Applying effective altruism to climate change

If you had $1 million dollars and your goal was to reduce as much CO2 in the atmosphere as possible by 2050, what would you spend it on? Buy a bunch of carbon credits? Protect a piece of the Amazon? Buy everyone in your neighborhood an electric car Oprah style? “YOU get a car, and YOU get a car…”

Effective altruism is a form of charity that asks: how can we use our resources to help others the most? The answers must be based on data about outcomes, not emotions. Applying effective altruism to climate change and the question becomes: how can we use our resources to most reduce climate change? In his book “The Most Good You Can Do”, the father of effective altruism Peter Singer suggests:

If your goal were solely to slow down climate change by reducing greenhouse gas emissions, you could do that more effectively by donating to organizations that are encouraging people to go vegetarian or vegan than by donating to leading carbon-offsetting organizations.

Ok, so donating to a company that encourages people to go vegetarian is a cheaper cost per ton of CO2 avoidance than buying a one ton carbon offset. Interesting… But what about other alternatives like buying solar panels, promoting biking, buying an EV?

Luckily, I’m not the first person to think of this idea. I found an organization called DrawDown.org that estimates the CO2 reduction potential of different technologies as well as the cost of implementing that technology. DrawDown was founded by environmentalist Paul Hawken and looks like it has a team of 8 full-time researchers with science related PhDs. On their solutions leader board they show the top 80 CO2 reducing technologies including predictable technologies like solar and wind to less predictable ones like bamboo, a “substituted for aluminum, concrete, plastic, or steel.” Here’s the top 10 technologies by total CO2 reduction:

If you sum the total CO2 reduction by sector, you find that the highest potential sectors to decarbonize are electricity generation, food, land use, materials, women & girls, buildings/cities, and transport, in that order.

Drawdown.org summary of CO2 potential by sector

Sources of CO2 by sector from 2014 IPCC Report

What’s interesting is that the CO2 reduction potential by sector is not a perfect match with the CO2 emissions by sector. Above are the sources of CO2 by sector as reported by IPCC in 2014. If you overlay the two graphs you’ll notice that transport is significantly under-represented with 4% CO2 reduction potential versus 14% actual global emissions. Food and land use, on the other hand, are 46% of total CO2 reduction potential versus 24% actual global emissions. The takeaway for me is that transportation is going to be a challenging sector to decarbonize, whereas there are more opportunities for food and land use.

What we are still missing is COST. The right question to ask is not how much CO2 reduction potential is there, but rather which sectors offer the best CO2 reduction per dollar spent. The good news is that Drawdown calculated this figure for 55 of the 80 solution types. Unfortunately, for 6 of the top 10 solutions solutions they wrote “ GLOBAL COST AND SAVINGS DATA TOO VARIABLE TO BE DETERMINED”. I understand that academics hold themselves to a certain level of scientific rigor before publishing any estimates. But certainly we should be able to do some napkin math to get an idea of the magnitude differences between solutions: $10 billion? $100 billion? $1 trillion? $100 trillion?

Since I am not held to the same level of scientific rigor, I did napkin math for the missing 25 solution types here (all feedback welcome). The purpose of my calculations was not to appear in a peer reviewed scientific journal, but rather to get a rough estimate of the cost. Here are two examples:

Reduced Food Waste: 35% of food is thrown away in high-income economies. Changing people’s habits is difficult. Another public awareness campaign that had to change people’s habits was anti-smoking. Estimates of the anti-smoking campaign cost per quitter was $480. Therefore, if we multiply $480 by the high-income economy population, we can have a rough estimate of changing food waste behavior via a public awareness campaign. There are 1.22 billion high-income citizens x $480/person = $586 billion.

