How to do Political Experiments

The Milwaukee Parental Choice Program started some twenty five years ago.  Some things have changed, but the basic idea has remained the same.     After 25 years of vouchers the jury is still out in many respects.  Anyone reading this blog can do some research to fact-check this entry.  There is much to be learned.   It is clear now that this voucher program has not produced the hoped-for educational paradise.  Comparing test scores between public and voucher schools in Milwaukee in 2011 (when the rules were changed to require testing of voucher students), 2012, and 2013 show that public schools out performed voucher schools to some degree. Milwaukee public and voucher schools both perform lower than the state as a whole, which suggests that urban poverty and family instability make it difficult for schools and students alike.

Those who think outside the box may want to notice that the failure of both public and vouchers schools to improve the performance of children of poverty is a big takeaway from this debate.  Parents seem to be very happy with the program.   But is the goal happy parents?  Or should we use this lack of performance to redirect the education reform effort.  Failing schools are not buildings that are substandard or teachers that do not teach.  Failing schools are full of failing students.  Indeed, many schools today that pass for outstanding schools do so by excluding poor students, not by better teaching.  One can turn any failing public school “around” just by letting the school pick the students instead of letting the parents pick the school.  As we see, this does not improve the educational process, it just fills schools with students that are easy to teach.

Improving instruction in failing schools will have to address the needs of the individual students that arrive at school unprepared to learn.   Students arriving at school stressed from family issues, hungry, lonely from not seeing parents, tired from lack of sleep (and the list goes on) will understandably have trouble learning.   There is a large body of research showing how adults do not perform well when their mental resources are depleted.   All of us have had days when we could not bring 100% to our job because of family or health situation.  How is a child supposed to cope with this when they arrive at school mentally depleted nearly every day?   (For more about the depleted brain concept read Thinking Fast and Slow by Daniel Kahneman.)

There have been other predictable problems.   Over time the accountability measures have been improved in Wisconsin. Many supporters and detractors agree that there needs to be better tracking of money and events.   Schools now must be approved to be part of the program.  They must  give the same tests in reading and math.   There is still a push for greater accountability.  Any time you are giving money away there must be tracking of how it is spent.   There would be no need for accountants in this world if everyone behaved themselves all the time.  My personal prediction was that schools would spring up in order to get the money.  “Joe’s Voucher School”  The two voucher schools that closed unexpectedly in Milwaukee were both rather like this.

But when do the accountability measures turn a private school into something else?  In Milwaukee and Louisiana private schools taking vouchers are called voucher schools.   The accountability measures (some would say “strings”) attached to the money are much more complex in Louisiana.   Over time any normally cynical student of government would expect voucher schools accountability measures, AKA regulations, to be come more detailed and complex, whether from changes in the political climate or from court decisions.  The voucher could be the death of both public and private education.

Now, to take this in a second direction, Wisconsin is mired in an experiment that did not work as promised. There is still a debate about the program, but everyone on both sides knows that it has not been simple magic solution. Why do we seem to be hellbent on joining in these experiments here in Oklahoma. Allowing entire school systems to become charter schools, Common Core (thankfully mostly gone), cutting income taxes to spur economic growth, we have either tried or have pending legislation in all these areas and I am sure others. Oh.. The teacher evaluation program comes to mind as well.

When are we going to let other states experiment on their citizens? That is the way Doctors experiment with drugs. They don’t take the drugs themselves.   They give them to other people. Let’s let other states do the trials.  If any of the ideas really work as promised then we can consider them here.

The hubris that allows us to think our theories must be right seems to be built into humans.  Part of my conservative political beliefs has to do with learning from the mistakes of the past.  In his book “Radical Son”,  David Hurewitz comments on the attempts by communists to get it right instead of admitting that communism is a failure. David was a proud member of the new left during his early life who had second thoughts. According to Hurewitz,   the prevailing attitude was (and is) that the various communists regimes failed because they did not get it right.  But now (hubris rears its ugly head) we will get it right because,  well, just because.  Because communism looks good on paper.  It agrees with their world view.  So it must work if we just do it right. Unlike those who have gone on before.

Conservatives have fairly criticized the left for governing by theory and refused to believe that the problem was embedded in the system.  But now conservatives are doing the same thing.  This masquerades for conservatism because they theories are rooted in conservative world views.  Milton Friedman first discussed the idea of school vouchers.  He was all about rational actors and rational choices.   But while he was still with us he never had the chance to do a large scale to see if his idea proved out.   And now this idea is still pushed very hard by the Friedman Foundation even though there is increasing evidence that it does not work.  Sometimes a theory, even a seemingly reliable one, is not enough.

