...

PDF

by user

on
Category: Documents
3

views

Report

Comments

Description

Transcript

PDF
Do Stronger Intellectual Property Rights Protection Induce More Bilateral Trade?
Evidence from China’s Imports
Titus O. Awokuse*
Department of Food & Resource Economics
University of Delaware
Newark, DE 19717, USA
Email: [email protected]
Hong Yin
Department of Economics
University of Delaware
Newark, DE 19717, USA
Email: [email protected]
Selected Paper prepared for presentation at the American Agricultural Economics Association
Annual Meeting, Orlando, FL, July 29, 2008
______________________________
*Copyright 2008 by Titus O. Awokuse and Hong Yin. All rights reserved. Readers may make
verbatim copies of this document for non-commercial purposes by any means, provided that
this copyright notice appears on all such copies.
1
Do Stronger Intellectual Property Rights Protection Induce More Bilateral Trade?
Evidence from China’s Imports
Abstract
Most of the previous studies on the effect of IPR protection on international trade have been
from the perspective of major industrialized nations. However, much of the current debate on
the effects of IPR protection involves large developing countries with high threat of imitation.
This study contributes to the literature by analyzing the impact of the strengthening of patent
laws in China on its bilateral trade flows. We estimate the effects of patent rights protection
on China’s imports at the aggregate and detailed product categories for both OECD
(developed) and non-OECD (developing) countries.
The empirical results suggest that
increased patent rights protection stimulate China’s imports, particularly in the knowledgeintensive product categories. Furthermore, while the evidence in support of the market
expansion effect is significant for imports from OECD countries, it is rather weak and mostly
insignificant for imports from non-OECD countries.
Key words: Intellectual property rights, patent laws, international trade
JEL classification: F13, 034
2
1. Introduction
In the past two decades, the nature of the linkages between intellectual property rights
(IPR) and international trade has been the source of much debate and controversy.
Disagreements persist on whether stronger IPR stimulate or discourage trade. Two key
developments contributed to the recent interest in this issue. In the political arena, the status
of IPR as a form of trade barrier became an issue of greater global concern after the enactment
of a special provision in the US Trade Act of 1988 which linked American trade policy to the
prevailing IPR regimes in bilateral trading partner nations. In addition, IPR became even
more important when increasing national disputes over IPR led to the multilateral World
Trade Organization’s (WTO) Agreement on Trade-Related Aspects of Intellectual Property
Rights (TRIPs) in 1994.
Since trade in knowledge-based goods is an important source of
innovation and technology for low-income countries, it is not surprising that one of the main
issues surrounding the IPR debate was centered on the need for greater IPR protection in
developing countries.
But despite the importance of IPR laws to developing countries,
relatively limited empirical evidence exist on the impact of IPR regimes on bilateral trade
flows with developing nations (Schneider, 2005).
This study contributes to the literature by analyzing the impact of the strengthening of
patent laws in China on its bilateral trade flows. China’s recent reform of its patent laws and
its status as a large developing nation with strong threat of imitation makes it an interesting
case study on the effect of patent protection on trade. As in previous studies, we also explore
the possibility that the trade effect of patent protection may vary by product sectors and by the
level of economic development in trading partner countries (Maskus and Penubarti, 1995;
Smith, 1999).
Since industrialized countries, i.e., major OECD countries, are the main
3
producers of new technology, we expect that the strengthening of patent protection in China
will have a stronger effect on bilateral trade (import) flows to China from OECD countries
relative to import flows from non-OECD countries. The results from this study provide much
needed empirical evidence on the current debate regarding policy reforms in IPR regimes and
its effects on technology transfer and trade with China.
This paper differs from previous studies in several ways.
First, this is the first
empirical study based on one developing country which experienced significant changes in its
IPR systems (including its patent laws) in the past two decades. Thus, this analysis from the
perspective of a large developing country provides an alternative to most previous studies
which usually emphasize export flows from a major industrialized nation to a diverse group of
importers (Maskus and Penubarti, 1995; Smith, 1999; Rafiquzzaman, 2002). Second, in
contrast to studies based on data from a single year, this study uses a panel data (1991-2004)
which covers a more extended time period and allows for the consideration of the dynamic
nature of the relationship between international trade and policy changes in patent regimes.
Third, we use an alternative measure of patent rights that may be more reflective of
actual patent activities. Previous studies typically use patent rights indices or scores based on
the works of Rapp and Rozek (1990) and Ginarte and Park (1998). These indices usually use a
scoring method that is often arbitrary in the choice of weights on the importance of various
criteria. Although useful in some cases, the index-based measures of patent rights may not
adequately capture the dynamic nature of the interaction between changes in patent laws and
standards over time and their potential impact on other economic variables (e.g., trade).
When possible, it may be more instructive to use actual data on the number of patent
applications over time as adopted in this study. Thus, we use annual patent applications from
4
importing countries as a measure of the strength of patent rights protection in China. The
growing number of foreign patents filed each year can be a good indicator of growing
confidence of foreign firms in the patent rights protection offered in China. This measure of
patent rights strength accounts for more variation across time and may be less susceptible to
measurement errors.
The main finding from this study is that the strengthening of patent laws in China led
to an increase in its import flows, particularly in knowledge-intensive goods. This paper’s
empirical results further support the hypothesis that the strengthening of patent laws has a
market expansion effect in China. However, we also find that market expansion effect is only
strong and significant for imports flows from major OECD countries, though the effect seems
to be declining over the years. In contrast, the evidence for market expansion effect is much
weaker for imports from non-OECD countries. Also, the results suggest that the effects of
patent rights protection on import flows vary by different product sectors and are strongest in
the knowledge-intensive sectors.
The rest of the paper is organized as follows. Section 2 contains a brief review of
literature on the trade effects of IPR. Section 3 describes the model specifications and data
sources. Section 4 provides a discussion of results and section 5 concludes the paper.
2. Patent rights protection and international trade: A Puzzle
2.1 What does theory predict?
Although the theoretical linkage between patent rights and innovation has received much
attention, relatively fewer theoretical studies exist on the relationship between patent rights
and international trade (Chin and Grossman, 1988; Helpman, 1993; Ethier and Markusen,
5
1996; Maskus and Penumbarti, 1997; Markusen, 2001; Qian, 2007). The existing theory
suggest that the strengthening of patent rights protection could have two opposing effects on
bilateral trade flows between countries: market expansion or market power effects (Maskus
and Penubarti, 1995).
A market expansion effect may occur if the strengthening of patent protection discourages
domestic firms from imitating the technologies embodied in imported goods. The resulting
reduction in the production of competing domestically produced imitation goods should
encourage foreign firms with better technologies to expand their exports to those markets as a
result of increased net demand for their products. In contrast, a market power effect may
occur if foreign firms respond to stronger patent laws in an importing nation by choosing to
reduce their foreign market sales in that market and take advantage of higher unit price. The
countervailing nature of the market expansion and market power effects implies that the
direction of the impact of patent rights strengthening on trade is ambiguous (Maskus and
Penumbarti, 1995; Smith, 1999).
