42
STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY Alice Amsden Memorial Lecture September 8th, 2021 José Antonio Ocampo Professor, Columbia University

STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

  • Upload
    others

  • View
    5

  • Download
    0

Embed Size (px)

Citation preview

Page 1: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

Alice Amsden Memorial LectureSeptember 8th, 2021

José Antonio OcampoProfessor, Columbia University

Page 2: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

THE MAJOR CONCEPT:DYNAMIC EFFICIENCY

❖ Successful development is essentially a process of structural change. It depends on dynamics of production structures and related policies and institutions.

❖ Basic rationale for industrial policy: conflict between dynamic (changes in the structure of production) and static efficiency (resource allocation).

❖ Analytical contributions of classical development economics, neo-Schumpeterian, structuralist and evolutionary economics: critical role of learning, externalities and economies of scale/agglomeration.

❖ Disappointment with effects of more open economic policies on growth (e.g., Latin America).

Page 3: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

STYLIZED FACTS

Page 4: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

STYLIZED FACT (1):

PERSISTENCE OF LARGE INEQUALITIES

❖ Persistence of large inequalities in the world economy: very few new entrants into the top club.

❖ High variance of growth experiences in the developing world: low and middle-income traps, truncated convergences, open divergence…

❖ … but also success stories, particularly in Asia → “dual divergence” in the last few decades.

❖ Agglomeration generated by static or dynamic economies of scale: the East Asian cluster.

❖ Major implication: economic opportunities are largely determined by the position that a particular country occupies within the world hierarchy

Page 5: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

FACTORS BEHIND THE

“MIDDLE-INCOME TRAP”

❖ High concentration of research and development, and thus of innovations, in developed economies → high entry costs into technologically dynamic activities.

❖ Asymmetries between firms that lead and suppliers in global value chains.

❖ Differences in domestic financial development.

❖ Volatility of external financing and the terms of trade.

❖ As a result, macroeconomic asymmetries, which are reflected in the quite different degrees of freedom to adopt countercyclical macroeconomic policies.

❖ Challenges associated with building up institutions that can reconcile the interests of diverse actors.

Page 6: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

A VISUALIZATION OF THE LOW AND MIDDLE-INCOME TRAPS

Page 7: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

STYLIZED FACT (2):

STRUCTURAL CHANGE IS THE ESSENCE

OF ECONOMIC GROWTH

❖ “Balloon” vs. “structural” views of economic growth.

❖ Ability to generate new dynamic activities/ innovations.

❖ Patterns of international specialization matter.

❖ Repetitive phenomenon of creative destruction.

Success in structural change has been the key to successful economic development.

Page 8: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

SPECIALIZATION PATTERNS MATTER

❖ Most countries that have failed in increasing market shares are exporters of primary goods and natural resource-intensive manufactures.

❖ Non-dynamic markets face “fallacy of composition” effects (typical of commodity markets).

❖ There are countries that have extracted fair growth out of a specialization pattern based on natural-resources or low-tech manufactures.

❖ But most developing countries that have grown fast have been increasing market shares in mid or high-technology exports

❖ The East Asian regional cluster has an effect on top of those captured by the patterns of export diversification (huge contrast with Latin America).

Page 9: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

SPECIALIZATION PATTERNS MATTER(Ocampo-Parra)

Per capita GDP growth according to specialization pattern

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

High-tech

manufactures

Mid-tech

manufactures

Low-tech

manufactures

Natural

Resource

based

Pimary goods

1980-2006

1990-2006

Page 10: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

SPECIALIZATION PATTERNS MATTER(Hausmann-Hwang-Rodrik)

e(

gro

wth

gdp

| X

,lexp

y19

92

) +

b*l

exp

y19

92

lexpy1992

Residuals Linear prediction

8.10487 9.83871

.31443

.429625

MDG

PRY

BGD

JAM

ECU

BOL LCA

LKA

COL

HTI

PER

KEN

IDN

BLZ

CHL

DZASAU

OMNTUR

TTO

IND

GRC

ROM

THA

CYP

CHN

HRV

PRT

MYS

BRA

HUN

AUS

MEX

ESP

KOR

NZL

SGP

NLD

CANUSADNKSWE

DEU

IRL

FINISL

CHE

Page 11: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

THE “PRODUCT SPACE”(Haussman-Hidalgo-Klinger)

Page 12: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

MANUFACTURING IS CRUCIAL FOR RAPID GDP GROWTH

Page 13: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

STYLIZED FACT (3):

PATH DEPENDENCE ASSOCIATED TO LEARNING PROCESSES

❖ Dynamic economies of scale associated with learning.

