A polite liberal response to policy frustration is to just say “well we just need more education.” Are we frustrated about Trump’s election? Well we must need more education to fix those Trump voters. What about the connected epidemics of obesity, suicide, and teen pregnancy? Well if we just had good gym class and better counselors that would fix it all, right?
On the face of it this sort of response seems polite and well-intentioned, and it kinda is. It replaces actual contempt for bad decisions with a sort of paternalistic pity — if only they had my education they would think like me. Ultimately though, I thinkIt ends up being pretty degrading, and, worst of all, there’s really not much reason to think that education has as much of an effect as we want it to. When we fall back on this response in political arguments we are really putting forward a polite but unhelpful idea that distracts from policies and problems that may be more important levers for change.
There are a lot of popular arguments for trying to show why education is such a central driver for change. There’s truth to those arguments, but our interpretation of them is almost always wrong. In particular, we like to show the massive disparities in income and unemployment depending on your education level (Figure 1). We also like to show how common measures of academic success are usually well correlated to income. This sort of works across the board. If you have a favorable trait regarding health, wealth, income, or even life expectancy, the odds are that there is a positive correlation between that trait and educational success as shown by test scores or educational attainment.
The really big problem at the heart of all of this is that school based factors explain only around 30% of academic outcomes (figure 2) (studiesarehere). So although academic outcomes have a big influence on things like employment, income, and health, the academic outcomes themselves are ~70% explained by outside factors like demographics and geography.
The school has a very small influence on the student’s success within the school itself. And to make matters worse, although educational attainment is predictive of different measures of success and well being, it isn’t perfectly predictive and it can have low predictability in a lot of cases. When you then take an imperfect measure like educational achievement and compound the fact that only a small fraction of that is in the control of the teachers and administrators in the school, it means that teachers as a whole likely have a very small impact on lifetime student outcomes.
Now it should be noted that, on the margins, there are some economists that make highly contested arguments that very effective teachers can have large impacts on the future earnings of their students. But just based on the nature of standardized testing, the broad improvement of teaching quality is a possibly zero-sum game. The standards for an 85th percentile teacher today are the standards of a 50th percentile teacher when everyone improves. The same goes for test scores. If an underachieving school is looking to improve in their test scores, they have to improve over-and-above the rate that everyone else is. Otherwise, if a 25th percentile school improves at the rate of everyone else in standardized testing, even though it’s new level may have gotten them 50th percentile 10 years ago, it’s still 25th percentile now. That’s the double edged sword of standardized tests. Because standardized tests are relatively measures that put you on a continuum within a population, wanting one group’s standardized scores to go up necessarily means that another group’s scores need to go down.
The main point is that education has less control over important social and economic outcomes than we want it to and that’s mainly because educational success is largely out of the hands of the school. This is not to say that education isn’t worth tremendous state investment and this is not to say that we shouldn’t still try to continually improve our schools. This is mostly about not scapegoating our social problems on education. We can’t ask schools to fix problems that need to be addressed by deeper and more complicated social policies that are related to the issues of poverty, housing, food, and health. By itself, education can’t save us.