Production | Economics for People with Ha-Joon Chang

Video Statistics and Information

Video
Captions Word Cloud
Reddit Comments
Captions
(intriguing orchestral music) - Today's dominant school of economics, the neoclassical school of economics, is an economics about market exchange. Unfortunately, this means that the other important domain of the economy, namely production, is almost completely neglected. Yeah, so there's some vague theory of production in which some abstract factors of production, factor in labor, are somehow mysteriously combined by some well-known formula called technology, and somehow something comes up at the end. So, there is hardly a theory. And this, I think, is a major problem, because when you think about it, production is where the wealth of nations, so to speak comes from in the end. Adam Smith was very clear about this point, the so-called father of economics. His famous book, The Wealth of Nations, it doesn't start with a market, despite it being the original text of market-based economics. It doesn't even start with a bank or the stock market. It starts with a factory. And at that, a very humble pin factory. So in the first chapter of his Wealth of Nations, Adam Smith says that even the most skilled artisan can produce probably 20 pins a day. But when you go to this pin factory, you have eight, 10 people working together practicing each of these steps. So someone draws the wire, someone cuts it, someone sharpens it, someone attach the thread and so on. And he says, in this way, these guys, eight, 10 of them could be producing literally thousands of pins a day. By making each worker conduct just one step, you are getting rid of this transition time. So if one guy is doing everything, he first has to do it and then rearrange the tools and do the second step. And so he says that is cut. But secondly, the workers become more skilled. They do the same thing over and over again. Then thirdly, he pointed out that this opens the door for mechanization. Each step is very simple. Actually, you don't need a whole set of human dexterity; you can just do it with the machines. Because of this, which he called increasing division of labor, productivity is increased, and that's where the wealth of nations come. So factory was, for him, the source of wealth of nations, the starting point of his whole thesis on the virtues of capitalist market economy. It's so important that, when the Bank of England decided to put Adam Smith on the 20 pounds note, they put the pin factory on the note, not the invisible hand for which is famous. Of course, I mean, I recognize the difficulty of drawing the invisible hand. There has always been this rumor that he used this expression, "invisible hand", only once in 700-800 pages of The Wealth of Nations. I couldn't verify it until Kindle came along and I could actually search for the expression, and it is true. I mean, he uses it only once, somewhere in the middle of the book. So the concept is one of the many metaphors that he use. I mean, it wasn't central to his thesis. Unlike the classical economists and, indeed, some other schools of economics, like the Marxist school and the developmentalist school and the Schumpeterian school, the fact that neoclassical economics ignores production on the place. Production is not just an academic quibble, because you know, a lot of economic policy issues are about production. So at the moment there's a great concern in the UK that productivity has been stagnating in the last decade. They are frantically searching for answers. And who do they ask? Neoclassical economists. So this is a bit like, I don't know, asking someone who has never even entered a kitchen in his life to write a cookbook. A lot of economists are advising on policies that they really don't understand because it's about production. They have no idea what goes in these sites of production, but then they are supposed to come up with answers to this. Production in neoclassical economics is seen mainly as a rather technical relationship. I mean, it's represented as production function in some versions of neoclassical theory. The form is the production function. So is that an entirely technical relationship? And indeed, the interesting thing is that, in neoclassical economics, you don't even need the firm. In theory, it will be a lot more efficient if, every day, capital and labor met in a marketplace and negotiated different combination according to changing relative prices. Why do you tie up certain groups of capital, that is machinery and so on, and certain groups of labor, workers, in a fixed relationship when they might, three months down the road, six months down the road, find that, actually, the relationship would be much more fruitful if they combined with different machines and different people. Of course, in the extended neoclassical theory of new institutional economics, you need the firm because, at that point, very well made by new institutional economists, is that market exchange actually is not costless. It requires what they call transactions cost, so, cost of search, finding out what is out there, cost of bargaining, cost of arranging contracts, cost of enforcing the contract. Because of this transaction cost, you cannot do everything with the market; you need the firm. This set up stable relationships between capital and labor that has more than kind of