Friday, May 3, 2013

Is the Fed's Able to Offset Austerity? Insights from the Employment Report

The April employment report came out today and is better than expected. The number of new jobs exceeded the median forecast, the previous two months job numbers were revised upward, and the unemployment rate fell to 7.5%. This report is not what one would expect if fiscal austerity were overwhelming the Fed's efforts to shore up the economy. But the report also is not what one would expect if the Fed were unloading both barrels of the gun at the economy. The April employment rate, therefore, reveals both the strength and weakness of the Fed's efforts.

On the first point, Mike Konczal and Paul Krugman claimed earlier this week that the contraction of U.S. fiscal policy in 2013 was trumping the Fed's QE3 program and, in so doing, undermining the views of Market Monetarists like Scott Sumner and me. The employment report and its revisions to the previous months throw some cold water on their claim. But this should not surprise anyone, since fiscal austerity has been happening from 2010 and it has yet to stop the steady progression of nominal GDP (NGDP) growth.

This can be seen in the figures below. The first figure shows that the cyclically adjusted (i.e. structural) budget balance as a percent of potential GDP has been shrinking since 2010. This is the budget balance due to policy changes, not from changes to the economy. It shows fiscal tightening over the past three years:

And what has this fiscal tightening done to aggregate demand (i.e. NGDP) growth since that time? The figure below shows it has done nothing:

This broader 3-year experiment of fiscal policy versus money policy is the one Konczal and Krugman should be examining. Instead, they focus on the first quarter of 2013 and miss the forest for the trees.

With all that said, the stable employment and AD growth is far from what is needed for the economy to reach escape velocity. Therein lies the shortcoming of QE3 and the other, previous asset purchasing programs of the Fed. They have been enough to stabilize growth, but not enough to shore up a robust recovery. And that is frustrating to watch.

This frustration led me to propose earlier this week that the Fed make the size of its monthly asset purchases under QE3 conditional on how fast the economy was reaching the Fed's targets. Ryan Avent came up with a similar proposal. And then the Fed announced later in the week that it was prepared to do something just like this:
The Committee is prepared to increase or reduce the pace of its purchases to maintain appropriate policy accommodation as the outlook for the labor market or inflation changes.
It is as if Fed officials were reading our blog posts! Okay, maybe not. More realistically, they too see the problems with QE3--it is applying the same pressure to the gas pedal irrespective of how the terrain on the road changes--and want to improve it. Unfortunately, the Fed did not get more specific than this statement so we don't know how or when this will happen. Josh Barro thinks it is unlikely it will ever happen. I hope he is wrong, otherwise we face a long journey to full employment.

Update: Robert Waldman replies to the post. Here is my response to Waldman. 


  1. As I see it, the Fed rate and Fed policy are basically designed to provide the economy with enough money (liquidity) so that aggregate demand can allow aggregate supply to reach full capacity.
    But there is enough liquidity. It is just in the wrong place. As I calculate labor income, it has fallen $400 billion since before the crisis (labor share is way down). Capital income on the other hand has risen $500 billion since before the crisis to new highs.
    The Fed is doing its part to provide liquidity. There is enough liquidity "on balance". But the liquidity is not making it to labor. Those with capital have the liquidity. Labor's income is the driving force behind aggregate demand.

    The problem is a failure to have a mechanism that directs the liquidity to labor away from capital. Obviously more labor income would translate into more jobs.

    The Fed is not the problem. In fact, the Fed rate is being made useless by the low labor share of income, which now requires a negative Fed rate to lift liquidity. But since the Fed rate can't go negative, labor's liquidity languishes from a Fed rate with no traction.

    There has to be another way to raise labor's share of income.

  2. "There has to be another way to raise labor's share of income." There is, increase the inflation rate! While not all of the benefit will accrue to labor income, most of it will.

    1. We can't just raise the inflation rate because liquidity in aggregate demand is insufficient. Also if you look at the equation for inflation...
      inflation(price level) = unit labor costs/labor share of income

      If you raise inflation, you either cause labor share to decrease relative to unit labor costs, or you cause unit labor costs to increase. Increasing unit labor costs could trigger a slowdown. Decreasing labor share would worsen the problem for the Fed rate.

      The best way is to raise unit labor costs and labor share together, but to make sure that unit labor costs rise a little faster than labor share. This will give room for aggregate demand to safely raise inflation without upsetting the economy. But it looks to be a careful process at this point because aggregate profit rates are peaking, which means business is more sensitive to changes in unit labor costs. It would have been better to make these changes a year or two ago.

  3. Tax change multipliers used used by private forecasting firms and by government models such as the Federal Reserve’s FRB/US suggest that about half of the ultimate level economic effect of the payroll and income tax increase should be felt by the second quarter. Similarly government purchase multipliers suggest that two thirds of the ultimate level economic effect will be felt during the first three months (in aggregate) of the sequester, which is now more than two months old. See Appendix A for example:

    In November 2012 the CBO estimated that the maximum level employment effect would be a decrease of about 200,000 jobs, 640,000 jobs (80% 0f combined payroll and UI effect of 800,000 jobs lost) and 800,000 jobs for the high income tax increase, payroll tax increase, and sequester respectively:

    In other words, according to these estimates, the sequester should already have decreased employment by over 500,000 jobs relative to baseline, and the tax increases should decrease employment over 400,000 relative to baseline by the next employment report at the latest.

    What happened to the liquidity trap?

  4. How can you say it's done nothing? You don't know that we wouldn't have had a more robust recovery if it weren't for fiscal austerity.

    Unless you want to claim that yesterday's job numbers would have been lower had we not had the sequester-Market Monetarist aren't going as far as claiming that austerity is expansionary-then why is this a defeat for what Sumner calls the "Keynesian multiplier?"

  5. 'There has to be another way to raise labor's share of income.'

    There is, but until Paul Krugman admits to the discentive effects of marginal tax rates of over 100% for low income earners the country has no chance of implementing things that would help. Say, cutting back from 99 weeks of UI benefits to the old 26 weeks. Eliminating both Federal and State minimum wages, repeal of Obamacare, stricter scrutiny of Food Stamp/Medicaid/Disability claims. Reform of state worker's comp laws.

    Until we make it profitable to employ people again, we'll have a lot idle human beings. People like Krugman (an easy target, I know) can't have it both ways.

  6. Say, cutting back from 99 weeks of UI benefits to the old 26 weeks

    Really. Since this an an NGDP blog, shouldn't one compare UI benefits to NGDP.

    What's your point? UI rises in recessions and falls in recoveries relative to NGDP.

  7. That NGDP graph is way to stable to be coincidental. It seems to me the Fed has been convinced that NGDP is the correct target, they just don't believe in trying to go back to the pre-2008 trend. If anyone wants to argue that any policy will have an effect on aggregate demand, they'll need to find a period where aggregate demand is at least a little unstable.