- Interview by
- David Zachariah
Anwar Shaikh’s Capitalism: Competition, Conflict, Crises (2016) has been widely acclaimed as one of the most important works of economic theory to have come from the Left in many years. Shaikh, a founding member of the Union of Radical Political Economists and the author of many influential essays (including a celebrated introduction to theories of capitalist crisis), has taught economics at New York’s New School for more than four decades.
In his magnum opus, Shaikh draws upon the writings of Adam Smith, David Ricardo, Karl Marx, and John Maynard Keynes to explain a wide range of patterns found in capitalist economies, from wage differentials and unemployment to technical change and the recurring cycles of economic crisis. Capitalism: Competition, Conflict, Crises dispenses with many of the concepts that underpin mainstream economics, but also challenges some of the most influential theories among latter-day Marxists, finding them to be poorly grounded in the work of Marx himself.
David Zachariah recently spoke with Anwar Shaikh about the dynamics of — and limits to — capitalism. Their conversation has been edited for clarity and length.
You initially trained as an engineer. What made you turn toward political economy?
My parents were always progressive. My brother, sister, and I all received a college education, which was very expensive for a Pakistani family. We were treated as equals. My mother was quite a feminist, and my father was also progressive in that respect, so we got a sense that everybody should have equal rights. That gives you some direction in life.
Engineers have a sense that they can fix anything in the world, but of course that’s not true. I began working in Kuwait: my father was stationed there as a Pakistani official. I worked in the desert, where the temperature was brutal. There were Indians, Pakistanis, Palestinians, Egyptians, and Jordanians, all laboring in the heat without any cover, while we engineers could at least retreat to the roasting shade of a cement-block building (which didn’t have any air conditioning).
It seemed to me that this was a deep illustration of the concern I had with the enormous levels of inequality in the world — only in this case there were no budget constraints, since Kuwait was extremely rich. That led me to think more about those questions, and at some point I happened to meet an economist who persuaded me to go into the field. Sometimes life depends on these little things!
You received your economics degree from Columbia University. What were the formative political activities during your time as a student?
I enrolled in 1967, when antiwar activism was in full bloom. There was a strike at the university, organized by Students for a Democratic Society, against a plan to build a university gym at a local park in Harlem that was also used by the African-American community on the other side of the park. I happened to be living and teaching in Harlem at the time, and I felt this plan was entirely inappropriate, so I joined the strike.
Do you think your prior training as an engineer gave you any advantage as an economist?
It gave me an advantage in trying to look at systems as a whole. I started off as an aeronautical engineer: you need to understand that you’re part of a big system, and you also need to understand how it works, if you’re going to operate within that system (and perhaps change it). That goes against the logic of orthodox economics, where you start from the individual elements and try to build an understanding from there.
Right from the beginning of my economics studies, the orthodox description of the world didn’t make sense to me. Coming from Pakistan, and after traveling to Malaysia, Africa, and North America, I just couldn’t accept the idea that people act in total selfishness, disconnected from each other, and are concerned only with commodities.
For most people, solitary confinement is the worst punishment you can imagine, other than direct physical torture. But in orthodox economics, the solitary, confined individual is the ideal subject. It was bizarre!
Let’s turn to some central issues of economic inequality and redistribution. Radical economists have typically explained persistent global inequalities in terms of unequal exchange, monopolistic capitalists, or extra-economic imperial coercion. By contrast, you have argued that the main causal force reproducing such inequalities is capitalist competition itself. Can you explain how your theory differs from the conventional accounts?
When you’re in the Global South, you think of the Center as this entity that has some sort of plan. That’s true to some extent, as the record of imperial powers like Britain makes clear. But the nature of capitalism subjects imperial planners and capitalists themselves to persistent pressures — what Smith, Ricardo, and Marx called the “laws of political economy” — which are beyond their direct control. And the resulting patterns also impose themselves in the Center itself. That led me to think, what are these forces to which even capitalists and their representatives in the state system are subject?
When you look at the history of capitalism you observe many patterns, such as the recurrence of economic depressions, which happen irrespective of anyone’s subjective intentions. Reading Marx gave me the sense that you could have a foundation to explain such things, and go on to explain regional and global inequalities from that base. That’s not to deny the heavy hand of imperialism, but I don’t believe that we need to explain it in terms of monopolization or unequal exchange in the traditional sense.
