information technology, venture capital and the stock market

1
INFORMATION TECHNOLOGY, VENTURE CAPITAL
AND THE STOCK MARKET
Ajit Singh, Alaka Singh and Bruce Weisse
University of Cambridge
Address for Correspondence:
Professor Ajit Singh, Faculty of Economics and Politics, University of Cambridge, Cambridge
CB3 9DE
Email: as14@econ.cam.ac.uk
Telephone: +44 1223 350434
This paper was prepared as a background paper for the International Labour
Organisation’s World Employment Report 2000-2001.
2
|Contents
I.
Introduction
II.
Bank-based vs. Stock market-based financial systems
1
and new technology
3
III.
Empirical evidence: A preliminary look
10
IV.
Multivariate analysis
14
V.
Venture Capital
17
VI.
Conclusions
21
Tables
Table 1.
Stock market performance: 1990-1999
Table 2.
Table 3.
Table 4.
Indicators of stock market development in selected countries
Indicators of ICT development in selected countries
Variables included in the ISI composite index
Table 5.
Medians of quartiles of full sample
Table 6a.
Table 6b.
Table 6c.
Table 6d.
Regression results: Dependent variable – Mobile phone
Regression results: Dependent variable – Internet hosts
Regression results: Dependent variable – Personal computers
Regression results: Dependent variable – High Tech Exports
as a per cent of Manufacturing
Regression results: Dependent variable – ISI scores
Correlation matrix
Venture capital investment
United States and Germany: Capital raised by venture
Table 6e.
Table 7.
Table 8.
Table 9.
finds by type of investor
Table 10.
United States and Germany: Venture Capital disbursements
by stage of financing
Table 11.
Information technology in India
Table 12.
India’s software exports by activity
Figures
Figure 1.
Figure 2.
Price earnings ratios
Amazon.com stock market valuation and profitability
3
INFORMATION TECHNOLOGY, VENTURE CAPITAL AND THE STOCK
MARKET
I. Introduction
This paper investigates the relationship between information technology and the capital
markets. The central analytical and policy question addressed here is what kind of
financial system or capital market arrangements are most conducive to fostering
information technology and its use in the economy.
This question is closely related to an old debate about the relative virtues of the AngloSaxon financial system based on stock markets, and the German/Japanese bank-based
financial model. Which system should developing countries attempt to emulate to
foster their economic growth and technological development? This debate has recently
taken a fresh turn with the apparent emergence of the “New Economy” in the U.S.
The U.S. has not only experienced fast growth of ICT industries but there has also
evidently been widespread successful adoption of ICT technology in many areas of the
economy. It is suggested that a major reason for the U.S. lead in this area, and the
apparent European and Japanese lag, has been the very important enabling and
stimulating role of the stock market.
As Larry Summers, the U.S. Treasury Secretary, points out, “financial markets have
played a central role, making available resources for guys who can raise a million
dollars before they can buy their first suit.” Similarly, Martin Feldstein, President of
the National Bureau of Economic Research, suggests that “it may be that the nature of
this technology is particularly favourable for the U.S., there are all kinds of facilitating
characteristics here – the venture capital market, incentive-based rewards for
managers.”1
The critics of the stock market are not, however, entirely convinced by these
arguments. They suggest the jury on this matter is still out. In their view, what the
recent developments on the stock market indicate is, to the contrary, the market’s
periodic irrationality. It is impossible, in their view, to have a rational explanation for
4
the astronomical price/earnings ratios of high technology firms that are not making any
profits or distributing any dividends. In other words, the stock market critics regard
the extremely high valuations of the dot.com companies during the period January
1996 to March 2000 as a classic bubble with potentially harmful consequences for the
economy when the bubble bursts.2
Summers, who in the past was critical of the short-term focus of the U.S. stock
market, now suggests that “increasing pressure for performance for shareholders” has
played a crucial role. “I think our financial markets should get a lot of the credit for
forcing money out of traditional management and entrenched corporations, and
preventing what would have been negative internal rates of return on investments.” He
goes on to point out that the pace at which companies mature has greatly increased:
“On conventional estimates it used to take five years to build a business to the point at
which venture capital would be entering. Now it’s less than a year.”
However, in the view of the critics, the evolution of the stock market valuation of
technology companies (the “New Economy”) and “Old Economy” companies does not
reveal a healthy situation. According to the Federal Reserve Bank of San Francisco,
technology companies in the U.S. accounted for only 7% of total stock market value in
1990 but by March 2000 this share had risen to 36%, a fivefold increase. However, the
share of employment accounted for by the technology companies rose from 6% in
1990 to only 9% in March 2000, while their share of sales increased from 6% to 10%
in the same period. Even though technology companies had faster sales growth than
old economy companies, the latter had faster earnings growth.3 Evidently, the very
high valuations given by the stock market to technology companies reflects investors’
willingness to pay a premium for future growth opportunities. From the point of view
of the economy as a whole, whether the implied diversion of resources from the old to
the new economy will lead to a superior or worse allocation depends on the extent to
which these valuations are justified by fundamentals.
1
All the quotations are from the Financial Times, “Winning ways: ready bucks and a flair for risk”,
14 December 1999.
2
Nasdaq rose faster than any other index between 1996 and spring 2000. However, the index has
been unusually volatile during the millennium year 2000, in November 2000, the index fell below
3000, a value last reached in November 1999. These issues and their significance are discussed more
fully in the following sections.
3
FRBSF Economic Letter, Number 2000-15; May 12, 2000.
5
Although the U.S. is apparently the leader in information technology, notable progress
has been made in this area by other industrial countries as well as by some developing
countries. These countries have different financial institutions, different industrial
structures and varying roles of the state in the economy. Such inter-country
differences provide us with an opportunity to see to what extent variations in ICT
development can be ascribed to differences in financial systems, controlling for other
relevant variables.
The paper is organised as follows. Section II briefly considers the theoretical,
conceptual and empirical issues concerning the relationship between technological
development and the financial system. It reviews the new literature on the subject and
assesses the controversy surrounding it. Section III provides a preliminary look at the
empirical evidence with respect to stock market development and the development of
ICT technology in various countries. The data pertain to 60 developed and developing
countries in the 1990s. Section IV carries out a multivariate analysis of the
determinants of ICT with particular reference to the role of stock market development.
The central hypothesis tested is that the more fully developed and mature a stock
market a country possesses, the faster, other things being equal, will be its
development and use of information technology. In this analysis, four individual
indicators of ICT development and one composite measure are used as dependent
variables while three indicators of stock market development are used as independent
variables together with relevant control variables. Section V considers the role of
venture capital in ICT development as well as, importantly, the relationship between
venture capital and stock market activity. It also comments on the significant role of
the government in the financing of venture capital for ICT development in many
developed as well as developing countries. Section VI examines the question of the
optimal size of the firm in relation to ICT expansion. This section also outlines case
studies of individual firms from both emerging markets and advanced economies.
Section VII concludes and draws out the policy implications of the analysis.
II. Bank-based vs. Stock market-based financial systems and new technology
6
The relationship between the financial system and technological development has long
been controversial. John Hicks (1969) argued on the basis of both economic history
and theory that Britain's industrial revolution was only made possible by the
development of financial institutions. The basic argument is that technology existed
long before the industrial revolution but could not by itself generate sustained
economic growth. The large-scale capital requirements of the industrial revolution
could only be met by the development of capital market institutions that permitted the
pooling of small individual savings into large funds for industrial development. An
equally famous theorist, Joan Robinson, however, suggested that the causation is the
other way around, finance follows where enterprise leads - in other words, financial
institutions will spontaneously develop and evolve where there is a demand for capital.
Technological and industrial development in this view is constrained by demand rather
than by the supply of finance (Robinson, 1952).
More recently, following the development of endogenous growth models, formal
models of finance and technological development have emerged in the neoclassical
tradition. King and Levine (1993) suggest that risk diversification made possible by
the development of stock markets positively aids innovation. Holding a diversified
portfolio of new technological products reduces risk and leads to greater investment in
new technology than would otherwise be the case.
As noted above, the apparently successful ICT revolution in the U.S. has brought to
the fore again the old debate on the relative merits of the stock market-based (U.S.,
U.K.) versus bank-based (Germany, Japan) financial systems. A large literature until
recently attributed the much greater success of countries like Germany and Japan in
international markets to the superiority of the bank-based system which promoted
close and long-term relationships between corporations and banks. These relationships
enabled Japanese and German corporations to have a long-term investment perspective
rather than being concerned overwhelmingly with the short-term profits expectations
of market analysts. A large number of theoretical contributions modelled the shorttermism of the stock market and its harmful consequences for the rate of investment, a
variable critical to technological development (see Stein (1988,1989) and Singh
(2000)). Under the leadership of Michael Porter, Harvard Business School carried out
a large empirical study of the U.S. financial system. On the basis of the findings of this
7
research, Porter (1992), summed up the situation as follows: “the change in nature of
competition and the increasing pressure of globalization make investment the most
critical determinant of competitive advantage … Yet the U.S. system of allocating
investment capital both within and across companies is failing. This puts American
companies at a serious disadvantage in global competition and ultimately threatens the
long term growth of the U.S. economy.”4
The U.S. financial system and ICT: The pros
Today the table appears to have turned. The American financial system is now
regarded as being particularly conducive to innovation, specifically in relation to ICT.
The huge investments in new technology firms in the U.S., despite their zero or
negative short-term profits, is an obvious refutation of the short-termism of the stock
market. Further, there are theoretical models that indicate that stock markets may be
better at choosing the technological winners than the bank-based systems (Allen,
1993). This is because it is argued that stock market prices reflect the collective
judgement of the public compared with the decisions of a small banking committee or
the loan officer in the bank-based system.
Other aspects of the American financial system which, it is thought, are highly suitable
for information technology include the system of incentives, rewards and punishments.
Specifically, it is suggested that the stock market-based U.S. financial system has
enabled the widespread use of stock options as a means of payment to those who work
for new technology companies. This helps to align the interest of the employees with
those of shareholders leading both to greater social efficiency and greater reward for
innovations. The latter derives in part from the existence of an exit mechanism which
the U.S. financial system provides in the form of IPOs and take-overs. Both these
avenues are thought to improve enormously the rewards for innovations (compared
with other financial systems that do not have such mechanisms).5 Last, but not least,
the take-over mechanism in the U.S. financial market which allows for hostile
acquisitions is regarded as being particularly helpful in the selection process, i.e. in
4
Porter (1992), p.65. This paper reports the findings of a large project sponsored by the Harvard
Business School and the Council on Competitiveness, a project that included 18 research papers by 25
academic experts.
5
See Black and Gilson (1998). This issue is discussed further in Section V.
8
being able to discriminate between useful technologies which benefit society by
increasing shareholder value and those which do not.
The U.S. financial system and ICT: The cons
These are indeed formidable advantages for the U.S. financial system in relation to
ICT; the claims for these have been further strengthened by the evidently successful
introduction and widespread usage of the new technology in the U.S. economy.6
However, the debate is far from being over. First and foremost it will be observed that
the above virtues of the stock market system depend crucially on the nature of the
stock market pricing process. If share prices always accurately and exclusively
reflected the true long term expected profitability of firms, the case for the virtues of
the stock market system will have a more solid basis. Orthodox financial economists
believe this would indeed be the end result of a pricing process based on rational
expectations of investors and that their similar beliefs about the future prospects of
companies. However, the prices may well be generated by altogether different
processes where investors base their decisions on irrational exuberance and are
motivated by speculative profits. The basic mechanism of such an alternative pricing
process is neatly described by Keynes’s famous beauty contest analogy. Keynes
(1936) observed in Chapter 12 of the General Theory that
professional investment may be likened to those newspaper competitions in
which the competitors have to pick out the six prettiest faces from a hundred
photographs, the prize being awarded to the competitor whose choice most
nearly corresponds to the average preferences of the competitors as a whole; so
that each competitor has to pick, not those faces which he himself finds prettiest,
but those which he thinks likeliest to catch the fancy of the other competitors, all
of whom are looking at the problem from the same point of view. It is not the
case of choosing those which, to the best of one’s judgement, are really the
prettiest, nor even those which average opinion genuinely thinks the prettiest.
We have reached the third degree where we devote our intelligence to
anticipating what average opinion expects the average opinion to be. And there
are some I believe who practise the fourth, fifth and higher degrees. (Keynes,
p.156)
6
The success of the ICT revolution in the US itself is not fully accepted by all analysts. Some regard
the recent upsurge in US productivity growth as simply a cyclical phenomenon. However, it will be
fair to say that increasingly it is being acknowledged that there has indeed been a trend rise in US
productivity growth over the last five years and that it is largely due to the adoption of ICT by
American enterprises. See further papers by Gordon (2000), Jorgensen et al (2000).
9
Which of the above two views of the stock market pricing process is more accurate is
therefore a crucial question. Not surprisingly, it is also a controversial one. In
interpreting empirical evidence on this issue, Tobin (1984) makes a useful distinction
between ‘fundamental valuation efficiency’and ‘information arbitrage efficiency’.
When financial economists claim that stock prices are ‘efficient’they are in fact
referring to the latter concept of efficiency. This simply refers to the fact that all
information is rapidly circulated in the market, any new information is more or less
immediately discounted by market players so that no gains are to be made from any
publicly available information. There is however no necessary correspondence
between this information arbitrage efficiency and fundamental valuation efficiency, and
it is the latter which is required if stock prices are to perform their task of efficiently
allocating resources in the economy as a whole. There are a number of theoretical
models as well as empirical evidence which suggests that share prices often depart
from fundamentals, being influenced by whims, fads, fashions and irrational pessimism
or exuberance.7
In terms of this critical analysis, the nature of the relationship between the new
economy and the stock market, rather than being regarded as a virtue of the American
financial system in facilitating the arrival of the new economy, becomes, instead, a
cause for concern. There is important evidence that suggests that technology stocks
are grossly overpriced. Shiller (2000) has carefully constructed data on real priceearnings ratios in the U.S. economy over a long time period, from 1881 to 2000 (see
Figure 1). During the present stock market boom that began in 1992 and gathered
pace during the last five years, the average real price-earnings ratio reached a value of
44.3 in January 2000. This compares with a peak value of 32.6, the highest ever
recorded before, reached in September 1929 on the eve of the Great Depression.
After this peak, the S&P index fell by 80 per cent for the next three years and did not
regain its 1929 value until 1958.
The unprecedented levels of price-earnings ratios for most of the 1990s are largely due
to the stock market’s very high valuation of the new economy stocks relative to that of
the old economy. If Dow Jones is regarded as representing mainly the old economy
7
See further Camerer (1989), Singh (1999), Shiller (2000), Shleifer (2000) and the Journal of
Economic Perspectives special issue on ‘bubbles’, 1990.
10
and Nasdaq as the representative index for the new economy, the price-earnings ratio
of the average new economy stock in April 2000 was 62 compared to 23 for the old
economy.8 This was so despite the fact that the Nasdaq index had fallen by about 15
per cent between March and April.
Most economists regard such valuations of technology stocks to be unrealistic and as
representing a classic case of a stock market bubble. This is perhaps brought home
more clearly by taking a closer look at some of the individual stocks rather than the
market averages. Figure 2 provides data for the last ten years on share prices and
profits of the foremost icon of the new economy, Amazon.com. The share price of the
company has been rising rapidly while it has been making increasing losses in each
successive year over the entire period. Another case that depicts even more vividly the
irrational exuberance and speculative character of stock market prices for technology
companies is that of a recent British IPO. The Financial Times (September 22, 2000)
observes:
It is often an amusing, if futile, exercise to read the ‘investment considerations’
section of a prospectus. In the case of Arc International, the customisable chip
designer, this ran to nine pages and advised investors that the company had never
made, and might never make, a profit; that if it did that profit might not be
sustainable; that revenues were likely to be volatile, unpredictable and subject to
factors outside the company’s control; that all manner of dreadful things could
happen that would have a material adverse effect on the company and its share
price; and that anyone buying the shares would ‘experience substantial and
immediate dilution in the net tangible book value of their investment.
Investors in technology stocks are made of stern enough stuff to set aside such dire
warnings, and the falls in the sector since the bookbuilding exercise began two
weeks earlier. Knowing what happened to other chip company flotations, they
were not going to miss this one. The issue was a great success and yesterday the
shares more than doubled in first dealings.
Shiller (2000) has rigorously considered a wide range of structural factors that could
justify the present high price-earning ratios in terms of fundamentals. He specifically
examines the role of the internet, the baby boom and other factors such as the decline
of inflation and the growth of mutual funds and finds that none of them individually or
collectively provides a satisfactory explanation for the observed rise in the average
price-earnings ratio.
8
Federal Reserve Bank of San Francisco (2000).
11
The prices of technological stocks, apart from being speculative, are also highly
volatile. This is shown by a comparison of the standard deviation of Nasdaq indices
with those of the Dow Jones, S&P 500, FT100 as well as the Nikkei indices (see Table
1). The volatility of Nasdaq is considerably higher than those of other index numbers
reported in Table 1, except for Nikkei. Share price volatility is however a negative
feature of stock markets for several reasons. First, it reduces the efficiency of the price
signals in allocating investment resources. Secondly, it increases the riskiness of
investments and may discourage risk-averse corporations from financing their growth
by equity issues and indeed from seeking a stock market listing at all. Thirdly, at the
macroeconomic level, a highly volatile stock market may lead to financial fragility for
the whole economy (Singh 1995, 1999).
Share prices of dot-coms, the high technology companies, have been particularly
volatile during the millennium year 2000. Since its peak in March 2000, the index had
fallen by 45 per cent by the end of November, trading below 3000 for the first time
since November 1999. By the end of October 2000 the index had fallen by 36% from
its record high. The volatility of dot-com companies share prices and their rise and fall
is best illustrated by considering cases of individual firms from various sectors of the
technology market. The share prices of the leading e-commerce company
Amazon.com fell from a high of $113 to $36 towards the end of October. The share
prices of a flagship B2B company, VerticalNet, rose to $140 a share, but on 27th
October, 2000 the share price had fallen to $25. The internet infrastructure company
Nortel Networks has halved in value during this year. The ultimate blue chip of the
technology world, Cisco Systems, has fallen by about 38 percent. 9
A main issue is whether these sharp falls represent the final bursting of the technology
bubble, or whether there is still some way to go before prices reach rock bottom.
Opinions differ on this score. The stock market historian David Schwartz notes in the
Financial Times10 that the bursting of the bubble image of a stock market collapse of
share prices is not accurate. He suggests that historical evidence is more in accord
with the alternative analogy of a slowly leaking tyre – prices ultimately fall to very low
levels but it may take a long time to reach that position. For example, the London
9
These data are from the International Herald Tribune, 30th October, 2000 in an article by David
Ignatius.
12
stock market downturn of 1973-74 was a result of a number of separate ‘waves’, each
one plunging prices even lower. This was the pattern of both the Wall Street )1929-32)
and the Japanese (early-mid-19990s) downturns as well. Thus the complacent view
that the technology stocks were overpriced and that since then the market has
corrected the position may turn out to be incorrect. The Economist (November 25
2000) reports that the 20 most valuable technology, media and telecom (TMT) still
have a price-earnings ratio of 55, down from 78 earlier in the year but still well above
the long term median for the sector of 33. On the other hand, in the old economy,
despite a smaller flow in the S&P index, the typical medium sized firm (market
capitalisation of £450 m. - $ 8.6 b.) has a price-earnings ratio of 16 which is close to
the historical median.
There is also a downside to the use of stock options as a means of payment to
employees in the new economy. This negative feature is represented by the increased
income inequality in the U.S. in the recent period. Although this rising income
inequality may be attributed to a number of factors (e.g. globalisation, skill biased
technology), the growing use of stock options has also been a contributing factor.
This is particularly true in relation to the widening income gap between the top ten per
cent and the median.11
It is significant that despite the large fall in share prices, particularly on Nasdaq the net
sellers of shares have been experienced institutional investors. Individual investors are
still continuing to invest in mutual funds and have not been scared away by the falling
stock market prices. However if and when the small investors flee the market, the
consequences for the economy could be quite serious. The U.S. may be lucky and
achieve a soft landing, unlike the Japanese who got trapped in a prolonged period of
slow growth following the end of the bubble in the late 1980s. However, many leading
economists believe that there are such serious imbalances in the U.S. economy (e.g. the
current account deficit, the negative savings of the private sector) that a soft landing
may not be possible. There is a certain irony in the leading U.S. officials attributing the
East Asian crisis in part to the absence of reliable price signals due to crony capitalism
and the nature of the relation between government, corporations and banks in these
10
11
Financial Times, October 28, 2000.
See Singh and Dhumale (2000).
13
countries. The above analysis indicates that the share prices in the new economy were
distorted during the boom years, not due to government intervention but rather
through the market processes themselves. These may also therefore not be a useful
guide for resource allocation.
One of the claimed virtues for the U.S. financial system is the availability of the takeover mechanism and its absence in Germany and Japan. However, both analysis and
evidence suggest that the stock market selection process in the real world is far from
being efficient in the sense that it does not select for survival high performing firms and
punish the poor performers. Evidence suggests that the selection in the market for
corporate control takes place only to a limited extent on the basis of profitability and
stock market valuation but mainly on the basis of size. A large but relatively
unprofitable firm has a greater chance of survival than a small profitable company
(Singh 1975, 1992; Meeks 1977 and Hughes 1991).
One issue which is raised by the extremely high valuation of the New Economy relative
to the Old Economy stocks is that the market has supplied so much capital to new
technology firms that they cannot use it productively in their own enterprises. To
some extent it will be conspicuously consumed or fuel a take-over binge on the part of
the New Economy firms, a good example being the takeover of Time Warner by
America Online. Some may think that is as it should be – by this means the New
Economy is able to increase the efficiency of the Old. However, it is far from certain
that the managers of the New Economy firms will even know how to run Old
Economy businesses, let alone enhance their efficiency. It is more than likely that the
net effect of the New taking over the Old may be considerably negative for the
economy as a whole.
To sum up, the theoretical case for a stock market economy as being particularly
conducive to fostering technical change is far from being unequivocal. The analysis and
evidence reviewed above suggests that the stock market based U.S. financial system
has both positive and negative features in relation to promoting technological change.
There is yet inadequate data to arrive at firm conclusions on this issue. However, the
broad controversy over the question of the superiority of one financial system over the
other cannot be based on the experience of the U.S. alone. It is necessary to consider
14
other countries both with systems similar to those in the U.S. (such as the U.K.,
Canada, Australia, etc.) and those that possess markedly different systems (Japan,
Germany, continental Europe). With respect to the policy question as to which
advanced country system, if any, is more suitable for developing countries, it is also
necessary to consider the actual experience of these countries so far. These
quintessentially empirical questions are considered below on the basis of data for a
large group of emerging and developed country markets.
III. Empirical evidence: A preliminary look
This section and the following one will empirically investigate the relationship between
stock market development and the development and usage of ICT technology on the
basis of data for a large number of developed and emerging market economies. The
present section will provide a preliminary univariate and bivariate analysis of the data
while Section V will carry out a multivariate analysis.
The full sample used in this survey contains observations from 63 developed and
developing countries on fourteen variables. Three variables, averaged over the 19901995 period, relate to economic output and growth (GDP, GDP growth and GDP per
capita); five variables, also averaged over the 1990-1995 period, relate to stock market
development (market capitalisation, market capitalisation as a percent of GDP, value
traded, the inverse of the turnover ratio, and the number of listed companies). ICT
development and usage are represented by six variables that are taken from the late
1990s: mobile phones, personal computers, and internet hosts, respectively, per 1000
people; high technology exports as a percentage of manufacturing exports, scientists
and engineers in R&D, and a composite index of ICT development and usage. A
smaller sample of 33 advanced and emerging market economies, summarised in Tables
2 and 3, has also been included covering the same variables and time periods.
The variables relating to economic production and growth are of importance in
unravelling the determinants of ICT development. It is clear that we would expect a
priori that countries with higher levels of per capita GDP to have a higher degree of
ICT development. Similarly, we might expect countries with a higher rate of GDP
growth to have higher investment rates and thus, ceteris paribus, greater ICT
development. We would also expect some disjuncture between the GDP per capita
15
variable on the one hand and GDP growth on the other. Theory suggests that less
developed countries further behind the technological frontier can achieve, ceteris
paribus, higher growth rates than countries at the frontier since they can take
technology “off the shelf” and dramatically improve their productivity and growth. At
the frontier, countries are limited to more marginal improvements in technology that
can generally be expected to have only smaller effects on the growth rate. The ICT
“revolution” may have a greater effect on labour productivity in advanced countries
but it remains to be seen how large this turns out to be.
Regarding the stock market data, empirical studies have shown that these indicators
are the best for revealing the extent and depth of equity market development. Market
capitalisation is the market value of all the companies traded on the stock exchange,
while value traded is the total value of equities traded on the exchange in a given year.
The turnover ratio combines both variables – it is defined as the ratio of value traded
to market capitalisation – and thus provides a measure of liquidity in the market
(please note that we have used the inverse turnover ratio in this paper, therefore the
higher the value the lower the turnover and liquidity). It is claimed to be a more
important variable than the market capitalisation to GDP ratio as a determinant of the
level of development of the stock market since it measures the degree to which easy
entry and exit from the market is possible. Given the greater sophistication of
developed country markets and in particular their more efficient and streamlined order
and payments systems, we would expect this variable to be negatively related to per
capita GDP (that is, the inverse turnover ratio decreases – markets become more liquid
– as per capita GDP increases). Market capitalisation as a percentage of GDP
indicates the relative size of the stock market in relation to the national economy, but
this variable is found to be a less reliable guide to the extent of stock market
development than the turnover ratio (see Levine 1997). The number of listed
companies is another indicator of stock market development and one that is
particularly important in the context of ICT development since it gives an indication of
the number of IPOs.
As noted above, the development and usage of ICT technology is reflected in a set of
six variables. The first three variables capture the use of mobile phones, personal
computers and internet hosts per 1000 people in the population. We would expect
16
these variables to be highly correlated with GDP per capita. High technology exports
as a percentage of manufacturing exports is a rough measure of the sophistication of
the country’s technological base. We would expect in general a positive – though not
necessarily linear – relationship between high technology exports and per capita GDP.
The relationship may not be exact because multinationals from OECD countries have
significantly expanded their production platforms in emerging market economies (such
as, for instance, Malaysia) from which they export to developed countries. The
number of scientists and engineers in research and development also provides a
measure of the sophistication of the country’s technological base and can be viewed as
an explanatory variable that helps determine the degree of ICT development.
The final variable, the “ISI score”, is a composite index based on four broad categories
measuring ICT infrastructure development and informational and social freedom
compiled by the Information Society Index. Table 4 presents the 23 variables in the
four main categories that comprise the index - computer, information, internet and
social infrastructure. The broad scope of this composite variable make it an excellent
measure of the relative standing of countries in ICT technology and thus an effective
dependent variable with which to test the determinants of ICT development.
The data presented in Table 2 for the small sample of countries conforms to our
expectations. It reveals the obvious differences in the median values of GDP and GDP
per capita between developed and emerging market economies. It also indicates the
far higher median growth rates of GDP in emerging markets over the 1990-1995
period (5.2% versus a median of 1.6% in developed countries). The data also indicate
that the inverse turnover ratio is both higher (a median value of 3.5 for emerging
markets versus 2.2 in developed markets) and more variable in emerging markets than
in developed country stock markets. The median value, however, somewhat disguises
the fact that many major emerging markets have very high trading values relative to
market capitalisation and hence very liquid exchanges. Brazil, China, Korea, Malaysia,
Thailand and Turkey each have (inverse) turnover ratios below the developed country
median. The other variables exhibit the expected behaviour, with developed countries
having higher median values for market capitalisation, market capitalisation as a share
of GDP, value traded and the number of listed companies. It should be noted,
however, that the values for developing countries exhibit a high degree of variability.
17
For example, Chile, Jordan, Thailand and South Africa all have market capitalisation to
GDP ratios far above the median developed country value. Malaysia had a ratio more
than four times higher than the developed country median.12 In terms of the number of
listed companies, several countries had numbers above the median for developed
countries. India was particularly noteworthy, having over nine times more listed
companies than the developed country median.
The data presented in Table 3 reflect the development and usage of ICT technology in
the small sample of countries. As expected, the results show a clear relation between
ICT development and per capita GDP. The usage of mobile phones, personal
computers and internet hosts are all substantially higher in developed countries.
Similarly, the percentage of high technology exports in manufacturing trade and the
number of scientists and engineers are significantly higher in developed countries. Not
surprisingly, the composite ISI score – which measures the degree to which a country
has developed ICT capabilities – reflects the global “digital divide” and is thus heavily
weighted in favour of developed countries (a median value of 2815 versus 214 for
emerging markets).
The same patterns can be seen in the summary of the findings for the full sample
contained in Table 5. The division of the sample into four categories reveals the nonlinear relationship between the growth rate of GDP and per capita GDP. Similarly, the
ratio of high technology exports to manufacturing exports also shows a non-linear
relationship, with “rich” countries having a higher percentage than “very rich”
countries (a median value of 20% versus 16%) and “very poor” countries having a
higher percentage than “poor” countries (a median value of 7% versus 4%). The
inverse turnover ratio shows a clear negative relationship with per capita GDP – the
higher the level of per capita GDP the more liquid the stock market. The ISI score is
also strongly positively related to per capita GDP, rising from a median value of 923
for “very poor” countries to 4174 for “very rich” countries. The other variables
related to stock market development and ICT development and usage are highly
positively related to per capita GDP. Thus, in testing the hypothesis that the greater
the development of a country’s stock market the greater will be its ICT development –
12
Note that this is an average of the 1990-1995 period – the period before the Asian financial crisis.
18
which will be carried out in the multivariate analysis in the next section - it is important
to control for per capita GDP.
IV. Multivariate analysis
The informal bivariate analysis of the last section indicated that there is a close
relationship between per capita income and most of the variables representing ICT
development. There also seemed to be a generally positive relationship between per
capita income and stock market variables.
Here we investigate the relationship between the variables measuring ICT development
and those pertaining to stock market development by means of multivariate regression
analysis. The latter controls for the effects of per capita GDP and economic growth.
The following regression model was fitted to cross-sectional data from 63 countries
including both emerging markets and developed economies. The choice of countries
was dictated entirely by the availability of data.
ICT development indicator
= β1constant
+ β2 GDP growth rate
+ β3 GDP per capita
+ β4 number of scientists and engineers per 10,000
+ β5 stock market capitalisation ratio
+ β6 reciprocal of the turnover ratio
+ β7 number of listed companies
+ U random error term
This equation was fitted successively with each of the five indicators of ICT
development as a dependent variable, i.e., (i) mobile phones per 1000 population; (ii)
personal computers per 1000 population; (iii) internet hosts per 1000 population; (iv)
high-technology exports as a per cent of manufacturing exports; and (v) ISI scores.
For each of these dependent variables, four regression equations were fitted - one with
all the independent variables listed above and the three others, each successively
keeping only one of the three stock market variables as an explanatory variable and
dropping the other two. The reasons for adopting this procedure is that the three
variables are correlated and it may be difficult to isolate the influence of each one when
they are considered together in the same equation. Tables 6a-6e therefore report
results on fitting twenty regression equations to the data.
19
Table 7 is a bi-product of the regression analysis and reports the zero order intercorrelation for all variables used in the analysis. The following notable points emerged
from this table:
(i)
there appears to be a relatively close positive relationship between four of the
five dependent variables and per capita GDP. For example, the simple
correlation coefficient between per capita incomes and personal computers per
1000 is 0.91, with mobile phones per 1000 is 0.68, with number of scientists
and engineers per 10000 is 0.91, with the composite ICT index (see last row) it
is 0.92. However, the correlation coefficients relating per capita GDP to hightech exports is only 0.20.
(ii)
the correlation coefficient between GDP per capita and the three stock market
variables is small but has the expected sign in each case.
(iii)
the relationship between the five dependent variables themselves is generally
quite close, with the composite index ISI having a correlation coefficient of
0.96 with personal computer per 1000, 0.94 with scientists and engineers, 0.84
with mobile phones and 0.74 with internet hosts. However, the value of ‘r’for
the correlation between the ISI score and high-tech exports is only 0.22. The
interrelationship among the three stock market variables is positive but small.
Another notable feature of Table 7 is that it indicates generally a negative
relationship between GDP growth and most of the other dependent as well as
explanatory variables.
Turning to the results of the multivariate analysis reported in Tables 6a-6e. Consider
first Table 6e where the dependent variable is the composite index of ICT development
- ISI scores. The table shows a good fit for the regression model with an adjusted R2
of 0.892. However, the only statistically significant variable at 5% levels
or less is the ‘scientists and engineers’. GDP per capita is significant at the 10% level.
None of the stock market variables are significant and one of them (number of listed
companies) has the wrong sign. The successive dropping of two of the three stock
market variables in the other three equations reported in Tables 6a-6e does not alter
this picture.
20
Overall, the results for Table 6a-6e reveal a broadly similar story. For some of the
dependent variables, the stock market variables are indeed statistically significant but
often have the wrong sign or changing values in alternative specifications of the
equation. The most consistently significant variables with correct signs in the
regression analysis as a whole for four of the five dependent variables is ‘scientist and
engineers’followed by ‘GDP per capita’. However, for ‘high-tech exports’as a
dependent variable, two of the three stock market variables are statistically significant
but the ‘scientist and engineers’is not. This result is curious from an economic point
of view as one would expect a priori ‘scientists and engineers’to be a more important
explanatory variable in this equation relative to the others. There are perhaps nonlinearities in the data that are not being captured by the linear regression model.
Thus, the multivariate regression analysis carried out so far in this study does not
suggest a robust relationship between stock market development and ICT development
and usage.
V. Venture capital, stock market and bank-based systems and the government
The U.S. venture capital industry has rightly attracted a great deal of attention in the
wake of the success of the new technology firms in that country. However, venture
capital, although normally associated with stock markets, is not peculiar to stockmarket-based systems. It is also found in bank-based systems and is at one level an
age-old phenomenon: people with money providing finance to high risk businesses to
gain large rewards when they back an enterprise that becomes subsequently successful.
In modern parlance, the term “venture capital” normally describes equity investments
in unquoted companies. In the U.K., continental Europe and much of the rest of the
world, the term private equity is used synonymously with that of venture capital. In
the U.S., venture capital usually refers to the provision of funds for younger, early
stage and developing businesses, whereas private equity also includes the financing of
leveraged management buyouts (MBOs) and buyins (MBIs) (BVCA, 27 January
2000). Therefore, statistics on venture capital comparing the U.S. to other economies
are not always directly comparable.
The U.S. venture capital industry has expanded enormously in the last five years as
there has been a share price boom for technology firms on the stock market. As Table
21
8 shows, U.S. venture capital investments have increased by a factor of four between
1995 and 1999. In 1999, according to the Financial Times, American venture capital
funds raised $56 billion, whereas ten years earlier, in 1990, the amount raised was only
$3 billion. In the first quarter of 2000, before the downturn in the Nasdaq, the venture
capital funds invested $22.7 billion in start-ups, nearly four times the amount of $6.2
billion dollars in the first quarter of 1999. This amount is an understatement as it does
not included money invested by wealthy individuals, the so-called “angels”, who are
thought to have invested twice as much in start-ups in 1999 as the traditional venture
capital firms. Although U.S. venture capital investments have been extremely volatile,
often moving in line with the Nasdaq index, there has been a huge trend increase since
the 1980s. There has been more venture capital investment in the last two years than
in the previous twenty. An obvious question is why was there so little venture capital
investment before the sudden boom of the last few years.
Nevertheless, it is important to keep a perspective here. Total investment by venture
capital funds accounts for only a small part of the $1.2 trillion invested by American
companies in gross non-residential investment. The European venture capital industry
is much smaller in absolute size, but larger in relation to GDP. In absolute amounts,
Manigart et al.(2000) report that investment by venture capitalists in the U.S. was
larger than the combined total of venture capital investments in the U.K., France, the
Netherlands and Belgium, while the amount invested by venture capitalists in the U.K.
was larger than the combined total of the other three. However, as a proportion of
GDP, U.S. venture capital investment in 1995 was only 0.05% of GDP, compared with
0.06% in Belgium, 0.07% in France, 0.05% in the Netherlands and 0.31% in the U.K.
Manigart et al. also suggest that the European venture capital industry invests much
less in start-ups than American venture capitalists. Investment by existing businesses is
preferred by European venture capital, with management buyouts representing nearly
two-thirds of the figure invested in the U.K. The degree of state involvement in the
provision of venture capital also distinguishes some European venture capital
industries from their counterparts in the U.S. For example, an important feature of the
Belgian industry is that the government-backed venture capital companies play a very
important role.
22
Black and Gilson (1998) provide a detailed analysis of the differences between the
U.S. and German venture capital industries. Their results are reported in Tables 9 and
10. A striking difference between the two countries in the sources of funds is that
pension funds account for the bulk of venture capital resources in the U.S. whilst they
contribute very little in Germany. Banks and insurance companies are more important
in Germany relative to the U.S. In the U.S., government agencies did not contribute
anything at all during the 1992-1995 period, while in Germany they contributed 8% of
total venture capital funds. In terms of the nature of investment, the differences
between the United States and Germany are very much like the differences noted
above between the U.S. and other European countries: the U.S. venture capital funds
invest more in seeds and start-ups than German venture capitalists. On the other hand,
LBO acquisitions are much more important in Germany.
Black and Gilson argue that the venture capital industry is much more vibrant and
successful in the U.S. compared to Japan, Germany and other continental European
countries because of its critical link with the stock market through the exit mechanism
of the IPO. This results in U.S. venture capitalists seeking to liquidate their portfolio
company investments as soon as they can rather than investing, like German and
Japanese banks, for the long term. Black and Gilson believe that the availability of an
exit mechanism from a successful start-up through an IPO is crucial to American
venture capital providers as it allows them to enter into implicit contracts with
enterprises (concerning future control of start-up firms). No such implicit contracts
are possible in bank-based systems. Black and Gilson’s argument is of course totally
plausible in a booming stock market. How credible it will be in conditions of falling
stock market prices is a moot question.
The high growth, high risk firms, which normally disproportionately consist of high
technology enterprises, of the kind financed by venture capital funds, have usually been
funded by governments in many developing and developed countries. The East Asian
developmental states have been prime examples of providing finance for such ventures.
Through an active and interventionist industrial policy, the Korean government has
obliged its firms to introduce new products and industrial processes through a mixture
of carrots and sticks. However, it took the view that in the context of
underdevelopment, such technical change is more likely to occur through the creation
23
and expansion of large firms rather than through small start-ups. The Korean
government effectively became a co-partner with these large enterprises (the chaebol)
in financing high risk projects, effectively socialising the risks involved (see further
Singh 1998). The Korean experience will become clearer when we discuss the case
study of Samsung Industries in the next section.
The government of Israel has also encouraged technological development through
direct aid to the venture capital industry. The state not only created the infrastructure
- a high quality labour force - but also provided direct assistance for promoting
technological change. In 1991 the government introduced a special program of
“technological incubators” which provided prospective entrepreneurs with physical
premises, financial resources, tools, professional guidance and administrative
assistance. Enterprises under this scheme could receive 85 per cent of the approved
budget subject to a maximum of $160,000. The money was, however, given for only
two years after which the companies had to leave the incubator and become selfsustaining (UN, 1999, p. 214).
A similar scheme was implemented in France with the government creating twelve
biotechnology incubators to commercialise research produced by government
scientists. Government policies, often taking the form of direct assistance, have also
played a major role in encouraging ICT development in Singapore and India. The
Indian case is particularly interesting because despite very low per capita income and
blanket import substitution policies, the country has managed to create a world class
information technology industry. Tables 11 and 12 show the very fast development of
the industry and its exports during the 1990s. Indian software exports rose from $128
million in 1991 to $2.9 billion in 1998-99 and are estimated to be on the order of $4
billion in 1999-2000. Some estimates project these exports to touch $50 billion in
2008 (Patibandla et al., 2000). The industry has now reached a level of development
where it is able to attract venture capital funds from abroad. Indian IT companies are
able to have IPOs on the London Stock Exchange and on Nasdaq. The government
helped the growth, development and maturing of the industry through a variety of
channels, the most important of which were: (i) creation of highly trained quality
manpower at elite technological institutions that the government had established; (ii)
having a selective policy to utilise multinational investment and encouraging exports;
24
(iii) the government provided finance, infrastructure, legal regulation and marketing
assistance to start-up technology firms; and (iv) the government set up software
technology parks in Bangalore and other Indian cities. One of the most successful
companies in India, Infosys, was set up with seed capital provided by government
financial institutions. The company had been refused funding by private banks and
without government assistance it may not have started at all. More light will be shed
on the Indian experience by an analysis of the optimal firm size for ICT development in
the next section.
To sum up, venture capital, IPOs and the stock market are not the only way of
promoting ICT development. Venture capital is perfectly compatible with bank based
systems and, indeed, in the developing country context, the government itself may well
be the best venture capitalist.
VI. New Technology and Large Firms
An important policy issue in developing countries is whether large or small firms are
the principal agents of technical change. This issue is particularly significant in the
context of information technology where it is generally believed that the best way of
developing and spreading this technology is through small firms and enterprising
individuals. Hence the emphasis on start ups, venture capital funds and incubators.
This portrayal of information technology is not, however, necessarily accurate in
relation either to developing or developed countries. In both groups of countries there
are many large firms which have been successful in ICT markets -- Samsung in Korea
and Nokia in Finland immediately come to mind. The question of the appropriate scale
of enterprise therefore requires further consideration.
We consider first the case of developing countries. In successful East Asian countries
such as Korea, large firms have generally been regarded as the main agents of technical
change (Amsden, 1989). This is ascribed to the fact that in late industrialising
countries, the chief objective is not so much to foster new technology but to adapt
existing technology obtained from abroad to national needs. In Korea the government
of General Park, instead of relying on foreign multinationals for this purpose, decided
25
to create huge domestic conglomerates, the chaebol. As noted above, it used a carrot
and stick industrial policy to make them invest in new technologies and new products
which would be competitive in international markets. The result is that the
manufacturing industry of Korea displays one of the highest levels of market
concentration anywhere – whether among developed or developing countries. The top
50 chaebol accounted for 15 per cent of the country’s GDP in 1990. Among the
largest 500 industrial companies in the world in 1990, there were 11 firms in Korea –
the same number as in Switzerland. A United Nations report observes in relation to
the industrial structure of Korea:
Such a structure is the deliberate creation of the government, which utilises a
highly interventionist strategy to push industry into large-scale, complex,
technologically demanding activities while simultaneously restricting FDI inflows
tightly to promote national ownership. It was deemed necessary to create
enterprises of large sizes and diversity, and to undertake the risk inherent in
launching investments in high-technology, high-skill activities that would remain
competitive in world markets. The chaebols acted as representatives and
spearheads of the government’s strategy: they were supported by protection
against imports and TNC entry, subsidised credit, procurement preferences and
massive investments in education, infrastructure and a science and technology
network (UN, 1993, p.43).
The chaebol have conspicuously succeeded in producing and exporting an ever
increasing range of new products as well as in propagating deep technical change in
the Korean economy.
Significantly, as indicated above, the conglomerate firms have not lagged behind in
information technology. The Samsung chaebol is now the largest producer and
exporter of Dynamic Random Access Memory (DRAM) microchips in the world. The
role of the government in Samsung’s success is noted by the World Bank, which does
not usually favour direct government intervention in the economy, in the following
terms:
The government’s Economic Development Board was a key player in
Samsung’s success. Government officials were keenly aware that the
Republic of Korea could not rely on forever on low wage
manufacturing. Just as the US had lost countless textile industry jobs
to Korea, they reasoned, so Korea would one day find it could no
longer compete for labour intensive manufacturing jobs with lower
wage neighbours such as China and Indonesia. To prepare for that
26
day, government officials, working with consultation with the private
sector, developed incentives for new knowledge - and capital intensive industry. Incentives varied widely and included the
government’s build industrial parks, subsidising utilities, giving tax
rebates for exports, and making cheap loans for investment in new
products. By 1980, urged forward by subsidies and incentives,
Korean industry had moved into steel, ships, and even cars and was
about to leap into world-class electronics.13
Samsung was established in 1969 and its major products range from home appliances
to sophisticated telecommunications equipment. In 1998, semiconductors, video tape
recorders, audio players and communication equipment accounted for 71 percent of
the company’s unconsolidated revenues; computer and monitors constituted another
18 percent and home appliances the remaining 11 percent. Samsung is the world’s
largest semiconductor maker, the fourth largest mobile handset maker and the largest
producer of thin film transistor liquid crystal displays. Samsung now accounts for 13
percent of the entire Korean stock market and attracts a large number of foreign
investors. Recently, the share price has fallen sharply by 58 percent from its peak in
mid-July owing to a selling of shares in leading chip-makers. Nevertheless, Samsung is
expected to report record profits of 6,000 billion won this year against 3,600 billion in
1999. The fall in Samsung’s share price has undermined the whole Korean market and
is a cause of serious concern to the authorities.
Other developing countries, such as India, have also found that in order to compete in
international markets and to export IT products, it is necessary to have large firms.
The exports of small software companies are no more than “body-shopping” engaging in low value-added data and code entry – for U.S. firms (Parthasarathi 2000)
Only large firms are able to export high value-added IT products which is what is
needed to maintain the dynamism in international competiveness of the industry.
As Table 11 indicates, software exports have become increasingly important to the
Indian economy as they increased more than five-fold during the 1995-1999 period and
this rapid growth is forecast to continue. There were 716 companies involved in
software exports but, as in other industries, the size distribution of export firms is
highly skewed. The largest company, Tata Consultancy Services, an off-shoot of one
13
World Bank (1993).
27
of India’s old economy’s leading corporations, the Tata Group, alone accounts for 15
per cent of India’s total software exports. The top ten software exporters include
Wipro, Infosys, NIIT and Satyam Computer Services and account for nearly half of the
exports. The top 25 account for 63 per cent of total exports of the sector.
As in the case with Tata, most of these indigenous software companies have emerged
from existing Indian industrial groups. However, the list of the top 25 exporters also
includes foreign joint-ventures and ‘captive’companies e.g. IBM Global Services,
Mahindra British Telecom and Hewlett Packard India.
Similarly, in advanced countries large firms are often at the cutting edge of the new
technology revolution, particularly in Europe. Large firms such as Nokia in Finland,
Ericsson in Sweden, and Intel and Motorola in the United States are outstanding
examples.
The prevalence and success of large corporations in the ICT market contradicts the
widely held view that the new technological revolution is fueled most effectively by
small, entrepreneurial firms. By extension, it also questions the assumption of many
policymakers that the best way to jump-start ICT development is through incubators
and start-ups. In this context we will briefly consider the case of Nokia. Following the
collapse of its business with the former Soviet Union in the late 1980s, Nokia, a large
and old Finnish company, decided to concentrate on producing mobile phones. It
turned out to be a fortuitous decision since the market for mobile phones has exploded
in the 1990s and Nokia has succeeded in capturing a growing share of this market
(Nokia held 22.1% of world handset production in 1995 and today accounts for 32%).
For the last eight years Nokia has been among the fastest growing companies in
Europe and this growth has been, in contrast to other rapidly-growing companies such
as the British telecoms firm Vodafone, organic, i.e. not based on acquisitions. Today,
the company accounts for a quarter of Finland’s exports and nearly four percent of its
GDP. Its research division employs 13,000 workers.
The reason why large companies tend to be significant in ICT activities is that
information technology encompasses industries such as mobile phones and computers
as well as software and the internet. The former two industries are quintessentially
28
old-economy type industries with huge old-fashioned economies of scale. For this
segment of ICT, the venture capital model and the small entrepreneur may not be the
most appropriate method for fostering rapid expansion. Successful ICT development
in these areas may require long-term relationships between finance and industry rather
than the greater uncertainties and instabilities inherent in the stock market and IPO
system.
Nevertheless, it may seem that at least in software and internet applications the small
firm and the small entrepreneur will be able to hold his own. Indeed, recent examples
of small IT companies being able to attract huge amounts of capital and being able to
overtake much larger (in terms of net assets) old-economy companies tend to confirm
this impression. But here again we need to pause. Analysts point out that in ecommerce there are huge economies of scale and scope deriving from concentration of
advertising and brand names. The Internet Advertising Bureau reveals that 71 percent
of US dotcom advertising revenue are going to the top 10 sites, 83 percent to the top
25 and 91 percent to the top 50 sites. According to one internet analyst, the internet
will end up as just an alternative channel of distribution for ‘real’corporations, because
they have the brands, the infrastructure, the expertise, the customers, the financial
resources and all the other things dotcoms lack. Some argue that what we have seen –
the rise and fall of dotcoms in the last three years – is a temporary phenomenon which
is not going to effect the long term prospects of the dominant old-economy companies
which did not get involved.
The fact that most of these corporations did not move fast enough to
position themselves for the internet, is irrelevant, because they now have a
second bite at the apple… it’s make versus buy - either they can buy a truly
troubled dotcom operation at a very depressed price, or they can build it
themselves. But most corporations have lost nothing by not becoming a
force on the internet three years ago. (Financial Times, “Dotcoms
devoured”, October 23, 2000)
Industrial economists have observed that in a large number of industries, the industry
generally starts out with many small firms but ends with a small number of stable
oligopolistic firms. For example, in the US, the second industrial revolution during the
last decade of the 19th century lead to the development of the railways network,
spawning hundreds of railroad companies but, today there are only a handful of
29
railroads in the US. It is arguable that the internet too is moving in the same direction.
When the dust settles down we may be faced with a few giant internet companies
dominating the world market.
VII. Conclusion
This paper has suggested that stock markets are neither a necessary nor sufficient
condition for promoting the development of ICT. Many countries, particularly in
Northern Europe but also elsewhere, have been able to achieve a high degree of ICT
development without a central role for venture capital, IPOs and stock markets. Other
countries, such as Britain which have flourishing stock markets, have failed to become
leaders in ICT development. Econometric analysis did not reveal any robust
systematic relationship between indicators of stock market development and those of
ICT development.
Even in relation to the U.S. where stock markets have been helpful in the diffusion of
ICT in the economy, the markets are not without considerable downsides.
Specifically, the extremely high price earning ratios of technology companies between
1996 and January 2000, compared with any previous period in stock market history,
raise the spectre of these valuations representing a speculative bubble. Even after a 45
per cent fall in the Nasdaq index since March 2000, many firms will still be overpriced. However, a further deflation may not turn out to be as benign for the US and
the world economy as the fall in share prices during the last nine months. The deflation
of this bubble could have serious implications for the U.S. as well as the world
economy.
There is, however, another very important negative feature of stock market-based ICT
development to which its enthusiasts do not pay sufficient attention, that is, the whole
question of the digital divide. The availability of venture capital, IPOs and stock
options no doubt create powerful incentives for budding entrepreneurs, but it also
creates a “winner-takes-all” culture and contributes towards increasing social and
economic disparity. The digital divide is not just a North-South phenomenon but also
very much a within-U.S. phenomenon as well, as the following suggests.
30
A study produced by the U.S. Department of Commerce detailed the growing digital
divide in the U.S. between different classes and ethnic groups. The survey found that
people with a college degree were eight times more likely to have a PC at home and 16
times more likely to have internet access at home than people with an elementary
school education. Similar divides were found between high income households in
urban areas and rural, low-income households in rural areas; white low-income families
and minority low-income families and between disabled people and people without
disabilities. Thus, although internet access has increased across all demographic
groups, the digital divide has mirrored persistent social and economic inequalities in
the country.
The Economist notes “it would be hard to find a better real-life symbol for the digital
divide than the gulf between Silicon Valley’s leafy Palo Alto, home to dot.com
millionaires where the average house sells for nearly $700,000, and East Palo Alto, the
desperate little town on the other side of Highway 101 that not long ago claimed
America’s highest murder rate.”14 Thus, contrary to the claims of IT enthusiasts that
the new technology is a great equaliser, it seems that by itself it perpetuates disparity if
not to worsen it. These disparities can only be addressed by the government. It is not,
therefore, surprising that in the absence of government intervention citizens groups in
the San Francisco and Silicon Valley area have been promoting ballots to stop any
further location of dotcom companies in their districts. It is not IT per se that creates
the digital divide but rather the nature of social limits which are imposed on it which
lead to greater or smaller disparities. It is significant that North European countries
like Sweden and Finland, where there has been an equally strong development of IT,
have not experienced the same kind of social divisions as the U.S.15
14
Economist, 24 June 2000, p. 22.
The United Nations Human Development Report 2000 shows that in terms of human poverty index
Finland and Sweden rank much higher than the United States or the UK. This index is based on
probability at birth of not surviving to age60, adult functional illiteracy rate, percentage of people
living below the income poverty line (50% of median disposable household income), and long-term
unemployment rate (12 months or more)
15
31
32
References
Allen, F. 1993. “Stock markets and resource allocation”, in C. Mayer and X. Vives
(eds.), Capital Markets and Financial Intermediation. Cambridge: Cambridge
University Press, pp.81-108.
Black, B. and R. Gilson. 1998. “Venture capital and the structure of capital markets:
banks versus stock markets”, Journal of Financial Economics, Vol.47, No.8, pp.243277.
Gordon, Robert, J. 2000. “ Does the ‘New Economy’measure up to the Great
Inventions of the past?”, working Paper, Northwestern University, April 25.
Forthcoming in Journal of Economic Perspectives.
Hicks, J. 1969. A Theory of Economic History. Oxford: Clarendon Press.
Hughes, A. 1991. “Merges and economic performance in the UK: A survey of
empirical evidence 1950-1990” in J. fairburn and J.A. Kay ed. 1989. Merger and
merger policy Oxford: Oxford University Press.
Jorgenson, Dale and Kevin Stiroh. 2000. “Raising the speed limit: US economic
growth in the information age”. Forthcoming in Brookings Papers in Economic
Activity.
Keynes, J.M. 1936. The General Theory of Employment, Interest and Money. New
York: Harcourt, Brace and Company.
IFC 1996. Emerging Stock Markets Factbook 1996, IFC, The World Bank.
Washington DC.
King, R. and R. Levine 1993. “Financial intermediation and economic development”,
in C. Mayer and X. Vives (eds.), Capital Markets and Financial Intermediation.
Cambridge: Cambridge University Press, pp.156-189.
Manigart, S., De Waele, K., Wright, M., Robbie, K., Desbrières, Sapienza, H. and A.
Beeman, 2000. “Venture Capital, Investment Appraisal and Accounting Information:
A Comparative Study of the USA, UK, France, Belgium and Holland”, European
Financial Management, vol 6, no. 3, pp. 389-403.
Meeks, G. 1977. Disappointing marriage: A study of the gains from merger.
Cambridge: Cambridge University Press.
Oliner, Stephen and Daniel Sichel. 2000. The resurgence of growth in the late 1990s:
Is information technology the story? Working Paper, Federal Reserve Board, May.
Parthasarthi, Ashok. 2000. “Opportunities and challenges of the emerging knowledge
economy” presented as the All India Association National Management Forum on
“Managing the Knowledge Economy”, Jawaharlal Nehru University, New Delhi, April
7, 2000.
33
Patibandla, M., D. Kapur and B. Petersen. 2000. “Import Substitution with Free
Trade: The Case of India’s Software Industry”, Economic and Political Weekly, 8
April, pp.1263-1270.
Porter, M. (1992). “Capital disadvantage: America’s failing capital investment
system”, Harvard Business Review, 70, (September-October), pp. 65-82.
Robinson, J. 1952. “The Generalization of the General Theory”, in The Rate of
Interest and Other Essays. London: Macmillan, pp.67-142.
Shleifer, Andrei. 2000. Inefficient markets: An introduction to behavioral finance.
Oxford: Oxford University press.
Shiller, R. 2000. Irrational Exuberance. Princeton, N.J.: Princeton University Press.
Singh, A. 2000. “The Anglo-Saxon Market for Corporate Control: The Financial
System and International Competitiveness”, in C. Howes and A. Singh (eds.),
Competitiveness Matters: Industry and Economic Performance in the U.S., Ann
Arbor: University of Michigan Press.
Singh, A. 1999. “ Should Africa Promote Stock Market Capitalism?”, Journal of
International Development, vol. 11, pp.343-365.
Singh, A. 1998. “Savings, Investment and the Corporation in the East Asian Miracle”,
Journal of Development Studies, vol. 34, no. 6, pp. 112-137.
Singh, A. 1995. “Corporate Financial Patterns in Industrialising Economies: A
Comparative International Study”, Technical Paper 2, International Finance
Corporation, Washington DC: The World Bank.
Singh, A. 1992. “Corporate Take-overs” in J. Eatwell, M. Milgate and P. Newman
(eds.), The New Palgrave Dictionary of Money and Finance. London: Macmillan, pp.
480-486.
Singh, A. 1975. “Take-overs, Economic Natural Selection and the Theory of the
Firm”, Economic Journal, September.
Singh, A. and R. Dhumale (2000). Globalisation, Technology, and Income
Inequality: A Critical Analysis. Mimeo, Cambridge.
Stein, J. 1988. “Take-over threat and managerial myopia”, Journal of Political
Economy, Vol.96, pp.61-80.
Stein, J. 1989. “Efficient stock markets, inefficient firms: a model of myopic corporate
behaviour”, Quarterly Journal of Economics, Vol.104 (November), pp.665-670.
Tobin, J. 1984. “On Efficiency of the financial system”, Lloyds Bank Review, 1-15
July.
United Nations. 1999. World Economic and Social Survey 1999, New York: United
Nations.
34
Table 4. Variables included in the ISI composite index
Computer
Information
Internet
Social
Infrastructure
infrastructure
Infrastructure
infrastructure
• PCs installed per
• Cable subscribers per
• Business Internet
• Civil liberties
capita
• Home PCs shipped
per household
capita
users per non-
• Cellular phone
agricultural
• Government and
• Cost for phone call
commercial PCs
• Fax ownership per
shipped per nonagricultural workforce
• Educational PCs
shipped per student
and faculty
• Percent of non-home
networked PCs
• Software vs. hardware
spending
workforce
ownership per capita
capita
• Home Internet users
• Newspaper
readership per capita
• Press freedom
• Secondary school
per household
enrolment
• Education Internet
• Tertiary school
• Radio ownership per
users per student and
enrolment
faculty
capita
• Telephone line error
• ECommerce
spending per total
rates
• Telephone lines per
Internet users
household
• TV ownership per
capita
Source: ISI website at http://www.worldpaper.com
Table 1 Stock
Market Performance:
1990-1999
Mean of percent
change
Standard
Deviation
Annualised
mean
Annualised
standard
deviation
Correlation
with S&P 500
NASDAQ
Dec. 1990 - Dec. 1999
Dec. 1990 - Dec. 1995
Jan.1995 - Dec.1999
2.31
1.82
3.11
6.40
4.08
7.26
28.74
21.84
29.90
22.16
14.12
25.14
0.95
0.98
0.92
Dow Jones
Dec. 1990 - Dec. 1999
Dec. 1990 - Dec. 1995
Jan.1995 - Dec.1999
1.33
1.16
1.81
3.93
3.05
4.64
17.37
13.90
17.36
13.63
10.56
16.06
0.99
0.99
0.99
S&P 500
Dec. 1990 - Dec. 1999
Dec. 1990 - Dec. 1995
Jan.1995 - Dec.1999
1.39
1.09
1.93
3.77
2.91
4.44
17.51
13.02
18.49
13.05
10.07
15.38
1.00
1.00
1.00
FT100
Dec. 1990 - Dec. 1999
Dec. 1990 - Dec. 1995
Jan.1995 - Dec.1999
1.06
0.99
1.39
3.94
4.02
3.75
14.01
11.85
13.36
13.65
13.92
13.00
0.99
0.92
0.98
35
FT- EuroPac
Dec. 1990 - Dec. 1999
Dec. 1990 - Dec. 1995
Jan.1995 - Dec.1999
0.68
0.68
0.96
4.42
4.64
4.19
9.64
8.15
9.20
15.30
16.06
14.51
0.93
0.74
0.92
Nikkei
Dec. 1990 - Dec. 1999
Dec. 1990 - Dec. 1995
Jan.1995 - Dec.1999
-0.09
-0.07
0.05
6.43
6.98
5.56
-0.15
-0.78
0.51
22.27
24.18
19.25
-0.49
-0.58
-0.32
Source: NASDAQ
website
(http://www.marketdata
.nasdaq.com)
Table 2:
Indicators of
Stock Market
Development in
Selected
Countries
(average 199095)
GDP
(US$ mill.)
Emerging markets
Argentina
Brazil
Chile
China
Greece
India
Indonesia
Jordan
Korea
Malaysia
Mexico
Pakistan
South Africa
Thailand
Turkey
Zimbabwe
Median
GDP growth
GDP per
capita
(US$ mill.)
Market
Capitalisation
(US$ mill.)
Market
Value T
Capitalisation
(US$ m
as % of GDP
218766.4
563643.6
44576.7
613707.2
96504.5
302950.2
153123.5
5265.8
350015.7
63753.3
341687.5
46178.3
125787.5
113025.3
157238.6
7405.7
139455.5
5.7
2.7
7.3
12.8
1.1
4.6
7.6
8.2
7.2
8.7
1.1
4.6
0.6
8.4
3.2
1.0
5.2
6478.8
3718.7
3245.2
513.0
9333.1
345.5
817.7
1083.3
7932.7
3333.4
3820.1
398.3
3309.8
2111.8
2726.1
703.4
2985.6
26504.0
90120.2
42991.7
29285.2
13689.2
77256.5
28953.5
3672.2
137928.0
140595.8
115262.5
8559.2
181293.3
86911.0
20762.0
1619.3
36138.4
12.1
16.0
96.4
4.8
14.2
25.5
18.9
69.7
39.4
220.5
33.7
18.5
144.1
76.9
13.2
21.9
23.7
312330.0
575706.2
147575.6
113751.4
1377927.5
2168252.0
67633.0
1101843.3
4032161.3
3.5
1.8
2.0
-0.5
1.0
18258.1
19656.8
30137.3
20651.8
23792.7
26623.6
13044.6
19287.7
32340.3
176049.8
293281.5
44824.3
25862.0
407125.3
434670.8
26506.0
164938.7
3139084.7
56.4
50.9
30.4
22.7
29.5
20.0
39.2
15.0
77.9
Developed markets
Australia
Canada
Denmark
Finland
France
Germany
Isreal
Italy
Japan
6.4
1.0
1.0
36
Netherlands
Norway
Singapore
Spain
Sweden
Switzerland
U.K.
U.S.A.
Median
Source: World
Development Report
(various issues) and
Emerging Stock Markets
Factbook (1996)
348140.6
124192.1
56616.3
507473.3
224920.0
251659.5
1032161.3
6534233.3
348140.6
1.8
3.5
8.7
1.1
-0.1
0.1
1.4
2.6
1.6
22631.0
28888.8
17334.3
12961.2
25871.8
36466.5
17764.2
25437.7
22631.0
202030.8
29071.7
91004.2
138368.5
113764.2
271718.4
1306715.0
5738594.2
176049.8
58.0
23.4
160.7
27.3
50.6
108.0
126.6
87.8
50.6
Table 3:
Indicators of
ICT
Development
in Select
Countries
(1998)
Mobile Phones
1000
Emerging Markets
Argentina
Brazil
Chile
China
Greece
India
Indonesia
Jordan
Korea
Malaysia
Mexico
Pakistan
South Africa
Thailand
Turkey
Zimbabwe
Median
Personal
Computers per
1000
Internet Hosts
per 1000
High Tech
Exports as %
of Mfg. Export
78.0
47.0
65.0
19.0
194.0
1.0
5.0
12.0
302.0
99.0
35.0
1.0
56.0
32.0
53.0
4.0
41.0
44.3
30.1
48.2
8.9
51.9
2.7
8.2
8.7
156.8
58.6
47.0
3.9
47.4
21.6
23.2
9.0
26.7
38.5
26.2
26.4
0.6
73.8
0.2
1.0
1.3
60.0
25.4
40.9
0.3
39.2
6.5
13.9
1.7
19.7
5.0
9.0
4.0
15.0
7.0
5.0
10.0
286.0
176.0
364.0
572.0
188.0
170.0
359.0
355.0
374.0
213.0
474.0
411.6
330.0
377.4
349.2
207.8
304.7
217.2
173.4
237.2
317.6
373.4
567.3
540.2
631.8
1218.4
1313.5
207.6
225.1
114.4
208.1
517.0
899.5
11.0
15.0
18.0
22.0
23.0
14.0
20.0
8.0
26.0
30.0
16.0
27.0
54.0
19.0
9.0
31.0
2.0
2.0
9.0
ISI scores
1651.0
1354.0
1677.0
915.0
2033.0
871.0
888.0
942.0
2931.0
1583.0
1286.0
719.0
1537.0
1010.0
1259.0
1286.0
Developed Markets
Australia
Canada
Denmark
Finland
France
Germany
Isreal
Italy
Japan
Netherlands
Norway
4129.0
4336.0
4336.0
4577.0
3140.0
3558.0
3140.0
2703.0
4093.0
4230.0
4481.0
37
Singapore
Spain
Sweden
Switzerland
U.K.
U.S.A.
Median
346.0
179.0
464.0
235.0
252.0
256.0
286.0
458.4
144.8
361.4
421.8
263.0
458.6
330.0
452.3
105.4
670.8
429.0
321.4
1940.0
517.0
59.0
7.0
20.0
16.0
28.0
33.0
20.0
4014.0
2533.0
5062.0
4174.0
3807.0
5041.0
4129.0
Source: World
Development Report
2000/2001
Table 5. Median values for
quartiles for full sample
Low income
GDP (US$m.)
(<$1566)
34511.8
Middle income
Rich
($1566< x <$4181) ($4181< x <$18181) ($18181
43264.8
82748.2
4.20
3.55
5.60
634.62
2959.79
9609.72
0.17
0.15
0.27
Market Capitalisation/GDP
(US$m.)
3477.58
6049.75
20547.17
World Trade Value (US$m.)
555.10
1879.67
5057.83
9.0
6.3
3.3
145.6
120.8
168.8
Mobile Phones/1000
4.0
41
227.5
Personal Computers/1000
5.7
34.8
150.8
Internet Hosts/1000
0.6
13.91
98.0
153.0
211.5
1243.5
7.0
4.0
20.0
923.0
1354.0
3035.5
GDP growth
GDP per capita
Market Capitalisation
(US$m.)
Turnover Ratio (inverse)
Number of Listed
Companies
Scientists & Engineers in
R&D/10000
High Tech Exports as % of
Mfg. Export
ISI scores
38
Table 6a.
Regression
Results:
Dependent
variable Mobile
Phones per
1000
Explanatory
Variables
(Constant) GDP growth
rate
B
Std. Error
t
Significance
R
R2
Ajusted R2
Standard error
B
Std. Error
t
Significance
R
R2
Ajusted R2
Standard error
B
Std. Error
t
Significance
R
R2
Ajusted R2
Standard error
B
GDP per
capita
Scientists and
Engineers in
R&D per
10,000
-2.847
4.978
-0.572
0.571
1.13E-03
0.003
0.369
0.714
3.31E-03
0.298
0.007
0.994
(Constant) GDP growth
rate
GDP per
capita
Scientists and
Engineers in
R&D per
10,000
0.526
4.935
0.107
0.916
4.21E-03
0.003
1.399
0.17
6.426E.02
0.024
2.66
0.012
(Constant) GDP growth
rate
GDP per
capita
Scientists and
Engineers in
R&D per
10,000
Market Turnover ratio
capitalisation
(inverse)
as % of GDP
-2.589
4.895
-0.529
0.6
2.92E-03
0.003
0.947
0.35
6.39E-02
0.023
2.73
0.01
-6.466
4.571
-1.415
0.166
(Constant) GDP growth
rate
GDP per
capita
Scientists and
Engineers in
R&D per
10,000
Market Turnover ratio
capitalisation
(inverse)
as % of GDP
2.69E-03
8.05E-02
92.397
44.949
2.056
0.048
Market Turnover ratio
capitalisation
(inverse)
as % of GDP
2.10E-03
4.464
-1.633
0.112
-7.29
0.024
3.47
0.001
0.871
0.758
0.716
82.0137
33.884
32.863
1.03
0.31
Market Turnover ratio
capitalisation
(inverse)
as % of GDP
-1.70E-01
0.311
-0.548
0.587
0.841
0.708
0.675
87.6628
79.283
46.542
1.703
0.097
0.849
0.721
0.69
85.6779
37.796
-0.221
39
Std. Error
t
Significance
R
R2
Ajusted R2
Standard error
30.897
1.223
0.229
4.46
-0.049
0.961
0.003
0.916
0.366
0.024
3.338
0.002
Explanatory
Variables
(Constant) GDP growth
rate
GDP per
capita
Scientists and
Engineers in
R&D per
10,000
-2.038
17.642
-0.116
0.909
1.16E-02
0.011
1.081
0.288
1.04E-01
0.084
1.232
0.227
(Constant) GDP growth
rate
GDP per
capita
Scientists and
Engineers in
R&D per
10,000
-1.183
17.968
-0.066
0.948
5.11E-03
0.011
0.465
0.645
1.90E-01
0.088
2.162
0.812
(Constant) GDP growth
rate
GDP per
capita
Scientists and
Engineers in
R&D per
10,000
Market Turnover ratio
capitalisation
(inverse)
as % of GDP
2.77E-03
0.011
0.243
0.81
0.19
0.087
2.199
0.486
-1.29E+01
18.366
-0.705
0.035
0.859
0.739
0.71
82.8674
Table 6b.
Regression
Results:
Dependent
variable Internet
Hosts per
1000
B
Std. Error
t
Significance
R
R2
Ajusted R2
Standard error
B
Std. Error
t
Significance
R
R2
Ajusted R2
Standard error
B
Std. Error
t
Significance
R
R2
Ajusted R2
Standard error
-12.112
159.999
-0.076
0.94
Market Turnover ratio
capitalisation
(inverse)
as % of GDP
-6.03E-01
1.07
-0.564
0.577
-6.651
17.26
-0.385
0.702
0.792
0.628
0.56
287.7884
-28.84
122.556
-0.235
0.815
Market Turnover ratio
capitalisation
(inverse)
as % of GDP
-2.73E-01
1.139
-0.239
0.038
0.718
0.515
0.46
319.0439
55.357
174.404
0.317
0.753
0.722
0.521
0.466
317.0623
-6.728
18.164
-0.37
0.713
40
(Constant) GDP growth
rate
B
Std. Error
t
Significance
R
R2
Ajusted R2
Standard error
GDP per
capita
Scientists and
Engineers in
R&D per
10,000
-2.557
15.214
-0.168
0.867
1.28E-02
0.01
1.277
0.21
9.69E-02
0.082
1.18
0.246
Explanatory
Variables
(Constant) GDP growth
rate
GDP per
capita
Scientists and
Engineers in
R&D per
10,000
2.366
4.064
0.582
0.564
6.93E-03
0.002
2.794
0.009
4.85E-02
0.019
2.501
0.018
(Constant) GDP growth
rate
GDP per
capita
Scientists and
Engineers in
R&D per
10,000
3.295
3.655
0.902
0.373
7.18E-03
0.002
3.216
0.003
5.04E-02
0.018
2.819
0.008
(Constant) GDP growth
rate
GDP per
capita
Scientists and
Engineers in
R&D per
10,000
Market Turnover ratio
capitalisation
(inverse)
as % of GDP
6.96E-03
0.002
2.881
0.007
5.47E+01
0.018
2.983
0.005
-1.05E-00
3.887
-0.27
0.789
-68.503
107.312
-0.638
0.527
Market Turnover ratio
capitalisation
(inverse)
as % of GDP
0.788
0.621
0.578
282.0256
Table 6c.
