What is the Net Impact Ratio (nIR) of educating a girl in India?

So why does Bill Gates focus on the developing world? He gets it. The impact ratio of changing education, or vaccines, or other small changes in the developing world are just huge. I am not suggesting that we only invest in social changes to save the environment, but we should not discount them either. We can get far more bang for the buck by changing people than by changing systems, but I think the optimal is changing both at the same time. 

If I spend $1,000 on primary school education for a girl in India, just between increased income and a reduction in carbon footprint, I can expect $7,120 back in value for an nIR score of 7.12. Just the value of the C02 decrease of investing in that education can be priced at $7.56 per tonne, which is a fraction of the cost of solar cells or most other green investments. 

Education fascinates me. I am going back to school in January, writing a Ph.D. in environmental economics around water use in copper. So I worked with a friend and former co-worker named Oshi on this; she is Indian. We came up with a list of impacts of education in India. 

More income: Family income goes up significantly if a woman in India has an education. This is the case in almost all outcomes, except if the women are rural, and in the informal economy, the data shows that education hurts rural women who don’t break into the formal economy. 

• This is Present Value of the earnings (modeled as attached)
• There is a carbon footprint of increased income (modeled)
• The economic multiplier of female-controlled money (not modeled)

Women, when they have their own money, tend to spend it very differently than men. My mom always said that if you want to empower society, women make of some wealth. I think there is a serious economic multiplier of female-controlled money versus male-controlled money, but I did not take the time to find papers on this topic. 

Fewer kids: There is a strong correlation between more education and fewer kids, and the impact of fewer kids can be priced in a few ways. I am going very simple with my model. 

• Reduced carbon footprint (I modeled this)
• More education per kid (fewer kids = more resources per kid = compounding effect of education) (I did not model this.) 
• More food/other resources per kid (if you have more resources you will grow more, be healthier, and be more productive)  (not modeled)

I come from a long 100+-year line of educated women. In 1908, my paternal great-grandmother got a Bachelor’s degree in Sweden. On my mother’s side, we have a history of advanced education measured in the 100’s of years. College for my family for generations was assumed, and I get and respect that privilege. 

What other impacts?

Life expectancy: Life expectancy goes up, as does the quality of life (not modeled)

Less abuse/child marriage: There is data on both of these points. Every year of added education for girls really shifts the child marriage issue, and the economic benefits of getting the last two years of secondary school get a significant bump from this (not modeled)

Rural vs. Urban: Where you live and how you work matters, and I ended up with a matrix of scores, Rural vs. Urban and Casual vs. Formal work  (modeled)

Cost of education: I got good primary school data, but I failed to obtain data for secondary or graduate (just way too much noise in the model I could find, so I put in a multiplier). This is the data point that needs to be fixed on the model.

Discount rate: I used a Western discount rate; I modeled this from the POV of me as an American deciding to fund a girl at school in India. The average Indian does not have the same discount rate (they pay more for money), but they also don’t have the same investment horizons.