I spent fifteen or twenty years teaching all three volumes of Marx’s Capital, trying to extract and develop his economic theory and apply it to the modern world. It seemed to me that the conventional account of “monopoly” was grounded in neoclassical theory. I used to be in contact with the Marxian economists Paul Sweezy and Harry Magdoff, who worked just two blocks away from the New School. I remember asking them what they meant by monopoly, and thus by competition?
Their answer was basically that competition is what neoclassical economics says it is. It seemed incredible to me that they would reduce Marx’s notion of competition to that of Milton Friedman. They had relegated competition to some long-distant stage of capitalism — a view that was advanced by the highly influential Marxian economist Rudolf Hilferding at the beginning of the twentieth century.
I believe that the same competitive forces that produce uneven development within national economies also produce regional or transnational inequalities. The difference is that within most advanced national economies, the state is under pressure to intervene in order to mitigate these imbalances, because they’re a threat to the whole system. In a world economy dominated by imperialism, powerful states can choose instead to suppress the struggles against inequality by force, for considerable periods of time.
Your recent work shows that the level and variation of income inequality within nationally organized economies can be predicted from the sources of incomes alone. How so?
This wasn’t entirely my own discovery: it came from the application of work by the physicist Victor Yakovenko and his coauthors, who had started looking at data on inequality. Yakovenko had the clever idea of dividing income earners into two distinct groups: the bottom 90 to 95 percent, and the top segment. It turns out that the income of the bottom segment comes mostly from labor, whereas the top layer derives its income almost entirely from property. They showed that labor income and property income have two distinct distributions, at all times and across all advanced countries.
I once asked Yakovenko how he picked out these two differential patterns in the data and their relation to labor and property income. He replied: “What do you mean? I was educated in Russia, we know about classes!” He also tried to explain these patterns in terms of particle physics — that is, he claimed that they were analogous to particle collisions under certain conditions. To me that made no sense. The relationship between wage labor and capital is certainly a “collision” of sorts, but not like that between colliding particles.
Initially, I expected that the forces giving rise to gender and racial differences would distort the income distribution for women and black people in the US, but that wasn’t the case. Gender and race only influenced the level of income distribution, not its shape. I wanted to explain how this could be so, and I started working with graduate students to derive these patterns from the basic notion in Smith, Ricardo, and Marx that wage and profit rates are subject to a process of equalization.
Workers move or “drift” from low- to high-wage areas. However, they encounter many obstacles along the way. Some make it, others don’t; some go back to where they started, some linger at points in between; others still end up with a wage lower than their previous income, and so on. This dispersion process is known as “diffusion.” Using the principles of drift-diffusion, we showed that the observed distributions of wage and property incomes can be derived theoretically.
Although labor organizations can shift the overall income distribution from profits to wages, differential wage rates still persist across the economy. Many economists hold segmented labor markets responsible for this. Your theory, in contrast, predicts that more capital-intensive firms will tend to have higher wage rates, at any given organizational strength of labor. What are the main reasons for such differentials?
This argument was developed by my former student, Professor Howard Botwinick, who after college became a steel worker. In his subsequent graduate research, he posed the question: “What are the limits for struggles over wage rates?” When you’re organizing, you know there are limits to what you can push for without killing the firm itself. Howard wanted to show that these limits can be derived from a theory of real competition. He summarized his efforts in a brilliant book called Persistent Inequalities.
Howard argued that you must distinguish between low-cost firms who are the price-setters and the other firms who are forced to accept those prices through the pressure of competition itself. Low-cost firms have more room to absorb higher wage costs. Moreover, the lower costs are often achieved through higher capital intensity. This means they have a proportionally lower amount of labor, so if wages rise, the total costs of such firms will tend to rise less than that of the rest.
Redistributive welfare states have often been characterized — from both left and right perspectives — as institutions that “tax the rich.” Your comparative analysis of different postwar welfare states suggests a different picture.
My student Ahmet Tonak and I began to work on this question because left-wing intellectuals — Samuel Bowles and Herbert Gintis, for example — had argued that taxation of the wealthy supported the welfare state. It was a very common assumption on the Left. I was surprised to discover that almost all redistribution took place within the working class itself: higher-income workers are taxed, and those funds are then redistributed to workers at the bottom.
