by Philip Pilkington
I’ve got quite a response to my last piece on the Job Guarantee program and it’s possible influence on wage-price spirals. Some of the kickback I received is, I think, based on a misunderstanding. A few people seemed to think that the JG program itself might induce a wage-price spiral. This is not what I was saying at all.
What I was really saying is that a JG program, as it provided continuous full employment, would lead to a sharp increase in worker bargaining power. If an external shock — like an oil price shock or a substantial currency devaluation — caused inflation to bounce workers may then try to push the costs of the inflation onto capitalists through increases in wages. The capitalists may then try to push these costs back onto the workers and other capitalists by raising prices. So, the causality runs,
JG Program => Tight Labour Markets => Increased Worker Bargaining Power => Potential Tinderbox Scenario Where a Wage-Price Spiral May Occur
Now, Neil Wilson said that I have no evidence that this would happen. Well, as with all applied economics I can never prove this beyond a shadow of a doubt. I am engaged in crystal ball-gazing to some extent here. But I can show historical correlations that imply that the mechanisms that I outline have the potential to do what I say that they might do.
First, let’s look at the relationship between labour markets and unemployment. As Nate Silver has shown there is a strong negative correlation between the unemployment rate and support for unions. You can see this in the regression plot posted below.
For those not use to reading regression plots, the closer the dots (observations) bundle around the line the more correlation there is between the two variables. An upward-sloping line indicates positive correlation while a downward-sloping line indicates negative correlation.
What the graph shows is that the higher unemployment is, the lower is support for unions. Conversely we can say that the lower unemployment then the higher the support for unions. So, if the JG program wipes out unemployment altogether we can say that there is a good chance that support for unions goes up. This is the increase in worker bargaining power that I outlined in the causal argument above.
But what about the theory that increased worker bargaining power leads to a potential for wage-price inflation? Well, first we must ask: how would this occur? The answer to that question is simple: workers would engage in strikes to try to get pay hikes. Do we have evidence to support this? Of course we do! Matias Vernengo has laid the whole thing out in a post here. But the key graph is probably this one,
If you don’t believe me that the increase in strike is strongly correlated with unionisation rates check out this graph.
As Vernengo notes,
That’s is why higher commodity prices in the 1970s, including the oil shocks, led to high inflation back then, but has had a marginal impact this time around. This also suggests that the low inflationary pressures in recent times – Bernanke’s Great Moderation – have less to do with Central Bank ‘credible’ policies than with the attack on unions and workers’ rights.
Indeed!
So, let’s recap. The causality runs like this,
JG Program => Zero Unemployment => Support For Unions => Unionisation => Potential For Wage-Price Inflation
Now, again I am saying: why risk this? If anything might unwind a JG program after it was put in place it would be inflation. It would surely be one of the first things on the board for cuts. Indeed, it would probably be pointed to as the cause of inflation by conservative economists. They would say,
“Oh, but we told you that silly program would cause demand to rise too high and spark inflation”.
And so it would likely be dismantled even if the source of the inflation was not the JG itself.
Why not instead just include measures in the JG package that minimise the risk of the above scenario playing out? What have we got to lose by tacking some combination of TIPS and MAPS onto the JG program? I really do not see a strong argument against considering this.