Innovation and the skill mix: chemicals and engineering in Britain and Germany.

Mason, Geoff ; Wagner, Karin

1. Introduction

In recent years continued disparities between countries in economic performance combined with the rapid pace of technological change have led to growing interest in comparisons of 'national systems of innovation'. The development of this concept reflects increased awareness of both the cumulative, informal nature of much innovative activity--especially the interaction between producers and prospective users of new products--and the ways in which national-institutional structures condition the decisions taken by key actors in the innovative process such as enterprises, universities, and individual engineers, scientists and other members of the labour force (Lundvall, 1988; Freeman, 1988; Westney, 1993; Sos-kice, 1993).

Analysis of 'national' innovation systems has obvious limitations: rapid growth in trans-national corporate activity continues to reduce the influence of national policy regimes and, conversely, recent studies of localised R&D spillovers in the United States and Italy suggest that the appropriate geographical unit of analysis of innovative performance for many purposes may be sub-national (Acs, Audretsch, Feldman, 1993; Audretsch and Vivarelli, 1994). However, it seems likely that at least some institutions operating at a national level do influence relative innovative performance, for example, institutions underpinning the domestic supply of finance to industry and the education and training of the workforce.

In this note we focus on the role of the latter set of institutions. Specifically, we draw on a recent study based on the chemicals and engineering industries in Britain and Germany to explore the links between inter-country differences in human capital formation and apparent disparities between the two countries in technological performance.(1) Long-term comparisons of investments in R&D and measures of innovative 'output' such as country shares of patents granted in the United States point to a marked German superiority over Britain in both respects (Patel and Pavitt, 1987, 1989) and studies such as Fagerberg (1989) suggest that such inter-country differences in 'technological competitiveness' have a strong effect on industrial competitiveness as measured by comparative export performance.

In keeping with the research methodology developed in earlier cross-country comparisons at the National Institute, detailed information on Anglo-German differences in industrial performance and the quality and utilisation of physical and human capital inputs was gathered primarily through visits to matched samples of chemicals and engineering establishments in each country. Both national samples comprised a mix of well-known 'leading' companies and small/medium enterprises. The visits were all carried out between December 1992 and July 1993. In each case extensive semi-structured interviews were carried out with technical and personnel managers and, in all but one establishment in each country, the visits included direct observation of the work taking place in production and technical support departments.(2)

The two industries were selected for investigation on the basis of their expected contrasts in respect of relative performance: in chemicals Britain's productivity and, to a lesser extent, trade performance is known to compare favourably with Germany whereas in engineering German superiority is well-documented. In all, some 30 production plants and eight separate research centres were visited in the two countries, focussing on the following main product areas:

Chemicals: paints & industrial coatings; specialised intermediates Engineering: vehicle components; specialised high-speed machinery

The ordering of the note is as follows: Section 2 reports on comparisons of the recent performance of closely-matched pairs of British and German production plants in the two industries. Section 3 assesses the mix of skills which companies need to respond effectively to intensifying competitive pressures to speed up the rate of new product development and innovations in production processes. In Section 4 we examine in detail recent changes in the employment of highly-qualified engineers and scientists in Britain and Germany and relate such changes to the high and growing level of technical uncertainty which characterises the modern innovative process. Section 5 summarises our main findings.

Table 1. Number of establishments visited by product area

 Britain Germany

Chemicals 6 5
Engineering production plants
 --high-speed machinery 4 4
 --vehicle components 6 5
Engineering R&D centres 5 0
Intermediate research institutes 0 3

Total 21 17

2. Comparisons of workforce skills, physical capital and industrial performance

As a result of the well-established 'Dual System' of apprentice training in Germany, some two thirds of the German workforce are qualified to craft level or above compared to just over one third so qualified in Britain. Recent studies suggest that there is a positive, and statistically significant, association between Anglo-German differences in the proportions of the workforce qualified to this level in individual industries and relative performance in respect of labour productivity (O'Mahony, 1992b) and export competitiveness (Oulton, 1993). Some of the mechanisms by which differences in workforce skill levels (as proxied by variations in formal vocational qualifications) contribute to relative performance have been identified in earlier comparisons of matched samples of production plants.

