Technological achievement, high technology exports and growth.
Gani, Azmat
The relationship between high technology exports and per capita
economic growth in countries with higher levels of technological
achievement is examined. Three groups of countries classified as
technological leaders, potential leaders and dynamic adopters are chosen
for empirical analysis on the basis of the technological achievement
index. The regression results reveal that high technology exports exert
a statistically significant positive effect on growth of the
technological leader category of countries and a positive but
statistically insignificant effect on the potential leader category of
countries. The main policy implication is that low-income countries with
lower levels of technological achievement and growth may need to focus
on new product development with high technological content so as to be
competitive in the global trading environment as well as to enhance
their growth and development.
1. Introduction
Technological diffusion and advancements as experienced largely by
several high-income countries seem to be the overriding cause of their
trade patterns in technology specific products. The rise in
technological capabilities is due to technological creation,
improvements and advances that have taken place in a range of countries
on several fronts. Given the importance of technology, its role in
influencing nations economic well being and the global economic growth
process has been addressed elsewhere (United Nations Development Program
(UNDP), 2001; Chan and Yang, 2005; Fu, 2005; and Kim and Park, 2006).
The ongoing strong emphasis and investments in research and
development (R&D), largely in production sectors, has led to the
creation of high technology products, leading to high technology exports
(Liu and Lin, 2005). High technology exports are products with high
research and development intensity such as computers, pharmaceuticals,
scientific instruments, electrical machinery, consumer electronics,
software, transportation electronics and military and civil aerospace
products. Information technology (IT) goods such as electronic data
processing equipment, sol, ware, electronic components and
telecommunications equipment are also part of the high technology
products.
It is noted that high-tech manufacturing has been the fastest
growing area of world trade and now accounts for over one fifth of total
trade (see for example, UNDP, 2001, p.31). Such high-tech exports are
likely to have a significant effect on the economic well-being of
exporting countries (Liu and Lin, 2005). In addition, Sen (2002) has
emphasised that "the importance of global contact and interaction
applies to economic relations among others and that there is much
evidence that global economy has brought prosperity to many different
areas of the globe and in overcoming pervasive poverty, both modem
technology and economic interrelations have been influential."
Similar sentiments are also expressed by Stiglitz (2002).
This paper examines whether or not high technology exports have
been a determining factor in per capita economic growth in countries
with higher levels of technological achievement. Cross country annual
data for 45 countries for 19962004 are used in reduced form equations
for empirical analysis. Separate equations are estimated for the full
sample of 45 countries, the leaders, potential leaders and dynamic
adopters of technology, using the cross sectionally hetroskedastic and
timewise autoregressive procedure. In addition, a fixed effects
procedure correcting for AR(1) errors are also utilised. These three
procedures are repeated with the full sample countries as well as the
three categories of countries based on their technological achievement.
The next section describes global patterns in high technology
exports. Section Three presents the analytical framework followed by a
discussion of sample countries and data in Section Four. Section Five
presents the empirical results. Conclusions and policy implications
follow.
2. Global Technological Achievement and the Shift Towards High
Technology Exports
A major factor contributing to technological advancement across
several nations is investment in R&D and human capital formation.
For example, in the new growth framework, it has been suggested that a
country's productivity depends on its own investment in R&D as
well as spillovers from R&D by other countries (Grossman and
Helpman, 1991 and Barro and Salai-i-Martin, 1995). Thus, R&D
activities have contributed to technological improvements and product
development in several areas: communication, transportation networks,
information technology; computers and peripherals; industrial
manufactures; pharmaceuticals; and military and civil aerospace
technology.
The level of technological achievement of a country had first been
put into perspective by Rodriguez and Wilson (2000) through the Index of
Technological Progress (ITP), using five components--personal computers,
Interact hosts, fax machines, mobile phones and televisions, for l l0
countries. Their ITP showed high-income economies as leaders in
technological progress and poor countries as laggards.
