Mary Daly (FRBSF), Bart Hobijn (FRBSF, VU Amsterdam, and TI)
DOWNWARD NOMINAL WAGE RIGIDITIES BEND THE PHILLIPS CURVE
Disclaimer: The views expressed in this presentation are those of the authors and do not necessarily reflect those of the Federal Reserve Bank of San Francisco or the Federal Reserve System as a whole.
Daly and Hobijn
DNWR bend the Phillips Curve
1
A primer
PHILLIPS CURVES ARE BENT
Daly and Hobijn
DNWR bend the Phillips Curve
2
Japanese Phillips Curve looks like Japan Japanese Phillips Curve Minus unemployment rate versus 12-month CPI inflation (Jan 1980 - August 2005)
Inflation (Percent) 10 8 6 4 2 0 -2 -4
-7
-6
-5
-4 -3 Minus the unemployment rate
-2
-1
0
Source: Based on Gregor Smith (2006)
Daly and Hobijn
DNWR bend the Phillips Curve
3
Japanese Phillips Curve looks like Japan Japanese Phillips Curve Minus unemployment rate versus 12-month CPI inflation (Jan 1980 - August 2005)
Inflation (Percent) 10 8 6 4 2 0 -2 -4
-7
-6
-5
-4 -3 Minus the unemployment rate
-2
-1
0
Source: Based on Gregor Smith (2006)
Daly and Hobijn
DNWR bend the Phillips Curve
4
Negative unemployment-inflation trade-off Japanese Phillips Curve Unemployment rate versus 12-month CPI inflation (Jan 1980 - August 2005)
Inflation (Percent) 10 8
6 4 2 0 -2 -4
0
1
2
3 4 Unemployment rate
5
6
7
Source: Based on Gregor Smith (2006)
Daly and Hobijn
DNWR bend the Phillips Curve
5
Magnitude of trade-off: Sacrifice ratio Japanese Phillips Curve Unemployment rate versus 12-month CPI inflation (Jan 1980 - August 2005)
Inflation (Percent) 10 8
6 4 2 0 -2 -4
0
1
2
3 4 Unemployment rate
5
6
7
Source: Based on Gregor Smith (2006)
Daly and Hobijn
DNWR bend the Phillips Curve
6
Sacrifice ratio higher at low inflation Japanese Phillips Curve Unemployment rate versus 12-month CPI inflation (Jan 1980 - August 2005)
Inflation (Percent) 10 8
6 4
Sacrifice ratio is higher at low rate of inflation
2 0 -2 -4
0
1
2
3 4 Unemployment rate
5
6
7
Source: Based on Gregor Smith (2006)
Daly and Hobijn
DNWR bend the Phillips Curve
7
Original Phillips Curve bent Britain (1913-1948): Nominal wage growth versus the unemployment rate. (Now known as wage Phillips curve)
Phillips (1958)
Daly and Hobijn
DNWR bend the Phillips Curve
8
Phillips conjectured it was DNWR Phillips (1958):
“…workers are reluctant to offer their services at less than prevailing rates when the demand for labour is low and unemployment is high so that wage rates fall only very slowly.”
Daly and Hobijn
DNWR bend the Phillips Curve
9
Outline
WHAT WE DO AND OUR CONTRIBUTION
Daly and Hobijn
DNWR bend the Phillips Curve
10
What we do in this paper… • Record-high fraction of U.S. workers with wage frozen in aftermath of Great Recession. • Even as unemployment rate has declined, wage growth has continued to slow. • Introduce simple model of monetary policy and downward wage rigidities • Show that dynamics of simple model of downward nominal wage rigidities are qualitatively consistent with facts. Daly and Hobijn
DNWR bend the Phillips Curve
11
Our main contribution is… • Simplify existing model of DNWR Benigno and Ricci (2011)
– Replicate existence of long-run Phillips Curve from other models in the literature. “Inflation greases the wheels of the labor market” Tobin (1972), Akerlof, Dickens, and Perry (1997), Fagan and Messina (2008), Benigno and Ricci (2011)
• Focus on transitional dynamics – Solve for non-linear path in response to a negative demand shock. – Track the evolution of the distribution of real wages along the equilibrium path. Solve non-linear transitional dynamics using extended path method by Fair and Taylor (1983) Daly and Hobijn
DNWR bend the Phillips Curve
12
Part I: Three facts about the U.S. labor market
INDIVIDUAL-LEVEL WAGE CHANGES AND THE U.S. WAGE PHILLIPS CURVE
Daly and Hobijn
DNWR bend the Phillips Curve
13
Fact 1
NON-NORMAL DISTRIBUTION OF LOG WAGE CHANGES
Akerlof, Dickens, and Perry (1996), Kahn (1997), Card and Hyslop (1997), Altonji and Devereux (2000), Lebow, Saks, and Wilson (2003), Gottschalk (2005), Dickens et al. (2007), Elsby (2009), Daly, Hobijn, and Lucking (2012).
