Want More Infrastructure? Invest in Human Capital

The kind of investment sustainability that’s getting the most love at the moment is environmental sustainability: we want our buildings, machines, and infrastructure to be low carbon (as we should). But financial sustainability still matters, too: if they are to be a success, solar power plants and wind farms need consumers who can pay for their electricity. That means those investing in renewable infrastructure should ask: what makes for more customers who can pay for our services? A big part of the answer is health and education.

George Yang and I have just published a paper that reviews the link between investment in physical capital (like power plants) and human capital (like education). The short answer is that investments are complementary: infrastructure creates demand for a more educated workforce that sees higher incomes if there is a good stock of infrastructure. That’s been true for a long time: when electricity networks first spread across America, firms that had more equipment and got more of their power from electricity started hiring more educated workers, who they paid more.

George and I explore some more recent evidence that points in the same direction: looking at the 151 countries with World Bank enterprise surveys reporting what firms cite as the most serious obstacle to their business, it appears that firm concern over electricity is higher in countries where education is not seen as a barrier and vice-versa. Take two measures: first, as an indicator of comparative satisfaction with human capital, the proportion of manufacturing firms in a country that do not suggest the quality of workforce education is the biggest barrier, and second, as a measure of infrastructure quality, the percentage of electricity not self-generated amongst firms with a generator. Countries are divided into three groups: (i) those with reliable electricity and with firms satisfied with education, (ii) those with reliable electricity but where firms are dissatisfied with education (iii) those with unreliable electricity but satisfaction with education. There is a fourth group that is utterly missing: countries with unreliable electricity and firms that are unsatisfied with education.  In countries where firms struggle to get even somewhat reliable power, enterprises don’t complain about the quality of education—because without electricity they don’t need a skilled workforce.

Some other results from the survey and analysis presented in the paper:

  • Adding a year of education to average years of schooling at a given income is associated with a 14 percent rise in electricity consumption.
  • A 17 percent increase in human capital is associated with a 17 percent decrease in dirty energy consumption and an 86 percent increase in clean energy consumption from the average share.
  • A one-year rise in average years of education in the adult population at a given income is associated with a drop in electricity output lost in transmission and distribution equal to 0.8 percent of total output.
  • A one-year rise in average years of education in the adult population at a given income is associated with a decline of between 3 and 5 percentage points in the proportion of firms reporting electricity outages.
  • A one-year rise in average years of education in the adult population at a given income is associated with somewhere between a 1.7 to 6.6 percentage point higher rate of return in World Bank-financed infrastructure projects.

It’s surely no surprise that you need a skilled, healthy workforce to make the most out of infrastructure investment just as much as you need quality infrastructure to ensure a skilled, healthy workforce can make the most of its abilities. But it does suggest that partisans of physical and human capital investment need to cooperate as much as they compete.


CGD blog posts reflect the views of the authors, drawing on prior research and experience in their areas of expertise. CGD is a nonpartisan, independent organization and does not take institutional positions.