In a similar vein, UNDP (2001, Table A2.1, p. 48-51) also
introduced a measure of technological achievement, the technology
achievement index (TAI) for 72 countries. The TA1 is a composite index
composed of (a) technology creation, (b) diffusion of recent innovation,
(c) diffusion of old innovation, and (d) human skills. According to the
TAI, there are 18 countries that are leaders in technological
achievement, 19 countries as potential leaders, 26 countries as dynamic
adopters and 9 countries as marginalized.
Building on similar attempts as that of UNDP (2001), Archibugi and
Coco (2004) devised a new indicator of technological capabilities for
developed and developing countries. These authors cite a wider number of
variables associated with technological change, with their measures
including three main components: the creation of technology,
technological infrastructures and the development of human skills.
While measures of technological capacity are now well established
as indicated above, technological improvements have led to the
development of industries that manufacture products of high
technological content. This has contributed to the growth of high
technology exporters. On the basis of the Heckscher-Ohlin trade theory,
nations' economic structures are likely to change, therefore,
changing factor endowments such as technological improvement would
result in shifts in the structure of trade. For example, Das (1998)
notes that "the product composition of exports would shift from a
predominance of natural resource intensive exports to unskilled labour
intensive exports, further to physical and human capital-intensive
exports, and then on to technology and knowledge-intensive
exports."
The expansion of high technology exports has given a quick push and
an expansion to growth in world trade. For example, technological
advancements have created new finished products, consumer and industrial
goods as well as new product market sectors. At the same time, improved
communications technology, continued improvements in efficiency of
international transport and lower levels of trade barriers have also
facilitated the rapid increase in global trade (The World Bank, 2000, p.
30). For example, Wolf (2003) notes that as the technologies of travel,
transportation and communications have improved, so inevitably have
returns to these activities, also noting that with the internet, the
cost of global communication is now close to zero. Falling costs of
global communication is also facilitating increased global trade.
Exports of high technology products are giving a quick boost to the
export incomes of several countries with high technological capability.
While export contributions to national growth and productivity is
obvious, imports of high technology goods can raise output directly as
inputs into the production process. For example, Connolly (2003)
provides empirical evidence that high technology goods imports from
developed countries not only positively affects domestic innovation, but
also leads to increased GDP growth as higher quality capital goods are
used in domestic production. In addition, Coe, Helpman and Hoffmaister
(1997) have noted that access to imported inputs facilitates the
diffusion of knowledge, which contributes to productivity while Wacziarg
(2001) notes that import competition increases not only the exit but
also the entry of domestic firms, spurring innovation.
The liberalization of trade has also contributed to the increased
internationalization of the world economies and expansion of the global
business environment (Edwards, 1993; International Monetary Fund, 2001
and Dreher, 2006). Combined with the ease of flow of goods and services
across several nations within the world communities, this has also
resulted in changing production structures of several nations (Mehta and
Parikh, 2005).
The developments in global exports are directly related to the
changes in sectoral production among several nations. Increases in
global exports have largely been concentrated between two sectoral
outputs: exports of manufactures and exports of services. For example,
trade in manufactures accounts for over 75 per cent of international
trade (The World Bank, 2000). There is a rapidly growing share of IT
goods in the world trade, which has risen from some 7.5 percent in 1990
to 11 percent in 1999, reflecting both growing demand for new technology
and the high price-to-weight ratio of IT goods, which contributes to
their greater tradability (International Monetary Fund, 2001).
Although the degree of dependence on high technology exports varies
considerably across the globe, the relative importance of high
technology exports has increased in almost all individual cases for all
of the countries selected in this study (Table 1). Among the
technological leader category, Singapore leads in high technology
exports while in the potential leader category; Malaysia outpaced all
other countries (Table 1). Among the dynamic adopters, Thailand is
ranked at the top in terms of high technology exports (Table 1).
