2

Measuring innovation in the business sector: beyond manufacturing and services

Prepared by: Vitaliy Roud1 and Leonid Gokhberg2

1 Senior researcher, Laboratory for Economics of Innovation, National Research University Higher School of Economics, vroud@hse.ru 2 First vice-rector, National Research University Higher School of Economics, lgokhberg@hse.ru

3

Content: ACRONYMS ................................................................................................................................... 4

1. Rationale for harmonized coverage of sectors beyond manufacturing and services ..... 5

2. Sectoral specificity as a methodological challenge ............................................................ 7

3. Case studies from national practice ................................................................................... 15

3.1 Mining and Utilities in Russia ...................................................................................... 15

3.2 Construction in the UK ................................................................................................ 19

3.3 ‘Low-tech’-services: the hospitality industry in the Netherlands .............................22

3.4 Agriculture: innovation at Australian farms .............................................................. 26

4. Implications for the OM framework and the survey practice ......................................... 29

References ................................................................................................................................... 31

4

ACRONYMS

ABARES Australian Bureau of Agricultural and Resource Economics and Sciences

BERD Business Expenditure on Research and Development

CIS Community Innovation Survey

ICT Information and Communication Technology

KIBS Knowledge-Intensive Business Services

NACE Nomenclature statistique des activités économiques dans la Communauté Européenne

NESTI OECD Working Party of National Experts on Science and Technology Indicators

NSO National Statistical Office

OECD Organisation for Economic Cooperation and Development

OM Oslo Manual

OM2005 Oslo Manual, 2005 (3nd edition)

R&D Research & Development

5

1. Rationale for harmonized coverage of sectors beyond manufacturing and services

Even though the existing revision of the Oslo Manual proposes concepts and definitions formally applicable to the whole business sector, the focus of the framework has been traditionally limited to the ‘core’ industries – manufacturing and selected services. To a certain extent, this is a consequence of the decades of studies that used to focus on the creation and adoption of innovation in these industries. This has strongly affected both academic and policy discussions, often in the form of the exaggerated attention to particular economic activities leading to a so-called ‘high-tech myopia’ effect. The subsequent methodological facilities as well as the existing statistical surveys propose limited means for treating the sectors in a harmonized way.

Such limited analytical practices have contributed to a commonly accepted but distorted perception of a number of other economic sectors, including agriculture, mining, utilities and construction, as those that are not associated with the generation of new knowledge and technology, even though active in adopting and implementing specific types of innovation. However, the evidence is far from being that obvious.

The aggregate statistical indicators demonstrate that for a number of countries, these sectors account for notable proportion of GDP and overall employment (Figure 1). At the same time, the flow of investments aimed at innovation in those sectors is often significant.

Figure 1. Contribution of sectors to GDP and employment (OECD Statistics, 2014 or closest available year)

Austra lia

Austria

Chi le

Czech RepublicDenmark

Estonia

EU-28

Finland

FranceGermany

Greece

Hungary

Ireland

Ita ly

Japan Korea

Latvia

Li thuania

New Zealand

PolandPortugal

Slovak Republic

Slovenia

SpainSwitzerland

Austra lia

Canada

Chi leNorway

Poland

Switzerland

Czech Republic

Estonia

Ireland

Latvia Poland

Austra liaCanada

Chi le

Japan

PolandSlovak Republic

Switzerland

0

5

10

15

20

25

0 5 10 15 20 25

Shar

e in

Labo

r (%

)

Share in GDP (%)

Agriculture Mining Utilities Construction Manufacturing

6

As recognized by the R&D statistics (Figure 2), in the particular economies, the ‘non-core’ sectors can contribute more than 10-15% to the national BERD and account for up to 20% of total R&D personnel employment. This does not include resources of the other industries and other types of institutions allocated for the sake of innovation in agriculture, mining, construction, etc.

Figure 2. Contribution of sectors to BERD and total R&D personnel (OECD Statistics, 2014 or closest available year)

The broader agenda (such as the ongoing debates on the Next Industrial Revolution, the adoption and diffusion of the best-available technologies, challenges of technological governance and technological development, the fostering of the productivity as well as the long-term sustainable economic growth) brings extra rationale to account for innovation beyond the ‘core’. Advanced technologies bring extended momentum of change and novel opportunities to a variety of economic sectors. It is unlikely that the nature and effects of innovation within the economy could be explained without a deeper understanding of the genesis and contents of these sectoral innovations. There are multiple examples of radical production technologies penetrating and transforming processes, market organizations, business models, demand for competences and occupation patterns in all the sectors, such as robotics and automation, smart ICT-driven systems, synthetic biotech, composite materials, additive technologies.

Chile

HungaryAustralia

Canada

ChileChina

Norway

Iceland

Australia

0

5

10

15

20

25

30

35

40

0 5 10 15 20 25 30 35 40

Shar

e in

Tota

l R&

D pe

rson

nel (

%)

Share in BERD (%)

Agriculture Mining Utilities Construction Manufacturing

7

The absorptive capacity of the ‘non-core’ sectors becomes an important driver of demand for innovation generated in other industries. In other words, the scale of innovation activities for a broad range of national economies and the overall assessment of the performance of the national innovation systems remains underestimated.

Even though there is a clear reason to cover those other sectors by regular innovation surveys, certain obstacles exist on this path. The major issues here include the applicability of the established conceptual frameworks as well as the design and organization of the surveys, followed by the proper international harmonization of the best measurement practices. The rest of the paper discusses methodological challenges and cases from the national measurement practice in order to outline possible methods.

2. Sectoral specificity as a methodological challenge

The literature distinguishes between three approaches to the exploration of innovation in the ‘non-core’ sectors3:

x Assimilation assumes that innovation across sectors shares principally the same nature with the original system of concepts centered on technological innovation and manufacturing. This means that new sectors can be studied by directly using the concepts and measurement tools developed for the latter.

x Demarcation implies that innovation in a given sector is highly distinctive and follows unique dynamics, thus requiring a new theoretical underpinning, conceptual framework and measuring facilities.

x Synthesis suggests that the measurement methodology inherits the baseline model and must develop it in order to account for previously neglected sectoral aspects of the innovation process.

These approaches find a balance between total unification and the ability to account for meaningful sectoral variations. One could argue that the ultimate objective for the innovation measurement frameworks is the development of the ‘synthesis’ methodology that effectively supports innovation surveys around all the industries, allowing for the production of comparable indicators but providing enough sector-specific insight. At the same time, studies that follow the ‘assimilation’ approach prove to serve as a reliable reference point, starting the process of integration into the general innovation discussion. The ‘demarcation’ studies support this process by revealing the most important challenges to account for during the expansion of sectoral coverage.

3 Originally formulated for the case of service sector and discussed e.g. in (Coombs & Miles, 2000; den Hertog, Rubalcaba, & Segers, 2008; Vergori, 2014).

