 Welcome to the second set of five lectures. And I'm going to try to talk about, I want to finish up a little stuff I ran out of time last time, and then talk about the theory of consumer behavior from a different point of view, from what I call real consumer behavior rather than perfect behavior. And then begin the theory of a firm where the key thing is the theory of real competition. Now, as I said last time, if I could erase from your mind the stuff you've learned would be very handy. Because I have no trouble explaining this to people in literature. I spoke at the Lillehammer Literature Festival in Norway. And they said, oh, that's pretty obvious. But in an economics department, I have to undo, undo things. So I ask you to think about these things as starting a different theoretical framework. I'm a big fan of the consistency of theoretical frameworks. I admire neoclassical theory because it has a general framework. I like post-keying theory because it has a general framework. I'm not so happy about the idea of mixing and matching. That is to say, this seems like a good conclusion. I like that I'll take it from that theory and then take a conclusion from another theory. And I realize that many people are comfortable with that, but I'm not. I started my teaching by accident in Kuwait. I was working in the desert. And I went blind, Desert Blindness, it's called. And when I recovered, I could not only see. I could see that working in the desert was not a good thing for me to do. So I happened by accident to get a job teaching high school at the Kuwait American High School. And I taught physics, math, and social science because they didn't have anyone teaching those subjects. And so I said, sure. I was 19 years old and had finished college. And had never taught before. And I discovered I really loved it. And among the things I was teaching in my physics class was about the analogy between physics and biology. Biology is also a general framework if you're coming from Darwin. And my students said to me, one of my students, or some of them said to me, but you say that biology is explained by evolution, but miss, I've forgotten her name. She was an Indian woman, actually, who studied in University of Iowa. She was a Christian. And she says that evolution is all wrong. That in fact, God created the Earth 5,600 some years ago. And you have the story of Adam and Eve. And so I went to this woman who was a biologist, by the way, trained biologist. I said, how could you do that? And she said, well, that's what I believe. And so I don't teach Darwin. Now, in some sense, you could be, and probably good for your career if you were in Texas, to do a little bit of evolution and a little bit of creation depending on who you're talking to and stuff like that. And maybe that's fine for politics. But I don't think theoretically you can be part creationism and part evolution, because they're really inconsistent with each other, despite the fact that people, oh, God created the universe and all the stars are destroying each other. Maybe. I doubt it. But anyway, that's not the point. The point is you can't explain evolution that way, except by making it up. And so there's a real conflict between the theories. And that's good, because the purpose of a theory is to tell you not only what is likely to happen, but what is not likely to happen. A theory is a device to explain things in a way that it can also have negative explanations. So I'm going to come to this later, but just to give you an idea. If you are a post-Kanginian economist, well, if you're a neoclassical economist, and you were to say, we should fight to raise wages, what would a neoclassical economist say? What would your microtech book say if you intervene in the market and you raise wages? What would happen? Unemployment, right? It's pretty clear. The market brings you to the real wage that gives you full employment. That's a very important part of the theory. It's really fundamental. So therefore, interference by raising the wage beyond what the market would do will create unemployment. Now, I'd like that about the theory. I haven't done it against wrong, but what I like about the theory is that it says what's not possible or what's not sustainable. That's a better way to put it. What's not sustainable? If you are a post-Kanginian economist and you were to say raise wages, what would be the answer? What would be the conclusion? What are the implications of a rise in wages? Almost every post-Kanginian textbook tells you. If you raise wages, what would happen? Exactly, higher aggregate demand. Why? Because wages go up, consumption goes up. Consumption will therefore create a higher output, higher aggregate demand, and that would cause employment to rise. So these are two different points of view. I like that. That's the great beauty of theory. It really gives you the conflict. There's some things of which they don't disagree, but there are some things that they do. So here's one that's clear. If you were to push for raising the wages of workers by 20%, then neoclassical would say you get unemployment and the post-Kanginians would get more employment. So one says the unemployment rate will go up, the other go down, right? That's a point. As an economist, you're responsible for that kind of conclusion. And so if you like the idea of higher wages, then you have to be responsible for the consequences. Now if you're post-Kanginian, there aren't any consequences, negative consequences there, or at least the limits are pretty wide. I mean, Keynes, I think, wouldn't say this, but post-Kanginians would. But I would argue that in fact there is a third way, and I can't use the word third way because Tony Blair irretrievably corrupted that word, so you have to think of something else. But there's another way. And that is to understand that in a different framework, limits operate. And I'm gonna show you this today. I'm gonna show you at the end of the graph, that's where I'm starting this. If you pump up the economy, then let's say by deficit spending, which as any Keynesian would tell you, one of the ways to pump up the economy, stimulate. You create a rise in aggregate demand, and the aggregate demand rise will create an employment rise, right? So unemployment should go down. If its stimulus is sufficient, unemployment should go down. If unemployment goes down, then from a classical point of view, or what I'm calling classical Keynesian political economy, you will enter the domain where the wage share might rise. Because if unemployment goes down, wages will begin to rise. Workers will be in a stronger bargaining position. If wages rise relative to productivity, then the profit rate falls. Because the profit share, the wage rises relative to productivity, the wage share goes up. The profit share is just a dual of the wage share. So the profit share falls, profit rate falls. And at some point, that is likely to inhibit growth because growth is driven by investment. And investment depends on the profit rate. So if the profit rate is falling, you're going to slow down the growth rate. The result is that if you transcend those boundaries, then you're likely to have a slowdown in growth, which would undermine the whole purpose of the stimulus, which is to increase growth, right? Now I'm gonna come back to this at the end of the fourth lecture, I think, but this is not an abstract problem. We can see episodes where there were massive stimulus, and that was, first of all, Hitler. Hitler, in the 1930s, created what is often called the first instance of military Cajunism, a massive stimulus to the economy, deficit finance, money printed and spent, and unemployment was eliminated in one year. And Hitler is on record saying that if you remember me for nothing else, remember me for this, I eliminated the problem of unemployment. Problems that arose from the wars and all didn't come from the consequences. And I'll explain why it didn't occur in Hitler's case. World War II, the United States, great depression, massive unemployment, misery, political turmoil, Communist Party with a million people marching in the streets, and by 1939, when the war effort is beginning and the economy is being pumped up, in the war, unemployment disappears. They have not enough people, so much so that they start inviting women to work. I mean, talk about really scraping the bottom of the barrel, that women were not allowed to work, and now suddenly going, Rosie the Riveter. I don't know if you've ever seen those images of a woman, you know, arm, and muscle saying, and why was that? It was to encourage women to thinking that work was good, as after 30 years of preventing them, and in fact, not even allowing them to vote, suddenly they needed them, and black people suddenly, oh, come on in, black people couldn't get in them before, and now suddenly they needed them. And yet that did not have negative consequences. They had, in fact, a tremendous increase in growth, tremendous increase in employment, and I'm gonna come back to why I didn't have. In the 1970s, so after World War II, everybody was a Keynesian. I mean, even Nixon famously said, we're all Keynesians now, right? And why? Because you showed here, in two instances, that you could eliminate unemployment and keep the economy at a high level of employment and growth through deficit spending. But in the 1960s, there was a modest goal that every government inherited, which was to make employment socially at socially desirable level. They didn't say we make unemployment zero, frictional unemployment, maybe three or 4%, that was the goal. And they were going to use Keynesian stimulus policies if they needed it to bring the unemployment down to three or 4%. So if we went above that, they would bring it down. Well, after World War II, unemployment began to reappear and following the theory of Keynesian economics, they stimulated the economy, the unemployment went back down, so that seemed to work, but then it started coming back up. So they stimulated it more, it went back down, and then it came back up. And every time they stimulated it, they got inflation. And now they were faced with a theoretical impossibility from their point of view, which is they were getting unemployment rising and inflation rising. But in Keynesian theory, you can't do that because in Keynesian theory, inflation, if you stimulate an economy and you are at unemployment, then the out-employment level rises until you reach full employment. And then after that, any more stimulus will give you inflation. So you can't have unemployment and inflation. And you certainly have unemployment rising, the employment rate falling, and inflation at the same time that's theoretically impossible. But that's what was happening in every advanced capitalist country in the world. Most recently, in the Lula government in Brazil, the leading economists there were from my university, Nelson Barbosa was a student of Lance Taylor's, and they had the same premise that you could stimulate the economy. And by stimulating the economy, you could pump it up, and they did pump it up in two successive governments. And then the growth began to fall, profitability fell, unemployment came back up, and in fact, they lost power actually. They lost an election because of the problems that were piling up. Now, one argument could be less external due to oil price changes, but the same thing happened in Argentina. If you look at the episodes where it worked, and the episodes didn't, I would argue that you see a dividing line. And that's why I'm gonna aim at these graphs here in a minute. The dividing line is that the episodes that worked were able to keep wages from not rising faster than productivity. Hitler had a very clear view, which is that the state controlled all of these things. So the state controlled wages. It also controlled inflation, controlled prices. I mean, you didn't wanna get into Hitler's bad side by raising your prices because there was scarcity in the thing. So you could, it also insisted that everybody work really hard and increase productivity. So length, intensity, the working day, as well as innovations of all sort were not just encouraged. If you didn't do it, you could be in big trouble. And what happened with the profit rate in Hitler's great stimulus was that the profit rate rose. So you had this big rise in employment, rise in the real wage, but productivity rising much faster and the profit rate rising. The very same thing happened in the United States in World War II. They controlled wages. They controlled prices. They insisted that everybody work as hard as they could until they dropped because this was the war effort, this patriotic. And they encouraged innovation of all sort because in a war you need the technology to change rapidly, to develop new weapons and all of that. So the same thing. So those two instances that worked were instances in which the real wage increase did not go beyond productivity increase. In the post-war period, it was the opposite. Real wages were rising relative to productivity or the wage share was rising. And that meant the profit rate was falling and as the profit rate was falling, the growth rate coming from private capital was being undermined. So we have the public stimulus and the private incentive working against each other. And that led to a slowdown in growth. It's famously called that episode in the 19th from 69 to about 1977, 1978. It's called