ephemeral life. So, some fixed relationship. When you think about it, this means that the firm actually exists because of the market. It exists only because market transactions are costly. It doesn't exist because of anything to do with the production process itself. So this idea is ... pithily summarized in the recently popular characterization of the firm within new institutional economics as a nexus of contracts. So it's not really an organization, it's not really a corporate body; it's just a natural group of contracts that happened to be kind of concentrated in certain areas. Herbert Simon, the father of behavioral economics, he's saying that instead of seeing the firm as nodes in a network of transactions, you have to see it as a producer. And once you do that, you actually need to go inside the skin of the firm. You can't just treat it as a shell. The shell has contents, and that is production, and that's what we need to look at. Actually, Herbert Simon, in this article probably, which is probably the last article he published before he passed away, makes a very important observation. He says people talk about the market economy, but actually are we living in market economies? No, we don't, actually; we live in overnight economies. His observation was if a Martian visited the Earth and tried to map Earth economic activity, if he painted all the market transaction in red and all the economic activities going on inside the firm or other organizations like the government green, the Earth will basically look green rather than red. So, on his reckoning, 80% of the US economy is run within organizations; market only represent 20%. So it's extremely important that we understand these organizations, mostly firm. So let's go inside the skins and discuss the nature of production and its evolution over time. The first point about production is that it's not simply a technical process, but it's a social process. It involves people with different goals and different capabilities. And it has internal power structure. The firm is a hierarchical organization. It's a planned organization. Karl Marx got his idea of central planning from the firm. When he was writing about capitalism, there's no planning. But then he observed that, actually, there are these islands of planning emerging in capitalist economy in this sea anarchy of the market. He reckoned why not plan the whole economy as if it's one firm. Being a organization of planning and hierarchy, there's relationships about dominance, power. I talk about these things in my lecture on the rise of the machines, rise of the robots, to put it a bit more dramatically. Secondly, production is a learning process. In standard neoclassical theory, most productive capabilities are conceived as something developed outside the firm. So, schools, universities, training institutes, in research labs, and then people come with it, and then somehow produce output. This is highly misleading, because there's a lot of what I and Antonio in our joint work have called learning in production. So production is not just a process of applying existing knowledge, but also a process of acquiring new knowledge. And a lot of this knowledge created during the production process involves what economists call tacit knowledge. This concept is especially important in the Austrian economics. Basically, the idea is that there is a knowledge that people know but is very difficult to codify and transform into a form that can be transmitted to other people. So it's something like, I mean, I don't know, you miss your mom's cooking, you call her up, and you try to get the recipe for your favorite dish, and then she'll be very vague. She'll be saying, oh, when it just bubbles up, you know, put a little bit of this and little bit of that. It's very difficult to learn this over the phone. You have to learn it with her, watching her cooking. So that's the most basic form of tacit knowledge. The fact that the all of our knowledge are created in the process of production is tacit means that this knowledge is very often specific to the organization, at most to the industry. Now, the interesting implication is that this means that, once an industry is destroyed, because of, I don't know, Chinese competition or whatever, it is almost impossible to put it back again. So actually this has a very important policy implication because a lot of ... policymakers advised by mainstream economists have said, ah, well, you know, de-industrialization, who cares, it's just a consequence of technological change and globalization. These people will find alternative jobs and we may revive the industry if the conditions are right, but it doesn't work like that. Once it's destroyed, it's irreversible, because the knowledge involved in the production process created in the production process is often tacit and very specific to the industry and to the firm. Thirdly, production is also on innovation process. A lot of innovation actually happens in the process of resolving particular problems that arise in the production process. If you look at some neoclassical so-called new growth theory models, some of these models even have a separate sector for R&D, meaning that all the innovation is happening outside production. So that they talk about the knowledge sector. Universities, independent research labs or corporate research labs, I mean, they all