Paper to be presented at the DRUID Summer Conference 2004 on
INDUSTRIAL DYNAMICS, INNOVATION AND DEVELOPMENT
Elsinore, Denmark, June 14-16, 2004
Theme
TRADEMARKS AS AN INDICATOR OF INNOVATION
AND INDUSTRIAL CHANGE
Sandro Mendonça
ISCTE University and Dinâmia
Dept. of Economics
Av. das Forças Armadas, 1649-026
Lisbon, Portugal
Tiago Santos Pereira
CES - Centro de Estudos Sociais
Universidade de Coimbra
Colégio de S. Jerónimo
Apartado 3087, 3001-401
Coimbra, Portugal
tsp@ces.uc.pt
Manuel Mira Godinho
ISEG/UTL and CISEP
Dept. of Economics
Rua Miguel Lupi 20, 1200-725
Lisboa, Portugal
1st draft: 13 November 2003, This version: May 2004
Abstract: As innovation becomes an ever more central issue for the development of human societies
and private firms so the need for improved assessments of innovative performance grows more
urgent. In order to reflect the innovation endeavour of countries and companies several internationally
comparable indicators have been created and developed since the 1950s. This paper suggests
trademark analysis can yield reliable results as a method for capturing relevant aspects of innovation
phenomena. We propose trademarks as a complementary indicator in the portfolio of empirical tools of
innovation studies. Our empirical exploration of trademarks is based on a study of Community Trade
Marks and draws on recent research on national trademarks of Portuguese firms. Quantitative as well
as qualitative data, including survey information from a representative sample of Portuguese
manufacturing and service firms, are used to identify the advantages and limitations of this new
research tool.
Keywords: trademarks; IPR; innovation; industrial dynamics
JEL - code(s).
Trademarks as an Indicator of Innovation
and Industrial Change
Sandro Mendonça
ISCTE University and Dinâmia
Dept. of Economics
Av. das Forças Armadas, 1649-026
Lisbon, Portugal
Tiago Santos Pereira
CES - Centro de Estudos Sociais
Universidade de Coimbra
Colégio de S. Jerónimo
Apartado 3087, 3001-401
Coimbra, Portugal
tsp@ces.uc.pt
Manuel Mira Godinho
ISEG/UTL and CISEP
Dept. of Economics
Rua Miguel Lupi 20, 1200-725
Lisboa, Portugal
1st draft: 13 November 2003
This version: May 2004
Abstract: As innovation becomes an ever more central issue for the development of
human societies and private firms so the need for improved assessments of innovative
performance grows more urgent. In order to reflect the innovation endeavour of
countries and companies several internationally comparable indicators have been
created and developed since the 1950s. This paper suggests trademark analysis can
yield reliable results as a method for capturing relevant aspects of innovation
phenomena. We propose trademarks as a complementary indicator in the portfolio of
empirical tools of innovation studies. Our empirical exploration of trademarks is
based on a study of Community Trade Marks and draws on recent research on
national trademarks of Portuguese firms. Quantitative as well as qualitative data,
including survey information from a representative sample of Portuguese
manufacturing and service firms, are used to identify the advantages and limitations
of this new research tool.
Keywords: trademarks; IPR; innovation; industrial dynamics
Trademarks as an Indicator of Innovation and Industrial Change1
1. Introduction
The business of branding products has been part of ordinary economic life for a long
time. Trademarks are the outcome of establishing recognisable designations and
symbols for material goods, services as well as firms’ identities. They are a crucial
part in the process of marketing an innovation, helping to differentiate the attributes
and content of goods and services in the marketplace. These characteristics make
trademarks a potential new indicator of product innovation and sectoral change.
Recent developments in the institutions of international filing and registration of
marks as well as the increasing availability of digital databases have increased the
potential use of trademark statistics as a new source of information for industrial and
innovation studies.
Trademarks are of interest for social science research for at least three reasons: they
confer an exclusive right of using a brand, therefore enhancing companies’ ability to
appropriate the economic returns of investing in new products or improving existing
ones; they are an important aspect of contemporary culture world-wide; and they are a
source of qualitative and quantitative information on socio-economic activity. This
paper focuses on the third of these features. It assesses the possibilities and problems
of using trademark data when analysing the introduction of new or improved
manufacturing and services’s products in competitive markets. Along with a
methodological reflection, the paper offers a concrete empirical application of the
indicator to EU-15 data on the basis of insights obtained through an in-depth study of
an intermediate European economy, Portugal, which we compiled national trademark
data for the period after 1980, survey data from a representative sample of 724 firms
collected during 2003 and information obtained from thematic workshops held with
entrepreneurs, managers and consultants (Godinho et al., 2003).
We argue that trademark-based indicators provide a partial measure of the innovative
output of profit-oriented organisations. In its most simple formulation, innovation can
be understood as the introduction in the market of a new idea, product or production
process. As an intellectual property right (IPR), trademarks are designed to identify
certain products as those provided by a person or a firm, therefore assisting
customers’ decisions to purchase something if they recognise its nature and the
quality meets their needs. The filing of new trademarks by economic actors partially
reflects the introduction of new offerings aimed at persuading potential buyers that the
range of their problems was not solved by the supply of solutions previously available
1
The authors acknowledge the National Patent and Trademark Office (INPI) for permission to use data
generated in a project on the use of industrial property rights by Portuguese individuals, firms and
public research organisations. One way or another the three of us were influenced by Keith Pavitt’s
sharp and provocative views as a Professor and a colleague. It is clear in our mind that the current work
would not have been the same without the experience of having known him. We wish to thank
participants of the Pavitt conference for their discussions, and especially to Scott Stern, our panel
discussant. Any faults in this paper are entirely our responsibility.
2
in the market. In this way, and since companies have to pay fees to register and renew
their rights in national and international offices, the compromise to register a new
brand or logo reveals an economic decision and an entrepreneurial initiative that are
worth investigating. Given the growing demand from governments, firms and
universities for more reliable information on innovation there is a case to test
trademarks as a complementary indicator to the more traditional measures of
technological activity, namely R&D investment and patents. Furthermore, trademarks
have an additional advantage in relation to patents because they are used by a wider
set of business firms, and are less dependent on the technological activity of firms,
capturing changes in service activities as well as small and medium sized firms
(SMEs).
Trademarks show promise in analysing (i) rates and directions of product innovations
in different industrial sectors, (ii) international patterns of sectoral specialisation, (iii)
links between technological activity and marketing advantages; (iv) evolution of
economic organisations and structures, not only in terms of entry and exit of firms,
but also in terms of firm growth, differentiation, and diversification. Evidence from
the Portuguese case on the first two points suggests that behaviour in filing
trademarks is largely as expected by tracking the innovative performance of firms,
industries and countries over time. However, simple counts of trademarks are affected
by various sources of bias, such as difficulties in data consolidation (e.g. one brand
can be protected simultaneously by a combination of words, symbols and 3D design),
sectoral differentiation (the classification follows the characteristics of the product
and not of the industrial sector and one trademark can be registered under several
classes; cf. Appendix) and weaknesses in international comparability (given, for
example, the different market niches targeted by firms of different countries). All of
these issues cannot be completely explored or solved in a single contribution. The full
assessment of trademarks as indicators of innovative activity and industrial
competitiveness requires further exploratory research, survey and econometric
research, as well as in-depth case studies.
The paper is organised as follows. The first section discusses what a trademark is both
from the IPR and managerial perspectives. The second tackles the conceptual and
analytical issues that arise when we consider trademark statistics as an innovation
indicator. This is followed by a presentation of the key patterns found on Community
Trade Marks for 15 EU countries and a review of a detailed study of the Portuguese
case. The fifth section critically reviews the findings and points out avenues for
further research. The final section concludes.
2. Brands and logos in business life
People show a tendency to label other beings and things surrounding them, as well as
to exert property over them. It is thought that the first marks served to indicate the
legitimate proprietor of livestock in Neolithic times, 7 000 years ago.2 Marks later
evolved to designate the author of an object and to establish her obligation in assuring
the quality of the product. The first documented example of this economic use of
2
http://www.lib.utexas.edu/engin/trademark/timeline/tmindex.html - website maintained by the Library
of the University of Texas at Austin which compiles and synthesises information on the history of
trademark from a wide collection of previous works on the topic.
