4.1 Introduction

Patents are among the most important indicators for the output of R&D processes and are frequently used to assess the technological performance of firms, technology fields, and entire economies (Freeman 1982; Griliches 1990; Grupp 1998). Thus, the key assumption is that patents reflect the knowledge capabilities or knowledge stocks of the patenting entities (mainly companies but also universities or public research institutes and single inventors) and – from a wider perspective – entire nations (Frietsch and Schmoch 2006). Although a patent may have no direct value for a firm or innovation system, it is at least a part of a technological trajectory from which a firm expects to generate economic or strategic value.

Because patents are used as output indicators of innovation, they fit into a system of several additional indicators to describe scientific and technological competitiveness and to analyse innovation systems. From this perspective, patents are an intermediate measure because they cover the output of R&D systems for which expenditures or human capital are the input. At the same time, patents can be regarded as an input into additional market activities, which are reflected by foreign trade, turnover, or qualified labour.

At its core, a patent is a legal intellectual right granted by an authorised government entity (patent office) to exclusively protect an invention from unauthorised use for a certain period of time (Frietsch et al. 2010), which is coupled with a disclosure requirement. Thus, all information that is covered by the respective patent must be disclosed after a given time period. Therefore, the patent system offers a temporary monopoly to inventors in exchange for their early disclosure of new technologies.

An invention, however, does not necessarily translate into an innovation. An invention is “[…] a research and development driven initial technical realisation of a new problem-solving mechanism” (Pleschak and Sabisch 1996). This definition implies that an invention only represents technical information, which may have an economic value in the future. An innovation “[…] is the implementation of a new or significantly improved product (good or service), or process, a new marketing method, or a new organisational method in business practices, workplace organisation or external relations” (OECD and Eurostat 2005). In many cases, an invention can be viewed as an initiator for an innovation. Because the innovation process encompasses all stages, including planning, research, invention, commercialisation and implementation, an invention alone is not sufficient to qualify as an innovation (Schubert et al. 2011). Only the result of a full innovation process, (e.g., a new product or new process can be viewed as innovation) (Grupp 1998). However, the successful completion of the innovation process alone is not a sufficient condition to obtain the expected benefits from innovation because firms must appropriate these benefits (i.e., to prevent its competitors from imitating their results) (Hanel 2008). This result can be achieved not only by patenting or other intellectual property rights (IPRs) but also by informal appropriation mechanisms, such as keeping an invention secret or utilising lead-time advantages (Blind et al. 2006; Neuhäusler 2012).

The growth of the world economy and increasing globalisation have led to a rapid expansion of access to information and new markets for inventors and resulted in greater international competition and new forms of organisation, which makes intellectual property protection increasingly important (OECD and Eurostat 2005). However, intellectual property protection via patents is not only a phenomenon of high technology. A large share of inventions originates from the non-R&D-intensive technology areas, which makes intellectual property protection via patenting indispensable to stay competitive at the national and international levels.

Furthermore, not all technological output of non-R&D-intensive industries can be considered non-R&D-intensive technology; instead, it may be medium- or even high-tech. Sectors are rather heterogeneous agglomerations of companies, which may themselves be heterogeneous in terms of products, services, or even R&D intensity. A non-R&D-intensive sector is defined as the share of total R&D expenditures over turnover below an average threshold of 2.5 %. Some companies may conduct research regardless of this threshold, and some companies may even invest greater than 2.5 % shares in R&D. However, we must stress the distinction between non-R&D-intensive sectors/industries and non-R&D-intensive technology fields or areas.

In summary, we aim to shed light on the technological output of non-R&D-intensive sectors with the help of patent indicators. In addition to investigating the absolute numbers and shares of patent filings compared with the high-technology areas, we will examine the positioning of non-R&D-intensive sectors within the innovation chain and assess their internationalisation trends within Germany over the last decade.

4.2 Data and Methodology

The patent data for this study were extracted from the “EPO Worldwide Patent Statistical Database” (PATSTAT), which provides information regarding published patents collected from 83 patent authorities worldwide. To differentiate technology fields according to their research intensity and International Patent Classification (IPC), a list of research-intensive industries and goods (NIW/ISI/ZEW-Lists 2012) is employed, where a distinction between low- and high-technology areas is established (Gehrke et al. 2013). High-technology sectors are defined as technologies for which an average investment in R&D of greater than 2.5 % of the turnover is required. Additionally, high-technology sectors are differentiated by high-level and leading-edge technologies. Whereas the high-level sector includes technologies that require R&D expenditures between 2.5 % and 7 %, the leading-edge area covers technologies that are beyond 7 % investment shares. The remaining technology fields are defined as non-R&D-intensive.

At the core of the analysis, the data applied follow the concept of transnational patents suggested by Frietsch and Schmoch (2010), which can overcome the home advantage of domestic applicants, thus enabling a comparison of the technological strengths and weaknesses beyond home advantage and unequal market orientations. In detail, all filings via the Patent Cooperation Treaty (PCT) at the World Intellectual Property Organisation (WIPO) are counted regardless of whether they are transferred to the European Patent Office (EPO) or not. Furthermore, the analysis includes all direct EPO applications without precursor PCT application, excluding double counts. All patents in our analyses are counted according to their year of first worldwide filing, which is commonly known as the priority year. The priority year is the earliest registered date in the patent process and is therefore closest to the date of invention.

Furthermore, the method of fractional counting by (inventor) countries and technology fields is applied to consider cross-classifications of patent filings within the IPC. If a patent is assigned three IPC codes of which two would fall into the category of high-level technologies and one to the category of non-R&D-intensive technologies, it would be counted as two thirds for the high-level technologies and one third for the non-R&D-intensive category.

Additionally, some of the patent statistics are differentiated by the type of patent applicant (i.e., if the applicant is a small or medium-sized enterprise (SME), large enterprise (LE), or academic applicant (university or public research institute)) (Frietsch et al. 2011). The company identifier (e.g., Inc., Corp., GmbH, AG) in the name of the patent applicants indicates whether it is a company. Single inventors, universities, and public research institutes were classified manually. In correspondence to the German SME definition (Günterberg and Kayser 2004), applicants with more than 500 employees and more than three patent filings in a 3-year time window between the priority years of 1996 and 2008 were classified as LEs. The remaining applicants that have less than three patent filings in the given time window and less than 500 employees were classified as SMEs.

4.3 Results

4.3.1 General Structures

First, we consider the technological perspective and compare R&D-intensive and non-R&D-intensive technologies. Technological innovation is a prominent phenomenon in non-R&D-intensive technology areas. Nearly 40 % of all transnational patent filings can be classified as non-R&D-intensive technologies (Fig. 4.1), whereas the remaining 60 % of patent filings in 2010 are classified as leading-edge (approximately 28 %) and high-level technologies (approximately 34 %). In absolute terms, the non-R&D-intensive technology areas account for more than 84,000 transnational patent filings in the year 2010. Aside from the major decrease in patent filings during the recent economic crisis, which has affected all of the technology areas to a similar extent, the absolute number of patent filings has been increasing over the years. Nevertheless, a slightly decreasing trend in the shares of patent filings in non-R&D-intensive technology areas for total filings can be observed since the year 2000. This trend is mainly due to an increase in the shares of patents from high-level technologies – which is especially prominent in the year 2010 - that leads to declining shares of filings from the leading-edge and non-R&D-intensive technology areas.

Fig. 4.1
figure 1

Absolute numbers and shares of transnational patent filings by technology areas, 2000–2010 (Source: EPO – PATSTAT; Fraunhofer ISI calculations)

Full size image

However, these trends become clearer when reviewing them by country (Fig. 4.2). Two BRICS countries, namely, South Africa and Brazil, have the highest shares of transnational patents in non-R&D-intensive technologies. In most European countries (e.g., Germany, France, Great Britain, Austria, Switzerland, and the Netherlands), the shares of patents from the non-R&D-intensive technology areas are slightly above the world average of 39.4 % (bold line). However, most Scandinavian countries have shares slightly below the worldwide average.

Fig. 4.2
figure 2

Country-specific shares of transnational patent filings in non-R&D-intensive technology areas, 2010 (Source: EPO – PATSTAT; Fraunhofer ISI calculations)

Full size image

North American countries (i.e., the USA and Canada, as well as Israel, which is technologically oriented toward the U.S. market) and Asian countries (i.e., Japan, Korea, and China) have comparably low shares of patents in non-R&D-intensive technology areas. Aside from Russia, which is located in the middle ranks with an above-average value, the BRICS countries have rather distinct technological strategies. Whereas South Africa and Brazil are highly patent-active in the non-R&D-intensive areas, China and India have the lowest shares of patents in these areas.

It is not only important to consider the shares within the technology fields but from a sectoral perspective (i.e., to determine which types of technologies are actually produced by firms from the non-R&D-intensive industries). Thus, aside from the technological view, a second perspective considers non-R&D-intensive sectors. Instead of classifying the patents/technologies according to their R&D intensity, we now use the sector classification (NACE) of the patent applicants and assign each company to one of these groups. Using a cross-tabulation of sectors and technologies enables us to observe whether companies in non-R&D-intensive sectors only file patents that are classified as non-R&D-intensive patents or whether they are active in high technology patenting as well. The shares of patents can also be calculated from non-R&D-intensive technology areas in high-technology sectors.