stagflation. Why? Because the growth rate was going down towards zero. That's stagnation. And yet inflation was rising, which is inflation. And the Keynesians lost that theoretical battle to Friedman. I mentioned this last time. Friedman said, well, there's no unemployment. I don't know what you guys are talking about. Those people that are working is because they choose not to work. So therefore actually there's full employment and every time you stimulate the economy, get inflation. Bingo, that was Friedman's Nobel Prize. Now I'm gonna argue that was wrong also, but what was right about it was a recognition that theory has consequences, has implications and you're responsible for the implications. So you can't just adopt a theory for the part you like and then not be responsible for the part you don't like. They are tied together. Now theory doesn't tell you everything. It tells you limits. That's the important thing. It tells you consequences, but how they play out is not necessarily given at an abstract level. So I'm gonna come to this shortly when I talk about the data. I wanna finish my data analysis. So that kind of thing I'm talking about is in chapter 13 of my book and chapter 14, which is these two chapters here on macro and there are lots of other things, stock prices, bond prices, and I'm gonna just come to where we were last time. I think I did that, that, that. Yes, okay. So this is another view of the price level in the United States from 1774 to 2019, for 2000 and something, 2010. I think. And you can see what is striking. Do I have to turn this thing on? Okay. What is striking is you can see these waves. You see them? These are the long waves of Kondratyev. And yet after, and here's the wave again, and then after the post-war period, instead of it coming back down, you see this rise in the price index and that's what we call inflation. That's modern inflation. So one of the theoretical questions you have to have explained is why. And it's not a standard one on the left is monopoly power. Well, there's lots of monopoly power actually here. Hillfitting and all the people, Lenin talk about monopoly power. They're not talking about 1940. They're talking about 1870s, 1850s. So if it was monopoly power, you'd expect inflation to be there 100 years before, but it wasn't. So something, we have to understand that. We have to confront it. And there are different ways of explaining it. I'm gonna come to one later on, but I wanna show you the standard way that used to be in place in the post-war period. Do people know what a Phillips Curve is? Okay, so Phillips Curve, when Keynesians were struggling with this issue of policy, one question is if you stimulate the economy, will you get inflation? Now in Keynesian theory, if you stimulate the economy and there's unemployment, the economy will rise until it reaches a full employment level. And if you stimulate it after that, the real economy can't go anymore, so you get inflation. So output increase until full employment, that's your 45 degree Keynesian curve, and then you get a rise in, any rise after that is nominal, not real. Is that familiar or should I do a diagram, not familiar? It is familiar, okay. All right, so Keynesian theory has a clear explanation. Unemployment comes about at full employment. Now everybody knows full employment is a little hard to define. I mean, how do you know what the people are? So the argument is transformed into a different argument, which is that as you approach full employment, by easy unemployment rate gets lower, you're likely to see inflation. So you should get a negative relationship. The Phillips curve looked like this. You have unemployment, and here you have the rate of change of prices, or rate of inflation. And the Phillips curve should look something like that, so that if you go towards lower unemployment, you should get higher inflation. And this was called the key question of macroeconomic policy in the 1970s was, how much inflation are you willing to tolerate to get to some level of unemployment? So the Phillips curve cut the axis somewhere, I don't remember exactly, but like 7%, and that was considered socially too high. So you wanna push it this direction, maybe to 4%, and 4% would imply some kind of inflation of around, as it turned out, about 4%. That was considered socially tolerable, because after all, you could create effective full employment here at 4%. Most of the people, the people who were moving from one job to another, or didn't really wanna work or whatever, so that was called frictional unemployment. And so you can see that if you are working in macro policy and macro theory, that was the curve. And that curve, when you plot it, as I show in my book, when you plot it from 1970, from 1947 to about 1973, you actually get a curve like this in the data. But the trouble is that by 1970s, the data no longer followed a curve. And in fact, if you look at this curve, which is the curve for the whole time period, inflation rate, unemployment rate, there is no pattern. If anything, the pattern says we're going the wrong way. And this was something that they could not explain. They could not explain. The Keynesians. There were many attempts to explain it. Expectation augmented Phillips curves. Every young, hot shot macro connoisseurs looking for an explanation that would make their reputation. But the fact is, there was no Phillips curve. One idea was, well, there is a Phillips curve, but it keeps shifting due to expectations, and you could look for other shift factors. Every time you invented a shift factor, the damn curve went somewhere else. And eventually, the Phillips curve fell out of favor. Some people are trying to bring it back. But then you can ask yourself, well, what's the question of the Phillips curve? What did Phillips actually do? It wasn't this. Phillips asked a different question, which was, what is the effect of unemployment on wages, on wage growth? Now, that is a more sensible question in this following sense, that you can see that unemployment would make it harder for workers to have, keep their wage level growing. And you'd expect that if unemployment went that way, if this was not in price growth, but a wage growth, you'd expect the curve to be downward, right? And the Phillips curve was originally based on wages, not inflation. So then, because the Keynesian policymakers needed a connection to inflation, they said, well, let's assume that prices are proportional to wages. This is a standard thing in post-Keynesian theory. Prices are costs plus a markup. A markup is fixed. So it depends on costs, but costs, what are costs? The costs are materials costs and wage costs. But materials costs are simply the cost of a bunch of goods, which is their materials and their wages plus a markup. And so on and so on. So if you add up all the materials, you keep decomposing into wages, you get just direct and indirect wages, and you get direct and indirect markup. If that markup is stable, then you have an explanation that prices depend on wages. And that gives you a Phillips curve. The trouble is that that construction disappeared. But you could ask Phillips question in a different way. You can say, what is the question, what is the effect of unemployment on wages, but then you have to ask, which wages? And after the way I started, you can see that there are three types of wages that you could ask that question of. I could ask, what is the relationship between unemployment and, I could say, money wages? So I would then expect to see some kind of curve like this. Or maybe, I mean, Phillips found that, by the way. Phillips does it. You should read the original Phillips article. It's quite amazing because he doesn't have the econometric devices we do now, so he has to figure out how to take cycles out of the data, and he invents a kind of technique and all that, and you can do it in a different way now. But basically, he found this downward-sloping curve which said that as unemployment decreases, the labor market becomes tighter, the rate of increase of money wages rises. Now, that's not hard to understand because that's something you can observe many, many times in history. But it's interesting that if you were to look at time discussion in Marx, for instance, or in Engels, or in that time, which these issues were discussed, then the question was not money wages, but real wages. Because after all, workers struggle for real wages, but there was no real difference between the struggle for real wages and the struggle for money wages for the simple reason that prices had no secular trend. I mean, look at this graph. This is the United States from 1774 to 1939, and there's really not a change in the price level. So it's not surprising that the relation between real wages and money wages was close because prices weren't changing very much. Real wages, after all, the rate of change of money wages minus inflation. If inflation is stationary, it's close to zero, then if you struggle to raise your money wage, you also are essentially working on your real wage. Is that point clear? But one particular emphasis, and you can see this in Ricardo and much more in Marx, is that the struggle is really not about wages, but about the share of value added. This is the rate of surplus value I did. The workers struggle for the division of value added. And so that's equivalent to saying the struggle is about the wage share. Now what is the wage share? The wage share is equal to the money wage divided by the price level, and that's the real wage, over productivity, which is output per worker. So you have a wage share struggle is equivalent to a struggle of wages relative to productivity. Now, you know, why is that relevant? Because this ratio at the level of the firm is your unit labor costs. And firms have a strong incentive to, if they have to guarantee your wage increase to make sure that productivity is also tied to it, because if they increase wages faster than productivity, their costs go up and they lose in competition against other firms or other countries and so on. So it's a micro founded, micro vested incentive for them to pay attention to the relation between wages and productivity. And by the way, workers know this too. They totally understand, if you're going to say to them, look, we want to raise real wage and we fight for that. Yes, if you say to them, we want to wage it to the point where your costs are such that your company will not be competitive and they're shut down, they go, well, no, that's not a good goal then, is it? So this is an old question. And that means that I could ask Philip's question in three different ways. I could do rate of change of money wages or I could replace this by real wages. That's another question. I could look at the data. If you do it in Philip's time, you don't see much difference because there's no inflation. Or I could say real wages relative to productivity. So that becomes then the rate of change of the wage share. All of these are legitimate questions. But you'll see that only one of them works all the way through the post-war period. And that's this one. And this is the one that is actually natural in the classical tradition. It's the one that's discussed in the argument in Marx about the reserve army of labor. Chapter 25, section one and two. He talks about how if wages rise as unemployment shrinks, the reserve army shrinks, labor market gets tighter, workers in a better bargaining position. So real wages begin to rise faster than what they were doing before. And that then he says at some point, cuts into profitability because the wage is rising, the wage share rises. From his point of view of his terminology, the rate of surplus value declines. If that's the case, the profit rate is falling at some point that inhibits accumulation. And that causes capitalists to begin to import workers to counter this, to attack the unions more than they would otherwise. And also to replace workers with machines more rapidly than they would otherwise. That was a point of the chicken run example. They always replace workers with machines. But this inherently, this accelerates that incentive. And so they have a built-in mechanism to counter this within limits. Now, the conventional Phillips curve, as I said, stop working. But you can derive another Phillips curve based on a different hypothesis. And I'm going to, I don't know if I should do this now or not, but anyway. Let me tell you, okay. So the point of the Phillips curve, the conventional Phillips curve, is that unemployment tells you when the economy is getting tighter, right? So the closer you come to full employment, the more the economy is unable to grow in real terms because of a lack of availability of labor. But that presumes that capitalism cannot recreate the availability of labor by importing workers or by displacing them through machinery. That already the assumption that the labor supply is something given to capitalists as a whole and they can do nothing about it. On the contrary, it's obvious from the point of view of individual capitalists and history that where there is a limitation of labor supply, the import of labor is like the import of any other thing that is in short supply. It can come from abroad, can come from people who are not working, women previously or non-white people and so on, that borders can become more porous. But then you have a problem. If the unemployment is not the limit to growth, what