What's the relationship between capitalists' spending and the profit share? This sounds like an abstruse question, but it is key to the question of whether predistribution and wage-led growth are feasible.

I say this for a simple reason. Measures to raise wages, such as a living wage or stronger collective bargaining would tend to depress the profit share. There are two possible responses to this:

- If workers have a higher propensity to spend than capitalists, aggregate demand would tend to rise. This in turn could encourage capitalists to invest, in anticipation of that demand. If this happens, we'll get wage-led growth.

- If workers save their higher incomes their demand won't rise much. And capitalists might cut their spending in anticipation of this. This might be exacerbated if they regard better conditions for workers as a threat to the business environment.

Theory, then, is ambiguous, as it always is. My chart provides some evidence. It shows the coefficient on a simple regression of three-year real GDP growth on the profit share lagged three years, for rolling five year periods. A negative coefficient indicates periods of wage-led growth - a below-average profit share leads to above-average GDP growth.

You can see that, for most of the last 50 years we have indeed had wage-led growth; the coefficient has been negative. In most five-year periods, lower profit shares did lead to better growth and higher profit shares to lower ones. And this has been true in recent years - including before the crisis - which vindicates Stewart Lansley:

In both the UK and the US, booming profits have been associated with falling investment. This is because the sustained squeeze on wages has created a number of highly damaging economic distortions. It has sucked out demand, encouraged debt-fuelled consumption and raised economic risk.

However, wage-led growth is not assured. In the 80s and late 90s, the coefficient of GDP growth on lagged profits was positive; a high profit share improved business sentiment and unleashed sufficient capital spending growth to raise GDP nicely. And in fact, looking at the wholse sample (since quarterly data begin in 1955), the relationship between the profit share and subsequent GDP growth has been insignificant.

Of course, I'm only looking at two variables here; controlling for other things that affect GDP - which is probably impossible given that there are so many - might yield different results.

I'm not sure what to make of this. You could read it as evidence that wage-led growth is probably feasible. Or you could read it as a sign that it's a high risk strategy, as the link between profit shares and subsequent GDP growth isn't that robust. What makes me fear the latter is the possibility that higher wages might not boost spending, to the extent that they are either offset by cuts in tax credits, or lead to people paying off debt.

There are, though, some things a government could do to mitigate these risks, such as:

- Tax and benefit reform to reduce the marginal withdrawal rates low-paid workers face. A basic income, anyone?

- A greater socialization of investment, to reduce the risk of an investment strike.

- More active fiscal policy, to give capitalists' confidence that aggregate demand will stay high; this might have been one reason why wage-led growth was possible in the 50s and 60s.

Perhaps, then, wage-led growth is feasible if it is accompanied by other policies.