8

Pavitt's taxonomy (Pavitt, 1984) being one of the first efforts to address the sectoral specificities of technological change in a systemic way, derived insight from an extensive analysis of case studies across a broad range of industries, including manufacturing, services, mining and agriculture. By classifying the strategies pursued by innovating firms, Pavitt emphasized four classes of sectors and proposed a view on relations between the sectors in terms of sources of new knowledge, technology and the underlying means to implement innovation (such as acquisition of machinery and equipment):

x Science-based sectors that consist of larger companies focused on R&D-driven innovation (intramural and extramural). Firms in this sector develop new products or processes and have a high degree of appropriability from patents, secrecy, and tacit know-how. Sectors included: electronics/electrical machinery, chemicals.

x Specialized suppliers: dominated by smaller, more specialized firms producing

technology to be transferred to other businesses, e.g. specialized machinery production and high-tech instruments. High level of appropriability due to the tacit nature of the knowledge. Sectors included: machinery, precision instruments.

x Scale-intensive: characterized by mainly large firms producing basic materials and

consumer durables, such as the automotive sector. Sources of innovation may be both internal and external to the firm with a medium-level of appropriability. Sectors included: bulk materials (steel, glass), assembly (consumer durables and automobiles).

x Supplier-dominated: firms from mostly traditional manufacturing which rely on

external sources of innovation. Sectors include agriculture, construction, private services, traditional manufacturing.

This taxonomy was the first in the tradition of empirically driven classifications of the sectoral propensities toward different types of innovation strategy. The follow up research (in particular, (Pavitt, Robson, & Townsend, 1989), (Soete & Miozzo, 1989) ) addressed the growing importance of the information and communication technology as well as the heterogeneity of services (originally being equally classified as supplier-dominated). However, the general perception of sectors beyond manufacturing as simple adopters of technology prevailed. A number of studies (e.g. (Cesaratto & Mangano, 1993)) used the data from the newly emerged innovation surveys that followed the original OM framework and provided the first quantitatively driven taxonomies as opposed to the qualitative case-study analyses. However, the source of data imposed a strict limitation on the scope of industries and strategy dimensions, considering only manufacturing and technological innovation. From a theoretical point of view, the revealed sectoral heterogeneity fully complies with a non-linear and multifaceted understanding of innovation processes often referred to as the chain-link model of innovation (Kline & Rosenberg, 1986). Providing enough degrees of freedom, this approach enables the uniform treatment of substantially different innovation strategies (and even remains feasible for the most recent phenomena, such as open innovation).

9

A complementary line of studies explores the factors influencing the sophistication of innovation strategies. Following the seminal contribution of Dosi (Dosi, 1982), a number of authors (e.g. (Breschi, Malerba, & Orsenigo, 2000) and others) elaborated upon the concept of firm-level technological trajectories as a function of endogenous (intra-firm) and exogenous (particularly, sector-specific) determinants: the level of competition, technological opportunities, the structure of the knowledge base, appropriability conditions. This emphasized the ambivalent nature of the sectoral innovation patterns: on the one hand, firms are dependent on the sectoral specificity, and on the other, this specificity shapes as the result of the firms' collective behavior. These studies also allowed Archibugi (Archibugi, 2001) to state that the relationships between sectors are subject to the emergence and advancement of novel technologies (shifts of technological paradigms) and thus each given moment is characterized by its own sectoral taxonomy of innovation. The application of these unified theoretical models (formulated at a high level of abstraction in order to maintain the generality) for the measurement practice was seriously challenged during the expansion of the OM framework towards the service sectors 4 . The widely recognized economic importance of services fostered studies of innovation in services. According to Coombs and Miles (Coombs & Miles, 2000), the original measurement effort inspired by the unified theoretical underpinning followed the assimilation approach, directly applying the technological innovation-related framework to the firms in services. The outcomes (presented by Sirilli & Evangelista (Sirilli & Evangelista, 1998; Evangelista, 2000) and others) were appreciated for bringing new empirical evidence on innovation through the prism of technological change. However, the critics argued 5 that the revealed 'similarity' of manufacturing and service innovation patterns is a consequence of insufficient metrics that fail to capture the unique nature of service innovation. This gave the impetus for a number of studies following an intentional ‘demarcation’ approach to emphasize the specificity of service innovation (as opposed to the production of goods). For example, Coombs & Miles (2000) outline the lower role of formal R&D, the importance of human capital, organization of work, intangibles, the mechanism of consumption (delivery) and the co-creation of service innovations with the clients. The revealed importance of non-technological changes led to a significant shift in the scope of innovation studies from the technology product and process innovation. Addressing the service innovation imposed a methodological challenge and it was successfully met by the integrative effort of the community, resulting in a new edition of the Oslo Manual (OM2005). The renewed methodological framework provided redesigned definitions of innovation as well as clarifications as to how to distinguish between different types of innovation in manufacturing and services, followed by an outline of the theories behind the non-

4 See e.g. (Miles, 2005). 5See (Brouwer & Kleinknecht, 1997), (Miles, 2000), (Hipp & Grupp, 2005) in line with other seminal contributions discussing innovation in services.

10

technological innovation processes. This is often considered an example of implementing the 'synthesis' approach to the sectoral specificity. The practice of measurement developed towards this synthesis of services in a gradual manner. The Community Innovation Survey (CIS) introduced several degrees of services integration since the second round and up to the CIS20086. CIS2 (which addressed activities of firms in 1994-1996) introduced a separate questionnaire for the selected service sectors, but followed the ‘assimilation’ approach. The central concept of the survey was technological innovation and for the services it was defined as those that ‘implemented new or significantly improved services and new or significantly improved ways of producing or delivering a service’ – never using explicitly the words ‘process’, ‘technological innovation’ or ‘technology’. These definitions allowed one to implicitly account for service innovation due to organizational change, however, there were no means to control for organizational innovation in manufacturing.

Box 1. Service innovation and servitization of manufacturing

The methodological achievement of the OM2005 went beyond pure integration of service sectors into the scope of innovation measurement. Providing facilities to capture service innovation in line with the development, production and delivery of new goods technically enabled the application of the framework to most of the sectors of the economy and it allowed one to address complex processes of blurring the intersectoral boundaries, such as servitization of manufacturing (see e.g. Lay, Schroeter, & Biege, 2009; Santamaría, Jesús Nieto, & Miles, 2012).

Innovation that implies the introduction of new services by the producers of goods is not directly captured using the conventional innovation surveys (they both consider product innovation and provide no means to go into detail). However, a special survey design explicitly addressing the manufacturing companies is able to produce indicators of interest. The European Manufacturing Survey (a sample-based survey executed by research centers and universities from 18 countries, coordinated by Fraunhofer ISI, Germany) addresses the list of product-related services, such as design, consulting, R&D for customers; technical documentation, setup, use, service; specialized software development; assembly, initial start-up; training; maintenance/repair; operation (remote operation); leasing, renting, other financial services. A service innovation is defined as offering any of these services for the first time for an enterprise over the last 3 years. In addition, the survey measured the share of the product-related services (and innovative services) in the turnover. These data enabled a series of evidence-based studies on the determinants and impacts of servitization in Europe (see e.g. Gustafsson et al., 2010; Dachs et al., 2014).