The difference between what workers as a class receive in health, education, income support, etc., and what they pay in taxes, is something we called the “net social wage.” We found that it was quite small in the US, while in Sweden, with its social-democratic welfare state and high social expenditures, the net social wage was effectively zero. That was quite surprising to us!
Mainstream policy debates between Right and Left are often anchored in competing schools of thought: neoclassical supply-side versus Keynesian demand-side economics. Your argument is that profitability of capitalist firms regulates both supply and demand and therefore a “profit-side” view is more appropriate. What do you mean by this?
Keynes first says that investment decisions govern aggregate demand but, unlike many of his followers, he goes on to argue that net profitability — the difference between the interest rate and the rate of return — regulates business investments. If you’re looking to make new investments, you’re seeking opportunities with the highest net rate of return. The interest rate is the minimal rate of return, since it’s what you would make if you simply put your investment funds in the bank.
Now it turns out that Marx says the very same thing: the “rate of profit enterprise” — the difference between the rate of profit and the interest rate — is what motivates investments. So the question then arises: how is Marx’s theory related to the theory of effective demand we can find in the work of Keynes?
I’ve tried to show how one can build a macroeconomic theory on the same foundation as microeconomics, which looks at the behavior of the capitalist firm. First, firms engage in production (create supply) on the basis of short-term profitability. To produce, they must buy raw materials, hire workers, purchase investment goods, and distribute dividends and interest to their owners and lenders. So profit-based decisions to create supply generate the demand for raw materials, and through the payments of wages, dividends, and interest, generate the consumption demand.
At the same time, long-term profitability regulates investment demand. In other words, profitability regulates both supply (production) and demand. Of course, large numbers of firms and consumers do this individually, so aggregate supply and demand only relate to each other through a process of errors and adjustment that I call “turbulent regulation.” Real macroeconomics is therefore neither supply side nor demand side: it is profit side.
Your recent book Capitalism develops a conception of “real” market competition that rejects conventional orthodox and heterodox foundations. Why is this issue so central to your book?
I’ve tried to show that you can explain a wide variety of phenomena from the point of view of competition. Of course, the question is: which idea of competition can you use? The theory of “perfect competition” is quite frankly absurd, and arguably designed to represent capitalism as an ideal social system. When you go beyond that, you still face the question of how the collisions between firms regulate their outcomes.
That’s the theory of real competition. Marx presented it in an elliptical way in Volume III of Capital, but it’s also implicit in the earlier volumes. I’ve elaborated upon it and developed it so as to explain the observed patterns of relative prices, exchange rates, international trade balances, interest rates, prices of bonds and equities, growth and effective demand, employment and unemployment, recurrent crises, and patterns of inequality. I always confront the theory with the facts.
You asked me earlier how being an engineer influences my approach to economic theory. Well, engineers are concerned with explaining empirical phenomena! It’s no good just having a theory of how airplanes fly. If you can’t make them fly in practice, then you’re not a good engineer.
Speaking of theoretical foundations, what do you see in the theories of Adam Smith, David Ricardo, and John Maynard Keynes that complements those of Karl Marx?
Marx acknowledged his debt to Smith and Ricardo. Just look at his theory of competition and of the appropriation of the product by capitalists and landlords, which shares the fundamental elements we can find in Smith and Ricardo.
There’s been a tendency on the part of some Marxists to dismiss the relation between these three thinkers. That means setting aside the forty years of work Marx did on the real patterns of capitalism. Marxists become complacently satisfied with the ideas of exploitation — which they understand as an abstract process, a seriously inadequate conception, in my opinion — and alienation.
But why did Marx then bother with things like the “reserve army of labor,” “circuits of capital,” “schemes of reproduction,” “prices of production,” “differential and absolute rent,” etc.? Couldn’t he have been a “good” Marxist and just stopped at “surplus value,” “exploitation,” and “commodity fetishism,” devoting the rest of his life to politics? In my view the conventional Marxist focus compresses and diminishes Marx’s work to a particular range of topics with which Marxists have become comfortable.
That’s partly because you don’t need to deal with competition and all the complex phenomena to which it gives rise if you start off with the assumption of monopoly. This is the orthodox line within Marxist theory, which I reject.