Daly and Hobijn
DNWR bend the Phillips Curve
14
Wage cuts are rare Distribution of 12-month change in log wages in 2006 All workers (hourly and salary, job-stayers and job-switchers)
Percent 18 16 14 12 10 8 6 4 2 0
-20
-10
0
10
20
Source: Current Population Survey and author's calculations.
Daly and Hobijn
DNWR bend the Phillips Curve
15
Rarity of wage cuts well-known Distribution of 12-month change in log wages in 2006 All workers (hourly and salary, job-stayers and job-switchers)
Percent 18 16 14
•
Individual-level evidence for the U.S. 12 Akerlof, Dickens, and Perry (1996), Kahn (1997), Card and Hyslop (1997), Altonji and 10 Devereux (2000), Lebow, Saks, and Wilson 8 (2003), Gottschalk (2005), Elsby (2009), Daly, Hobijn, and Lucking (2012). 6
•
Individual-level evidence across countries 4 Dickens et al. (2007). 2
• -20
-10
0
Source: Current Population Survey and author's calculations.
Daly and Hobijn
Survey evidence 0 Kahnemann, Knetsch, and Thaler (1986), 10 20 Bewley (1995, 1999), and Bonin and Radowski (2011)
DNWR bend the Phillips Curve
16
Shift in distribution of wage changes Distribution of 12-month change in log wages All workers (hourly and salary, job-stayers and job-switchers)
Percent 18 16
2011
14 12
2006
10 8 6 4 2 0 -20
-10
0
10
20
Source: Current Population Survey and author's calculations.
Daly and Hobijn
DNWR bend the Phillips Curve
17
Increase of spike at zero Distribution of 12-month change in log wages All workers (hourly and salary, job-stayers and job-switchers)
Percent 18 16
2011
14 12
2006
10 8 6 4 2 0 -20
-10
0
10
20
Source: Current Population Survey and author's calculations.
Daly and Hobijn
DNWR bend the Phillips Curve
18
Compression of wage increases Distribution of 12-month change in log wages All workers (hourly and salary, job-stayers and job-switchers)
Percent 18 16
2011
14 12
2006
10 8 6 4 2 0 -20
-10
0
10
20
Source: Current Population Survey and author's calculations.
Daly and Hobijn
DNWR bend the Phillips Curve
19
Not many more wage cuts Distribution of 12-month change in log wages All workers (hourly and salary, job-stayers and job-switchers)
Percent 18 16
2011
14 12
2006
10 8 6 4 2 0 -20
-10
0
10
20
Source: Current Population Survey and author's calculations.