This ambiguity could be attributed to the complex
interactions between local market demand, the degree of imitative production, and the nature
of trade barriers (Maskus, 2000, p.113). Thus, the debate is on-going regarding the nature of
the relationship between IPR and international trade. Since this is an empirical question, most
of the previous investigations have been data-driven.
2.2 Review of previous empirical studies
In a seminal study, Maskus and Penubarti (1995) used an augmented version of the
Helpman-Krugman model of monopolistic competition to estimate the effect of patent
protection on international trade flows. Their results, based on 1984 bilateral trade data, show
6
that the market expansion effect dominate the market power effect as they found that higher
levels of patent protection have a positive impact on manufacturing exports of OECD nations
to developing countries. Subsequent studies provide some qualifications on the evidence in
support of market expansion effect. Smith (1999) further extends this line of inquiry by
exploring the effect of the threat of imitation in the importing countries. The threat of
imitation is weakest in countries with weak imitative abilities and strong patent laws and is
strongest in countries with strong imitative abilities and weak patent laws. The market
expansion effect is expected to be more pronounced in the market with high threat of
imitation, while the market power effect should be more likely in importing countries with
low threat of imitation. Using US manufacturing exports data, Smith (1999) showed that the
link between patent rights protection and international trade depends on the ability of the
importer to imitate the exporter’s technology. She found empirical evidence supporting the
market expansion effect for US manufacturing exports to countries with high imitation threat,
but found the market power effect to be more prevalent for exports to countries with weak
threat of imitation.
In a recent study based on Canadian exports data, Rafiquzzaman (2002) found similar
results indicating that stronger patent laws induced more Canadian exports to countries with
strong threat of imitation and less exports to those markets with weak threat of imitation. In
addition, Fink and Braga (1999) examined the IPR and trade nexus using 1992 data for a
cross-section of 89 countries and found that stronger patent rights increase bilateral flows of
manufactured non-fuel imports. They noted that the positive link is weaker for trade in the
high-technology sectors. Other studies on the relationship between patent rights and trade
7
flows to developing countries draw similar conclusions (Lesser, 2001; Park and Lippoldt,
2003).
2.3 Patent rights protection in China
Among developing countries, China is particularly suitable as a case study for
analyzing the impact of IPR reforms on bilateral trade flows. Although China is one of the
largest economies in the world and operates a significant trade surplus with most other nations,
it is a net importer of capital-intensive manufacturing products. For example, about 80% of
China’s imports are used as intermediate inputs in its growing manufacturing sectors
(Tongzon, 2001). Surprisingly, China had no patent rights protection before 1985. However,
since the establishment of its first patent law in 1985 and its substantial revision in 1992,
China has undergone a gradual reform of its patent systems in order to achieve compliance
with international laws. Although the 1985 version of China’s patent law contained its most
comprehensive requirements hitherto, it still lacked several important components.
In 1992, mainly due to the pressure from the United States, China’s patent law was
substantially amended to bring it closer to those of many industrial nations (Allison and Lin,
1999). The strengthening of China’s patent laws was also accelerated by its membership in
several IPR-related international treaties, such as the World Intellectual Property Organization
(WIPO) in 1995. Most recently, the latest revision of China’s patent law was made in 2000 as
part of the preparations for China’s entry into the World Trade Organization (WTO).
Furthermore, as a recent member of the WTO, China has also made significant efforts toward
aligning its IPR laws (i.e. patents, copyrights and trademarks) with the requirements of the
TRIPS agreement and other major international IPR conventions (Maskus, 2004, 2006).
8
Since 1992, foreign patents have increased steadily with an annual growth rate of 19% (see
Figure 1).1 The majority of Chinese patents (invention patent applications) are filed by foreign
firms and the most important foreign patentees are from Japan, the US, EU countries, and
South Korea.
3. Econometric methods
3.1 Model Specification and Hypotheses
We apply a conventional gravity model of bilateral trade to estimate the effects of
China’s patent rights protection on its import flows. The gravity model has a long history in
theoretical and empirical studies of international trade (Anderson, 1979; Bergstrand, 1989,
1990). Typically, it estimates bilateral trade flows as a function of exporter and importer
characteristics and various determinants of trade, including factors that distort trades (e.g.,
tariff). The gravity model could be represented by the following equation:
IMPORTijt = α(GDPit)β (GDPjt)δ (DISTijt)γ (Aijt)ηεijt,
(1)
where i, j, and t are indexes for the source country, the destination country and year,
respectively. IMPORTijt denotes the imports of the source country to the destination country
(China). Gross domestic product (GDP) is employed as a proxy for national income or
market size. GDPit and GDPjt denote the GDP of the source and destination countries,
respectively.
The variable DISTij measures the distance between the source and the
destination countries. Aijt represents trade distortion factors, and εijt is a normally distributed
error term with zero mean and constant variance.
1
The figure is computed by the authors, based on the patent data collected directly from Chinese patent database
CNPAT ABSDAT (http://www.sipo.gov.cn).
9
Equation (1) could be augmented with the inclusion of Chinese tariff rate (TARIFFijt)
and a measure of patent rights protection in China (PATENTijt). After taking the natural logs,
the resulting model is expressed as:
Log (IMPORTijt) = α + β 1 Log (GDPit) + β 2 Log (GDPjt) + β 3 Log (DISTijt) +
β 4 Log (1+TARIFFijt) + β 5 Log (PATENTijt) + εijt
(2)
where β 1 and β 2 represent the income elasticities of import demand and the expected sign is
positive. β 3 is expected to have a negative sign as distance is a proxy for transportation cost
which should vary inversely with imports. for β 4 is the import tariff coefficient and the
expected sign is negative indicating that higher tariff rate should discourage imports. As
discussed previously, the existing theory suggests that the sign of the coefficient on the
patents variable, β 5 is ambiguous. A positive sign on this parameter estimate will provide
support for the market expansion effect while a negative sign will suggest support for the
market power effect. Nevertheless, since China is commonly perceived as a large country
with strong threat of imitation, it is reasonable to hypothesize that the market expansion effect
will dominate such that stronger patent protection will result in more imports into China .
3.2 Model estimation and data issues
Hsiao (2003) argues that OLS estimates of equation (2) may yield biased estimates if
certain specification issues are not adequately addressed. First, for a large and diverse crosssection of countries as in this study, OLS is subject to unobservable heterogeneity bias. A
common remedy is the specification of a model that includes country-specific effects in the
panel regressions. Hausman (1978) specification test result suggests that the random effects
10
model is more appropriate than the fixed effects model. The random effects estimator, which
accounts for the unobserved country-specific effects, requires the transformation of equation
(2) so that each variable for each country is normalized such that the time-invariant countryspecific effects are removed.
The mean of unobserved and specific country effects is
reflected in the single constant term.