One reviewer of your book questioned your adoption of the classical-Marxian theory of value, in which labor requirements in production are a source of profit. According to the reviewer, this theory appeared to have no relevance to your economic framework in general. Was he wrong?
In fact, a good portion of the book is built on the labor theory of value. I show that relative prices á la Smith and Ricardo are dominated by labor values. What they called direct and indirect labor time, Marx called abstracted labor time — I’ve written a lot about that before. I also address the issue raised by Marx himself, of the two different sources of aggregate profit: profit based on surplus value and profit based on transfers of wealth and value. If you start from the theory of monopoly power, you don’t have to deal with these issues.
Your analysis of capitalist dynamics is centered on patterns and constraints that emerge from capitalists and their firms. Critics could argue that state institutions and labor organizations merely play a secondary role in your analysis. Consequently, it lacks strategic relevance to political movements engaged in the class struggle. What would you say in response to that point?
In order to talk about state intervention and labor organization, you first have to lay out the laws of gravity with which they have to reckon. If you believe that the system is founded on monopoly — which has become a sacred nostrum of Marxian economics — then it’s all about the power of the state and the power of capital against labor.
From my point of view, nothing — not even the capitalists themselves — has that sort of power, because the rules imposed on labor and capital stem from the creation of profit and the competition of capitals, which Marx specifically links to each other. A state can intervene to redistribute income and oppose both capital and labor. Pushed by the struggles of workers, it can also intervene to construct a welfare system. But these interventions are still fundamentally constrained by their impact on the profitability of firms.
Let me illustrate those limits by reference to Marx’s own argument about the reserve army of labor. He begins by asking: suppose growth is high enough that this reserve army of unemployed workers begins to dry up — what happens then? Labor markets will tighten, and wages will tend to rise, which is good for workers.
But if wages rise relative to productivity, then profitability falls, which accelerates the ongoing mechanization in firms and slows down overall rates of growth. The slowing recruitment of labor through growth, and the rising displacement of workers through mechanization, will then replenish the reserve army. These are feedback effects inherent in capitalist operations themselves.
I’ll give you another classic example. You can create employment by pumping up effective demand through state expenditure. In fact, this is what Hitler and Roosevelt did in the 1930s, and both governments were enormously successful in bringing down unemployment.
This might suggest that all the talk about profitability and the reserve army of labor is irrelevant. But when we look more closely, we can see that war conditions allowed both the Nazi regime and the Roosevelt administration to block the normal feedback effects: wages and prices were not permitted to rise, productivity rose dramatically, and interest rates were kept low. Hence the leap in deficit-financed demand and employment, which gave a significant boost to profitability and investments.
However, in the 1960s and ’70s, similar policies designed to stimulate growth stopped working after a while: profitability fell, unemployment returned, and inflation took off. Indeed, this can be predicted by using Marx’s theory of the reserve army of labor and the growth limits implied by his schemes of economic reproduction.
In other words, if you want to intervene in support of labor, then you have to choose between two options: you must either keep productivity rising faster than real wages (what I call the “Swedish example”), or prevent wages, prices, and interest rates from rising (by suspending the normal functioning of the market). That gives you some breathing room for a while, but if you don’t understand what you’re doing, workers end up paying the price! This kind of analysis explains the parameters or boundaries for struggle.
The history of capitalism has been characterized by long waves of economic growth. Why does this growth exhibit an exponential rate, rather than a slower trend?
Growth in capitalism is a self-feeding process: it’s not simply exponential, but also cyclical. To the extent that the system is fueled by an internal dynamic, you’ll find that it periodically reaches definitive limits. One sign of this is when the cost of gold and other safe assets rises faster than the rest. This is the process that Nikolai Kondratiev first observed, and it’s ultimately regulated by the profitability of capital.
The question is, what fuels the compound growth rate of output? Well, the profit rate is the objective of capital, and regulates the growth rate of the capital stock. As firms compete and grow, they reinvest a portion of their profits, so output will then tend to follow the compound growth rate of the capital stock.
In a world with finite resource constraints, exponential growth can only persist if resource efficiency grows at least as fast as output. Do you think capitalist market economies can overcome this barrier by internal means of technical change?
Orthodox economics tend to view the supply of labor as a fixed and finite resource constraint. I think this view is empirically wrong, since there are always people outside the reserve army of labor who can be brought into play: for example, the recruitment of women workers during World War Two, or migrant workers from across the globe today. Nevertheless, the labor supply does have implications for the form that technological change assumes in particular contexts.
