Environment Directorate 1
Structural feed-back mechanisms supporting convergence:Some observations
Philip Bagnoli, OECD Environment Directorate
ORGANISATION FOR ECONOMIC CO-OPERATION AND DEVELOPMENT
ORGANISATION DE COOPÉRATION ET DE DEVELOPMENT ÉCONOMIQUES
Global convergence scenarios: structural and policy issues
Paris, 16 January 2006
Environment Directorate2
New angle for interest in convergence
• Want to be able to provide a good basis for policy analysis where long horizons are necessary.
• Baumol (1986) interested in US growth (vis-à-vis Japan)
• Want to know whether the rich are getting richer and the poor are getting poorer – a social-welfare issue.
• Want to know which theories are best: AK models versus the neo-classical framework.
Environment Directorate3
Audience for workshop results
• Outcome of two OECD workshops (June 2004) could contribute to IPCC work.
• IPCC-sponsored meeting in Seville in March. – How much involvement should IPCC have in scenarios generating
process and how best to implement that.
• IPCC was advised to make a distinction between just looking out 25-30 years, and looking out much farther.
• Question raised by McKibbin’s point about time horizons:– If we know that current actions will impact on people 50 or 100
years from now, shouldn’t we take those impacts into account in today’s policies?
– Shouldn’t economists be involved in those discussions with policy-makers?
Environment Directorate4
Convergence: The contextTable 2: Growth Miracles and Disasters, 1960-90. Annual Growth Rates of Output per Worker
Miracles Growth Disasters Growth
Korea 6.1 Ghana -0.3
Botswana 5.9 Venezuela -0.5
Hong Kong 5.8 Mozambique -0.7
Taiwan 5.8 Nicaragua -0.7
Singapore 5.4 Mauritania -0.8
Japan 5.2 Zambia -0.8
Malta 4.8 Mali -1.0
Cyprus 4.4 Madagascar -1.3
Seychelles 4.4 Chad -1.7
Lesotho 4.4 Guyana -2.1Source: Temple (1999)
Environment Directorate5
Structural Feedback: Growth
• Endogenous world growth– How do the externalities cut? “backwash” or “shoulders of giants”
(i.e. increasing cost of R&D or decreasing?) Policy and issues not clear.
– Leads to models that are highly esoteric: does this belong in the baseline or in an analysis
– First question: How robust are the findings?– Second question: How would they be implemented? AJR (2001)
suggests that changes in institutions can have a large effect (property rights and expropriation risk), but how would that be implemented without having something that was ad hoc.
– Is endogenous growth ready for “prime time”
Environment Directorate6
Structural Feedback: Capital Movements
FDI in China contributed about 10 percent of fixed domestic investment between 1992 and 1997 (OECD, 2000).
Is this a large or small contribution?
How much slower than 11.5 % would growth have been?
A significant part of that 10 % is likely to have been in technologies that China would not otherwise have had. Appropriating technology.
Oddly enough, most of the FDI did not come from OECD countries. It came from Hong Kong and Taiwan!
Literature linking FDI and technology transfer has to be part of the convergence story.
Environment Directorate7
Structural Feedback: Convergence as part of the Baseline
• Need clarity: distinction between a structural feedback and an empirical regularity being put in the baseline.– As McKibbin implied: there is likely a Lucas Critique to be made
between the processes of interest and some reduced-form relationships being represented.
• Policy analysis for environmental issues needs a baseline• Scenarios may or may not be suitable for policy analysis
– E.g. Globalisation Scenario (where increased future inter-dependence among countries is embedded in the baseline) is difficult to use for analysing the impact of trade policy.
– When the scenario comes from a “storyline” it is unlikely to specify in detail the policies and drivers that make globalisation come about.
– Thus, attempting a study of new trade-related policies becomes difficult since the new policy being analysed can’t be distinguished from implied policies in the scenario.
Environment Directorate8
Purchasing Power Parities, 2002 Benchmark Year
Source : OECD (2004b)
Comparative price levels and indeces of real and nominal GDP per head
0
20
40
60
80
100
120
140
160
180
200
220
Lux
embo
urg
Nor
way
Uni
ted
Sta
tes
Irel
and
Sw
itze
rlan
dA
ustr
iaD
enm
ark
Net
herl
ands
Can
ada
Icel
and
UK
MB
elgi
umS
wed
enA
ustr
alia
Fin
land
Fra
nce
Japa
nIt
aly
Ger
man
yS
pain
Isra
elN
ew Z
eala
ndC
ypru
sG
reec
eP
ortu
gal
Slo
veni
aK
orea
Mal
taC
zech
Rep
Hun
gary
Slo
vak
Rep
Est
onia
Pol
and
Cro
atia
Lit
huan
iaL
atvi
aM
exic
oR
ussi
an F
edB
ulga
ria
Rom
ania
Tur
key
FY
RO
M
Per capita volume (PPP) indices Per capita value (MER) indices Comparative price levels
Environment Directorate9
Structural Feedback: Endogenous PPPs
• Differences in sectoral productivity lead to structural change. • Essential difference between most services and virtually all
manufactured goods: manufactured goods are tradable, services are mostly not.
• With high productivity in tradables and low productivity in non-tradables, we would expect relative prices between them to be changing.
• A PPP exchange rate necessarily measures the difference in prices between non-tradables (tradables generally sell at similar prices across countries after accounting for taxes and shipping costs).
• High productivity in tradables will lead to wage gains in those sectors which, in a closed labour market, must lead to wages increasing in the non-tradables sector. But increasing wages in the non-tradables sector must lead to increases in the PPP exchange rate: development will bring about an endogenous change in PPP.