produce knowledge, but that's not all of it. A lot of knowledge, a lot of innovation, is created through the production process. And the interesting implication is that this means that because all of this knowledge is in the production process gained by the workers engaged in it, it makes a huge difference if you treat the workers differently. The people who have had the most success with different approaches to incentivizing workers in the production process are the Japanese. They basically understood that, actually, if you give a lot of autonomy and incentives for workers to come up with new ideas, suggestion for improvements, you can actually enormously improve the production process, and may even eventually lead to some radical innovation, whereas in the traditional American-style, mass-production system, the workers were just told to do whatever was in the instruction sheet. As I discussed in the other lecture on the rise of the machines, there were actually some structural impediments for workers doing anything creative. So the Japanese put it on top of his head and said, no, actually, these are people who know how these things work, these are people who can suggest improvement, and gave them a lot of autonomy, and listened to them. So that's how Toyota and other Japanese auto companies beat the American and European companies in the '80s and '90s. And now, I mean, American and European companies actually have adopted a lot of Japanese practices. So, until the beginning of capitalism, basically you had these independent workshops or independent artisans. So I make, I don't know, the violin, I make the cheeses or whatever. And they would do everything in their little corner. What Adam Smith witnessed was the rise of the so-called putting-out system. So now workshops became more specialized. So you produce a pin, you produce the blade, you could sell the pin to different workshops, you could sell the blade to different workshops. So there was this network of workshops that were highly specialized that were feeding into each other. And subsequently, this gave rise to the factory system, which is where they brought these different elements of production together under one roof. So you had this big workshops called a factory that did all the things related to your final product. By the late 19th century, the factories, they started initially with, I don't know, factories employing 20 people, 30 people, and then it expanded to employ 200, 500, 2,000. And then there was a serious problem of financing, because in the early days of capitalism these workshops and factories were owned either by individual capitalists or at most a partnership of a small group of people. That was your arrangement, but then now, when you are trying to create your factory with 2-3,000 people, you just cannot find enough partners to do this, especially because, at the time, being a capitalist, being a partner in a capitalist firm involved unlimited liability. So if your venture fails, you have to pay the creditors with all the money you have, selling your own pots and pans. If you still cannot repay, they might, your creditors might send you to a debtor's prison. It was very risky. So, what was invented at the time was this idea of limited liability. Actually, this first started with colonial ventures; so, the East India Company of Britain, East India Company of the Netherlands, when they are sending these ships to colonize different parts of the world, it was very risky, involved huge amount of money, and people realized that you cannot raise enough money if it meant ruin for everyone who's involved. So they said, okay, you are going to lose only what you have invested in if the venture goes wrong. Yeah, and then a lot of people willing and able to contribute certain money so companies with shares were born. So this idea of limited liability enabled mobilization of large-scale finance to support this large-scale factories. Another legal innovation at the time was this idea of corporate personhood. So the corporation was treated as a separate legal person. So what the corporation did was separated from what the people who work in the corporation, who invested in the corporation did. So that way you could separate the risks involved in this business ventures from the individuals involved in that. The first economist who saw the potential in the corporation was Karl Marx. Adam Smith and many so-called capitalist economists actually were against that limited liability, and for understandable reasons. They said, well, especially Adam Smith said that this is giving license to managers to play with other people's money. What's the incentive of this manager to be careful with this money. This will create what modern economists call moral hazard. I mean, knowing that you are not going to lose out of it, you're going to be lax in managing risk and so on. Karl Marx, however, said no, this is the future. This is the future of capitalism because this will enable capitalists to expand their scale of production without almost limit. Of course, he had an ulterior motive. We thought this will eventually lead to a society dominated by a few monopolies at which point people will see the sense of central planning. He had that ultimate political motive to sing praises of the corporation. But most other allegedly pro-capitalists economists at the time were against