However, it is difficult to combine the sectoral and technological perspectives because patents are classified based on their technological content according to IPC, which does not correspond to industry definitions (e.g., by NACE codes). Therefore, a concordance between industries and technology fields is required. To reach this concordance, a probability matching of German patent applicants with company names from the Hoppenstedt database of German companies (www.hoppenstedt.de) was applied. This matching allows for a combination of the two data sources at the micro-level of companies/patent applicants, through which patent filings can be assigned to individual sectors of the economy,Footnote 1 and enables us to calculate the shares of patents in non-R&D-intensive (manufacturing) industries by technology field. The non-R&D-intensive manufacturing sectors are defined according to their R&D intensity and are based on the sector lists by Gehrke et al. (2013) at the level of 3-digit NACE (Rev. 2) codes.Footnote 2 The results are displayed in Fig. 4.3.

Fig. 4.3
figure 3

Shares of transnational patent filings from non-R&D-intensive manufacturing industries (NACE Rev. 2) by technology field, German applicants (Source: EPO – PATSTAT; Hoppenstedt; Fraunhofer ISI calculations)

Full size image

Approximately 65 % of all patent filings from German companies in the non-R&D-intensive sectors are filed in technology fields that are defined as non-R&D-intensive. However, approximately 7 % of their filings are classified as leading-edge technologies and 28 % are classified as high-level technologies. A comparison of the trends over time indicates that since 2006, patents from the non-R&D-intensive sectors are increasingly filed in high-level and leading-edge technologies. Thus, firms from the non-R&D-intensive industries have increasingly entered the high-technology scene in terms of patenting in recent years, and they increasingly rely on R&D-intensive technologies in their daily business.

As stated above, we can view the argument from the inverse perspective and consider the shares of patent filings from high-level and leading-edge manufacturing industries in non-R&D-intensive technology areas (Fig. 4.4). The companies from high-level sectors have a higher share of patent filings in non-R&D-intensive technology areas than companies from leading-edge sectors. Over the years, the shares for the high-level sectors have increased, implying that not only companies from non-R&D-intensive industries are increasingly entering the high-technology scene, but also that more patents in non-R&D-intensive technology areas are filed by companies from the high-level technology sectors. For the leading-edge sectors, the share of patents in non-R&D-intensive technology areas remains relatively stable over the entire time period, with only a slight increase of 2 % points between the years 2000 and 2010.

Fig. 4.4
figure 4

Shares of transnational patent filings from high-level and leading-edge manufacturing industries (NACE Rev. 2) in non-R&D-intensive technology areas, German applicants (Source: EPO – PATSTAT; Hoppenstedt; Fraunhofer ISI calculations)

Full size image

To gain a complete perspective of the trends in patenting in non-R&D-intensive technology areas, we review the shares of patent filings differentiated by the type of patent applicant (i.e., SMEs, large enterprises, and academia, including both universities and public research institutes). In addition to the worldwide average, the trends for Germany are displayed in greater detail (Fig. 4.5).

Fig. 4.5
figure 5

Applicant-type specific shares of transnational patent filings by technology area, in Germany and total, 2010 (Source: EPO – PATSTAT; Fraunhofer ISI calculations)

Full size image

Although large enterprises are responsible for the largest number of filings in non-R&D-intensive areas, the shares of SME filings in these areas are higher (i.e., 45 % of all SME filings are categorised as non-R&D-intensive). As expected, academia has the lowest shares of patents in non-R&D-intensive technology areas. Thus, the focus is clearly on research in leading-edge technologies. No clear specialisation can be observed for large enterprises. All three technology areas are nearly equally represented in their patent filings.

The figures for Germany are slightly different than those for the worldwide scale. Although the basic trends across the types of patent applicants are similar (i.e., SMEs have the largest shares of non-R&D-intensive patents in their portfolio, whereas the smallest shares can be found for academia), the general focus on non-R&D-intensive technology areas is greater in Germany than in the world average. This result resembles the country-specific trends shown in Fig. 4.2, where Germany has an above-average number of filings in non-R&D-intensive technologies. However, this greater focus on non-R&D-intensive technology areas mainly comes at the expense of a smaller focus on leading-edge technologies. High-level technologies, of which most mechanical engineering sectors belong and for which Germany is has its technological strengths, are slightly over-represented in the German technology portfolio, which can mainly be attributed to the large share of filings in high-level technologies from large enterprises.

