With all the recent discussion of how hard it is for journalists to read academic articles, I thought I’d provide a little service here and ‘translate’ the recent NBER working paper by Daron Acemoglu, David Laibson and John List, “Equalizing Superstars” for a general audience. The paper contains a ‘light’ general equilibrium model that may be difficult for some to parse.
The paper is interested in what the effect of MOOCs or, in general, web-based teaching options would be on educational outcomes around the world, the distribution of those outcomes and the wages of teachers. It does this by starting with a model whereby different countries have different levels of student ability (at pre-school) that are correlated with different teacher skills (or available time/funding) across countries. That means that there is one country that is the ‘top of the heap.’ The assumption made is that the top country’s teachers are superstars and the web means that some of their skills can be transmitted to students around the world.
The key model element, however, is what the teachers do. They do a number of tasks (think transmitting base information and reinforcement). The web technologies allow some of those tasks to be done at the skill level of the top country (base information) and free up the teacher time to allocate more of their energy to other tasks (reinforcement). This is a pure gift to education in all countries other than the top country. They get world’s best practice on some tasks and freed up teacher time on the rest. Of course, the top country gets none of these benefits as they already had the best. This is one tick for equality but the authors also show that the ‘bottom’ countries benefit the most from the boost and moreover, the ‘next to top’ country might actually have students out-performing the top country — think of a country just below the top country, they get a whole lot of almost top teachers freed up to devote their times to non-web tasks and in aggregate could end up outperforming the top country.
Basically, what the model is saying is that ‘free gifts are good especially for receivers.’ However, there is nothing unique about the web here. What the model is saying is that whenever some teaching tasks can be moved costlessly to best practice, that is good news for educational outcomes. So it could equally apply for web technologies deployed across schools or, for that matter, textbooks which basically substitute some teacher tasks to a non-rival public good. Of course, so long as lower cost education means more education, such innovations will lead to more education. But if the educational model is provide a given amount of education at the lowest cost, this may simply mean a displacement of teachers.
So let’s focus on the teachers as we academics do really care about them. What happens to their wages? If their wages are related to the marginal productivity of the students being produced in their country, there are two offsetting effects. First, yes teachers can reallocated effort to other tasks but there are diminishing returns to that effort on each task — this tends to depress teacher impact on marginal productivity and hence, their wages. The authors call this ‘crowding out.’ On the other hand, there is a complementary effect. As a country now has some tasks at the world’s best practice, the productivity of the other tasks actually rises. This makes teachers more important and increases their pay. Not surprisingly, the bigger the boost the higher the complementary effect and so the teachers in the bottom country almost surely get paid more as a result of this. But the teachers close to the top don’t have that far to go and their wages would likely fall.
Now that conclusion may leave some consternated — but? but? surely if you are close to the top and of high skill things can’t be that bad? This is where the model’s key assumption is at work: there are a fixed number of tasks and you are always doing some of them all both pre and post-Internet. But, as any teacher would know, there aren’t a fixed number of tasks. While it is not considered in the paper, what you are able to do with students is more fluid than that and, moreover, adding other tasks requires some upfront fixed costs. That means that you won’t add a task unless you have time to do it in the future. As already mentioned, the web-based technologies are a free gift to teacher time allowing them to reallocate effort to other tasks. If those tasks are limited, the model applies. But if those tasks can be expanded, then there is increasing returns and not diminishing returns. In that case (a) all teacher wages will rise; (b) the next to bottom is more likely to outperform the top and (c) there will likely be some amplification of inequality in educational outcomes and teacher pay.
The good news is that we can actually look at past educational innovations (such as textbooks and better teaching methods) to see these things at play. Were educational outcomes more equal or more unequal as a result of these past technologies? Actually, we might be able to see these effects in the earnings of those providing the technologies.