Social disparity image via shutterstock. Reproduced at Resilience.org with permission.
Recently, in this blog, Jacopo Simonetta raised a very important question: Would a fairer distribution of income worldwide diminish the damage humans are doing to the earth? His answer, that it would not and would actually make matters much worse, intrigued me. So, I decided to look at the best available data.
Simonetta specifically looked at the question of whether a fairer distribution of income would reduce global CO2 emissions. In 2015, Lucas Chancel and Thomas Piketty (henceforth C-P) wrote a paper and posted online a related dataset that dealt with the global distribution of household consumption and CO2e (carbon dioxide equivalent = CO2 and other greenhouse gases) emissions in 2013. The data are not perfect, but they are the best that exist. The C-P dataset captures the Household Final Consumption Expenditures (HFCE) values provided by the World Bank, using the distribution of income in Branko Milanovic’s dataset (for the bottom 99 percent) and in the World Wealth and Income Database (for the top 1 percent). (Income is not the same as consumption, and C-P assume that the distribution of income is the same as that of consumption. Also, they assume that the same distribution of income that existed in 2008 also existed in 2013. Like I said, the dataset is not perfect).
The C-P dataset includes 94 countries, which cover 87.2 percent of the earth’s population, about 6.2 billion people, who are responsible for 88.1 percent of global CO2e emissions. Generally speaking, C-P divide each country in “11 synthetic individual observations (one for each of the bottom nine deciles, one for fractile P90-99, and one for the top 1%).”
The following chart shows consumption per capita and CO2e emissions per capita in 2013 from the C-P dataset.
Figure 1. Consumption and CO2e emissions per capita by world consumption percentile in 2013. (Some percentiles are missing due to the fact that the country quantiles vary in size and sometimes extend beyond a given global percentile.) (Source: own elaboration from data of Chancel and Piketty (2015).)
The top 1 percent on the consumption scale spend an average $135,000 (2014 PPP dollars) and emit an average 72 tCO2e per person per year. The threshold for belonging to the top percentile is $54,000. Their consumption is equal to 18 percent of all the money spent by households around the world. Let’s assume, for the sake of argument, that consumption equals income. If one were to take all the income of the top 1 percent and distribute it among the 99 percent, each person in the 99 percent would get about $1,400.
C-P assume a CO2e emissions to consumption spending elasticity of 0.9. A 10 percent increase in consumption means a 9 percent increase in CO2e emissions. This is a broad generalization, and C-P have a range of elasticities, but they chose that one because it is the median value of the estimates. Using that elasticity in the C-P dataset, if each person in the bottom 99 percent got $1,400 and those in the top 1 percent were left with nothing, global CO2e emissions would increase by 9 percent.
But, of course, the top 1 percent are only part of the problem. About 22 percent of the world’s population lives with a consumption level above the global mean of about $8,000 per year. Let’s assume that everyone had a level of consumption equal to the mean. Going back to the C-P dataset, if one averages the CO2e emissions of everyone within a consumption bracket ranging from $7,700 to $8,300, one gets an average emission of 6.15 tCO2e per person per year. If everyone had that kind of emission, global CO2e emissions would be practically the same they are today, but, needless to say, that would improve the lot of more than three-quarters of the world’s population.
In short, a perfect distribution of income would have a negligible effect on global CO2e emissions.
There remains the question: At what level of consumption would CO2e emissions be reduced dramatically and would this level be compatible with a decent existence?
Cuba offers an interesting example. Moran et al. (2008) looked at the UN’s Human Development Index (HDI) and the Ecological Footprint of 93 countries for 2003, and worked on the assumption “that an HDI of no less than 0.8 and a per capita Ecological Footprint less than the globally available biocapacity per person [one planet earth] represent minimum requirements for sustainable development that is globally replicable.” Their survey showed that only one country met both of those requirements, Cuba.
Cuba also has the second lowest fertility rate of the Americas, 1.61 births per woman. Only Canada’s is lower. This means that a low-consumption society can be compatible with no population growth. The average Cuban eats 3,277 calories a day. Cubans have a life expectancy at birth of 79.4 years. This is above the United States and only 1.5 years below Germany. And Cuba’s mean years of schooling are above Finland’s. And only Monaco and Qatar have more doctors per capita than Cuba. Clearly, a level of consumption compatible with the finite planet that we have does not have to equal penury and destitution for everyone. I’m not saying life in Cuba is easy for everyone. It isn’t, but at some point in the near future, those who live in the developed world and in the rich enclaves of the developing world are going to be faced with a choice between a Cuban lifestyle and, to quote Noam Chomsky, the destruction of “the prospects for decent existence, and much of life.”