❖ Opportunities are determined by production experience.

Comparative advantages can be created.

The loss of productive experience can have adverse long-term effects on growth (which may come with a lag, particularly in economies experiencing a commodity boom).

Page 14: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

DYNAMICS OF PRODUCTION

STRUCTURES

Page 15: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

DYNAMICS OF PRODUCTION STRUCTURES

Interaction between two basic forces:

❖ Innovations: New activities and new ways of doing previous activities (“new combinations”) and the learning processes associated with their development.

❖ Complementarities (linkages, networks, value chains) among firms and production activities and the institutions required for the full development of these complementarities.

❖ A complementary factor: Elastic factor supplies for innovative activities.

The combination of these factors determines the dynamic efficiency of a given production system

Page 16: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

INNOVATIONS AND ASSOCIATED

LEARNING PROCESSES (1)

❖ Critical mix between creation and destruction…

❖ …and between substitution vs. complementary effects of innovations: creative destruction or destructive creation?

❖ In the industrialized world, technical change is the engine; in the developing world, transfer of sectors from the industrialized world plays that role.

❖ No innovative process is passive, as it requires investment and learning. So, technology transfer is positive to the extent that it generates the accumulation of technological capabilities.

Page 17: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

INNOVATIONS AND ASSOCIATED

LEARNING PROCESSES (2)

❖ Climbing up the ladder in the world hierarchy entails shortening transfer periods and gradually becoming a more active participant in the generation of technology.

❖ Following the evolutionary school: attributes of technical change, organizational and commercial knowledge:

✓ Incompletely available and imperfectly tradable.

✓ Proficiency cannot be detached from production experience.

✓ Private-public attributes.

Page 18: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

COMPLEMENTARITIES AND

INSTITUTIONAL DEVELOPMENT (1)

❖ Development of networks of suppliers of goods and specialized services, marketing channels (value chains)…

❖ … and organizations and institutions that disseminate information and provide coordination among agents

❖ Demand effects: macroeconomic multipliers

❖ Supply effects: positive externalities, basis of mesoeconomic dynamic economies of scale that determine competitiveness of production activities

Page 19: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

COMPLEMENTARITIES AND

INSTITUTIONAL DEVELOPMENT (2)

❖ Efficient provision of non-tradable inputs and specialized services is crucial

✓ Knowledge, logistic and marketing services

✓ Specialized financial services

✓ Adequate infrastructure

❖ Institution-building is a non-tradable factor (at best imperfectly tradable thanks to imitation), closely associated with experience and has dominant public good attributes

Page 20: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

INTERPLAY OF INNOVATIONS AND

COMPLEMENTARITIES

Learning

process

Complementarities

Strong Weak

Strong Deep Short breath

Weak Labor

absorbing

Shallow

Page 21: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

ANCHORED VS. SHALLOW INDUSTRIES

❖ The development impact of the strategy of a given country depends on the capacity to capture a high or small share of value added.

❖ This is, in a sense, obvious and even tautological, as GDP is nothing else but “value added”.

❖ But it has broader implications, as those activities with limited value added (e.g., maquila) are likely to be footloose.

❖ Unless the industries are firmly “anchored” in the domestic economy, their growth-enhancing capacity evaporates: “shallow” specialization.

Page 22: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

ELASTIC FACTOR SUPPLIES

❖ Crucial role of availability of finance for innovative activities.

❖ Structural heterogeneity (coexistence of firms with different productivity levels) guarantees an elastic supply of labor in the developing world.

❖ Rapid development is the result of reallocation of labor towards high-productivity activities subject to increasing returns to scale.

❖ Kaldor-Verdoorn growth-productivity links.

❖ This implies that the dynamics of aggregate productivity is a largely a result of dynamic economic growth, rather than a cause.