is the growth limit? So if you're coming from this perspective, from a classical perspective, unemployment is not the limit to growth. And in fact, that's exactly the wording in the discussion of the Reserve Army in Marx. He says capitalism creates the unemployment it needs. How does it do that? If the pool gets too low, then it starts to open the floodgates, open the gates so that more people are drawn in or pulled in from abroad, but it also increases the displacement of labor through mechanization which goes faster and so the pool of unemployed people goes back up. Those are his words. So then you have a theoretical problem. What is the limit to growth? And the answer is profitability. I mean, in the classical tradition, it's clear that investment depends on profitability. It's also in Keynes. And so the limit comes from the effects on profitability. There's also a physical limit to growth. Profit has a different dimension and this comes up in Ricardo. Ricardo comes up with a simple explanation of the relationship between the profit rate and the growth rate. And it's very beautiful actually, which is that imagine that I were making, he does it in a case of what came famously called the corn-corn model. Imagine that I take a certain amount of corn as seed, let's say 20 bushels of corn. And I have to pay 30 bushels to my workers. So my inputs are 20 bushels of corn and 30 bushels for the workers and I get an output of 70 bushels. So 20 plus 30 is the inputs of corn and the output is 70, so the surplus product is 20. The money value of the surplus product is profit. The ratio of the money value of profit to the capital inputs is a profit rate at this level of abstraction. So it turns out that the profit rate is a price of the surplus product divided by the price of the inputs, the average profit rate in the economy. But it's also true from a physical point of view that you cannot grow faster than putting all that corn back in. In other words, that 20 bushels of corn, which is a surplus, can be eaten, in which case your growth rate is zero. Some of it can be invested, so your growth rate is above zero, but you can't invest more than the profit rate. And therefore, you get from Ricardo the point that the growth rate is less than or equal to the profit rate. Or to put it indifferently, the maximum growth rate is equal to the profit rate. Now this is Ricardo's, you can see intuitively, that's true and you can easily show it. The nice thing is almost all the great ideas were done without algebra, but this is von Neumann's growth path. The von Neumann ray is the maximum growth rate in an economy in balanced growth using multi-sectoral analyses. And that is the growth rate, which is equal to the rate when you put all the surplus product back in. Again, this sounds like an abstract example in von Neumann's balanced growth and all that, but this problem came up concretely in Russia. After the revolution in this Soviet Union, they had a political issue and an economic issue. They had lots of unemployed workers, but so they weren't limited by the supply of labor, but they couldn't grow at a rate beyond a certain level. And the question was why is that? And they discovered quickly, because the Russians were great theoreticians, they discovered quickly that that's because you can't grow faster than the amount of surplus product that you generate, which you have to put back in in order to grow. I mean, if you want to have more corn in the next period, you have to put more seed in, right? And you have to put more higher-more workers, so you have to put more inputs in. But the surplus is the limit to how much you can expand the inputs. So they basically invented input-output analysis. They had to. Leon Tev brought it over, but they invented it. And they ran into the same problem. So the von Neumann problem was actually politically fundamental to the existence of Soviet society in the early years, and they struggled with that problem. Well, I would argue that, in fact, the same mechanism operates in any country, that was von Neumann's point, does not mind, that any system cannot grow faster than the surplus product it generates. That tells you that the rate of growth, the maximum growth rate, is equal to the profit rate, and that can be shown more formally. As an economist, I always have to throw in a little algebra, a nonlinear algebra, just to keep people convinced that they're doing economics rather than logic. And this is fairly easy to show. But then it follows that the rate of growth, actual rate of growth relative to the maximum growth rate is a kind of measure of the tightness of the economy. Is that point clear? If this is the maximum growth rate, then the ratio of the actual to the maximum is a measure of the degree to the economy is tight in its growth capacity, unconstrained by labor, because there's always a labor supply, except in rare conditions like wars and so on. So if that's true, what is this measure? This is investment over capital, that's the actual growth rate, and the maximum growth rate is profit over capital. So you can actually measure the degree of the tightness of the economy by the ratio of investment to profit. Now that was an important thing for me to understand because it means that I had here a measure of the utilization of the growth potential. I call this the throughput ratio. From engineering, you know, if you put fluid through a pipe, there's certain maximum flow that you can get through the pipe. And that's the throughput limit. The flow you have is usually less than that. And so the throughput ratio is a ratio of the actual throughput to the maximum throughput. You can see the analogy there, right? So I called it the throughput rate. And we began to explore in seminars about the relation between that and the Phillips curve. Now the Phillips curve is inflation on this side and unemployment on this side. Unemployment is the employment rate minus one. It's what you do employ minus assumption that you employ everybody. So it's employed over the labor force. One minus that is the unemployment rate. So the equivalent measure for me would be one minus the throughput rate on the axis. And if you look at the data, this is a conventional Phillips curve unemployment rate and do I have it twice? No, that was a conventional Phillips curve, yes. This is what happens if you replace, keeping the vertical axis the same, you replace the bottom by the growth tightness rather than by the labor market tightness. And you can see that you get very much something like a Phillips curve. Now that's only part of the explanation. You have to also add in a very important factor, which is the degree to which you stimulate the economy by deficit finance. It's creation of new purchasing power. Because obviously, if you are close to the upper limit, you may be some inflation is creeping in from that, but if you jump to a World War II situation and pump up the economy, you're gonna get inflation, which is not dependent on just some kind of tradeoff, but inflation due to the pumping. So you want an equation, so to speak, and this is what we did. If you have the growth utilization rate as being the growth of capital over the profit rate, then demand pull comes from the creation of new purchasing power, new public and private credit and net exports, which you can measure. The IMF has data on this for every country. The real output growth will be harder and harder for real output to grow more, the closer you are to the limit. Think of it as a spring that you're pulling on, right? The more you pull on the string, the more resistance there is for growth. So you need some idea of output growth, resisting further output growth as it gets nearer to the top part. And the creation of purchasing power is pulling prices up. The tightness of the economy is restricting the growth rate. So inflation will be the result of these two variables, the demand pull through new purchasing power and the tightening of the economy through the degree of utilization of the growth potential. And that, the form of the equation is not obvious. We don't have any a priori. You could try linear or non-linear. We try different things. But what we found is that this data for the US and then for 13 other countries was very well explained by just these two variables. And that's in the book. But here's an example of the data applied to the United States. This is the US. Some reason this is not working. Well, I'll try and have it fixed. It must be the battery may be dead. But we look at the equation at the top. Pi is the profit, is the inflation rate. Depends on the lagged inflation rate. It depends on two variables. One which is a function of the combination of demand pull and growth resistance and the other is another function of those two variables. It's a non-linear combination. But you can see that you can get a very close explanation of inflation from just two variables and those are familiar variables. If you're a Keynesian, you would say, oh, yeah, the first variable is demand pull. Everybody knows that. The second one, however, they would say inflation. And I'm making the argument, the second one, the classical point of view is the tightness of the growth potential. Now, why do I go this long about way to do that? The point is that if you're going to do theory, you need to understand that different theories have different ways of explaining the same phenomena. And one way to compare them is to look at the same pattern and ask how they do it. What's the commonality? Keynesian and this classical explanation have the commonality that demand pull will pull on output. But when does it become inflation rather than output increase? That depends on the response of supply. And the supply response depends on the tightness of the economy. So that's, again, the same as Keynesian. The difference is that Keynesians measure tightness by the unemployment rate. And I'm measuring tightness by the Ricardian and von Neumann tightness of growth potential. But that one substitution gives you a different theory of inflation and a theory that actually works extremely well across a bunch of countries in different places, different time periods. OK? Now, I may or may not get, I won't get to the theory of inflation, so that's why I'm trying to sneak it in here. But I want to show you then, remember what started all this. What happens if you stimulate the economy? Well, as Marx points out, if you stimulate the economy, you reduce unemployment. As you reduce unemployment, the labor market gets tighter and workers are in a better position to have wage increases relative to whatever they were doing before. So the rate of growth of wages tends to rise. It tends to be growing faster than it is. So the tighter the economy, the more rapid the rate of growth of real wages. But in the argument that Marx is not real wages, it's real wages relative to productivity that's relevant. So you want to look at the wage share growth. And the verbal argument is actually wages rise and the wage share begins to rise. So we look now at the rate of change of the wage share. And that becomes very important because we want to know what happens to the wage share when the economy is in a boom phase or a stimulated phase as opposed to in a non-stimulated phase. Everybody with me? So what I'm plotting here is a rate of change of the wage share, real wages relative to productivity growth, growth of wages relative to productivity growth, or the growth of real wages relative to productivity. Everybody clear on that? Yeah? And here I'm measuring unemployment in a standard way except I make an adjustment for something which you observe in the data plays a very big role, which is that in the neoliberal era, unemployment, duration of unemployment rises dramatically. Again, you have to know what your data tells you. As I said yesterday, unemployment tells you that you're employed if you've got two hours of work in the last couple of weeks. That's not a very good measure, is it? So you want to use other measures. But the other measures don't go back to 1947, except for one. And that is the duration of unemployment goes back. So I combined the unemployment rate with the length of time that you're unemployed to create a measure of unemployment intensity. It's intuitively plausible because after all, being unemployed is one part of the problem. How long you're unemployed is another part of the problem. And the idea is that if you're unemployed for a long time, that is definitely going to have an impact on your ability, on workers' abilities, to bargain. If they know they're going to be unemployed for three years, then they're going to be bargaining a different frame than if they think they could be unemployed for three months. So with that little change, which I described in the book and the data is there, this is the actual economy, US economy, coming out of World War II. So it begins in 1949, because that's where my data begins. Oops. Yeah, there it goes, 1949. Now the arrows tell you the direction. So it's beginning in 1949, and the wage share is growing at about 1%. So wages are rising 1% more rapidly than productivity. Is that clear? That's because World War II is a huge stimulus. So they're in a very strong position after World War II because of this full employment. And you can see they're down here at a very low level of unemployment intensity. They'll never get that low again, actually. So 1950, 51, 52, 