Three Reasons Why The Fed Should Not Worry Too Much About Inflation Now

Transcription

1 Three Reasons Why The Fed Should Not Worry Too Much About Inflation Now William T. Dickens Northeastern University and The Brookings Institution Peterson Institute for International Economics conference on labor market slack September 24, 2014

2 Why So Bold? As we have already seen slack measures are mixed Some are at or approaching normal levels (short term unemployment, LFP if you believe the more extreme versions of demographic adjustment) But the rest all suggest substantial slack Do any suggest we have overshot?

3 Justifying Boldness (continued) If you believe in downward nominal wage rigidity then you cannot believe in a vertical Phillips curve (and you really should believe in DNWR after what we ve seen the last few years) DNWR implies an downward sloping long run Phillips curve at low rates of inflation as unemployment drops so a sustained uptick in inflation does not mean accelerating inflation any more than high unemployment means deflation!

4 Justifying Boldness Finally The one measure most people are pointing to that indicates imminent inflation is the unemployment rate for those out of work for less than 27 weeks Even if Kruger etc. are right about STU being the right measure for the Phillips curve are we going to accept that a large number of people who are out of work should stay that way? Larry Ball s fine work on boldness vs. timidity in monetary policy suggests that LTU can be made to go away and the natural rate brought down by accepting a point or two more inflation. WE SHOULD DO THAT If WE HAVE TO!

5 How much slack? We have already heard a lot about this and I don t have a lot to add except The version of the Akerlof-Dickens-Perry Phillips curve model with downward nominal wage rigidity that I maintain suggests for a number of reasons that the short term unemployment rate is the best slack measure That model does a great job of tracking inflation forever but has been grossly over predicting inflation for the last 6mo

6

7 You Should Believe in DNWR My model embodies all the current explanations for the lack of deflation over the last 6 years Anchored expectations Shifting NAIRU Wrong slack measure DNWR The only model that works consistently over the post WWII period (and the best fitting and the one with the best out of sample predictions) is a model with DNWR

8 LOTS OF EMPIRICAL WORK SUPPORTING THIS BUT Here is a new picture from Shahriar Sadighi

9

10

11 Percent Earn Total 2013 Rigidity N=

12

13 But Suppose that Larry Ball Has Been Right All Along and the Short Term Don t Count Does the Natural Rate ratchet up? Maybe so, but is this Milton Friedman s Natural Rate? It is hard to put the long term unemployed back to work but is that really a problem with the long term unemployed we have right now?

14 Another Very Important Picture from One of My Graduate Students Rand Ghayad

15

16 Do We Really Believe that Human Capital Depreciates Abruptly at 6mo?

17 Of Course Not! During normal times you don t want to hire someone who has been out of work more than 6mo So employers carry that prejudice into the current period when many very good workers have been out of work for more than 6mo We need a tight labor market so that they have to sample those long term unemployed so we can put them back to work That is worth a few points of inflataion

18 END

19

20

21

22 NAIRU For Short Term Unemployment Only with DNWR (with 90% Confidence Interval) 9 break 1997:4 2001: NAIRU (percent)

23 12 NAIRU For Short Term Unemployment Only (with 90% Confidence Interval) break 1997:4 2001: NAIRU (percent)

24 Figure 19: Monthly Vacancy and Unemployment Rates Using Proportion of White-Collar Unemployed Workers Job openings as a fraction of labor force Unemployed as a fraction of labor force Jan Aug 2009 Sep Dec 2012 Source: CPS and JOLTS. Data are seasonally adjusted.

25 Figure 20: Monthly Vacancy and Unemployment Rates Using Proportion of Blue-Collar Workers Job openings as a fraction of labor force Unemployed as a fraction of labor force Jan Aug 2009 Sep Apr 2013 Source: CPS and JOLTS. Data are seasonally adjusted.

26 Figure 18: Monthly Vacancy and Unemployment Rates Using Unemployed Persons with a Bachelor Degree or Above Job openings as a fraction of labor force Unemployed persons as a fraction of labor force Jan Aug 2009 Sep Apr 2013 Source: CPS and JOLTS. Data are seasonally adjusted.

27 Figure 17: Monthly Vacancy and Unemployment Rates Using Unemployed Persons with a High School Degree Job openings as a fraction of labor force Unemployed as a fraction of labor force Jan Aug 2009 Sep Apr 2013 Source: CPS and JOLTS. Data are seasonally adjusted.

28 Figure 13: Monthly Vacancy and Unemployment Rates Using Age Group Years Job openings as a fraction of labor force Unemployed as a fraction of labor force Jan Aug 2009 Sep Apr 2013 Source: CPS and JOLTS. Data are seasonally adjusted.

29 Figure 15: Monthly Vacancy and Unemployment Rates Using Age Group 55 + Job openings as a fraction of labor force Unemployed as a fraction of labor force Jan Aug 2009 Sep Apr 2013 Source: CPS and JOLTS. Data are seasonally adjusted.

30 Figure 3: Monthly Vacancy & Unemployment Rates Using Unemployed Persons With Duration Less Than 27 Weeks Job openings as a fraction of labor force Short-term unemployed as fraction of labor force Jan Aug 2009 Sep Apr 2013 Source: CPS and JOLTS. Data are seasonally adjusted.

31 Figure 4: Monthly Vacancy & Unemployment Rates Using Unemployed Persons With Duration Greater Than or Equal to 27 Weeks Job openings as a fraction of labor force Long-term unemployed as a fraction of labor force Jan Aug 2009 Sep Apr 2013 Source: CPS and JOLTS. Data are seasonally adjusted.

32 Figure 21: Monthly Vacancy and Unemployment Rates Using Unemployed Job Losers Job openings as a fraction of labor force Unemployed as a fraction of labor force Jan Aug 2009 Sep Apr 2013 Source: CPS and JOLTS. Data are seasonally adjusted.

33 Figure 22: Monthly Vacancy and Unemployment Rates Using Job Leavers, New Entrants, and Re-entrants Job openings as a fraction of labor force Unemployed as a fraction of labor force Jan Aug 2009 Sep Apr 2013 Source: CPS and JOLTS. Data are seasonally adjusted.