this. Limited liability companies ... became the norm from the late 19th century. Before that there was some limited liability companies, like the East India Company, but they were all subject to royal or governmental charters. So there were only a handful of maybe 100, 200 limited liability companies in the whole country, whereas from the mid-19th century, the pioneering countries like Britain started generalizing limited liability so that you could set up limited liability corporations only if you meet some minimum standard and other countries followed. Then, by the late 19th century, this became the norm, and the culmination of this system was the so-called mass production system that emerged in the early 20th century, especially represented by the automobile industry in the United States: General Motors, Ford. Huge factories producing tens of thousands, and later millions, of cars. Basically, cutting down the unit cost by hugely increasing the number of units they produce. But in order to do that, you needed very expensive machines, dedicated machines, and workers dedicated to the use of that machine. And this created a problem with increasing prosperity in the second half of the 20th century, when consumers wanted greater variety. Henry Ford, when he first produced his famous Model T cars, he insisted all cars have to be painted black, so that when one of his own managers told him that consumers might need different colors, Henry Ford famously, oh yeah, it can be any color, as far as it's black. This idea of producing the same thing on a massive scale hit the wall in the '70s and '80s, when people said, no, I want different colors. I mean, of course, by the '70s and '80s there were more than one color of car, but still people wanted greater variety. So in order to meet that increasingly diversifying consumer demand, but without running high cost of small-scale artisanal production, Toyota came up with this brilliant system of production called the lean production system. I mean, it wasn't the only company, but it kind of pioneered it, so sometimes it's called the Toyota system. And the two main elements of the system was flexible automation and just-in-time delivery. So flexible automation meant that machines were not as dedicated as before. The story went that, in the late '80s, early '90s, when Toyota was riding high and American companies were going down the drain, it took General Motors factory two to three days of shutdown to change over from one model to another. In Toyota factories, it took just two, three hours. So Toyota could very flexibly respond to changing demand. It could produce many more varieties of cars than the American companies. And this reduction of, if you like, down time needed for switching between different models, enormously contributed to its competitiveness and cut costs. Another element in the Toyota system was the so-called just-in-time delivery. In the ordinary mass production system, you had, because you are producing on a massive scale, you had huge inventories, hundreds and thousands of units of, I don't know, the engine plug, or the tires or whatever. Toyota said, why do you need to keep all this? You just deliver whatever is needed just before they are used before production, and you can more or less eliminate inventory costs. Of course, this required a very high-quality subcontracting network, because the subcontractors had to make sure that all these parts are without defect, so they had this idea of zero defect movement. All these have to be produced and delivered in time. And in order to do that, Japanese companies invested a lot in their subcontracting companies. So instead of just squeezing them, they would make investment, they would second their engineers and skilled workers to the subcontractors to develop and improve parts together. They would have frequent exchange of opinions. Once they got these two elements right, flexible automation and just-in-time delivery, it was impossible to beat them. So, for a while, car companies in other countries really struggled. Now, a lot of them have adopted Toyota practice so that they can survive. But you know, today, you think that the rise of Toyota was predestined or something, but you have to realize what a great surprising event this was. Back in 1955, the United States produced seven million cars a year. All the 11 or 12, I forget the exact number, Japanese car companies put together produced 70,000 cars. And Toyota, the biggest company, produced 35,000. I mean, this is one of the most significant changes in the history of capitalism and shows the power of organizing production right. Together with the ... Toyota system there was another element in the so-called flexible production system. I mean, of course it wasn't as important as the lean, or Toyota, production system, but this is known as industrial clusters. Geographically compact areas populated by companies doing very similar things. In Italy, you go to certain area, all the companies in the area are producing chairs of some kind. In another area, they're all producing handbags of some kind. Being geographically close to each other, they understand each other, they help each other. They share knowledge, they share workers. If one company becomes too popular and cannot fulfill all its orders, it might ask other companies to make what they make normally. And through that