3. The Analytical Framework
The estimation framework adopted here complements the more
macroeconomic oriented discussion on economic growth (Mankiw, Romer and
Well, 1992; Fischer, 1993; Barro and Sala-i-Martin, 1995; Yao, 2006;
Awokuse, 2006; and Tang, 2006). An important aspect of the empirical
framework is that key issues relating to high technology exports and
growth are unfolded.
Growth in output is presumed to be primarily a function of the
growth of factor supplies as well as several other variables affecting
the efficiency of resource allocation and factor productivity. The
variables tested are predominantly the main conventional variables as
used in many cross-country studies. These are: (a) the growth rate of
population to account for the growth in the productive labour force; (b)
the ratio of gross domestic investment to GDP to account for the growth
in the stock of physical capital; (c) the starting level of income per
capita to probe if it actually correlates with per capita growth over
the entire period (convergence hypothesis); (d) the number of
researchers involved in research and development per million people as a
proxy for human capital; and the rate of inflation. Jones (1998) and
Temple (1999) provide a comprehensive review of these conventional
variables while Rogers (2003) provides an in depth survey on economic
growth that focuses on a wide range of models and empirical results.
Exports have been regarded as an engine for growth. Past studies
examining export growth and growth in income have suggested that they
were significantly positively correlated. There are numerous studies
examining the possible link between exports and economic growth
(Balassa, 1978; Edwards, 1993; Fosu, 1996; Awokuse, 2006; Tang, 2006;
and Yao, 2006). On the other hand, Sharer (1999) points out that in
recent years, no country with an inward focused policy has proved
successful in attaining or sustaining a high internal growth rate of
GDP. High technology exports as a share of manufactured exports is the
prime variable, added to test whether it has any impact on per capita
growth. It is expected that a high technology exports coefficient would
show with a statistically significant positive sign, in the estimation
phase of the structural equation.
The estimation framework consists of the following generic forms:
[g.sup.ALL.sub.ir] = [l.sub.it][gamma] + [htx.sub.it] [pi] +
[v.sub.it] (1)
[g.sup.L.sub.ir] = [l.sub.it][gamma] + [htx.sub.it] [pi] +
[v.sub.it] (2)
[g.sup.PL.sub.ir] = [l.sub.it][gamma] + [htx.sub.it] [pi] +
[v.sub.it] (3)
[g.sup.DA.sub.ir] = [l.sub.it][gamma] + [htx.sub.it] [pi] +
[v.sub.it] (4)
where g is growth rate of gross domestic product per capita. The
generic form includes two types of regressors, l and htx. l represents
the standard variables, commonly known to influence per capita growth
and htx, in contrast, is high technology exports. The standard variables
in equations (1) to (4) are the growth rate of population (grp); the
ratio of gross domestic investment to GDP (inv); the real per capita GDP
(PPP) at the start of the period (pcgdp); the GDP deflator inflation
(ifn); the human capital, measured by number of researchers per million
people (hc); and htx is high technology exports as a share of
manufactured exports. ALL is a sample of 45 countries (leaders,
potential leaders and dynamic adopters); L is the leader's category
(17 countries), PL is the potential leader's category (18
countries) and DA is the dynamic adopter's category (10 countries).
The subscripts i and t are countries and time period respectively.
The error term in the above equation is [v.sub.it] with the
assumption that [v.sub.it] [approximately equal to] iid(0,
[[sigma].sup.2])
While equations (1) to (4) can be estimated with ordinary least
squares, the result is likely to be biased if the error terms are
correlated within each time series unit and are heteroscadastic across
each cross sectional unit, given that the data utilised here is
cross-sectional. Combining these assumptions means estimating a
cross-sectionally heteroscdastic and time-wise autoregressive model.
Hence, the initial estimation procedure begins with the estimation of a
cross-sectionally heteroskedastic and time-wise autoregressive model.
