The Future of MacroeconomicsA Discussion of a paper by John Muellbauer

Prepared for the ECB Colloquium in Honour of Vítor ConstâncioBy

Roger E. A. FarmerUniversity of Warwick and NIESR, Wednesday 16th May 2018

5/12/18 UCLA, University of Warwick, and NIESR. (c) Roger E. A. Farmer 1

Three Themes of John’s Paper1. DSGE models are deeply flawed2. Insights from the ‘information revolution’

have been ignored particularly with regard to financial markets

3. John provides some insights from his own work on forecasting inflation and business cycles

5/12/18 UCLA, University of Warwick, and NIESR. (c) Roger E. A. Farmer 2

Three Themes of my Discussion• How should we incorporate information

theory?• Not just amplification! Continuum of steady

state unemployment rates! i.e. hysteresis!• Forecasting: Is hysteresis due to demand or

supply?

5/12/18 UCLA, University of Warwick, and NIESR. (c) Roger E. A. Farmer 3

The Great Depression

20,000

30,000

40,000

50,000

60,000

70,000

80,000 0

5

10

15

20

25

301929 1930 1931 1932 1933 1934 1935 1936 1937 1938

The Real Value of WealthUnemployment Rate

Unemployment and Wealth During the Great Depression

Shaded areasare recessions

5/12/18 UCLA, University of Warwick, and NIESR. (c) Roger E. A. Farmer 4

The financial markets mattered in 1929!

The Great Recession

12,000

14,000

16,000

18,000

20,000

22,000

24,000

26,000

28,000 4

5

6

7

8

9

10

11

122001 2002 2003 2004 2005 2006 2007 2008 2009 2010

The S&P 500 Measured in Wage UnitsUnemployment Rate

Unemployment and the Stock Market During the Great Recession

Shaded areasare recessions

5/12/18 UCLA, University of Warwick, and NIESR. (c) Roger E. A. Farmer 5

The financial markets mattered in 2008!

Post-War Unemployment and the Stock Market

6

1.2

1.4

1.6

1.8

2.0

2.2

2.4

2.6

2.8

3.0 0.8

1.0

1.2

1.4

1.6

1.8

2.0

2.2

2.4

2.650 55 60 65 70 75 80 85 90 95 00 05 10

The Unemployment Rate (transformed)The Log Ratio of the S&P to GDP

Unemployment and the Stock Market

Shaded areasare recessions

UCLA, University of Warwick, and NIESR. (c) Roger E. A. Farmer5/12/18

Unemployment and the stock market are non-stationary and co-integrated

Wicksell’s Rocking Horse Model

• This picture shows how the New-Keynesian model responds to a shock

• It is an example of a rocking-horse model

5/12/18 UCLA, University of Warwick, and NIESR. (c) Roger E. A. Farmer 7

A VAR converges to a point

5/12/18UCLA, University of Warwick, and NIESR. (c)

Roger E. A. Farmer8

Δ"

Δ#

Δ" = 0

Δ#=0

Conventional models are represented by VARs

The Windy Boat Model

5/12/18UCLA, University of Warwick, and NIESR. (c)

Roger E. A. Farmer9

• This picture shows how model with hysteresis responds to a shock

• It is an example of a windy-boat model

A VECM converges to a set

5/12/18UCLA, University of Warwick, and NIESR. (c)

Roger E. A. Farmer10

Δ"

Δ#

Δ" = Δ#=0

Any point on

the green line

is a possible

equilibrium

We need

models

represented by

VECM’s where

unemployment

may be non-

stationary

This is What Happened in the Data

5/12/18UCLA, University of Warwick, and NIESR. (c)

Roger E. A. Farmer11

• GDP per person still has not returned to trend eight years later after the collapse of Lehmann Brothers

If it Looks Like Duck….

5/12/18 UCLA, University of Warwick, and NIESR. (c) Roger E. A. Farmer 12

…swims like a duck and quacks like a duck…

Conclusion• Yes DSGE models are flawed: • Yes we can learn from the information

revolution but tweaks and frictions are not enough

• Key question for policy makers: Can we affect the long-run unemployment rate through monetary and fiscal policy?

5/12/18 UCLA, University of Warwick, and NIESR. (c) Roger E. A. Farmer 13

Thank You for Listening

Hmm… Perhaps I need to buy his book

5/12/18 UCLA, University of Warwick, and NIESR. (c) Roger E. A. Farmer 14