process, they have created this very interesting mixture of competition and corporation that enable these small and medium sized firms to become globally competitive. I mean, they are small, but in some areas they are the world-leading company. There's this Italian company in the region of Emilia-Romania, that's the region near Apollonia, which has world leadership in dialysis machine. So, we may not have heard of this company because they are very specialized and they are not covering big markets like automobile, but you have these companies competing and cooperating with each other in close geographical proximity. And yeah, Emilia-Romania is the most famous region of industrial clusters, but you can also find it in the United States, the so-called Route 128 are near Boston. By cooperating with each other, they are more efficient, because they can do certain things together, things that are too expensive. So they do export marketing together, they do design together, they do research and development together. But then when it comes to making things and developing the techniques, they compete with each other. So these industrial clusters have the mix, the virtues of the old craft system with virtues of new mass production system in this unique way. Since the, really, 1990s, a lot of companies have ... started to de-diversify, if you like. So a lot of companies throughout the late 20th century became very diversified doing lots of different things. Phillips, that transformed itself into a medical equipment company. It doesn't make computers, it doesn't make video players, all the other electronics equipment anymore, because they decided that this is what they are best at and are going to concentrate on it. And, yeah, a lot of things have been outsourced, and then offshore to other countries. This has basically spread the so-called global value chain. So, you might be ... looking at a Toyota car, but then the engine might have been made in Thailand. The chassis might have been made in the Philippines. The shell might have been made in Japan. So you pull all these different things together, similarly to the putting-out system that Adam Smith was observing. So you have all these specialized workshops producing only engines, only chassis, only the bodies, and then you put them together. So there is an interesting reinvention of this old putting-out system, albeit on a global scale. One note of caution is that people often think that this global value chain was invented in the 1990s. No, actually, it has existed at least from the 1950s. So already in clothing industry, in the shoes, toy industry and so on, and also electronics industry, there existed this global value chain from the '50s and '60s. So, Korean companies, Chinese companies got into the electronics industry at the start as the lowest ... assembler in the global value chain led by, say, American TV company, RCA, or semiconductor company like Texas Instrument. And of course, these guys continuously climbed up the value chain, and now they control these chains themselves. I mean, the ultimate irony was that, when RCA went bankrupt, part of it was bought by LG, the Korean electronics giant, which started out as the lowest subcontractor to RCA. So, I mean, this global value chain has existed for at least half a decade, but it has gained prominence in the '90s, because a lot of corporations started to focus on what they perceived to be their core competencies. A corporation create value in the production process through knowledge accumulation and innovation. Once that value is created, that needs to be distributed. Part of this distribution is distribution between capital and labor. But another part is distribution between the insiders of the corporation and the shareholders as the owners. Because it's the property of the shareholders, those who are managing the corporation, the profession managers need to basically serve the interest of the shareholders. They are the agents of shareholders. How do you do that? Of course, you have to maximize profit. Once profit is generated, you can do two things with it. One is to retain it in the corporation and use it to invest or do whatever you want, or you can give away the profit to the shareholders. The dominant form of this used to be dividends, which you all know. But recently, especially since the 1990s, this practice of share buybacks has become very important. Share buybacks is basically companies using their own profit to buy their own shares so that the share prices go up. This gives the shareholders an opportunity to cash in on the rising value of the share. They don't have to sell the shares, but they are given the opportunity. This is done by company using the profit to buy its own shares to prop up the price. Anyway, so in this view, you have a mechanism basically to serve the shareholders, maximize profit, and maximize distribution of profit, both dividends and share buybacks. These managers, according to the orthodoxy, should be internally supervised by a board of directors with a high profession of outsiders to prevent insider collusion. So ideally, it should be 100%. And some companies, yes, that practice this, I mean, General Motors, I don't know about today, but before it went bankrupt, basically all the members of the board of directors were outsiders. Now, I mean, this cuts both ways. I mean, these are people who