53. What's happening here? Every one of these wages are still rising, but they're rising more and more slowly relative to productivity. So the wage share is growing, but it's growing at a slower rate. As long as you're above the zero line, the wage share is growing. Below it, it's falling. But you can see the growth rate is falling. Now, I lost track of where I was, but this is about 57, 58, 59, 58, 59, 60. Now 1960, 61. What happens in 61 in the US economy? Pardon? Yes, but also the Vietnam War. Huge stimulus. And Johnson takes that over and creates what he calls a great society program, which was a stimulus program, pure financing of output growth and deficit financing and all that. So we're entering a Keynesian stimulus. That's the whole point of this regression. You were entering a Keynesian stimulus. So what does that Keynesian stimulus do? It starts to make you move up the same curve. These are the actual points. They're not fitted. They're not graphed. I'm just plotting the actual points. And you can see you're basically moving up. And 68 is pretty much the peak of the Vietnam War stimulus. We have lots of measures of that. And the war is winding down. The US is losing the war. The population draft was extremely unpopular. And Johnson had the economy begin to cool off as a stimulus package was cut back, as a war was cut back. And so you're now moving back down the curve. So by 1973, you've made a 13-year round trip due to Keynesian policies. And what did they do? They pumped up output. They pumped up employment. They reduced the unemployment rate. But they also raised the rate of growth of the wage share until the stimulus package peaked. And then as that stimulus faded away, the rate of growth of the wage share began to fall again. And you came back in 1973 to where you were in 1960. After that, you're in the period of stagflation, which I mentioned before, and we're going to come to again. You have unemployment rising, inflation rising. Stimulus packages are now forbidden because governments are not able to maintain that in the face of inflation. So the wage share, rate of growth of the wage share declines. And by 1977, we're now at a period where the wage share is growing at a zero rate. You've reached stable wage share. But unlike mathematical models, economies don't have to stay at the equilibrium point. And so the wage share continues to fall because we are in a crisis now. We're in a great depression in the 1970s. And they keep falling. So now we're in negative territory. This is now the wage share is falling at 0.2%, not very great fall, but it's falling, whereas previously when we started, it was rising at 1%. So it's falling at 1.5% percent rate here. And then comes on the picture, neoliberalism. Reagan, I showed last time how the wage gets smashed and gets flattened out by neoliberal policies. So the wage share gets declined. But something interesting happens. The curve itself is shifted. This is easier to see in the scatter plot, but it's in the book. But the curve gets shifted to another curve. And now we are in 1993, and you get another stimulus. It's called the dot-com boom. All kinds of money was being spent on through borrowing and credit and deficit finance at a personal, private level. And you get, again, the wage share rate of growth is negative, but less and less negative. Then the dot-cott boom dies. And we go back to the wage share growth going negative and more and more negative. And here now you hit the Great Depression, the first Great Depression of the 21st century, which is 2007. And by 2011, the wage share is falling. Now these are two curves. In the data, as you can see that clearly, you fit the data, you have one curve here, and you fit the data, as Phillips does, a second curve. And what Reagan did and what Thatcher did in England is to shift the curve by changing the institutional and political balance of power. So this is an important point, the role of institutions and power in determining these relationships. This is the golden age of labor, a relationship between wages and productivity, which seemed to depend on unemployment. But there's also a tightness there, which came from the fact that unions had power. And there was a state, a welfare state, that supported certain institutional practices. That was smashed here, and the curve actually shifts down. So you get two things. The curve gets shifted due to a change in the power structure, but movements along the curve reflect stimulus and it's dying away. Any questions on this? Now I'm setting up a problem, which I'll come back to later, which is how do we understand the limits to stimulus? Because clearly we see from this data and we see from other data in the book, which I talk about, that a stimulus pumps up output in employment. You can see that here, because where they had the Vietnam War stimulus, unemployment drops, which is pumping up the economy. But as unemployment drops, the rate of increase of the wage share rises. In this lower period here, on this new curve, as the dot-com bubble, you pump up the economy. You go in this direction, which means unemployment drops. The wage share is not rising, but it's falling at a slower rate. So it's pumping up again the pressure on the wage share. So stimulus clearly has an effect, but it also has a counteracting effect, which is the effect on the wage share, which is going to be an effect on profitability. And you can see this when you look at the data. I can never get out of this. Yeah, there it is. This is the data on profitability. Here's the profit rate, calculated from the BEA. It's falling until about 1960, and you get the Vietnam War boom. If you adjust for the deficit finance, you're going way above the trend and coming back. But then the trend comes back, and now you enter this period here, which is the great stagflation. There are business failure rates, bank failure rates, unemployment was rising. So this is a very bad time. This is the stagflation depression. And then comes Reagan, and the profit rate gets stabilized and actually rises a bit for some time period. So this wasn't just a populist movement about having more moral backbone and all that. It was a movement that succeeded in one of its core things, which is restoring the strength of capital relative to labor. And that's something you have to understand, that economics is about power. It doesn't mean that power can determine anything, but power and institutions have an influence, and you have to know, therefore, the limits to that influence. And the whole point of this earlier diagram was that stimulus can make it seem, as if you can pump up the economy to any level you want. But if it causes a negative impact on profitability, and hence on investment, and hence on growth, then it's not true. So you have to know what the limits are. Any questions on this? Yes, sure. There's any other way? So your films go away? Yeah. Yeah. So I guess you said, in 1960, stimulus had a peak, right? Peak, yeah. The rate of growth of stimulus or something like that, yeah. Doesn't profit rate go as well, right? Well, all the way here, the wage share is rising, the profit rate is falling, because the rise in the capital intensity is very slow. So the movements are going to come from wage share. I have to explain to people what this means. The profit rate is simply, the rate of profit is profit over capital, but that's the same as profit over output, which is a profit share times capital over output, which is a capital intensity, right? So if this is going up or down, this goes upward modestly through the whole post-war period, when you adjust for capacity utilization. We'll come to that in the last lecture. But the big movements are going to take place in here. This rises at like 1 and 1.5% a year pretty much over the whole post-war period. This, on the other hand, goes up and down with what's happening with politics and conjunctural factors and stimulus and all that. So the big movements in the short run and medium run come from that. So what's happening up here is you're coming off World War II, the wage share is rising, the profit share is falling. And you can see this movement and the profit rate from that. But if it's falling at a slower and slower rate, so to speak, as the wage share growth becomes smaller and smaller. I'm not sure I answered your question to that. So I guess I'm confused because is like capital reacting to the wage share for like? Absolutely. And then dynamic through debt. OK, very good question. So there is no such thing as capital as a person. There are a lot of capitalists, right? And so they have differently situated reactions to this. Some are hurt more than others. But very big and powerful interests were behind the Reagan Revolution. We know this. We know from the funding and the Koch brothers and all these people. And there are some interests which are general to most capital and profitability is a general one. There are other more local ones. If you're an oil capitalist versus a mineral capitalist or something, you have more specific things. But wages is a general one. So the struggle between capital and labor is a general struggle. And this was in fact one where many people on the left say, oh, these were wonderful times, the golden age of labor. But then they see this collapse of those wonderful times as a political change. I'm trying to point out that we have to understand that that political change had a motivation also, which was shifting the decline in profitability and a reaction of the capitalist class, which was to attack unions and attack the welfare state, which they held responsible for this golden age. And they were right. They were responsible for the golden age. So they attacked it. And they made it the golden age of labor into the golden age of capital. And they're open about this. It's not like a mystery. Yes, you had a question. So I understood the relationship between the wage share and the ratio between the growth and capital growth and maximum growth, because it's the good question. Very good question. But I didn't find the match between the relationship of that ratio with the equation. We're on the same last part. That is the difference that you showed in the other slide. I didn't have the relation. What is the reason behind that ratio of capital growth with maximum growth within a person? I didn't do that here, because that is a whole chapter on inflation. So yes, you're absolutely right. I can tell you this, though, an interesting problem arises. If, in fact, the profit rate is falling, if the profit rate is falling, then the maximum growth rate is falling. That means that any given growth rate, the economy becomes more inflation problem. That's a very big clue. Because what you see is that as the profit rate is falling, the growth rate actually doesn't change very much, because of the stimulus packages. In fact, sometimes it gets closer. So it means it's tighter from a growth rate perspective. The economy is tighter. Even though the profit rate is falling, and the growth rate is falling, but the profit rate is falling faster, the economy can become tighter. And the stimulus is exactly what you're supposed to do if you're a Keynesian at that time. Growth rate is falling, unemployment is rising, you need to stimulate. But if the profit rate is falling, the top is falling, and you're closer to the ceiling, and you're going to get more inflation. And that's exactly what we find econometrically and also graphically. So you can see that, again, this is an illustrative of how a different theoretical framework will read the events historical and economic in a different way. And the point I'm trying to push is that you should not make up your theory for each event that you like. I am strongly opposed to that. I went to Catholic school, so I don't believe you should make up a new version of a story every time. I do believe you should have some responsibility for having a consistent theoretical framework. And I know that many people believe just the opposite. If you like outcome, you want workers' wages to rise and you look for a theory that tells you it's good. The trouble is that if you are wrong, they suffer. In the old days, if you were an astrologer and you were wrong, they would hang you from the front of the turret just to remind astrologers that there was a feedback mechanism. But nowadays, you just get another grant and move on to another job, that's all. And there's something to be said for severe feedback mechanism, not that bad. I'm an economist, I don't want to be hanging from any place, but there's something to be said for people being responsible for screwing up other people's lives. Remember Jeffrey Sacks? So he didn't get punished for that. He went out to New Heights. Now he says it was because they didn't follow his reasoning, shock therapy and all that. Well, not true, or whether it's true or not, there were terrible consequences for other people. But my point is that Keynesian policy also had that consequence. And it was not because of the nastiness of the people involved, they were the best intentions. They believed strongly that this is what works, but they were not sufficiently connected to the whole story. And let me say one more thing. Keynes is really the source of the Keynesian policy, right? So how come he didn't see this? There are many reasons Keynes doesn't really have any good theoretical foundation for profitability. The driving mechanism