This case demonstrates the umbrella-like expansion of the baseline OM concepts. It maintains the consistency with the framework and at the same time provides the required level of specificity to answer the immediate needs of the researchers and policy makers.

See also: http://www.isi.fraunhofer.de/isi-en/i/projekte/fems.php

CIS3 (1998-2000) introduced a unified questionnaire for manufacturing and services. The central concept was ‘innovation’ (as opposed to ‘technological innovation’ earlier), defined as ‘a new or significantly improved product (good or service) introduced to the market or the introduction

6 See e.g. Vergori, 2014.

11

within the enterprise of new or significantly improved process. The innovation is based on the result of new technological developments, new combinations of existing technology or utilisation of other knowledge acquired by the enterprise.’ The non-technological innovation was considered only in terms of contributing to the product and process innovation (through a set of yes-no questions on different organizational and marketing practices). CIS4 (2002-2004) and CIS2006 (2004-2006) introduced definitions of organizational and marketing innovations, while CIS2008 updated the definitions to match the OM2005. Thus, through several waves of the surveys, the measurements for the non-technological innovations moved around the questionnaire. Their role changed from the implicit options to support the definitions of innovation in services to the full-fledged types of innovation relevant for all the covered sectors. The emergence of harmonized data facilitated another wave of taxonomy studies and helped to raise the theoretical discussion on the sectoral patterns to a new level of empirical evidence. Table 1 presents a taxonomy developed by Castellacci (2008) that operates the firm-level technological trajectories in order to construct the joint taxonomy of manufacturing and service sectors through the bottom-up approach (based on the distribution of the specific firm strategies within each sector). These studies prove that the perception of sectors as homogenous in terms of innovation patterns is misleading. There are examples of highly innovative firms all through the economy regardless of the expected sectoral labels.

Figure 3. Sectoral coverage of business sectors by national innovation surveys (based on OECD NESTI Innovation Metadata Survey, 2013).

The success of the OM2005 and of the corresponding surveys resulted in the implicit consensus that the developed frameworks and measurement tools are fully applicable to a range of sectors broader than those discussed in the Oslo Manual. The community was encouraged to experiment with measuring innovation in the sectors beyond manufacturing

22.9

94.3

100

97.1

54.3

88.6

42.9

0 20 40 60 80 100

Agriculture

Mining

Manufacturing

Utilities

Construction

Knowledge-intensive business services

Other services

% of countries

12

and services. According to a (very limited) number of studies, this national practice is rich but poorly discussed. The OECD NESTI Innovation Survey Metadata analysis (2013) reveals that of 35 countries under consideration, 23% cover agriculture, 94% cover the mining and quarrying sector, 97% observe utilities within the framework of national innovation surveys. Since then, national practice emerged even further. Particularly, most of these sectors are now addressed by the European CIS. Peneder (2010) exploits the data on innovation in Mining and Utilities from several EU countries in order to create a general taxonomy that equally treats the innovation intensity in the ‘traditional’ and new sectors (see Table 2). However, in the absence of the harmonized sectoral frameworks, the national practice of measuring innovation in these sectors has to follow one of two options. One is implementing the ‘assimilation’ approach, which often implies the straightforward distribution of the unified questionnaire among the enterprises from the new sectors. As discussed above, the OM framework per se appears to be suitable for accounting for innovation in any sector of commercial activity. However, using the general definitions of innovation imposes the additional burden of interpretation on the respondents, thus lowering the motivation for proper data provision and decreasing the quality of the resulting information.

Another option is to spend considerable effort and develop a proper adaptation of the survey methodology. This practice is honorable, however, beyond the excessive burden on national statistical offices, this leads to limited global exposure of the produced indicators. There is a number of analytical reports and academic papers addressing innovation in mining, agriculture, utilities, etc. within the OM-compatible innovation survey framework is scarce compared to the evidence collected on manufacturing and services.

The next section introduces several cases from national measurement practice that follow either of these approaches.

Table 1. Data-driven taxonomy of sectoral patterns of innovation in manufacturing and service industries (Castellacci, 2008)

Sectoral category Sectors Technological regimes Technological trajectories

Personal goods and services

Supplier-dominated goods

Food and beverages; textiles; clothing; leather; wood and related products; pulp and paper; printing and publishing; furniture; recycling

Opportunity levels: medium External sources: suppliers and end users Appropriability: trademarks; design; know-how Dominant firm size: SMEs

Type of innovation: process Innovation expenditures and strategy: acquisition of machinery

Supplier-dominated services

Sales, maintenance and repair of motor vehicles; retail trade and repair of personal and household goods; hotels and restaurants

Opportunity levels: low External sources: suppliers Appropriability: non-technical means Dominant firm size: SMEs

Type of innovation: process Innovation expenditures and strategy: acquisition of machinery

Supporting infrastructure

services

Physical infrastructure Wholesale trade and commission trade; land, water and air transport; supporting and auxiliary transport activities

Opportunity levels: medium External sources: suppliers and users Appropriability: standards; norms; design Dominant firm size: large

Type of innovation: mixed process, service and organizational innovation Innovation expenditures and strategy: R&D; acquisition of software; training

Network infrastructure

Post and telecommunications; financial intermediation; insurance and pension funding; activities auxiliary to financial intermediation

Opportunity levels: low External sources: suppliers Appropriability: standards; norms; design Dominant firm size: large

Type of innovation: process Innovation expenditures and strategy: acquisition of machinery and software

Supporting infrastructure

services

Science-based manufacturing

Chemicals; office machinery and computers; electrical machinery and apparatus; radio, TV and communication equipment

Opportunity levels: high External sources: universities and users Appropriability: patents; design; copyright Dominant firm size: large

Type of innovation: new products; organizational innovation Innovation expenditures and strategy: R&D; cooperation

Scale-intensive manufacturing

Rubber and plastic products; other non-metallic mineral products; basic metals; fabricated metal products; motor vehicles; other transport equipment

Opportunity levels: medium External sources: suppliers and users Appropriability: design; process secrecy Dominant firm size: large

Type of innovation: mixed products and process innovation Innovation expenditures and strategy: R&D; acquisition of machinery

Advanced knowledge providers

Knowledge-intensive business services

Computer and related activities; research and development; other business activities

Opportunity levels: very high External sources: users and universities Appropriability: know-how; copyright Dominant firm size: SMEs

Type of innovation: new services; organizational innovation Innovation expenditures and strategy: R&D; training; cooperation

Specialized suppliers,manufacturing

Machinery and equipment; medical, precision and optical instruments

Opportunity levels: high External sources: users Appropriability: patents; design know-how Dominant firm size: SMEs

Type of innovation: new products Innovation expenditures and strategy: R&D; acquisition of machinery; software purchase

14

Table 2. Classification of sectors by innovation intensity based on unified CIS data: Mining, Manufacturing, Utilities, Services (Peneder, 2010)

Sectoral category Description of technological regimes based on clustering results Sectors

High innovation intensity

High share of creative firms focused on product innovations (either alone or in combination with process innovations) and many firms performing high intramural R&D. Typically, the appropriability regime depends on the use of patents (frequently applied together with other measures) and knowledge is highly cumulative.

