Download Charts Related to Labor Market Development

Charts Related to Labor
Market Developments
Robert J. Gordon, Northwestern and
NBER
CBO Advisers Meeting
November 13, 2009
Recent Performance of UI
New Claims as an Indicator

Back in March, discovery of a tight lag
between the cyclical peak in new
claims and the NBER trough
– Four of last five recessions, lag had been
4-6 weeks (1991 -3 weeks)
– This time turned out to be 10 weeks, not
a big miss for this leading indicator

How bad will the jobless recovery be,
according to this indicator?
New UI Claims Relative to
Peak Value, red line is now
Initial Unemployment Claims as a Percentage of Peak Value During Recession,
1967-2009 (4 Week Moving Average)
100%
Mar-75 (6)
90%
Jul-80 (6)
Nov-82 (6)
80%
Mar-91 (-2)
Nov-01 (4)
???-09
70%
peak date
60%
50%
-52 -48 -44 -40 -36 -32 -28 -24 -20 -16 -12 -8
-4
0
4
Weeks From Peak
8
12
16
20
24
28
32
36
40
44
48
But the Labor Force Has
Grown; UI Claims / LF
Initial Unemployment Claims as a Percentage of Labor Force,
1967-2009 (4 Week Moving Average)
1.0%
0.9%
0.8%
0.7%
0.6%
0.5%
0.4%
0.3%
0.2%
0.1%
0.0%
-52 -48 -44 -40 -36 -32 -28 -24 -20 -16 -12 -8 -4
0
4
8 12 16 20 24 28 32 36 40 44 48
Weeks From Peak
What do Buoyant
Productivity Numbers
Imply for Trends?




Output Identity Y = Y/H * H
Per Capita Version Y/N = Y/H * H/N
To make the identity work, Y/H is total
economy output per hour, not NFPB sector
Method of Obtaining Trends:
– Kalman Filter with a Cyclical Adjustment based
on the unemployment gap derived from my
“triangle” inflation equation
– Relation of U gap and GDP gap has shifted
– Results here based on detrending for 1986-2009,
not the alternative of 1962-2009
Kalman GDP Trend based on
Ugap Lags vs. Leads
Figure 2a(2). Kalman Trended Y using alternatively UDEV leads and UDEV lags,
1986:1 - 2009:3
6
5
4
Value
3
C8 Y
Y Trend using UDEV Leads
2
Y Trend using UDEV Lags
zero
1
0
-1
-2
1986
1991
1996
2001
Year
2006
Right Side of the Identity:
Trends for Y/H, H, and H/N
Figure 4a(2). C1LP vs C1 Hours, 1986:1 - 2009:3
Figure 4(2). C1LP vs C1 Hours per Person, 1986:1 - 2009:3
3
3
2.5
2.5
2
2
C1 Hours
1.5
C1 Productivity
Value
Value
1.5
C1 Hours Per Person
1
C1 Productivity
zero
0.5
1
0
0.5
-0.5
0
1986
-1
1991
1996
2001
Year
2006
1986
1991
1996
2001
Year
2006
Conclusions about Trends
in Y, Y/H, H, and H/N



Results reflect a twist compared to previous recessions
– 2007-09 is worse for H/N than it is for output
– Big U gap implies output trend is growing at 3.0 percent
vs. the previous 2.5 consensus
– Of the 3.0, 2.4 is the productivity growth trend and 0.6 is
the growth trend of total economy hours
Robust Y/H trends contingent on continuing decline in H/N of
0.5 percent per year
Qualifications on robust productivity trend of 2.4 in the total
economy, ~2.7 to 2.8 in NFPB sector
– Contingent on an ongoing train wreck in the labor market
– Contingent on this recession’s imbalance between labor
market weakness vs. relative output market strength