First Post访谈

Luck vs Pluck: The man who could have been Bill Gates Dec 20, 2014

The first line in your book goes "my career was launched by a trash can". Can you take our readers through that story?

When I was a senior in college, I didn't really know what I wanted to do with my career, but I knew I needed a job. Drexel Burnham Lambert, an investment bank that was very successful at the time, came on campus to interview and I did well enough to be invited to New York City for a final round. So I put on my best suit and made the trip.

The day of the interviews, we candidates were told that we would have six long interviews and just 10 minutes with the executive who ran the division. My interviews went fine, and then it was my turn to meet with the executive. Upon walking into his office, I noticed that he had a trash can that carried the emblem of the Washington Redskins, a professional American football team. Being a sports fan and having spent my last four years in Washington, DC, I complimented him on the trash can. That comment hit in an emotional spot, and he launched into a discussion of his time in DC, the virtues of sports, and the link between athletics and business. I sat and nodded, as 10 minutes stretched to 15.

You got the job?

I got the job, and accepted it. Indeed, my time at Drexel Burnham was extremely formative.

After about six months into the programme, one of the leaders pulled me aside. "You're doing fine in the programme," he started, "but I have to tell you something. The six interviewers voted against hiring you. But the top guy came down and insisted that we bring you in. I don't know what you said, but it sure worked." So I like to say that my career was launched by a trash can, and that was pure luck.

What was the broader point that you were trying to make through that example?

The broader point is that luck permeates many aspects of our lives and we're frequently unaware of its role. So this book is about skill and luck, and includes the definition of each term, tools and methods to quantify the role of each, and what to do about it.

"Most of the successes and failures we see are a combination of skill and luck that can prove maddeningly difficult to tease apart," you write. Can you explain that in detail?

I open a chapter with the story of an entrepreneur who was born near Seattle who was a brilliant programmer and wrote code that effectively launched the personal computer revolution. He started a company that by 1980 had a dominant market share in the software that ran on the Intel chip. But the company's fate was sealed in 1981 when IBM came calling and sealed a deal.

Now if you know a little about Bill Gates, you can see how that series of facts fits him pretty well. But then I share the end of the story: this tech pioneer walked into a bar in California in 1994 and hit his head bluntly as a result of a fight or a fall-the details were never clear. He died three days later. His name was Gary Kildall, and he has a floppy disk etched on his tombstone.

Chances are you've never heard of Gary Kildall but you have heard of Bill Gates.

That's very interesting...

When IBM executives first approached Microsoft about supplying an operating system for company's new PC, Gates actually referred them to Digital Research (Kildall's company). There are conflicting accounts of what happened at the meeting, but it's fairly clear that Kildall didn't see the significance of the IBM deal in the way that Gates did.

And what happened then?

IBM struck a deal with Gates for a lookalike of Kildall's product, CP/M-86, that Gates had acquired. Once it was tweaked for the IBM PC, Microsoft renamed it PC-DOS and shipped it. After some wrangling by Kildall, IBM did agree to ship CP/M-86 as an alternative operating system. IBM also set the prices for products. No operating system was

included with the IBM PC, and everyone who bought a PC had to purchase an operating system. PC-DOS cost $40. CP/M-86 cost $240. Guess which won. But IBM wasn't the direct source of Microsoft's fortune. Gates did cut a deal with IBM. But he also kept the right to licence PC-DOS to other companies. When the market for IBM PC clones took off, Microsoft rocketed away from competition.

So what is the point?

The fact is, Kildall played his cards much differently than Gates did, and hence did well but enjoyed financial success vastly more modest than Gates. But it's tantalising to consider the possibility that with a few tweaks, Kildall could have been Gates. Now the book acknowledges that untangling skill and luck can be imperfect, but even some sense of the relative contributions of the two can really help you understand history and, more importantly, make better predictions of the future.

You write that "most people have a general sense that luck evens out over time. That may be true in the grand scheme of things. But the observation doesn't old for any individual, and the timing of luck can have a large cumulative effect." What do you mean by that?

In some activities, the outcomes are largely independent: what happened before doesn't affect what happens next. This is true, of course, in classic games of chance such as dice throwing or roulette wheels, but it also applies to relatively stable systems like sports. You can model the batting average of baseball players, for example, using a simple, independent model. In these cases, luck does tend to even out over time. In other activities, the outcomes are path dependent. What happens next is affected by what happened before. This is relevant in realms that are socially driven such as sales of books, music, and movies. The fact is if you ran the universe over again, it is very unlikely that the same products would be commercial smash hits. We know this through examining the results of some clever sociological experiments.

Can you explain it through an example?

One example I give in the book is the income of college graduates. It turns out that men who graduate during times of relative prosperity earn more than those who graduate during more challenging conditions. That is not so surprising. What is more surprising is that that effect remains in place for 15 years (and perhaps more) following graduation. So two students of identical ability can have substantially different incomes over a long period by dint of when they graduated. So luck is not evening out in these cases. There is a large cumulative and apparently irreversible effect.

Why do we vastly underestimate the role of luck in what we see happening around us?

Especially when social processes are at play, it's really hard to know how things are going to unfold. In other words, luck is a very large variable. You often hear executives in the entertainment industry lament how hard it is to manufacture hits. And that is true. When social processes are at play, there's an inherent lack of predictability and generally high inequality.

Any examples?

One story that captures this is that of a young singer named Carly Hennessey. Music executives were looking for the next Brittney Spears, and Hennessey had everything they were looking for-a great voice, charisma, and drive. They spent millions on her first album, which ended up a commercial flop. In the first three months the album sold a grand total of 378 copies, earning less than $5,000. There is simply no easy formula to predict success.

The flipside is true as well. When Michael Eisner, then CEO of Disney, saw the pilot of the show "Lost," he gave it a 2 on a scale from 1 to 10, with 10 being the highest. He later called it "terrible." But "Lost" was a huge success and was very profitable for Disney.

When the top executive at Disney has no idea what's going on, it's easier to accept that it's really hard to anticipate hits.

Why are we so bad at distinguishing luck from skill?

There's a module in your left hemisphere that neuroscientists call "the interpreter." Its job is to create a narrative that explains cause and effect. For most of mankind's existence, cause and effect was a pretty straightforward affair: a rustle in the bushes likely signalled danger, for instance. This module has conferred extraordinary advantage, and some scientists argue it is at the core of what distinguishes humans from other species. Now, here's the fascinating component. Studies of split-brain patients reveal that the interpreter will fabricate a cause whenever it sees an effect, even when the cause makes no sense. For example, researchers would feed information into the right hemisphere-largely absent of language-and ask the subject to explain what is going on. Since the hemispheres in these subjects are severed, there's no way to communicate. The interpreter simply makes up a story.

So here's the answer to your question. The interpreter doesn't know anything about luck.

When it sees an effect, it searches for a plausible cause. Once a cause is found, your mind puts the issue to rest. In fact, you start to believe your own story and dismiss any other possibility, a concept psychologists call "creeping determinism." So once something has happened, we tend to grossly underestimate the role of luck.

Could you give us an example?

I'll mention quickly a paper by Professor Andrew Lo at MIT. He studied about 20 accounts of the recent financial crisis - half of them by journalists and the other half by academics. He found that there was no common explanation for the crisis, and in fact some of the explanations contradicted one another.