Furthermore, for a cross-section of countries of various sizes as in this study and given
the notable growth in China’s bilateral trade, it is plausible to expect the presence of
heteroskedastic errors. Likelihood ratio test for heteroskedasticity was applied and the null
hypothesis of homoskedastic errors was easily rejected at the 5 percent significance level
suggesting the presence of heteroskedasticity (Green, 2000). Hence, the parameter estimates
reported in the tables are based on robust standard errors from Huber-White’s
heteroskedasticity-consistent covariance matrix.
In addition, there is the potential for simultaneity bias because the causal link between
imports and some of the regressors may be bi-directional. For example, while import growth
may affect GDP, the reverse is also possible. Furthermore, countries with higher per capita
income tend to export more on average and have higher likelihood of producing patented
knowledge-intensive products. If endogeneity problem exists, then an instrumental variable
(IV) estimation method is more appropriate. Results from the Hausman test for the
disaggregate product level data indicate that the null hypothesis of no simultaneity bias could
not be rejected at the 5 percent significance level. Thus, in order to account for crosssectional heterogeneity, a feasible generalized least squares (FGLS) estimation technique was
applied to the data (Baltagi 2001, p. 79-80).
11
The analysis involves the estimation of equation (2) for both aggregated import and
for twenty manufacturing product sectors, using a panel data set covering 1991-2004. In
order to remove short-term cyclic fluctuations of patent filings, three-year averages were
used.2 The sample consists of 36 countries which include 21 OECD countries and 15 nonOECD countries. These are the top trading partners of China with significant number of
patent applications over the estimation period. The summary statistics of variables used in
this study are provided in Table 1.
Gross domestic product (GDP) of both source countries and China are taken from
World Development Indicators 2006. Trade data (IMPORTijt) for the twenty product sectors
are organized according to the 2-digit classification of the Standard International Trade
Classification (SITC) and were obtained from the United Nations Comtrade database.3 These
sectors can be further grouped into knowledge-intensive products (mainly outputs from
science-based industrial sectors) and non-knowledge-intensive products (mainly outputs from
traditional or low-tech industrial sectors). All import values are deflated to 2000 constant US
dollar using China’s CPI index. As in previous studies, the tariff rate (TARIFFijt) is measured
as the ratio of total import duties to the total value of imports (Maskus and Penubarti, 1995;
Fink and Braga, 1999; Rafiquzzaman, 2002). The data on tariff duties were obtained from
various issues of China Statistical Yearbook. The distance variable is measured as the
distance
in
miles
between
capital
cities
of
importing
countries
and
China
(http://www.indo.com/distance).
2
The data used the following averages: 1991-1993, 1994-1996, 1997-1999, 2000-2002, and 2003-2004. Thus
we have a total of 180 observations for model estimations.
3
UN Comtrade database is available at http://unstats.un.org/unsd/comtrade.
12
3.3 Measurement of patent rights protection
In several previous studies, the strength of patent rights protection was measured using
IPR indexes constructed by Rapp and Rozek (1990) and Ginarte and Park (1997). These
popular IPR rankings are usually based on five components: (1) the extent of patent rights
coverage, (2) membership in international treaties, (3) enforcement mechanisms, (4) duration
of protection, and (5) provisions against loss of protection.
While this approach for
measuring IPR strength has its merits, it also has some limitations. For example, the indexes
are based on laws on the books and may not accurately reflect dynamic changes in actual
patent protection practices in these countries. The criteria used in constructing the indexes
could be quite subjective with potential for large measurement errors and may not adequately
reflect the effect of diversity in levels of economic development (Fink and Braga, 1999).
In contrast to other studies, we measure the strength of patent protection as the actual
number of patent applications (PATENTijt) from importing countries. We assume that the
steady and growing number of patent applications from industrialized countries in the last
decade closely reflects the growing confidence of foreign firms in the improvements in
China’s patent rights protection.
Thus, we employ patent applications from importing
countries as a proxy for the strength of patent rights in China. Relative to the available IPR
scores, this measure of patent protection provides more variations and less measurement
errors from more subjective considerations in constructing IPR indexes. The data for patent
applications were obtained directly from official Chinese patent database. This database is
maintained by the State Intellectual Property Office (SIPO) of China (http://www.sipo.gov.cn).
13
4. Empirical results
4.1. All Countries Combined
Table 2 presents the empirical results for the estimated random effects model from
equation (2) using data for all countries combined. The first row reports parameter estimates
for aggregate imports while subsequent rows contain estimates for disaggregated 2-digit SITC
product categories. In general, the coefficients on the control variables (incomes for exporters
and China, tariff, and distance) have the expected signs and are significant at the 10 percent
level. The remainder of the discussion of results will focus on the effect of the patent variable
on imports. At the aggregate product level, the effect of foreign patents on imports has the
expected positive but is statistically insignificant at the 5 percent level (see Table 1, first row).
It is possible that the high level of data aggregation (product categories and level of
development differences across countries) is hiding much of the impact of patents on imports.
It may be more interesting to examine the effect of patents at a less aggregated product level.
At the disaggregated product sector level, the impact of patents appears to be
relatively stronger. The results (shown in Table 2) indicate that strengthened patent rights
protection have a positive and significant effect on imports in seven product categories and
negative and significant effect for one sector (beverages and tobacco). This finding implies
strong support for the market expansion hypothesis. Furthermore, the market expansion
effect tend to be stronger and larger on average in knowledge-intensive sectors (i.e.,
electronics, instruments, machinery, pharmaceuticals, rubbers and transports) where the
manufacturing process usually involves significant investment in research and development.
The impact of patents on pharmaceuticals (where the patent coefficient equals 1.07) is the
largest among the sectors. Since the majority of patents are filed by knowledge-intensive
14
product sectors, it is not surprising to find that the impact of patents is relatively larger in
these sectors. The results at the sector level are supportive of the theory that the market
expansion effect is predominant in a large market with high threat of imitation (Smith, 1999).
A notable exception to the finding of market expansion effect is in the beverages and
tobacco sector where the patent coefficient value is -0.716.
This sector’s negative and
statistically significant estimate provides some evidence in support of the market power effect.
This finding is consistent with Smith (1999) who also found evidence for market power effect
of increased patent protection in the US exports of tobacco products.
4.2 OECD countries versus non-OECD countries
Next, we consider the possibility that the impact of patent protection on imports may
vary by the level of economic development.
OECD countries are the major sources of
advance technologies and consequently majority of foreign patents in China are filed by those
countries. In contrast to imports from non-OECD nations, technologies embodied in imports
from OECD countries are on average more patent-related. Most developing (or non-OECD )
countries still lack sufficient innovation ability to develop high level technologies which are
worth patenting. Thus, we expect the effect of patent rights protection on imports to be more
pronounced across all sectors for OECD countries, but should be relatively weaker and less
significant for non-OECD nations.
Using the World Bank Development Indicators
classification scheme, the countries in our sample were divided into two groups: high-income
OECD (developed) and low-income non-OECD (developing) countries.4
4
The OECD countries used in the analysis includes: Australia, Austria, Belgium, Canada, Denmark, Finland, France,
Germany, Ireland, Israel, Italy, Japan, Korea, Netherlands, New Zealands, Norway, Spain, Sweden, Switzerland, United
Kingdom, USA.
