Standardisation and particularisation in services: Evidence from Germany PDF

Preview

Summary

Research Policy 30 (2001) 1115–1138 Standardisation and particularisation in services: evidence from Germany Bruce S. Tether a,∗ , Christiane Hipp b,c , Ian Miles d aESRC Centre for Research on Innovation & Competition (CRIC), University of Manchester and UMIST, Tom Lupton Suite, Precinct Centre...

Description

Research Policy 30 (2001) 1115–1138

Standardisation and particularisation in services: evidence from Germany Bruce S. Tether a,∗ , Christiane Hipp b,c , Ian Miles d a

d

ESRC Centre for Research on Innovation & Competition (CRIC), University of Manchester and UMIST, Tom Lupton Suite, Precinct Centre, Oxford Road, Manchester, M13 9QH, UK b Mannesmann Pilotentwicklung, Chiemgaustr. 116, 81549 Munich, Germany c Department of Technology and Innovation Management, Technical University of Hamburg Harburg, Schwarzenbergstr. 95, 21073 Hamburg, Germany CRIC and PREST, University of Manchester, Mathematics Building, Oxford Road, Manchester, M12 9PL, UK Received 16 September 1999; received in revised form 10 September 2000; accepted 12 September 2000

Abstract Services have been widely neglected by economists and analysts of innovation, who have instead focused on manufacturing. One of the widely supposed features of services outputs is that they are often highly tailored to their clients. In practice, however, services are sometimes mass-produced and sometimes customised versions of standard products, but can also be produced on a one-off basis. This paper examines the pattern of service activities using German evidence with respect to the structure of service firms’ income from ‘standardised’, ‘partially customised’ and ‘bespoke’ services. The analysis then relates the revealed patterns of ‘standardisation–particularisation’ in the output of the firms to their size and broad sector of activity, and considers the relationship with innovation. Our analysis lends support to previous theoretical studies which provide useful taxonomies of service activities and innovation processes in services. However, our analysis also confirms that services are tremendously diverse both between and within sectors. Mapping and understanding this diversity is a major challenge for future research on service firms and their (innovative) activities. © 2001 Elsevier Science B.V. All rights reserved. Keywords: Standardisation; Particularisation; Services; Innovation

1. Introduction It is well known that in the advanced economies of the OECD services account for roughly two-thirds of GDP and employment (Eurostat, 1999), and that these shares are increasing, whereas those of manufacturing are in decline. In 1970, services accounted for half the ∗

Corresponding author. Tel.: +44-161-275-7376; fax: +44-161-275-7361. E-mail addresses: [email protected] (B.S. Tether), [email protected] (C. Hipp), [email protected] (I. Miles).

European Union’s GDP and less than half (46%) of its total employment, whilst by 1997 ‘market services’ alone accounted for 46% of employment in the EU and 52% its GDP. ‘Non-market services’ accounted for a further 21% of employment and 15% of GDP (Eurostat, 1999). Thus, in 27 years services have increased their share of total employment in the European Union by 21 percentage points; they are the only broad sector of the economy that has expanded in terms of employment, and this trend will undoubtedly continue into the foreseeable future. Even in Germany, a country famous for its manufactured products, services account for a significantly larger share of GDP than manufacturing

0048-7333/01/$ – see front matter © 2001 Elsevier Science B.V. All rights reserved. PII: S 0 0 4 8 - 7 3 3 3 ( 0 0 ) 0 0 1 3 3 - 5

1116

B.S. Tether et al. / Research Policy 30 (2001) 1115–1138

Table 1 The service sector in the West German economy in 1994

All services Services surveyeda Other servicesb Manufacturing

Percentage of the workforce (%)

Growth since 1981 (%)

Percentage of firms (%)

57 32 25 33

2.0 1.9 2.1 −0.8

70 42 28 17

a Wholesaling, retailing, transport and communications, banking and insurance, technical services, software and other business oriented services. b Public administration, education, health care, hotels and catering, services to households, other non-business oriented services. Source: Licht et al. (1995).