I wanted to find out if the findings of Moran and colleagues were still true today, but I made one change. The HDI is built using three dimensions: life expectancy, education, and per capita income. This has always bothered me. A long, healthy life and an educated population are no doubt hallmarks of human development. But, is driving a Lexus a sign of human development? I think not. Therefore, I used the UN’s data to build an index that looks only at life expectancy and education, which I’m calling the truncated human development index (THDI). (The calculation of the HDI is explained here. The THDI follows the same procedure used from 2010 onward, but it only takes the geometric mean of the first two variables.) In the following chart, I plot the THDI versus the Ecological Footprint, measured in the number of planet earths the inhabitants of a given country consume, using the most recent data.
Figure 2. THDI and Ecological Footprint of 176 countries. The red dot represents Cuba. (The THDI corresponds to 2014, the Ecological Footprint to 2012.) (Source: own elaboration from data of the UN and Global Footprint Network.)
There are only two countries in the vicinity of one earth that have an THDI higher than 0.8, Georgia and Cuba, the red dot. Of the two, Cuba has the highest THDI. It’s interesting that Cuba has practically the same THDI as Chile, but Chile uses 2.5 earths. And it has practically the same THDI as Lithuania, but Lithuania uses 3.4 earths. Furthermore, Cuba uses as many earths as Papua New Guinea, but Papua New Guineans have an average of 4 years of schooling, Cubans 11.5. This is just to show the possibilities that exist for an egalitarian, sustainable society. As of late, inequality in Cuba has been on the rise. However, according to the World Bank, CO2e emissions per capita in Cuba are not substantially different today than they were in 1986, when Cuba’s Gini coefficient was very low, 0.22 (Mesa-Lago 2005, page 184). In any case, I’m not advocating that we copy the Cuban model completely. I’m not defending Cuba’s crackdown on individual liberties, freedom of speech among them. All I’m saying is Cuba is an interesting example of the possibilities that an egalitarian society offers. I, for one, would like to live in a society that is even more egalitarian than Cuba. It seems to me that there is no reason in principle why humans cannot build a society that is more egalitarian than Cuba and just as sustainable, especially when the alternatives are dire.
Cuba is not in the C-P dataset. It is hard to estimate the level of consumption of Cubans in dollars, because the statistics the Cuban government publishes are not comparable with those of the rest of the world, but last year the UN published a GNI per capita number for Cuba for 2014 that seems to be solid and comparable to other countries, 2011 PPP $7,301. That number is not directly comparable to the C-P data, because C-P looked at household consumption. Assuming that the share of GNI for household consumption published by Cuba’s National Statistics Officeis correct, one can estimate household consumption per capita in Cuba to be at around 2011 PPP $3,900. It’s hard to translate that to 2014 dollars, because I don’t trust the PPP conversion factor published by the World Bank, but let’s assume that the consumption of the average Cuban is around 2014 PPP $4,000.
Going back to the C-P data, one can find that the average CO2e emission for a consumption bracket ranging from $3,700 to $4,300 is 3.14 tCO2e per person per year. If everyone in the world had that level of emissions, global CO2e emissions would be cut by half. And in a social system similar, but not identical, to Cuba’s, no one would starve or be unschooled, and the lot of 61 percent of the world’s population would improve.
To recap, an equal level of consumption for everyone around the world at the level of today’s Cuba offers the possibility of substantially lowering human impact on the biosphere while at the same time maintaining a rather decent standard of living for all.
According to the Global Carbon Project, in 2014, “the ocean and land carbon sinks respectively removed 27% and 37% of total CO2 (fossil fuel and land use change), leaving 36% of emissions in the atmosphere.” If CO2 emissions were cut by half, all of them would be removed by the earth’s sinks, and there would be no net addition of CO2 to the atmosphere.
It is worth pointing out that global mean consumption will reach the level of today’s Cuba eventually. The question is will that happen before humans increase the global temperature to dangerous levels. Cubans today consume 6 barrels of oil equivalent per person per year of fossil fuels, which is what Laherrère (2015, page 20) forecasts humans will consume around 2075, after the peaks of oil, natural gas, and coal production. But, by that time, according to Laherrère’s forecast (2015, page 22), humans would have emitted about 2,000 GtCO2 since 2015, 800 GtCO2 more than the maximum Rogelj et al. (2016) estimate we can emit to have a good chance of avoiding the 2 °C threshold. (Laherrère is skeptical about anthropogenic climate change, but I’m not endorsing his conclusions, just looking at his data.)