15
Table 3 contains the results from the estimation of equation (2) for high-income
OECD (developed) countries only. In general, the evidence indicate that patent protection
have a positive impact on China’s imports, further confirming the support for a market
expansion effect. At the aggregated product level, the model fits improved significantly from
the combined countries results in Table 2. The patent coefficient (0.343) is positive and
statistically significant. Similarly at the disaggregated product sector level, the estimated
patent coefficients are positive and significant for nine sectors and negative and significant for
one sector (i.e., beverages and tobacco). The market expansion effect is more pronounced in
knowledge-intensive sectors: nine of the knowledge-intensive sectors are significant. For
knowledge-intensive sectors, the estimated patent coefficients range from 0.574 in plastics to
0.926 in instruments. These results show again that the effects of patents on imports vary not
only between knowledge-intensive sectors and non-knowledge-intensive sectors but also
within these two categories. Similar to previous results, the beverages and tobacco sector has
a negative and statistically significant estimated patent coefficient value of -1.310, suggesting
evidence for market power effect of increased patent rights protection in this secto.
The variations in the effects of patents across product sectors can be attributed to
various reasons. For instance, the propensity to obtain patents for a product varies across
different industrial sectors, due to different technology opportunities, and market conditions.
In some industries firms have alternative means to appropriate the returns to their investment,
such as lead-time advantages or brand name reputation, or their technologies may be difficult
The non-OECD countries are: Argentina, Brazil, Chile, Czech, Hungary, India, Indonesia, Malaysia, Mexico , the Phillipines,
Poland, Russia, Singapore, South Africa , Thailand.
16
to imitate or reverse engineer. In this case, firms may not seek IPR protection but rather rely
on natural protection of their innovations.
Furthermore, within the knowledge-intensive
sectors, patent protection are generally considered to be very important for those industries
where their technologies are relatively easy to imitate, such as chemicals, pharmaceuticals and
computer industries In contrast, patent protection are viewed to be of moderate importance in
other industries (e.g., metals, machinery and transports) where the technologies are relatively
more difficult to imitate (Park and Lippoldt, 2003).
Table 4 presents the empirical results for non-OECD (developing) countries. This
group of countries includes the most important emerging economies, such as India, Brazil,
South Africa and Southeast Asian countries. As expected, the effect of patents on trade is
significantly weaker for these non-OECD countries. For example, at the aggregate product
level, the estimated patent coefficient is not statistically significant at conventional levels. In
sharp contrast to the results for OECD countries (see Table 3), only two disaggregated
product categories have statistically significant patent coefficients: pharmaceuticals (1.09)
and wood products (-1.20).
In the former case, a comparison of the estimated patent
coefficients across OECD and non-OECD countries suggest that while 1% increase in total
patent filings leads to a 1. 09% increase in pharmaceutical imports from non-OECD
(developing) countries; it only led to 0.532% increase in imports from OECD (developed)
countries.
This finding confirms earlier conclusions in previous studies which suggest that firms
in rich industrial nations may respond to increase in patent protection by significantly
increasing their level of FDI or licensing as an alternative mode of entry instead of just
increasing international trade. As discussed in Smith (2001), there might be simultaneous
17
effects of patents on imports, FDI and /or licensing for developed countries. Hence, the
patent coefficients tend to be smaller as some of the market expansion effect is offset by FDI
or licensing.5 In contrast, when the effect is significant, large developing countries appear to
increase trade volumes to a country with stronger patent protection.
4.3
The dynamic effect of patents on trade
The results discussed above reflect the average effects of patents on import flows into
China over the years 1991-2004. However, there were significant reforms in patent laws
during these years. In order to better understand how the effects of patents on imports varied,
this section analyzes the data by patent regimes that coincide with major patent policy reforms
in China. To properly identify the different patent regimes in China, we use five elements
used in computing IPR scores (Ginarte and Park, 1997) as a reference. Beginning with
China’s first patent law in 1985, China’s patent policy could be classified into three different
patent regimes: (1) substantial revision of patent law in 1992 (this includes the extension of
coverage to pharmaceuticals and chemicals; and the explicit specification of standard duration
of protection); (2) joining the WIPO in 1995 (most important membership in international
agreements); and (3) WTO accession in 2001(conformity with TRIPs agreements).
Based on these points of major patent policy changes, the data was divided into three
sub-periods as follows: pre-WIPO period (1991-1995), pre-WTO period (1996-2000), and
post-WTO period (2000-2004). In our view, the most important period is the pre-WIPO
period as China’ s patent laws were considered to be comparable to the world standard only
5
According to World Investment Report 2005 (UNCTAD, 2005), multinational firms in developing countries are
less transnationalized than their counterparts in the developed countries. Our results here seem to support this
view. However, due to the constraint of data, further analysis of the effect of patents on FDI can not be carried
out at the this point.
18
after 1992. To assess how the trade effects change over different patent regimes, we estimate
equation (2) for the three sub-periods separately, both for the OECD and non-OECD countries.
Table 5 contains the results from the estimation of equation (2) using pooled annual
data for each of the three sub-periods reported separately for both OECD (developed) and
non-OECD (developing) countries.6 For developed countries, there is a clear pattern showing
that patents have a very significant effect on China’s imports in the pre-WIPO period (19911995) when the initial wave of patent laws were established. However, in subsequent years,
the effect of patents on imports notably declined. During the first period of 1991-1995, the
patent coefficients are not only more significant but also larger. For example, 16 out of 20
product categories have statistically significant patent coefficients with values close to or
larger than 1. This indicates that the market expansion effect was very strong during the
earlier years when China began to strengthen its patent laws. The weakening of the effect of
patents on imports from developed nations could be explained by the increasing role of FDI in
China in recent years. China’s further revision of its patent laws to conform with the TRIPs
agreement sent a signal to foreign firms of its willingness to further strengthen and enforce its
patent laws. Previous studies have shown that foreign firms tend to respond to stronger patent
rights by shifting from exporting commodities to establishing foreign affiliates and increasing
licensing and FDI sales (Park and Lippoldt, 2003). This is especially true in the case of China,
a large developing nation with high threat of imitation.
In contrast, for non-OECD (developing) countries, patents have a weak effect on
imports across the three IPR regimes.
At the aggregated level, while patents have a
significant effect on imports during the first time period, the estimates became much smaller
6
For brevity, only the estimates of patent elasticity of imports are reported.
19
and statistically insignificant in subsequent sub-periods. At the sectoral level, the results are
mixed. In most sectors, the effects of patents on imports are not significant over the three
periods. Relative to the results for China’s imports from OECD countries, increased patent
protection have a significant effect in much fewer sectors and no consistent pattern exists for
these non-OECD countries. It is only in the case of food imports where patents have a
significant effect in all three sub-periods.
