Mary OMahony

1
Mary OMahony

• Senior Research Fellow
• National Institute of Economic and Social
  Research
• London

2
Measuring MFP - comments

• Important conclusions from the papers
• Canada mirrors US findings on acceleration in MFP
  growth in late 1990s
• but some differences, notably a lower
  contribution of ICT capital deepening and less of
  an impact from ICT producing sectors in Canada
• Less evidence of aggregate productivity bias
• Problems remain in measuring service outputs
• This discussion also considers evidence for Europe

3
Chart 1. TFP growth in the EU and US, 1989-95
and 1995-99
4.00
3.50
3.00
2.50
1989-95
2.00
1995-99
1.50
1.00
0.50
0.00
US
Italy
UK
Spain
Austria
Greece
France
Belgium
Ireland
Sweden
Finland
Denmark
Total EU
Germany
Portugal
Netherlands
4
Chart 2. MFP growth in the US and UK, 1995-1999
100
80
MFP-Other
60
MFP-ICT
K-NICT
40
L-NICT
K-ICT
L-ICT
20
0
US
UK
-20
5
The impact of ICT on MFP

• Industry level estimates
• Divide into ICT producing, ICT using and
  (sometimes) other
• The evidence on balance suggests an acceleration
  of MFP growth in ICT using industries

6

• But need econometric analysis to give definitive
  answers
• First question
• Is there a direct impact of ICT on TFP?
• Second question
• Is this effect common across countries?
• (Are there impediments to reaping full benefits
  from ICT in Europe?)

7
Chart 2. MFP growth in the US and UK, 1995-1999
100
80
MFP-Other
60
MFP-ICT
K-NICT
40
L-NICT
K-ICT
L-ICT
20
0
US
UK
-20
8
Chart 3. UK MFP growth by sector, 1995-1999
9.00
8.00
7.00
6.00
5.00
4.00
Output
3.00
MFP
2.00
1.00
0.00
-1.00
MIN
UTIL
MANF
CONS
TC
DIST
FIN
BUS
PSERV
-2.00
9
The measurement of output in service sectors

• BLS papers suggest there may be a problem but
  less so now than in the past in the US
• Negative MFP growth rates are apparent also in
  similar sectors in European countries
• Raises the possibility that the US lead over
  Europe is a statistical mirage
• Europe lags the US in its measurement of real
  output

10
Measuring MFP - comments

• Conclusions
• The decomposition of output growth into growth of
  inputs and MFP is useful in outlining differences
  across countries
• But needs to be supplemented by econometric
  analysis
• MFP growth can be negative even in
  well-measured sectors
• Our understanding of the impact of information
  technology requires more attention to measuring
  output in services