4. Same Countries and Data
The selection of countries for empirical analysis is based on the
Technological Achievement Index (TAI) developed by the UNDP (2001). The
TAI is numbered 0.0 to 1.0, indicating the level of innovation in a
society, with 0 being low and 1.0 being high. The TAI is a composite
index composed of(a) technology creation, (b) diffusion of recent
innovation, (c) diffusion of old innovation, and (d) human skills. Based
on the TAI, the UNDP (2001) classifies countries into four categories:
leaders (TAI of 0.50 and above); potential leaders (TAI between 0.35 and
0.49); dynamic adopters (TAI between 0.20 and 0.34); and marginalized
(TAI below 0.20). In this analysis, the countries are chosen from
leaders, potential leaders and dynamic adopter categories.
The sample of countries selected for the leader category include
Finland, United States of America, Sweden, Japan, Republic of Korea, the
Netherlands, United Kingdom, Canada, Australia, Singapore, Germany,
Norway, Ireland, New Zealand, Austria, France, and Israel.
The sample of countries selected for the potential leader category
include Spain, Italy, Czech Republic, Hungary, Slovenia, Hong Kong,
Slovakia, Greece, Portugal, Poland, Malaysia, Croatia, Mexico, Cyprus,
Argentina, Romania, Costa Rica, and Chile.
The sample of countries selected for the dynamic adopter category
include Bolivia, China, Colombia, Ecuador, El Salvador, India,
Indonesia, Panama, Thailand and Tunisia.
The sample period covers the years 1996-2004. The data source for
all variables is the Worm Bank's World Development Indicators
CD-ROM 2006 (The World Bank, 2006). Tables 2 and 5 in Section 5 present
the results of the contribution of high technology to growth of real GDP
per capita.
5. Estimation Procedure, Results and Discussion
Equation (1) includes 45 countries and 9 time periods; equation (2)
includes 17 countries and 9 time periods; equation (3) includes 18
countries and 9 time periods; and equation (4) includes 10 countries and
9 time periods. A cross-sectionally heteroskadastic and time-wise
autoregressive model is estimated first. This procedure of estimation is
also equivalent to the generalised least squares (GLS) estimation
(Kmenta, 1986). The results of this estimation procedure are recorded in
the in the second column of Tables 2 to 5.
The GLS equivalent estimation does not take into account
country-specific factors. While the sample of countries share somewhat
similar economic structures, the extent of the development of their
export sector differs from one country to another. To take into account
country-specific differences, a fixed effects estimation procedure
including country-specific dummy variables is adopted. In total there
are 45 dummies for equation 1, 17 for equation 2, 18 for equation 3, and
9 for equation 4. The no-constant option is adopted in the estimation
procedure so as to avoid the commonly known dummy variable trap. Tables
2 to 5 do not report results of these country effect dummy variables due
to space constraints.
The initial estimation results suggested autocorrelated errors
within the cross-sections in the fixed effects estimation procedure.
Given the nature of data, the possibility of AR (l) errors are likely
and so a third procedure is adopted: the fixed effects estimation
procedure corrected for AR (1) errors. The results of this third
procedure of estimation (AR(I) errors) is reported in the fourth column
of Tables 2 to 5 and is considered to be most robust of the three
estimation procedures adopted here.
Turning to the central focus of this paper, namely the impact of
high technology exports (htx), the robustly positive and statistically
significant coefficients in the full sample (Table 2) and leader
category (Table 3), and the positive but statistically insignificant
coefficients in the potential leader category (Table 4) are obtained.
The results, particularly of the leader category, confirm the strong and
positive impact of high technology exports on per capita growth. This
suggests that, insofar as the expansion of the technological base of the
export oriented industries entails a greater overall productivity and a
more efficient resource allocation within the economy, such effects are
favorably and strongly contributing to the growth in per capita income
in the leader economies category of countries. In the potential leader
category of countries, the coefficient of high-technology exports has
the expected positive sign but is statistically insignificant across the
three estimation methods adopted. The results of this category are not
surprising given that the sample countries in this category have low
levels of technological achievement and a limited range of
high-technology export products compared to the technological leader
category of countries.