In recent years, several large developing countries have benefited from globalization
and the accompanying internationalization of knowledge generation via offshore research and
development (R&D) by multinational corporations in developing countries (e.g., India,
Thailand, and Malaysia). The internationalization of R&D not only facilitates the transfer of
technology to the host countries, but could also generate spillovers which may stimulate
domestic innovation in host countries. Although past studies have shown that spillovers from
developed countries to developing countries are substantial (Coe and Helpman, 1995; Coe,
Helpman, and Hoffmaister, 1997; Keller 2002), empirical evidence is still sparse. The results
from current analysis suggest that China’s imports from developing countries are not
significantly patent-related.
5. Conclusion
Although several studies have investigated the effect of IPR protection on international
trade from the perspective of major industrialized nations, to our knowledge, no previous
analysis have examined this issue from the perspective of a developing nation with strong
threat of imitation. In this study, we examine the impact of patent rights protection on
China’s aggregate and sectoral imports.
Specifically, the analysis provides empirical
20
evidence for addressing these three questions of interests: (1) to what extent do the effects of
patent protection vary across various manufacturing product sectors? (2) does the impact of
patent protection differ between OECD (developed) and non-OECD (developing) countries?
(3) to what extent does the impact of patents vary across time as the strength of China’s IPR
protection increased? This analysis provides new insights on these empirical questions.
The empirical results suggest that exporters respond positively to the strengthening of
patent laws in China. This finding further confirms the results from earlier studies which
found evidence in support of the market expansion effect of stronger IPR protection in
countries with strong threat of imitation.
In general, our results show that the market
expansion effect is more pronounced in knowledge-intensive sectors than in non-knowledge
intensive sectors. Although the effects of patents on imports from OECD countries are
positive and statistically significant, the evidence for non-OECD countries is rather weak and
insignificant in many cases. These results are consistent with the fact that OECD countries
are the major producers of knowledge-based outputs and are the major exporters of patentsensitive products. Nevertheless, an examination of the dynamics of recent changes in patent
regimes in the past fourteen years indicates that the impact of patent rights protection on trade
appears to have weakened over time. This result may be reflecting the increasing role of FDI
by multinational corporations as the preferred alternative to international trade in commodities
(Smith, 2001). Recent data shows that China is a leading destination of FDI from
industrialized nations to developing and transition economies.
Overall, this study shows that IPR reforms in China tend to attract more knowledgeintensive imports and thus facilitates technology transfer to China. However, it should be
emphasized that while the implementation of stronger IPR regimes is necessary, it is not a
21
sufficient condition for trade expansion and economic growth. In addition to stronger IPR
protection, other complementary factors such as higher levels of R&D expenditures, quality
legal institutions, and improved physical infrastructure are also needed for narrowing the
technology gap between China and most developed countries.
Given the diversity in size and economic structure of developing economies, current
empirical evidence for China should not be automatically generalized to all developing
countries. Nevertheless, an important policy insight from this empirical analysis is that a
large developing nation with strong threat of imitation could reap increased bilateral trade
benefits from harmonizing and strengthening its IPR laws to conform with international
standards as China has done in recent years.
Future research would be beneficial in
unraveling the complex interactions between trade, FDI, licensing and IPR protection. As
more industry and firm level data become available, future analyses may be able to provide
more definitive conclusions on the trade effects of IPR protection.
22
References:
Allison, J. R., and Lin, L. (1999). “The Evolution of Chinese Attitude toward Property Rights
in Invention and Discovery.” SSRN Electronic Paper Collections.
(http://papers.ssrn.com/abstract=22539)
Anderson, J. E. (1979). “ A Theoretic Foundation for the Gravity Equation.” American
Economic Review. 69, P106-116.
Baltagi, B. (2001) . Econometric Analysis Panel Data. 2nd Ed. Chichester, England: John
Wiley & Sons.
Bergstrand, J. H. (1989). “The Gravity Equation, Monopolistic Competition and the Factor
Proportions Theory in International Trade.” Review of Economics and Statistics, 71,
P143-153.
Bergstrand, J. H. (1990). “The Heckscher-Ohlin-Samuelson model, the Linder Hypothesis and
the Determinants of Bilateral Intraindustry Trade.” Economic Journal. 100, P12161229.
Chin, J. C. and Grossman, G. M. (1988). “Intellectual Property Rights and North-South
Trade.” NBER Working Paper No.2769, Cambrige, MA.
Coe, D. T. and Helpman, E. (1995). “International R&D Spillovers.” European Economic
Review, 39, P859-887.
Coe, D. T., Helpman, E., and Hoffmaister, A.W. (1997). “ North-South R&D Spillovers.” The
Economic Journal, 107, P134-149.
Either, W. J. and Markusen, J.R.(1996). “Multinational Firms, Technology Diffusion and
Trade.” Journal of International Economics, 41, P1-28.
Fink, C. and Braga, C.A. P. (1999). “How Stronger Protection of Intellectual Property Rights
Affects International Trade Flows?” Policy Research Working Paper 2051, The World
Bank.
Greene, W. (2000). Econometric Analysis. 4th edition, Prentice Hall.
Ginarte, J. C., and Park, W. G. (1997). “ Determinants of Patent Right : a Cross-National
Study.” Research Policy. 26, P283-301.
Helpman, E. (1993) “Innovation, Imitation, and Intellectual Property Right” Econometrica,
61(6), P1247-1280.
Hausman , J. A. (1978). “Specification Tests in Econometrics”, Econometrica, 46, P12511271.
23
Hsiao, C. (2003). Analysis of Panel Data, Second Edition, Cambridge: Cambridge
University Press.
Keller, W. (2002). “Geographic Localization of International Technology Diffusion.” The
American Economic Review. 92(1), P120-142.
Lesser, W. (2001). “The Effect of TRIPs-Mandated Intellectual Property Rights on Economic
Activities in Developing Countries.” Prepared under WIPO Special Service
Agreements. WIPO.
Markusen, J.R. (2001). “Contract, Intellectual Property Rights, and Multinational Investment
in Developing Countries.” Journal of International Economics. 53, P189-204.
Maskus, K.E. (2000). Intellectual Property Righst in the Global Economy, Institute for
International Economics, Washington, D.C.
Maskus, K.E. (2004). “Intellectual Property Rights in the WTO Accession Package:
Assessing China’s Reform.” In Bhattasali, D., Li, S., and Martin, W. (eds). China and
the WTO. World Bank and Oxford.
Maskus, K.E. (2006). “ Assessing Coherence of the Intellectual Property Rights Regime in
China.” Indian Journal of Economics and Business, Special Issues China & India,
P175-187.
Maskus, K.E. , and Penubarti, M. (1995). “ How Trade-related are Intellectual Property
Rights?” Journal of International Economics. 39, P227-248.
⎯ (1997). “Patents and International Trade.” In Quiet Pioneering: Robert M. Stern and His
International Economics Legacy. (Eds.) Maskus, K.E. et al. Ann Arbor: University of
Michigan Press.
Park, W. G and Lippoldt, D. (2003).“The Impact of Trade-Related Intellectual Property
Rights on Trade and Foreign Direct Investment in Developing Countries” Working
Paper of the Trade Committee, OECD.
