Grouping Countries by

1
Grouping Countries by National Models of
Technological Learning
Tatyana P. Soubbotina Consultant, ST
Program HDNED Presentation to STI Thematic
Group November 10, 2005
TSoubbotina_at_comcast.net
2
Practical Questions

1. Should the WB develop some standard guidelines
   on ST assistance to client countries?
2. Should these guidelines be customized for groups
   of developing countries with similar STI
   capacity-building needs?
3. Should the WB rely on one of the existing
   classifications of countries by ST capacity or
   develop some new approach?

3
Composite Indices of ST Capacity

• UNDP Technology Achievement Index
• UNIDO Competitive Industrial Performance
• Index
• WEF National Innovative Capacity Index
• WB Knowledge Economy Index
• UNCTAD Innovation Capability Index
• Francisco Sagasti ST Capacity Index

4
UNDP Technology Achievement Index
Human skills Diffusion of old innovations Diffusion of new innovations Creation of technology
Mean years of schooling Tertiary enrolment in science, math and engineering Electricity consumption per capita Telephones per capita Internet hosts per capita High- and medium-tech exports as of total exports Patents granted per capita Receipts of royalty and license fees from abroad per capita
5
UNIDO - Competitive Industrial Performance Index
1. Manufacturing value added (MVA) per capita
2. Manufactured exports per capita
3. Share of medium and high-tech activities in MVA
4. Share of medium and high-tech products in manufactured exports
6
WEF National Innovative Capacity Index
1. Share of scientists and engineers in population
2. Innovation policy
3. Cluster innovation environment
4. Innovation linkages
5. Operations and strategy
7
WB Knowledge Economy Index
Economic incentive institutional regime Education and human resources Innovation system ICT infra-structure
Tariff non-tariff barriers Adult literacy rate Number of researchers in RD Telephones per 1,000 population
Regulatory quality Secondary enrolment rate Patent applications granted by USPTO Computers per 1,000 population
Rule of law Tertiary enrolment rate ST journal articles Internet users per 1,000 population
8
UNCTAD Innovation Capability Index
Human capital Index Technological Activity index
Literacy rate as of population X 1 RD personnel per million population
Secondary school enrolment as of age group X 2 US patents granted per million population
Tertiary enrolment as of age group X 3 Scientific publications per million population
9
F. Sagasti ST Capacity Index
SCIENCE TECHNOLOGY PRODUCTION
Internal capacity
RD expenditure as of GDP Number of scientists engineers per million people High-tech exports as of total exports
External linkages
Number of scientific publications (in log.) Number of patent applications by residents and non-residents (in log.) Infrastructure, communications, and technology index
10
Different ST indices can be used depending on
the task

• Because they all have different focuses
• UNCTAD underlying technological capacity
  (focus on inputs education and RD)
• UNIDO revealed technological capacity in
  industry only (focus on manufacturing
  competitiveness)
• UNDP revealed technological capacity across
  the economy (focus on broad diffusion of old and
  new technologies)
• WEF institutional and policy environment for
  innovation
• WB KAM the advantage is in its flexibility,
• select indicators at your own risk!

11
Considerations in selecting ST capacity
indicators

• Select input or output indicators depending on
  whether you want to measure technological effort
  or technological achievement , underlying
  (potential) technological capacity or revealed
  ST capacity.
• Absolute size of inputs can matter no less than
  input intensity because of economies of scale
  and critical mass effect (e.g. Number of
  researchers or Total RD expenditure vs. their
  shares in population and GDP)
• Some indicators reflect present-time capacity,
  others reflect expected but still uncertain
  future capacity (e.g. Mean years of education of
  adults vs. Secondary and tertiary enrolment
  rates)
• Indicators of knowledge sales (e.g. Share of
  high-tech exports or Receipts of royalty and
  license fees) reflect quality of knowledge rather
  than just its quantity (e.g. as reflected by
  Share of high-tech industries in MVA or Number of
  patent applications).
• However, exports indicators should be compared
  to similar MVA indicators, because fast
  improvement in exports often reflects enclave FDI
  activities rather than national ST capacity
  growth.