are somewhat detached from the corporation, so they are not going to get involved in the internal issues. But this also means that usually these are people who have no idea about the car industry. So there's an issue there. But anyway, the view is that basically you prevent the insiders, especially managers and workers, colluding to increase wages or increase worker welfare. Externally, the profession managers will be kept on their toes by the stock market, because there's always the threat of hostile takeover. If you don't generate enough profit, if you don't give away enough of it to the shareholders, shareholders will begin to sell the shares. And then, stock prices will sink, and then someone can buy them up and take over the company and run it in a different way so that you increase profit and give away more of it. So this is today's orthodoxy. Of course, the consequence has been, first of all, that it has squeezed the workers, because in order to maximize profit, one thing you can do is basically to keep the wages down, to sack as many workers as possible. This is why, in this country, in the U.S., when you try to get through some service company, it takes 30 minutes before anyone comes to the phone, because they are running these things with a minimal number of people. The squeeze on the workers have come from both inside the corporation and from outside, because there were a lot of labor market regulation that enabled new practices that squeeze the workers. So especially in countries like the U.S. and the U.K., this shareholder value-oriented management has meant that wage growth has been low, intensity of work has increased, and employment has become more precarious. In another lecture on inequality, mentioned that the medium wage in the Unite States has been stagnant since the 1970s. It's a direct consequence of these changes in corporate governance. In many countries, the profession of agency workers who are employed by some employment agency, not directly by the company, who will be doing more or less the same work as the regular workers but get paid much less and have much lower job security. In many countries, the profession of these workers has increased. Also think of low-level services, like catering and security guarding, and cleaning. This has created tightly squeezed supplier firms using highly exploited workers, including many immigrant workers with extreme job insecurity. So you miss your work once, you are fired kind of arrangement. And with labor market deregulation, new forms of precarious work has emerged, like the famous zero-hours contract in the UK. The company might employ you, but it doesn't have any obligation to give you any work. Zero hours. So you could be employed, in theory, by the company, but you have no work. So as a result of these changes in the corporate governance system, which then changed labor hiring and composition procedures, labor shares in GDP has declined quite substantially, especially in the rich countries. In developed economies in 1980, just of 61% of GDP went to the workers. Developing countries started lower, at about 52.5%, rose a bit, but then it's back to the level. But in the rich countries, the decline has been very clear and consistent. Now, that is not the end of it. I mean, the shareholder value maximizing general approach not only hurts the workers, but the bigger problem is that it hurts the future of the corporations themselves. The trouble is that shareholders are footloose and they can move very quickly, especially with financial deregulation the last few decades and opening of financial markets around the world, now they have so many options they have become very impatient. So for the average period of shareholding in Britain in the 1960s used to be five years. On the eve of the financial crisis, this had fallen to something like seven months. Basically, this is why people have been talking about quarterly capitalism. If you don't produce result in the next quarter, at most two, then they'll sell their shares, there'll be a bid for hostile takeover, you might lose your job. So, what do you do as the manager? You do your best to maximize short-term profit. In order to do that, you, A, have to squeeze the workers, as I've just explained, but, B, more importantly, you should not do anything that has returns in the long run. Basically, you do not invest, because investment has time horizon. If you invest in certain machines, R&D, return comes three, five, even 10 years later. You don't have that time, because you have to survive the next six months, at most next year or two. So basically you minimize investment, squeeze the workers, generate the maximum possible amount of profit, and this is the more important part, you have to give it away massively through dividends, share buybacks. So the share of profit retained by the corporations as a result of these changes has fallen dramatically. So for example, between the 1950s and '70s in the United States, which already had more shareholder-oriented approach to corporate management than other countries, like Japan and Germany, during those days the U.S. corporations were giving away 35-45% of their profit through dividends. Share buybacks were virtually unknown before the 1980s, so this was mostly dividends. The 35-45% going to the shareholders, the remaining 55-65% remaining in the corporation and basically