Transcript Crises and Joint Employment-Productivity Dynamics: A

Motivation
Motivation
0 Business cycle theory:
0 any change in labor usage can be decomposed into
0 a movement along a marginal productivity
schedule
0 and a shift of the schedule
0 Mulligan (2009) decomposes this differential for the
current crisis and compares to the previous ones
Intuition
0
0
0
At the „hip” of the cycle two things coincide:
a.
high demand for products
b.
high demand for labour and high labour productivity
Just after the „hip” demand no longer increases => adjustments
necessary
a.
either on labour demand with unchanged demand/productivity
schedules
b.
or on productivity – upward shift of the demand/productivity
schedules
Over longer horizon
0
adjustment of type (a) leads to permament reduction in
employment (till next positive demand shock)
0
adjustment of type (b) implies growing labour productivity =>
efficiency and TFP
Outline
0 Method
0 Data
0 Findings
0 Conclusions
Method
1. Have to find a „hip” in the cycle…
2. … anchor on the „hip” the changes in productivity and
employment
0 how to measure productivity?
0 how to measure employment?
3. At the end we expect rather coherent:
0 shifts OF the demand/productivity schedules and
0 shifts ALONG the demand/productivity schedules
4. We do that for EU countries for as many countries as
there is data
Method – step by step
1. Identify the boom and the bust
2. Find peak and anchor there
3. Compute cummulated change of labour use and
productivity
4. Graph them one against the other
5. Repeat for all countries for all cycles
Method – 1. finding the hip
0 Short term and long term fluctuations - HP filter
Second
0,03
First
0,02
Third
270000
8
250000
?
0,01
6
230000
0
4
210000
-0,01
190000
2
-0,02
-0,03
170000
1996q3
1997q3
1999q3
2000q3
2001q3
GDP yoy
GDP HP filter (lambda=16)
-0,04
-0,05
1995Q01
1998q3
2002q3
2003q3
2004q3
2005q3
2006q3
2007q3
Output gap (HP filter)
2008q3
GDP HP filter (lambda=1600)
GDP HP filter (lambda=160)
0
150000
GDP sa
130000
1997Q01
1999Q01
2001Q01
2003Q01
2005Q01
2007Q01
2009Q01
0 After filtering away the trend – identify peaks
Method – 2. measuring
productivity and employment
0 Employment
0 Ideally: hours worked in economy
0 Reality: not available for any of the EU countries
0 Working/employed/FTE – huge differences
0 Eventually: working, no FTE
0 Labour productivity:
0 Essentially: output per hour worked
0 Reality: no output (GDP) no hours worked (heads)
0 Eventually: GDP/working
Method – 3. link labour
productivity and employment
0 Refresh Mulligan
0 In reality for European
economies?
0 Nice RBC-coherent pattern
Data
Country
Belgium
Czech Republic
Denmark
Estonia
Finland
France
Germany
Hungary
Italy
Latvia
Lithuania
Netherlands
Poland
Spain
United Kingdom
Beginning of sample
1990q1
1995q1
1990q1
1995q1
1990q1
1990q1
1991q2
1995q1
1990q1
1995q1
1995q1
1990q1
1995q1
1990q1
1992q2
End of sample
2009q4
2009q4
2009q4
2009q3
2009q4
2009q3
2009q4
2009q4
2009q2
2009q4
2009q4
2009q3
2009q4
2009q4
2009q3
Total no. of obs.
80
60
80
59
80
79
75
60
78
60
60
79
58
80
70
Results - identified „hips”
Country
No of peaks in sample
Peaks identified
Belgium
3
1992q1, 2000q4
Czech Republic
2
1998q2, 2007q3
Denmark
2
2000q4, 2008q2
Estonia
2
1997q4, 2007q4
Finland
3
1990q4, 1998q2, 2007q4
France
3
1992q1, 2001q1, 2008q1
Germany
3
1995q3, 2002q1, 2008q2
Hungary
2
1998q1, 2008q1
Italy
3
1992q1, 1996q1, 2001q1
Latvia
2
1997q4, 2008q1
Lithuania
2
1997q4, 2008q2
Netherlands
3
1992q1, 2001q1, 2008q1
Poland
2
2000q2, 2008q1
Spain
3
1992q1, 2001q1, 2007q4
United Kingdom
2
1999q3, 2008q1
Results – identified „hips”
Results – identified „hips”
Germany
0.04
Labour productivity
0.03
0.03
2008q1
2001q4
0.02
1995q2
0.02
0.01
0.01
-0.08
-0.05
-0.03
0.00
0.00
-0.01
Labour demand
0.03
Results – groups of countries
0 Analysis reveals few adjustment patterns in the last crisis
0 (a) a canonical RBC;
0 (b) reduced (or absent) change in employment and reductions
in productivity;
0 (c) a contemporaneous reduction in both productivity and
employment (profound recession);
0 (d) a growth or preservation of both employment and
productivity.
0 These reactions are not like in the past.
Group 1 – RBC consistent
0 Note: solid black line (late 2000s);
dashed black line (late 1990s for
Hungary and early 1990s for Spain)
Group 2 – eventually RBC
consistent
Group 3 – productivity drop
Group 4 – both drop
Group 5 – all quiet
Conclusions
0 We applied simple tool to see if recent crisis is RBC type in Europe
0 We also compare it to previous crises
0 Many of the past were RBC in nature
0 Now – only few countries have „adjustment” patterns as economists
would like it
0 Policies to stop employment adjustment at the expense of
productivity
0 Some countries „on the safe side” even if GDP contracts
0 Some countries in deep recession