12
Country rankings on 3 indices differ quite
radically
UNCTAD (117 countries) UNIDO (87 countries) UNDP (72 countries)
China 72 37 45
Russia 23 44 -
Malaysia 67 22 30
Mexico 59 23 32
Philippines 60 25 44
Singapore 30 1 10
Sweden 1 7 3
13
All of these groupings focus on ST levels
achieved or expected to be achieved by various
countries, but fail to account for

• Different speed of ST progress, and
• Different sources of ST progress.

That is what grouping countries by models of ST
learning can add
14
Concept of National technological learning
National technological learning is the process
of creating or acquiring from foreign sources of
new (for this particular learner) ST knowledge
skills, as well as adapting, disseminating,
and using those for improving the technological
structure of national production and exports.
15
National technological learning occurs at all
levels

• and implies acquiring different kind of
• knowledge skills, e.g. at the level of
• national labor force science, math,
  engineering education training life-long
  learning,
• enterprises firms learning to innovate by
  absorbing foreign and investing in own new
  technologies,
• governments learning to receive expert advice,
  develop ST strategies and create enabling
  stimulating conditions for national technological
  progress.

16
Factors of national technological learning
ST learning capacity
ST learning opportunities

ST co-operation
Knowledge generation capacity
Knowledge absorption capacity
Licensing
RD
Internet
Education
Inward FDI
Capital imports
17
Crystals of ST Learning -graphical/statistical
illustrations
18
Crystals of ST Learning -graphical/statistical
illustrations

• Human capital accumulated / human capability for
  ST learning (see indicators 11, 12, 1),
• The most accessible opportunities for learning
  from foreign sources created by capital goods
  imports and FDI (indicators 9, 10),
• The more demanding opportunities for learning
  from domestic and foreign sources through
  domestic RD (indicators 2, 3),
• The most demanding opportunities for learning
  through knowledge markets and international ST
  cooperation (indicators 4, 5, 6),
• Success in using ST knowledge for improving
  technological structures of a countrys MVA and
  manufactured exports (indicators 7, 8).

19
Crystals can grow, but only in the right
(learning) environment
20
6 models of national technological learning

• Traditionalist slow learning,
• Passive FDI-dependent,
• Active FDI-dependent,
• Autonomous,
• Creative-isolated,
• Creative-cooperative.

21
Traditionalist slow ST learning

• Relying mostly on traditional technologies,
• low ST learning capacity,
• minimal ST learning opportunities,
• low international competitiveness,
• high risk of further economic marginalization,
• most urgent need of international ST
  assistance.

22
Crystals of sample Slow-Learning Countries
23
Passive FDI-dependent learning

• passively relying on FDI to bring in new
  technologies,
• low ST learning capacity,
• no or week government technological strategy,
• limited opportunities for technological
  learning,
• high risk of losing in economic competition
  with poorer, lower-wage countries.

24
Active FDI-dependent learning

• relatively high ST learning capacity,
• active government strategy aimed at building
  national human capital and accelerating national
  technological learning from FDI,
• active targeting of the most beneficial FDI,
• much wider opportunities for technological
  learning from FDI,
• lower risk of losing in economic competition
  with lower-wage but lower-skill countries.

25
Crystals of sample Passive and Active
FDI-dependent learners
26
Autonomous ST learning

• High ST learning capacity and favorable
  international environment,
• active government strategy aimed at building
  national human capital and accelerating national
  technological learning via open sources, foreign
  consultants, contract manufacturing, licensing,
  copying re-engineering, own RD, even outward
  FDI,
• minimal reliance on FDI or international ST
  cooperation,
• aspiring to compete with technological leaders.

27
Creative-cooperative ST learning

• Capacity for both, generating and absorbing ST
  knowledge among the highest in the world,
• global technological leadership in at least some
  niches of the global economy,
• active government ST strategy directly linked
  to global competitiveness strategy,
• extensive RD and efficient NIS,
• active participation in and control over
  international ST cooperation,
• the fastest ST learning.

28
Creative-isolated ST learning

• High ST learning capacity, but unfavorable
  international environment or isolationism,
• limited opportunities for ST learning from
  foreign sources,
• aspiring to produce most of the needed
  technologies inside the country,
• low international competitiveness of high-tech
  industries,
• high risk of lagging further behind in
  technological and economic development.

