MEASUREMENT PROBLEMS OF TFP PERFORMANCE IN AN UNSTABLE ECONOMY:

1
MEASUREMENT PROBLEMS OF TFP PERFORMANCE IN AN
UNSTABLE ECONOMY ARGENTINA 1990-2004 A Case of
the Tyranny of Methodology ARIEL COREMBERG ECLAC
Buenos Aires CONICET/IDES acorem_at_fibertel.com.ar a
coremb_at_yahoo.com.ar The 2008 World Congress on
National Accounts and Economic Performance
Measures for NationsMay 1217, 2008Key Bridge
Marriott (Arlington, VAminutes from Washington
D.C.)
2

• (T. J. Koopmans 1947) This research applies the
  famous Koopmans motto It is necessary to avoid
  Measurement without theory
• (J. Hicks 1981) The measurement of capital is
  one of the nastiest jobs that economists have set
  to statisticians.
• Z.Grilliches (1990) procyclical fluctuations in
  productivity do not make sense if we want to
  interpret them as a measure of the growth in the
  level of technology or the state of economically
  valuable knowledge of an economy. The US.
  Economy did not forget 4 of its technology
  between 1974 and 1975.
• D. Jorgenson (1995) the productivity term A
  reflects all those effects on output growth that
  are not investment, where investment is
  understood as the commitment of current resources
  in the expectation of future returns, implying
  that these returns can be internalised by the
  investor.
• Heymann, Galiani y Tomassi (2004) Agents make
  economic decisions taking the cycle as a trend a
  correct identification of TFP is important in the
  context of sustainability of macroeconomic
  configuration, wealth perceptions and
  competitiveness of an economy

3
OBJETIVE

• IDENTIFICATION OF TREND AND GROWTH PROFILE
• SUSTAINABILITY OF GROWTH AND COMPETITIVENESS IN
  THE LONG RUN STRICT TFP
• EMERGING UNSTABLE ECONOMIES WITH RELATIVE PRICE
  VOLATILITY AND INCONCISTENT MACROPOLICIES
• ARGENTINA 1990-2004 WHAT ABOUT STRICT TFP?

4
ARGENTINA-STYLIZED FACTS 90S

• 90s deregulation privatization, trade
  financial openness
• apparent productivity gains compensate real
  appreciation due to convertibility plan
• Lower relative capital costs ?K intensity
• Previous Underutilization of K ? Ku (60-75)
• After Brazilian devaluation, inconsistent macro
  policy could not generate S and TFP in l.r.

5
ARGENTINA-STYLIZED FACTS 02-08

• 2002 Devaluation and Corralito double real
  exchange rate import substitution and increase
  in exports growth in aggregate demand
  improvement in terms of trade as permanent
• High decreases in labor costs and Previous
  Underutilization of K L ?Ld(ul 20-10) and
  ?Ku (55-73)

6
SOURCE OF GROWTH

1. Increase in capital intensity
2. Improvements in productive organization
   independently of factor endowments (TFP)
3. Changes in the composition of GDP, K, L
4. Quality changes in primary inputs
5. Cyclical Changes in the Utilisation of Productive
   Factors Labour intensity L and K Utilization

7
BASIC SOURCE OF GROWTH TFP OR k
y
B
Eb
yb
A
E
y
E
ya
Ea
0
k
kb
ka
8
TFP OR KU
y
A
E0
y0
B
E1
y1
E1
0
k
k0
k1
9
TFP TYPOLOGIES

• APARENT TFP INCLUDE QUALITY, COMPOSITION
  UTILIZATION EFFECTS
• STRICT TFP SHIFTS IN PRODUCTION FUNCTION DUE TO
  ORGANIZATION IMPROVEMENT (excludes Q, C, U
  effects)

10
Identification of Basic Source of Growth

• Intensive Form
• Extensive Form

• Abramovitz Solows Residual is the measurement
  of our ignorance
• Grilliches non-measure factors and error
  measurement are in the Residual

11
Index Numbers

• Tornquist Index

12
Composition Effects
Quality Effects
Utilization Effects
13
EXHAUSTIVE GROWTH ACCOUNTING
14
MAIN RESULTS
15
GDP
16
(No Transcript)
17
(No Transcript)
18
(No Transcript)
19
LABOR INPUT
20