Page 23: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

LINKS BETWEEN STRUCTURAL AND

MACROECONOMIC DYNAMICS:

SINGLE EQUILIBRIUM

Productivity

growth

GDP growth

T

T

G

G

Page 24: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

MULTIPLE EQUILIBRIA

Productivity

growth

GDP growth

T

TG

G

B

A

Page 25: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

EFFECTS OF A NEW WAVE OF

INNOVATIONS

Productivity

growth

GDP growth

T

T

G

G

T'

T'

Page 26: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

EFFECTS OF A FAVOURABLE

MACROECONOMIC SHOCK

Productivity

growth

GDP growth

G

G

T

T

G'

G'

Page 27: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

COMMODITY SPECIALIZATION

❖ Long debate on the “Dutch disease”

❖ Problems associated with the economic structure. The clue here is the exploitation of the production linkages and technological upgrading.

❖ Macroeconomic vulnerability due to commodity price volatility and associated fluctuations in capital flows. The clue here is counter-cyclical policies, and in particular avoiding exchange rate overvaluation and volatility.

❖ Political economy problems: rents generate a perverse political economy + rent-seeking and possibly corruption + conflict in oil/mining areas.

Page 28: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

THERE IS STRONG EVIDENCE OF LONG-TERM COMMODITY PRICE CYCLES

-0.4

-0.3

-0.2

-0.1

0.0

0.1

0.2

0.3

0.4

1875 1900 1925 1950 1975 2000

Non-tropical super cycle Tropical super cycle

Metal super cycle

Super Cycle Components for Non-oil Subindices

-0.8

-0.6

-0.4

-0.2

0.0

0.2

0.4

0.6

0.8

1875 1900 1925 1950 1975 2000

Non-oil total super cycle

Oil super cycle

Super Cycle Components for Non-oil and Oil Prices

Page 29: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

STRONG COMMODITY PRICE FLUCTUATIONS IN THE EARLY

TWENTY-FIRST CENTURY

0

20

40

60

80

100

120

140

160

180

20

03

m0

1

20

03

m0

7

20

04

m0

1

20

04

m0

7

20

05

m0

1

20

05

m0

7

20

06

m0

1

20

06

m0

7

20

07

m0

1

20

07

m0

7

20

08

m0

1

20

08

m0

7

20

09

m0

1

20

09

m0

7

20

10

m0

1

20

10

m0

7

20

11

m0

1

20

11

m0

7

20

12

m0

1

20

12

m0

7

20

13

m0

1

20

13

m0

7

20

14

m0

1

20

14

m0

7

20

15

m0

1

20

15

m0

7

20

16

m0

1

20

16

m0

7

20

17

m0

1

20

17

m0

7

20

18

m0

1

20

18

m0

7

20

19

m0

1

20

19

m0

7

20

20

m0

1

20

20

m0

7

20

21

m0

1

Commodity Prices (2010=100)(CPB Netherlands Bureau)

Fuels Non-fuels

Page 30: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

THE CASE FOR NATURAL RESOURCE-BASED EXPORTS TODAY

❖ It may be the main (only) strategy available for low-income and some middle-income countries + Some middle-income countries may have lost the train in manufacturing ICT products (Latin America).

❖ Complementarities with East Asia.

❖ Opportunities in natural-resource based industrialization✓ Technological: biotechnology, nanotechnology,

environmentally-friendly products.

✓ New market opportunities: differentiated and customized products.

✓ High Asian demands + prices (now questionable)

❖ Need to exploit the whole value chain + fair trade and environmentally-friendly opportunities.

Page 31: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

MACROECONOMIC AND FINANCIAL

DIMENSIONS

Page 32: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

VOLATILITY OF EXTERNAL FINANCING

❖ In developing economies, external financing may not be available, or is procyclical (including, possibly, sudden stops).

❖ External financial cycles directly affect domestic spending, the growth of credit and asset prices…

❖ … but they also reduce the margin for countercyclical macroeconomic policies, and even generate incentives to adopt procyclical policies.

❖ With capital mobility, economic policy faces a strong undesirable trade-off between a procyclical monetary policy and a procyclical exchange rate policy.

❖ The latter has strongly negative effects on economic diversification, both because of uncompetitive rates during booms (“Dutch disease” effects) and volatility.