in Keynes is a marginal efficiency of capital, which is the profit rate on investment. Yet he has no explanation for it, except to saying that in the short run is dependent on the animal spirits and all that. Keynes is very clever. Whenever you corner him, he shifts to another place. He's like the Muhammad Ali of economists. He dances out of the way. But the problem is it doesn't give you an answer. And Keynesians have been split ever since. Is the marginal efficiency of capital linked to the real economy, profitability? Or is it purely arbitrary and conventional? Now, obviously coming from the framework I do, and I'm going to make this point more concretely, when we look at the data, the marginal efficiency of capital is linked to the actual economy. We can measure it, and we can explain it from exactly the movements of wages and productivity. So that's a key point. Any other questions in the back here? In that profit rate figure that you just throw over there, you talked about the profit share, right? Yeah. Peelware outcome. So you said when the wage share goes up, the profit share goes down. So that the way you're picturing it is that there is profits and wages there. So rents would go into profits, rents would be excess profit. Yes, in fact, I spend quite a lot of time in the book on how to decompose national income accounts, so that you take into account rents' interests. But broadly speaking, this is what national income counts are called gross operating surplus, or net operating surplus, which is very close to the classical idea of a surplus. So you can map these figures, and profit comes out of that. Profit is essentially profit before payments of interest in rents, and you can decompose that in national accounts. I hold detail mapping from NIPAA accounts to this kind of framework. Okay. One more question, and I should move on. Yes. So again, in your Phillips curve, you only work on the taxes you measure in the unemployment density. It's like a measure through, it's a sort of unit to measure the strength of your G-class, and it's like that's the kind of point. Yes, let me just clarify that a little. Marx has this wonderful phrase in the discussion of the Reserve Army of Labor. He says that when labor market is tight, the workers feel confident and they feel strong, and that very confidence, which is the ability to raise wages, creates conditions where unemployment rises again, and he says it puts a check on their pretensions. Now the point is that duration of unemployment is a very important check on your pretensions, on your ability to be confident about struggling for wages, because you get thrown in the labor pool and you're out of work for several years, that's a big hit. And so I wanted to take that into account in when measuring unemployment, and so I make a combination of the unemployment rate and the length of time of unemployment. Basically I'm measuring the number of unemployment hours over some kind of normal unemployment hours. So if that number gets bigger, then the unemployment rate is worse. It's the impact, I'm interested in the impact on the ability to bargain, but clearly you have to take that into account. So does it capture the union density or the impact of the low-missive fluctuation of production of the workers? I didn't do that because I wanted a nice clean graph, but of course I have graduate students who have done that and they're doing what they do. Yeah, you can do that, yeah. You want to show essentially the relationship with the duration of unemployment between unemployment, duration of unemployment, openness to the world market. But if I did that, then I'd have to say, well, I explain a very complicated econometric thing and this is such a lovely graph because it's so clear of what's happening at least in the US, but it doesn't mean that it would necessarily work in other countries that, and I could track year by year of what was going on and related to well-known pumping stimulus program things, which is one of my interesting questions. I think the question is, is too sensible that unemployment rate to the nation? Like we see this graph in Korea and very much. I just said that again. This graph would be too sensible to just the unemployment because instead of, we kind of find the original. Unemployment intensity, you can. No, but the same, we only see this graph with unemployment rate and not density. Yeah. It would be even better? No, it looks completely different. And because, first of all, I'm looking at the rate of change of the wage share, not Phillips curve. But it looks somewhat different because you're not picking up the pressure on the labor force that comes from being unemployed for, unemployment intensity doesn't change very much till the neoliberal era. So unemployment rate and unemployment intensity are pretty much the same. But then the duration of unemployment shoots up because capital is going abroad, people are losing jobs and are getting them again. And so the picture looks somewhat different. One last point, because I didn't do that here. I don't know how I see this. You guys are so green. Nice being here and held by a magnet. Okay, I wanna just make one more point, which I mentioned in the book and since I may not come back to it. I'm saying that the rate of change of the wage share, I'm gonna use this to mean the rate of change. So it's the wage share, wage bill over the wage, over value added is a function of unemployment intensity. That's a picture we're looking at in there. But that has an implication for the Phillips curve because the wage share is equal to the wage rate times labor over output, which is the wage rate over nominal output per worker. And you can also then write this as the real wage over productivity because the real wage is the wage rate, this is the wage rate over prices. Now, why is that relevant? Because if I have this relationship, then I can explain the movements of the money wage, which is the Phillips curve and the real wage, which is the Friedman curve. Because from here, I can write that the real wage, the rate of change of the real wage, I'm sorry, of the money wage, of the money wage plus inflation, which is the rate of change of the real wage, that's the top here, here, minus the rate of change of productivity is equal to this function. And that means I can explain the Phillips curve by saying that the Phillips curve depends on this unemployment part plus, minus, this is minus, sorry, minus, yeah, minus, minus, minus plus inflation, plus productivity growth. So the rate of money wages put a little dollar sign here, just so this is money wages per worker depends on the degree of intensity of unemployment plus inflation plus productivity growth. Why is this relevant? Because inflation shifts the Phillips curve. So that's kind of like the expectation augmented thing, but so does productivity growth. And I can track, I can show in the book that I can explain the movements of the Phillips curve, which is this big mess that we saw before from these two variables, in addition to the unemployment intensity. So this allows me to explain what happened, not only from this alternate framework, explains what actually happened. So I wanna do one more thing that I'm gonna stop and switch to a different framework, which is different question. I wanna make the point that capitalism has created wealth. I made that in the beginning and I wanna emphasize that here. Capitalism creates wealth and it creates wealth per person. And this is data from Madison, which is the per capita GDP, real per capita GDP in different parts of the world. So here we have Western Europe, beginning 1600. I don't know how reliable these numbers are in 1600, but anyway, this is the Madison data, beginning in 1600. And so you see Western Europe going along and now beginning in the 18th century, the effect of capitalism, so it's beginning to rise. And then the 19th century, it really takes off. And here is this huge growth of per capita income, real per capita income from capitalism. The Western offshoots are U.S., Canada, Australia, New Zealand. So they are obviously much poorer than Western Europe in 1600. And then as they literally get settled and become capitalist producers, you see this tremendous growth. They have great resources, they have a lot of land and you see that their per capita GDP overtakes that of their home countries or the home countries of the bulk of their population, the invading Europeans. Asia remains sort of dormant for a long time from 1600 all the way up to about 1850, 1875 and then begins to take off, but no wait, Asia, here, sorry, I'm sorry, Asia begins to take off in the 20th century and this is a famous development decade, so you see a little bit here and then you see this rise here. Latin America, including Caribbean, the data doesn't go all the way back, but you can also see the take off point in the late 19th century. And Africa, the latest of all, starting from the same level really as Western offshoots, which is really interesting, comes to the development decade much later. Some growth in the late 19th century, but stagnation here and so the point, I started the whole set of discussions with, the graphs with this, the argument that capitalism creates wealth, but the other side is that capitalism also creates inequality and one way of looking at, this is Madison's data, I took in the data, the top four countries at any moment of time from 1600 to 2000 and whatever, and the poorest four and here you see the top four countries rise from a GDP per capita of about 1,000 in international dollars and you could see the tremendous increase, these are log scales, so this is really a tremendous increase. You can see the bottom poor, on the other hand, have some kind of increase in the late 19th and early 20th century and then you get actually a fall. So it's important to be concrete. Capitalism creates wealth, it also creates poverty. It discovers new resources and it destroys old resources. It creates new culture and it destroys existing culture. That's the point of the beginning of the communist manifesto actually, which is that it has this double-sided character. So I took the ratio, so in this graph, it's the top, richest four at the top and the poorest four at the bottom. I took the ratio of their GDP and here you see a very characteristic phenomenon of capitalism that it creates wealth, but it creates it more and more unequally. Okay, and that's not a surprise now after Piketty, but I've been doing this graph for about 20 years and it is a feature that needs to be understood theoretically, socially, structurally, politically, but also theoretically. Why? What is it that creates the intrinsic inequality across nations? To do that, we need to talk about international trade, which is a very important feature. I won't get a chance to do that, but it's in the book. You can take a look at some of those things. Okay, so let me stop there because I'm gonna switch gears. Well, I'm sorry, let me stop here. One last. I showed you all these graphs for a reason, which is that these represent the patterns of actual capitalism, turbulence, fluctuations, but underlying order, structure, order. And so if you're going to do analysis from this framework, the first lesson you have to understand is that the underlying principles are regulative principles, but they're gravitational, turbulent gravitation. The proper analogy is water flowing in a stream around rocks. You can see what's called turbulent flow, not equilibrium as a state. Again, if I could zap you every time you use a word equilibrium, most of you would be burning cinders by that point, but you even think equilibrium because there's no such thing as equilibrium. There are tendencies that pull you towards some center and there are things that move you away from it. So if you wanna think about this as a differential equation system, for instance, you can have differential equation system that'll convert to an equilibrium. But if you add noise to them, representing all the details and complications and interactions, then you get a stochastic differential equation, guess what? You don't get it settling to equilibrium, you get constant fluctuations around whatever the center of gravity is and you get a distribution of outcomes. And that alone is an important question because that distribution of outcomes is a big issue in physics. These are the Fokker-Planck equations, it's the simplest way to think of these and they're concerned not so much with the way it settles but the distribution. We tend to be concerned with the center, but when we look at this, we have to ask both, which is how are we going to see? What are we going to see? We're going to see a tendency towards an average. We're also going to see a distribution. Okay, the tendency towards the average we did before. This is the incremental profit rates in the United States. Incremental, meaning the change in profits over investment. That is actually the definition of the marginal efficiency of capital. So here is Keynes's marginal efficiency of capital. This is what it looks like. You can see why Caplus would be very uncertain about it. It fluctuates, this is the actual data, but you can also see that it converges around a common center. This is gravitation. This is turbulent equalization, but in no moment of time will it ever stand still. And that's a very, very important point because then when you look at the real data and