Daly and Hobijn
DNWR bend the Phillips Curve
20
Fact 2
SPIKE AT ZERO COUNTERCYCLICAL
Card and Hyslop (1997)
Daly and Hobijn
DNWR bend the Phillips Curve
21
Spike increases in recessions Workers reporting same wage as one year prior All types of workers (hourly, salary, and job switchers, and job stayers)
Percent 18
16
14
12
10
8
6 1986
1989
1992
1995
1998
2001
2004
2007
2010
2013
Source: Current Population Survey. 12-month centered moving averages
Daly and Hobijn
DNWR bend the Phillips Curve
22
Record-high spike after Great Recession Workers reporting same wage as one year prior All types of workers (hourly, salary, and job switchers, and job stayers)
Percent 18
16
14
12
10
8
6 1986
1989
1992
1995
1998
2001
2004
2007
2010
2013
Source: Current Population Survey. 12-month centered moving averages
Daly and Hobijn
DNWR bend the Phillips Curve
23
Fact 3
U.S. WAGE PHILLIPS CURVE IS BENT
Phillips (1958), Samuelson and Solow (1960), Galí (2011)
Daly and Hobijn
DNWR bend the Phillips Curve
24
Composite measure of wage growth Four Measures of Nominal Wage Growth Nominal wage growth is 4Q change
Nominal Wage Growth (%) 9 CPH ECI MWE AHE Principal Component
8 7 6 5 4 3 2 1
0 1986
1991
1996
2001
2006
2011
Source:Bureau of Labor Statistics
Daly and Hobijn
DNWR bend the Phillips Curve
25
High unemployment low wage growth Nominal wage growth and unemployment gaps 4-quarter moving average Percentage points 3
Percentage points 6
Nominal wage growth gap
2.5
Unemployment gap
2 1.5
3
1 0.5 0
0
-0.5
-1 -1.5 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012
-3
Source: Bureau of Labor Statistics and authors' calculations
Daly and Hobijn
DNWR bend the Phillips Curve
26
U.S. wage Phillips curve bent U.S. Wage Phillips Curve: 1986-2012 Nominal wage growth is 4Q change
Nominal wage growth gap
3
Before 2007 recession
2
2008
1 Since 2007 recession 2009
2011
0
2010
2012
-1
-2 -2
-1
0
1 2 Unemployment gap
3
4
5
Source: Bureau of Labor Statistics and authors' calculations
Daly and Hobijn
DNWR bend the Phillips Curve
27
Wage growth leveled off when unemployment rose U.S. Wage Phillips Curve: 1986-2012 Nominal wage growth is 4Q change
Nominal wage growth gap
3
Before 2007 recession
2
2008
1 Since 2007 recession 2009
2011
0
2010
2012
-1
-2 -2
-1
0
1 2 Unemployment gap
3
4
5
Source: Bureau of Labor Statistics and authors' calculations
Daly and Hobijn
DNWR bend the Phillips Curve
28
Wage growth decelerated and unemployment declined U.S. Wage Phillips Curve: 1986-2012 Nominal wage growth is 4Q change
Nominal wage growth gap
3
Before 2007 recession
2
2008
1 Since 2007 recession 2009
2011
0
2010
2012