Regression
Results:
Dependent
variable Personal
Computers
per 1000
B
Std. Error
t
Significance
R
R2
Adjusted R2
Standard error
B
Std. Error
t
Significance
R
R2
Ajusted R2
Standard error
-13.833
36.855
-0.375
0.71
Market Turnover ratio
capitalisation
(inverse)
as % of GDP
3.91E-01
0.246
1.587
0.122
-2.443
3.976
-0.615
0.543
0.92
0.847
0.819
66.291
-29.346
24.927
-1.177
0.247
Market Turnover ratio
capitalisation
(inverse)
as % of GDP
3.66E-01
0.232
1.581
0.123
0.919
0.844
0.826
64.8911
B
Std. Error
t
Significance
R 0.913
-15.385
36.908
-0.417
0.679
4.606
3.844
1.198
0.239
41
R2 0.833
Ajusted R2 0.814
Standard error 67.0979
(Constant) GDP growth
rate
B
Std. Error
t
Significance
R
R2
Ajusted R2
Standard error
GDP per
capita
Scientists and
Engineers in
R&D per
10,000
4.949
3.605
1.373
0.179
7.50E-03
0.002
3.16
0.003
5.04E-02
0.19
2.587
0.014
Explanatory
Variables
(Constant) GDP growth
rate
GDP per
capita
Scientists and
Engineers in
R&D per
10,000
-0.171
0.903
-0.189
0.851
-4.70E-04
0.001
-0.902
0.375
2.90E-03
0.004
0.745
0.462
(Constant) GDP growth
rate
GDP per
capita
Scientists and
Engineers in
R&D per
10,000
7.23E-05
0.001
0.141
0.889
1.35E-03
0.004
0.34
0.726
-24.175
25.428
-0.951
0.348
Market Turnover ratio
capitalisation
(inverse)
as % of GDP
0.914
0.835
0.816
66.828
Table 6d.
Regression
Results:
Dependent
variable High Tech
Exports as
a % of
Manufacturi
ng
B
Std. Error
t
Significance
R
R2
Ajusted R2
Standard error
B
Std. Error
t
Significance
R
R2
Adjusted R2
Standard error
17.285
7.362
2.348
0.026
Market Turnover ratio
capitalisation
(inverse)
as % of GDP
2.19E-01
0.47
4.658
0
-2.161
0.771
-2.804
0.009
0.732
0.536
0.436
12.3411
2.461
5.59
0.44
0.663
0.637
0.405
0.326
13.4937
0.971
0.88
1.103
0.279
Market Turnover ratio
capitalisation
(inverse)
as % of GDP
1.81E-01
0.049
3.705
0.001
42
(Constant) GDP growth
rate
B
Std. Error
t
Significance
R
R2
Ajusted R2
Standard error
B
Std. Error
t
Significance
R
R2
Ajusted R2
Standard error
GDP per
capita
Scientists and
Engineers in
R&D per
10,000
Market Turnover ratio
capitalisation
(inverse)
as % of GDP
1.388
1.078
1.288
0.208
-1.57E-04
0.001
0.244
0.809
3.57E-03
0.005
0.761
0.453
-1.15E-00
0.948
-1.214
0.234
(Constant) GDP growth
rate
GDP per
capita
Scientists and
Engineers in
R&D per
10,000
Market Turnover ratio
capitalisation
(inverse)
as % of GDP
1.596
1.017
1.863
0.072
1.60E-04
0.001
0.251
0.804
2.22E-03
0.005
0.437
0.665
Explanatory
Variables
(Constant) GDP growth
rate
GDP per
capita
Scientists and
Engineers in
R&D per
10,000
-28.334
34.082
-0.831
0.412
3.95E-02
0.02
1.975
0.057
6.30E-01
0.149
4.225
0
(Constant) GDP growth
rate
GDP per
capita
Scientists and
Engineers in
R&D per
10,000
4.90E-02
0.018
2.729
0.01
6.06E-01
1.706
4.373
0.274
14.017
9.342
1.5
0.144
0.417
0.174
0.064
15.9044
5.582
6.687
0.835
0.41
0.37
0.137
0.022
16.2529
Table 6e.
Regression
Results:
Dependent
variable ISI scores
B
Std. Error
t
Significance
R
R2
Ajusted R2
Standard error
1238.544
280.834
4.41
0
Market Turnover ratio
capitalisation
(inverse)
as % of GDP
2.66E-00
1.82
1.46
0.155
-41.175
30.518
-1.349
0.187
0.954
0.91
0.892
473.3509
B
971.037
Std. Error
195.934
t
4.956
Significance
0
R 0.951
R2 0.904
Adjusted R2 0.892
-7.748
30.257
-0.256
0.8
Market Turnover ratio
capitalisation
(inverse)
as % of GDP
1.90E-00
0.139
1.114
0
43
Standard error 472.0871
(Constant) GDP growth
rate
B
Std. Error
t
Significance
R
R2
Ajusted R2
Standard error
B
Std. Error
t
Significance
R
R2
Ajusted R2
Standard error
GDP per
capita
Scientists and
Engineers in
R&D per
10,000
Market Turnover ratio
capitalisation
(inverse)
as % of GDP
-9.122
31.382
-0.291
0.773
4.34E-02
0.019
2.27
0.03
6.38E-01
0.14
4.567
0
-2.74E+01
28.927
-0.949
0.35
(Constant) GDP growth
rate
GDP per
capita
Scientists and
Engineers in
R&D per
10,000
Market Turnover ratio
capitalisation
(inverse)
as % of GDP
5.05E-02
0.019
2.675
0.012
6.09E-01
0.15
4.068
0
1194.206
276.474
4.319
0
0.95
0.903
0.891
474.528
1001.634
198.244
5.053
0
2.175
29.512
0.074
0.942
0.949
0.901
0.888
480.7237
Table 7.
Correlation
Matrix
GDP
(US$
m.)
GDP
growth
-0.14
GDP (US$ m.)
GDP per
Market
Market
Wor
capita capitalisation/ Capitalisation Valu
GDP
(US$m.)
0.42
0.15
0.96
-0.32
0.23
-0.13
0.33
0.36
GDP growth
-0.14
GDP per capita
0.42
-0.32
Market
capitalisation/
GDP
0.15
0.23
0.33
Market
Capitalisation
(US$m.)
0.96
-0.13
0.36
0.24
World Trade
Value (US$m.)
0.95
-0.10
0.34
0.20
0.97
Number of
listed
companies
0.83
-0.07
0.22
0.21
0.88
0.24
44
Turnover Ratio
(inverse)
-0.16
-0.15
-0.19
-0.12
-0.12
Mobile
Phones/1000
0.24
-0.21
0.80
0.30
0.23
Personal
Computers/
1000
0.39
-0.22
0.91
0.36
0.38
Internet Hosts
per 1000
0.57
-0.28
0.68
0.16
0.57
Scientists and
Engineers in
R&D per 10000
0.52
-0.44
0.91
0.20
0.48
High Tech
Exports as % of
Mfg. Export
0.19
0.25
0.20
0.61
0.24
ISI scores
0.38
-0.40
0.92
0.21
0.39
Table 8.
Venture
capital
investmen
t
Number
of
companie
s
1995
1369
1996
1882
1997
2493
1998
3446
1999 (Q12664
Q3)
Financial
Times,
Invested
(US$billio
ns)
5.8
9.9
14
19.1
28.6
45
December
14, 1999
Table 9. United States and
Germany: Capital raised by
venture capital funds by type
of investor
Percentage of capital raised by
venture capital funds in the
United States and Germany,
by type of investor, for 19921995.
1992
1993
1994
1995
United States
Corporations
Private individuals and families
Government agencies
Pension funds
Banks and insurance companies
Endowments and foundations
Other
Total
3
11
0
42
15
18
11
100
8
8
0
59
11
11
4
100
9
9
0
46
9
21
2
100
2
17
0
38
18
22
3
100
Germany
Corporations
Private individuals and families
Government agencies
Pension funds
Banks and insurance companies
Endowments and foundations
Other
Total
7
6
4
0
53
10
17
100
9
7
6
0
52
12
14
100
8
8
7
0
55
12
10
100
10
5
8
9
59
6
2
100
Source: Black and Gilson (1998),
p.249, Table 3.
Table 10. United States
and Germany: Venture
capital disbursements by
stage
of financing
Percentage of capital
disbursed by venture capital
funds in the United States
and
Germany, by nature of
investment, for 1992-1995.
United States
Seed
Start-up
Other early stage
Expansion
1992
1993
1994
3
8
13
55
7
7
10
54
4
15
18
45
1995
46
LBO Acquisition
Other
Total
7
14
100
6
16
100
6
12
100
Germany
Seed
Start-up
Expansion
LBO Acquisition
Other
Total
1
6
45
24
25
100
1
7
66
25
0
100
2
8
54
36
0
100
2
6
65
18
8
100
Source: Black and Gilson
(1998), p.250, Table 4.
Table 11. Information
Technology Industry in
India
(US$ million)
Software
Domestic
Exports
Hardware
Domestic
Exports
Peripherals
Domestic
Exports
Training
Maintenance
Networking and others
Total
1994-95
1995-96
1996-97
1997-98
1998-99
350
485
490
734
670
1083
950
1750
1250
2650
590
177
1037
35
1050
286
1205
201
1026
4
148
6
107
142
36
196
6
143
172
710
181
14
183
182
156
229
19
263
221
193
329
18
302
236
237
2041
2886
3805
5031
6052
1990-91
1995-96
1997-98
90
18
72
48
19
29
26
10
4
12
100
25.2
48.9
25
23
30
Note: Domestic market
includes both domestic
production and imports
Source: NASSCOM, New
Delhi, cited in Patibandla et
al. (2000)
Table 12.
India's
software exports by
activity
(Percent of total software
exports)
Professional services
(a) offshore
(b) onsite
Consultancy and Training
Data processing
Other services
Products and packages
Total
Rs billion
0
5
5
100
2.5
3
8.8
100
109
47
Source: Ghemawat and
Patibandla (1999), cited in
Patibandla et al. (2000)
Figure 2. Amazon.com stock market valuation and profitability