[MUSIC] Hi, welcome back to the course of nexus and complexity. We are starting the second week, and this week will be dedicated to an analysis of the challenges that quantitative analysis faces because of the complexity of the work. This is the first lesson and it will provide examples of the problems you get when you try to simplify too much. So there are three sessions, the first session provides you with examples of bad indicators. The second session is about the fragility, the intrinsic fragility of numbers. And the last one, Session C, is dedicated to the handling of the issue of scale. So let's go with the first session, Session A is about giving examples of bad indicators. In this session, we will try to show you that at the moment, a lot of the indicator that we trust, that are used by the establishment body, Serious societies, in reality, are not that good. And then for this reason, we should be aware that if we want to do better, we have to try to understand what is wrong at the moment with the use of numbers. So we tend to use, to trust numbers because you say, if they're putting numbers, they must know what they are doing. Unfortunately this assumption is not always true. So I will give you a few examples of popular indicators to show that as a matter of fact, these indicators are quite fragile. Let's start with a classic one, food consumption per capita. So this is an indicator that is used to compare the status of the diet, of the, Energy nutrient intake in the diet. And this is the standard narrative, the mantra that goes, no? In poor countries, they are supposed to have a very low calorie intake, because they have 2,200 kilocalories a day. Whereas in the rich countries, we have a very high level because we have 3500 kilocalories per person per day. And this is, I took this number from a document of the FAO, so there's no doubt that this is the state of the art of the understanding of the indicator. Let's see what happens if we're looking at the population structure of developing countries. So these are the number of, out of 100 people, the number of kids below, very, very young people, below five years, then you have 30 youngsters, then you have 20 adults, and 10 elderly. If you try to see why you get, what are the amount of kilos such a population is expressing per capita, you have to have 3,500 kilos, 100 people with this structure population. That translates into having 31 kilos average person. And then we are looking at 2,200 kilocalories per day, you see that in this society they are eating 71 kilocalories per kilogram of mass. That is what is relevant for calculating the calories getting into the metabolic part of the body. So let's look at the population structure of a developed country. And we will have very little, very few person below five years, a lot of adults here and a growing quantity of elderly. Then if you try to look at the structure of population in relation to the weight, you will see that 100 people in the developed country weigh much more than in the developing countries. This means 50.7 kilos as average. Then if you are looking at the quantity of kilocalories per kilo of body mass, you will get the 62 kilocalories per day. So let's now check this indicator. In developed countries the average person weighed 50 kilos, in developing countries the average person weighed 30 kilos. In developed countries they eat 3500 kilocalories per day, in developing countries 2200 kilocalories per day. How does it look if we are looking at the kilocalories per kilo of biomass, I mean a kilo of body mas of humans? You would see that in developing countries they are eating more kilocalories per kilo. This is normal because there are more more young people out there, and the young people have a higher metabolism than the old people, and also they do much more physical exercise. So what is the point here? That we are making a lot of analysis about inequity on the distribution of food based on an indicator that is completely false. It is not true that developed countries consume much more than developing countries. Of course they consume much more. In terms of animal protein, three times more. In terms of the calories required to produce the food, to consume the food, is probably ten times more. But I mean, the point is that the indicator that is used now to make the calculation is not particularly good. Let's give other examples, another example, this is very, very popular. Everybody knows the the GDP per capita, the gross domestic product per capita, and then this is used to see whether an economy is doing better than another. Again, let's have a comparison, let's imagine the structural population of Italy and China. You can see that in Italy, because of the very large part of elderly population, there is a small share of the population which is in the workforce. For those who are not familiar with this, this is a structure of the demographic structure of a population. So these are men and women, and these are elderly, and this is people below five or ten. And then these blocks tells you who are the age class, the young and the working class, over 15 and below 65. And if you go above 65, then these people are retirees who are dependent. If you are making a comparison, China, because of the One Child Policy, is at the moment in a very, very peculiar situation, in the sense that they have a lot of adults in the workforce. For this reason, if you are comparing, the economic active population in Italy is 40%, in China it's 60. And then, if you are looking at the workload per year, not only in Italy, there are less people working, but the working people work for less work than in China. These are all data, I think these are in the 19, I don't remember, but I mean basically, now in China, they are working less. They would be 2,000, a few hundred, 200 or 300. But I mean, the difference with Italy is still significant. What is the problem that if you are then reconsidering how many hours of work you have in the economy per capita per year, you will have in Italy 608 hours of work per capita, whereas in China, you have 1600. So of course you cannot beat this economy, because they have an enormous