(68% versus 31% in 1996 — OECD, 1998), and the same is true of employment, as Table 1 shows. Yet, despite their economic importance, services have received relatively little attention from economists (whether orthodox or Schumpeterian) or from other analysts of innovation, particularly at the micro level. At the macro level the growing importance of services is often related to the alleged productivity slowdown, and the ‘productivity paradox’ (e.g. Roach, 1988, 1991), but whether this slowdown is real, or the result of mis-measurement, is contentious (Grilliches, 1992). The poor state of statistical data on services in many countries underlies this debate — there are grounds for thinking that output and productivity gains in services may be underrepresented in existing statistics. But these measurement problems reflect, in turn, an inadequate understanding of the nature of service activities, and limited efforts to measure innovation, productivity, and indeed many other features of services. The neglect of services, and innovation in services, relates to the widely held view that services are undynamic, even moribund. Arguably, this itself reflects a strong tradition within economics and innovation studies that privileges scientific and technological knowledge, particularly R&D and the development of new tangible artefacts, over other forms of knowledge or change — for example, organisational innovations or changes in routines and procedures — and which therefore sees manufacturing as the key driving sector. Because most services do not undertake R&D and are not producers of new or technically improved tangible

artefacts, 1 analysts have had difficulty in applying the received understanding of innovation to services, and much of the literature is concerned with services as adopters and users of new technologies, particularly information and communication technologies (ICTs), rather than as creative innovators in their own right. Until recently, the only substantial evidence for dynamic activities in services came from case study work. But in the 1990s considerable efforts have been made, through the use of wide-scale surveys and broad research programs, to gain a fuller understanding of the nature of service activities, and a handle on innovation in services. This paper explores some results derived from one such pioneering effort — the 1995 survey of innovation in German service companies. This survey was one of the first large-scale surveys of services, and innovation in services. 2 In this paper we consider the relationships between the classic variables of firm size and sector of activity, together with the issue of the standardisation or particularisation of the firms’ outputs, which is especially relevant to services. We also discuss innovation, although briefly. In a second paper (Hipp et al., 2000), we discuss innovation in services at greater length. In presenting this analysis, we emphasise two things. Firstly, the theoretical understanding of services and especially the extent to which service outputs are subject to standardisation or particularisation (that is whether they are undifferentiated between customers, or are adapted to particular customer needs) is still relatively immature, as is the wider understanding of innovation in services. Secondly, the empirical assessment of the extent of standardisation or particularisation in services, and indeed innovation, is also 1 Services are so diverse that practically any generalisation about them will come up against numerous exceptions. Rapid prototyping services, dentistry, and the ‘hushkitting’ of aircraft are all examples of activities classified as services which may well produce improved technological artefacts. 2 Similar surveys, have been carried out in Denmark (Sundbo, 2000), Finland (Marklund, 2000), Italy (Evangelista and Savona, 1998; Sirilli and Evangelista, 1998) and the Netherlands (Brouwer and Kleinknecht, 1995). More recently, 13 European Union countries included some market services in the second round of the Community Innovation Survey (CIS-2), results from which should be available in due course. Unfortunately CIS-2 did not address the issue of standardisation–particularisation that we assess here, nor does it provide information on the structure of employment by qualifications.