We will now change our focus and review the internationalisation trends in the respective areas and then assess the non-R&D-intensive sectors' positioning within the innovation chain.

4.3.2 Internationalisation of Non-R&D-Intensive Technology Areas

The analysis of transnational filings offers the opportunity to assess trends beyond home advantage effects, national idiosyncrasies, and different market orientations. We can develop statements regarding the international orientation of countries for specific technologies or technological areas by relating the number of transnational filings to the total number of filings that are targeted toward a national market. Because patent filings via the EPO or PCT system are more expensive than purely national filings, we can assume that the patent applicant is expecting to sell products that incorporate the protected invention at an international scale (i.e., at several international markets). Thus, the ratio of transnational filings to national filings provides an estimate of how many inventions will be commercialised internationally.

To construct our indicator, we count all patent filings that are filed at the German Patent and Trademark Office (GPTO), whether they were filed directly at the GPTO via the EPO system or PCT system, excluding double counts. This method of counting patents is associated with the assumption that all PCT filings and patents granted by the EPO are forwarded to the GPTO, which is true for the majority of patents, at least when German inventors are named on the patent application. Thus, we limit the analysis to patents filed by German inventors.

By relating the number of transnational filings to the number of patents that are targeted toward the German market, differentiated by technological areas, we can estimate the internationalisation trends within these areas, which is plotted over time in Fig. 4.6. The orientation toward international markets has increased over the study period for all technology areas. Additionally, there was a strong decline in the share in transnational filings during the crisis of the new economy in 2003 and 2004 as well as during the recent economic crisis due to the cost-saving IP-strategies of firms within these periods (Neuhäusler et al. 2014).

Fig. 4.6
figure 6

International orientation of German patent filings by technology area, 2000–2010 (Source: EPO – PATSTAT; Fraunhofer ISI calculations)

Full size image

A review of the different technological areas demonstrates that leading-edge technologies are most strongly oriented toward international markets across the entire time period. Approximately 63 % of all filings from leading-edge technologies targeting the German market are filed via the PCT or EPO system. The trends for the high-level technologies and non-R&D-intensive areas are even more interesting. Whereas R&D-intensive areas were slightly less internationalised than high-level technologies at the beginning of the century, the high-level technologies lost some ground until 2007. In addition, the effect of the economic crisis is considerably more pronounced in the high-level technologies because non-R&D-intensive technology areas are approximately 5 % higher in internationalisation on this indicator than the high-level technologies. Between 2009 and 2010, the shares decreased slightly in the non-R&D-intensive technology areas, which may imply a convergence of the numbers in the near future.

4.3.3 Position in the Innovation Chain

Citation-based measures were applied to assess the non-R&D-intensive sector’s position in the innovation chain. Citations are provided by either the patent applicant or patent examiner and listed on a patent document. They reflect references that are made to prior art, most commonly to other patents, but also to scientific literature.

Citations can be counted from both forward-looking and backward-looking perspectives. The number of citations a patent receives from subsequent patent filings (i.e., the forward-looking indicator) is commonly referred to as patent forward citations. The basic assumption is that the number of forward citations measures the degree to which a patent contributes to further developing advanced technology and is an indicator of basicness, novelty, or technological significance of a patent in terms of spill-over effects (Carpenter et al. 1981; Trajtenberg 1990). However, patent backward citations refer to previous patents that are mainly used as an indicator of technological breadth and can provide hints on the scope of a patent (Harhoff et al. 2003). Additionally, patent backward citations can be interpreted as a measure of “originality”: Patents with a large number of backward citations can be assumed to build on a larger given pool of already existing knowledge, whereas patents with only a few backward citations have a small existing knowledge stock on which to build (Rosenkopf and Nerkar 2001). In addition to previous patents, patent applicants and patent examiners have the option of citing scientific publications. These references to non-patent literature (NPL) can be used to indicate the closeness of a patent applicants’ R&D activities to science or basic research (Deng et al. 1999). A large number of references made to scientific literature implies that the patent builds upon a comparably large scientifically used knowledge stock.

By building on these indicators of the spill-over potential, originality, and scientific linkage, we gain a deeper understanding of the non-R&D-intensive technology areas compared with high-technology and gain insights on their positioning within the innovation chain.