Plant Rich Diet: In economics there is a concept known as price elasticity of demand which measures how much demand for a product decreases if price increases. Meat, for example, has a price elasticity of 0.7. Therefore, if the price of meat increased 10%, demand would decrease by 7%. To estimate the price increase needed to reduce global meat consumption by 50%, we could calculate what the price increase would have to be: 50% / 0.7 = 71% price increase. The global meat market is about $1 trillion dollars, so reducing meat consumption 50% would cost $710 billion per year.

Once we have the estimates of cost, we can divide the cost by the CO2 reduction potential to get cost per Gigaton (GT) of CO2. When you view the technologies by cost per GT of CO2 avoidance, the picture completely changes. X-axis is the amount of CO2 reduction and y-axis is the cost per GT (both axes on log scale). The green area marks solutions that have high impact and are also cost effective. The yellow areas are either cost effective or high impact. The red box shows solutions that are both expensive and not very impactful. Keep in mind that only cost was considered so any savings/revenue generated by these solutions was not considered. For example, wind and solar would generate electricity revenue.

The results above surprised me in two ways:

1. Land use is the most impactful and cost effective sector with almost all solutions in the green box. Surprisingly, these solutions get little press. I had never even heard of the most cost effective solution of Silvopasture.
2. Transportation, for all its hype, is the sector with the most solutions in the red box. Spending money on high-speed rail, electric bikes, and ridesharing does not appear to be an effective way to combat climate change. EVs are in the yellow box for having high-impact, but are one of the most expensive solutions. These topics are considered sexy in the popular press, but they are shiny objects that distract from more cost-effective solutions.

Silvopasture example

The top 5 solutions for impact and cost effectiveness in the bottom right corner are:

1. Silvopasture ($1/ton): practice of integrating trees, forage, and the grazing of domesticated animals in a mutually beneficial way.
2. Tropical forests($4/ton): restoring 751 million acres of degraded land in the tropics
3. Educating girls and family planning($5/ton): Women with more years of education have fewer and healthier children, and actively manage their reproductive health.
4. Reduced food waste($8/ton): 50 percent of food waste is reduced by 2050.
5. Refrigerant management($10/ton): Every refrigerator and air conditioner has refrigerants that have 1,000 to 9,000 times greater capacity to warm the atmosphere than carbon dioxide. Phasing out use of these refrigerants would avoid leaks when these appliances are disposed.

Green roofs are an expensive way to address climate change

And which solutions are the least cost-effective?

1. Green Roofs ($1,800/ton): Green roofs are roofs covered in plants; cool roofs are roofs that reflect more sunlight back into outer space.
2. Trains ($1,500/ton): Electrification of train fleets from internal combustion.
3. Electric Vehicles($1,300/ton): Compared to gasoline-powered vehicles, emissions drop by 50 percent if an EV’s power comes off the conventional grid. If powered by solar energy, carbon dioxide emissions fall by 95 percent.
4. Biochar($1,000/ton): waste disposal method of burying and burning organic waste which traps CO2 in soil.
5. High-speed rail($700/ton): building new rail infrastructure for high-speed trains, or bullet trains, like the one proposed from San Francisco to LA.

So what does this all mean?

If we compare the most cost-effective solution with the least cost-effective solution we can see that Silvopasture costs $1/ton while green roofs costs $1,800/ton. Therefore, green roofs are 1,800x less cost-effective than Silvopasture.

I think the results of this analysis will be surprising to many who assumed that buying an electric vehicle and advocating for high-speed trains/green roofs are good policies. In fact, these “feel good” solutions are actually a drain on public resources that could be used towards more cost-effective but less obvious solutions. The size and scope of addressing climate change is so massive that all resources must be put to their most efficient use for us to have a chance of limiting global warming to below 1.5 degrees Celsius.

Returning to the central question of this post, “If you had $1 million dollars and your goal was to reduce as much CO2 in the atmosphere as possible by 2050, what would you spend it on?” Based on the preliminary results presented here, you should pick one of the top 5 solutions listed above, or any of the solutions in the green box above.

Good luck and I’d love to hear your comments on applying effective altruism to climate change.

Part II: Top charities for climate change