Marx noted that a shortage of cheap labor in the US led to the adoption of stone-breaking machinery. In contrast, there was a seemingly infinite supply of poor Irish workers who could be enlisted to do this backbreaking work in England — not to mention Ireland itself — so the pace of change lagged behind the US. Cases like that illustrate the flexibility and variability of capitalism.
What are the internal limits to the growth of capitalism? Marx’s schemes of reproduction lay this out. The level and trend of profitability regulates both demand and supply; the more surplus value is reinvested, the faster the system can grow. The (abstract) maximum growth rate of capital is the point at which all the surplus is reinvested.
But the ratio of surplus value to capital is simply the profit rate. So the upper limit to growth is set by the profit rate! This maximum growth rate is abstractly possible only when growth rates are equal across different sectors of the economy, which in practice is impossible. Therefore, as the economy approaches maximum growth, bottlenecks start to arise in different sectors and constrain actual growth.
When it comes to finite resources such as fossil fuels, I think capital can overcome such barriers. Take coal, for instance, which was the primary fuel for capitalism for a long time. Capitalist firms adapted by moving to other fuels, long before environmental pressures kicked in. A similar shift to nuclear and solar power would also be feasible, but that depends on state-led development that can act in the collective interest of capitalism even when individual capitalists don’t recognize that interest.
In that sense, I have more confidence in the adaptability of capitalism than most Marxists do. Now that doesn’t mean that capitalists won’t destroy the planet in the process, but that would be a limitation of a different kind.
The waves of economic up- and downturns — the longest waves spanning about forty years each — are in your analysis governed by secular changes in profitability. You argue that profitability, in turn, is determined by three central factors: capacity utilization, class struggle, and technical change. Are the long waves of capitalism primarily driven by technical change that emerges and exhausts itself? What role do the other two factors play?
First of all, capacity utilization plays no major role in the long run, since firms continuously adapt supply to demand by expanding output and adjusting inventories. That’s not a limit. Also, class struggle over wages and working conditions (as opposed to struggle over the existence of capitalism itself) is constrained by the supply of labor, the rate of technical change, the mobility of capital, etc. Therefore it is the level and direction of profitability that regulates the long-term patterns of capitalist development.
Building upon Marx’s theory of unemployment, you have argued that unemployment cannot be eliminated under capitalism in a stable manner. That is, if unemployment declines, the increased bargaining power of workers affects profitability and the willingness of capitalists to invest. Unemployment will then return, unless wage demands are kept below the rate of productivity growth. Such wage restraint could either come about through intensified class struggle by the ruling class, or through highly centralized collective bargaining, such as was formerly pursued by the Swedish trade-union movement. How feasible do you think the latter strategy would be in an advanced capitalist economy today?
Postwar Sweden and Nazi Germany exemplified the two opposing strategies for combating unemployment, through voluntary coordination or brute coercion. All of my theoretical work is based on careful study of the empirical evidence — the same approach taken by Smith, Ricardo, Marx, and Keynes. It’s been a while since I had a look at the Swedish case, but I would like to go back to it with two things in mind: one is how Sweden tackled growth, output, and employment from the point of view of real competition; the other is how it dealt with transfers within the working class.
Your theory of unemployment runs counter to the advocates of Modern Money Theory (MMT), who argue that the state can expand economic output and eliminate unemployment merely by creating and spending state money. The only limitation in this scenario would be price inflation on goods and services, which according to MMT advocates could get out of hand once full employment has been achieved.
Leaving aside the implication of coordinated wage discipline to sustain such a trajectory, your analysis suggests that increased growth can raise cost-driven inflation well before the point of full employment is reached. Why do you reach such different conclusions?
MMT’s core insight, as many people have observed, is unfortunately not new: when you have fiat money, the state can print as much as it wants. This is not a “modern” discovery but rather an invention of the North American settler colonists, who found that they couldn’t buy things because they had no gold and were prevented by Great Britain from obtaining it.
So they invented a process whereby they convinced people to accept printed money on the basis that it was convertible into gold at some future point in time. This also led to one of the first hyper-inflations in history. J. K. Galbraith wrote a beautiful book called Money: Whence It Came, Where It Went. I have a chapter in my book that provides a critique of MMT, with respect not only to the origins of money but also to its operations and effects.