you don't realize that and you just look at this period of time here and you say, oh my God, they're not equal. Then I look over here, they're not equal and they're not equal. I don't look at the time thing, I just take snapshots. Then I say, well, equilibrium doesn't exist as all kinds of monopoly power and imperfections, but this is a perfection. This is how perfection looks in turbulent gravitation. So that's the first point. Just one second, let me do the, so then the issue is that if you're looking about turbulent gravitation, you can't use something like the long period method. That's a complete fiction. It's an analytical device. Yes, I could ask if nothing else changes except to profit rates are unequal, then if they go towards gravity and equalization, we call that the long period because they move slowly. But that's not going to describe anything. It's just because you are not complicated enough, mathematically we're not sophisticated enough to ask the question of how they actually move. So we get rid of all the important issues, which is the actual path and just focus on where they come together. But that's not useful for actual analysis because as you can see from the fluctuations, the fluctuations are very big and very relevant compared to the movements of the center of gravity itself. So you have to have the appropriate mathematical tools. If you're gonna do this, learn about the tools that physicists use. So I started by saying, I don't believe that the problem in neoclassical economics is that it's too mathematical. I think the problem in neoclassical economics is it's a long theory and its math reflects the theory, its theory. If you have a different theory, you're gonna need a different math, different econometrics too. Talk about non-linear. That's what I call a non-linear process. This is non-linear. These are non-linear processes. Dynamic, stochastic, non-linear processes. So what's the econometrics you're gonna use for this? In the book we use some simple econometrics but it's clear that the question is different when you have a different theory. Okay, let me just finish that. The second point, I'm gonna come to, so forget about that, and the role of agency, the third point we've already talked about, for instance, how the wage share's path was changed by political agency, by class struggle, but also the agency of workers struggling for wages and the agency of capitalists resisting unions and attacking workers, importing workers. So agency is part of the story but if there are mechanisms that override certain processes, then agency does not produce the outcome. Let me just explain what I mean by that. Suppose that this side of the room has higher profit rates than this side of the room, right? So now capitalists are gonna pour money more rapidly into there, maybe the same people are gonna pour their money more rapidly and some of these people are gonna be moving some of their money over there. Why are they doing that? They're doing it in order to get a higher profit rate. The ones here in order to preserve it, here to get a higher profit rate. What's the consequence of that? Equalization in this turbulent sense of profit rates. So the incentive is to get a higher profit rate, but the result is an equalization of profit rates. It's an emergent property. It's not intentional. Same thing for wages. If wages are higher over here than there, then workers start to migrate, stochastically migrate over there and that puts pressure on the labor supply, brings the wage down here, reduces the labor supply relative to demand, raises wages, equalization. But on the equalization as inequality, it just means fluctuation around similar means and that is a process that is not desired by anybody. No one going there saying, hey, we wanna equalize wages. They're going there saying we want higher wages. And there they're going, we don't want these people because they're gonna lower our wages. There's just this whole immigration displacement of people in the labor force comes from the same incentive structure. They understand perfectly well. If you increase the labor supply, they lose their bargaining power. So that's a real process. So again, they're always winners and losers. The stage on which these historical and political events are played out is always moving because capitalism moves. Just like Galileo supposedly said when he was threatened with being burnt at the stake, he recanted, but then supposedly he muttered under his breath, but still it moves. So we have to remember it moves. So let me take, stop there. Open up for questions and I wanna switch to the theory of the firm. Yeah. I'm sorry if you could expand a bit about the power of the labor force and the difference between the two. I mean it's a welcome story that I have to explain is that the two different ways to deal with the political happens and the same delivery. Just that you see what you suggest so that it's being covered with the imperfections. I'm gonna actually get to that in the next part. So remind me, because what I want to show you now is how competition works and what its dynamics are. And then I'm going to show you can derive all the empirical patterns that Kolecki derives from monopoly from just the movement of competition. But just to anticipate that, if I were looking at data on profit rates, and I want to ask it myself, are profit rates equalized? And I just took slices of this graph. I didn't do the time graph, right? And I just took slices. I would find the answer is no, profit rates are never equalized. If I make this a time graph, then I say, oh my god. I didn't realize that what I saw here was a dispersion. I saw another dispersion, but this one is now down here. And this one is back up there. That I didn't see, because I didn't ask the question of dynamics properly. And Kolecki does not ask that question, actually. He has a very static framework. So I'm going to try to show you that, but yet, the problem is that Kolecki identifies some problems. And every alternate theory has to explain what the other side explains. And has to hopefully explain it in a way that's linked to a different theoretical framework. Just like Galileo and Kepler had to explain orbits. They couldn't just say there's no orbits, or maybe we can solve the thing that had to explain it. And when they did, they had to explain everything that had been explained in a different way. They had to unify it. So I'm pushing for a unified field theory, basically. And I'm not claiming that I invented it. It's been there. But it's been there, buried in the unfinished works of Marx and comments in Ricardo that always agree. So you have to extract from it a coherent framework. And that's a task I set myself. And I believe you can do a lot with it. Of course, you can't do everything with it. But the point is that what you can't do with it, you have to ask, how do you do that? If it can't be done with it, then that's a problem for the theory. If it hasn't been done, that's a different issue. And of course, every theory has many things that haven't been done, hopefully. Okay, so let me now switch to, we have a half an hour and I'll do as much as I can before we take, we stop and I'm running behind. You may have noticed, but where is it? Let's see if this, normally I print out my lecture so that you can just see the graphs. But now I can't do that because I forgot. So make sure that I, I think I can do it from here. So, okay. So before I go there, let me just, oh, I'm sorry, that's the wrong graphs. I want this one. Okay, I've jumped, I forgot that I didn't do the consumer theory, which I need to do now. Okay, so let me pose a problem to you. You've all taken micro, you know that if you want to analyze a consumer, you have to endow them with absurd properties, transitive preferences, knowledge of the past and the future, perfect information and all of, and the idea that they are selfish. You have to start with that. When I first came into graduate school, coming from Pakistan and Kuwait, I was offended. I was culturally offended by this. This is such a stupid idea. Why are you doing this? Well, because we assume rationality in, it's not true of the West, by the way. You should be as offended if you are a European or you're a Native American. You have to be offended by this because it's a stupid idea. But you have to ask yourself, why is it that it is constructed? Let me just explain what I mean by that. Suppose I take any individual, I say, okay, I'm gonna portray you. The first thing I wanna do is to draw a kind of network of your connections. I don't know how many people see network analysis, but there's a node there, person, and that person is connected to a whole bunch of other things. You're connected to your family. You're connected to your culture, to your football team, if you're a male, something. You're connected to your job. You're connected to your history. You're connected to your friends. You're connected to the home you own. So I can draw a lot of these connections. And this is done, by the way. Anthropologists often do this when they're talking about structures of tribes and all that. They show the interconnections. Biologists do this because we know our neural network is a network of connections. So imagine that each person is a part of this social neural network with things, commodities there too. So I'm connected to a bunch of commodities, but I'm also connected to a lot of the things. And I come along, and as an economist, I cut all the other connections. I throw out connection to your family, mother and father out, your family, your nationality out, your race out, your location out, your neighborhood out. And I leave only one connection between you and a list of commodities, the commodity layer, so to speak. And I say, that's the only thing you care about. And we're gonna start from there. You think, why, that is a really stupid way to proceed because you've taken out of the play all the important things that people do. They care about each other, they care about the passage, willing to die to sacrifice, to fight for a piece of soil. I mean, you can't have a greater commitment than that. And yet, we leave that out and we say the only connection to commodities. And that type of representation is what Marx long ago identified is characteristic of bourgeois economics, which is commodity fetishism. Because it's not false, it's incredibly one-sided. And so you ask yourself, why does a utility function not have those other things? Why is there not other regarding and family caring? And the answer is though, you can create a local thing like that. All the optimality things disappear. You can't do it. You cannot include caring for others and talk about Pareto optimality, which is another stupid idea in my opinion, but that kind of reference. So then you have to ask yourself, how come? How can you do this for 100 years, get by with this representation, which no philosopher, no sociologist, no anthropologist, no normal human being, because economists were not normal, let's face it, no normal human being would accept. And the answer is because it gives you a picture of capitalism, which is ideal and perfect and derived from this foundation. Smith is supposedly the great source of this. Smith talks about, if you look at textbooks, Smith talks about the invisible hand. And Smith talks about people being, have tendency to truck and barter. But I read you that quote from Smith in the beginning for a reason. Smith does not say that this is the only motivation or even that it's a dominant motivation or even that it's a desirable motivation. As a social scientist, he understands that this is a part of a motivation, especially in a particular social structure. Capitalism teaches you to make this central. It teaches you to focus on this. But even so, it does not make you into that kind of being. And everybody, I mean, Amartya Sankar's rational fools, many others, Wesley Clare Mitchell considered this a mental aberration to portray people this way. So it's been said for a long time. But yet we continue to do it. We continue to do it because this is, in our church, in the orthodoxy of the church, this is a portrayal of the capitalist system as ideal. What astonishes me is that's picked up in newly developing countries. I'm astonished to see China moving in that direction, given its history and its politics and all that, adopting this as a framework. And that shows you its power, its way of selling capitalism. Now, I'm interested in its analytical consistency. I think that's a virtue. But if you're going to replace the argument, you have to explain the same phenomena. So what other phenomena that this theoretical framework designed as it is, explains. If you take a textbook like Maskelel and you cut it into parts, those parts that describe empirical phenomena and those parts that talk about the theory and welfare implications and optimality and all of that, you get a picture like this. Oops, I forgot to move it. I'll find it next time because it's somewhere else. See where would it be? Hang on, I think I can find it. I guess it's in the previous course, or where would be the construction? I'll bring it back. I mean, I have it someplace, but I realize I don't have it where I thought I did. So anyway, Maskelel, here, yeah. What we did in the class is we went through Maskelel and looked at pages. Which pages talk about empirical phenomena and which pages talk about