The results show that the growth of population, grp, has a negative
and statistically significant effect in all but the dynamic
adopter's category. This might be viewed as counter intuitive at
first, as population growth is often hypothesized to have a negative
impact on per capita growth (Kelley, 1988). But the empirical evidence
on this relationship has been less than robust at best, if not
altogether inconclusive. The correlation between population growth and
per capita growth generally varies by the level of economic development
(in terms of both the direction and the size of the correlations with
the correlation likely to be negative in relatively poor countries and
positive in relatively wealthy countries. The result of this variable is
also consistent with some past empirical studies controlling for
population growth while testing growth equations, for example, Miller
(1996) and Burney (1996).
As expected, the coefficient investment (inv) is positive and
statistically significant in the all countries category (Table 2), the
dynamic adopter category (Table 5) and positive but statistically
insignificant in the potential leader's category (Table 4). The
results obtained for these categories of countries indicate that
physical capital accumulation is indeed an important factor for per
capita growth.
The results of variable pcgdp provide evidence in favour of the
convergence hypothesis in the all countries category (Table 2), leaders
and potential leader's categories and dynamic adopter categories
(Tables 3 to 5). The estimated coefficient of pcgdp is statistically
significant across all specifications in all cases. This expected effect
shown by pcgdp is also consistent with some of the previous growth
studies and confirms the pattern noted by many others.
As for human capital (hc), the coefficient is positive and
significant in all countries (Table 1), and positive but insignificant
in the potential leaders and dynamic adopter categories (Tables 4 and
5). In general, the modeling of the relationship between human capital
and economic growth has produced very mixed results. Engelbrecht (2003)
provides a good review of the empirical literature pertaining to this
issue. Existing empirical work shows both positive and negative effects
of human capital measures.
Tables 2 to 5 also show that the inflation rate variable, ifn, is
negative and statistically significant in the all countries (Table 2),
potential leaders and dynamic adopter's categories across all the
specifications tested. The results here for this variable are also
consistent with this and with other cross-country studies (Alexander,
1997). This suggests that high inflation rates are potentially
disruptive to per capita growth in these economies.
5. Conclusion and Policy Implications
Using methodologies common in the literature, this paper examines
whether or not high technology exports have recently been a determinant
in per capita economic growth in countries with higher levels of
technological achievement. The analysis used data from a sample of
countries based on the technological achievement index. The empirical
results of the technological leader category of countries provides
strong evidence of the positive impact of high technology exports on per
capita growth. In the potential leader category of countries, the
coefficient of high-technology exports has the expected positive effect
but is statistically insignificant. The result of this category is not
surprising given the limited range of high-technology export products.
The major policy implication is that countries aiming for high
growth may wish to expand trading into technology specific products.
However, in order to capitalise on this, technological capabilities will
have to be improved in areas of technology creation, improvement of the
technological base and product development. Such developments also call
for investments in R&D and human capital formation and a more open
trading environment. In addition, countries need to develop products in
industries that show high technology growth potentials.
There is likely to be a lower level of disagreement among policy
makers that trade can have beneficial effects on nations' growth
and development. By changing the export structure, that is, exporting
goods with greater growth potential such as those with high
technological content, nations can gain from a rapidly expanding trading
environment. Thus, countries with low technological achievement need to
focus on new product development with high technological content. This
in itself calls for more investments in research and development and
expansion of the human capital base.
It is also worthy of note that countries with a protected trading
environment have much to benefit from the import of goods with high
technological content if an open trading regime is established. It has
been noted that "trade is a mechanism by which more advanced
foreign technology can be used to the advantage of a less developed
country, not only to boost domestic innovation, but also as a means of
benefiting from continued foreign innovation." (Connolly, 2003). In
a further study, Yanikkaya's (2003) results provide strong evidence
in favour of the hypothesis that countries that have more trade with the
United States of America (one of the most highly innovative countries in
the world), are likely to grow faster and this phenomenon is especially
important for developing countries. Thus, even if countries cannot
produce and export goods with high technological content, they have much
to gain from imports of goods with high technological content and
establishing trade links with countries that have higher levels of
technological achievement.
