Skills lost due to COVID-19 school closures will hit economic output for generations (hmmm)

Snippet from OECD report on covid-19 and education This CBC report is one of many dozens of articles in the world’s press highlighting one rather small but startling assertion in a recent OECD report on the effects of Covid-19 on education – that the ‘lost’ third of a year of schooling in many countries will lead to an overall lasting drop in GDP of 1.5% across the world. Though it contains many more fascinating and useful insights that are far more significant and helpful, the report itself does make this assertion quite early on and repeats it for good measure, so it is not surprising that journalists have jumped on it. It is important to observe, though, that the reasoning behind it is based on a model developed by Hanushek and Woessman over several years, and an unpublished article by the authors that tries to explain variations in global productivity according to amount and  – far more importantly – the quality of education: that long-run productivity is a direct consequence of the cognitive skills (or knowledge capital) of a nation, that can be mapped directly to how well and how much the population is educated.

As an educator I find this model, at a glance, to be reassuring and confirmatory because it suggests that we do actually have a positive effect on our students. However, there may be a few grounds on which it might be challenged (disclaimer: this is speculation). The first and most obvious is that correlation does not equal causation. The fact that countries that do invest in improving education consistently see productivity gains to match in years to come is interesting, but it raises the question of what led to that investment in the first place and whether that might be the ultimate cause, not the education itself.  A country that has invested in increasing the quality of education would, normally, be doing so as a result of values and circumstances that may lead to other consequences and/or be enabled by other things (such as rising prosperity, competition from elsewhere, a shift to more liberal values, and so on).  The second objection might be that, sure, increased quality of education does lead to greater productivity, but that it is not the educational process that is causing it, as such. Perhaps, for instance, an increased focus on attainment raises aspirations. A further objection might be that the definition of ‘quality’ does not measure what they think it measures. A brief skim of the model used suggests that it makes extensive use of scores from the likes of TIMSS, PIRLS and PISA, standardized test approaches used to compare educational ‘effectiveness’ in different regions that embody quite a lot of biases, are often manipulated at a governmental level, and that, as I have mentioned once or twice before, are extremely dubious indicators of learning: in fact, even when they are not manipulated, they may indicate willingness to comply with the demands of the powerful more than learning (does that improve GDP? Probably).  Another objection might be that absence of time spent in school does not equate to absence of education. Indeed, Hanushek and Woessman’s central thesis is that it is not the amount but the quality of schooling that matters, so it seems bizarre that they might fall back on quantifying learning by time spent in school. We know for sure that, though students may not have been conforming to curricula at the rate desired by schools and colleges, they have not stopped learning. In fact, in many ways and in many places, there are grounds to believe that there have been positive learning benefits: better family learning, more autonomy, more thoughtful pedagogies, more intentional learning community forming, and so on.  Out of this may spring a renewed focus on how people learn and how best to support them, rather than maintaining a system that evolved in mediaeval times to support very different learning needs, and that is so solidly packed with counter technologies and so embedded in so many other systems that have nothing to do with learning that we have lost sight of the ones that actually matter. If education improves as a result, then (if it is true that better and more education improves the bottom line) we may even see gains in GDP. I expect that there are other reasons for doubt: I have only skimmed the surface of the possible concerns.

I may be wrong to be sceptical –  in fairness, I have not read the many papers and books produced by Hanushek and Woessman on the subject, I am not an economist, nor do I have sufficient expertise (or interest) to analyze the regression model that they use. Perhaps they have fully addressed such concerns in that unpublished paper and the simplistic cause-effect prediction distorts their claims. But, knowing a little about complex adaptive systems, my main objection is that this is an entirely new context to which models that have worked before may no longer apply and that, even if they do, there are countless other factors that will affect the outcome in both positive and negative ways, so this is not so much a prediction as an observation about one small part of a small part of a much bigger emergent change that is quite unpredictable. I am extremely cautious at the best of times whenever I see people attempting to find simple causal linear relationships of this nature, especially when they are so precisely quantified, especially when past indicators are applied to something wholly novel that we have never seen before with such widespread effects, especially given the complex relationships at every level, from individual to national.  I’m glad they are telling the story – it is an interesting one that no doubt contains grains of important truths – but it is just an informative story, not predictive science.  The OECD has a bit of track record on this kind of misinterpretation, especially in education. This is the same organization that (laughably, if it weren’t so influential) claimed that educational technology in the classroom is bad for learning. There’s not a problem with the data collection or analysis, as such. The problem is with the predictions and recommendations drawn from it.

Beyond methodological worries, though, and even if their predictions about GDP are correct (I am pretty sure they are not – there are too many other factors at play, including huge ones like the destruction of the environment that makes the odd 1.5% seem like a drop in the barrel) then it might be a good thing. It might be that we are moving – rather reluctantly – into a world in which GDP serves as an even less effective measure of success than it already is. There are already plentiful reasons to find it wanting, from its poor consideration of ecological consequences to its wilful blindness to (and causal effect upon) inequalities, to its simple inadequacy to capture the complexity and richness of human culture and wealth. I am a huge fan of the state of Bhutan’s rejection of the GDP, that it has replaced with the GNH happiness index. The GNH makes far more sense, and is what has led Bhutan to be one of the only countries in the world to be carbon positive, as well as being (arguably but provably) one of the happiest countries in the world. What would you rather have, money (at least for a few, probably not you), or happiness and a sustainable future? For Bhutan, education is not for economic prosperity: it is about improving happiness, which includes good governance, sustainability, and preservation of (but not ossification of) culture.

Many educators – and I am very definitely one of them – share Bhutan’s perspective on education. I think that my customer is not the student, or a government, or companies, but society as a whole, and that education makes (or should make) for happier, safer, more inventive, more tolerant, more stable, more adaptive societies, as well as many other good things. It supports dynamic meta-stability and thus the evolution of culture. It is very easy to lose sight of that goal when we have to account to companies, governments, other institutions, and to so many more deeply entangled sets of people with very different agendas and values, not to mention our inevitable focus on the hard methods and tools of whatever it is that we are teaching, as well as the norms and regulations of wherever we teach it. But we should not ever forget why we are here. It is to make the world a better place, not just for our students but for everyone. Why else would we bother?

Originally posted at: