*{Innovation and Growth
A Discussion Paper for the
OECD Council at Ministerial Level
26-27 June 2000}

Discussion on the emergence of a "new economy" can lead down various
paths. At BIAC, we understand our subject of discussion to be focused around
the following questions: How does the interaction between innovation, be it
technological or managerial, and productivity vary across the OECD countries?
Has there been a fundamental change in the kind and strength of innovation
affecting productivity growth significantly in some countries? To what extent
does that account for better growth performance and to what extent are other
factors responsible? What can other countries do in order to experience a
similar impact from technical innovation?

Innovation is the centre of focus for both business and the OECD.

At a more popular level, much of the "new economy" discussion confines itself
to the impact of newly invented Internet-based business models on the
economy. Important and fascinating as the latter discussion may be, it
addresses only a limited facet of the ongoing technological and managerial
transformations. The OECD's endeavour to distil evidence for what may be
new factors affecting economic growth and a new understanding of how some
of the other factors operate is an extremely welcome ray of light illuminating a
debate that otherwise risks capturing only one part of a larger picture.

The real meaning of the term "new economy" is broader than the Internet
technology itself, and has a scope that reaches everywhere in the "old
economy" where the new technology is applied. The rest of the economy will
grow more or less well in terms of vigour or performance to the extent that
market incentives permit and innovation emerges. The key development to be
studied is the transformation of commercial activity in general, especially
business processes with attention to the raw speed with which change is
occurring.

This velocity of innovation is enabled by technology, feeding on itself,
permitting the management of the private enterprise model to alter its
composition with remarkable speed. At the same time, that invention is
opening new product and marketing possibilities for the customers of those
goods and services, sometimes shifting demand so quickly that even the most
sophisticated modellers cannot predict tomorrow. In the technology business,
they have taken to thinking in “web years”, that is, periods of three months.

Empirically, we might observe that the micro-economy has increased its
“revolutions per minute” and could postulate that its ability to right itself and
stay on course has improved. This is the element that would be “new”.

The key driver of the technology-based part of the economy has been price
performance of hardware and software, which is improving at a 20 percent per
year pace. The rate and remarkable persistence of downward price movement
in ICT (information and communications technology) producing industries is
perhaps one of the key distinguishing characteristics which sets this
technology revolution apart from many others. Thus those firms may be
responsible for something “new” in their impact on the rest of the economy.

It should be pointed out that any development that will arrest or slow down the
phenomenal quality/price performance in the ICT industry will jeopardise the
spread of ICT-based services throughout the rest of the economy. "Ecommerce"
and "new economy" have become major political topics and, in
some circles, there is an effort to craft a new form of industrial policy to
nurture them. The single most important factor enabling widespread
digitisation of economic activity - that phenomenal quality/price performance
of ICT equipment - was made possible in a market environment that was
characterised by fierce competition and a high degree of internationalisation. *{1}

Although government has been an important enabler mostly as a customer of
products, efforts to steer the direction of technical innovation in this industry
and "pick winners" have generally led to disasters. While important, the
successes among those efforts have not been responsible for most of the
commercially viable technological breakthroughs in the semiconductor and
computer industries. *{2} In this context, the current intensification of some of the
political debates on e-commerce and related issues as bones of contention
between states is not a good omen for the spread of this technological
revolution from the ICT goods sector to the rest of the economy, the services
sector in particular. On the other hand, the OECD can be credited for much
work countering discrimination on the basis of technology.

On the other hand, it is undeniable that research sponsored or conducted by
government has provided some of the crucial ingredients of current
technological innovations. In the case of the Internet, such ingredients may
include the original development of Arpanet and the World Wide Web.
However, it is important to realise that such developments have been turned
into innovation by actors other than their developers and in ways not
anticipated by the original R&D effort.

It is also a potent question to ask the counter-factual question of what the
impact of such inventions might have been on the development of electronic
networks usable in a pervasive way, in the absence of telecommunications
liberalisation and rapidly declining price of computing equipment. One would
not want to place all of one’s eggs in government-controlled technology
development basket, certainly not as the principal driver of innovation. It is a
significant ingredient, which can deliver results, only where it can be combined
with entrepreneurship and favourable market conditions. Many policy makers
who have recently turned their attention to the "new economy" need to hear
this aspect more loudly and clearly than the well known, and well accepted
merits of government funded and controlled basic science and R&D.

According to OECD studies, there is evidence for a significant secular speedup
of productivity growth in the United States around 1995, considered unusual in
a country at the productivity frontier in many sectors. Business economists
reckon that US economic growth was fully 100 basis points higher in 1999 due
to ICT spending per se (representing a growth rate of 4.2 percent, instead of
3.2). However, when we turn to measuring the impact of ICT on the
productivity in the rest of the economy, conceptual and empirical issues render
conclusions difficult. This is precisely where OECD's help is needed. In
addition to the United States, only a small number of OECD countries have
shown improved or continuing good performance in terms of productivity
and/or GDP per capita growth. These countries populate the upper echelons of
the ranking in Table 1 below.

Much of the available evidence regarding the linkage between ICT, innovation,
productivity and growth is skilfully captured in the two main analytical papers
produced by the OECD in the first half of 2000. *{3} The specific work on the
impact of innovation and ICT on growth provides a good assessment of their
contribution that is significant in some countries.

Given the limitations in the availability of data and its comparability, the OECD
staff has done a job which is as good as it can possibly be on this subject at
this time. Given the nature of some of the variables tracked, such as
productivity or per capita real GDP, which behave in stochastic fashion (i.e.,
involving randomness), it will be some years before one can know whether
there was a long term acceleration in the US productivity. The time period
under study (1995-99) is just too short even if we had complete data for it. It is
therefore incumbent on the OECD to lace enthusiasm with some caution, while
continuing to promote policies which facilitate innovation.

It is crucial to understand what real combination of factors is accounting for
the changing fortunes of some of the OECD countries in the area of
productivity and growth, so that proper and helpful recommendations can be
given to other countries who wish to emulate the former. In that context, there
is every reason for the OECD to be clear and reliable, once again, by providing
a factual assessment exploring the frontiers of current knowledge.

Productivity revolutions occurred before. The task at hand is to monitor and
understand if there is indeed one taking place at present with the most current
economic and industrial analysis methods. Unlike some past productivity
gains brought by improved transportation technologies such as railways, that
produced by ICT can be directly embodied in virtually every production
process, including that of services.

Ultimately the current transformation of innovation and production processes
in the advanced economies may be explained by a favourable combination of a
number of factors, of which technological change is only one. Such a finding
would make the present era not unlike the productivity improvements of the
period 1950-73, which can not be accounted for by a single driver. In that
period, a favourable combination of a number of factors, including the
accumulated technical innovations of the first half of the century in a range of
industries, a rapid and decisive movement to liberalise trade and otherwise
widen the range of the economy subject to competition, and an international
economic co-operation framework with clear common objectives and ideals all
helped create the potential for a business environment which was broadly
forward-looking, optimistic, as well as growth and innovation-oriented.

Despite the empirical and conceptual limitations in interpreting data, of what
may only be the beginning of a return to a robust pace productivity growth (as
in the period 1950-73), the OECD studies nevertheless chart very useful
avenues for continuing and deeper analysis with a better coverage of
countries.

Innovation is the enabler in the market and, at the OECD, there is
opportunity for innovation in analytical tools.

OECD studies go a considerable way to clarify that, in the last decade, a
significant portion of higher GDP growth experienced in some of the member
countries is due to higher factor utilisation. This aspect should not be lost in
the final political message. Ability to increase labour input clearly accounts for
the major part of higher growth per person in a number of countries (such as
Ireland, Korea, Netherlands), but it is also a significant contributor even in
countries with accelerating multi- (or total-) factor productivity (MFP) growth.

However, it is much more difficult to specify long-run growth paths for physical
and human capital. OECD studies explicitly recognise these difficulties and
carry out reasonable approximations, including (in the Economics Department
papers) an excellent treatment of the distinction between growth of capital
stock and growth of services of capital. The resulting finding that capital
services have grown at a higher rate than the capital stock (at least in the G7
countries) focuses attention on ICT investment, which has tended to have a
higher depreciation. The extent to which that is so largely depends on the
types of price depreciation used. However, these considerations should not
lead to a bias in the choice of price indices that generate higher multi-factor
productivity for short-lived assets. In other types of technology, such as
pharmaceuticals or biotechnology where investment cycles are long, this type
of measure might underestimate productivity improvements.

Innovation in policy initiatives should be derived from fact-based
analysis.

OECD studies still identify a significant role for MFP growth in most of the
(small number of) countries, which are reckoned as having increased their GDP
per capita growth rate. In turn, innovation is correctly recognised as the
primary vehicle for the growth of MFP. But, typically, innovation does not take
the form of a frontal wave of improvement starting in and moving through
simultaneously in all enterprises producing a given good or service. Instead,
technical innovation, usually accompanied by the appropriate managerial
innovation and adaptation, tends to be nurtured in a small number of firms and
once it bursts, or proves its success, tends to spread to the rest of the sector
by emulation.

To do that, leading firms must show initially a significant performance gain
above their competitors and at the expense of them. It is at this stage of the
innovation-productivity linkage that the fortunes of countries are likely to start
diverging, given widely different approaches prevalent within the OECD area to
the treatment of success and failure in business by public authorities. Indeed,
to the extent that government policy and practice has a strong bias towards
automatic and open-ended assistance to unprofitable enterprises, this is
bound to act as a brake on incentives both to emulate innovation and respond
by counter-innovation.

Innovation replaces the obsolete with opportunities for faster
growth.

Many executives in BIAC's membership have a strong impression that the
diffusion of ICT has been playing a major role in driving innovation and growth
performance in recent years. The study by the Directorate for Science,
Technology and Industry provides an eloquent treatment of the various
channels through which ICT and digital networks impact upon R&D and
business itself ultimately enabling higher productivity. But, it is crucial to
recognise that the impact of ICT can fuel improvement in productivity in the
economy as a whole only when that is juxtaposed with a business environment
which privileges ability to adapt.

This is especially true in the case of the new or newly popularised network
technologies, the Internet in particular and new business models based upon
them. At the initial stage, the most obvious and measurable impact of these
tends to take the form of dramatic cost reductions in the business process.

By necessity, the "sector" which is subject to technology-driven innovation
has to grow, albeit fast, from a small initial base. At that stage, therefore,
economy-wide improvements in productivity are correlated with the ability of
that economy first to permit the dissolution of some activities and second,
ability of the rest of the economy to move resources thus temporarily "freed"
to another uses. To the extent that public policy is focused on the protection
of existing activities, the process of new business and employment creation is
likely to be made slower.

Let us clarify our argument at this stage. Business is all too aware of the acute
shortages of labour experienced currently in the ICT producing industries per
se. There may even be some evidence for net job creation by ICT-led business
transformation is some places. The argument being made here is to focus
attention on the appropriate policy environment for channelling and using
creative destruction, rather than pretending it does not exist, for lack of
political expediency in it.

It is also clear that, due to new technologies but also the new market
environment (lower barriers to international movement of trade, capital and
production processes and greater competition in most markets), the velocity of
change with which innovation can spread in the economy is much higher
today. That makes it even more important to get the public policy and attitude
to "destruction" right. In this light, many companies traditionally thought to be
"old economy" are rapidly becoming "new economy" companies as they turn
to e-commerce to advance their business models and gain competitive
advantage. The object of policy discussion ought to be to identify the right
combination of factors which make that transformation more or less difficult.

Innovation in fathoming the complex market dynamic is a key role
for the OECD.

In comparison to twenty years ago, today's market dynamic occurs in a
significantly different environment. The degree if openness to trade is much
higher in most countries and so is the degree of globalisation of production
processes. Most product markets are characterised by a higher degree of
competition or, where this may be lacking, by intensifying political pressures
to open them up to competition, which may have a similar effect on the longerterm
expectations of managers. Regulatory reform is now widening the range
of competitive pressures to sectors which hitherto enjoyed significant
immunity from them.

Many sectors are experiencing a rapid transformation through mergers and
acquisitions and other forms of link-ups between enterprises, often across
political frontiers. Indeed, the overwhelming majority of foreign direct
investment is now reckoned to be associated with mergers and acquisitions, a
rapid rise in share in the past few years. *{4} Irrespective of whether each and
every merger is a success, this suggests a need to reorganise industrial
capacity, including services, at geographical scales independent from the
nation-state.

In some countries factor markets have been subject to significant amount of
reform and there is significant political pressure brought to bear on countries
which lag behind in this respect. Intuitively all these factors are bound to have
major effects on business climate and innovation climate. The very noticeable
contribution of ICT and network technologies is being added to the equation as
an additional - powerful and catalytic - factor.

However, there are important gaps between countries in terms of their status
and pace of movement along the various dimensions in which the business
environment is changing. It would be desirable to centre the analysis of a
phenomenon so central to the economy as the evolution of innovation and
productivity squarely within a framework which takes account of each of these
dimensions and the interactions between them.

The OECD is well suited to conduct such an analysis, and indeed focus much
of the follow up work of the "Growth project" in this fashion. In particular, the
Organisation is well positioned to draw upon the results of three of its wellknown
"horizontal" projects: the Jobs Strategy, Regulatory Reform, and
Electronic Commerce. BIAC believes that there may be a high return to a
project which synthesises the knowledge and understanding gained in these
separate analyses in terms of how their combination and interaction affects the
prospects for innovation and economic growth. Indeed, the Growth Project
Phase II should be primarily a beginning of this. In the following section we
attempt to suggest a few directions in which such a project might be developed
and explain further the motivation behind the idea.

Innovation in measurement is possible by means of a scoreboard.

Table 1. synthesises information from a number of different OECD studies to
lay out a suggested path towards a "scoreboard" on readiness for growth in
the new economic environment. As such, it provides a very rudimentary
panorama. Ideally, one would prefer to have a composite indicator for labour
market "adaptability" and another for "ICT/e-business readiness.” An
additional one should be on regulatory reform, or the conduciveness of the
product market regulatory environment for business creation. While not
spelled out in this paper, there has been valuable work done in PUMA and
elsewhere as part of the Regulatory Reform project. Performance in terms of
GDP per capita or MFP growth would then be compared with respect to these
three "dimensions" to reveal inter-relationships of causality between the
former and the latter.

We separate several groups of countries with respect to GDP per capita growth
performance, comparing 1980s and 1990s. The first group from the top (Korea,
Turkey, United States, Denmark, Luxembourg and Spain) is countries which
have maintained a relatively high growth rate from the 1980s to 1990s. The
next group (Australia, Norway and the Netherlands) has increased their growth
rate from lower levels to an above average rate in the 1990s.

The subsequent groups contain, respectively, high performers who have lost
some ground in the 1990s, and the rest. In each segment of the Table "better"
performance is indicated in bold. Across the two decades under consideration
only two countries in the entire OECD area have experienced both a speedup
of MFP growth rate, and increased labour utilisation and labour productivity,
Australia and the United States. These countries are experiencing
consequently the highest rates of per capita income growth in mature OECD
economies of any appreciable size.

Both MFP growth and labour utilisation have increased in Denmark and
Norway - in the latter, excluding the oil output outside the mainland, but not
necessarily the impact of oil wealth on the rest of the economy. Netherlands
has significantly improved GDP growth performance despite a fall in MFP
growth, thanks to an increase in both labour utilisation and productivity. In
Ireland, the star performer of the OECD in terms of increasing growth, the large
increase in GDP per capita growth rate is mostly explained by an increase in
labour utilisation, MFP growth was maintained unchanged at a high rate.

Lack of MFP data for Korea and Turkey make further analysis difficult, although
both countries have relatively high per capita GDP growth, Korea slowing
down from an even higher level. Perhaps Portugal could even be included in a
special category of declining but still high GDP per capita growth with
improving MFP growth.

Attempting to compare these observations with a measure of labour market
adaptability, we then look at the structural unemployment rate and its change
(columns (7) and (8) respectively) in the 1990s, as estimated by the OECD
Secretariat. *{5} Column (9) provides, for information, the absolute number of
country-specific recommendations made by the OECD Jobs Strategy and not
yet sufficiently acted upon according to the assessment of the OECD
Secretariat as of 1999. *{6} Each recommendation made for a given country does
not point to a similar amount of labour market inflexibility, but in general, a
large number of recommendations imply that a country's labour market
conditions leave much to be desired.

As a third dimension the ICT is glimpsed, quite superficially, by measures of
Internet penetration rates (column (12)) and the average price of 20 hours
Internet access as a ratio of monthly GDP per capita at purchasing power
parities (PPPs). There is at present, unfortunately, no comprehensive set of
indicators for assessing general e-commerce- or information-society readiness.
BIAC wholeheartedly supports significant and prioritised OECD
investment in developing such indicators and looks forward to its results.

In the meantime we have chosen to look at one facet of this dimension which
may be more crucial than others. Internet is certainly the most popularised
aspect of new information technology. As ICT hardware and software prices
are experiencing rapid decline, it is increasingly the cost of access to services
which is likely to be the key determinant of uptake in this technology - making
the need for effective liberalisation of basic and value-added telecommunications
services even more important. Even if the measure used here
is quite simplistic - business usage of Internet, which is more important for
innovation, and growth will not depend on the same price structure - we
nevertheless postulate that a higher price indicates higher inhibitors in the way
of e-business.

By and large, all countries which are associated in some way with good or
improving performance in GDP per capita growth are the ones which have
relatively lower NAIRU (less than 8 percent, which is roughly the simple OECD
average) and have lowered it in the 1990s. Some countries combine
acceleration in MFP with a reduction in NAIRU. Those could be declared the
stars of the new economy. All three countries, which have reversed their
growth fortunes (Australia, Norway, the Netherlands), have also improved their
labour market conditions appreciably. All countries with continuing high
growth performance (for which there is NAIRU data) have also lowered their
structural unemployment rate. Although in Spain the NAIRU still remains very
high, and Spain is the only country in this group with a deteriorating MFP
growth.

Turning to the Internet market readiness, there does not seem to be a clear
correlation between cost of access and any growth indicator, though it is
interesting to note that all three countries which have reversed their growth
performance in the 1990s (Australia, Norway, the Netherlands) have relatively
low Internet access costs. This is by no means shown as a proof of causality
from Internet costs to growth. But there may be an element of policy symmetry
in the sense that governments and countries which have been wise enough to
implement appropriate policies for economic growth and labour market
adaptability have probably also been good at creating the right conditions for a
dynamic Internet market.

Nordic countries are associated with higher ICT penetration rates and greater
presumed readiness for an e-business driven economy, which is illustrated by
lower access costs in the Table 1. But, only Denmark and oil-rich Norway
among them have been able to achieve a high or improved performance in
terms of GDP per capita growth. Whereas both Finland and Sweden have
experienced an accelerating MFP in the 1990s, this was coupled with a major
reduction in labour utilisation manifest in a sharp rise in structural
unemployment. However, empirical problems associated with trend
adjustment after the early '90s recession may be greatest for Finland and
Sweden and there may be serious risks in evaluating new-economy hypothesis
using data for these countries in the 1990s. In other words, the jury may still
be out regarding the impact of ICT on the economic performance of the Nordic
area. Regarding the high-performing Denmark, let us note in passing that,
although the country may share much of the welfare state tradition of the
Nordic area, it is distinguished by one of the most flexible labour markets when
it comes to job protection.

The Table is not free of paradoxes either. Canada is characterised by
improving MFP and a labour market which has at least not worsened. Indeed,
Canada also has relatively good indicators in Internet access and pricing.
Nevertheless, Canada does not seem to be able to break through a modest
GDP per capita performance in the last generation. Given the highly pro-active
posture of Canada in world-wide e-commerce policy dialogue it will be
interesting to see further understanding on the impact of ICT on productivity
and growth in that country. Canada's case also implies a deeper look into
some of the other factors which are associated in recent studies with the
success of a "new economy", namely a relatively younger and faster growing
labour force and an open immigration policy. At least in theory, Canada has a
greater ability to utilise a "US-style" policy and approach to innovation and
growth. A similar story could be told about New Zealand and partly the United
Kingdom.

Innovation must diffuse throughout the economy.

The examination of growth performance and market conditions provided above
is very sketchy and inconclusive from a scientific point of view. Its humble
intention is to suggest an avenue for more robust and in-depth analysis of
interaction between factor markets, product markets and technology-readiness
on the one hand and innovation and productivity on the other.

In any case, OECD growth studies this year only lead one on to better analysis
of ICT-led technical innovation and its relationship to productivity growth. The
treatment of the ICT embodiment in productivity growth and MFP-R&D linkages
requires considerable faith in econometrics to constitute conclusive evidence.
Under these circumstances, it would be wise to bring the Growth Study
endeavour to a broader context and explore the hypothesis that, "yes
technology-driven innovation does improve productivity and economic growth,
but only when other conditions affecting the adaptability of the economy to
change are also conducive."

If the line of reasoning of this discussion paper so far is generally accepted,
then it should be clear that there are certain inherent risks in adopting the "new
economy" as a Deus ex machina. Policymaking communities of OECD
countries are not immune from such temptations. Therefore, it is crucial for
the OECD to clarify the inter-relationship between what may be "new" factors
and the "old" knowledge and understanding that the organisation has been the
purveyor of on fundamental aspects of economic prosperity such as labour
market adaptability or regulatory reform.

In current debates regarding the old and new economy, we characterise the
new as the old with a turbocharger, the turbo being not just a technological,
static device but also the velocity of investment in innovation. Next step for
the research community is to identify the necessary architecture of the test
ring surrounding the turbocharger, so that the new machine does not run into a
wall of social resistance or indifference.

*partie=titre Better data *partie=nil

Business recognises the value of data produced by national statistical
organisations and international ones such as the OECD. Serious data, i.e.
blessed by governments who use up-to-date, often expensive but thorough 
methodologies are valuable and merit the expenditure. This is fundamental. To
be able to develop the technology dimension of this debate, serious and
internationally comparable data on such new indicators as software, ebusiness,
ICT services, biotechnology and science and technology in general
are necessary for a wide range of countries. Such data will enable productivity
measures reflect reality better. In particular, international harmonisation of ICT
measures is essential to a good understanding of underlying trends.
Internationally comparable data on services sector productivity is also a
priority investment.

As new and topical issues such as e-commerce, new economy or
biotechnology arise in the policy agenda and benefit from concentrated
attention of policy makers, the OECD should remain vigilant in terms of
reminding them how crucially the correct understanding policy choices
necessitates continuing investment in less glamorous work on statistical and
other indicators. That is the bedrock of the OECD.