3
trademarks is found in the Roman Empire. Roman bricks had the stamp of the
manufacturer, the date and the place of production. During the Middle Ages
corporations of craftsmen started to identify their workmanship with a mark. In the
absence of modern means of advertising the reputation of trade guilds that function
was carried through marks inscribed in the products. A century following the dawn of
the industrial era a series of explicit trademark laws were enacted in France (1857),
UK (1862) and the US (1870). In this field, as well as in others, framing institutions
co-evolved with the actors’ behaviours and goals. The remainder of this section brings
in the key institutional features of trademarks as a part of the IPR system and briefly
examines the economic rationale and the historical practise of their use by business
firms.3
The IPR institution
According to the World Intellectual Property Organization (WIPO) a trademark is
defined as a “distinctive sign used to distinguish the goods or services of an enterprise
from those of another” (WIPO, 2003). The two objectives of protection and
dissemination built into this definition are practically indistinguishable. Like patents a
trademark provides legal protection to the owner by granting the exclusive right to use
it to identify goods or services, or to licence its use to another in return for payment.
Rights are granted at the national level but, unlike patents and copyrights, once
registered marks can be renewed indefinitely in time on payment of additional fees.
The common expectation in trademark regimes is that a registered mark is used; an
unused trademark is implicitly regarded as a barrier to economic initiative and can be
applied for by another company after a grace period.
Today, trademarks may consist of one or a combination of words, slogans, letters, and
numerals; they may also be drawings, symbols, three-dimensional signs such as the
shape and packaging of goods, audible signs such as musical or vocal sounds, or
distinguishing fragrances, colours and holograms. Therefore, when compared to
patents, they might seem like a lower order of intellectual creation. However, as stated
by Doern (1998, p. 72), who carried out over 70 interviews with officials of seven
agencies, trademarks generate the second highest volume of IPR revenue after
patents. Notwithstanding the traditional neglect, trademarks constitute a crucial part of
innovation and commercial processes helping to attract scarce attention from the
public and differentiate the nature and quality of products.
As a creation of the mind, a trademark does not reveal technological inventiveness but
rather marketing creativity. Patents are granted to inventions on the basis of nonobviousness, inventiveness in the face of prior art and the potential for industrial
application. A commercial sign, on the other hand, may also be denied registration but
only if judged deceptive to consumers (e.g. if it can be confused with other marks, if it
contains a misleading description of the character or quality of the goods or services,
etc.), if it is deemed contrary to morality or if it denotes symbols reserved to the state
or public organisations.
Trademarks are an IPR issued by an authorised national government agency following
an examination process that is dependent on the criteria just alluded to and on a mix
3
For a historic accounts of trademarks see Diamond (1983) and Wilkins (1992).
4
of limited human and technical resources administrated by a patent and trademark
office. Once an application has been filed, examiners search available databases to
detect any other marks in use that can come into conflict with the applicant’s. A
successfully registered trademark is recognisable by having one of these two symbols
attached: “” or “”. The lag between the filing and formal registration is much
shorter than that for patents. For instance, while it takes up to a year to register a
Community Trade Mark (CTM) for a mark for which there is no grounds of refusal, it
can take over five years to obtain a patent from the European Patent Office.
The first international trademark agreement was reached at the Paris Convention of
1883 whereby countries involved agreed to provide the same protections regarding
marks to foreign applicants as those provided to nationals. In this context, WIPO
eventually emerged as the global coordinating institution promoting the development
of IPR laws and facilitating the international registration of trademarks. This role
emanates from the 1891 “Madrid Agreement Concerning the International
Registration of Trademarks” that opened up the scope for protection of marks beyond
their market of origin. Another important international development was the
establishment of the CTM in Europe, which came into being with the establishment of
the Office for Harmonisation in the Internal Market (OHIM) in 1994, a EU institution.
The sources for trademark information have to be found elsewhere than the OECD
and national offices of Science & Technology. Amongst the national sources, the
Patent and Trademark Offices are the public agencies traditionally entrusted with
registering and keeping the records of trademarks, but unfortunately not necessarily
with easy access to the data or easily usable for statistical treatment. The major public
international agencies responsible for trademarks are WIPO and OHIM. OHIM
publishes monthly and annual surveys of trademark activity, broken down by country
of applicant, year of filing and registration, trademark classes of goods and services
and trademark types (word mark, figurative mark, three-dimensional mark, etc) since
1996. Finally, amongst private sources there are an increasing variety of firms
compiling and selling databases on trademarks, mostly for consultancy purposes.
Brands and business strategy
Brands are commercial signatures that firms create, maintain, protect and reinforce for
a number of different marketing objectives. Brand names have multiple origins. Some
are linked to the names of the company founders (Ford, Nestlé), others show a
connection to a particular line of business (Microsoft, Airbus), other still are pure
neologisms (Kodak, Xerox). Brands become genuine cultural references especially
since the late XIX century with the rise of consumer product industries oriented
toward mass markets in western societies. Brands such as Coca-cola, Campbell’s
Soup or McDonalds are liked (or disliked) by many but surely they mean something
to everybody. Many brands became synonym in many languages of iconic
innovations of the XXI century (Hoover, Gillette, Black & Decker). Many others
established themselves as true national symbols (GM, Ferrari, Nokia) or even came to
signify the achievement of transnational partnerships (Airbus).
For better or for worse brands are with us, and their influence on our lives is not
bound to diminish in the near future. Judging from newspaper interviews to CEO’s
and popular business literature, branding is becoming an ever more central concern in
5
corporate strategy. Evidence from everyday life tells us that hardly anything goes
unbranded. Even fruits are branded – Chiquita bananas. But, where does all that
interest in brands come from? Surely brands are obviously of interest to buyers.
However, the ways in which such awareness and loyalty commanded by brands
translates into a competitive asset are not completely clear.
The rationale and actual influence of trademarks is a relatively neglected aspect in the
study of the development of modern corporations (Wilkins, 1992). For one,
economics has notoriously little to say about the activity of branding. From the point
of view of many economists prices and quantities are still the most important
economic variables. The complex deliberation process concerning other attributes,
such as quality, features, reputation, support services, and so forth, is still scarce dealt
with in economic theory, not to mention applied research (Trajtenberg, 1990, p. 8).
Standard industrial organisation textbooks tend to cluster issues such as product
differentiation and advertising in the same chapter and focus on the relationship
between market structure and the advertising-to-sales ratio. Brand decisions are
implicitly considered part of advertisement expenditures. Luís Cabral, an editor of the
Journal of Industrial Economics and author of a popular textbook on the field, goes
on to make an interesting declaration on this subject. Under the heading “Price and
non price strategies” of a teaching note available in the companion site of his book, he
states that: “(a)dvertising is not one of the core topics of IO ... this chapter may
therefore be omitted from a course directed primarily at economics majors”, and he
goes on to admit that “(n)ot much economics research has been done on this topic. As
a consequence, most of the points presented are rather tentative: more questions are
raised than answered.” (Cabral, 2000).
The only paper explicitly on the “economics of trademarks” we found (an interesting
but rather under-cited paper published in a technical journal) gives us the usual cost
and benefit perspective (Economides, 1987). The author points out that the key reason
why brands are indispensable for the efficient provision of products is because they
are a device which facilitates consumer’s choice in the wide rage of variety and
quality combinations available in a modern economy. Nevertheless, the conventional
reasoning goes on, monopoly power embedded in the IPR is bound to cause
inefficiencies and allocative distortions. Part of such anti-competitive effects consists
of barriers to entry to new competitors. The treatment of this dynamic competition
issue is very hard to grasp using the traditional analytical framework of mainstream
economics.