While this analysis has been cross-sectional, future research
should concentrate on more country-specific studies of high technology
exports and growth of individual countries so as to develop more precise
policy instruments with regard to stimulating economic growth at a
country-specific level.
References
Alexander, W. R. 1997. Inflation and economic growth: evidence from
a growth equation. Applied Economics, 25: 233-238.
Archibugi, D. and Coco, A. 2004. A new indicator of technological
capabilities for developed and developing countries (ArCo). World
Development, 32(4): 629-54.
Awokuse, T. O. 2006. Export-led growth and the Japanese economy:
evidence from VAR and directed acyclic graphs. Applied Economics, 38(5):
593.
Balassa, B. 1978. Exports and economic growth: further evidence.
Journal of Development Economics, 5: 178-89.
Barro, R. J. and Salai-i-Martin, X. 1995. Economic Growth. Boston:
McGraw Hill.
Burney, N.A. 1996. Exports and economic growth: evidence from
cross-country Analysis. Applied Economics Letters, 3: 369-373.
Chen, J-R and Yang, C-H. 2005. Technological knowledge, spill over
and productivity: evidence from Taiwanese firm level panel data. Applied
Economics, 37(20): 2361.
Coe, D. T., Helpman, E., and Hoffmaister, A. W. 1997. North-South
R&D Spillovers. Economic Journal, 107:134-49.
Connolly, M. 2003. The dual nature of trade: measuring its impact
on imitation and growth. Journal of Development Economics, 72(1): 31-56.
Das, D. K. 1998. Changing comparative advantage and the changing
composition of Asian exports. The World Economy, 21 (1): 121-140.
Dreher, A. 2006. Does globalization affect growth? Evidence from a
new index of globalization. Applied Economics, 38(10): 1089.
Edwards, S. 1993. Openness, trade liberalization and growth in
developing Countries. Journal of Economic Literature, 31(3): 1358-93.
Engelbrecht, H-J. 2003. Human capital and economic growth:
cross-section
evidence for OECD countries. The Economic Record, S40-S51.
Fischer, S. 1993. The role of macroeconomic factors in growth,
Journal of Monetary Economics, 32 (3): 485-512.
Fosu, A. K. 1996. Primary exports and economic growth in developing
countries, World Economy, 19: 465-475.
Fu, X. 2005. Exports, technical progress and productivity growth in
a transition economy: a non-parametric approach for China. Applied
Economics, 37(7): 725.
Grossman, G. M. and E. Helpman. 1991. Innovation and Growth in the
Global Economy. Cambridge, MA: MIT Press.
International Monetary Fund. 2001. World Economic Outlook. October
2001. The Information Technology Revolution. Washington, D.C.
Jones, C. 1. 1998. Introduction to Economic Growth. New York:
Norton.
Kelley, A. C. 1988. Economic consequences of population change in
the Third World, Journal of Economic Literature, 26(4): 1685-728.
Kim, T. and Park, C. 2006. Productivity growth in Korea: efficiency
improvement or technical progress? Applied Economics, 38(8): 943.
Kmenta, J. 1986. Elements of Econometrics. Macmillan: New York.
Liu, W-H and Lin, Y-C. 2005. Foreign patent rights and high-tec
exports: evidence from Taiwan. Applied Economics, 37(13): 1543.
Mankiw, N. G., Romer, D. and Weil, D. N. 1992. A contribution to
the empirics of economic growth. Quarterly Journal of Economics, 107(2):
407-437.
Mehta, R. and Parikh, A. 2005. Impact of trade liberalisation on
import demands in India: a panel data analysis for commodity groups.
Applied Economics, 37(16): 1851.
Miller, S. M. 1996. A note on cross-country growth regressions.
Applied Economics, 28: 1019-1026.
Rodriguez, F. and Wilson, E. J. III. 2000. Are Poor Countries
Losing the Information Revolution? University of Maryland at College
Park. A paper prepared for infoDev. Available at
http://infodev.org/library/working.html
Rogers, M. 2003. A Survey of economic growth. The Economic Record,
79(244): 112-135.