As an additional issue, there is a general tendency for OECD analyses on many
aspects of the economy and the society to omit many of its member economies
when it comes to more sophisticated indicators. For instance, there may be
simple unemployment data for all member countries, but very often the more
sophisticated measures such as standardised unemployment rates or the
NAIRU are not calculated for many member states such as Korea, Mexico,
Turkey or the new members from Central/Eastern Europe. Likewise, general
GDP data may be available for all countries, but more detailed measures of
GDP per hours worked, capital deepening, or MFP are unavailable for many
member countries.

These are only illustrations of a general trend. Some member countries are
covered in OECD statistics just about as well as non-member countries! This
is not a desirable situation from the point of view of those who need to
understand economic trends in the world, especially because some of these
countries are experiencing more rapid change. To the extent that these
omissions are due to the inability of the countries in question to provide the
necessary underlying information, BIAC calls upon their ministers in charge of
the economy to recognise the importance of connecting their countries to the
OECD better in this regard and prioritise statistical work. The cost of ignorance
is rising.

*{SOURCES OF DATA IN TABLE 1.

By Column:
(1) - (3): ECO/WP1(2000)6/ANN1, Economic Growth in the OECD Area: Recent
Trends at the Aggregate and Sectoral Level, 2000, Table 1 and ANN4, Table A4.5.
(4) - (6): Knowledge, Technology and Economic Growth: recent Evidence from
OECD Countries, Andrea Bassanini, Stefano Scarpetta, Ignazio Visco, May 2000,
mimeo, Table 4.
(7) - (8): ECO/WP1(2000)2/ANN1, The Concept, Policy Use and Measurement of
Structural Unemployment, 2000, Table 1.
(9): OECD, Implementing the OECD Jobs Strategy. Assessing Performance and
Policy, Paris, 1999. Table B.5., pp. 182-183.
(10) and (12): DSTI/ICCP/TISP(2000)1, Local Access Pricing and E-Commerce,
2000, background data of Figure 5 kindly provided by the OECD Secretariat.
(11): Column (10) recalculated as a percentage of monthly GDP/capita at PPPs from
OECD in Figures, 1999, p. 79.

NOTES

1 An approximate calculation shows that international trade in ICT goods (including
software) is at least 59 percent of world-wide ICT production (including software and ICT
services). This is almost double the ratio of world trade (exports plus imports) in goods to
world GDP, which is 37 percent. Excluding services from the denominator (world-wide ICT
production) as well would increase the share international trade even further. All data are for
1998. Sources: trade in ICT from OECD IT Outlook 2000, tables 3 to 7; ICT production from
European Information Technology Observatory 2000; World trade and GDP from World
Bank, World Development Indicators 1999.
2 For a recent recapitulation of evidence on this theme the following is highly
recommended: Richard N. Langlois and Edward Steinmueller, "The evolution of competitive
advantage in the worldwide semiconductor industry, 1947-1996", and Timothy F. Bresnahan
and Franco Malerba "Industrial dynamics and the evolution of firms' and nations' competitive
capabilities in the world computer industry", both in David D. Mowery and Richard R. Nelson,
eds., Sources of Industrial Leadership. Studies of Seven Industries. Cambridge University
Press, 1999.
3 "A New Economy?: The Changing Role of Innovation and Information Technology in
Growth" (DSTI/ICCP/IND/STP(2000)/FINAL) and " Economic Growth in the OECD Area: Recent
Trends at the Aggregate and Sectoral Level "(ECO/WP1(2000)6), and a number of supporting
analyses such as "The Contribution of ICT to Output Growth" (DSTI/EAS/IND/SWP(99)4).
4 Source: UNCTAD World Investment Report 1999, pp. xxi-xxii.
5 Source: "The Concept, Policy Use and Measurement of Structural Unemployment"
(ECO/WP1(2000)2/ANN1/CORR1) Table 1. Structural unemployment rate refers to the "Nonaccelerating
inflation rate of unemployment" (NAIRU).
6 OECD, Implementing the OECD Jobs Strategy. Assessing Performance and Policy, Paris,
1999. Table B.5., pp. 182-183.}