The average number of forward citations (in a 4-year time window) and backward citations by technology area is presented in Fig. 4.7. The leading-edge technologies are the most highly cited and thus have the highest spill-over potential. In addition, patents within this technology area cite the smallest number of previous patents and can thus be regarded as quite original. Patents from the non-R&D-intensive areas are at the other side of the scale. They build on a rather large pool of existing knowledge as indicated by the patent backward citations, and are cited less frequently (i.e., less subsequent patents are building on those technologies). High-level technologies are located between leading-edge and non-R&D-intensive areas. Although these patents receive a comparably large number of citations on average, they cite a larger number of previous patent applications.

Fig. 4.7
figure 7

Average number of forward and backward citations by technology area, 2006 (Source: EPO – PATSTAT; Fraunhofer ISI calculations)

Full size image

Before we reach a final conclusion regarding the positioning in the innovation chain of the different technology areas, we review the scientific linkage as indicated by the average number of NPL citations, which is shown in Fig. 4.8 and is compared to the number of patent backward citations. Patents from the leading-edge areas are most heavily citing scientific literature and have therefore can be said to have the largest science linkage. The science linkage of the non-R&D-intensive technology areas is rather limited, although approximately three scientific publications are cited in those patents on average. Patents from high-level technologies cite an average of four scientific publications and thus are located in the middle between the leading-edge and non-R&D-intensive technology areas.

Fig. 4.8
figure 8

Average number of NPL citations and backward citations by technology area, 2006 (Source: EPO – PATSTAT; Fraunhofer ISI calculations)

Full size image

In summary, leading-edge areas provide the innovation system with the most upstream-oriented technologies, which are characterised by a close link to science and a high-potential for further inventions to build on the knowledge generated within this area. Thus, these technologies position themselves on an early stage within the innovation chain. Additionally, high-level technologies have a rather high potential for spill-over effects. However, within this area, inventions build on a relatively large pool of existing knowledge and can thus be considered less original in nature. The non-R&D-intensive areas are mainly positioned at the end of the innovation chain (i.e., they provide rather downstream or market-oriented inventions). The patents originating from these areas are less linked to science and are building on rather large pools of existing knowledge.

4. Conclusions

The non-R&D-intensive technology areas are an integral part of the development of research and technology within the world economy. Nearly 85,000 transnational patent filings originated from non-R&D-intensive technologies in the year 2010. However, although the share of patents from the non-R&D-intensive areas has declined slightly over the study period, mainly due to the increasing share of filings from the high-level technologies, patents from the non-R&D-intensive areas still constitute approximately 40 % of all transnational filings. Yet, the size and importance of the non-R&D-intensive technology areas is highly dependent on national idiosyncrasies as well as the industrial structure and field specific specialisations within different countries. Whereas the non-R&D-intensive technology areas play a relatively large role in Europe, they are less important in North America and Asia. Interestingly, the BRICS countries have rather distinct national profiles. Whereas South Africa and Brazil are highly patent-active in the non-R&D-intensive areas, China and India have the lowest shares of patents in non-R&D-intensive technologies and are more highly specialised in high-level and leading-edge technologies.

From a sectoral perspective (i.e., looking at which types of technologies are actually produced by firms from the non-R&D-intensive industries), the non-R&D-intensive industries produce more than only patents that are classified as non-R&D-intensive; approximately 7 % of filings from non-R&D-intensive industries are classified as leading-edge technologies, and 28 % are classified as high-level technologies. Because these shares have increased over time, firms from the non-R&D-intensive industries have increasingly entered the high-technology scene in terms of patenting. In addition, the share of patents in non-R&D-intensive technology areas filed by companies from the high-level technology sectors has increased in the last decade.

Additionally, the differentiation by the type of the patent applicant indicated that although large enterprises are responsible for the largest number of filings in non-R&D-intensive areas, SMEs file the largest number of patents in these areas on average. Academia has the lowest shares of patents in non-R&D-intensive technology areas and is highly focused on leading-edge technologies.

The internationalisation trends reveal that the non-R&D-intensive technology areas are more strongly targeted toward international markets than high-level technologies, although leading-edge technologies are still the most strongly internationally oriented. Regarding the position in the innovation chain, the patent citation indicators demonstrate that inventions generated within the non-R&D-intensive technology areas are less commonly linked to science and generally build on a large pool of existing knowledge. In addition, inventions from the non-R&D-intensive areas are less frequently the basis for further technological developments. Thus, the non-R&D-intensive sectors are largely positioned at the end of the innovation chain (i.e., they provide rather downstream or market-oriented inventions).

In summary, the non-R&D-intensive technology areas are an important and highly dynamic sector at the international scale that is mainly building on the R&D that has been performed in the high-technology areas and typically provides downstream innovations for directly marketable applications.