Machinery, n.e.c..; Computers, office machinery; Electrical equipment, n.e.c.; Communication technology; Precision instruments; Computer services; Research and development

Intermediate-to-high innovation intensity

Intermediate share of creative firms mostly involved in process innovations, and many firms performing R&D, high share of creative firms with product innovations with a strong dependence on the external acquisition of new technology. Cumulativeness of knowledge is high or intermediate and firms frequently use patents for appropriation.

Textiles; Ref. petroleum, nuclear fuel; Chemicals; Rubber and plastics; Mineral products; Basic metals; Motor vehicles, parts; Other transport equipment; Post, telecommunications

Intermediate innovation intensity

Sectors with large number of firms pursuing opportunities through the acquisition of external innovations. Appropriability measures are relatively weak, with some importance ascribed to strategic means.

Wood, -products, cork; Pulp/paper, -products; Fabricated metal products; Manufacturing nec.; Air transport; Financial intermediation; Other business services

Intermediate-to-low innovation intensity

The main characteristic of this group is the high share of firms with adaptive behavior, pursuing opportunities through the adoption of new technology. Accordingly, the prevalent mode of innovation is the acquisition of new technology. For most firms, appropriability conditions are weak and the cumulativeness of knowledge is low.

Mining: coal, peat; Mining: petroleum, gas; Food products, beverages; Tobacco products; Publishing, reproduction; Electricity and gas; Water supply; Insurance, pension funding

Low innovation intensity

A relatively homogenous group is characterized by a predominance of firms pursuing opportunities other than from new technologies, typically performing no innovation activities nor applying any measures for appropriation. For the majority of firms, the cumulativeness of knowledge is low or irrelevant, since no information regarding innovation is pursued.

Mining: other; Clothing apparel, fur; Leather, -products, footwear; Recycling; Wholesale trade; Land transport, pipelines; Water transport; Auxiliary transport services; Auxiliary financial services

3. Case studies from national practice

3.1 Mining and Utilities in Russia Innovation surveys traditionally used the assimilation approach to address mining and utilities. These sectors heavily rely on technological innovation and often benefit from formal R&D. As in manufacturing, the production of tangible goods generally comprises the core component of the business model. This makes the original technological product/process innovation framework relevant as well as the later revisions. Thus, many statistical offices initiated the measurement of innovation in mining and utilities starting from the first innovation surveys. However, the results of these exercises are poorly communicated and discussed. Unlike manufacturing and services, no harmonized indicators are widely circulated. The empirical research work on innovation in mining and utilities is scarce and these sectors are outside the focus of emerging cognitive testing studies. This section discusses insights from the Russian innovation surveys that account for mining and utilities since their launch in 1994 (see Gokhberg & Kuznetsova, 1999). Designed as an annual and mandatory statistical survey targeting medium and large enterprises, the contemporary survey provides data on the innovation activities for roughly 1,600 mining enterprises and more than 9,000 utilities enterprises. The survey uses a unified questionnaire for all the sectors accompanied by extended guidelines that clarify the definitions of key concepts and provide instructions on constructing measures based on conventional statistics. The specificity of mining and utilities is addressed by providing relevant examples of innovation in the supplementary instructions. The assimilation approach allows one to construct all the conventional statistical indicators of innovation input and output (expressed for example as a share of innovation-active companies, a share of innovation sales in total sales, a share of innovation expenditure in total sales, see Figure 4). The intensity indicators position mining and quarrying slightly above low-tech manufacturing. Utilities lag behind all other sectors in terms of engagement in innovation and innovation output but appear to have a higher share of innovation-related expenditure in their turnover compared to mining and low-tech manufacturing. In terms of innovation expenditure priorities (Figure 5), mining follows a similar strategy to medium low to medium high-tech manufacturing. Enterprises in mining are characterized by a higher share of R&D, while the dominant component of the innovation strategy remains the purchase of the machinery and equipment. Utilities stand in line with the low-tech manufacturing, also accounting for a higher share of the R&D expenditure.

16

The sectors have an explicit preference for specific types of innovation (Figure 6). Both mining and utilities have the strongest focus on process innovation. New products are of less importance as opposed to manufacturing industries. Interestingly, there is much less sectoral specificity in terms of the propensity toward organizational innovation. Russian innovation survey includes a set of open questions to collect examples of the most important innovations. The provided answers allow for studying questionnaire comprehension. Table 3 presents examples of the derived innovations. In line with the findings of the cognitive testing for the manufacturing sector, the respondents in mining and quarrying easily identify product and process innovations. As for the most significant inconsistencies, the respondents tend to classify geological exploration and other procedures that accompany the launch of new deposits as product innovation. Utilities introduce other complications as the production of goods and delivery of services is often equally important for the firms. This brings certain difficulties in classifying product, process and organizational innovations, especially subject to environmental protection initiatives or IT-enabled advancements, such as automated tariff and billing systems, online payment and control. These cases of confusion are to be addressed in the updated questionnaires and guidelines. Broader dissemination of the national cognitive testing outcomes in these areas could save the community's effort and promote the data quality and comparability.

Figure 4. Innovation activities (% of total companies) in Mining and Utilities vs. Manufacturing (HSE, 2015)

6.5 6.5

10.9

17.4

30.6

4.5

7.2

4.1

8.8

14.8

17.7

0.71.4 0.7

2.6 2.4

6.1

1.8

0

5

10

15

20

25

30

35

Mining Low Tech Medium Low Tech Medium High Tech High Tech Utilities

Share of innovation-active firms Share of innovation sales in total sales Share of innovation expenditure in total sales

17

Figure 5. Composition of innovation expenditure by innovation activity (percentage) in Mining and Utilities vs.

Manufacturing (HSE, 2015)

Figure 6. Types of innovation in Mining and Utilities vs. Manufacturing (HSE, 2015)

17.46.3

20.8 21

39.8

9.3

1

10.1

8.8 8.2

8.9

3.2

64.6 70.6 48.6

37.2

34.1

67.4

8.5

4.79.8 3

9.710.9

6 9.15.5 8.5 10.9 12.7

5.1 1.6

Mining Low Tech Medium Low Tech Medium High Tech High Tech Utilities

R&D Design and development Purchase of machinery and equipmentNew technologies Acquisition of software Other start-upTraining Marketing Other

25.7

58.455.8

74.877.9

24.4

81

54.8

63

46.4

52.2

82.6

35.4

26.331.1 29.7 29.4

33.3

0

10

20

30

40

50

60

70

80

90

Mining Low Tech Medium Low Tech Medium High Tech High Tech Utilities

Product innovation Process innovation Organisational innovation

Share (%) of innovation-active companies that introduced:

18

Table 3. Examples of innovation in Mining and Utilities (Russian innovation survey-2015)

Mining and Quarrying Utilities Product Examples:

Cubiform crushed stone New grade of sweet crude oil Short-grain coke Production of hexan-isopentan-pentan fraction Coal slurry pipeline Common errors: Development of a new site as an innovation project with the consequent classification of all costs as an innovation expenditure Misclassified process innovations – adoption of energy and resource saving technologies, ICT-enabled mobile monitoring systems

Examples: Gas pipeline to a new location Polythene pipes-enabled outdoor water distribution Common errors: Misclassified process innovations: increased output, energy and resource efficiency Misclassified organizational innovations: accounting software Misclassified incomplete innovations: design and construction documentation

Process Examples: Zero-waste technology for iron-ore concentrate production Fuel control system Sorbate desalination facility Increased oil recovery technology Thermal stabilization of grounds using bored piles Automatic system to control volumes of production depending on the market demand Common errors: Misclassified organizational innovations: outsourcing of accounting and ICT Misclassified product innovations (e.g. the development of products with significantly improved qualities) Repairs and reconstruction Development in progress as opposed to implemented innovations.