amount of supply of working hours, compared with European countries. So what would happen is that because of this difference, you have that in Italy, you have 13 hours of consuming for one hour of producing, whereas in China you have 5 hours consuming per one hour of producing. If we do this comparison, this difference and we try to interpret this in terms of economic productivity, again, these are old data and probably this is in the 90s or something, whatever, doesn't matter. You have a GDP for Italy, a GDP for China, but the GDP per se doesn't tell you how well the economy does because how much, how productive is the labor. Because in reality, what you should look at is the amount of GDP produced per hour of labor. And then in this case, the difference is much more than the difference in GDP because since Italy has a much lower amount of working hours and higher GDP, they what is called economic job productivity. That is, the amount of gross added value of the GDP produced per hour of labor is much higher than the other. Why this is important? Because if you're looking at the structure of the population in movements, let's imagine this is what happened in time. You can see that the change in the demographic structural implies change in parameters that are completely very, very relevant for economic reasons. The ratio would be who is paying the taxes to pay for the pensions, and the dependency ratio not only implies that you have less workers, but you have more services to provide, because you have more people to take care. So what is the point here? That if you are making a comparison on historical series on how the economy did between 1990 and 2015, in reality, we're not considering that the economies are not the same, because of the different structure of the population. Another important problem is generated, you can see now, these are two population structure of Shanghai in 2000 and 2010. Another big problem is generated by immigration. Because the very very high level of adults in Shanghai is not only due to the structure of the One Child Policy in China, but it's rather due to the fact that they had a quick migration into the city, and of course immigrators are only adults at the beginning. So what happened is that immigration worsened the situation in the sense to doing an analysis of the country, like a black box per capita, really implies that you cannot understand how the economy works. Finally, there is another example, one of the energy intensity of the economy, that is an indicator that is often used to compare the economies of different countries. And this is based on the amount of energy consumed per dollar. So you have the Finland has 12.7 megajoule per dollars. And you can get another country, let's have El Salvador. You will find that El Salvador has again 12.7 megajoule per dollar. Why is that? How is this possible, that there are completely different countries that have the same values of energy density? In reality this is very easy to explain because the energy density, which is energy consumed by the country, and the GDP of the country, in reality, comes from ratio of two ratios. The energy per person or energy per hour or whatever, and the GDP per person. Now it is obvious that if you are consuming more energy to produce and consume goods, you will have a higher GDP. And then if you are consuming less energy to produce and goods and services, you will have a lower GDP. So if you are dividing a high quantity of energy per hour by a high GDP per hour, and a low energy per hour and GDP per hour, you'll have the same ratio. So what does it mean? If you're making a ratio of two variables that are correlated, their ratio is what? No it doesn't have any meaning, doesn't have an external referent, is not reflecting the existence of any mechanism that is consistent. What would happen if we would, rather than using a ratio, using a plane? So in this case you have energy per hour on the left and GDP per hour on the right. You would see that El Salvador remained, in the period of 1998 to 2004, exactly in the same position, it didn't even work. So, using a plane you can see that El Salvador is using less energy, making less GDP, and is not changing in time. Whereas if you are looking at Finland, you can see that it uses much more energy making more GDP, but also that the economy is moving in time from 1998 to 2004. We can see that European countries are moving in the diagonal and I'll generalize is this correlation between energy per hour and added value per hour. Let's see all the countries in the world, so this is the energy intensity of all countries in the world. There is on the right countries that are very high energy intensity, but this is due either because they are partially market economies that are There's a lot of energy is used in self-sufficient activity, so they are using energy, not generating the equivalent of the value. Or because they are oil producers, so oil producers consume a lot of energy to extract and refine oil that they are not using, so they are then exporting, so in a way they are a special case of society. But let's focus on the other group of countries. They more or less have a similar quantity of value for energy intensity. So we have three cluster, let's see the cluster. One cluster of those have a low energy intensity. So this would be Guatemala, Germany, the Netherlands, Angola, Norway, Chile. So you can see that the correlation is very, very high. But the countries are completely different from each other, in the sense this is not an indicator that makes it possible to understand how countries are different and why. There's also Japan, Brazil. Look at the cluster of medium energy intensity. Still, you would have Sweden, Macedonia, Azerbaijan, France, Argentina, whatever. In the sense, again you have a situation in which the, per se, the value of energy intensity doesn't map in any typology of country. And finally, if we go to our very high energy intensity, you find Thailand, Australia, United States, Malaysia, and Turkey again. So what is the lesson here? That in reality this indicator, like the other we saw before, is not particularly useful unless put this way,