• Hs. Worked vs Jobs
• Inclusion Self Employed
• Another Input
• Imputation of Labor Compensation of Self Employed
  with the Wages of Employees by Sector
• Imputation of Internal Rate of Return of the
  Industry

21
Weights wages by industry
22
(No Transcript)
23
(No Transcript)
24
(No Transcript)
25
CAPITAL SERVICES
26
Measurement Problems of K services in Unstable
Economies

• Cohorts Aggregation Efficiency Growth Profile
• User cost
• Utilization Indicators

27
(No Transcript)
28
(No Transcript)
29
(No Transcript)
30
(No Transcript)
31
(No Transcript)
32
SUMMARY
Effects Excluded Bias on Q Bias on Inputs Bias on Q/L TFP
QC - -
LCLQ (02-04) -
L Intensity Cyclical Phase -
L Intensity -Cyclical Phase -
Ku Cyclical Phase -
-Cyclical Phase -
KCKQ -
33
(No Transcript)
34
(No Transcript)
35
(No Transcript)
36
(No Transcript)
37
(No Transcript)
38
(No Transcript)
39
(No Transcript)
40
(No Transcript)
41
INTERNATIONAL COMPARISONS Q, Q/L, TFP INTERNATIONAL COMPARISONS Q, Q/L, TFP INTERNATIONAL COMPARISONS Q, Q/L, TFP INTERNATIONAL COMPARISONS Q, Q/L, TFP INTERNATIONAL COMPARISONS Q, Q/L, TFP INTERNATIONAL COMPARISONS Q, Q/L, TFP INTERNATIONAL COMPARISONS Q, Q/L, TFP
USA USA EU EU ARG
70-95 95-04 70-95 95-04 95-04
Q 2.8 3.2 2.2 2.2 1.5 (1.9)
Q/L 1.3 2.4 2.4 1.4 -0.2 (0.3)
80-95 95-04 80-95 95-04 95-04
TFP 0.7 1.6 0.7 0.3 0.3a (-0.1S)
Q K L TFP
Hong Kong 66-91 7.3 8 3.2 2.3
Singapore 66-90 8.5 11.5 5.7 -0.3
Corea 66-90 10.4 13.7 6.4 1.6
Taiwan 66-90 9.6 12.3 5.1 2.4
42
(No Transcript)
43
EXTENSIVE GROWTH PROFILE

• This Results are analogous to Young (1997) y
  Timmer and Van Ark (1990) for Nics experience.

1990-2004 1990-2001 2002-2004
K 57 56 50
L 30 25 53
Strict TFP 13 19 -3
Apparent TFP 40 43 47
44
CONCLUSIONS

• K estimation was done following OECD and SNA
  econometric estimation of depreciation, hedonic
  prices, and rental equivalent taking into account
  the reliability of Argentine statistics K/Q
  K/L lower greater dynamism than developed
  countries.
• Adjustment by User Cost Greater Performance than
  KW excluding this effect could overestimate the
  TFP

45
CONCLUSIONS

• Non-measure of Composition and Quality Effects in
  L Substitution in Q underestimate Q/L Strict
  TFP (2002-2004)
• The adjustment by Labour Intensity and K
  utilisation reduce the pro cyclical behaviour of
  TFP

46

• The importance of competitive gains of the
  Argentine economy through improvements in
  apparent cyclical TFP, generated both during the
  nineties and after 2002 devaluation, are
  unquestionable. However, doubts arise about the
  ability of the Argentine economy to generate the
  necessary productivity gains in the strict sense,
  independently of composition and quality effects
  and cyclical variations in factor utilisation,
  that allow to maintain a sustainable long run
  growth.
• The extensive growth profile diagnosed for the
  Argentine economy, especially during the
  nineties, contrasts with assessments of other
  authors and institutions based on the traditional
  methodology without adjusting by relative price
  effects and factor utilisation. On the contrary,
  our results are analogous to the evidence found
  by Young (1995) and Timmer and Van Ark (2000) for
  Nics countries.
• This conclusion is based not only in what Young
  (1995) called the tyranny of numbers, by
  assessing strictly the consistency of the country
  statistical information, but also a consequence
  of the tyranny of the economic cycle,
  macroeconomic and methodological consistency.