Examples: Automation of power grid management, tele-automation Modernized combined cycle gas turbine Automated sewage monitoring and control system Queuing system at the client offices GPS-enabled vehicle control system Common errors: Misclassified organizational innovation: accounting software Misclassification of new services as processes (e.g. online payments). Misclassified upgrading of ICT tools: Modernized office equipment and computers

Organizational Examples: Complex asset management Enterprise resource planning system Team-based supervisory control system Integrated quality management system (ISO 9001, 14001) Lean production New scheme of watch-based labor organization Common errors: Business expansion, launch of new sites

Examples: Outsourcing of ICT services Individual staff training program Enterprise resource planning system Certification of services Common errors: Misclassified marketing innovations: Internet web-site development, upgrading of office machinery.

Marketing Examples: Corporate portal, website Electronic trade platforms Entry to new geographical markets Common errors: Misclassified process or organizational innovations.

Examples: Corporate portal Mobile-based information for clients New ergonomic packaging for gas Flexible discount and delayed payment systems Common errors: Misclassified process and organizational innovations: e.g. integrated environment control system, ICT-outsourcing

19

3.2 Construction in the UK Innovation in construction is significantly underexplored even though for many countries this sector accounts for a high share of GDP and provides considerable input for overall growth (also by creating capital goods for other sectors). The intensity of change in practices, materials, technologies over the past several years is outstanding, at the same time few studies attempt to approach these processes using the systemic innovation frameworks. The UK Innovation Survey provided the basis for several studies of this sector (Reichstein, Salter, & Gann, 2005; Reichstein, Salter, & Gann, 2008). The studies mention the limitations of the assimilation approach but choose to neglect the possible inconsistencies and compare the innovation performance of construction against other sectors using harmonized indicators.

A cross-sectoral comparison of innovation activity positions construction below all sub-sectors of manufacturing and services (Figure 7). Innovation strategies have an explicit specificity regarding the sources of information (Figure 8). The regression results indicate a high reliance of product innovation on customers and clients. Process innovation draws upon the technical standards, health and safety regulation, environmental standards and others. Suppliers facilitate both types of innovation. The companies very rarely engage in formal R&D.

These indicators of lower intensity comply with the existing theoretical underpinning. The literature proposes a number of explanations for the poor innovation performance of the construction sector including longer product lifecycles (estimated at more than 30 years on average) and the geographical immobility of the final goods as factors greatly hampering propensity toward innovation.

However, the assimilation approach does not allow one to differentiate between the ‘true’ lag behind other sectors and the systemic undervaluation of the creative effort. Table 4 presents examples of product and process innovations derived from the UK Innovation Survey. While the list of process innovations is rather solid, there is less coherence in the proposed product innovations. Reichstein et al. (2005) emphasize the project-based nature of sectoral business models that limit the direct application of a product innovation definition and in general challenges the concept of ‘implemented’ innovation.

The unified definitions do not support the respondents in a number of complex situations. For example, the adoption of new materials is a common innovation strategy. However, depending on the case, the resulting innovation could be classified either as process or as product innovation: consider, for example, the adoption of light and durable roof coating that is easy to lay as opposed to coating the roof with solar cells. From the firm’s point of view, both of the cases are about choosing the supplier of the materials and providing training to the workers. At the same time, the framework considers these innovations to be fundamentally different.

20

Table 4. Examples of innovation in construction (Reichstein, Salter, & Gann, 2005)

Product innovation Process innovation

Development of a composite fire door Electronic communications with clients for exchange of data

Lancing table for improved heat exchange during the cleaning process

CAD and electronic data links with some of our customers

Conditioning and monitoring systems for the railway points system Automatic delivery of concrete

External solar shading for new buildings Establishment of intranet for knowledge exchange, etc.

Square dill bits - new advanced mains boards More modern woodworking machinery

Development of a ‘‘sobo’’ system for a specific application in the manufacturing industry

Computerized systems - computerized timesheets stock control

Multiple temperature cabinet built into wall and house to receive house delivery Work identification and control processes

Installation of new structural lining within failed underground structure Asset-based maintenance management

Implementing lean manufacturing quality procedures in construction

Direct cost control system with integrated design, buying, invoicing, processing

Design and construction of welding machine We introduced a new manufacturing line and manufactured more components in-house

Figure 7. Innovation in construction compared to other sectors (Reichstein, Salter, & Gann, 2005)

16.2

33.4

30.3

47.6

14.7

40.7

23.3

36.9

9.1

14.4

17.2

31

17.5

36.5

10.7

18.8

18.2

26.2

8.2

16.6

0 5 10 15 20 25 30 35 40 45 50

Size <50

Size >=50

Size <50

Size >=50

Size <50

Size >=50

Size <50

Size >=50

Size <50

Size >=50

Low

-Tec

hM

anuf

actu

ring

High

-Tec

hM

anuf

actu

ring

Trad

ition

alSe

rvic

es

Know

ledg

e-In

tens

ive

Serv

ices

Cons

truc

tion

Product innovation Process innovation

21

Figure 8. Sources of information in the construction sector compared to other industries (Reichstein, Salter, & Gann, 2005)

63.3

35.8

64.9

62.2

50

34.5

22.4

22.4

15.5

21.9

16.5

37.5

45.9

48.5

53.5

51.6

59.8

56.6

78

49.8

75.2

74.9

62.5

46.2

37.1

38.2

23.6

29.1

22.2

51.6

53.1

64.2

68

69.9

71.2

69.5

51.3

30.9

52.4

48.6

42.2

33.6

11.5

10.7

9.7

16.2

9.8

34.4

41.4

42.3

37.6

35.7

42.8

39.4

60.9

31.4

58.4

53.2

45.3

45

15.2

22.4

15.9

22

14.7

56.9

39.4

49

42.4

42.4

34.7

31.4

41.9

22.8

48.4

45.9

34.3

35.8

13.4

16.6

14.5

20.7

12.4

33.6

40.2

41.5

31.1

45

52.1

48.3

0 10 20 30 40 50 60 70 80 90

Within enterprise

Other enterprises within the enterprise group

Suppliers of equipment, materials, components or software

Clients of customers

Competitors

Consultants

Commercial laboratories/R&D enterprises

Universities or other higher education institutes

Government research organisations

Other public sector e.g. business links, government offices

Private research institutes

Professional conferences, meetings

Trade associations

Technical/trade press, computer databases

Fairs, exhibitions

Technical standards

Health and safety standards and regulations

Environmental standards and regulations

Inte

rnal

Mar

ket

Inst

itutio

nal

Oth

erSp

ecia

lised

% of enterprises

Low-Tech Manufacturing High-Tech Manufacturing Traditional Services KIS Construction

22

3.3 ‘Low-tech’-services: the hospitality industry in the Netherlands

Although OM rev. 3 provided the general framework for measuring innovation in services, the diversity of activities within the service sector sometimes imposes an excessive interpretation burden on the respondents from particular industries and hampers the process of data collection. Den Hertog, Gallouj, Segers (2011) discuss this using the case of the hospitality industry in Netherlands.