Sen, A. 2002. Globalisation, inequality and global protest.
Development, 45(2): 11-16.
Sharer, R. 1999. Trade: an engine of growth for Africa, Finance and
Development, 36(4): 26-29.
Stiglitz, J. E. 2002. Globalization and its Discontents. Penguin
Books: London.
Tang, D. 2006. The effect of financial development on economic
growth: evidence from the APEC countries, 1981-2000. Applied Economics,
38(16): 1189.
Temple, J. 1999. The new growth evidence, Journal of Economic
Literature, XXXVII:112-156.
The World Bank. 2006. World Development Indicators CD ROM 2006. The
World Bank, Washington, D.C.
The World Bank. 2000. World Development Report 2000. Oxford
University Press: Washington, D.C.
United Nations Development Programme. 2001. Human Development
Report. New York: United Nations.
Wolf, M. 2003. Is globalization in danger? The World Economy,
26(5): 313-412.
Wacziarg, R. 200 I. Measuring the dynamic gains from trade. World
Bank Economic Review, 15(3): 393-429.
Yanikkaya, H. 2003. Trade openness and economic growth: a
cross-country Investigation. Journal of Development Economics, 72:
57-89.
Yao, S. 2006. On economic growth, FDI and exports in China. Applied
Economics, 38(3): 339.
by Azmat Gani
The University of the South Pacific, Suva, Fiji Islands
Table 1: High Technology Exports as a Percent of Manufactured Exports
Country 1995 2000 2004
Leaders
Australia 15.7 15.2 13.6
Austria 8.6 13.0 12.1
Canada 15.1 18.6 13.6
Finland 14.7 27.3 20.9
France 18.7 23.8 19.1
Germany 12.9 18.0 17.2
Ireland 45.8 47.5 33.8
Israel 15.8 25.1 18.8
Japan 26.1 28.3 23.7
Korea, Rep. 25.9 34.8 32.8
Netherlands 23.8 35.4 29.1
New Zealand 11.6 14.8 13.7
Norway 14.0 17.1 18.3
Singapore 53.9 62.6 58.9
Sweden 16.1 22.1 17.2
United Kingdom 27.3 30.0 24.1
United States 32.8 35.3 32.3
Potential Leaders
Argentina 3.5 9.0 7.6
Chile 3.3 3.4 4.8
Costa Rica 6.1 51.6 36.8
Croatia 5.9 8.5 13.0
Cyprus 13.4 6.4 22.1
Czech Republic 5.0 8.1 12.9
Greece 5.7 13.3 11.4
Hong Kong, China 16.4 23.6 32.0
Hungary 6.8 26.4 28.9
Italy 8.0 9.2 7.7
Malaysia 46.1 59.5 55.4
Mexico 15.1 22.4 21.2
Poland 2.7 3.3 3.2
Portugal 5.4 6.4 8.7
Romania 2.3 5.5 3.4
Slovak Republic 3.8 4.1 5.2
Slovenia 3.4 4.7 5.5
Spain 7.1 7.6 7.0
Dynamic Adopters
Bolivia 14.0 40.0 9.2
China 10.0 18.6 29.8
Colombia 7.0 7.7 5.6
Ecuador 6.0 5.6 7.3
El Savador 8.0 6.0 4.1
India 4.0 5.0 4.9
Indonesia 7.0 16.2 16.1
Panama 1.0 0.1 2.1
Thailand 24.0 33.3 30.2
Tunisia 2.0 3.4 4.9
Source: The World Bank (2006).
Table 2: Regression Results -All Countries
Fixed Effects
Fixed Corrected for
Variable GLS Effects AR(1) Errors
constant 1.667 ... ...
