PowerPoint slides

Postwar Metrics
Session 15
Problems with the Phillips curve:
• Trade-off between inflation and
unemployment / real output
• Used as a (Keynesian) policy instrument
• Appeared not to be that stable as originally
was supposed (not by Phillips!)
• Stagflation: increase of both inflation and
Rational expectations
Expectations and the Neutrality of
Money (1972):
• No usable trade-off, notwithstanding
econometric evidence
• Rational agents + inadequate price
signals + equilibrium system
Econometric policy evaluation: A
critique (1976)
• No invariance under policy
Robert Lucas
Lucas Critique (1976)
y  F ( x1 ,  , x n ;  1 ,  ,  m )
invariant for policy variations
Tastes and Technology
Debate in and about econometrics
CC econometrics has failed
• Hendry, David F. (1980). Econometrics –
Alchemy or Science? Economica.
• Sims, Christopher A. (1980). Macroeconomics
and Reality. Econometrica.
• Leamer, Edward E. (1983). Let’s take the con
out of econometrics. American Economic
• August 19
A Symposium on Econometric
Methodology at the World Congress of the
Econometric Society. David Hendry, Edward
Leamer and Christopher Sims.
• December 29 Joint Session with the HES at
the Annual Meeting of the AEA: ‘First Forays
Into the History of Econometrics’. Roy Epstein,
Mary Morgan and Chris Gilbert. Edward
Leamer was discussant.
• Foundation of Econometric Theory by Peter
C.B. Philips
Editorial policy of ET
5. To publish historical studies on the evolution
of econometric thought and on the subject's
early scholars.
6. To publish high-level professional interviews
with leading econometricians.
David Hendry
LSE inaugural lecture
Econometrics: Alchemy or Science?
LSE approach: General to Specific
Keynes according to Hendry
Problems of the linear regression model:
• using an incomplete set of determining factors (omitted variables bias)
• building models with unobservable variables (such as expectations)
• estimated from badly measured data based on index numbers
• obtaining spurious correlations from the use of proxy variables and
• being unable to separate the distinct effects of multicollinear variables
• assuming linear functional forms not knowing the appropriate dimensions
of the regressors
• mis-specifying the dynamic reactions and lag lengths
• incorrectly pre-filtering the data
• invalidly inferring causes from correlations
• predicting inaccurately (non-constant parameters)
• confusing statistical with economic significance of results
• failing to relate economic theory to econometrics.
Additional problems:
stochastic mis-specification
incorrect exogeneity assumptions
inadequate sample sizes
lack of structural identification
inability to refer back uniquely from observed
empirical results to any given initial theory
Keynes’s critique still relevant:
“It is difficult to provide a convincing case for
the defence against Keynes’s accusation almost
40 years ago that econometrics is statistical
alchemy since many of his criticisms remain
Don Patinkin (1974 Presidential Address to the
Econometric Society): “I find it somewhat
depressing to see how many of them are, in
practice, still of relevance today.”
Hendry’s methodology
“The three golden rules for econometrics are
test, test and test; that all three rules are broken
regularly in empirical applications is fortunately
easily remedied. Rigorous tested models, which
adequately described the available data,
encompassed previous findings and were
derived from well based theories would greatly
enhance any claim to be scientific.”
Christopher Sims
VAR approach
(vector auto regressive)
Response against Koopmans critique of
Measurement without Theory: CC
methodology implies too much a priori
“I think the most reliable way to do
empirical research in macroeconomics is to
use assumptions drawn from ‘theory,’
which actually means intuition in most
cases, as lightly as possible and still develop
Edward Leamer
• Economic tricks, econo-mystics,
• Bayesian approach
• Statistical approach too much based
on statistics developed in
agricultural experimentation (RCT,
R.A. Fisher)
Non-econometric response
• Kydland and Prescott carrying out the Lucas
program to localize invariance not in the
relations but in the parameters.
• Stable facts = stylized facts = NBER facts
• Calibration
NBER Working plans
Observable characteristics
“The way we have chosen is to observe the
business cycles of history as closely and
systematically as we can before making a fresh
attempt to explain them.”
Woodward: Data - Phenomena
• Phenomena are relatively stable and general features of
the world and therefore suited as objects of explanation
and prediction.
• Data, that is, the observations playing the role of
evidence for claims about phenomena, on the other
hand involve observational mistakes, are idiosyncratic
and reflect the operation of many different causal
• Theories are not about observations but about
Stylized Facts
“Since facts, as recorded by statisticians, are
always subject to numerous snags and
qualifications, and for that reason are incapable
of being summarized, the theorists, in my view,
should be free to start off with a ‘stylized’ view
of the facts – i.e. concentrate on broad
tendencies, ignoring individual detail”
- Nicholas Kaldor (1961), 'Capital Accumulation
and Economic Growth.’
Kaldor’s Facts
The continued growth in the aggregate volume of production and
in the productivity of labour at a steady trend rate;
A continued increase in the amount of capital per worker;
A steady rate of profit on capital;
Steady capital-output ratios over long time periods
A high correlation between the share of profits in income and the
share of investment in output; a steady share of profits (and of
wages) in societies and/or in periods in which investment
coefficient (the share of investment in output) is constant.
There are appreciable differences in the rate of growth of labour
productivity and of total output in different societies, the range of
variation (in the fast-growing economies) being of the order of 2-5
per cent.
Stylized Kaldor’s facts
Solow: “There is no doubt that they are stylized, though it
is possible to question whether they are facts.”
1. Output per worker grows at a roughly constant rate
that does not diminish over time.
2. Capital per worker grows over time.
3. The capital/output ratio is roughly constant.
4. The rate of return to capital is constant.
5. The share of capital and labor in net income are
nearly constant.
6. Real wage grows over time.
Lucas’s Program
“A ‘theory’ is not a collection of assertions
about the behavior of the actual economy but
rather an explicit set of instructions for building
a parallel or analogue system – a mechanical,
imitation economy. A ‘good’ model, from this
point of view, will not be exactly more ‘real’
than a poor one, but will provide better
imitations. Of course, what one means by a
‘better imitation’ will depend on the particular
questions to which one wishes answers.”
Computational Experiment
pose a question
use a well-tested theory
construct a model economy
calibrate the model economy
run the experiment
how much-questions instead of why-questions
In our business cycle studies, we do not try to fit or
explain anything. Rather, we derive the business
cycle implications of growth theory. […] We
emphasize that we deduce the quantitative
implications of theory for business cycle fluctuations.
Well-tested Theory
set of instructions for building a mechanical
imitation system to answer a question
to provide reliable answers to a class of
data are used to calibrate the model economy
so that it mimics the world as closely as
possible along a limited but clearly specified,
number of dimensions
Stylized Facts of Economic Growth
Stylized Facts
1. Real output grows at a more or less constant rate.
2. The stock of real capital grows at a more or less
constant rate greater than the rate of growth of the
labor input.
3. The growth rates of real output and the stock of
capital tend to be about the same.
4. The rate of profit on capital has a horizontal trend.
Real Business Cycles
y t  a t F ( k t , ht )
max E {   (1   ) u ( c t ,1  h t )}
c , x ,h
ct + x t  y t
(1+)kt+1 = (1-)kt + xt
Calibration I: Shaping
Stylized Facts of Economic Growth
F ( k t , h t )  (1   )
u ( c t ,1  h t ) 
t (1   )
kt h
(1  h t ) )
1 
1 
What is the quantitative nature of fluctuations
induced by technology shocks?
at  e
zt+1 = zt + t
 t ~ N (0,   )
Calibration II: Parameter values

similar documents