-1
-2 -2
-1
0
1 2 Unemployment gap
3
4
5
Source: Bureau of Labor Statistics and authors' calculations
Daly and Hobijn
DNWR bend the Phillips Curve
29
Similar pattern across recessions U.S. Wage Phillips Curve: 1986-2012 Nominal wage growth is 4Q change
Nominal wage growth gap
3
2 2001 recession
1
2007 recession 0
-1
1990 recession
-2 -2
-1
0
1 2 Unemployment gap
3
4
5
Source: Bureau of Labor Statistics and authors' calculations
Daly and Hobijn
DNWR bend the Phillips Curve
30
Part II: Sketch of model
MODEL OF MONETARY POLICY AND DOWNWARD NOMINAL WAGE RIGIDITIES
Daly and Hobijn
DNWR bend the Phillips Curve
31
Firms • Production function 𝑌𝑡 = 𝐿𝑡 – 𝐿𝑡
Labor composite
Dixit-Stiglitz (1977)
• Perfectly competitive goods market
• Equilibrium: Price equals unit production cost 𝑃𝑡 = 𝑊𝑡 – 𝑊𝑡 Daly and Hobijn
Composite of wages set by workers DNWR bend the Phillips Curve
32
Dixit-Stiglitz aggregates • Continuum of types of workers 𝑖 ∈ 0,1 • Labor aggregate 𝐿𝑡 =
1 𝜂−1 𝜂 𝐿𝑖𝑡 𝑑𝑖 0
𝜂 𝜂−1
, where 𝜂 > 1
• Wage aggregate 1
𝑊𝑡 =
0
1 𝑊𝑖𝑡
• Equilibrium condition 𝑊𝑡 1= = 𝑃𝑡 Daly and Hobijn
1 0
𝑃𝑡 𝑊𝑖𝑡
1 −𝜂−1
𝜂−1
𝑑𝑖
1 −𝜂−1
𝜂−1
𝑑𝑖
1
= 0
DNWR bend the Phillips Curve
1 𝑤𝑖𝑡
1 −𝜂−1
𝜂−1
𝑑𝑖
. 33
Wage setting under DNWR • Wages set by workers Erceg, Henderson, and Levin (2000)
• DNWR: Fixed probability, 𝜆, of a worker not being allowed to adjust wage downwards. Calvo (1983)
• Idiosyncratic shocks to labor supply, 𝑍𝑖𝑡 . Benigno and Ricci (2011)
– Productivity shocks give similar representation of equilibrium dynamics of aggregates. Fagan and Messina (2008) Daly and Hobijn
DNWR bend the Phillips Curve
34
Household • Chooses 𝑌𝑡 ∞
∞ 𝑡=0
1,∞ 𝑖=0,𝑡=0 to maximize 𝛾+1 1 𝛾 𝛾 𝑍𝑖𝑡 𝐿𝑖𝑡 𝑑𝑖 where +1 0
and 𝑊𝑖𝑡 , 𝐿𝑖𝑡
𝑡−1 𝐷 𝑡 − 𝛽 𝑒 𝑠=0 𝑠
𝑡=0
ln 𝑌𝑡 −
𝛾
• Subject to budget constraint
1
𝐴𝑡 = 1 + 𝑖𝑡−1 𝐴𝑡−1 + • … to labor demand curve
1 𝐿𝑖𝑡 = 𝑤𝑖𝑡
𝛾 > 0.
0
𝑊𝑖𝑡 𝐿𝑖𝑡 𝑑𝑖
𝜂
𝐿𝑡
• … path of 𝐷𝑡 and process of 𝑍𝑖𝑡 : 𝑙𝑛 𝑍𝑖𝑡 ~𝑁
𝜎2 − ,𝜎 2
• DNWR constraint Daly and Hobijn
DNWR bend the Phillips Curve
35
Split household’s problem up • Staggered wage setting – What wage would individual household members, 𝑖, set? – How is it affected by DNWR constraint? – Their idiosyncratic consumption risk is insured. Erceg, Henderson, and Levin (2000)
• Savings decision – Determines interest sensitivity of household’s demand. – Consumption Euler equations is IS-curve. Daly and Hobijn
DNWR bend the Phillips Curve
36
Wage-setting by workers under DNWR • Bellman equation for wage-setting decision ∞
𝑉𝑡 𝑤 = 1 − 𝜆 ∞
+𝜆 0
–𝜆 –Ω
0
max Ω 𝑤𝑖𝑡 ; 𝑍𝑖𝑡 , 𝐿𝑡 + 𝛽𝑒 −𝐷𝑡 𝑉𝑡+1 𝑤𝑖𝑡
𝑤𝑖𝑡 ≥0
max Ω 𝑤𝑖𝑡 ; 𝑍𝑖𝑡 , 𝐿𝑡 + 𝛽𝑒 −𝐷𝑡 𝑉𝑡+1 𝑤𝑖𝑡
𝑤𝑖𝑡 ≥𝑤
1 + 𝜋𝑡+1
1 + 𝜋𝑡+1
𝑑𝐹 𝑍𝑖𝑡
𝑑𝐹 𝑍𝑖𝑡 .