Qian, Y. (2007). “ Do National Patent Laws Stimulate Domestic Innovation in a Global
Patenting Environment? A Cross-County Analysis of Pharmaceutical Patent
Protection, 1978-2002.” Review of Economics and Statistics, 89(3) , P436-453.
Rafiquzzaman, M. (2002). “The Impact of Patent Rights on International Trade: Evidence
from Canada.” Canadian Journal of Economics. 35, P307-330.
Rapp, R. T., and Rozek, R. P. (1990). “ Benefits and Costs of Intellectual Property Protection
in Developing Countries.” Journal of World Trade. 24, P75-102.
24
Schneider, P.H. (2005). “International Trade, Economic Growth and Intellectual Property
Rights: A Panel Data Study of Developed and Developing Countries.” Journal of
Development Economics, 78(2), 529-47.
Smith, P.J. (1999). “Are Weak Patent Rights a Barrier to U.S. Exports?” Journal of Internal
Economics. 48, P151-77.
Smith, P.J. (2001). “ How Do Foreign Patent Rights Affect U.S. Exports, Affiliate Sales and
Licenses?” Journal of Internal Economics, 55, P441-439.
Tongzon, J. L. (2001). “China’s Membership in the WTO and the Exports of the Developing
Economies of East Asia: A Computable General Equilibrium Approach.” Applied
Economics, 33, P1843-1959.
UNCTAD (2005). World Investment Report 2005, United Nations.
25
12 0 0 0 0
10 8 0 0 0
Number of Invention Patents
96000
84000
Domestic filings
Total filings (domestic and foreign)
Foreign filings
72 0 0 0
60000
48000
36000
24000
12 0 0 0
0
1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004
Year
Figure 1. Patent Filings in China from 1985-2004
Source: Chinese patent database, CNPAT ABSDAT (http://www.sipo.gv.cn)
26
Table 1. Summary statistics of main variables
In Levels
Variables
Mean
Std.Dev.
Min
IMPORTS
430652.5 904085.9
0
(All products, in 100 millions 2000US$)
GDPIM
7.10E+11 1.60E+12 3.64E+10
(GDP of importing countries, in 100 millions 2000 US$
GDPC
1.02E+12 3.67E+11 4.86E+11
(GDP of China, in 100 millions 2000 US$)
TARIFF
1.03
0.0098
1.02
(Average tariff rate )
DIST
5015.6
2512.4
598.0
(Distance between China and Importing countries)
PATENT
894.6
2947.3
0
(Number of patents filed by importing countries)
27
Max
8994883
1.08E+13
1.72E+12
1.06
11957.0
33187
Table 2. Estimation results of Gravity equations, with all the countries included
Product Sectors
ALL PRODUCTS
APPARELS
BEVERAGE,TOBACCO
CHEMICALS
ELECTRONICS
FOOD
FURNITURES
GLASS
INSTRUMENTS
IRONS
LEATHER
MACHINERY
METALS
PAPER PRODUCTS
PETROLUM, COAL
PHARMACEUTICALS
PLASTICS
RUBBER
TEXTILES
TRANS. EQUIPMENT
WOOD PRODUCTS
GDPIM
(Importing)
0.843 ***
(6.010)
1.492
(3.910)
2.891
(4.630)
0.940
(3.350)
0.761
(2.420)
1.066
(3.030)
0.823
(1.970)
1.265
(3.370)
0.883
(2.980)
1.872
(5.120)
2.316
(4.130)
0.761
(2.610)
0.859
(2.890)
1.183
(2.590)
1.731
(2.780)
0.415
(1.010)
1.358
(4.470)
0.810
(1.790)
1.309
(4.410)
1.017
(2.380)
1.124
(1.800)
**
***
***
**
***
*
***
***
***
***
**
**
**
**
***
*
***
**
*
GDPC
(China)
1.525 ***
(8.870)
3.592
(5.030)
6.228
(6.460)
3.506
(4.500)
3.957
(5.000)
3.000
(3.970)
4.602
(6.220)
2.997
(3.440)
3.600
(5.050)
2.969
(3.450)
2.518
(3.220)
1.828
(2.300)
3.443
(4.410)
3.066
(4.130)
4.560
(4.540)
1.626
(1.730)
3.226
(4.230)
3.843
(5.460)
1.991
(2.720)
2.262
(2.790)
3.249
(3.490)
***
***
***
***
***
***
***
***
***
***
**
***
***
***
*
***
***
***
***
***
TARIFF
PATENT
DIST
CONST
-22.556 ***
(3.260)
0.080
(1.080)
-0.832 ***
(2.980)
-6.713
(0.200)
82.253 *
(1.830)
10.109
(0.290)
34.658
(0.980)
12.990
(0.380)
42.119
(1.230)
-9.955
(0.220)
11.341
(0.370)
-78.888 *
(1.870)
-23.199
(0.610)
-13.663
(0.390)
-11.412
(0.320)
1.265
(0.040)
-36.868
(0.720)
-7.186
(0.150)
6.268
(0.180)
-3.896
(0.120)
3.772
(0.110)
-36.446
(0.990)
20.880
0.500
0.168
(0.960)
-0.716
(2.190)
0.165
(1.150)
0.464
(2.720)
-0.151
(0.860)
0.600
(2.570)
0.275
(1.420)
0.580
(3.420)
-0.017
(0.100)
-0.293
(1.140)
0.574
(3.990)
0.142
(0.880)
0.083
(0.390)
-0.076
(0.230)
1.070
(3.750)
0.024
(0.160)
0.629
(2.610)
0.037
(0.240)
0.626
(2.630)
-0.113
(0.370)
-3.919
(7.060)
-1.583
(1.580)
-1.644
(3.220)
-2.893
(4.470)
-1.002
(1.570)
-2.437
(3.460)
-3.234
(4.920)
-2.170
(3.310)
-2.398
(4.170)
-1.818
(2.180)
-1.691
(3.050)
-1.673
(2.790)
-2.246
(3.000)
-4.555
(4.970)
-0.970
(1.220)
-2.428
(4.480)
-3.251
(4.790)
-2.346
(4.500)
-1.879
(2.510)
-3.255
(3.620)
**
***
**
***
***
***
***
***
***
***
***
***
***
***
***
**
**
***
***
***
***
***
***
**
***
R^2
-45.298 ***
(7.300)
-94.308
(4.010)
-223.314
(6.580)
-91.599
(3.910)
-90.879
(3.820)
-86.015
(3.770)
-120.874
(4.700)
-75.544
(2.790)
-91.949
(4.080)
-93.117
(3.410)
-99.206
(3.650)
-40.784
(1.690)
-87.731
(3.680)
-81.814
(3.300)
-118.767
(3.510)
-38.864
(1.280)
-88.624
(3.820)
-90.326
(3.480)
-54.529
(2.410)
-59.818
(2.140)
-79.239
(2.370)
0.625
***
0.663
***
0.377
***
0.306
***
0.475
***
0.184
***
0.525
***
0.495
***
0.524
***
0.420
***
0.335
*
0.402
***
0.281
***
0.371
***
0.375
0.458
***
0.412
***
0.571
**
0.382
**
0.431
**
0.318
*Significant at 10% level. ** Significant at 5% level. *** Significant at 1% level. t-statistics are computed based on the robust standard
error and reported in parentheses. Dependent variable is Log (IMPORTS).