29
Sample crystals of Autonomous,
Creative-Cooperative, Creative-Isolated learners
30
Rules of national technological learning

• National ST learning requires a certain
  minimal stock of human capital and a favorable
  economic institutional learning environment.
• Government ST policies and international aid
  should target both prerequisites.
• Different models of ST learning can be also
  seen as consecutive stages in the same countrys
  development (crystals are growing from 9 a.m.
  to 6 p.m.).
• But there are some policy choices, e.g. active
  FDI-dependent vs. autonomous and
  creative-isolated strategies.
• The higher a countrys underlying ST capacity,
  the broader its choice of ST learning
  strategies.

31
Tree of national technological learning
Human capital accumulation
Creative-cooperative
Autonomous
Creative- isolated
Active FDI-dependent
Aid supported
Passive FDI-dependent
Slow learning
Time
32
5 major learning paths

• 1. From slow-learning traditionalism to passive
  and active FDI-dependent learning,
• 2. From passive FDI-dependent to active
  FDI-dependent or autonomous,
• From active FDI-dependent to more autonomous or
  creative-cooperative,
• From autonomous to creative-cooperative,
• 5. From creative-isolated to creative-cooperative
  learning.

33
Prioritization Table of Policies for
Transitioning from Non-learning Traditionalism to
Passive/Active FDI-dependent ST Learning
34
Prioritization Table of Policies for
Transitioning from Passive to Active
FDI-dependent ST Learning
35
How to help the majority of slow-learning
countries?

• What should be the main features of
  international aid-supported ST learning?
• What can be learned from previous international
  aid projects with ST components?
• What should be the role of the World Bank in
  these countries?

36
Crystals assessment Modified indicators for
SSA
37
The advantages of ST Learning Models approach

• compared to any ST capacity indices are that it
  
• Looks forward, helps predict future difficulties,
• Allows for diversity of learning paths,
• Underlines the importance of policy choices made
  by developing countries themselves.

38
First of all, I think that sense of
assuming responsibility by developing country
governments is really critical. We often talk
about building institutions or building
capacity. And my feeling is that sort of
suggests you can come in like an outside
contractor and bring some bricks and mortar and
you construct capacity. It doesn't work that
way. You grow it. Its got to be indigenous.
It's got to have indigenous roots. You can
fertilize it. You can water it. You can rip
the weeds out, which I think is part of fighting
corruption. Or you can help people do it. But
they need to do it themselves. Paul Wolfowitz
on capacity building vs. capacity growing
at his first Town Hall Meeting in the World
Bank, 2005.
39
School teachers and university professors know
the advantages of active teaching and learning
methods. Should the World Bank aim to help all
client countries turn into active learners of
modern science and technology?
40
Models of ST Learning approach is an alternative
to

• Regional models of development e.g. East Asian
  vs. Latin American
• High-tech model vs. low-tech Latin model

41
High-tech success stories are obviously too
different to be treated as one model
Source W.F. Maloney. 2005. Patterns of
Innovation. Innovation Policies II Regional
Study, World Bank.
42
Further improvements to crystals indicators are
needed, e.g.

• A brain drain/brain gain statistics instead of
  brain retention survey results
• Taking into account strong economies of scale
  and critical mass effect in RD
• A better indicator of benefits from
  participation in cross-border RD cooperation
• Building data bases for historical and
  sub-national crystals of ST learning

43
Practical application of crystals assessment

• Is the countrys ST learning likely to be fast
  enough compared to its major competitors?
• Is national ST learning constrained mainly by
  the lack of human capital or the lack of learning
  opportunities?
• Which additional learning opportunities could be
  available but are currently underused?
• How successful is this country in using its ST
  capacity for improving technological structure of
  its production and exports?

44
Crystals assessment Mauritius
45
Crystals assessment Malaysia
46
Is there a need in an on-line interactive data
base and an automatic graphing tool? (similar to
KAM)
47
Crystals for further discussion
Creative-cooperative leaders
48
Crystals for further discussion high-income
Slow learners
49
Crystals for further discussion former
Creative-isolated learners