Diversification, Innovation and Growth

Diversification, Innovation and Growth

W. F. Maloney, World Bank

Knowledge Economy ForumBerlin, May 7th, 2010

Three Issues

1. Diversification vs. innovation and growth2. Doing what we do better

1. Natural Resources2. Export Quality

3. 2 queries on thinking about the innovation environment

Pasteur: “Chance Favors the Prepared Mind”

DIVERSIFICATION: WHERE ARE WE?

ECA: Some countries likely to diversify, some not

$US 1985

What does this relationship mean?

1. From diversification to income:

Specialization leads to gains from trade (Ricardo, Krugman) Diversified portfolio dampens sector specific shocks

Only clear negative effect from Natural Resources For small countries, there must be a trade off of the two

Other ways of managing shocks? Sovereign wealth funds?

2. From income to diversification:

Taste for diversity-if opening new sectors is costly then economy will become more diverse with income.

Diversification an outcome of development

What does this relationship mean? 3. Co-movement of diversification and income?

Discovery of new goods (ideas) embodies TFP growth Theoretically, no better or worse than productivity gains in

existing sectors. But, becomes partly an issue of innovation

Diversification as a specific innovation policy is tricky Diversification an outcome of difficult to predict

productivity shocks/opportunities? Exports very concentrated and unsystematic (Easterly et al

2009) Following existing patterns not obviously helpful

Market saturation? Past as prologue? There are market failures in experimentation- discovery

But these appropriation externalities held in common with ongoing products as well

Often emerge from existing industries -Nokia Innovation policy

Be prepared to take advantage of advances in what we do Be prepared to take advantage of new products Mixed support products (e.g.Tekes)

Eventually, development may be about doing fewer things better

]

W. F. MaloneyDevelopment Economics Research GroupWorld Bankwww.worldbank.org/laceconomist/wmaloney

Source: Klinger and Lederman 2002

DOING WHAT WE DO BETTER

1. Natural Resources: No curse, but lots of heterogeneity in performance

Log

GD

P pe

r cap

ita 1

990

Log Natural Resources (Leamer)-11.5041 11.7949

6

7

8

9

10

Algeria

Argentin

AustraliAustria

Banglade

Benin

Bolivia

Brazil

Burkina Burundi

Cameroon

Canada

Cape Ver

Chad

Chile

China

Colombia

Comoros

Congo, D

Costa Ri

Cyprus

Denmark

Dominica

Ecuador

Egypt, AEl Salva

Fiji

FinlandFrance

Gabon

Gambia,

Germany

Ghana

Greece

Guatemal

GuineaGuinea-B

Guyana

Honduras

Hong Kon

Hungary

Iceland

India

Indonesi

Iran, Is

IrelandIsraelItaly

Cote d'I

Jamaica

Japan

Jordan

Kenya

Korea, R

Madagasc

Malawi

Malaysia

Mali

Mauritan

MauritiuMexico

Morocco

Mozambiq

NetherlaNew Zeal

Nicaragu

Nigeria

Norway

Pakistan

Panama

Papua Ne

ParaguayPeru

Philippi

Poland

Rwanda

Senegal

Sierra L

South Af

Spain

Sri Lank

Sudan

SwedenSwitzerl

Syrian A

Thailand

Togo

Tunisia

Turkey

United K

United S

Uganda

Uruguay

Venezuel

Zambia

Zimbabwe

Leamer Measure: Net Exports of NR/Worker

Resource AbundantResource Scarce

2. Export Quality: A new lens on resources and growth Quality measured by price (unit value)

Huge variance within products (Schott 2004)

Standardize:

90itc

itcit

uR

u

Quality rises with development

CANADA

GERMANUKINGDOM

CHINA

JAPANITALY

FRANCE

MEXICO

KOREA_S

NETHLDS

SWITZLD

SPAIN

BEL_LUX

HONGKONGINDIA

BRAZIL

THAILAND

SWEDEN

AUSTRAL

ISRAEL

INDONES

AUSTRIA

MALAYSIA

SINGAPR

DENMARK

PHIL

S_AFRICA

TURKEY

IRELAND

ARGENT

RUSSIA

FINLAND

COLOMBIA

PORTUGAL

POLANDCZECHREP

NORWAY

NEW_ZEAL

VENEZ

CHILE

HUNGARY

DOM_REP

COS_RICA

PAKISTAN

GUATMALA

PERU

SRI_LKA

GREECE

ROMANIA

EGYPT ARAB_EM

HONDURA

SLOVENIA

SALVADR

MACAU

BNGLDSH

ECUADOR

YUGOSLAVBULGARIA

PANAMA

UKRAINE

JAMAICA

MOROCCO

VIETNAM

SLOVAKIA

URUGUAY

ST_K_NEV

SD_ARABMRITIUS

HAITI

BURMA

TRINIDAD

CAMBOD

KENYA

CROATIANICARAGA

JORDON

TUNISIA

ICELAND

NEPAL

LITHUANI

OMAN

SYRIA

BELARUS

BOLIVIA

IVY_CST

FIJI

MACEDONIMONGOLA

ESTONIA

LATVIA

BAHRAIN

LEBANON

GUYANA

MADAGASZIMBABWE

N_ANTIL

y = 8E-06x + 0.3482R² = 0.4874

0.2

0.3

0.4

0.5

0.6

0.7

0 5000 10000 15000 20000 25000 30000 35000

Rel

ativ

e Q

ualit

y (C

ount

ry M

edia

n)