In practical terms, a trademark should protect a brand when its expected value
(probability of being accepted times the revenue associated with the brand) exceeds
the cost of applying for it (plus the present value of the fees needed to maintain the
mark). In the management literature the question “to brand or not to brand” is treated
by marketing research. In terms of the classic marketing framework, brands are
included in the Product variable of the marketing mix. The “marketing mix” is the
conventional view of the set of tools that a firm blends to influence the demand for its
product – it comprises the well-known “four Ps”: Product, Pricing, Promotion and
Placement (Kottler et all, 1999). Contrary to what one would expect, brands are not
seen as a tactical tool for promotion management, which instead refers to
communications activities including advertisement, sales-people approach to
customers, special promotions, public relations, etc. This is evidence that marketeers
6
regard brands as a key characteristic that helps delivering the core benefits of the
product to the consumers. The implication of this understanding is that brands do
more that just transmitting persuasive signals (horizontal differentiation), they also
have a deeper role in transforming customers’ experience of using the product
(vertical differentiation).
Customer benefits such as trust and satisfaction have a translation into the strategic
motivations that underlie trademark management. The issue of whether or not to go
ahead with branding a product depends on a complex list of potential motivations that
have to be balanced with the cost of the branding strategy – see Box 1. In deciding to
create a brand, and whether or not protecting it legally through a registered trademark,
a company is expected to define its overall brand strategy. According to the marketing
literature there is a proliferation of strategic intents when a company introduces
additional items in existing products under the same brand, such as new flavours,
ingredients, colours, packages, etc., which eventually evolve to become sub-brands
(e.g. Diet Coke, Pepsi Blue). Another common strategy, called brand-stretching or
brand extension, consists of using a successful brand image to launch new or modified
products into new categories (a classic case being the way Honda penetrated the US
car market having achieved a great reputation for motorcycle engineering). Since
there are no established taxonomies we give our own, incomplete, list of motives and
strategies behind the uses of brands.
Box 1. Strategic motivations behind the creation of a brand
•
•
•
•
•
•
•
•
•
Building inelasticity around the product and achieving a premium pricing (differentiation, line
extensions)
Opening up opportunities to enter new product segments or entirely new lines of business (brandstretching)
Penetrating new geographical markets (geographical market diversification)
Improving the conditions for appropriating the returns on innovation whenever other means are
not effective
Entering the market for trademarks (licensing)
Saving on promotion expenditures (building loyalty)
Achieving greater bargaining power against suppliers (supply chain coordination)
Signalling changes in strategy or changes in corporate identity (internal and external marketing)
Extending the protection conferred by other IPRs after their expiration date (namely patents)
Naturally, as in the case of patents, trademarks are not all identical in their value. The
value of patents is equated to the net money equivalent stemming from the
exploration of the rights associated to a particular idea or innovation (Grilliches, 1990,
p. 1690). The value of a patent is generally considered proportional to be the
magnitude of the innovation protected (Trajtenberg, 1990, pg. 5). In management
literature, what counts to determine trademark equity is a set of characteristics such as
name awareness, customer loyalty, perceived quality and associations to the brand
that add value to the product being offered (Aaker, 1991). Trademark or brand equity
influences the market value of a firm, namely of the large corporations acting on the
global market. Interbrand, a specialised consultancy firm regularly releases reports
where it estimates which are the world’s most important brands – see Figure 1.
Summing up, in seeking and maintaining trademark rights, companies make important
economic decisions. Companies decide to create or enhance a protected brand to
7
distinguish themselves from other suppliers and to rise above the competition.
Therefore, something deeper than horizontal differentiation is at stake. By monitoring
what happens to trademarks we can learn a great deal about the entrepreneurial
dynamics of given firms or aggregates of firms as well as gain a lever for making
inferences about their marketing capabilities. This is the reason why trademark
statistics are interesting in spite of all the difficulties involved in their use and
interpretation.
Table 1. The world’s most valuable brands
1990
1992
1996
1997
1999
2000
1
Coca-cola
Marlboro
McDonald’s
Coca-cola
Coca-cola
Coca-cola
2
Kellogg’s
Coca-Cola
Coca-cola
Marlboro
Microsoft
Microsoft
3
McDonald’s
Intel
Disney
IBM
IBM
IBM
4
Kodak
Kellog
Kodak
McDonald’s
GE
GE
5
Marlboro
Nescafé
Sony
Disney
Ford
Nokia
6
IBM
Budweiser
Gillette
Sony
Disney
Intel
7
Amex
Pepsi-Cola
Mercedes
Kodak
Intel
Disney
8
Sony
Gillete
Levi’s
Intel
McDonald’s
Ford
Sources: www.interbrand.com, Kotler et al (1999), Granstrand (1998)
3. Towards a trademark indicator
This section will argue that trademarks constitute a unique and under-used source for
the analysis of the process of innovation and industrial evolution. Trademarks rival
with patents in terms of accessibility and quantity of available data. True, they do not
exhibit the same detail of information; patents show the description of a technological
invention, the individual inventors that contributed to it and the citations made of
prior art. It is also not fully clear what trademarks really indicate. However, the
promise of adding a new indicator of innovative economic activity to the limited
portfolio of output innovation indicators is a strong motivation for researching
trademark data. If that promise is fulfilled, the potential applications to fields such as
innovation studies, industrial dynamics, business history and international trade can
be very rewarding. Of course, implications for public policy and management strategy
should follow.
Indicators in innovation studies
Indicators are means to obtain quantitative information about certain aspects of a
phenomenon. There are, of course, no perfect or complete indicators of innovation as
there are no perfect indicators of other socio-economic phenomena. The particular
problem with the innovation phenomenon is its multidimensional nature involving
qualitative change of economic, strategic, organisational and institutional factors.
8
Thus, in order to operationalise a trademark indicator it is important to get to terms
with the ontological requisites for achieving reliable knowledge.
Innovation indicators are expected to convey behavioural information on social
entities. They detect and register “levels” and “dynamics” of individuals, companies,
institutions and countries. The imperfect nature of the figures is always present; these
are ‘indications’. Indicators capture, but only partially so, some aspects of the object
at stake. One needs, therefore, to be aware of the intrinsic limitations of an indicator;
it is not an objective, complete measurement. Indicators are institutionally created and
maintained. They are socially constructed, growing more out of practice e rather than
out of theory. There is nothing sacred about them. For instance, there are no a priori
reasons to think that registered trademarks on brands do not involve a process as
judgemental and non-technical as patents on inventions. They often turn out to be
given other uses from those originally intended (Patel and Pavitt, 1994).
Numbers can be as misleading as words. As Smith (2004) stresses, indicators are not
simple numbers. They tend to come with strings attached, they imply associations to
given theories or views of the world which shape the way quantitative information is
produced and/or interpreted. In the case of trademarks, even though we take the view
of innovation as an evolving process of interactions and feedbacks, this indicator
provides information on one specific event (filing) with no information on such
interactions, inputs or differential impacts. This invites us to follow the
recommendation made by Martin and Irvine (1983) and take trademarks as a partial
indicator of performance that should be combined with other information in order to
produce reliable results.
The study of the sources and patterns of technological change has progressed
enormously since the mid-1960s when the first internationally comparable statistics
on R&D activities were published by the OECE on the basis of work by C. Freeman
and A. Young. Despite having such a long history S&T indicators remain contested
even today, one paramount reason being the emphasis on inputs rather than on outputs
and impacts (Godin, 2003). In order to realise the potential of trademarks we need to
identify the basic procedures that should be involved in counting and comparing.
Only with scepticism and methodological care can trademarks be used in certain cases
as an indirect measures of innovation and a tool to assess structural changes in the
economy.
Trademarks as an indicator
To our knowledge the first explicit reference to trademarks as an indicator of
innovation is a report on the 2001 Germany’s Technological Performance, a report
written on behalf of the Federal Ministry of Education and Research by a group of 8
German research centres (Velling, 2002). The team of authors argue that trademarks
can no longer be considered a subordinate industrial right. The fact that the number of
trademark registrations filed in Germany tripled during the 1990s area a “clear
indication that trademarks are being assigned considerably more importance than in
the past” (Velling, 2002, p. 20). Although novelty is not a requirement for registering
a trademark, the authors say one can safely assume that trademarks are filed primarily
for new products and services. They also point out that commercial enterprises
constitute a large number of applicants. Another positive aspect emphasised in this
9
pioneering contribution relates to the degree the indicator keeps pace with the market;
in Germany it takes just six months for a trademark to be entered into databases after
application. However, the trademark section of the report is rather short (4
consolidated pages in a total content of 174 pages) and lets many questions both
unasked and unanswered. For example, why do the authors prefer to count trademarks
applications and not registrations? Our answer would be that due to the lack of
technical content in trademarks, applications reflect much better the trademarking
intentions, and the revealing underlying motives of protecting innovation and
marketing capabilities of companies.
In a recent study, Schmoch (2003) has concluded for the suitability of service marks
as an indicator of innovation in this sector “as they meet essential preconditions, in
particular correlation to innovation, good data access by electronic databases, and the
possibility of operationalising them in relevant dimensions of desegregation” (p. 155).
We go in the same direction of this recent work, using EU data to assess both
manufacturing and service products trademarks. In comparison with some widely
used innovation indicators, trademarks appear to be complementary. Trademarks have
a comparative advantage in capturing small firms due to the relatively small fees
involved in filing and maintaining marks. They are an additional output indicator, to
the existing limited number of output indicators. In this context, when compared to
patents, they are closer to commercialisation and cover a broad range of activities
from manufacturing product classes to service classes.
10
Table 2. Different approaches and measures of innovative activities
Indicator
Type
R&D
expenditure
input
Technical
personnel
input
Patents
output
Advantages
Limitations
Level of
analysis
Sources
availability
lack of detail
country
OECD
economic variable
industry
captures different
formal competencies
overestimates large
electrical and chemical
firms
qualifications are not
homogeneous
firm
country
national
statistical
offices
OECD
complement to R&D
statistics
does not capture
informal competencies
industry
extremely detailed
uneven propensities to
patent
regular and very
long-run
Trademarks output captures small firms
covers both
manufacturing and
service industries
S&T field
firm
country
S&T field
underestimates small
firms, design,
mechanical, software
and services activities
industry
Nice classes are
considerably aggregate
and heterogeneous
country
captures innovations
coming out of
universities and public
research organisations
product
class
firm
firm
national
statistical
offices
OECD
national patent
and trademark
offices
European
Patent Office
OHIM
WIPO
national patent
and trademark
offices
From trademark data to information on innovation and industrial phenomena
Given the experience with using patents as proxy measures of technological activities
(Pavitt, 1985; Griliches, 1990) we take advantage of this accumulated knowledge and
techniques to explore the potential of trademark data.
There are strong reasons to believe that a correlation between innovation and
trademark use exists. In his study Schmock (2003) finds a highly significant
correlation between innovation and marks, namely in the manufacturing sector.
Focusing on service marks he found considerable differences between sectors, namely
regarding the trade sector, technology-oriented services, or knowledge-intensive
services, with the indicator being particularly relevant in the latter case. In a study on
the use of Portuguese IPRs by residents Godinho et al. (2003, p. 154) also found sharp
and highly statistically significant differences across manufacturing sectors according
to their technological intensity. Trademark use is not randomly distributed across
sectors at the 1% significance level, with high and medium-high technological
intensive sectors being heavy users of trademarks. In what concerns services, the class
of information intensive service sectors was also found to be associated to a higher
pattern of use of trademarks at the 5% significance level.
As with patents, trademark statistics have the strength of reasonably unambiguous
legal definition, being collected and classified by (the same) specialised institutions,
in accordance to international agreements, and long time series are available. The
11
basic classification follows from the 1957 Nice Agreement Concerning the
International Classification of Goods and Services for the Purposes of the
Registration of Marks. The Nice Classification system distinguishes between goods
and services. It is regularly revised and is now on its 8th edition, which is in force
since January 1, 2002, has 34 classes of manufactured goods and 11 classes of
services (three new service classes were added in the last edition). One difficulty,
however, Nice classes do not have a direct connection sectoral nomenclatures such as
the NACE (Statistical Classification of Economic Activities in the European
Community).
For the purpose of the analysis of innovation and industrial dynamics, the greatest
limitation of the classification is that the different classes are much too aggregate. For
example, Class 5 covers the product categories of pharmaceutical, medical and
veterinary preparations, mixed with dental wax, disinfectants, fungicides, herbicides,
and even food for babies.
Another characteristic of trademark data is that a given trademark for a word or
symbol can be asked just for one or for several or even all Nice classes. This means
that the number of counts in the total classes will be much larger than the total number
of trademarks filed, even if it is possible to identify such multiple classifications. This
is a limitation for cross-sectoral analysis, and is unlike other output indicators.
Furthermore trademark applications are not classified according to the main product
line or productive sector of the applicant company.
A given product or supplier can also be protected by more than one trademark. For
instance, the Coca-Cola beverage is protected by the word mark “Coca-Cola”, by its
distinguishing logo (the stylised letters composing the word underlined by a ribbon)
and by a three-dimensional mark protecting the distinctive shape of the bottle. Intel
Corporation not only protected its Intel name with a word mark and a corresponding
logo, it filed a sound mark as well (see Box 2).
These considerations mean that there is no clear-cut correspondence between a new
product and a new trademark. Raw data can lead us to over evaluate the patterns
observed. However, the multiplication of trademarks in certain product categories
Box 2. Examples of overlapping protection using trademarks and other IPRs
According to the Coca-Cola corporation its flagship brand is the most recognised trademark in the
world, with 94% of the world’s population recognising it, and it is the most widely recognised word
after “OK”. The trademark for the word “Coca-Cola” was obtained in 1887. The type of print used
for the logo is called “Spencerian Script”. In an effort to safeguard the bottle contour it was
registered in 1977 as a three dimensional mark. The argument was that the packaging distinguished
the beverage from others.
In another example, Intel has also
registered its well known jingle. The
musical notation describing the registered
Intel’s jingle is pictured on the side.
The relationship between the trademarks
and other IPRs, namely industrial design
and copyright, is evolving. Another
aspect of these times of change is
institutional tension created by the
Sources: www.coca-cola.pt, www.dww.com/articles/protect.htm
12
surely constitutes evidence of increased competition, whether horizontal and/or
vertical.
One further limitation is the anecdotal observation that there are many unregistered
brands in use in the marketplace. This is similar to the problems found in patents: not
all inventions can be patented and not all patentable inventions are patented. In the
case of trademarks, brands take the place of inventions. Unlike inventions, however, a
given brand might be protected by many trademarks (words, logos, 3D mark, sound,
etc.) whereas an invention is supposed be protected by just one patent. The effect of
this on different companies, product categories, industries and countries is not yet
fully clear, and other limitations are possibly not yet identified. One lesson taken from
patent analysis is that decisions to file an IPR vary among different companies,
technologies, industries and countries. Likewise, there is no reason that decisions to
trademark do not vary as well.
On the positive side the large and increasing numbers of trademarks allow us to be
confident that many aspects of corporate commercial activities can be revealed
through this indicator. Because they are cheaper and do not require a technological
breakthrough, a much wider range of SMEs are likely to be involved in applying for
trademark rights compared to patent rights. Moreover, given the known limitations of
patents this indicator is an important complement for covering service activities. This
allows us to cover a wider spectrum of the industrial structure.
In this paper we use CTMs as source of internationally comparable trademark
statistics. Why use OHIM data? This EU trademark system is certainly biased towards
members-states’ commercial activities, generating therefore an over-representation of
European trademarks. However, as the EU is the world’s largest market, it is certainly
an important space for protecting brands from counterfeiting. Moreover, given the
fact that CTMs are attractive to non-EU filers and that applications from all countries
are considered on an equal footing we have, henceforth, a useful basis for
international comparisons. Furthermore, different trademark applications are judged
against precisely the same criteria using the same database. From an instrumental
perspective, OHIM presents the researcher with a wealth of comparable trademark
data and reports are freely accessible online.
4. Empirical exploration
This study asks two main questions: (1) can innovation and industrial change be
assessed through trademark-based indicators?; (2) more precisely, can significant
differences and evolution in performance be identified? We will answer positively
both questions. Although the methodological understanding for using trademarks is
far from mature, our results show that this indicator can generate useful insights for
researchers, policy-makers and managers. In addressing these research questions we
hope not only to demonstrate the power of this type of data for studying innovation
and industrial change but also to shed light on the case of Portugal in the European
context.
Trends in Community Trade Marks (CTM) applications
13
In this sub-section we present data on the rate and direction of CTM applications. We
start by the basic observation of aggregate trademark filings. Figure 1 shows the
number of total CTM filed since it became instituted. In 7 years 294 625 CTMs were
filed by companies from the EU and around the world, corresponding to an average of
about 42 000 trademarks a year. Of this total 65% were word marks and 34% were
figurative marks (logos).
Figure 1. Total CTMs filed, all countries, 1996-2002
70,000
60,000
50,000
40,000
30,000
20,000
10,000
0
1996
1997
1998
1999
2000
2001
2002
Source: OHIM (2003), Annual Report 2002
Note: 2000 data point is based on estimation by interpolation.
The high level of trademark applications in the first year (1996) is due to the start of
new registration system. Many marks were already in existence in other regimes and
decisions of economic agents were waiting for the start of the new system. If we
compute the average rate of growth using the extreme years of 1996 and 2002 we
have a 0,6% change. But if we disregard the figure for the first year, there has been a
growth of 8,7% in annual applications. Later years reveal a downward trend in the
number of applications, a phenomenon probably caused by the general
macroeconomic downturn and a readjustment of business investment expectations.
What about the territorial origin of those applications? As expected, Table 3 shows
applications are dominated by the 15 EU Member States.
Table 3. – Origin of CTM applications, 1996-2002
Annual average 1996-2001
Year 2002
Number of applications
%
Number of applications
%
EU-15
Non-EU Countries
25 679
15 908
62
38
29 435
15 669
65
35
Total
41 587
100
45 104
100
Source: OHIM (2003), Annual Report 2002; Own calculations
14
Within the EU Member States exhibit different performances as portrayed in Table 4.
Trademark applications are strongly concentrated. Germany, UK and Italy filed 60%
of the total CTM filed in the period from 1996 to 2001.4 As expected the first five
countries in trademark applications correspond to the largest European economies and
are responsible for 80% of all the filings. Among the countries outside the EU the US
represents 78% of these CTM filed between 1996 and 2001. In fact, the US is the
biggest single user of the CTM system; it applied for more CTMs than any European
country. As such, this suggests that “trademarking” in the EU’s large and
sophisticated market is a comprehensive reflection of the trends and patterns in the
development of marketing activities of firms.
The normalisation of these figures according to population and the size of the
economy provides further information. Beyond Luxembourg, where a high number of
MNC’s headquarters are located, a number of small countries appear at the top,
including the Nordic countries, Ireland and Austria. Their pattern of business activity
appears to be trademark intensive in the European context, relative to population and
to the GDP. The UK, and to some extent Germany, performs particularly well among
the largest countries.5
Interestingly enough, European patent data point to a similar pattern (Eurostat, 2002).
Within the EU, Germany accounts for the largest share of the EU’s total patent
applications at the European Patent Office (EPO) with 42.4%. France and the UK
accounted for 14.4% and 12.9 % respectively, showing that European patent
applications to the EPO are largely skewed towards the large economies. However,
when taking population into consideration, we have again the small Scandinavian
economies outperforming the large economies: Sweden and Finland recorded the
highest rates.
Among the smaller countries, Portugal and Greece have little trademark activity, even
when considered in relative terms. This is not totally surprising under the light of
other innovation indicators.
Table 4. Trademark performance among the EU, 1996-2001
1996 - 2001
MC/Pop
MC/PIB
DE
41554
LU
2852
LU
GB
32664
DK
734
SE
61
25
IT
17713
SE
623
DK
25
ES
15679
IE
619
GB
22
FR
14944
GB
547
IE
21
NL
6586
DE
506
ES
20
SE
5530
AT
456
DE
20
DK
3915
FI
453
FI
18
AT
3700
NL
415
AT
17
BE
3655
ES
397
NL
15
FI
2347
BE
356
BE
14
IE
2344
IT
310
IT
12
4
Detailed 2002 data not available.
Conclusions must be prudent. As trademarks we are counting my types of distinctive signs such word
marks, logos, 3D, etc. Moreover, many trademarks can protect specific goods and services at the same
time. Strategies to protect products may vary from country to country.
5
15
PT
1543
FR
254
FR
10
LU
1252
PT
154
PT
9
GR
648
GR
61
GR
4
154074
EU
410
UE
17
EU
Sources: OHIM (2003), Statistics of community trade marks, Situation at the end of 2002;
OCDE, Labour Force Statistics; Own calculations
Note: BE – Belgium ; DK – Denmark; DE – Germany; GR – Greece; ES – Spain; FR –
France; IE – Ireland; IT – Italy; LU – Luxembourg; NL – Netherlands; AT – Austria;
PT – Portugal; FI – Finland; SE – Sweden; GB – United Kingdom
On dynamic terms EU’s “trademarking” performance has been ahead of the “rest of
the world”. Between 1996 and 2001 the CTM applications for the EU-15 taken
together grew annually 2.4%; excluding 1996 EU applications grew at an average
annual growth of 9.8%. Within the EU, Spain and Greece, followed by Luxembourg
and Portugal, have been increasingly active. Sweden and Netherlands have had lower
growth rates.
16
Table 5. Detailed annual data for the EU-15, 1996-2002
1996
BE
DK
DE
GR
ES
FR
IE
IT
LU
NL
AT
PT
FI
SE
GB
UE
643
706
7 714
85
2 794
1 610
282
2 211
137
1 069
678
161
358
855
5 705
1997
418
473
3 911
65
1 468
1 891
217
2 070
112
908
393
183
282
713
3 659
1998
562
526
4 811
78
1 965
2 131
289
2 378
148
877
537
232
330
787
4 234
1999
574
621
7 040
111
2 609
2 735
436
3 263
227
994
615
274
387
870
5 303
2000
771
822
9 793
161
3 417
3 406
636
4 221
349
1 632
766
396
507
1 346
7 622
2001
687
767
8 285
148
3 426
3 171
484
3 570
279
1 106
711
297
483
959
6 141
2002 Growth rate Growth rate Growth rate Growth rate
96-01
97-01
97-02
96-02
598
727
7 113
197
3 759
3 411
429
3 719
233
1 055
717
378
438
801
5 860
1.1%
1.4%
1.2%
9.7%
3.5%
12.0%
9.4%
8.3%
12.6%
0.6%
0.8%
10.7%
5.1%
1.9%
1.2%
10.4%
10.2%
16.2%
17.9%
18.5%
10.9%
17.4%
11.5%
20.0%
4.0%
12.6%
10.2%
11.4%
6.1%
10.9%
-1.0%
0.4%
-1.2%
12.8%
4.3%
11.3%
6.2%
7.7%
7.9%
-0.2%
0.8%
13.0%
2.9%
-0.9%
0.4%
6.2%
7.4%
10.5%
20.3%
17.0%
10.3%
12.0%
10.3%
13.0%
2.5%
10.5%
12.9%
7.6%
2.0%
8.2%
25 008 16 763 19 885 26 059 35 845 30 514 29 435
3.4%
12.7%
2.4%
9.8%
Sources: OHIM (2002), Detailed Applications Statistics. Situation at the end of June 2002;
OHIM (2003), Annual Report 2002; Own calculations
Note: Data for 2000 are estimations.
Heterogeneity is also evident in terms of the distribution of applications by Nice
classes. Table 6 shows the ten most “trademarked” product classes between 1996 and
2002. These absorbed 53.2% of the total applications. In 2002 the most demanded
product categories were: Instruments (Class 9), Research (Class 42), Business
Consultancy (Class 35), Paper Products (Class 16) and Education (Class 16).6 It is
interesting to note that the service categories in the top five (Classes 42, 35 and 16)
are broadly correspondent to what Miles (2004) knowledge-intensive business
services, or KIBS, and to what Godinho et al. (2003) and Schmock (2003) called and
knowledge-intensive and information intensive service sectors.
6
It is still not clear is this is due to a great dynamism of innovative activities in this product classes or
rather a manifestation of toughness of competition.
17
Table 6. Most “trademarked” NICE product classes, 1996 to 2002
Top 10
1
2
3
4
5
6
7
8
9
10
Total
NICE
Class
9
42
16
35
41
25
38
5
36
3
Applications
1996-2001
79.374
64.171
44.927
41.535
32.935
29.659
29.292
21.276
20.618
17.864
%
12.1
9.8
6.9
6.3
5.0
4.5
4.5
3.3
3.2
2.7
Top 10
1
2
3
4
5
6
7
8
9
10
NICE
Class
Applications
2002
9
42
35
16
41
25
5
38
3
36
58.3
12.961
8.490
8.029
6.941
6.237
5.681
4.797
4.362
3.797
3.407
%
10.7
6.9
6.6
5.7
5.1
4.7
3.9
3.6
3.1
2.8
53.2
Source: OHIM (2003). Statistics of community trade marks. Situation at the end of 2002; Own
calculations
How do, on the whole, intangible products measure up to tangible ones? Table 7
shows that while classes of goods still count as the major segment, services are rising
fast in importance. This might be interpreted as evidence of structural change in the
EU economies. The latest OECD figures point out that the service sectors accounts of
70% of OECD’s GDP (OECD, 2003). The study by Velling and his colleagues (2002)
finds the same pattern for German home trademark applications and Godinho et al.
(2003) draw similar conclusions in the Portuguese case, as further explored in section
6. They also suggest that in the 1990s one of the possible contributing factors behind
the spectacular increase of service marks in the increase of service marks applications
by manufacturing companies.
Table 7. Structure and sectoral dynamics of CTM “trademarking”
Goods
Services
1996 1997 1998 1999 2001 2002
74
72
69
65
58
66
26
28
31
35
42
34
Growth rate 96-02 Growth rate 97-02
0.4%
9%
6%
14%
Fonte: IHMI (2002), Detailed Applications Statistics, Situation at the end of June 2002;
OHIM (2003), Statistics of community trade marks. Situation at the end of 2002; Own
calculations
Note: Desagregated data for 2000 nor available
If services are a dynamic category, how have services themselves changed in terms of
structure? Looking for the individual service classes can give further insight into the
direction of development in services and to the more dynamic services. Table 8 shows
the composition of the service trademarks by different service classes, while table 9
presents the rates of growth7 of the different classes. The services with more
trademark applications are Research (Class 42), Business Consultancy (Class 35),
7
Classes 43, 44 and 45 were added only in 2002, therefore their growth rates were not included. The
inclusion of these new classes may have differentiated effects in the other classes.
18
Education (Class 41) and Telecoms (Class 38). Knowledge-based services again are
prominent.
Table 8. Composition of CTM service applications
Nice Class
35
36
37
38
39
40
41
42
43
44
45
Descriptor
1996-2002
%
Business Consultancy
Finance
Construction
Telecommunications
Transportation
Processes
Education
Research
Restaurants and Hotels
Medical Care
Personal Services
49564
24025
17814
33654
17346
6607
39172
72661
1883
1501
506
19
9
7
13
7
2
15
27
1
1
0.2
Services
264733
100
Total
Source: OHIM (2003), Detailed Applications Statistics, Situation at the end of June 2002
The following table presents trademark application evolution by type of service for
the available years. Nice classes growing above of the average of the service sector
were again knowledge-based services: Business Consultancy (Class 35) and Telecoms
(Class 38).
Table 9. CTM growth within service product categories
NICE Class
Descriptor
Growth rate 1996-2002Growth rate 1997-2002
35
36
37
38
39
40
41
42
Business Consultancy
Finance
Construction
Telecommunications
Transportation
Processes
Education
Research
12.5
5.3
2.2
9.1
6.0
6.5
6.3
1.7
20.0%
14.4%
14.4%
16.5%
15.3%
15.2%
15.2%
8.6%
Total
Services
7.4
16.2%
Source: OHIM (2002), Detailed Applications Statistics. Situation at the end of June 2002;
OHIM (2003), Statistics of community trade marks. Situation at the end of 2002;
Own calculations
Two difficulties limit, however, our capacity to produce more precise conclusions. On
the one hand, as already mentioned, Nice classes are very aggregate, containing many
different kinds of products under the same heading. On the other hand, the short life
of this IPR regime in Europe makes us cautious on the comparative dynamics of
product classes.
5. Patterns of use of national trademarks by Portuguese firms
19
In this section we explore in greater detail data from the Portuguese case, drawing on
a wide study on the use of IPR in Portugal.
The oldest trademark in Portugal still standing today was registered February 20,
1890. The trademark Real Companhia Vinícola do Norte de Portugal given to the
company with the same name for the specific uses of stamping it on Port wine casks.8
Among the ten oldest registered trademarks the majority of them correspond to the
wine and olive oil business, corresponding to the historical productive profile of the
Portuguese economy.
The rest of this section reports evidence from the period from 1980 to 20019. With a
yearly growth of about 10% a year, trademarks granted by the Portuguese National
Intellectual Property Office (INPI) to resident companies surpassed annual GDP
growth. Portuguese trademarkers are mostly business firms (80%), the rest are private
individuals (most of whom are likely to be owners of small firms). Concession rates
(trademarks registered on the base of trademarks applied) were, during this period,
86% for residents and 96% for non-residents. Non-residents tend to have trademarks
that last longer: total dropout in the period was 11% where as residents dropped 19%
of the trademarks registered in the period. Non-residents have a lower growth rate of
applications; these companies’ applications peaked just after Portugal became a
member of the European Economic Community. Among non-residents USA is the
largest filer with 9% of all non-resident applications, followed by the UK (15%),
Spain (12%), Japan (6%), France (4%), Germany (3%) and Brazil (3%). On the whole
EU-15 filings account for 42% of the total non-resident trademarking. From the mid1990s Spanish and Brazilian companies became the most dynamic foreign
trademarkers.
National marks are also classified on the basis of the Nice Agreement. The most
demanded classes by residents consist of Pharmaceutical and Hygiene Products (Class
5), Instruments (Class 5), followed by Detergents and Cosmetics (Class 3), Paper
Products and Clothing and Footwear (Class 25). This pattern discloses some
differentiation by applications of residents and, above all, a predominance of products
in relation to the services. However, the last years of the period saw some changes,
with classes typically of services be more demanded, namely Class 41 (Education),
Class 42 (Research) and Class 35 (Business Consultancy), and filings divided equally
between goods and services.
Trends for both residents and non-residents follow a path of sectoral change in line
with processes of economic development. These transformations are essentially
characterised by a gradual fall of the proportional importance of the tangible goods
and by increases in the service products. Table 10 presents this pattern.
8
9
This mark was registered as an international trademark in 1925
This sub-section draws substantially on Godinho et al. (2003).
20
Table 10. - Distribution of national trademark applications by goods and services
Residents
Non-residents
Goods
Services
Total
Goods
Services
Total
80-84
89.7%
10.3%
4665
87.8%
12.2%
4435
85-89
87.1%
12.9%
9285
87.4%
12.6%
9971
90-94
80.5%
19.5%
22669
84.6%
15.4%
22563
95-99
66.3%
33.7%
31279
78.9%
21.1%
17700
51.0%
49.0%
27751
72.4%
27.6%
7575
68.4%
31.6%
95649
82.2%
17.8%
62244
2000-02
Total
10
Source: INPI; Own calculations
Also as expected, non-residents do not trademark as much services as the residents
given the fact that services have been traditionally more locally based.11 It is also
interesting to note that the weight of the national wine industry is reflected in the
detailed data as Class 33 (Beverages) has much more filings by residents than by nonresidents.
The information obtained from a survey on the use of IPR by resident firms shows
that the interest for and actual use of IPRs is still very limited. Only 39% of business
firms state having acquaintance with INPI activities, while a smaller proportion (19%)
acknowledges actual use of IPRs or intention of their use in the near future. These
values have a statistically significant variance across sectors and display a positive
association with firm size.
In terms of attitude, more than half of firms express a “low interest” in IP issues with
responses varying between 54.0% and 86.7% for different IPR types. Trademarks are
the more attractive IPRs, with 17.8% of the firms showing “high” interest for it and an
additional 28.2% “medium” interest. In terms of actual use 4.2% of the surveyed
firms state that they have applied for Patents or Utility Models while 17.7% applied
for Trademarks and 2.6% for protection related to Technical Designs and Industrial
Models.
These results are not surprising or particularly different from what has been observed
elsewhere. They stem from a variety of reasons, the most important ones being: (1)
economic structure (relative low weight of both high tech industries and information
intensive services and virtual absence of large firms acting on global markets); (2)
low supply of critical competencies (required to explore commercially IPRs and
manage complementarities within IPR portfolios); (3) low accessibility both to
demanding users and to technology markets (this inhibits the capacity to explore
commercially any IPR undertaken by firms, inventors or public research
organisations).
10
11
June 2002.
Negotiations on a General Agreement on Trade in Services (GATS) are now under way.
21
A number of variables revealed a statistically significant degree of association with
the importance attributed to trademarks and their use in the past. These factors
included the existence of an autonomous marketing department in the company, the
relationship with specialised consultancies, the size of marketing budgets and the
knowledge of the national IPR agency’s services and role. A variable such as size was
always strongly significant in the analysis. However, whether a firm is geared to
intermediary products or final consumer products does not discriminate the firms in
terms of importance attributed to trademarks but only in their actual use: consumer
goods firms tend to use more trademarks.
Since we found a great correlation between the uses of different IPR rights, it is not
unexpected that there are clear distinctions between manufacturing industries
aggregated by technological intensity. High and medium high-technology industries
tend to have an integrated strategy regarding both patents and trademarks. The
technological intensity taxonomy seems to discriminate well between industries in
relation to their use of trademarks. Services, which were classified as high or low
information intensive, exhibit a clear divide: high information intensive service
sectors such as consultancies, telecoms and banks use much more trademarks in their
business strategies than sectors such as real-estate, restaurants, transportation and
personal service industries. These survey observations are in line with the data on the
filing of trademarks, from the database analysis.
6. Avenues for further research
There are a variety of ways in which trademark data can be further analysed. The
performance of different countries can be compared in terms of their Revealed
Marketing Advantage (RMA) indicator, as has been done for other indicators. The
RMA calculates a country’s industry share in each of the trademark classes divided by
the share of total trademarking in the same class. It indicates the relative trademarking
activity of a country (j) in a specific product category (i), normalised to the total
trademarking of the relevant population in that class. Preliminary analysis using CTM
data has shown that Portugal comparative marketing advantages lie in Metal Goods
(Class 6), Textiles (Class 24), Business Consultancy (Class 35) and Personal Services
(Class 45). While the first two results are unproblematic give the weight of
automotive components and clothing products in the international trade profile of the
Portuguese economy, the observation of service strengths is more difficult to interpret
given their essentially non-transactional nature.
TM ij
RMA ij =
∑ TM kj
k
∑ TM il
l
∑ ∑ TM kl
k
l
22
Other aspects that can be further analysed include the life cycle of brands, the factors
and trends in trademark licensing. The relationship between trademarks and other
IPRs in the context of integrated IPR management can be further explored using
quantitative data. The cross analysis of trademark data, regarding the ways different
trademarks (word marks, logos, 3D, sound, etc.) are combined to protect given
products or applications across different Nice classes (requiring extensive data
cleaning and database management) is promising and not well understood. In this
perspective trademarks can be understood as indicators of the marketing capabilities
of profit-oriented firms. Studying how firms exploit trademark positions in certain
classes to make inroads into other classes could yield valuable insights into the
dynamics of product diversification.
7. Concluding remarks
The implicit question this paper tried to answer was “what can welearn from
trademarks as an indicator of innovation and industrial evolution?”. The objective was
therefore to test trademarks as an indicator of product innovation activity as well as a
measure correlated to structural change in contemporary economies.
Brands are a very important part of firms’ marketing plans and strategies, they are
used to protect firms’ products and business identity but also for other purposes like
business diversification. Firms make a huge (and increasing) use of brands as a
(dynamic) competition tool. As a result applications for service trademarks have
boomed in the 1990s, and this trend was led by knowledge-based services: Business
Consultancy, Telecoms and Education. Within the EU-15, a number of small
countries seem to exhibit very strong marketing capabilities, namely the Nordic
countries, Ireland and Austria.
The analysis of more detailed data on Portugal revealed it is lagging behind in terms
of such marketing capabilities. However, Portugal shows clear signs of structural
change. It goes along the general trend towards increasing services trademarking.
Moreover, the composition of this trend is biased towards Education, Research and
Business Consultancy categories. Evidence from the Portuguese case suggests that
companies that tend to use one kind of IPR tend to use other IPRs. That implies that
highly technological intensive sectors, which use more patents, also make a more
intensive use of trademarks. In the case of services, evidence from a survey in
Portugal shows that industries usually classified as intensive users of information are
those that use more trademarks. This observation is compatible with the evidence
from the CTM database on the dynamism of knowledge services in recent years.
Combined with the increasing availability of electronic data, these results indicate an
interesting opportunity for using trademarks as indicators of innovation and industrial
change. We argue that trademark data can serve the purpose of being a partial output
indicator of new products introduced in the market and, therefore, can be used as an
empirical yardstick for measuring overall change in the patterns of economic activity.
This can prove to be especially useful for advancing research in innovation studies,
industrial dynamics, international economics, as well as economic and business
history. Acknowledging the limitations of this type of data, and having soughed to
contribute to the methodological reflection on this indicator, this paper argues that
23
substantial knowledge about innovation and industrial change can be acquired with
trademarks analysis. The fact that much remains to be done is the best of the good
news.
References
Aaker, D.A. and E. Joachimsthaler (2000), Brand Leadership, The Free press: New York.
Aaker, D.A. (1991), Managing Brand Equity: Capitalizing on the Value of a Brand, The Free press:
New York.
Cabral, L. (2000a), Introduction Industrial Organization, The MIT Press: Cambridge, Massachusetts.
Cabral, L. (2000b), www.luiscabral.com
Christopher, M. and M. Mcdonald (1991), Marketing: An Introdutcion, Pan...
Cohen, W., R. Nelson and J. Walsh (2000), “Protecting their intellectual assets: Appropriability
conditions and why U.S. manufacturing firms patent (or not)”, NBER Working Paper Series
7552.
Dearlove, D. and S. Crainer (1999), As 50 Maiores Marcas, Biblioteca Executive Digest,
Abril/Controljornal Editora: Linda-a-velha.
Diamon, S.A. (1983), “The historical development of trademarks”, Trademark Reporter, vol. 73, no. 3,
pp. 222-47.
Doern, G. B. (1999), Global Change and Intellectual Property Agencies, Pinter, Londres.
Economides, Nicholas (1987), ‘The Economics of Trademarks’, TradeMark Register, vol. 78, pp. 523539.
Eurostate (2002), Statistics in Focus: Science & Technology, Theme 9, 1, European Communities:
Brussels.
Godin, B. (2003), “The emergence of S&T indicators: Why did governments supplement statistics with
indicators?”, Research Policy, vol. 32, pp. 679-691.
Godinho, M.M., T.S. Pereira, V.C. Simões, S. Mendonça and V. Sousa (2003), Estudo Sobre a
Utilização da Propriedade Industrial em Portugal, INPI: Lisboa.
Granstrand, Ö. (1999), The Economics and Management of Intellectual Property – Towards
Intellectual Property, Edward Elgar: Cheltenham, Northampton.
Griliches, Z. (1990), “Patent statistics as economic indicators: A survey”, Journal of Economic
Literature, vol. 27, pp. 1661-707.
Kotler, P., G. Armstrong. J. Saunders e V. Wong (1999), Principles of Marketing, Second European
Edition, Prentice Hall Europe.
Martin, B. and J. Irvine (1983). “Assessing basic research: Some partial indicators of scientific progress
in radioastronomy”, Research Policy, vol. 12. pp. 61-90.
Narin, F. and D. Olivastro (1988), “Technology indicators based on patents and patent citations”, in A.
van Raam (ed.), Handbook of Quantitative Studies of Science and Technology, North Holland:
Amsterdam, pp. 465-5.