The Dutch Innovation Survey (2004-2006) presents the hospitality sector as the least innovative among all the services. Only 9% of firms reported technological innovation, 11% of all firms introduced non-technological innovation over the 2004–2006 period (compared with 24% of all services), of them 10% of all hospitality firms report organizational innovations (compared with 22% for all services) and 3% marketing innovations (compared with 9% for all services).

The authors addressed the specificity of innovation process with a highly tailored demarcation approach. The initial construction of an innovation concept that is acceptable to the horeca (hotels, restaurants, cafes) firms used 14 deep interviews to identify core areas of creative effort (Table 5). The follow-up survey collected 613 filled-in questionnaires. The data was weighted to represent the population of horeca companies. Firms indicated the amount of creative effort they spent over the various aspects of main and supportive business processes (Figures 9, 10). Tailored concepts help to reveal much higher innovation activity than captured by the unified survey. The authors proceed with outlining the specificity of the innovation strategy (Figure 11), cooperation patterns (Figure 12), the impact of innovation (Figure 13). The proposed operationalization of innovation effort demonstrates the amount of undervalued creativity caused by excessive focus on technological innovation in services. In this line, there is also another important illustration. The horeca companies indicate the difference between innovating by new goods and innovating by new services (while the first is pretty much connected with technological innovation, the second is mostly non-technological). That means that the delivery of new goods and new services is characteristic of all business enterprises regardless of sectors, in reality mixing the two components of product innovation. The study proposes detailed instruments to capture innovation in the horeca segment. The tailored concepts and the special design of the questionnaire reduces the burden of interpretation to a minimum. Thus, the higher quality of data helps to explore service innovation in great detail. However, the proposed categories are inconsistent with the OM2005 framework – an understandable limitation of ‘demarcation’ studies.

23

Table 5. Key innovation targets in the hospitability industry (den Hertog, Gallouj, & Segers, 2011).

Primary business processes Supporting business processes

Service formula or concept (e.g. a new franchise formula or a new highly customized service approach)

Day-to-day operational management (e.g. innovative ways of management)

The service interaction level offered or service experience offered to the guest (e.g. highly personalized services or electronic reservation system)

Procurement or supply management (e.g. electronic ordering)

Assortment of products and services directly related to food, drinks and sleeping (e.g. menu engineering or the choice between various types of pillows at a hotel)

Marketing and sales management (e.g. channel management and loyalty programs)

Serviscape or location/building and how it is decorated (e.g. outlets in unexpected locations or the look and feel of a particular café, restaurant or hotel such as a designer hotel)

Accounting (e.g. the use of advanced cash registers or administrative software)

The equipment or technology used in the primary production process (e.g. remote ordering devices on terraces or the latest kitchen technology)

Human resources management (e.g. training of personnel for new type of functions)

The actual primary service production in the area of food, drinks and sleeping (e.g. innovative approaches to keeping a location tidy and clean or the way food and drinks are prepared)

Use of innovative equipment/technology in generic supporting activities

Figure 9. Innovation on six key aspects of the primary business process in the Dutch hospitality industry during 2002– 2004, by level of innovativeness (den Hertog, Gallouj, & Segers, 2011)

38

46

54

62

62

66

43

29

20

22

21

17

13

16

14

10

12

6

6

9

12

6

5

11

0 10 20 30 40 50 60 70 80 90 100

Assortment

Technological innovation

Serviscape

Organisation primary service production

Service interaction level

Service concept

Not innovated Somewhat innovated Considerably innovated Completely innovated

24

Figure 10. Innovation on six key supporting business processes in the Dutch hospitality industry during 2002– 2004, by level of innovativeness (den Hertog, Gallouj, & Segers, 2011)

Figure 11. Practices used to realize innovations over the period 2002– 2004, by sub-sector (den Hertog, Gallouj, & Segers, 2011)

49

54

64

65

65

69

32

27

24

16

20

17

10

14

9

12

11

9

9

5

3

7

4

5

0 10 20 30 40 50 60 70 80 90 100

Technological options secondary processes

HR management

Marketing and sales

Administration

Purchasing

Operational management

Not innovated Somewhat innovated Considerably innovated Completely innovated

93

70

50 52

16

4

96

64

39

48

15

91

69

56

48

16

7

92

75

5659

16

95

76

61 61

24

8

0

10

20

30

40

50

60

70

80

90

100

Own initiative Asked collaboratorsto contribute

Made collaboratorsresponsible

Made plan andarranged budget

Bought specializedknowledge on the

market

Other

% o

f ent

erpr

ises

All Horeca Café Fastfood Restaurant Hotel

25

Figure 12. Co-operation with third parties to realize innovation in the Dutch hospitality industry over the period of 2002-2004, by innovation intensity (den Hertog, Gallouj, & Segers, 2011)

Figure 13. Share of Dutch hospitality firms reporting various types of impacts from innovation in their industry introduced over 2002-2004 period, by sub-sector (den Hertog, Gallouj, & Segers, 2011).

1817

12

76

43

9

13

6

3 32 2

23

20

16

89

54

31

18

1615

7

3

1

0

5

10

15

20

25

30

35

Suppliers Bank,accountants and

other advisors

Other firms inindustry

Firms in otherindustries

Industryassociations

Headoffice, sameconcern

Other partners

% o

f ent

erpr

ises

All Horeca Low innovative Medium innovative Highly innovative

69

60

53 52 50

40

70

62

56 55

45 44

6459

46 48 48 46

70

57 55

48

60

36

80

63

50

61

50

23

0

10

20

30

40

50

60

70

80

90

100

Qualityimprovements

Meet regulationbetter

Welcome newcategory guests

Higher turnover Lower costs Higher capacity

% o

f ent

erpr

ises

All Horeca Café Fastfood Restaurant Hotel

26

3.4 Agriculture: innovation at Australian farms In the first place, agriculture challenges the unified OM-based questionnaires with the inconsistent terminology. It is very unlikely that the firms engaged in agricultural production would be robust at differentiating their innovations between the conventional types of product/process/organizational innovation without any guidance. The next layer of complication implies the exclusion of regularly repeated practices (e.g. annual shifts in the crop choice at the fields in order to maintain fertility), the complicated criteria of ‘successful’ innovation (delayed effects caused by the fuzzy boundaries of production cycles and the complexity of the living systems). The high dependence on the external knowledge base hampers the calculation of the costs of innovation, especially measuring the R&D component, which is usually done outside of the sector (by universities, public organizations, biotech companies, etc.)