GDP per capita PPP (Country Median, U$ 2005)

Relative Quality vs. GDP (HS>250)

Schott (2005) Krishna and Maloney 2010

Quality Growth: Are we are converging to the quality frontier?Figure 3: Quality Growth by Region 1990-2001

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

OECD(high-

income)

EASIAP(high-

income)

LAC MENA EASIAP(low -

income)

SASIA EUROPE(non-

OECD)

SSAFRICA CASIA

Region

Med

ian

Qua

lity

Gro

wth

Krishna and Maloney 2010

Figure 1: Unit Values: Drift and Standard Deviation, Unconditional Quantile Regression 1990-2001

ALBANIA

ALGERIA

ANGOLA

ARAB_EM

ARGENT

ARMENIA

ASIA_NES

AUSTRAL

AUSTRIA

AZERBAIJ

BAHAMAS

BAHRAIN

BARBADO

BEL_LUX

BELARUS

BELIZEBENIN

BERMUDA

BNGLDSHBOLIVIA BOSNIA-H

BRAZILBULGARIA

BURMA

BURUNDI

CAMBOD CAMEROON

CANADA

CHILE CHINACOLOMBIA

CONGO

COS_RICACROATIACYPRUS

CZECHO

CZECHREP

DENMARK

DOM_REPECUADOR

EGYPT

ESTONIA

ETHIOPIA

FIJIFINLAND

FRANCE

GABONGEORGIA

GERMAN

GHANA

GILBRALT

GREECE

GREENLD

GUADLPE

GUATMALA

GUINEA

GUYANA

HAITIHONDURA

HONGKONG

HUNGARY

ICELAND

INDIA

INDONES

IRELANDISRAEL

ITALY

IVY_CST

JAMAICA

JAPAN

JORDON

KAZAKHST

KENYA

KIRIBATI

KOREA_S

KUWAITLAO

LATVIA

LEBANON

LIBERIALITHUANI

MACAUMACEDONI

MADAGAS

MALAWI

MALAYSIA

MALI

MALTA

MAURITN

MEXICO

MOLDOVA

MONGOLA

MOROCCO

MOZAMBQ

MRITIUS

N_ANTIL

NEPAL

NETHLDS

NEW_CALE

NEW_GUIN

NEW_ZEAL

NICARAGA

NIGER

NIGERIA NORWAY

OMAN

PAKISTAN

PANAMA

PARAGUA

PERU

PHIL

POLANDPORTUGAL

QATAR

ROMANIA RUSSIA

RWANDA

S_AFRICA

SALVADRSAMOA

SD_ARAB

SENEGAL

SIER_LN

SINGAPR

SLOVAKIA

SLOVENIA

SPAINSRI_LKA

ST_K_NEV

SUDAN

SURINAMSWEDEN

SWITZLD

SYRIA

TAIWAN

TAJIKIST

TANZANIATHAILAND

TOGO

TRINIDAD

TUNISIA

TURKEY

TURKMENI

UGANDA

UKINGDOM

UKRAINE

URUGUAYUS_NES

USSR

UZBEKIST

VENEZ

VIETNAM

YEMEN_N

YUGOSLAV

ZAIRE

ZAMBIA

ZIMBABWE

-0.075

-0.025

0.025

0.075

0.125

0.05 0.15 0.25 0.35 0.45 0.55

Standard Deviation (country dummies)

Drif

t (co

untr

y du

mm

ies)

Convergence appears to be related to ability to place risky bets Growth in Unit Value vs Standard Deviation of Growth

Standard Deviation

Gro

wth

Rat

e

Krishna and Maloney 2010

This appears to be related to:

Financial Intermediation: (credit by deposit money banks as a share of GDP)

Innovation Effort: (R&D/GDP) inability to resolve market failures/indivisibilities around

innovation and R&D leads to less complex, less risky products

2 QUERIES ON THE INNOVATION ENVIRONMENT

00.5

11.5

22.5

33.5

44.5

5

0 1 2 3 4

Source: Rodriguez y Maloney (2005)

1. A problem of accumulation of knowledge, or of accumulation in general?

K/L

TFP vs Capital Intensity

TFP

Returns to R&D vs Distance to the Frontier:Complementary factors are key

Distance to the economic frontier (z)

Poorer countries- low complementarities

Middle income

Advanced innovators

Fuente: Goñi, Lederman y Maloney (2010)

Distance to the technological frontier

Per

u 19

96-2

000

2. What type of innovation is necessary?

2

21&

CAPGDP

CAPGDP

GDPDR

Czech Rep.HungaryPolandRomaniaRussia Turkey Ukraine

The national innovation system needs to be viewed very generally

Universities/Think tanks/CTs

The firm

Barriers to DemandMacro ContextTrade RegimeInternational Marketing ExternalitiesCompetitive StructureEntrepreneurship

Innovation “supply” Demand Side

Knowledge

Accumulation

Barriers to AccumulationCredit Entry/Exit barriersBusiness/Regulatory Climate

Barriers to InnovationMarket Failures (&IP)Seed/Venture capitalPoorly articulated S&T system (including discovery, oversight)Labor regulationDeficient human capital

Capital

Fin