Does Capitalism Require Growth?

Recently I made a post in response to a passage from David Graeber’s Neoliberalism, or the Bureaucratization of the World.

Capitalism itself – industrial capitalism at least – has had a very brief historical run. During a mere 200 years, however, it has nonetheless shown a remarkable ability to come up with threats to the very existence of the species: first nuclear destruction, now global climate change. There are good reasons to believe it is simply not a viable long-term system: most obviously, because it is premised on the need for continual growth, and economic growth cannot continue forever in a planet with finite resources.

In response, I mused:

Why do so many critiques of capitalism start with “capitalism requires continual growth”, where did this weird idea come from? Maybe capitalism will deliver continuous growth, but to say it requires it seems just straightforwardly wrong.

The post got a lot of attention, and I learned that lots of people basically agree with Graeber’s claim, but understand it in a huge variety of mostly mutually incompatible ways. So rather than letting my replies stand in the deep recesses of back-and-forth tweet threads, I’ll collate responses to the broad categories of answers here, in no particular order.

Can economic growth continue forever on a finite planet?

First to address Graeber’s substantive point. No one can quite agree what’s actually supposed to be growing in the quote, but if we gloss it as “economic growth” in the usual sense, the claim is straightforwardly wrong. Economic growth is not growth in physical output. Real GDP is a measure of value added, in dollar terms. That is to say, in willingness-to-pay terms. The fact that markets align inframarginal buyers’ marginal utilities into an “objective” price shouldn’t obscure the fact that dollar values are fundamentally values in the same subjective sense as any other value.

For this reason, there’s no law of conservation of value, and this basic point is why economics always starts with value theory. Indeed, much economic growth consists in fulfilling values using fewer resources, not more. Many electronics today are far less resource-intensive than their predecessors a few decades ago, and nevertheless satisfy many more values. Even in the aggregate, a great number of economically developed countries have managed strong real economic growth while at the same time reducing carbon emissions. The limit on economic growth is not finite physical resources, but technology and institutions to facilitate the more efficient use of finite physical resources.

Wouldn’t capitalism be zero-sum without economic growth? (1, 2)

Profit depends on growth, and who would make capital investments without profit? (1, 2, 3, 4, 5)

Both of these are fallacies of composition. Profit does not depend on aggregate growth, and capitalism is not necessarily zero-sum without it.

Economic profit is defined as revenues minus costs, where ‘revenues’ refers to a stream of want-satisfaction (not necessarily money), and ‘costs’ refers to all foregone opportunity costs (again not necessarily money). Profit is not an absolute rate of return; it’s always relative to the next best alternative.

If we understand profit this way, it’s clear that one can both make money but fail to make a profit (for example if one pulls in a 1% return on a business venture, but the risk-free market interest rate is 3%), and lose money but make a profit (for example if the real rate of return is -5%, an investment with a -2% ROI is still profitable!)

What sort of world would this be? Clearly there would be some sort of heavy depreciation going on. Perhaps a natural disaster, a pandemic, or an aging population. But as long as you’d rather invest to lose 2% than do nothing and lose 5%, investment is still both profitable and positive-sum, even in a negative-growth world.

Growth is necessary to cover interest payments in a debt-based economy. (1, 2, 3, 4, 5, 6, 7)

There are two separate claims being conflated here, I think.

First: lots of debt has been entered into on the expectation of economic growth. If that growth fails to materialize, there will likely be a big debt crisis and a lot of defaults. This would be bad, and potentially an existential threat to the legitimacy of capitalism.

But this is a different question from whether capitalism is compatible with low growth per se, as opposed to a major expectational shock. Without economic growth, a lot of debt – perhaps most – simply wouldn’t be entered into in the first place. To the extent that current patterns of debt require growth, it’s because expectations of growth made it worthwhile to borrow. Without any expectational shocks, there’s no reason capitalism should inherently lead to the accumulation of unsustainable – or even growth-premised – patterns of debt.

Second: if lenders charge interest, then there needs to be growth in the money supply to ensure the ability to pay back.

This would gloss the claim as being about nominal growth, not real growth. This might be true for certain behaviors of the price level. But there’s no reason why lenders wouldn’t be satisfied with zero nominal interest if prices are falling (which would entail positive real interest rates). Milton Friedman even proposed exactly this! Provided it’s anticipated and adequately adjusted for, and doesn’t jeopardize the economy’s monetization in general (which can happen with sufficient hyperinflation or deflation), in principle there’s no reason why any positive or negative rate of growth of the money supply should be incompatible with capitalism.

This is true of programs like social security. (1, 2)

One could make a sensible claim like “governments of capitalist economies tend to finance benefit schemes in a way that does indeed require continual growth.” To gloss this as “Capitalism requires continual growth” seems like an abuse of categories.

Actually it’s just fiat/fractional-reserve/Keynesianism/neoliberalism that requires continual growth. (1, 2, 3, 4, 5, 6, 7)

Can’t make heads or tails of this one, to the extent it’s different from the previous two. Never comes with an actual argument.

What do we mean by “capitalism” anyway? (1, 2, 3, 4, 5, 6)

Fair point, but the original claim still doesn’t go through no matter whether we understand it as “the ideal type of a free market” (libertarian definition) or “an economic system where economic gain is purposefully pursued by the issue & acquisition of financial capital” (Weberian definition) or “whatever monstrous corporatist hybrid currently exists” (pop definition). The latter definition especially would cast doubt on whether “capitalism” is even the sort of thing one could make any generalizations about.

Because the rate of profit falls, capitalists have to push into new markets to seek return. (1)

This one I think is the closest to the original Marxist contention, and it would understand growth not as economic growth, but as growth in size of the market. Well, the “rate of profit” does tend to get arbitraged to zero over time, but this is a ceteris paribus condition for any individual investment, not a law of history for some aggregate “rate of profit”. Firms can stay ahead of zero-profit equilibrium in any number of ways, not necessarily expanding market size. They can innovate, they can reduce costs, they can experiment with bundling or other ways of increasing willingness to pay, and so on.

“So aren’t capitalists always pushing for growth one way or another” Yes, but now we’re back to considering economic growth in general, on which see the following:

Shareholders/capitalists demand continual growth. (1, 2, 3, 4, 5, 6, 7, 8, 9)

And they don’t always get it. Companies stagnate, shrink, and fail all the time, and maybe shareholders don’t like this, but it’s no problem for capitalism as an economic system.

Interest leads to concentration of wealth, and growth is necessary to manage the redistribution of it. (1, 2)

I think this one also comes from Marx, and the premise just seems straightforwardly factually wrong. In any case, capitalists are mostly debtors in advanced economies, not creditors, and the workers are largely creditors, by virtue of retirement and savings accounts.

The point is just that prioritizing growth-over-all leads to bad outcomes. (1, 2, 3, 4)

On the macro level, this is again a misunderstanding of what economic growth is. Economic growth is not one value among many, it’s the realization of all values.

Now are there values that actually-existing measures of real GDP fail to capture? Of course. This can even happen systematically in cases like quality change, or externalities like environmental degradation. But that’s a measurement critique of GDP, not a systemic critique of economic growth. In practice, these sorts of modifications – while potentially valid – don’t show anything drastically different from regular ol’ GDP.

On the micro level, doesn’t an obsession with growth drive firms to optimize short-term metrics at the expense of long-term sustainability? No, and it’s the point of financial markets to punish exactly that. A company clearcutting its forest to juice quarterly revenue will reduce its capital in present-value terms, and the stock market will punish it. Investors are quite willing to wait a long time, especially with low interest rates, before realizing a positive return, provided they confidently expect it. If for no other reason than that an investor who’s not willing to wait can always profitably sell to an investor who is.

Indeed, one of the biggest factors in exploding CEO pay over the past few decades is the fact that CEOs are compensated in stock more and more over time. This aligns CEO incentives better than if they drew a fixed paycheck. When the bulk of CEO compensation is in stock, short-termism directly and adversely affects CEOs’ own net worth.

So, provided we have reasonably thick financial markets, and barring rent-seeking strategies, growth-over-all is actually the most prosocial and longtermist approach that a firm could take.

Growth is important for the political legitimacy of capitalism. (1, 2, 3)

Plausible, though I don’t think a necessary feature of capitalism, or a clear disadvantage compared to any other system.

Economic growth is a positive feedback loop between growth and the division of labor. So just like growth is self-feeding, regress would be self-feeding too. (1)

The single serious defense of the original claim as stated, I think. And it does seem like the depopulation of the developed world is likely to proceed somewhat along these lines. Even so, the process shouldn’t be taken to refer to short-term GDP movements: clearly the self-feeding process of economic growth isn’t thrown into reverse by recessions or wars, for example. And if regress does become self-feeding, it’s not clear (1) that the end result is an inevitable spiral toward zero, or (2) that capitalism is any worse at handling regress than any other potential system.

Graeber said it’s “premised on the need for” growth, not that it “requires” growth. (1, 2, 3)

“Premised on the need for” is the sort of circumlocution one uses when trying to avoid the actual meaning of one’s claim being nailed down. Judging from the variety of interpretations, Graeber’s sentence was successful in what it seems to have set out to do.