OECD (2003), STI Scoreboard 2003, OECD: Paris.
24
Patel, P. (1998), “Indicators for systems of innovation and system interactions”, IDEA Paper series 11,
STEP Group.
Patel, P. and K. Pavitt (1994), “Patterns of economic activity: Their measurement and interpretation”,
in Stoneman. P. (ed.), Handbook of Economics of Innovation and Technical Change.
Blackwell: Oxford.
Pavitt, K. (1985), “Patent statistics as indicators of innovative activities: Possibilities and Problems”.
Scientometrics, vol. 7, nos. 1-2, pp. 77-99.
Scherer, F.M. (1983), “The propensity to patent”, International Journal of Industrial Organisation, vol.
1, pp. 107-28.
Schmoch, Ulrich (2003), “Service marks as novel innovation indicator”, Research Evaluation, Vol. 12,
2, pp. 149-156.
Smith, K. (2004), “Measuring innovation”, forthcoming in J. Fagerberg, D. Mowery and R. Nelson
(eds), Understanding Innovation - Oxford Handbook of Innovation, Oxford University Press.
Trajtenberg, M. (1990), Economic Analysis of Product Innovation: The Case of CT Scanners, Harvard
University Press: Cambridge, Massachusetts.
Trajtenberg, M. (2001), “Innovation in Israel 1968-1997: A comparative analysis using patent data”,
Research Policy, vol. 30, pp. 363-89.
Velling, J. (ed.) (2002), 2001 Germany’s Technological Performance, Bundesministerium für Bildung
und Forschung, available at …
Wilkins. M. (1992), “The neglected intangible asset: The influence of the Trade Mark on the rise of the
modern corporation”, in Harvey and Jones (eds). ... . pp. 691-707.
Appendix
NICE CLASSIFICATION SYSTEM
1.
2.
3.
4.
5.
6.
7.
8.
Chemicals used in industry. science and photography. as well as in agriculture. horticulture and
forestry; unprocessed artificial resins. unprocessed plastics; manures; fire extinguishing compositions;
tempering and soldering preparations; chemical substances for preserving foodstuffs; tanning
substances; adhesives used in industry.
Paints. varnishes. lacquers; preservatives against rust and against deterioration of wood; colorants;
mordants; raw natural resins; metals in foil and powder form for painters. decorators. printers and
artists.
Bleaching preparations and other substances for laundry use; cleaning. polishing. scouring and
abrasive preparations; soaps; perfumery. essential oils. cosmetics. hair lotions; dentifrices.
Industrial oils and greases; lubricants; dust absorbing. wetting and binding compositions; fuels
(including motor spirit) and illuminants; candles and wicks for lighting.
Pharmaceutical and veterinary preparations; sanitary preparations for medical purposes; dietetic
substances adapted for medical use. food for babies; plasters. materials for dressings; material for
stopping teeth. dental wax; disinfectants; preparations for destroying vermin; fungicides. Herbicides.
Common metals and their alloys; metal building materials; transportable buildings of metal; materials
of metal for railway tracks; non-electric cables and wires of common metal; ironmongery. small items
of metal hardware; pipes and tubes of metal; safes; goods of common metal not included in other
classes; ores.
Machines and machine tools; motors and engines (except for land vehicles); machine coupling and
transmission components (except for land vehicles); agricultural implements other than handoperated; incubators for eggs.
Hand tools and implements (hand operated); cutlery; side arms; razors.
25
9.
Scientific. nautical. surveying. photographic. cinematographic. optical. weighing. measuring.
signalling. checking (supervision). life-saving and teaching apparatus and instruments; apparatus and
instruments for conducting. switching. transforming. accumulating. regulating or controlling
electricity; apparatus for recording. transmission or reproduction of sound or images; magnetic data
carriers. recording discs; automatic vending machines and mechanisms for coin-operated apparatus;
cash registers. calculating machines. data processing equipment and computers; fire-extinguishing
apparatus.
10. Surgical. medical. dental and veterinary apparatus and instruments. artificial limbs. eyes and teeth;
orthopedic articles; suture materials.
11. Apparatus for lighting. heating. steam generating. cooking. refrigerating. drying. ventilating. water
supply and sanitary purposes.
12. Vehicles; apparatus for locomotion by land. air or water.
13. Firearms; ammunition and projectiles; explosives; fireworks.
14. Precious metals and their alloys and goods in precious metals or coated therewith. not included in
other classes; jewellery. precious stones; horological and chronometric instruments.
15. Musical instruments.
16. Paper. cardboard and goods made from these materials. not included in other classes; printed matter;
bookbinding material; photographs; stationery; adhesives for stationery or household purposes; artists'
materials; paint brushes; typewriters and office requisites (except furniture); instructional and teaching
material (except apparatus); plastic materials for packaging (not included in other classes); printers'
type; printing blocks.
17. Rubber. gutta-percha. gum. asbestos. mica and goods made from these materials and not included in
other classes; plastics in extruded form for use in manufacture; packing. stopping and insulating
materials; flexible pipes. not of metal.
18. Leather and imitations of leather. and goods made of these materials and not included in other classes;
animal skins. hides; trunks and travelling bags; umbrellas. parasols and walking sticks; whips. harness
and saddlery.
19. Building materials (non-metallic); non-metallic rigid pipes for building; asphalt. pitch and bitumen;
non-metallic transportable buildings; monuments. not of metal.
20. Furniture. mirrors. picture frames; goods (not included in other classes) of wood. cork. reed. cane.
wicker. horn. bone. ivory. whalebone. shell. amber. mother-of-pearl. meerschaum and substitutes for
all these materials. or of plastics.
21. Household or kitchen utensils and containers (not of precious metal or coated therewith); combs and
sponges; brushes (except paint brushes); brush-making materials; articles for cleaning purposes;
steelwool; unworked or semi-worked glass (except glass used in building); glassware. porcelain and
earthenware not included in other classes.
22. Ropes. string. nets. tents. awnings. tarpaulins. sails. sacks and bags (not included in other classes);
padding and stuffing materials (except of rubber or plastics); raw fibrous textile materials.
23. Yarns and threads. for textile use.
24. Textiles and textile goods. not included in other classes; bed and table covers.
25. Clothing. footwear. headgear.
26. Lace and embroidery. ribbons and braid; buttons. hooks and eyes. pins and needles; artificial flowers.
27. Carpets. rugs. mats and matting. linoleum and other materials for covering existing floors; wall
hangings (non-textile).
28. Games and playthings; gymnastic and sporting articles not included in other classes; decorations for
Christmas trees.
29. Meat. fish. poultry and game; meat extracts; preserved. dried and cooked fruits and vegetables; jellies.
jams. fruit sauces; eggs. milk and milk products; edible oils and fats.
30. Coffee. tea. cocoa. sugar. rice. tapioca. sago. artificial coffee; flour and preparations made from
cereals. bread. pastry and confectionery. ices; honey. treacle; yeast. baking-powder; salt. mustard;
vinegar. sauces (condiments); spices; ice.
31. Agricultural. horticultural and forestry products and grains not included in other classes; live animals;
fresh fruits and vegetables; seeds. natural plants and flowers; foodstuffs for animals. malt.
32. Beers; mineral and aerated waters and other non-alcoholic drinks; fruit drinks and fruit juices; syrups
and other preparations for making beverages.
33. Alcoholic beverages (except beers).
34. Tobacco; smokers' articles; matches
35. Advertising; business management; business administration; office functions.
36. Insurance; financial affairs; monetary affairs; real estate affairs.
37. Building construction; repair; installation services.
26
38. Telecommunications.
39. Transport; packaging and storage of goods; travel arrangement.
40. Treatment of materials.
41. Education; providing of training; entertainment; sporting and cultural activities.
42. Scientific and technological services and research and design relating thereto; industrial analysis and
research services; design and development of computer hardware and software; legal services.
43. Services for providing food and drink; temporary accommodation.
44. Medical services; veterinary services; hygienic and beauty care for human beings or animals;
agriculture. horticulture and forestry services.
45. Personal and social services rendered by others to meet the needs of individuals; security services for
the protection of property and individuals.
27