28
Table 3. Estimation results of Gravity equations, with OECD (developed) countries
Product Sectors
ALL PRODUCTS
GDPIM
(Importing)
0.591 ***
(3.850)
GDPC
(China)
0.691 ***
3.760
TARIFF
-15.251 **
(2.090)
APPARELS
PATENT
0.432 ***
(4.160)
DIST
-0.484 **
(2.550)
CONST
-20.927 ***
(2.940)
R2
0.855
1.189 ***
1.270
-2.676
0.488
-1.776 **
-41.567
0.615
(2.810)
(1.460)
(0.070)
(1.020)
(2.540)
(1.590)
BEVERAGE,TOBACCO
4.072 ***
6.646 ***
73.904
-1.310 **
-0.782
-269.136 ***
0.500
(4.860)
(5.280)
(1.440)
(2.140)
(0.740)
(5.890)
CHEMICALS
0.744 ***
2.728 ***
16.623
0.355
-0.741 **
-73.755 ***
0.328
(3.050)
(3.220)
(0.390)
(1.260)
(2.230)
(3.090)
ELECTRONICS
0.454
1.904 **
31.145
0.809 **
-0.835 **
-44.912 *
0.406
(1.490)
(2.240)
(0.740)
(2.340)
(2.410)
(1.820)
FOOD
0.692 **
2.277 ***
2.912
0.256
0.234
-68.515 ***
0.200
(1.910)
(2.600)
(0.070)
(0.660)
(0.410)
(2.740)
FURNITURES
0.922 **
2.821 ***
18.435
0.745
-0.228
-92.700 ***
0.562
(2.310)
(3.500)
(0.550)
(1.590)
(0.330)
(3.820)
GLASS
1.430 ***
2.585 ***
-4.667
0.036
-1.082 *** -85.266 ***
0.477
(5.650)
(2.690)
(0.100)
(0.130)
(4.840)
(3.090)
INSTRUMENTS
0.374
2.092 ***
-2.528
0.926 ***
-0.346
-53.265 **
0.457
(1.490)
(2.700)
(0.070)
(3.080)
(1.060)
(2.410)
IRONS
1.206 ***
1.885 *
-32.860
0.599
-1.618 *** -57.302 **
0.515
(3.060)
(1.970)
(0.720)
(1.450)
(2.960)
(2.000)
LEATHER
1.005 *
1.562
-11.279
0.431
-1.080
-48.353
0.307
(1.780)
(1.470)
(0.270)
(0.740)
(1.060)
(1.480)
MACHINERY
0.535 *
0.798
-12.267
0.829 **
-0.344
-19.086
0.371
(1.840)
(0.910)
(0.280)
(2.490)
(0.920)
(0.750)
METALS
0.631 **
2.082 ***
5.588
0.683 **
-0.675 *
-56.193 **
0.432
(2.270)
(2.600)
(0.140)
(2.110)
(1.680)
(2.430)
PAPER PRODUCTS
0.391
1.183
23.010
0.832 *
-0.497
-28.247
0.354
(1.060)
(1.270)
(0.590)
(1.730)
(0.700)
(1.070)
PETROLUM, COAL
1.474 **
2.724 **
24.455
0.500
-1.259
-94.121 **
0.400
(2.450)
(1.760)
(0.390)
(0.680)
(1.390)
(2.030)
PHARMACEUTICALS
0.508 *
1.981 **
31.343
0.584 *
0.163
-58.466 **
0.286
(1.620)
(1.990)
(0.680)
(1.620)
(0.440)
(2.080)
PLASTICS
0.908 ***
1.793 **
4.960
0.574 **
-1.243 *** -50.349 **
0.502
(4.200)
(2.110)
(0.120)
(2.110)
(3.220)
(2.090)
RUBBER
0.767 ***
2.332 **
11.523
0.797 ***
-0.823 *
-69.109 ***
0.636
(3.240)
(3.320)
(0.360)
(2.600)
(1.860)
(3.400)
TEXTILES
0.909 ***
1.233
6.642
0.541
-1.650 *** -32.002
0.475
(2.720)
(1.450)
(0.170)
(1.510)
(3.050)
(1.280)
TRANS. EQUIPMENT
1.000 ***
0.627
-19.911
0.885 **
-0.217
-30.266
0.498
(2.900)
(0.650)
(0.480)
(2.030)
(0.460)
(1.060)
WOOD PRODUCTS
0.755 *
0.658
19.007
0.543
-0.594
-23.343
0.402
(1.950)
(0.650)
(0.410)
(1.330)
(0.770)
(0.780)
*Significant at 10% level. ** Significant at 5% level. *** Significant at 1% level. t-statistics are computed based on the robust standard
error and reported in the parentheses. Dependent variable is Log (IMPORTS).
