Tuesday, May 27, 2014

Parlez-vous "Trolley"?

Or maybe I should say, Sprechen Sie Trolley?

A fascinating article in last week's Economist suggests that we think about moral issues differently in our native language than we do in another language in which we are competent but not fluent.

For those of you unfamiliar with the philosophical game of "Trolleyology", here's the simple version: A runaway trolley is speeding down the track. From a bridge above, you can see that five people around the next bend, unsuspecting, will be struck and likely killed. You can run and pull the switch that will turn the trolley onto another track, where it will hit "only" one person. Do you pull the switch?

What if, instead of a switch, there were only one, very fat, man, standing next to you on the bridge. If you push him over, his mass would stop the trolley and save the five. Would you push the fat man?

Most people say, "Yes" to the first example, and "No" to the second. But why? In both cases, one person dies, involuntarily, so that five others may live (And there are endless variations to the problem).

The fascinating new study reported in the Economist is that "when people are asked the fat-man question in a foreign language, they are more likely to kill him for the others' sake." The study was conducted by Albert Costa of the Universitat Pompeu Fabra in Spain and his colleagues, and was published last month in the journal PLOS ONE.
When asked in their native language, only 20% of subjects said they would push the fat man. When asked in the foreign language, the proportion jumped to 33%.
 As the Economist reporter noted, "Morally speaking, this is a troubling result."

Why would the language used affect the decision made? Was the difference cultural (the problem was posed in English, Spanish, Korean, and French; half of the group were asked to consider the "fat man problem" in their native language, and half in the second language)? No, because the results were the same no matter what combination of languages was used. Note that those who considered the problem in a foreign language were competent in it, not fluent. And that's where the crucial difference seems to lie.
Several psychologists... think that the mind uses two separate cognitive systems -- one for quick, intuitive decisions and another that makes slower, more reasoned choices. These can conflict, which is what the trolley dilemma is designed to provoke: normal people have a moral aversion to killing (the intuitive system), but can nonetheless recognise that one death is, mathematically speaking, better than five (the reasoning system).

Since the subjects who considered the problem in a foreign language were not fully fluent, they had to think harder, more slowly, than did the native-speaker subjects. And were able therefore to achieve "psychological and emotional distance" -- which made it easier to sacrifice one for the good of the five.

The Economist is hopeful that, because more and more firms are making English their de facto language "even if it is not the native tongue of most of the workers", they may start making better, more rational decisions.

"More rational", of course, as long as you're not the fat man being tossed over the bridge.

Thursday, May 15, 2014

How Much Is Too Much?

Or is there even such a thing, when you're talking about CEO pay?

There are certainly some people who will argue that there's no such thing as too much; that the big guys earn the big bucks because they're so smart and so capable. But maybe attitudes are changing, just a little bit.

This morning's New York Times carried a DealBook article by David Gelles on the remarkable pay package proposed for Chipotle Mexican Grill co-chief executives Steve Ells and Montgomery Moran. Chipotle is arguably one of the hottest chains out there today, and the company rewarded its chiefs last year with cash and stock in excess of $24 million. Each.

As Gelles noted,
Each man individually made more than the chief executives of larger companies like Ford, Boeing and AT&T. Together, they made more than all but the highest-paid chief executive among the country’s biggest 100 companies, Lawrence J. Ellison of Oracle.
I'm not going to say that running Chipotle is easy, but it strikes me as a lot less complex than running Boeing.

And last year was not an aberration: "Since 2011, Mr. Ells and Mr. Moran have each made more than $100 million on top of their salaries through a complex mix of stock awards."

Most troublesome is the huge gap between executive pay and front-line pay:
...[Although] general managers at Chipotle can earn upward of $100,000 a year, the average starting salary at one of the company’s 1,600 restaurants is about $21,000 annually. Earning that wage, a Chipotle employee would have to work for more than a thousand years to equal one year of the co-C.E.O.s’ pay.

Ells and Moran hit on a great concept and have rolled it out brilliantly. They deserve to be well-compensated. But the skills to be great entrepreneurs aren't necessarily the skills needed to be great managers. And if what they are now is managers, shouldn't they be paid as managers?