For example, German machine operators and other shopfloor workers have been found to perform, on average, a wider range of tasks than their British counterparts and to be better able to meet (and consistently maintain) high quality standards (Daly, Hitchens, Wagner, 1985; Steedman and Wagner, 1987, 1989; Mason, van Ark, Wagner, 1994). Similarly, Meister-qualified supervisors in Germany have been shown to be better-equipped than most supervisors in Britain to take detailed responsibility for production management tasks and to liaise with technical support departments (Prais and Wagner, 1988; Steedman, Mason, Wagner, 1991).

These same studies have also identified important complementarities between management and workforce skill levels in each country and the extent to which physical capital equipment is effectively selected and utilised. In consequence, it is not easy to evaluate the impact of skill differences on relative performance in isolation from differences in physical capital. A recent 'growth accounting' calculation suggests that differences in the two types of capital have 'about equal weight' in explaining the Anglo-German manufacturing productivity gap (O'Mahony, 1992a, p 54).

For the present study (as described in Mason and Wagner, 1994, Section 6), we undertook a detailed comparison of the physical and human capital inputs used by seven pairs of production plants which were closely matched for principal product area. In respect of performance criteria such as international and domestic market shares and export and employment growth, these matched pairs of plants--four in engineering and three in chemicals--in many ways exemplified the conventional wisdom regarding Anglo-German performance in the respective industries.

In engineering, for instance, surviving British producers have--after several rounds of rationalisation and reorganisation in the last 15-20 years--achieved significant improvements in labour productivity but have not yet succeeded in closing the gap with Germany. In two customised small-batch machinery comparisons, both British plants were much smaller than their German rivals (reflecting their loss of market share in recent decades) and tended to be 'followers' in technical innovation. In vehicle components the German plants were still ahead in respect of process innovations and shopfloor efficiency even though their productivity advantage over British competitors has narrowed greatly in recent years. British price competitiveness still depends heavily on favourable movements in exchange rates.

By contrast, in chemicals, the three matched plant comparisons all pointed to a much more favourable performance by Britain relative to Germany. Two British manufacturers of paints and surface coatings were as securely placed in their domestic markets as their German counterparts and, in a specialised intermediate chemicals comparison, the British producer was found to typify the substantial export success of this sector of the industry. In two of the three comparisons, the German plants had until recently succeeded in expanding employment in contrast to cutbacks in Britain but, at the time of our visits, a sharp decline in domestic demand had led to a considerable increase in short-time working and 'labour-shedding' in the German industry.

In both industries relative competitiveness was associated with variations in a range of factors pertaining to both physical and human capital inputs. For example, British engineering plants attributed a great deal of their recent productivity growth to new investment in capital equipment and improved layout of machinery on the shopfloor but their German rivals were typically making even greater use of new equipment such as automated assembly lines (including the use of robotics) as well as using more purpose-built and linked machinery. In chemicals, again by contrast, most British plants were at least equal to and sometimes ahead of their German counterparts in the use of new, sophisticated capital equipment.

In respect of workforce qualifications there were sharp disparities between the two countries in both industries, and particularly (as expected) at intermediate level. In the full sample of chemicals plants visited some 45 per cent of German shopfloor (process) workers had received a craft apprentice training as against 23 per cent in Britain. In engineering the gap was even larger with 57 per cent of German shopfloor workers holding craft-level qualifications compared to only 20 per cent in Britain. (In both countries shopfloor skill levels in the specialised machinery plants were notably higher than in the vehicle components sector). The German skills advantage in direct production areas was further enhanced by the widespread deployment of Meister-trained supervisors, whereas only 5 per cent of British supervisors had received any form of specialised training for their role.