29
Table 4. Estimation results of Gravity equations, with non-OECD (developing) countries
Product Sectors
ALL PRODUCTS
GDPIM
(Importing)
0.941 ***
(2.770)
GDPC
(China)
2.181 ***
(7.080)
TARIFF
-31.417 ***
(2.840)
APPARELS
PATENT
-0.023
(0.200)
DIST
-0.999 **
(2.090)
CONST
-63.765 ***
(6.360)
R^2
0.323
1.836 ***
7.352 ***
-16.798
-0.476
-6.349 *** -185.436 ***
0.745
(2.860)
(5.230)
(0.290)
(1.490)
(9.280)
(4.310)
BEVERAGE,TOBACCO
1.247
7.486 ***
84.606
-0.964
-2.164
-210.674 ***
0.208
(1.040)
(3.520)
(0.950)
(1.390)
(1.180)
(3.440)
CHEMICALS
1.294 *
3.999 **
3.305
0.235
-2.587 *
-106.392 **
0.266
(1.730)
(2.130)
(0.050)
(0.390)
(1.980)
(2.020)
ELECTRONICS
0.947
6.591 ***
39.569
0.220
-4.988 *** -150.734 ***
0.566
(1.100)
(4.280)
(0.700)
(0.440)
(3.740)
(3.410)
FOOD
1.465 *
3.296 *
29.311
-0.239
-2.198
-94.569 *
0.230
(1.620)
(1.820)
(0.460)
(0.400)
(1.500)
(1.880)
FURNITURES
0.536
7.322 ***
72.829
0.340
-4.613 *** -170.931 ***
0.555
(0.590)
(4.430)
(1.080)
(0.700)
(4.400)
(3.200)
GLASS
2.322 ***
2.572
-12.722
0.843
-5.961 ***
-69.319
0.667
(3.390)
(1.400)
(0.150)
(1.490)
(5.700)
(1.340)
INSTRUMENTS
1.347 *
5.816 ***
29.374
0.148
-4.245 *** -147.707 ***
0.543
(1.600)
(4.130)
(0.570)
(0.310)
(3.100)
(3.640)
IRONS
3.172 ***
1.976
-132.628 *
0.495
-3.116 **
-92.190
0.428
(3.990)
(0.950)
(1.670)
(0.810)
(2.510)
(1.520)
LEATHER
4.056 ***
2.937 *
-36.336
-0.676
-3.046 *
-144.240 ***
0.433
(3.930)
(1.830)
(0.490)
(1.240)
(1.900)
(2.890)
MACHINERY
0.989
2.789
-14.057
0.538
-3.034 **
-61.841
0.344
(1.260)
(1.530)
(0.230)
(0.940)
(2.380)
(1.190)
METALS
0.271
4.799 **
-33.929
-0.037
-2.068
-105.531 **
0.185
(0.320)
(2.510)
(0.490)
(0.060)
(1.380)
(2.000)
PAPER PRODUCTS
1.347
5.271 ***
-28.497
-0.411
-3.790 *
-132.144 ***
0.464
(1.330)
(3.130)
(0.470)
(0.760)
(2.560)
(2.640)
PETROLUM, COAL
2.389 **
5.761 ***
-119.305
-0.099
-7.856 *** -138.413 **
0.506
(1.670)
(3.010)
(1.300)
(0.160)
(4.310)
(2.410)
PHARMACEUTICALS
1.675 **
1.665
-63.323
1.091 *
-3.162 **
-52.877
0.416
(1.930)
(0.780)
(0.640)
(1.790)
(2.090)
(0.850)
PLASTICS
1.724 **
3.592 **
14.961
0.284
-3.298 **
-101.400 *
0.377
(2.060)
(2.010)
(0.240)
(0.490)
(2.370)
(1.950)
RUBBER
1.351
5.280 ***
-23.751
0.694
-5.898 *** -121.307 **
0.626
(1.180)
(3.240)
(0.370)
(1.280)
(5.140)
(2.350)
TEXTILES
1.484 **
1.843
4.600
0.148
-2.706 **
-51.905
0.260
(2.170)
(1.090)
(0.070)
(0.270)
(2.160)
(1.050)
TRANS. EQUIPMENT
0.822
3.694 **
-56.283
0.815
-3.252 *
-82.514
0.362
(0.740)
(2.050)
(0.850)
(1.450)
(1.910)
(1.490)
WOOD PRODUCTS
0.230
7.955 ***
15.930
-1.200 *
-5.552 *** -165.106 **
0.403
(0.160)
(3.680)
(0.200)
(1.700)
(2.960)
(2.460)
*Significant at 10% level. ** Significant at 5% level. *** Significant at 1% level. t-statistics are computed based on the robust standard
error and reported in the parentheses. Dependent variable is Log (IMPORTS).
30
Table 5. Changing effects of patents on imports:
different patent regimes
Product Sector
1991-1995
ALL PRODUCTS
APPARELS
BEVERAGE,TOBACCO
CHEMICALS
ELECTRONICS
FOOD
FURNITURES
GLASS
INSTRUMENTS
IRONS
LEATHER
MACHINERY
METALS
PAPER PRODUCTS
PETROLUM, COAL
PHARMACEUTICALS
PLASTICS
RUBBER
TEXTILES
TRANS. EQUIPMENT
WOOD PRODUCTS
Observations
0.886
(1.750)
0.933
(1.710)
-1.051
(1.480)
1.901
(1.800)
1.781
(1.730)
1.281
(1.360)
1.062
(1.650)
1.139
(1.060)
1.527
(1.600)
1.650
(2.270)
1.263
(1.640)
1.930
(2.000)
1.444
(1.740)
1.036
(2.100)
0.725
(1.960)
1.399
(1.740)
1.409
(1.670)
1.415
(2.510)
1.814
(2.270)
1.332
(1.750)
0.059
(0.070)
105
Subsample:OECD Countries
1996-2000
2001-2004
*
*
*
*
*
*
**
*
**
*
**
*
*
*
**
**
*
0.102
(0.860)
0.377
(0.970)
0.078
(0.090)
0.531
(1.440)
0.910
(2.150)
0.162
(0.370)
1.176
(1.850)
0.057
(0.170)
0.968
(2.690)
0.521
(1.350)
0.085
(0.180)
0.085
(0.180)
0.654
(1.840)
0.705
(1.610)
-0.335
(0.390)
0.402
(1.020)
0.539
(1.610)
0.568
(1.770)
0.163
(0.440)
0.995
(2.420)
0.753
(1.330)
105
**
*
***
*
*
*
*
**
0.005
(0.070)
0.162
(0.710)
0.334
(0.600)
-0.241
(1.350)
0.157
(1.130)
-0.393 *
(1.800)
-0.037
(0.160)
-0.241
(1.350)
0.520 ***
(2.690)
-0.113
(0.970)
-0.349
(0.880)
0.070
(0.510)
-0.109
(1.280)
0.057
(0.370)
-0.058
(0.160)
0.267
(1.180)
0.142
(1.040)
0.107
(0.630)
-0.122
(1.180)
0.498
(1.360)
0.090
(0.510)
84
estimated patent coefficients of
1991-1995
0.749
(2.560)
-0.235
(0.440)
-0.303
(0.650)
0.698
(1.720)
0.315
(0.710)
0.598
(2.140)
0.580
(1.020)
-0.075
(0.140)
0.489
(1.160)
0.668
(1.170)
-0.277
(0.760)
0.210
(0.410)
0.306
(0.770)
-0.261
(0.440)
-0.579
(0.880)
0.665
(1.480)
0.218
(0.560)
0.486
(1.210)
0.131
(0.320)
1.057
(1.610)
-0.222
(0.570)
75
Subsample:non-OECD Countries
1996-2000
2001-2004
**
*
**
*
0.097
(1.260)
-0.293
(0.760)
-1.046
(2.120)
-0.238
(0.700)
0.168
(0.340)
-1.023
(2.300)
0.412
(0.640)
0.739
(1.610)
0.035
(0.120)
-0.156
(0.300)
0.024
(0.040)
-0.243
(0.710)
-0.380
(0.920)
-0.207
(0.590)
-1.023
(2.300)
0.031
(0.050)
-0.647
(1.590)
-0.001
(0.000)
-0.219
(0.730)
0.333
(0.620)
-0.502
(0.720)
75
**
**
*
**
0.067
(1.050)
0.095
(0.440)
-0.082
(0.130)
0.086
(0.870)
0.295
(1.210)
-0.325
(1.980)
0.299
(0.780)
0.619
(2.250)
0.046
(0.180)
0.914
(2.320)
-0.591
(1.060)
0.179
(1.140)
0.027
(0.150)
0.158
(0.680)
-0.219
(0.410)
1.216
(1.550)
0.216
(1.680)
-0.005
(0.010)
0.023
(0.180)
0.338
(0.880)
-0.255
(0.810)
60
*Significant at 10% level. ** Significant at 5% level. *** Significant at 1% level. t-statistics are computed based on the
robust standard error and reported in the parentheses. Dependent variable is Log (IMPORTS). Other variables are all in logs.
*
**
**
*
Fly UP