A notable example of a successful methodological effort in this area is presented in (Nossal & Lim, 2011; Nossal & Sheng, 2010). These papers describe the outcomes of the exercise undertaken in 2008 by the Australian Bureau of Agricultural and Resource Economics and Sciences (ABARES). Employing the OM2005 as a framework, ABARES executed the survey within the system of existing surveys that cover farming on the regular basis. The farm innovation study was implemented as a one-time exercise independent from the ongoing innovation survey in manufacturing and services (which is closely OM- and CIS-based). This allowed a full translation of innovation-related concepts to the agriculture-specific terminology.

One of the project’s key methodological achievements is the detailed umbrella-like specification of the definitions for all four major types of innovation discussed in the OM2005 (product, process, organizational and marketing innovation, Table 6). The resulting definitions provide a useful specification of the general innovation concepts relevant to agricultural production. This type of specification provides clear guidelines for identifying the occurrence of innovation at the agricultural enterprise (or a farm).

The survey produced indicators of innovation activity, revealing a high propensity toward process and product innovation in Australian farming (Figure 7). However, the survey design provided fewer metrics for evaluating the outcomes or inputs of the innovation process, mainly focusing on the qualitative registration of the innovation activities. It was proposed that the farms estimate the extent of the effort they invested in the particular types of innovation based on a 3-point scale (innovations were adopted: not at all, to some extent, to a great extent, Table 7). This approach avoided a discussion of innovation implementation for the sake of emphasizing the innovation activity within the agricultural sector using detailed indicators and producing the knowledge that meets the demand of both of the sectoral specialists (e.g., it is possible to identify the focus of innovation activities of the farmers) and of the community interested in general innovation dynamics.

27

Table 5. Umbrella-like definitions of innovation in agriculture (Nossal, Lim, 2011)

Product innovation

New crop types New crop cultivars New livestock types New livestock breeds

Process innovation

Cropping equipment Fertilizer practice Weed, pest and disease management practices Soil management practice Weed-related natural resource management Pest-related natural resource management Soil-related natural resource management Other crop practices Livestock feeding practice Livestock handling practice Livestock health practice Grazing management practice Other livestock practices Pasture type

Irrigation and water management practices

Organisational innovations New approach to labor use New members for farm management

Marketing innovation New approach to marketing farm’s products

Figure 7. Proportion of farms innovating, by innovation type (Nossal, 2011)

0

10

20

30

40

50

60

70

80

Cropping Mixed Beef Sheep Sheep-beef Total broadcare Dairy

% o

f far

ms

Product Process Organisational Marketing

28

Table 6. Extent of innovation adoption, broadcare and diary farms in Australia, ABARES, 2008 (Nossal, 2011)

Broadcare farms Dairy farms

not at all

to some

extent

to a great

extent

not at all

to some

extent

to a great

extent

Product innovation New crop types % 78 18 3 67 26 7

New crop cultivars % 71 26 3 79 18 3

New livestock types % 90 8 2 94 6 0

New livestock breeds % 80 16 4 87 7 6

Process innovation Natural resource management

Weed-related NRM % 78 18 4 68 30 2

Pest-related NRM % 85 12 3 75 25 0

Soil-related NRM % 77 18 5 72 23 5

Cropping

Fertilizer practice % 77 17 6 66 20 15

Soil management practice % 71 23 6 na na na

Weed, pest and disease management practices % 74 22 4 na na na

Equipment for cultivation, planting, fertilizing, spraying and harvesting

% 66 26 8 na na na

Other crop practices % 92 6 2 85 8 7

Livestock

Livestock feeding practice % 83 14 3 66 18 16

Fodder conversation and use practice % 89 10 1 58 25 17

Livestock handling practice % 87 10 3 89 8 3

Livestock health practice % 85 12 3 77 22 1

Grazing management practice % 84 13 3 71 23 6

Other livestock practices % 93 5 1 94 6 0

Pasture type % 81 16 3 63 26 11

Irrigation

Irrigation and water management practices % 79 15 6 58 18 24

Organisational innovation

New approach to labor use % 82 14 4 76 21 2

New members to farm management % 92 6 3 91 6 3

Marketing innovation

New approach to marketing farm’s production

% 75 19 6 90 5 5

29

4. Implications for the OM framework and the survey practice

There is a clear rationale for including sectors beyond manufacturing and services into the harmonized framework of innovation surveys. Growing demand for these indicators is partly satisfied with the emerging national practice, although the results of these exercises are not properly disseminated and discussed. The absence of the relevant guidelines in the OM2005 imposes excessive burden on the national statistical offices willing to expand the scope of statistical observation and limits the international comparability of results. Evidence of the latter is the low number of analytical reports and academic papers addressing the issues of innovation in the ‘non- core’ sectors through the prism of the innovation surveys. Key implications for the OM development and for the survey design:

(1) The OM framework provides enough conceptual power to account for innovation in the whole business sector (and has the potential to expand further, i.e., towards public services). This was enabled by adopting the service innovation concept as a complement to technological product innovation. The combined system of definitions that distinguishes innovation in both that provide tangible goods and services makes virtually all industries accountable.

(2) However, the generality of the framework imposes an excessive burden on the respondents when applied without the necessary adjustments. This decreases the robustness of data collected and hampers the comparability.

(3) The ultimate objective of the OM development is to provide definitions that on the one hand provide enough generality and enable cross-sectoral and cross-country comparisons, and on the other hand are able to capture the important sectoral specificities thus not decreasing the value of information collected.

(4) In the survey practice this can be implemented using a balance between assimilation (the ultimate unification of the questionnaires) and demarcation (a high extent of specificity) approaches. This process can be gradual and iterative, combining the results of large-scale surveys with specialized studies.

(5) For a number of sectors, the concepts proposed in the OM are fully applicable on the condition that they are translated into the appropriate language. Addressing particular sectors (e.g. agriculture, low-tech services) with the general definitions is counter-productive. Constructing an umbrella-like system of definitions that are based on the relevant terminology is an effective strategy in this case.

(6) The lack of an explicit discussion of the sectoral specifics (beyond manufacturing and general services) is one of the weakest points of the existing OM. A proposed solution could be to follow the best practice of the Frascati Manual (2015) that introduced the section 2.7 “Examples of R&D, boundaries and exclusions in different areas”. This section explicitly addresses complex cases of defining R&D, e.g., in Arts, in Software Development, in Education and Training, in Services, etc. In the updated OM, a similar section could provide a harmonized view on the most common areas of methodological complexity, thus presenting the best-available practice of conceptualization and fostering the comparability of data produced through the surveys.