Secondly, over the years MMT has changed its theory of inflation. Initially, it relied on the standard monopoly-capital notion that prices are created through stable mark-ups on money wages. From this perspective, inflation is a product of rising wages. Subsequently, the MMT school seems to have reverted to the Keynesian notion that prices only rise when increased demand meets supply that is constrained by full employment.
Now, as I show in my book, that simply doesn’t work empirically! So you need to ask what sets the limit to expansion, and that leads me back to the internal limits to growth set by profitability that we discussed before: if the profit rate is rising, then the growth rate also has more room to rise before it starts to reach these limits.
On the other hand, if the profit rate is falling, which it certainly was in the 1970s, then the limits become tighter, which leads to inflation if aggregate demand is continually being stimulated. So when you use state spending and credit — which are in principle unlimited — to pump up the economy, you start hitting limits to growth even if there is persistent unemployment. Moreover, the drying up of the reserve army of labor affects profitability in turn, which means those limits are even tighter.
The Future of Political Economy
How would you describe the current state of Marxian political economy, especially compared to when you were a graduate student?
When I was a graduate student, it was expanding into a lot of different topics, like imperialism and military spending, exploitation, uneven development — very crucial questions. On the other hand, it had already gotten stuck on the foundational idea of monopoly. It seemed to me that I had to explain the same things — which I’m certainly not finished doing — from a different foundation. The more I read Marx, the more distant his notion of competition seemed from that of neoclassical economics. In the process, my dissatisfaction with the tenets of existing Marxian political economy grew.
I think this field stagnated for quite some time and largely retreated to a few well-worn tropes. You can see that the resurgence of neoliberal thought suppressed much of the energy and interest that was stimulated by people like Sweezy, Magdoff, Paul Baran, Ronald Meek, Samir Amin, Celso Furtado, and others. Nevertheless, the fundamental questions seem to be coming back.
What areas of research in political economy do you think are strategically most important for socialist movements today?
Well, I’m sure that what is strategically most important varies across countries and historical moments. But I do argue that a section of the socialist movement should try to find a coherent foundation for strategy, and that the theory of monopoly is not the foundation it needs, since that leaves so many things open and unexplained.
I remember going to a conference organized by economists from the Caribbean and Latin America. The resurgence of competition shocked them, since they had taken it for granted that the global economy was run by monopolists in the Center. They had thought the US ruled the world, but now it was experiencing trouble because it couldn’t compete with the likes of Japan, Korea, China, Vietnam, Malaysia, etc.
For me, it’s important that the US can’t beat the lower costs of the Chinese economy, even though it has tremendous military power. Therefore you shouldn’t assume that the power in the Center is a power exercised over capitalism — rather it is a power of capitalism. It is a mistake to associate all power with political power. Radical economics often seem to vest all power in either state or capital, but I’ve always argued that profit has power over both, because it enforces their limits.
At one point, a group of us sat down and read the platforms of all of the major left parties in the US during the twentieth century. What we were looking for was their economics — that is, how do they understand the functioning of the capitalist economy (a question that is not always posed in political platforms!). We found that in almost every case, the implicit economic theory of these parties was either Keynesian or Post-Keynesian. That’s most obvious when you see the influence of Baran and Sweezy, who built on the work of Michał Kalecki.
What advice would you give young socialists who would like to study political economy?
My advice is to take Marx — and his predecessors — more seriously. Marx did not claim that he was independent of Smith and Ricardo, as should be clear from his Theories of Surplus Value. He read everything, not to mention all the data about industries, wages, etc.
I recently sneaked out of a conference in Manchester to go to the Chetham Library, one of the oldest libraries in the English-speaking world. There you can find a table with a small plaque that says, this is where Marx and Engels sat when they were writing the Manifesto. I have to tell you, that’s the closest I’ve ever come to a religious experience, because I could feel the power of these two young people who dedicated themselves to changing the world!
But how did they do it? They studied capitalism, which was the dominant force in that world, then as now, as well as studying other countries. These two young people were just unbelievable characters — they read everything! They weren’t doing it for entertainment, but for a scientifically grounded understanding of the world. They founded an analysis in a way that others had not done before. We should take the economic aspect of their legacy more seriously.