If I hear once more that the reason for the wage gap is that women don't negotiate, I may just blow a gasket.

Linda Babcock herself, the author of the studies that gave rise to the "women don't ask" industry, has shown that women don't negotiate for a very simple reason: they sense—correctly—that it will hurt them if they do.

Babcock and her colleagues found that women don't negotiate their initial salaries as much as men. No doubt you’ve heard that. This finding has received a wild amount of coverage in the press.

What you probably haven’t heard is what happens when women do negotiate. Often they end up worse off than if they’d kept their mouths shut.

A 2006 study Babcock did with Hannah Riley Bowles and Lei Lai helped explain why women are less likely to negotiate their starting salaries (referred to as the Bowles study). When they do, both men and women are less likely to want to work with, or to hire, them. The effect size is large. Women who negotiated faced a penalty 5.5 times that faced by men.

Women who negotiate often trigger backlash for gender deviance. Women are supposed to be nice and not overly demanding. Women who negotiate flout these mandates. The Bowles study found that women’s tone didn’t matter much. Whether women asked simply or assertively mattered very little. A recent New York Times article advising women to simply smile when they negotiate is highly misleading, unless I am overlooking something. (I hope I am; let me know.)

“How to Attack the Gender Wage Gap? Speak Up,” we hear yet again from the newspaper of record. The WAGE program discussed in the article is a good idea, as long as its trainings take into account the complexity of teaching women how to negotiate without triggering backlash.

But it’s complicated. If women act too feminine and don’t ask, they end up with lower salaries. If they act too masculine and ask, then people don’t want to work with them. Women walk a Tightrope between being too feminine and too masculine. Men don’t, which is one reason why office politics are trickier for women than for men.

This is one of the central points of The New Girls' Network, and my book What Works for Women at Work (forthcoming 2014), co-written with my daughter Rachel Dempsey. We document the everyday ways the Tightrope plays out in women’s lives and provide proven strategies for navigating offices still shaped by subtle gender bias. For the Tightrope, the solution is to mix the masculine and the feminine. But that’s not as easy as it sounds. Remember that women who did not ask assertively still got penalized.

Women don’t negotiate because they’re not idiots. They sense a startlingly traditionalist substrata: that men are entitled to be “hard-driving” but women are not. A “successful man” is one with a big salary. It’s unseemly for women to want a “big one” too.

Organizations need to take responsibility here, rather than continuing to tell women they have only themselves to blame.

The key problem here, never discussed, is with the business processes that artificially advantage men. If your organization relies on incoming employees to negotiate for themselves, men will be able to negotiate without holding back for fear of alienating people but women will not.

The solution is a new system for setting starting salaries. Some universities fix the problem by having the department chair negotiate start-up packages for all incoming professors. This solves the problem of having women scientists fall behind the men because the men can negotiate for fully-equipped labs and the women find they can’t.

Another approach is suggested by a study by Hannah Riley Bowles, Linda Babcock, and Kathleen L. McGinn. This study found that women do better in situations with “low structural ambiguity”: where there is more information and lower uncertainty about the potential salary range and appropriate norms for negotiation. What if everyone—both the candidates and those hiring them—were told that incoming candidates are supposed to negotiate their salaries, along with a brief reference to the research on biases sometimes triggered when women do negotiate?

The point is that redesigned business processes to control for implicit bias are rarely discussed. What we have today are implicit bias trainings that have limited effectiveness, according to studies by Frank Dobbin and others.

Here’s a proposal. What if we link every program designed to fix women with a program redesigning basic business processes to stop artificially advantaging men? Food for thought.