30

(7) A specified set of definitions of innovation would not solve all the issues of intersectoral differences. The systemic and synthesis effort would imply a careful analysis of existing data (for example, the answers to open questions about the most important innovations introduced in various sectors), as well as undertaking serious theoretical work. A study (potentially done using the demarcation approach) would help determine the importance of the following factors for making the framework fully applicable:

a. Specificities of products, services and processes and the underlying technological base (e.g., a stronger dependence on external factors such as environmental conditions and the nature of living systems that result in risk redistribution which together with longer product lifecycles alter the understanding of sector-specific successful innovation);

b. Specificities in defining innovation activities and the expenditure on innovation (certain sectors, like agriculture, imply repetitive practices which comply with the formal definition of innovation activity but are not connected with the development of novelties or other types of organizational change), as well as knowledge production/dissemination patterns (institutional locus of formal R&D, the impact of the appropriability conditions on innovation-related rent, typical schemes of cooperation, etc.);

c. Business models and industrial organization (heterogeneity of industry structures, dominant sizes of actors, role of regulatory frameworks, and presence of the state).

(8) The most successful national measurement exercises follow the umbrella-like specification of these concepts, by developing tailored and clarified definitions suitable for capturing the innovation activities in the particular sectors. The OM framework could support these activities by emphasizing this approach explicitly and providing umbrella-like general definitions accompanied by sector-specific additional details suitable for further adoption. Particular options for implementing the sectoral surveys may include either generally universal cross-sectoral designs with tailored methodologies or highly specialized initiatives and the intermediate options implying modular questionnaires.

(9) The interpretation burden is the key factor hampering data collection. The respondents could greatly benefit from the explicit lists of sectoral examples of all the innovation types formulated using the terminology from the sectors. This reference material can improve the quality of data collection and the degree of harmonization even if the questionnaire remains unified.

(10) The appropriate input for this process is a full-scale stocktaking Eurostat/OECD exercise to review the national practices, generalize the methodological findings and cognitive testing results across countries and sectors.

31

References

Archibugi, D. (2001). Pavitt’s taxonomy sixteen years on: a review article. Economics of Innovation and New Technology, 10(5), 415–425.

Breschi, S., Malerba, F., & Orsenigo, L. (2000). Technological regimes and Schumpeterian patterns of innovation. Economic Journal, 110(463), 388–410.

Brouwer, E., & Kleinknecht, A. (1997). Measuring the unmeasurable: a country’s non-R&D expenditure on product and service innovation. Research Policy, 25(8), 1235–1242.

Castellacci, F. (2008). Technological paradigms, regimes and trajectories: Manufacturing and service industries in a new taxonomy of sectoral patterns of innovation. Research Policy, 37(6–7), 978–994.

Cesaratto, S., & Mangano, S. (1993). Technological profiles and economic performance in the Italian manufacturing sector. Economics of Innovation and New Technology, 2(3), 237–256.

Coombs, R., & Miles, I. (2000). Innovation, measurement and services: the new problematique. In Innovation systems in the service economy (pp. 85–103). Springer.

Dachs, B., Biege, S., Borowiecki, M., Lay, G., Jäger, A., & Schartinger, D. (2014). Servitisation of European manufacturing: evidence from a large scale database. The Service Industries Journal, 34(1), 5–23.

den Hertog, P., Gallouj, F., & Segers, J. (2011). Measuring innovation in a ‘low-tech’service industry: the case of the Dutch hospitality industry. The Service Industries Journal, 31(9), 1429–1449.

den Hertog, P., Rubalcaba, L., & Segers, J. (2008). Is there a rationale for services R&D and innovation policies? International Journal of Services Technology and Management, 9(3–4), 334–354.

Dosi, G. (1982). Technological paradigms and technological trajectories:: A suggested interpretation of the determinants and directions of technical change. Research Policy, 11(3), 147–162.

Evangelista, R. (2000). Sectoral Patterns Of Technological Change In Services. Economics of Innovation and New Technology, 9(3), 183–222.

Gokhberg, L., & Kuznetsova, I. (1999). Specificities of innovation activity in Russian industry. In Innovation and Structural Change in Post-Socialist Countries: A Quantitative Approach (pp. 291–305). Springer.

Hipp, C., & Grupp, H. (2005). Innovation in the service sector: The demand for service-specific innovation measurement concepts and typologies. Research Policy, 34(4), 517–535.

HSE. (2015). Indicators of Innovation in the Russian Federation: 2015. Data Book. Moscow: National Research University Higher School of Economics.

32

Kline, S. J., & Rosenberg, N. (1986). An overview of innovation. The Positive Sum Strategy: Harnessing Technology for Economic Growth, 275–305.

Lay, G., Copani, G., Jäger, A., & Biege, S. (2010). The relevance of service in European manufacturing industries. Journal of Service Management, 21(5), 715–726.

Lay, G., Schroeter, M., & Biege, S. (2009). Service-based business concepts: A typology for business-to-business markets. European Management Journal, 27(6), 442–455.

Miles, I. (2000). Services innovation: coming of age in the knowledge-based economy. International Journal of Innovation Management, 4(04), 371–389.

Miles, I. (2005). Innovation in Services. In J. Fagerberg, D. Mowery, & R. Nelson (Eds.), The Oxford Handbook of Innovation. Oxford; New York: Oxford University Press.

Nossal, K., & Lim, K. (2011). Innovation and productivity in the Australian grains industry. Canberra: ABARES.

Nossal, K., & Sheng, Y. (2010). Productivity growth: Trends, drivers and opportunities for broadacre and dairy industries. Australian Commodities, 17(1), 216–230.

OECD. (2013). Innovation survey metadata: Wave 2006-2008. OECD NESTI Room Document.

Pavitt, K. (1984). Sectoral patterns of technical change: Towards a taxonomy and a theory. Research Policy, 13(6), 343–373.

Pavitt, K., Robson, M., & Townsend, J. (1989). Technological accumulation, diversification and organisation in UK companies, 1945–1983. Management Science, 35(1), 81–99.

Peneder, M. (2010). Technological regimes and the variety of innovation behaviour: Creating integrated taxonomies of firms and sectors. Research Policy, 39(3), 323–334.

Reichstein, T., Salter, A. J., & Gann, D. M. (2005). Last among equals: a comparison of innovation in construction, services and manufacturing in the UK. Construction Management and Economics, 23(6), 631–644.

Reichstein, T., Salter, A. J., & Gann, D. M. (2008). Break on Through: Sources and Determinants of Product and Process Innovation among UK Construction Firms. Industry & Innovation, 15(6), 601–625.

Santamaría, L., Jesús Nieto, M., & Miles, I. (2012). Service innovation in manufacturing firms: Evidence from Spain. Technovation, 32(2), 144–155.

Sirilli, G., & Evangelista, R. (1998). Technological innovation in services and manufacturing: results from Italian surveys. Research Policy, 27(9), 881–899.

Soete, L., & Miozzo, M. (1989). Trade and development in services: a technological perspective. Merit.

Vergori, A. S. (2014). Measuring innovation in services: the role of surveys. The Service Industries Journal, 34(2), 145–161.