One investor, who was planning to oppose the pay proposal, said, "It's a reckless pay structure that does nothing to appropriately incentivize management to create long-term value.... Their pay is out of whack however you measure it."

At last year's shareholders meeting, "27 percent of shareholders... voted against the company's compensation package." Gelles surmised that the number might be higher this year.

And he was right. In a follow-up story posted online later today, Gelles reported that
More than 75 percent of investors voted against Chipotle’s say-on-pay measure, which asked investors to ratify a compensation plan that would continue such payments to Steve Ells, Chipotle’s founder, and his co-chief, Montgomery Moran, over the next few years. That was the highest vote against any say-on-pay measure among the country’s largest 3,000 companies this year.

The motion is of course non-binding, but "Chipotle said it was taking investor sentiment into consideration." We'll see....

Tuesday, May 6, 2014

Trust: How Far Should You Take It?

Even casual readers of this blog should know my twin mantras of Trust and Transparency. But a "DealBook" article in today's New York Times is getting me to think more about what I mean by Trust.

Is it, to follow Ronald Reagan, "Trust but Verify"? Pretty much.

Andrew Ross Sorkin, in the above-mentioned article (full story, here), quotes Berkshire Hathaway vice-chairman Charlie Munger as saying,
By the standards of the rest of the world, we overtrust. So far it has worked very well for us. Some would see it as weakness.

He was arguing, Sorkin wrote, that "Instead of filling your ranks with lawyers and compliance people [you should] hire people that you actually trust and let them do their job." This explains why Berkshire Hathaway, one of the largest companies in the United States, has neither a general counsel nor a human resources department (many of its subsidiary companies, however, have their own general counsels and HR departments).

Munger, and his boss, famed investor Warren Buffett, allow that this doesn't always work. Three years ago, Buffett's then heir-apparent David Sokol resigned over questionable stock trades in a company that Sokol had invested in and then recommended that Berkshire Hathaway purchase (for more details, see my 2011 blog post on the story, here; Sokol denied illegal insider trading and Buffett apparently agreed; the SEC investigated, but did not file charges). As Sorkin reported from last weekend's Berkshire Hathaway meeting,
"We will have a problem of some sort at some time," Mr. Buffett said to his faithful audience. He added, "300,000 people are not all going to behave properly all the time."
That's a risky position to take, I think. It's not just a question of money made or lost; it's a question of reputation, which is a lot easier to wreck than to repair.

How scalable is the Berkshire Hathaway model? Sorkin asked two Stanford University researchers: 
A trust-based system can be more efficient than a compliance-based system, but only if self-interested behavior among employees and executives is low. The risk is that the board makes an incorrect assessment of an executive’s ability and integrity and selects the wrong C.E.O.

So as long as your hiring techniques and intuition are perfect, you should be fine. (How many of us are willing to say that of ourselves? Not me, for sure.)

Worse yet, if you are confident in your methods, will it make you too confident?

I was surprised when Buffett was quoted as saying last weekend that he wasn't bothered by the $4 billion error that Bank of America reported late last month (Times article on the subject, by Peter Eavis and Michael Corkery, here). Apparently, Buffett trusts the bank's management.

I was disappointed when Buffett, who has spoken out publicly against excessive corporate compensation, "punted" (to quote Joe Nocera; full Times column, here) an opportunity to press Coca-Cola on what he himself thought was an excessive equity compensation plan. Buffett's explanation:
I love Coke. I love the management. I love the directors. So I didn’t want to vote no. I didn’t want to express any disapproval of management. But we did disapprove of the plan.

Really? That's the best you could come up with. Oh, but see: "I love the management." In other words, he trusts the management.

I was surprised and disappointed in January when Buffett said, of JPMorgan Chase CEO Jamie Dimon's huge jump in pay, "Over all, I think the shareholders of JPMorgan and the American people should be happy that Jamie Dimon has been running the bank over this period." This despite multi-billion dollar fines and more (my blog post on the issue, here).

Maybe there's such a thing as Too Much Trust?