Tuesday, November 15, 2016

A Big Data Evaluation on the Causes and Policies Regarding Upward Mobility

286 days until I start classes.

I hope.

Trump's victory will likely make it the case that I have no option but to be a serf working in some corporate-feudal empire, as that will become the only way to get access to health care. Of course, the corporate-feudal lords and masters will love the power that this gives them over workers. Unfortunately, I am not in the corporate-feudal class. I am in the worker class.

And since, unlike some very wealthy but very ignorant media pundits expressing opinions that have no basis in morality, I do not say the things that corporate-feudal lords want said, so I am not in a position to get much money from them.

Well . . . that is my rant for today.

If you like your economic and social policies grounded on hard data, I would like to recommend a series of three lectures from the London School of Economics that I have just finished. I needed to wait until the end of the third lecture to make sure that my biggest objections to the first two lectures were addressed. He addressed them directly and to my satisfaction.

Professor Raj Chetty presents a "big data" examination of facts and policies regarding upward mobility. The conclusions that he draws comes from a huge data set - and he explains how he and others use this huge data set to try to figure out how to improve the upward mobility of people born into households in the bottom 20%.

Upon discovering what aids in upward economic mobility - what allows the children of the poorest families to be better off than their parents - he looks at policies that should improve upward mobility.

His data yields conclusions that traditional conservatives and traditional liberals both will not like.

In the second lecture, he discussed education policies, and makes the case that children would be better off if principles had the power to fire the worse teachers and replace them - even replacing them with average teachers would be an improvement. He also argued that principals are quite adept at identifying the best teachers so that allowing principles to base teacher pay on merit rather than experience would likely result in significant improvement.

The quality of teachers, as it turns out, has a massive effect on the upward mobility of students - far more than class size or, for example. Consequently, we can gain more by giving an excellent teacher a larger class than we can by cutting the size of the class and assigning some of the students to a worse teacher.

His lecture focused not so much on policies that would require that we spend more money, but argued that the money we spend on policies can be spent more efficiently. In fact, some of our policies to help lower-income households are counter-productive.

For example, the way we handle programs that allow families to move into better neighborhoods is built in such a way that, for many children, it does more harm than good. Without spending any more money - but instead by redesigning the policies such that the money spent in them are spent more productively - we can get much better results.

One of the conclusions that he supports from this big data project is that children that are born and raised in integrated communities - as opposed to segregated communities - tend to experience greater upward mobility.

In the final lecture, he addressed some of the concerns I was having with the first two lectures such as: if somebody goes up in social mobility doesn't somebody else have to fall down to take their place? Here, he talks about two social goals - equality and overall wealth - and shows how the policies that he discussed earlier promote both goals at the same time. They both increase the overall size of the pie, and gives each person a more equal slice.

He even ventures into taboo territory and argues that there may be some genetic influence on economic success - though it is far from being the whole story.

I still saw a couple of inconsistencies in his argument. For example, while he talks about giving the students with the best math grades some type of assistance that would allow them to have the type of exposure that promotes upward mobility, he argues against using school vouchers to remove "the creme of the crop" from lower quality schools.

As I said, if you want your policy opinions to be based on data rather than repeating the "de dicta" beliefs of your social group, then this might be worth looking into.

There are three lectures.

The Geography of Intergenerational Mobility.

Policies to Remove Upward Mobility

Upward Mobility, Innovation, and Economic Growth.

No comments: