Is capitalism a positive-sum or zero-sum game? The answer is: both: Smith and Marx both had a point. However, the extent to which it is either varies from time to time.

For example, from the mid-40s to the mid-70s, high wage growth and full employment were in capitalists' interests. Rising wages sustained aggregate demand not only via consumer spending growth, but also because higher wages gave firms incentives to invest in labour-saving technology.

In the 70s, though, this ceased to be the case. Wage growth then began to squeeze capitalists' profits. The positive-sum game became a zero-sum one, as Marglin and Bhaduri have described.

The solution to this was Thatcherism, or if you prefer neoliberalism. Policies aimed at restoring profit margins by weakening trades unions and the welfare state and creating job insecurity helped to raise productivity, profit rates and growth.

But we might now be back in a phase of a positive-sum game. A recent paper by Servaas Storm argues that secular stagnation is due in large part to "the deliberate creation after 1980, through economic policies, of a structurally low-wage-growth economy." Low wages, and low wage growth deter firms from investing in new technology, and thus causes low productivity growth. This echoes a point recently made by Ben Chu, and is consistent with the fact that the neoliberal era has seen falling productivity growth.

In this sense, low wages aren't just bad for workers, but for capitalists too - because low productivity means falling profit rates (pdf) and low real interest rates and hence low returns on savings. Of course, low wages are good are any individual capitalist. But they are not necessarily good for all.

In this sense, Tories today might be in a similar position to social democrats in the 70s. Back then, Labour was playing a positive-sum game in a zero-sum economy. The result was both an economic crisis and intellectual disarray in the party from which it took years to recover. Today's Tories are playing a zero-sum game in a positive-sum world, and the results are similar: economic crisis and intellectual confusion.

But there's another parallel with then. Thatcher's project of shifting us from positive-sum to zero-sum policies was a dangerous and uncertain one. The same might be true of the shift back. Not all policies to increase wages would necessarily be beneficial. As Servaas says, a general wage rise is not a magic bullet. I have three problems here:

- One reason why higher minimum wages haven't so far had much adverse effect upon employment is that they have diminished employers' monopsony power. There'll come a time, though, when this low-hanging fruit will have been plucked, and so further rises in minimum wage would hurt jobs. Seattle at least might be around this point.

- Many low-wage industries are also low-profit industries; think of hand car-washes or care homes. Higher wages in such industries might not stimulate innovation so much as just close them down.

- With savings low and debt high, households might respond to higher wages by saving more. If so, we'll not get higher aggregate demand and hence incentives to invest. There's a warning here from the 70s. One reason why the positive-sum game broke down then was that workers saved increasing proportions of their wages.

These doubts don't make me side with neoliberalism. They just make me think that wage-led growth is nothing like sufficient.