theory and welfare implications and those ideology, from my point of view. And what we found is that the observed phenomena take up about 30% and 70% is based on developing utility function, optimizing and so on. The new classical foundations, 70%, either the foundation is about 50% and interpretation, which is very important, interpretation, is about 16%. So one third of a standard micro textbook is talking about empirical phenomena. And two thirds is talking about the theory and social implications of that theory. And you can do this exercise, it's a very good exercise. I don't know what textbook do you guys use here? Maskelel or Varian? Pardon? Sam is different, because Sam doesn't actually talk about these things, at least in the last version of the book. It has game theory, and I wanna come to game theory. But a standard textbook, yes. Okay, let me leave that and go into the question I wanna address. I haven't read the latest version, but I have Sam's micro economics. But you do have neoclassical, you don't do it? We do. Yeah, I'm sure you encountered, and if you teach undergraduate, I'm sure you don't teach Sam's book, so. Okay, I wanna show you that a simple framework, which was actually developed first by Armin Alkian, neoclassical economist, looking at it from a stochastic point of view, is able to explain all the observed phenomena that Maskelel talks about. All of them, without any reliance whatsoever on optimality, micro foundations, or any of that stuff. So, let me start in an obvious way. I wanna draw the parallel to the standard micro thing, so I'm gonna take a notion that income is partitioned into two uses, which is spending on goods X1 and X2, and their relative prices are P1 and P2. If anything I say is not clear, let me know, they should be familiar, right? X1, however, I want to call a necessity, and the reason is because necessities have a lower limit, you can't go to zero, so I want a threshold below which you can't go, it creates a non-linearity, which has a very useful effect. So it has a positive minimum, X1 minimum, and the feasible range of the budget constraint then is that for X is a distance between, if you spend all your money on X1, the necessary good, you could be here. If you have less and less X1, you can't go below X minimum, because you need that as your minimum level. So it means the feasible part of the budget line is here. And obviously this is luxury good, you spend all your goods on the luxury good, you go up here, and I'm sorry, you can't go up here, the all money on your luxury good over your X1 is here. Now you could recast this in terms of disposable income. Disposable income is the money that you have over what you need to buy the minimum good. But then because the price changes, it's not so simple, the mapping can be done, it's very easy, but this is more familiar. So this is your disposable income measured in terms of X1, and this is, well, it's a little bit different on the other side, but formally it's very simple to do. And I wanna ask, what happens if you increase the price of X1? Now you know, for micro, we should know the answer should be that there'll be less of a demand for X1. I haven't specified what this point is. But the point itself can be characterized by the propensity to consume out of income on necessary goods. In other words, this point here can be characterized by the average propensity to consume necessary goods out of disposable income. And that's convenient to do it that way. I mean, it's basically a ratio of A this part to this part, I think, A to B, and that's a propensity to consume. And if we assume that that propensity to consume is stable for a group, the whole group, I don't really care how you do what you do. I say that that's a minimum requirement that there be a complicated set of influences that affect you, that means individually you could change. The stability comes from the fact that there's a distribution of these patterns and they are stable in the large. Then if I define this propensity to consume that I was talking about, which is the ratio of X in excess of X one over the maximum X in excess of X one, and that's basically the actual consumption above the minimum for good X one divided by the disposable income. So it's a propensity to consume X one above the minimum. Then if I define it that way, which is one of the ways you can define that point, any point, then I can rewrite those to say the demand for X one depends on this propensity to consume on the minimum and on income and on price. Now you recognize this right away. If this propensity to consume is given and the minimum is given, this is a demand curve. This is a linear demand curve in the simplest form. So that tells you that if the price of X one goes up, the demand for X one goes down. Same thing for X two, you can rewrite those equation, the same equation in terms of X two because X one and X two are tied together by the budget line. So you can substitute and you get the demand for X two now depends on negatively on the price of X two and also on the relative price of X two. It's a cross elasticity. And you can immediately from here derive the elasticities. These are downward sloping demand curves up at the top. You can see that because increasing the price of either good will cause a demand to fall, which is a definition of a downward sloping demand curve. And yet I've made no assumptions whatsoever about micro behavior except that the collective group has a stable pattern of margin of average propensity to consume of a particular good. And that is a very weak one. I'm gonna show you that there are many, many different models that'll give you that weak one. Now it doesn't, you don't even have to assume that it's true for all time. It's what I call the revolution model, Tahir square model where people can decide to try to overthrow government or overthrow society or whatever and they stop doing what they're doing. And we understand that we should be able to accommodate that. So as a result of social decisions, but the stability in the large is the fact that people make decisions on lots of micro complicated reasons and you get an average which is a result of these. It doesn't require you to do the same thing every day. You could eat a different food every day, but someone else is eating a different food every day too. And in the aggregate these variations produce some kind of stable pattern. That's all the requirement is. Now, so what happens if you have this stable propensity to consume is that if you raise the price of X1, as I said the demand will go down, in effect you're moving here to a lower quantity of X1 and to a lower quantity of X2 also because the demand for X2 depends on the price of X1. And this is down a sloping demand curve, that's all. If you write these out formally, algebraically, you can derive all the price elasticity of X1 with respect to P1 is that term and it's less than one. The absolute value is less than one. That is an important point. X1 is a necessary good and from this simple apparatus, the elasticity of demand, price elasticity demand of necessaries is less than one. That's one of the best known phenomena and in micro you have to create some kind of curves to make it come out right to show this, but this follows simply from the fact that you have a constraint on X1. X2, the elasticity, price elasticity will be one in this simple example and then you have cross elasticities. I want to get to income elasticities. I haven't done anything so far except change, change the, I accept to work with that single assumption that the propensity to consume is stable for a group. You can also get income elasticity since the propensity to consume depends on the sense that the demand depends on income also. You can calculate the income elasticities and the income elasticities turn out to be exactly what you expect. Income elasticity of necessary goods is positive but less than one. Income elasticities of luxury goods is positive and greater than one. Again, the best known empirical phenomena going back hundreds of years, but this follows simply from the fact that necessary goods have a minimum. So there's a non-linearity in the structure. Now, if I take that theory and I look at the expenditure share of necessary goods, I would expect it to fall because the elasticity is less than one. So the amount of money spent on necessaries will fall as income rises and the amount of, this is the share will fall but the amount spent on necessaries will rise. It will rise more slowly than income so the share will fall. This is called Engel's Law in microeconomics but it doesn't require cob-douglas production utility functions or anything. It follows simply from the sole requirement that the group has a stable propensity to consume. That's it. Depending on what other assumptions you make about what happens to this propensity with income itself you can have the propensity to consume necessary goods declines with income. Then you get a slightly different shape but it's pretty much the same as before. But this is the actual data. This is data on working class budget expenditures. Allen and Bowley in 1935 and here you see the expenditure share on food declining as income rises. On the other hand the amount of money spent on food rising. So the money spent on food rises with income but not as fast as income so the share falls. This is Engel's Law and it's often considered microeconomics as a puzzle. But my point is it follows automatically from this single assumption which has nothing about your motivation only about your collective stability. So I wanted to then show that you can derive that motivation oh and one other point. I can change the axes and call this the luxury good I can call savings and then the necessary good I call consumption and what do I get? I get a linear demand curve which is the consumption function. The New York, the Keynesian consumption function follows straight from the same thing except I've now got income on the vertical axis and I've got savings on the vertical axis and consumption on the horizontal axis and you get exactly a neoclassical consumption function. I mean a Keynesian consumption function. I'm rushing a little bit here but it's chapter three. What I wanted end by showing is that this does not depend on any particular assumption about individual behavior. And this is a general point. I cite in the book a physicist by the name of Robert Laughlin who wrote a very interesting book and Laughlin says that though physics has many models of micro behavior, the aggregate behavior is not derived from the micro behavior. It's not true. In fact, much of physics is just patterns at the aggregate level. Einstein's laws for instance are not derived from micro. Einstein in fact believed that quantum mechanics was wrong because it didn't end up with his theory. Quantum mechanics believes Einstein is wrong because he doesn't have the micro foundation but the Laughlin's point is those two are, there's no necessary connection between two. Laughlin says many aggregate laws can be made consistent with different micro foundations. So the fact that they exist as aggregate laws gives them an autonomy because a micro foundation, there are many different ones. And to just give you an example, I think most people know this famous law in physics called Boyle's law which is that PV is equal to RT. P is pressure, V is volume, R is a constant, so and T is temperature. Now this was discovered empirically in 15 something by Boyle and Gay-Lussac and others have discovered this. And this is a macro law. And they didn't need to have a micro foundation for it because there was no micro foundation in those days for macro physics. But then along comes the idea that atoms, a gas is composed of atoms. Now they have a problem because they already have the law but they have to find how it's consistent with the micro foundation. And this gives rise to statistical thermodynamics which says, oh think of all these atoms as bouncing around in a container and they keep hitting the wall and if I take a circle on the wall, I count the number of hits. Now if I increase the pressure, the hits will get more. If I lower the volume, they're gonna hit more often. So I get, and you can derive this from statistical thermodynamics. I get this relationship which is just log linear. Log of P plus the log of V is equal to a constant plus the log of T. That's what you can derive from statistical thermodynamics. And the law of law says, so we got the micro foundations for a law that we already had, so we didn't derive the law from the micro foundations. We just tried to make it consistent but there is a problem. And the problem is that when quantum mechanics comes along, it says, wait a minute, atoms are not little billiard balls. There are actually things that may or may not exist. Sometimes they're in and sometimes they're out. If you look at them, they may change their state. So how can you make this? And Loughlin says, whenever I give this example, some student's gonna raise my hand and say her hand and say, look, didn't you just tell us that actually atoms don't exist? So you can't tell and he said, yeah, I lied. Because I can't get from atoms in quantum phenomena to Boyle's law. So I assume they're billiard balls and tell you the story we used to tell. But in fact, these are autonomous domains. Now, I'm not arguing that micro is unimportant. I'm saying it's so important that