Bill Gates has loomed large in education for the past decade. The reason is obvious: his foundation is the largest in the world, and districts are more than willing to accept his conditions in return for his money.

When anyone asks Gates whether it is right that one man and one foundation should have so much influence, he says that the money he gives is minuscule compared to the hundreds of billions spent annually by American schools. But he is being disingenuous, and he knows it. Almost all of those billions are fixed costs, whereas his money is discretionary. A district with a huge budget–often facing budget cuts—will dance to Bill Gates’ tune. All he need do is dangle $50-100 million dollars, and district leaders will do as he asks.

But what happens when he is wrong? In the first decade of this century, he said that small high schools were THE answer, and districts lined up to get money and break up large high schools. It wasn’t a bad idea, but he decided that it wasn’t THE idea, and in 2008, he decided it wasn’t producing the miraculous results he wanted (ROI–return on investment), and he dropped it.

Since he can’t tolerate being without answers, he next placed his bets on raising teacher quality. A good idea poorly executed. Instead of changing working conditions or coming up with other ways to make teaching a rewarding profession, Gates chose to go the punitive route. He decided that all of American education was broken, and that teacher evaluation was the most broken part of it. For whatever reason, administrators were not weeding out the incompetents, and he decided to make that his mission. He never stopped to ask why 40% or so of new teachers left teaching within five years of starting.

How to evaluate millions of teachers? Gates had the answer. Use the test scores of their students to a significant degree to find out who was best and worst.

Given Gates’ unusual power, the U.S. Department of Education decided that he must be right, even though the research was thin and speculative. No need to conduct experiments to see if Bill was right. He is so rich, he must be right. So, Race to the Top required states to include Bill’s idea– judging teachers by their students’ test scores to a significant degree–if they wanted to be eligible for any part of the $4.35 billion prize, or later, if they wanted a waiver from NCLB’s punishments for failing to make 100% of their students proficient by 2014.

Some districts have now experimented with “value-added assessment” for four years, and no miracle is in sight. Most researchers say the methodology is flawed that it will never work. The most recent study, conducted by Andy Porter, dean at the University of Pennsylvania, and Morgan Polikoff of the University of Southern California, found little or no correlation between teacher quality and VAM ratings. This study was funded, ironically, by the Gates Foundation.

The question now is, will Bill Gates have the courage to admit he was wrong, as he did in 2008?