Most all of the current school reform agenda is based on theory.  Some, like value-added teacher evaluation, is based on some impeccable research.  But the interpretation and implementation of the research has not resulted in good public policy.   Some is based on a misunderstanding of statistics.

Misunderstanding of the law of small numbers is too common.  Bill Gate’s failed Small High Schools initiative is  a good example.  Gates noticed that there were an unusual number of small high schools in the top performing group and spent a large amount of money busting up big schools into smaller schools because he neglected to see that there were also an unusual number of small high schools on the low performing group.  Since testing of smaller numbers results in more variation of performance the smaller schools are more likely to trend to either end of the curve.

Another common misuse is the confusion about causality versus relationships.  Someone pointed out that there was a relationship between student test scores and their evaluations of their teachers.   This is not a surprising discovery.  But this does not mean that it is a good idea to have students evaluate teachers and use that on their professional evaluation.  Anyone that has taught for any length of time has had a class that was very hard to teach.  All of the normal nice-teacher demeanor had to be thrown out the window in order to get anything at all completed.  Everyone, the principal, the counselors, the cafeteria ladies, the teacher, and the custodian, knows that the average score of this class on tests at the end of the year is going to be awful.    Some students may do well, but for the most part it will be a big fail.    To teach and maintain order in this class the teacher is not going to be giving out warm fuzzy strokes all the time.  The teacher may very well have to seek medical treatment for vocal problems.   This is not going to be a fun class for teacher or students.   The students will most likely give the teacher poor marks and the class will not give this teacher a good evaluation.  And this evaluation should never be used to determine how the teacher is doing her job.

If there are any studies showing the value of teacher merit pay please link them in the comments.  This idea is one of the oldest ideas in the reform movement.   There are some position papers that source many references that support merit pay. But controlled studies that collect data on merit pay do not support this idea.   A study of math teachers in Nashville from 2006 to 2009 compared student performance depending on their teacher’s eligibility for merit pay.    The students taught by teachers who were eligible for bonus pay did not out perform the students taught by teachers who were not offered rewards.   Several times at widely spaced intervals in history merit pay has been tried and abandoned.

Common Core is no longer the law of  the land in Oklahoma.  It was ad0pted into law before the standards were even released.  There was never a pilot study to see how it worked.   It was rushed to market.  It was another experiment, and a damned expensive experiment, at that.  We shall see how it plays out. But I predict another fail.   This because the main reason that students fail in schools has never been curriculum standards.

All this brings us to one of Oklahoma’s biggest and most ill-considered experiments.   Arthur Laffer published a well-known but study comparing the seven states with no income tax with the seven states with the highest income tax rates.  Many economists have sliced this paper to bits.   Laffer cherry picked data – He chose to highlight factors that supported his case although those factors were not causal.  He chose population growth as a measure of economic growth because it matched up well with the story he was selling.   And then he wrote a “report” about how wonderful things would be in Oklahoma if we could just do away with our income tax.

This effort to insure economic growth through tax cuts was bound to fail eventually. I was sitting in Senator Brown’s office in the Oklahoma State capitol a little over two years ago discussing this issue.    He pointed out that in the first year of the tax cut the tax collections went up.   And I pointed out that the income tax collections when up because of the oil boom.   And I promise I said, “Eventually the price of oil will fall and you are going to be in a lot of trouble.”   I have no record of it.  I also had no clue that the drop in oil was going to be so quick and so far.  But I knew it was going to happen.    I wish I had written it in a blog post or a letter to the editor so I could have appeared prescient. The price of oil is cyclical.  And hard to predict.   Anyone as old as I am has lived through more than one cycle.  We should know better.  It goes up for awhile, just like Joseph’s seven fat cows, and then down for awhile, just like Joseph’s skinny cows.

Today I read that a state senator wished he had delete an email instead of sending it.  He was obviously vexed that he was being blamed for a revenue failure.  How was he supposed to know the price of oil was going down.  He should have known.  Anyone paying attention to the price of oil in Oklahoma, which should be a lot of people, should have known.

The revenue failure is a based on a tax and fiscal policy that was a failed experiment.  The experiment failed because Art Laffer’s recent studies have not been accurate.  Many economists in Oklahoma sounded the alarm.  But they were accused of having an agenda.  The legislature owes economists in Oklahoma a big “I’m Sorry!”

The legislature enthusiastically destroyed our state’s fiscal health by basing it on two inaccurate papers that were misleading at best.

Now it is up to them to man-up and fix it.

One last thing:  There is no reason that supporting any of these ideas should be a requirement for being conservative.  Conservative ideas are intelligent ideas that are informed by the past.  None of these reforms pass this test.

It is time to let some other states be cutting edge with their unproven reforms and see how they do.

If you want to read a dissection of Art Laffer’s work which I reference you can find it here:


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s