Krugman´s bias Part II

In his NYT column Krugman also says:

In the United States, government spending programs designed to boost the economy are in fact rare — F.D.R.’s New Deal and President Obama’s much smaller Recovery Act are the only big examples. And neither program became permanent — in fact, both were scaled back much too soon.

What this may be indicating is that it´s not a good idea to design government spending programs to boost the economy. Maybe the powers that be get ‘cold feet’ and become frightened of the possible consequences.

Anyway, the logic of the ‘austerity – anti austerity’ debate eludes me. This research from The Brookings Institution has an ‘arithmetic’ or accounting take:

There are more than 2 million unemployed Americans who might have jobs today if not for austerity.

That’s the conclusion of a new study by Michael Greenstone and Adam Looney at the Brookings Institution. In the 46 months since the Great Recession ended, state, local and federal governments have cut about 500,000 jobs. In contrast, in every other U.S. recession since 1970, the government hired approximately 1.7 million people, on average. That means the U.S. is an estimated 2.2 million jobs in the hole.

I don´t think you can talk about the power, or lack thereof, of fiscal stimulus without taking into account what´s going on with monetary policy. Unfortunately this is how the issue is stated, proceeding along the lines of “since we are at the ZLB, monetary policy has lost power so we can only rely on fiscal policy”.

This reminds me of a 1969 debate between Milton Friedman and Walter Heller (Chairman of President Kennedy CEA in 1961-64) on Monetary vs. Fiscal Policy, having as backdrop the 1964 tax cut. According to Friedman:

…So far as I know, there has been no empirical demonstration that the tax cut had any effect on the total flow of income in the US. There has been no demonstration that if monetary policy had been maintained unchanged…the tax cut would have been really expansionary on nominal income…

In order to satisfy my own curiosity I set up the following exercise: I took the recession phase (peak to through) of each postwar cycle (1948, 1953, 1957, 1960, 1969, 1973, 1981, 1990, 2001 and 2007) leaving out the 1980 recession which was quickly followed by the 1981-82 recession. I compare the behavior of three variables: Real Government Purchases (a measure of the fiscal stance), NGDP (a measure of the monetary stance) and Employment (a measure of real activity).

I then took the ‘expansion phase’ (not exactly because I simply used the 15 quarters since the 2007-09 recession ended which is not necessarily the through to peak of the previous cycles) and observed the behavior of the same three variables.

Panel 1 shows the recession phase of the cycles. Note that the vertical axis denotes the percentage accumulation of the variables under scrutiny and the horizontal axis the number of quarters the recession lasted in each cycle. The scales are not the same and in the vertical axis the variations can be large.

Krugman Bias Part II_1

Observe that in the cycles above, despite the wide variations in the fiscal stance (Real Government Purchases) during the downturn, employment (our gauge of the real economy) appears to be much more closely related to what happens to monetary policy (NGDP).

Take the 1948 cycle, for example. There is a strong fiscal expansion (remember this is the immediate post war period of big infrastructure investments, among other things). So there was no lack of fiscal expansion but since monetary policy was contractionary, the real economy ‘suffered’.

The 1957 cycle is a double whammy, with both fiscal and monetary contraction, although fiscal contraction was much bigger.

In the 1960 cycle, despite some fiscal expansion, monetary policy remains ‘put’, so there is some loss of employment.

Krugman Bias Part II_2

Over the cycles above, things are less clear cut. Nevertheless indications are that monetary policy is the major determinant of what happens to employment.

In the 1969 downturn, fiscal policy is somewhat contractionary while monetary policy is ‘reasonably’ expansionary. When monetary expansion weakens towards the end of the downturn employment falters.

In the 1973 downturn, this time associated with a negative supply shock from the rise in oil prices, both monetary policy are significantly expansionary. Employment is supported but falls when, despite a strengthening of the fiscal expansion, monetary expansion is reduced. (Note. The fiscal and monetary expansion in this cycle to try to offset the impact of the oil price shock ended up stocking inflation).

The 1981 cycle needs an explanation because although both monetary policy and fiscal policy appear expansionary, nevertheless employment falls significantly. Although real government purchases rise significantly (mostly defense related), monetary policy is in fact contractionary. In his quest to bring inflation down, the Volcker Fed greatly reduced the growth rate of NGDP. Initially the impact on the real economy was strongly negative.

In 1990, both monetary policy and fiscal policy are ‘weak’, but the recession is short and shallow.

Krugman Bias Part II_3

During the 2001 downturn fiscal expansion cannot offset weak monetary expansion (in fact, monetary policy is contractionary since it´s growing significantly below trend). The result is falling employment.

In 2007 fiscal policy is significantly expansionary during the downturn. Since monetary policy is initially weakly and then highly contractionary employment tumbles more so than in any other downturn. It´s not the often alleged case that “fiscal expansion wasn´t enough” but that the monetary policy practiced was “extremely cruel”. From all the previous charts we can (almost) safely infer that more expansionary fiscal policy wouldn´t have made much difference!

Now for the ‘expansion phase’.

Krugman Bias Part II_4

I´ll be more succinct here. The expansion phase of the 1953 cycle is illustrative. While fiscal policy is mostly contractionary, a highly expansionary monetary policy stokes employment which falters exactly when monetary policy weakens.

In the 1960 expansion we observe the seeds of the high inflation that was to come. Both monetary policy and fiscal policy are expansionary. Employment expands robustly. We are reminded of Friedman´s remarks about the effectiveness of the tax cut (expansionary fiscal policy). Would that have made much difference if monetary policy had not been as exuberant?

Krugman Bias Part II_5

In the 1969 expansion phase although fiscal policy is somewhat contractionary, monetary policy is strongly expansionary. The result would be the high inflation of the 1970s, reinforced by the monetary behavior of the 1973 expansion phase.

The recovery phase of the 1981 cycle is ‘fantastic’ in as much as it was accompanied by a drastic fall in inflation. Fiscal expansion was associated with ‘Star Wars” defense expenditures, but that appears to be a coincidence with results being much the same if those purchases hadn´t materialized.

Krugman Bias Part II_6

Since the recovery began in mid-2009, monetary policy has shown the smallest increase compared to all previous expansions. And this despite the fact that during the 2007 downward phase, it had been the most contractionary. But even so, note that employment is gaining (after the large losses registered in the down phase) even though fiscal policy is becoming increasingly contractionary.

Krugman says the fiscal contraction is the result of ‘austerian hysteria’. That may be so, but the important thing to note is that the economy remains depressed or, more congenially, the recovery remains weak, because monetary policy so ‘desires’ and not because fiscal policy is contracting.


10 thoughts on “Krugman´s bias Part II

  1. My national pride is injured.

    The best economics historian, and analyst of the USA, is Brazilian.

    Is this fair?

  2. ‘In the United States, government spending programs designed to boost the economy are in fact rare — F.D.R.’s New Deal and President Obama’s much smaller Recovery Act are the only big examples.’

    Krugman is forgetting about the big Hoover expansion of govt. spending during the worst of the Depression, 1929-33.

  3. Great post Marcus. Of course the rebuttal from Krugman is “We are at ZLB and your examples are from times when rates are above zero, so they don’t matter.”

  4. Pingback: Is Krugman Ignoring the Evidence on Fiscal Stimulus’s Impact on Employment? « Economics Info

  5. Pingback: Marcus Nunes Again Accues Krugman of Bias | Last Men and OverMen

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.