I get lots of commenters complaining that monetary stimulus would not work because it would lead to inflation, which would reduce real wages. This would (they claim) lead workers to buy less stuff. They are confusing real hourly wages with real income. In standard sticky wage models, monetary stimulus will reduce real wages and boost real income. I can’t be sure, but I wonder if Steve Waldman might have made a similar error:
A recession, in the New Keynesian telling, occurs when this stabilization policy is not sufficient. If changes in supply and demand are so great that “sticky downward” prices must fall faster than the targeted rise in the price level, markets won’t clear. If the “sticky downward” price is workers’ wages, then it is employment markets that won’t clear, and we will experience mass joblessness. If this occurs, a cure would be to increase the targeted rate of inflation until real wages fall relative to other goods and services. When real wages fall enough, employment markets will clear again and the recession will end.
In the post-Keynesian story, a depression is driven by an decrease in agents’ willingness or ability to carry debt. Agents “pay for” decreased indebtedness by devoting their income to the purchase of safe assets (including especially their own outstanding debt) rather than spending on real goods and services. Unfortunately, money spent on financial asset purchases does not create income (they are asset swaps), and may not be cycled back into income for producers of real goods and services. So, in aggregate the attempt to reduce indebtedness can lead to a reduction of income that sabotages the attempt to pay down debt. This is the famous “paradox of thrift”. We simultaneously experience unemployment (reduced spending and income to real goods and service providers plus sticky wages means that people get canned) and financial distress (reduced income and fixed debt makes prior debt ever more burdensome). In this story, reducing real wages is not a solution. Real wage reductions might mitigate unemployment temporarily, but they also engender financial distress. Financial distress then causes agents to redouble their efforts to satisfy debts, reducing aggregate income and requiring further reductions in real wages ad infinitum. The only way out of a post-Keynesian depression is to increase real wages relative to the real burden of debt. In the post-Keynesian story, inflation is helpful only if real incomes hold steady, or, at very least, fall more slowly than the real value of prior debt.
That criticism of the New Keynesian model seems slightly misleading, as most versions don’t assume wage stickiness is the key, or that real wages will be countercyclical. But let’s put that issue aside for the moment. It seems to me that the final sentence is a non-sequitur, or perhaps is confusing wages with incomes.
Keynes argued that wage and price stickiness were factors in the transmission of nominal shocks into real output changes, but also famously argued that wage cuts probably wouldn’t help, because they would simply push you deeper into deflation. But that argument has no merit if the central bank is targeting inflation or NGDP (as in Britain today.) In that case wage cuts can increase employment. Now that doesn’t mean that a fall in real wages will necessarily be correlated with higher employment—we’ve known for decades that there is no reliable cyclical relationship between real wages and the business cycle. Indeed I’ve argued that we should stop talking about inflation and real wages entirely, and instead focus on the ratio of hourly wages to NGDP. That ratio rose sharply in the US during 2009, and I’m almost certain the same happened in Britain.
In fact, in every single macro model ever developed (including Austrian, RBC, monetarist, Keynesian, and Marxist) expansionary policy initiatives are only successful if real incomes don’t fall, because real income is the variable you are trying to expand! Now I suppose some readers are thinking that I’m nitpicking, and that Steve obviously meant real wages, not real incomes, in that final sentence. But that won’t work either, as it would make the rest of his claim incorrect. It is real incomes that determine consumption spending, it is real incomes that determine ability to serve debt. Not real hourly wages. Even shifting to an income distribution argument won’t help; as long as the central bank targets inflation of NGDP, it’s almost impossible to tell a story of high unemployment and downward flexible wages.
And yet I oppose policies aimed at reducing wages, as they will not work, and will merely provide a distraction from the need for greater NGDP. So in policy terms I’m right with my “frenemy” Steve Waldman. I just don’t think the real wage patterns observed in the UK tell us anything about the relative validity of New or Post–Keynesian models. I do agree with Richard Williamson, however, that this data suggests supply-side problems play a greater role in the British recession than the US recession. He has a new post showing Krugman looking at the distorting effect of the VAT increase without first accounting for the decrease that occurred slightly earlier. I’ve been watching the British situation with great interest and Williamson is clearly right and Krugman is clearly wrong.
That’s not to say Britain doesn’t also have major demand-side problems, as Britmouse has ably demonstrated.