Bill the engineer wants to become a teacher.

He has 10 years of experience working in the engineering division of Lockheed Martin, and he'd like to share some of his extensive knowledge with high school students in Northern Virginia, where he lives. He'd prefer to take a couple of hours each day to teach a class on physics or calculus, which would enable him to stay in his current job. Bill imagines that this part-time teaching job will give him the opportunity not just to teach, but to mentor local students aspiring to science careers.

So Bill goes to the principal of the local public high school with his proposal. Before we detail the vast array of statutes and regulations governing who is allowed to teach in public schools, let's pretend--for a moment--that those regulations don't exist. Just consider how, in an ideal world, the principal would react to Bill's offer.

First, the principal needs to verify that Bill can be an effective teacher. How might the principal do that? Perhaps require him to give practice presentations of difficult material. Then maybe Bill should shadow seasoned teachers for a period of time to get a feel for classroom management and lesson planning. When Bill does get his own classroom, the principal will want to check each year that his students are learning what they're supposed to learn.

Bill is not likely to consider these requirements overly burdensome or bureaucratic. But in the real world, what would it actually take to allow Bill in the classroom? Let's examine the tangled web of teacher certification law.

First of all, it will be difficult for Bill to find a part-time teaching job. Teaching a single class period is rare in a regular public school, so it may be full-time or nothing for him. That's the first hurdle. So let's hope that Bill is open to giving up his engineering job entirely.

The most straightforward option for Bill is to get a master's degree in teaching, which includes semester-length courses of questionable value or relevance to STEM education -- adolescent development, lesson planning, and contemporary issues in education. This will set him back about $30,000 after two years of full-time study, which seems like an awfully time-consuming and expensive commitment for someone who is already an expert in his field.

Thankfully, the Virginia state government has tried to make it easier for mid-career professionals such as Bill to become teachers without getting another master's degree. He could participate in Virginia's Career Switcher Program. That can't be too complicated, right?

To go this route, Bill must first apply to the Career Switcher Program through one of the state-approved institutions that offers it -- for instance, George Mason University or the Virginia Community College System. To apply, prospective teachers must have at least five years of work experience in their field, hold a bachelor's degree, possess sufficient subject-matter coursework, and have passed the teaching exams approved by the State Board of Education.

Once admitted into "Level I" of the career switcher program, Bill must complete 180 hours of courses in, among other things, human growth and development, differentiated instruction, and classroom behavior, along with in-service teaching.

After that process, a provisional license is awarded, and Bill can enter the classroom. Now he is in "Level II". But during that second year (which is a full-time paid teaching position), he must continue amassing credentials by attending five instructional seminars. After successfully fulfilling those requirements -- which, combined with the Level I courses, may cost upwards of $6,000 -- a prospective teacher can apply for a 5-year renewable license, which typically takes another four weeks to process.

If all of that sounds overly complicated, Virginia also offers a route that does not go through the Career Switchers program. But the alternative route has similar requirements. And upon completion, prospective teachers receive licenses that are good for three years and are not renewable. Bill would still have to obtain traditional certification to continue as a teacher after three years.

And there's more. Once Bill has obtained his provisional license through this alternative route, he must satisfy any remaining "endorsement deficiencies" in the content area in which he is teaching, and must complete 15 hours of professional development classes to qualify for that five-year renewable license.

It's not hard to imagine that such a time-consuming process might dissuade Bill from trying his hand at teaching. This is highly unfortunate: A 2010 report published by the President's Council of Advisors on Science and Technology noted that there are "approximately 20 million people in the United States who have degrees in STEM- or healthcare-related fields [who] can potentially be a tremendous asset to U.S. education." But by smothering teacher hiring in regulations and requirements, governments can actually dissuade the most talented people from trying to become teachers in the first place.

These regulations are always couched in terms of assuring teaching quality. Only with strict standards, we are told, can we be confident that the best teachers are in our children's classrooms. The American Federation of Teachers (AFT) has recently taken this view to an extreme, calling for a new "bar exam" for teachers. Its recent report, "Raising the Bar," argues that "teaching must raise standards for entry into the profession through a process similar to the bar process in law or the board process in medicine."

It's not surprising that the AFT would endorse such onerous entry requirements. Limiting entry into the profession means less competition for current teachers -- and maybe even a profession-wide pay increase, if schools begin struggling to find applicants willing to jump through all of the new hoops.

Nonetheless, it seems hard to dislike the AFT's idea at first blush. After all, won't the bar exam help ensure we get the smartest and most qualified teachers? Even Joel Klein, an enterprising reformer when he was school chancellor in New York City, has fallen for this view.

But it's wrong. This rationale fails for two main reasons.

First, barriers to entry deter people from pursuing teaching in the first place. We have already discussed Bill the engineer, for whom becoming a teacher would require a commitment so time-consuming and expensive that it's hard to believe it would be worth his effort. A bar exam for teachers would likely discourage him even more.

And it's not just mid-career professionals like Bill who are deterred by onerous licensure requirements. High-ability college students must sacrifice time spent studying math and science in order to take required education courses and bone up on the latest trends in pedagogy. Smart students on the fence about whether they want to become teachers will likely choose the math and science courses (which have broad labor market value) rather than wasting time on education courses (which have value only if they pursue teaching).

The second reason a bar exam will not improve teacher quality is more fundamental: Scores on these tests are simply not good predictors of teacher effectiveness. In fact, economists and education scholars have known for decades that the standard resume characteristics -- level of education, certifications or licenses, and experience beyond the first few years of teaching -- have essentially zero power to predict how much students learn from a given teacher. Even raw intellectual ability as measured by IQ tests has only a small positive effect on how much knowledge teachers are able to impart to their students.

To illustrate more vividly, a teacher with a doctorate degree, every certification and license available, and 15 years of experience is no more likely to be a high performer than a teacher with a B.A., the minimal certification, and five years of experience.

On reflection, this should not be too surprising. Clearly teachers need to be intelligent and knowledgeable, but effective teaching requires a rare blend of patience, empathy, articulation, and motivation -- qualities that cannot be easily measured on a bar exam or other standardized test.

For all of these reasons, a bar exam is not any more likely to put effective teachers in the classroom than existing certification tests are. This is especially true if the bar exam covers faddish pedagogical theories that often lack a scholarly foundation.

There is a better way to improve teacher quality. It's essentially the opposite approach of the one espoused by the AFT. School districts should be less discriminating in choosing new teachers, but more discriminating in deciding which veteran teachers receive tenure. This approach may seem counterintuitive, but for teachers, the best predictor of future results is past performance. Teachers who show strong performance -- as measured by student tests and principal evaluations -- should quickly move up the pay scale, while those who perform poorly should be let go or denied raises.

When economists Douglas O. Staiger and Jonah Rockoff simulated how this kind of system might work, they found that it would involve a rather high rate of turnover. In their view, only the top 20 percent or so who performed best during their tryout period should be kept on.

Whether Bill the engineer would make the cut is unknown. But lowering barriers to entry into the classroom would at least give him the incentive to try.