Short people get the short end of the stick when it comes to wages, with added height worth nearly $600 per inch annually.

Now two Penn economists have done some research that shows, in short, that the cause for the wage shortfall for short people was neither employer discrimination against shortness nor unfair admiration for height.

Much to their surprise, Professor of Economics and Finance Andrew Postlewaite and Professor of Economics Nicola Persico, along with graduating Ph.D. Dan Silverman, found that height at age 16 was the significant factor.

Their findings, though strongest for age 16, seem to apply to the teen years, and definitely exclude ages 11 and younger and 23 and older. “It was teen height that mattered,” said Persico.

The study confirmed that the tall guys earned the most, but the research team then found that it was the tall guys who were tall at the age of 16 or thereabouts who earned more. Tall adults who measured below average at 16, on the other hand, earned less. And conversely, short men who were above average at 16 earned as if they were tall.

Looking for a cause, they concluded that people who were short in their teen years did draw the short straw as far as feeling accepted. Shorter teens were less likely to feel comfortable participating in social activities—like athletics, clubs, student government—that are associated with things like self-esteem and social adaptability, the report stated. And so taller teens eventually come to the job market, whatever their adult size, with higher self-esteem, broader social skills, discipline and motivation. And the study suggests that’s what they eventually get paid for, not for their adult height.

Looking at a number of other factors that might affect height at age 16 and earning later in life, Persico and Postlewaite found nothing to change their focus on height and age. “It doesn’t matter how far through puberty you are,” said Postlewaite. “What matters is the height.”

They eliminated poor nutrition as an issue. “If people are malnourished, what should matter is the early height [at ages 7 or 11].” But early height came up short, as did family background, schools attended, wealth and native intelligence.

The economists based their findings on two enormous data sets, the National Longitudinal Study of Youth (NLSY) in the United States, and the National Child Development Survey (NCDS) in Britain. Both studies surveyed thousands of people, and then surveyed them again, periodically. The data in both surveys were consistent.

The NCDS, which surveyed its subjects at ages 11, 16 and 23, allowed the researchers to focus in on the age of 16. They focused on white males to avoid the confusion that the effects of gender and race discrimination might have on the height discrimination data.

“The study would indicate that social environment has long-term social and economic consequences,” said Postlewaite. But not, it appears, on the researchers. “All of us claim to have been short as teenagers. None of us are tall,”he said.