Page 33: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

VOLATITY OF PORTFOLIO FLOWS HAS BEEN PERSISTENT

-20,0

0,0

20,0

40,0

60,0

80,0

Jan

-14

Ma

y-1

4

Sep

-14

Jan

-15

Ma

y-1

5

Sep

-15

Jan

-16

Ma

y-1

6

Sep

-16

Jan

-17

Ma

y-1

7

Sep

-17

Jan

-18

Ma

y-1

8

Sep

-18

Jan

-19

Ma

y-1

9

Sep

-19

Porfolio flows, 2014-2019 (IIF, billion dollars)

Emerging Asia Latin America Emerging Europe Africa and Middle East

Page 34: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

CURIOUSLY, THE SUDDEN STOP WAS VERY SHORT DURING THE COVID-19 CRISIS

-30.0

-20.0

-10.0

0.0

10.0

20.0

30.0

20

15

20

16

20

17

20

18

20

19

Jan

-20

Feb

-20

Mar

-20

Ap

r-2

0

May

-20

Jun

-20

Jul-

20

Au

g-2

0

Sep

-20

Oct

-20

No

v-2

0

De

c-2

0

Jan

-21

Feb

-21

Mar

-21

Ap

r-2

1

May

-21

Jun

-21

Jul-

21

Capital Flows towards Emerging Markets, JPMorgan (billion dollars)

Hard-currency bonds Other bonds Equity

Page 35: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

A CRITICAL FINANCING INSTRUMENT: NATIONAL DEVELOPMENT BANKS

❖ Private finance unwilling to fund activities with uncertain returns, strong learning effects and externalities key for structural transformation and sustainable development.

❖ Basic functions:

✓ Provide counter-cyclical finance.

✓ Support activities that lead structural transformation.

✓ Deepen and improve financial markets for development-friendly instruments.

✓ Support greater inclusion of small firms.

✓ Finance global public goods (climate change).

❖ NDBs should work very closely with private sector.

Page 36: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

THE ROLE OF EXCHANGE RATES

❖ Competitive and stable real exchange rates always play an essential role in the development of new production sectors. Large empirical evidence in this regard.

❖ This role is stronger if we want to overcome two constraints: the possible rent-seeking effects of industrial policy, and limits imposed by international rules (lack of sufficient policy space).

❖ At the same time, tax sectors with no learning spillovers or externalities.

❖ This leads to effectively multiple real exchange rates: different real exchange rates for sectors with diverse spillovers, while maintaining the commitment of IMF members to avoid multiple exchange rates.

Page 37: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

SUMMARY:

POLICY IMPLICATIONS

Page 38: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

STRUCTURAL

TRANSFORMATION POLICIES

❖ High quality infrastructure and education systems serve as basic “background conditions”

❖ Support for structural transformation of production✓ Support for new industries and production clusters

✓ Diversification of the export base

✓ Domestic production linkages of exports and activities with FDI presence

❖ Innovation systems that accelerate the development of technological capacities

❖ Appropriate international rules / “policy space”

❖ Structural transformation is not a once and for all process, it is not smooth (creation/destruction), and structural heterogeneity is a persistent feature.

Page 39: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

EXPORT STRATEGIES

❖ Domestic conditions (existing specialization patterns, size, location) determine the opportunity that an individual country has.

❖ Diversifying into higher technology products is crucial, but opportunities may be limited.

❖ Individual countries can succeed in any of these strategies, but as a group developing countries can only succeed if the demand is elastic (it may require developed countries losing market shares).

❖ Different markets provide different opportunities (N-S, S-S with China at the center, intraregional).

❖ Domestic and regional markets are attractive!

Page 40: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

STRONG SLOWDOWN OF INTERNATIONAL TRADE MAKES DOMESTIC/REGIONAL

MARKETS MORE ATRACTIVE

7,4%

3,7%

7,3%

2,9%

4,8%

3,2% 3,1%

2,4%

0,0%

1,0%

2,0%

3,0%

4,0%

5,0%

6,0%

7,0%

8,0%

1950-1974 1974-1986 1986-2007 2007-2019

World GDP and trade growth

Exports (quantum) GDP (at market prices)

Page 41: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

ACCOMPANYING MACROECONOMIC

AND FINANCIALPOLICIES

❖ Combine strategies of structural transformation with countercyclical macroeconomic policies that help manage commodity and external financial cycles.

❖ Commodity stabilization funds and capital account management are crucial instruments of an appropriate counter-cyclical policy.

❖ An active exchange rate policy to guarantee a competitive and relative stable real rates is important in a structural transformation strategy.

❖ Taxing sectors that do not generate learning externalities is a complementary instrument.

❖ At as a domestic policy, National Development Banks play an essential role.

Page 42: STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

STRUCTURAL DYNAMICS AND INDUSTRIAL POLICY

Alice Amsden Memorial LectureSeptember 8th, 2021

José Antonio OcampoProfessor, Columbia University