B.S. Tether et al. / Research Policy 30 (2001) 1115–1138

relatively undeveloped. In the main, research tools developed through studies of manufacturing have simply been adapted to services. A debate continues as to how appropriate or adequate these tools are for assessing service activities. Our aim in this paper is to provide some insights into the extent and pattern of standardisation and particularisation in services, and to relate these to innovation activities. On occasion, due to the stage of development of this field of study, our interpretation of the empirical results will be tentative. We begin the paper with a brief review of the literature concerning standardisation and particularisation in services (Section 2), before introducing the empirical evidence (Section 3). The empirical analysis begins with a discussion of the size and (broad) sector characteristics of the surveyed firms. Section 4 then focuses on the standardisation–particularisation of the firms’ outputs. Section 5 provides some further evidence on the structure of employment by qualifications in the surveyed firms, their propensities to innovate, and the propensities of the innovating firms to undertake R&D. The paper ends with a concluding discussion. 3 2. A brief review of the literature We begin with a brief review of the literature on standardisation, variation, specialisation and innovation in services. In particular, we discuss two sets of models: firstly, life-cycle models which relate to the standardisation and destandardisation of service outputs, and secondly, some taxonomic models which broadly relate the nature of the service activities to the extent to which their outputs are standardised or particularised and, less directly, to sector and firm size. 2.1. Life cycle models — standardisation and innovation in services For our purposes, life cycle models are of interest because they relate service activities, and innovation 3

A second paper (Hipp et al., 2000) builds upon this analysis to consider the interaction between the characteristics of the firms which are examined in this paper and their reported innovation activities, and to consider the effects of the innovations on their producers, on the service(s) provided, and on the service users.

1117

in service activities, to the standardisation or particularisation of service outputs. Barras (1986, 1990) saw such models as particularly relevant to services, and related innovation in services to a reverse product life cycle model. 4 In the conventional industry or product life cycle (PLC) model, the first period of activity of the industry is characterised by entry and variety, with a large number of (mainly small) producers providing various offerings to the market (see Klepper, 1996, for a discussion of the PLC model). Competition is in large part focused on the ability to supply superior product designs, with suppliers coming to understand user requirements whilst users grow accustomed to the features and capabilities of the product offerings. During this phase the firms in the industry may supply standardised outputs, although generally in small volumes; they may customise their outputs, adapting them to the requirements of individual customers; at the extreme each output may be individually designed, i.e. bespoke. During this early phase, the emphasis is on product innovation, both within individual firms and in the industry as a whole. Whilst most firms active in the new industry will be small, this picture is confused if large firms from other industries enter the market. The second phase of the PLC is characterised by a convergence on a dominant design, such that variety, in terms of product offerings, declines, standardisation of output increases, and a relatively small number of dominant producers emerge after a period of ‘shake-out’. Not enough is known about the processes by which dominant designs emerge, but, within individual firms, the dominant producers tend to standardise their outputs, and concentrate particularly on process rather than product innovation, whilst fringe producers might also standardise their outputs — producing them in low volumes for niche markets. Alternatively, fringe producers may adapt their outputs to suit niche markets or individual needs. Competition becomes more focused on price, with suppliers seeking to produce the established product types by cheaper methods than their rivals. The latter 4

In this paper we consider standardisation–particularisation to be the result of strategic choices made by the management of the firms in relation to their outputs. Standardisation may also occur because of cultural norms, or may be imposed through legislation, but these are not discussed in this paper.

1118

B.S. Tether et al. / Research Policy 30 (2001) 1115–1138

stages of the PLC are generally characterised by the dominance of a few large producers providing generally standardised products, although some commentators stress the scope for product differentiation, or for ‘dematuration’, following changes in underlying technologies or changes in market environments. Although a useful starting point, the PLC model is more controversial than its widespread adoption by analysts of industrial dynamics might suggest. 5 Even within manufacturing there are doubts about its generalisability and it has been suggested that the model may be particularly applicable to mass-production and is less valid in the ‘Post-Fordist’ era of ‘flexible specialisation’ (Piore and Sabel, 1984). One particular problem concerns the issue of how the industry/product life cycle relates to the life cycle, if such a thing exists, of the underlying technology or production platform upon which each product is based. Barras (1986) adapted the standard PLC model in an attempt to theorise the evolution of services, and thus innovation in services. Barras suggested that innovation in services has largely been driven by the adoption of technological innovations, particularly information and communication technologies (ICTs), developed in the manufacturing sector. He proposed a three-stage model of services evolution. First services adopt, for their own use, technologies developed elsewhere (especially in manufacturing). Innovation in this phase is largely process orientated, and centred on improving the efficiency of service production or delivery, because outputs are essentially standardised (i.e. undifferentiated between customers, with rival firms providing readily comparable services). In the second phase service providers develop new produc5 Teece (1986) contends that the model is most suited to mass markets where consumer tastes are homogeneous, and several authors (e.g. Abernathy, 1978; Pavitt and Rothwell, 1976; Porter, 1983) have also argued that the model does not hold for industries in which there are not rich opportunities for both product and process innovation. Su`arez and Utterback (1995) emphasise that the model may only hold for assembled manufactured goods, and it is clear that the definition of an industry is central to the model. Su`arez and Utterback (1995, footnote 2) admit that ‘the notion of an “industry” is somewhat obscure, for its limits are difficult to define.. . . We think of an industry as composed by a product class, i.e. a group of similar products that serve the same market need and thus compete directly in the market place’. Klepper (1996) subscribes to the same view, making several references to ‘the (industry’s) product’.

