In one of the most memorable scenes in the movie Glengarry Glen Ross— a classic among salespeople on commission—the sales manager played by Alec Baldwin announces a new motivational plan: "As you all know, first prize is a Cadillac Eldorado. … Second prize is a set of steak knives. Third prize is you're fired." Reward structures like this exist in reality as well as fiction. Former General Electric CEO Jack Welch famously used a "20-70-10" system, promoting the top 20 percent of GE employees, keeping the middle 70, and firing the bottom 10 percent.
The motivating effect of incentives like this can cut either way, however. While the top prize is big (a Cadillac, a promotion), your chance of getting it is relatively slim, and your prospects depend in large part on the quality of the competition. Strong competitors are generally thought to bring out the best in everyone, but what if the competition is so strong it makes the top prize feel out of reach? Can strong competition actually undermine a reward structure? A new study by Jennifer Brown of Berkeley provides an answer to this question by looking at the world of professional golf, with Tiger Woods playing the role of the strong competitor.
Tiger is the player of his generation. For PGA Tour events between 1999 and 2006, the "nonexempt" players (that is, the ones who had to qualify for each tournament) averaged 1 to 4 strokes under par. The "exempt" players (the top 125 money winners, who qualify automatically) averaged 3 to 6 strokes under par. Tiger averaged between 10 and 14 strokes under par. Another way to measure Tiger's dominance is his world golf rating. Golfers receive rating points based on their performance in tournaments, earning more points for higher finishes. The first place finisher in the U.S. Open gets 100 points, for example, while the second place finisher gets 60. The world golf rating is a player's average points per tournament. In 2000, Tiger had nearly 30 "world ranking points"—more than twice his closest competitors (Phil Mickelson, David Duval, and Ernie Els each had about 12 points).
The purse at PGA tournaments varies between $4 million and $8 million. The prizes for the top finishers are large, but they fall sharply as you move down the leader board. First place gets 18 percent of the purse ($1.44 million at an $8 million major), second place gets 10.8 percent ($864,000), third place gets 6.8 percent ($544,000), and fourth gets 4.8 percent ($384,000).
If money motivates, then the prospect of winning the top prize should bring out extreme effort in golf. But when Tiger is playing and you're not Tiger, you face a depressed prize schedule. If you assume Tiger is going to win, then the top prize available to you is $864,000 rather than $1.44 million. That beats the heck out of steak knives, but it's significantly less than the winner's take. Second place—among players who are not Tiger—gets $544,000 rather than $864,000, and so on. While Tiger certainly doesn't win every tournament he enters, he does frequently shift the reward schedule for most of the field. Of the 219 tournaments he's played in during his first professional decade, Tiger collected 54 PGA wins, finished in the top three in 92, and in the top 10 in 132.
Tiger is thus formidable even if he doesn't always take first, which gets us to the study's question: How does his participation in a tournament affect other players' performance? It's almost a given that other players will rank lower when Tiger gets first place, but what the study asks is whether other players shoot more strokes on a given course when Tiger's in the mix.
Analyzing data from round-by-round scores from all PGA tournaments between 2002 and 2006 (over 20,000 player-rounds of golf), Brown finds that competitors fare less well—about an extra stroke per tournament—when Tiger is playing. How can we be sure this is because of Tiger? A few features of the findings lend them plausibility. The effect is stronger for the better, "exempt" players than for the nonexempt players, who have almost no chance of beating Tiger anyway. (Tiger's presence doesn't mean much to you if the best you can reasonably expect to finish is about 35th—there's not much difference between the prize for 35th and 36th place.) The effect is also stronger during Tiger's hot streaks, when his competitors' prospects are more clearly dimmed. When Tiger is on, his competitors' scores were elevated by nearly two strokes when he entered a tournament. And the converse is also true: During Tiger's well-publicized slump of 2003 and 2004, when he went winless in major events, exempt competitors' scores were unaffected by Tiger's presence.
A skeptic might ask whether a golfer can really try harder, but in fact there are many ways for players to improve their performance in a given tournament. They can study the course. They can hit more balls on the driving range. They can arrive at the tournament a few days early and play more practice rounds.
Of course, there are other possible explanations for the study's finding besides effort, such as distraction or intimidation. Maybe other golfers are just intimidated when Tiger is playing. Or maybe they are put off by the media attention lavished on the PGA's premier player. But you'd expect these alternative mechanisms to be stronger for golfers playing near Tiger, and Brown's study found that being in Tiger's foursome has no additional negative impact on performance.
What does this mean for the nongolfing world? It's generally agreed that people work harder when they are paid for performance. Anyone who has ever languished in a Paris cafe—where service compris translates roughly as "the Republic of France mandates a minimum 15 percent tip regardless of service quality"—can appreciate the power of incentives. But the effects of incentives appear to be muted when the incentives are based on relative performance and the competition is tough. We're taught that quitters never win, but if the evidence from golf is any indication, it might be more accurate, if less pithy, to say that expected losers are more likely to quit, or at least not perform as well. If you're running a business, and you have the opportunity to hire the Tiger Woods of office work, you're not going to pass up the chance. But Brown's study suggests you might want to consider its effect on your other workers' performance. Steak knives might not cut it as second prize.