Monday, November 23, 2015

Automatic Destabilizers

Timothy Taylor has a typically reasonable and interesting post at Conversable Economist today about counter-cyclical federal budget policies - policies that create federal budget deficits during downturns and surpluses during expansions.

There are some issues that I would take with this framing by looking at this with Scott Sumner's monetary offset point of view and also with Arnold Kling's "patterns of sustainable specialization and trade".  First, it seems that many of the effects of deficit spending can be better achieved through monetary policy.  Second, to the extent that counter-cyclical policy is based on temporary spending increases, it seems like these may frequently be an impediment to the adjustments that will pull us into future expansion rather than an aid.

But, leaving all that aside, with regard to automatic stabilizers - ways in which the budget naturally goes into deficit during a downturn - I think there may be more here than meets the eye - or less, as it were.  Here is a graph posted by Taylor, showing automatic stabilizers on the revenue side:

The fall in national income leads to proportionately larger falls in federal revenues.

But, we need to keep in mind what happens in a typical downturn.  Equity holders earn profit by exposing themselves to cyclical fluctuations.  This means that when national income begins to fall, equity is the first victim - the canary in the coalmine, if you will.  So, corporate profits are a relatively small portion of national income, but they are a large portion of the variance in incomes as we move through the business cycle.  Here is a graph comparing corporate profits to labor compensation. (They are on different scales to show proportional changes, because labor compensation is a much larger portion of national income.)  A decline in corporate income, relative to labor compensation is a leading indicator of a coming contraction.

So, what we are really seeing in the automatic stabilizer of federal revenues is the excess volatility of corporate profits.  Corporate assets are forced by competitive pressures to target, via both operating and financial leverage, expected revenue levels.  When there is a shock to national spending, the disequilibrium created by the mismatch of this leverage with actual revenues pushes profits down sharply as a proportion of revenues.

Here is a graph of domestic corporate taxes as a proportion of national income.  Comparing this to Taylor's graph, we can see that most of the change in federal revenues is coming from changing corporate taxes.  This isn't really a stabilizer, because firms are still paying the same tax rates.  Taxes are simply falling with revenues.  The automatic stabilizer isn't measuring a decline in the federal tax burden on corporations.  It is simply measuring the outsized negative economic shock that corporations absorb because of nominal instability.

But, it's actually worse than that.  Here is a graph comparing the effective corporate tax rate (red) to the domestic corporate profit share of domestic income (green).  What we see is that the tax rate is actually highest when profits hit their cyclical lows, and then drops as the economy recovers.  I expect that much of this has to do with the non-deductibility of losses.  In the early part of the contraction, when profits are falling, some corporations take losses which lead to a drop in total corporate profits, but not a drop in total corporate taxes.  Tax rates rise.  Then, as recovery takes hold, corporations can start claiming tax deductions on those losses against their subsequent profits, so effective tax rates in the earliest part of the recovery are especially low.

The non-deductibility of corporate losses is pro-cyclical fiscal policy.  At the beginning of economic contractions, when the incentives for investment and the stability of corporate incomes are most important, our effective corporate tax rates are arranged to rise to cyclical peaks.  Then, after trends have recovered and the process of new expansion is in place, corporate tax rates are arranged to fall to cyclical lows.

We don't have automatic stabilizers.  We have automatic destabilizers.


  1. "the same way that kangaroos fly." Great line.

  2. Apparently Larry Summers is saying that recessions have longer term damage to GDP than expected which would mean your automatic destabilizers have higher damage to long term GDP growth than expected.

    1. Thanks for the link. The part of my story that is probably the hardest to document may be the part that is the most important. That is the monopolistic profits that flow to firms and highly skilled workers because of the limitations of access to networks of highly skilled workers, which allows tech and financial firms and their workers to claim economic rents that would otherwise flow to consumer surplus. Everyone is looking to federal policy for answers to this issue, and some may lie there. But, I think local population limits might explain a lot of it. And, we can only account for a small part of it by measuring returns to real estate, firm profits, or high wage incomes. The wider net we toss to try to measure this, the harder it is to firmly confirm that this housing issue is the source. I wonder how much of the productivity slowdown mystery is explained by this.

    2. Related: Bank of England blog on real-estate-values/biz-collateral/borrowing-ability/investment mechanism.

    3. Interesting. Thanks.

      I feel a little bit like Scott Sumner, that everyone forgot what they knew when this crisis hit. I think the mania we had about housing markets and our error in attributing high prices to excess credit caused us to demand destructive policy. I think before the 2000s, if you would have said that part of the Fed's job in maintaining financial stability was to manage inflation in such a way that home values didn't collapse, most people would have replied, "duh." In fact, this is what led to the great expansion of the late 90s. As the article you linked to points out, the early 90s saw a sharp housing contraction. We had one in the US, too, and the inflation rates of around 4% or so at the time helped to soothe the potential debt and real estate equity disequilibria that would have come from sharp nominal declines. But, in the 2000s, we all decided that nominal stability was dangerous, because it would trigger risky behavior, so, really, we demanded that the Fed manage the money supply for instability. Now, if you would say that the Fed should manage inflation with nominal home prices in mind, to prevent sharp nominal declines in price, the likely reply will be "Are you nuts? That will just encourage risk taking."