tion systems, focused on improving the quality of the services provided — a greater variety of service offerings are provided by each firm, and competition between firms turns to quality rather than just price. Eventually, in the third phase, many new services are produced, so that there is more emphasis on product innovation than on process innovation — thus the reverse product cycle. This last phase is also associated with the entry of new entrepreneurial firms, which adds to the diversity of service offerings to the market. Standardisation of outputs across the industry consequently declines and it becomes more difficult to make direct comparisons between the service offerings of different firms. Within individual service firms the services provided are increasingly tailored to individual clients or niche markets, although for larger niches standardised outputs are still provided. Barras developed his theoretical ideas in the course of empirical investigations, primarily in the banking sector and other information-based services. The reverse life cycle model has been widely criticised (see for example, Buzzacchi et al., 1995, and especially Uchupalanan, 1998, 2000). One reason for caution in accepting the model is that it effectively generalises from a limited set of case studies mainly in one sector to all service activities, but another is that it is focused on just one period in the evolution of services — that associated with the adoption and use of ICTs. 6 Beyond this, it also accords almost no role for strategy and differences between firms, implying that all firms in a sector adopt basically the same strategy, for technologically determinist reasons. Non-technological innovation is also overlooked (Gallouj, 2000; Sundbo, 2000). Despite these criticisms, there are some features of life cycle models that are worth noting for our own analysis. The first is the relationship between (average) firm size, variety and standardisation. Standardised outputs tend to be associated with large firms, due to scale economies, and from this perspective it is notable that service firms tend to be small, or very small, both 6 Arguably, Barras’ model is more robust as a model of technology adoption than of innovation — it has marked resemblance to classic accounts of the technology transfer process — although even here it may be rather specific to a particular era and set of technologies, rather than forming a general model of technology adoption in services.

B.S. Tether et al. / Research Policy 30 (2001) 1115–1138

in absolute terms and relative to manufacturing firms (Eurostat, 1999). There are of course several important exceptions, notably in banking and finance, transport and communications, and to some extent, retailing, but the preponderance of small firms would lead us to expect a relatively high level of heterogeneity in service outputs. This said, there are several reasons for qualifying this statement. Firstly, the predominance of small firms is in part a reflection of services traditionally supplying very local markets and the need to physically locate in close proximity to the customer base. Telecommunications are being used to break down this relationship, particularly where standardised services are concerned. Traditionally, relatively few of the small firms were suppliers of highly specialised niche services. The dominance of small firms in services may also reflect the prevailing appropriation conditions for ‘product innovation’ in services, in that it may be more difficult to protect these from imitation, compared with product or process innovations in manufacturing, or process innovations within the service sector. More generally, this problem of copying may mean that firm growth on the basis of service (as opposed to process) innovation may be relatively rare in service activities. 7 Arguably, this also relates to the nature of product life-cycles in services, for a service standard may emerge, but it is less clear that a small number of firms will dominate the market (unless scale economies and process innovation are important and lead to the ‘selection’ of the most efficient producers). 2.2. Taxonomic models of specialisati...