being used for investment. Today, this has changed dramatically. This graph shows the total payout ratio, which is basically share of dividends, denoted as TD, and share buyback or cold share repurchases in this table as a proportion of total profit, which is denoted as net income here. And you see that, basically, today the American corporations are giving away 90% of its profit. They used to give away 35-45%; now it's 90%. Between 2000 and 2008, this used to be almost 100%. They U.K. corporations are in a similar position. Yeah, after the financial crisis, the payout ration has fallen just below 80%, but you look at this absurd column between 2000 and 2008, U.K. corporations are giving away more than what it earned in profit, because the payout ratio was over 100%. In contrast, European corporations have been giving away just under 70%. Throughout the whole of this period, payout ratio of European corporations was 68%. So basically, American and British corporations give away 90% of its profit. No wonder American and British corporations don't have any money to invest. This difference is not accidental. It comes from the differences in their corporate governance structure. In the U.S., in the U.K., corporate governance is basically geared towards maximizing short-term interest of the shareholders. In other countries, it's different because they have different systems. Germany has what is known as the co-determination system, which mandates all large companies, basically companies with more than 2,000 employees to have an extra board called the board of supervision, which is half made up of representatives appointed by the workers. So actually, they have equal number of representatives from labor and capital, and this supervising board doesn't run day-to-day business, but they have the final say on big decisions, like mergers and acquisitions, factory closure, and so on. Now, the chairman of the supervisory board, who has the casting vote, is actually appointed by the management. So if it really comes down to a showdown, then management will always win, because they'll always have +1 vote. But this is not how these supervisory boards have been managed. They usually run on consensus through, of course, negotiation. And this is why hostile takeovers have been so rare in Germany. Because, when you have hostile takeover, it usually leads to job losses, and workers do not like that, so they block all these in this supervisory board. And when there was ... a merger between voter firm and management, the German company in 2002, this was supposed to be the first ever hostile takeover in Germany since 1945. On top of the co-determination system, the Germans have the main bank system, which basically means that the main lending bank also owns shares in your company and act as long-term stakeholder. So in many German companies, when the company needs restructuring, restructuring is usually done at the behest of the main bank, which usually has a seat in the managerial board and has a lot of say in the management of the company. When it comes to small and medium sized enterprises, they have a lot of collaboration with regional financial institutions that are either non-profit seeking cooperative or that are hard owned by the government and therefore act as a semi-public bank. So these cooperative and semi-public banks are very keen to preserve the production and employment in the region, so they try to save the company rather than just disbanding it, strip the assets, sell them, and move on. So all of this together means that the German managers do not need to please short-term shareholders as much as their Anglo-American counterparts and can therefore more willingly invest for the long-term. Yeah, maybe in the future people will begin to look more closely to alternatives to the Anglo-American corporate governance system. As you see in this graph, the American and British corporations have been basically destroying their own future by basically giving away everything they make. They are not investing. How are they going to keep up with evolving technologies and raise productivity? Soon after the financial crisis of 2008, Jake Welch, who actually the former CEO of General Electric, who actually invented the term "shareholder value maximization" admitted that it was the stupidest idea in the world, although, a few years later, he recanted on that. Warren Buffet, one of the leading financiers, has been very critical of this short-term oriented shareholder value maximization system. So, yes, the point is that ... the Anglo-American system, which is considered to be the best practice by many people these days is actually not the best practice. There are lots of different systems, yeah, all with its own limitations and problems, but we need to be open to this existence of diversity and explore different possibilities in terms of corporate governance, because that's, in the end, where our wealth of nations come from.
Info
Channel: New Economic Thinking
Views: 15,759
Rating: 4.9632182 out of 5
Keywords: ha-joon chang, inet, institute for new economic thinking, matthew kulvicki, economics, economics for people
Id: 8iHeh0iRykw
Channel Id: undefined
Length: 51min 36sec (3096 seconds)
Published: Tue Dec 10 2019
Related Videos
Note
Please note that this website is currently a work in progress! Lots of interesting data and statistics to come.