see if it is still there."

Imagine what would happen, I said, if the city council did define performance measures. "Then you could focus on measuring performance, not as individuals inventing your own measures, but using agreed-upon measures. That would create alignment at the top - and lead to a clear understanding of everyone's role."

The city councilman grinned. "That would be a good trick." He said his goodbyes and left.

Yes, I thought to myself. It is a good trick.

What makes good people be bad partners? Over the past month I've witnessed the dissolution of a law firm's 15-year partnership. It began when one of the senior partners filed for divorce. The timing was unfortunate. It came just weeks before two high-powered associates were scheduled to buy shares. A buy-in signals the value of the stock. Fearing the repercussions, the senior partner (the one with the looming divorce) announced he wanted to put the deal on hold. "We have to wait," he told his colleagues. Secretly, he was holding out for more money.

Fast forward two months. Five partners split away, forming a new firm, taking several associates with them, including the two who were scheduled for the buy-in. The senior partner became one of four shareholders in the firm. Three months later, the firm filed for bankruptcy, citing an excess of debt and an inability to draw in new investors.

Could this have been prevented? Of course. With the appropriate governance mechanisms, the firm could have put in place systems to deal with conflicts such as these. It requires trust to build those types of systems - and a desire to make those decisions long before trouble occurs. Most important, everyone needs to assume responsibility. In this case, they hadn't. And that made all the difference.