Larry Ferlazzo’s Websites of the Day…

…For Teaching ELL, ESL, & EFL

“Goodhart’s Law” & Education Policy


Source: via Larry on Pinterest



The World Policy Institute has just published an interesting article titled “Brave New Math.” It’s focus is question whether the GDP (Gross Domestic Product) of a country is the best metric to use to judge a nation’s well-being. However, I think a number of points made can be applied to education policy issues, too, and the emphasis on being data-driven instead of being data-informed and our narrow focus on standardized test results — for measuring teacher quality and student learning.

Here are some excerpts:

Even its creator, however, realized the limitations of GDP. In 1934, Kuznets warned, “the welfare of a nation can scarcely be inferred from measurement of national income.” He wrote again in 1962, “distinctions must be kept in mind between quantity and quality of growth, between its costs and return, and between the short and the long run.” In other words, GDP and its components can and do give us a measure of how much we produce and consume—but reflect none of the qualitative aspects of the economy. GDP cannot answer such essential questions as whether we are consuming too much of the wrong things or saving too little. To any government statistician tallying GDP, $100 spent on textbooks is sadly no more valuable to society than $100 spent on cigarettes….

GDP as a statistic may have fallen victim to the phenomenon of Goodhart’s Law. Devised by an adviser to the Bank of England in the 1970s, the law states that as soon as an indicator is relied upon for policy decisions, it stops being effective. For example, the police can reduce the rate of shoplifting by diverting more resources from other crime-fighting activities. Shoplifting rates go down, but other crime rates go up. As a result, shoplifting becomes a useless indicator of overall crime trends. In this respect, when a particular yardstick like GDP is used as a performance indicator of a complex system—like a national economy—the government may choose to target the measure, improving its value but at other costs to the country. As such, GDP may improve, but it becomes less useful as a measure of the broader economy and national well-being….

By going beyond simple GDP and looking at a diversity of timely data we can better diagnose our economic health. In the global age, new economic thinking needs to be oriented around developing human capital, not blindly stoking GDP.

What do you think — are there some parallels here?


Author: Larry Ferlazzo

I'm a high school teacher in Sacramento, CA.


  1. Not only does GDP not differentiate between good spending and bad spending, but it totally ignores a lot of the work that is done in an economy. When a woman gives up a career to raise kids that will one day become productive members of society, that work doesn’t factor into the calculation. When a man builds his own porch over a weekend instead of hiring a team to do it, that work doesn’t factor in either.

    In other words: the very measure is biased only towards work done for pay. So if we use that as a benchmark for policy decisions, we are going to create policies that favor only paid work as well.

  2. There are blindingly obvious parallels between Goodhart’s Law trends in economics and the same trends in education because of policy emphasis on high-stakes tests scores. In the world of social sciences, we call it Campbell’s Law but it boils down to the same things: so much of education policy now revolves upon performance for standardized testing regimens that standardized tests have lost all value for measuring achievement in our students, teachers and schools. As soon as a school’s success ranking or a teacher’s value as an educator or a child’s worth as a student is measured by a standardized test, scoring well on that state or national test becomes of utmost importance to all parties. Instead of learning to reason logically or think critically, our students only have time to learn how to pick out the best answer from a list of four options. There is a very big difference between answering a question and using logical reasoning and previous knowledge to arrive at a conclusion based on a set of given factors but because of the punitive measures which are laid down by education policy when a school fails to meet certain “standards” (number of students passing standardized tests), there isn’t much choice about what must be taught. The data says our kids are getting an education, but what good is it to be educated in the art of filling in the right bubble?

Leave a Reply

Required fields are marked *.

Skip to toolbar