Probability worker faces DNWR constraint Difference between utility value of income from working and disutility of working Erceg, Henderson, and Levin (2000)
– 𝑍𝑖𝑡
Idiosyncratic shock to disutility of working. Benigno and Ricci (2011)
Daly and Hobijn
DNWR bend the Phillips Curve
37
Implications of DNWR for wage setting • DNWR constraint: Workers who, in absence of constraint, would have lowered their wages keep them fixed. • Subdued wage increases: Workers who change their wages set them lower than under flexible wages. Elsby (2009), Ball and Mankiw (1994)
Daly and Hobijn
DNWR bend the Phillips Curve
38
Optimal savings decision: IS-curve • IS-Curve: Consumption Euler equation with log preferences 1 1 −𝐷𝑡 = 𝛽𝑒 1 + 𝑟𝑡 𝑌𝑡 𝑌𝑡+1
Daly and Hobijn
DNWR bend the Phillips Curve
39
Very standard policy rule • Monetary policy rule: Standard Taylor rule 1 + 𝜋 𝑌𝑡 𝑖𝑡 = 𝛽 𝑌
𝜑𝑌
1 + 𝜋𝑡 1+𝜋
1+𝜑𝜋
−1
– 𝜑𝑌 = 𝜑𝜋 = 0.25, 𝜋 = 2% (annualized),𝑟 = 2% (annualized) Taylor (1993)
– No zero nominal lower bound on 𝑖𝑡 .
Daly and Hobijn
DNWR bend the Phillips Curve
40
inflation, p
Long-run Phillips curve due to DNWR LRPC
unemployment, u Daly and Hobijn
DNWR bend the Phillips Curve
41
inflation, p
Aggregate demand side: IS and MPR LRPC AD
unemployment, u Daly and Hobijn
DNWR bend the Phillips Curve
42
inflation, p
Recession: Drop in aggregate demand LRPC AD
unemployment, u Daly and Hobijn
DNWR bend the Phillips Curve
43
inflation, p
Results in short-run trade-off LRPC AD
SRPC
unemployment, u Daly and Hobijn
DNWR bend the Phillips Curve
44
inflation, p
Traced out by transitional dynamics LRPC AD
SRPC
Short-run Phillips curve (SRPC): Equilibrium path in response to negative demand shock that generates a recession.
unemployment, u Daly and Hobijn
DNWR bend the Phillips Curve
45
inflation, p
SRPC bent in two ways LRPC AD
SRPC
Short-run Phillips curve (SRPC): 1. Initial response non-linear in size of shock due to DNWR. 2. Bent path back to steady state due to pent up wage deflation.
unemployment, u
Daly and Hobijn
DNWR bend the Phillips Curve
46
Part II: Results
LONG-RUN (STEADY-STATE) AND SHORT-RUN (TRANSITIONAL DYNAMICS) TRADE-OFFS
Daly and Hobijn
DNWR bend the Phillips Curve
47
Steady State
REPLICATE LONG-RUN INFLATION-UNEMPLOYMENT TRADE-OFF
Daly and Hobijn
DNWR bend the Phillips Curve
48
Distortion due to downward wage rigidities Steady-State Density of Log Wage Changes Quarterly changes in log wage under flexible wages and DNWR Frequency (percent) 10
Spike at zero (percent) 70
9 60
DNWR Spike (right axis)
8
50
7
6
40
5 30
4 Flexible
3
20
2 10
DNWR
1 0
0 -10 -9 -8 -7 -6 -5 -4 -3 -2 -1
0 (
1
2
3
4
5
6
7
8
9 10
)
Source: Authors' calculations
Daly and Hobijn
DNWR bend the Phillips Curve
49
Long-run unemployment-inflation trade-off Long-Run Phillips Curve for Parameter Calibration By various degrees of downward nominal wage rigidities
Inflation (annualized rate) 10 9 8
l = 0.8
7 6 l = 0.5
5
l = 0.2 4 3 2 1 0 0
1
2
3 Unemployment rate
4
5
6
Source: Author's calculations
Daly and Hobijn
DNWR bend the Phillips Curve
50
Long-run sacrifice ratio in line with other studies Long-Run Phillips Curve for Parameter Calibration By various degrees of downward nominal wage rigidities
Inflation (annualized rate) 10 9 8
Long-run sacrifice ratio: 10 percentage point l = 0.5 increase inl =inflation target 0.2 leads to 2.2 percentage point decline in natural rate of unemployment. Similar to Akerlof, Dickens, and Perry (1997) 0
1
2
10 percentage points
l = 0.8
7 6 5 4 3 2 1
2.2 percentage points
3 Unemployment rate
0 4
5
6
Source: Author's calculations
Daly and Hobijn
DNWR bend the Phillips Curve
51
Transitional Dynamics
SLOPE AND CURVATURE OF SHORT-RUN PHILLIPS CURVE
Daly and Hobijn
DNWR bend the Phillips Curve
52
Result 1
NON-NORMAL DISTRIBUTION OF LOG WAGE CHANGES
Daly and Hobijn
DNWR bend the Phillips Curve
53
Shift in distribution of wage changes Distribution of quarterly log wage changes In steady state at time
and after shock at
at
Frequency (percent) 10
.