(3.808) *
grp -1.061 -0.984 -0.989
(18.420) * (9.069) * (8.868) *
inv 0.133 0.107 0.088
(7.996) * (4.204) * (3.649) *
pcgdp -0.0001 -0.0001 -0.0002
(6.056) * (5.884) * (6.092) *
ifn -0.063 -0.088 -0.086
(5.608) * (6.823) * (7.349) *
he 0.0007 0.0002 0.0002
(0.909) (1.770) *** (2.068) **
htx 0.037 0.025 0.033
(4.979) * (1.902) ** (2.642) *
N 405 405 405
R-square 0.54 0.47 0.53
DW 1.75 2.19 ...
Von Neumann 1.76 2.19 ...
LM (Hetroskedasticity) ... 273.64 ...
B-P ... 1182.9 ...
t-statistics are in parentheses.
*, ** and *** indicates statistically significant at the 1, 5 and 10%
levels respectively.
Table 3: Regression Results--Leaders Category
Fixed Effects
Fixed Corrected for
Variable GLS Effects AR(1) Errors
constant 4.636 ... ...
(4.694) *
grp -1.078 -0.930 -0.0934
(17.610) * (7.257) * (7.152)
inv -0.006 -0.037 -0.043
(0.353) (1.403) (1.625) ***
pcgdp -0.0002 -0.0002 -0.0002
(5.202) * (3.411) * (3.408)
ifn 0.215 0.838 0.077
(3.577) * (1.116) (1.032)
hc 0.0004 -0.0006 -0.0007
(0.419) (0.464) (0.599)
htx 0.087 0.075 0.077
(7.898) * (5.268) * (5.427) *
N 153 153 153
R-square 0.71 0.63 0.65
DW 1.77 2.10 ...
Von Neumann 1.78 2.12 ...
LM (Hetroskedasticity) ... 87.76 ...
B-P .... 149.16 ...
t- statistics are in parentheses.
* and *** indicates statistically significant at the 1 and 10% level
respectively.
Table 4: Regression Results--Potential Leaders Category
Fixed Effects
Fixed Corrected for
Variable GLS Effects AR(1) Errors
constant 2.099 ... ...
(2.087) *
grp -0.803 -1.108 -0.965
(4.858) * (4.727) * (4.786) *
inv 0.115 0.064 0.045
(3.439) * (1.322) (0.885)
pcgdp -0.0002 -0.0001 -0.0001
(4.451) * (2.083) ** (2.835) *
ifn -0.069 -0.078 -0.081
(4.180) * (7.060) * (7.473) *
he 0.0004 0.0007 0.0009
(1.237) (1.098) (1.530)
htx 0.019 0.032 0.029
(1.236) (1.539) (1.411)
N 162 162 162
R-square 0.39 0.38 0.56
DW 1.86 2.48 ...
Von Neumann 1.88 2.50 ...
LM (Hetroskedasticity) ... 63.63 ...
B-P ... 245.3 ...
t-
statistics are in parentheses.
* and *** indicates statistically significant at the 1 and 5% level
respectively.
Table 5: Regression Results--Dynamic Adopters Category
Fixed Effects
Fixed Corrected for
Variable GLS Effects AR(1) Errors
constant 1.053 ... ...
(0.413)
grp -1-198 0.690 0.908
(1.365) (0.532) (0.676)
inv 0.267 0.321 0.323
(4.945) * (4.327) * (4.542) *
pcgdp -0.0004 -0.0004 -0.0004
(2.772) * (2.188) ** (2.512) **
ifn -0.114 -0.144 -0.146
(3.871) * (3.840) * (4.086) *
he -0.0003 0.002 0.002
(0.249) (1.132) (1.273)
htx -0.033 -0.022 -0.0001
(1.520) (0.608) (0.003)
N 90 90 90
R-square 0.63 0.59 0.63
DW 1.65 2.25 ...
Von Neumann 1.66 2.27 ...
LM (Hetroskedasticity) ... 66.58 ...
B-P ... 55.60 ...
t- statistics are in parentheses.
* and ** indicates statistically significant at the 1 and 5% level
respectively.