you need to take it seriously. And that means that you have to have a foundation that explains how people really behave and then you have to show how that foundation is consistent, at least with the aggregate patterns. So what we did, thanks to my former student, Amar Raghav, is we created four models in NetLogo, just agent-based simulation. One model, the standard neoclassical model, everybody has a Cobb-Douglas utility function and you have the same utility function so that everyone is a representative agent because you're all alike. And you may start with different points of different incomes, but your utility functions are all the same and we simulate the model. Then a second one, each one of you is neoclassical but you have different utility functions. So your parameters of the Cobb-Douglas are different. Now there's no representative agent. It's been shown formally that you cannot have a representative agent, your agents are different. So we simulate that model. The third one is each one of you randomly chooses from your budget line, from the feasible part of your budget line. So that's Becker's irrational consumer. I prefer to think of it as an impulsive consumer. You notice they're completely different from the rationality assumptions of the first two. And the fourth one is a model we sort of copied or developed from Dosey's work, which is the idea that you have two groups of agents. One are adapt their preference, say inherit the preference structure like a genetic structure, but they adapt their preference structure to those in their social neighborhoods. So they look around, as you can make agents do. You can tell the turtle to look and see what others are consuming and sort of become more like them. And the second set are consumers who innovate, they mutate. So their behavior changes. Now people are a mixture of both of those. Younger people are more on the mutation side, older people more on the imitation side, but you can see that and we simulate that. And the question is, what happens? And the answer is, what happens is, all four models give you exactly the same pattern with some minor variations. This is the demand curve estimated for necessary goods, not estimated, calculated for necessary goods, from four different models. The theoretical model, which is my linear curve, the homogeneous agents, everybody's alike. The heterogeneous agents, everybody's neoclassical, but they're all utility function maximizers and all of that, commodity fetishism. The whimsical one where you change your behavior every iteration and the imitate, innovate where some mutate and others adapt towards. All four of them give you demand curves, they're like this. We tune it to give you the average the same, the average propensity is the same, but the fact is they give you stable average propensity to consume and when you shock it by changing prices, you get this kind of demand curve. The same thing for demand curve for necessaries, the four totally different, inconsistent psychologically will give you the same shape. You cannot distinguish empirically between these, by the way, there'd be no way. If I give you the data and to ask you which is which, you can't do it. Hang on, I just wanna get to the end of this. All you calculate, all the elasticities from the four models and they're all essentially the same. So I'm just gonna skip over that, the income elasticities, the cross price elasticities, the demand, the price elasticities, essentially the same. And so my point is that you have four absolutely opposing models and you get, identical statistically identical models and outcomes and that tells you that you cannot justify your micro from just having some outcomes of the cement and this throws away Friedman's whole attempt to say, well, I start from here and look, I get data that fits, yeah, but I start from here and I get the same thing. So therefore you are responsible for the micro assumptions because another way to put it is, if I take these four models and I look at individual behavior and I say please tell me which corresponds to individual motivation and behavior, it's clear that these three are not right because we have lots of evidence on that from micro observation. We don't need new economics to tell us that, psychology, sociology, anthropology. So you cannot justify the fact that you have some patterns that are observable and you can't justify your foundations from that. Or another way to put it is you meet someone in Rome and you can't tell where they came from because all roads lead to Rome. So you can't tell by just the fact that you met them in Rome, which country or which region or whatever they came from. So that's a philosophical problem but it's a theoretical problem. If we distinguish among these, we have to look at their micro behavior. We have to post test to them to say, is that really a good micro foundation? You can't do it by saying, yeah, I get demand curves. Okay, I cut off a question somewhere. I can't remember, I saw a hand. Question? Assumptions are relevant as long as they explain the theory. So in this case, is it okay if the assumptions are wrong? Well, but then which assumption should we choose? I have four, I can make up a million. So if this is accurate, then surely this is accurate too. So what do we mean accurate? We mean accurate here. Not accurate if I observe people and ask them what they do. If I look at their social context, I look at their reaction, I ask them, do you care about your mother or you care about your country? Then that's gone. So it depends on the reference point. And Samuelson, by the way, says this. He says about Stephen, about Freeman. He said that's really a stupid argument. No physicists, no scientists would accept that argument because that implies that you can carve out a domain where you can test with it. You absolutely refuse to test elsewhere. That's not true. You can't do it in any science. You can't do it in physics and say, oh, I don't like those particles, they're the wrong color. I'm not gonna deal with that. I'm just gonna deal with the part that works for me. I mean, if Newton could have done that, there'd be no Einstein. Because 90% of what Newton says carries over to Einstein. So Newton could say, look, I can explain 90%. What's this stuff about light? Oh, I don't care about that. You can't do it. Not philosophically possible. Yeah. Like, who shows the necessity of my differentiation? Why are they important then, you mean? Yeah. Oh, yeah. We get the same question. Excellent question. It's what I call the Chris Christie question. Chris Christie was the governor of New Jersey. He was at New Dooloo after the whole business with the bridge and he's trying to punish his opponents. Politicians know that it's not the outcome that matters, it's how people feel about the outcome that matters, right? You know that. So if we are asking is not just the demand curves go down, but how do people feel about that? Now, suppose I say to you, Christie says to people, okay, I got caught in a few things in the cookie jar, but look, I've made New Jersey richer because I gave a whole lot of money to some friends of mine and you didn't lose any money. So therefore you should reelect me. They come lynching because of course, people care about other people, even if they don't like other people, they care about what happens. They care about what happens to them, they care about others, they care about outcomes, they care about their nation. So when you want to be a micro economist, you have to talk about how people feel and there's no stupider representation than Pareto optimality because it's designed to eliminate exactly that kind of question. No politician would ever fall for that. Why would you do that? Then I come back to my original question, why would you do that? And the answer is because with these assumptions is the only way you can prove, it can create a fictional vision of capitalism as optimal. It's a religious vision in my view, but that's what you create. And once you introduce connections, then we know from general equilibrium theory all the ideological parts fall out. You can't even generally find a point or a set of points that satisfies general equilibrium. Now that's not the same thing anyway of asking does capitalism have patterns? Does it have structure, a distribution? That we can do with this kind of approach. But you can't make it seem optimal. In fact, I would argue capitalism is surely not optimal. The incentive structure of capitalism is driven by the profit motive. First of all, self-interest is created and driven. And of course it's reasonable of your capitalist. You create something with some poisonous chemicals and then you dump them in the river. What's wrong with that? It lowers costs and it gives you a higher probability of survival. And you can't go around saying, oh that's bad, that's bad, it's easy for you to save. They go out of business if they take care of it. So you have to impose social limits on the incentive structure which comes from their individual incentive. And to do that, you have to go against their profit. All this environmental stuff we just persuade people, don't pollute. Well, Trump knows that that's not the point. The point is to make money. So you wanna go to the Antarctic because you can make a ton of money from there. Forget this pollution stuff. This is the expression of the logic of capital. And if you wanna operate on capitalism you have to understand how people's motivations are including their motivations as capitalists or all people also. Okay, I don't know if I answered your question. Yep, are you been going back to? No, because Lucas Critique is the idea that everybody is super rational. And if I do something, they all anticipate it and negate it or at least adapt to it in some way. I'm not making that argument. I'm saying that people care about outcomes. It doesn't mean that they know what's gonna happen in the future. On the contrary, Lucas himself didn't have a clue. He's on record saying in 2000 something we've solved the problem of business cycles as he was speaking in the ground we're shaking under his feet and Sam Wilson did it before and all of that. So it's not true. The local critique is a critique that claims that somehow people know in advance soon it's going to happen. There's no evidence for this, including Lucas himself. So what's the reason for this ideology? Why this fixation? In my opinion, it's ideological. They need to have a vision of capitalism. That's perfect. And Lucas himself says as well, if we're going to assume rational consumers who have knowledge, then you have to build it into your model. That's valid. But I say, well, why do you want to make that assumption? There'd be no rationale for it at all. That's one more question we're over here. I'm keeping you from your pizza. We're just getting cold. Yes. It's setting the moral framework from the beginning so that it actually creates a setting for how people think and then they don't go back and object the whole structure because that was the moral setting in the beginning. There is no moral economy taught in those schools so that you understand the things that are going to be your foundation. I'm saying something different. I'm not saying that there's a moral framework. I'm saying that people inherit moral frameworks and frequently these moral frameworks are wrong. The role of women in history, the place of women history is part of a moral framework that every culture justifies and then wrong. But you have to understand that that motivates how people behave. And if you're going to change that, you're talking about changing really fundamental assumptions. So the first step is to know what people know. And then we can ask whether we approve of it or not. The South slave owners had a moral framework. They saw themselves as highly moral as a matter of fact. But it's a morality that we don't accept now. And they did then. If I want to explain how a Southerner thinks in that time period, my first step is to find out how they think. And then I have to show how that explains their behavior. It's a different step to say I don't like that framework or I don't like their behavior and I want to attack it but you've got to know it before you attack it in my opinion. Or you should know it before. And anyway, our job as economists is to understand how it operates. And that is not independent of how we feel about it but it is something different. We have to know what the effect on the environment is even though we may be opposed to that effect. We have to know how it comes and why it comes and why it's so persistent. And it's not because they don't understand. It's because we don't understand. It's our problem. To change the world, I started as an aerospace engineer. Now one thing I know for sure, if you want to go from here to Mars, you don't want to just build a rocket, point in the right direction and light the fuse and say I'll figure it out on the way over there. But the future is more complicated than Mars. So you should get the best tools that you can. And even then, most people don't make it to Mars but at least if you want to change the world you're required to understand it in my opinion in a way that understands how it really works rather than to ideologically represent it or misrepresent it. Okay, I never got to the theory of the firm so I'm already like half a lecture behind but I'll try and speed up next time. Thank you.