Spike at zero (percent) 70 Spike at 0 after shock at (right axis)
9 8
60
7
50
Spike at 0 in steady state (right axis)
6
40
5 30
4 3
20
Steady State
2 10
After shock at
1 0
0 -10 -9 -8 -7 -6 -5 -4 -3 -2 -1
0
1
2
3
4
5
6
7
8
9 10
Source:
Daly and Hobijn
DNWR bend the Phillips Curve
54
Increase of spike at zero Distribution of quarterly log wage changes In steady state at time
and after shock at
at
Frequency (percent) 10
.
Spike at zero (percent) 70 Spike at 0 after shock at (right axis)
9 8
60
7
50
Spike at 0 in steady state (right axis)
6
40
5 30
4 3
20
Steady State
2 10
After shock at
1 0
0 -10 -9 -8 -7 -6 -5 -4 -3 -2 -1
0
1
2
3
4
5
6
7
8
9 10
Source:
Daly and Hobijn
DNWR bend the Phillips Curve
55
Compression of wage increases Distribution of quarterly log wage changes In steady state at time
and after shock at
at
Frequency (percent) 10
.
Spike at zero (percent) 70 Spike at 0 after shock at (right axis)
9 8
60
7
50
Spike at 0 in steady state (right axis)
6
40
5 30
4 3
20
Steady State
2 10
After shock at
1 0
0 -10 -9 -8 -7 -6 -5 -4 -3 -2 -1
0
1
2
3
4
5
6
7
8
9 10
Source:
Daly and Hobijn
DNWR bend the Phillips Curve
56
Not many more wage cuts Distribution of quarterly log wage changes In steady state at time
and after shock at
at
Frequency (percent) 10
.
Spike at zero (percent) 70 Spike at 0 after shock at (right axis)
9 8
60
7
50
Spike at 0 in steady state (right axis)
6
40
5 30
4 3
20
Steady State
2 10
After shock at
1 0
0 -10 -9 -8 -7 -6 -5 -4 -3 -2 -1
0
1
2
3
4
5
6
7
8
9 10
Source:
Daly and Hobijn
DNWR bend the Phillips Curve
57
This is the data-equivalent Distribution of 12-month change in log wages All workers (hourly and salary, job-stayers and job-switchers)
Percent 18 16
2011
14 12
2006
10 8 6 4 2 0 -20
-10
0
10
20
Source: Current Population Survey and author's calculations.
Daly and Hobijn
DNWR bend the Phillips Curve
58
Result 2
SPIKE AT ZERO COUNTERCYCLICAL
Daly and Hobijn
DNWR bend the Phillips Curve
59
Spike at zero peaks after shock Fraction of workers with same wage as 4 quarters ago After
and under DNWR
Percent 2.5
2
1.5
1
0.5
0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 Daly and Hobijn
DNWR bend the Phillips Curve
60
Remember the data?... Workers reporting same wage as one year prior All types of workers (hourly, salary, and job switchers, and job stayers)
Percent 18
16
14
12
10
8
6 1986
1989
1992
1995
1998
2001
2004
2007
2010
2013
Source: Current Population Survey. 12-month centered moving averages
Daly and Hobijn
DNWR bend the Phillips Curve
61
Result 3
WAGE PHILLIPS CURVE IS BENT
Daly and Hobijn
DNWR bend the Phillips Curve
62
Wage Phillips curve bent Short-run Phillips Curve under DNWR For three different sizes discount rate shocks
Wage inflation (annualized) 3 2 1 0 -1 -2 -3 -4
4.0
5.0
6.0 7.0 Unemployment rate
8.0
9.0
Source: Authors' calculations
Daly and Hobijn
DNWR bend the Phillips Curve
63
Deep recessions more adjustment through unemployment Short-run Phillips Curve under DNWR For three different sizes discount rate shocks
Wage inflation (annualized) 3 2 1 0 -1 -2 -3 -4
4.0
5.0
6.0 7.0 Unemployment rate
8.0
9.0
Source: Authors' calculations
Daly and Hobijn
DNWR bend the Phillips Curve
64
Pent up wage deflation realized during recovery Short-run Phillips Curve under DNWR For three different sizes discount rate shocks
Wage inflation (annualized) 3 2 1 0 -1 -2 -3 -4
4.0
5.0
6.0 7.0 Unemployment rate
8.0
9.0
Source: Authors' calculations
Daly and Hobijn
DNWR bend the Phillips Curve
65
Here is the empirical equivalent U.S. Wage Phillips Curve: 1986-2012 Nominal wage growth is 4Q change
Nominal wage growth gap
3
2 2001 recession
1
2007 recession 0
-1
1990 recession
-2 -2
-1
0
1 2 Unemployment gap
3
4
5
Source: Bureau of Labor Statistics and authors' calculations
Daly and Hobijn
DNWR bend the Phillips Curve
66
Target inflation and shape of curve Short-Run Wage Phillips Curves at different For -0.03 discount rate shock
Wage inflation (annualized) 10
(annualized)
8
6
4
Long run PC (annualized)
2
0
-2
-4 3.0 Daly and Hobijn
4.0
5.0
6.0 Unemployment rate
7.0
DNWR bend the Phillips Curve
8.0
9.0 67
Sacrifice ratio higher under low inflation Short-Run Wage Phillips Curves at different For -0.03 discount rate shock
Wage inflation (annualized) 10
(annualized)
8
6
4
Long run PC (annualized)
2
0
-2
-4 3.0 Daly and Hobijn
4.0
5.0
6.0 Unemployment rate
7.0
DNWR bend the Phillips Curve
8.0
9.0 68
Inflation barely accelerates recovery Short-Run Wage Phillips Curves at different For -0.03 discount rate shock
Wage inflation (annualized) 10
(annualized)
8
6
4
Long run PC (annualized)
2
0
-2
-4 3.0 Daly and Hobijn
4.0
5.0
6.0 Unemployment rate
7.0
DNWR bend the Phillips Curve
8.0
9.0 69
Conclusion • Record-high fraction of U.S. workers with wage frozen in aftermath of Great Recession. • Even as unemployment rate has declined, wage growth has continued to slow. • Dynamics of simple model of downward nominal wage rigidities is qualitatively consistent with facts. • Downward nominal wage rigidities have played a role in shaping dynamics of wage growth and unemployment from mid-2009 through 2012. Daly and Hobijn
DNWR bend the Phillips Curve
70
REFERENCE SLIDES
Not part of the main presentation
Daly and Hobijn
DNWR bend the Phillips Curve
71
Compensation per hour gets revised a lot Compensation per hour: Initial and final releases Annualized quarterly growth
Percent 20 min-max Initial
15
Final 10
5
0
-5
Correlation = 0.19 -10 1997
1999
2001
2003
2005
2007
2009
2011
Source: Bureau of Labor Statistics and ALFRED
Daly and Hobijn
DNWR bend the Phillips Curve
72
Facts not CPS-specific Distribution of 1-year change in log wages (SIPP) All Types of workers (hourly and salary)
Percent 60
2011 wage distribution
50
40 2006 wage distribution 30
20
10
0
-0.5
-0.4
-0.3
-0.2
-0.1
0
0.1
0.2
0.3
0.4
0.5
Source: Survey of Income and Program Participation and author's calculations.
Daly and Hobijn
DNWR bend the Phillips Curve
73
SIPP survey question suspect Workers reporting same wage as one year prior All types of workers (hourly, salary, and job switchers, and job stayers)
Percent 60
SIPP
50
40
30
20 CPS
10
0 1986
1989
1992
1995
1998
2001
2004
2007
2010
2013
Source: Current Population Survey and Survey of Income and Program Participation and author's calculations.
Daly and Hobijn
DNWR bend the Phillips Curve
74