At graduate level and above the differences in work-force qualification levels in the national samples of establishments strongly reflected Anglo-German differences in the annual output of highly-qualified personnel. In engineering and technology subjects, Germany produces approximately two thirds more Bachelor degree graduates per head of population than does Britain, more than five times as many people qualified to MSc level and a third more PhDs. By contrast, in physical sciences such as chemistry and physics, the proportions of the population gaining higher education qualifications are much the same in each country but the mix of awards in Germany is more oriented towards higher degrees than in Britain.(3)

Thus in the chemicals industry the combined total of Bachelor and higher degree graduates in technical (science, engineering and technology) subjects represented similar proportions of total employment in both samples of establishments (11 per cent in Britain, 10 per cent in Germany) and similar large proportions (80-85 per cent) of these highly-qualified personnel were employed in the main 'technical support' departments. The main difference between the two TABULAR DATA OMITTED samples lay in the mix of higher education qualifications in these technical areas: the British chemicals plants employed roughly 3 PhDs (nearly all chemists) for every 7 First degree graduates (fairly evenly divided between engineers, on the one hand, and chemists and other scientists on the other); in the German plants there were roughly equal numbers of PhDs (predominantly chemists) and Fachhochschule graduates (mainly engineers). In neither chemicals sample were there significant numbers of MSc-holders or their German equivalent.

By contrast, in the engineering industry inter-country differences at intermediate skill level appeared to be compounded by disparities in high-level qualifications. In the case of our two national samples the comparison was complicated by several plants in the British sample belonging to companies which operate detached corporate R&D centres. This form of deployment contributed to a smaller share of employees in the British plants attached to technical support departments. With this reservation, it is notable that the average 12 per cent share of technical graduates and post-graduates in total employment in the German engineering sample was more than double that in the British engineering plants.

In the British production plants there were only half a dozen PhDs in total and, even in technical support departments, there was only one MSc-holder for every 9 First degree graduates. In the specialist R&D centres in TABULAR DATA OMITTED Britain (which of course were responsible for serving all plants owned by their parent company) the ratio of higher to First degrees was 1:4 and some 10 per cent of technical graduates held PhDs.

In the German engineering plants the relatively large employment share of technical graduates as a whole primarily reflected the widespread availability of Fachhochschule graduates who were employed not just in technical support departments but also, frequently, in a 'first-line' production management role (thus augmenting the technical skills and knowledge of shopfloor Meister). However, the proportion of employees qualified to University diploma (MSc) level was also larger than in Britain (just under a third of all German graduates in technical support areas were so qualified) and, although PhDs accounted for only 4 per cent of all technical graduates in our German engineering sample, in absolute terms their numbers were significant as well: the five largest engineering production plants visited in Germany employed about 30 per cent more doctorates than did the five corporate R&D centres belonging to leading British engineering companies.(4)

In the next section we explore in greater detail the ways in which these distinctive Anglo-German differences in workforce qualification levels affected the ability of plants in each country to compete effectively in new product development and innovations in production processes.

3. Innovation and skills

In both Britain and Germany a majority of companies we visited were under growing pressure to improve the quality and performance of their products and shorten product lead-times while simultaneously having to compete on price as well. This 'double squeeze' was particularly common in the vehicle components sector where a relatively small number of large end-users seek to pass on the cost pressures resulting from their own intensely competitive battles for market share, but we also heard of similar pressures in many of the machinery and chemicals plants.

An inherent feature of the specialised high-speed machinery sector is that each individual machine, or very small batch of machines (rarely more than four or five at a time), has to be made to precise customer-specific requirements, with the manufacturer's responsibility typically extending to installation and commissioning of the equipment in customers' factories as well. This customisation of the machinery is now increasingly combined with efforts to improve its performance in respect of a range of factors such as speed, reliability, energy efficiency, quietness and compactness.

The pace of change is such that, for example, in some branches of the industry average machine speeds have increased tenfold over the last 20 years. In general terms the pressure to innovate is sustained by the certain knowledge that competitors in world markets are working on the same problems and by increasingly tight product specifications sought by customers, for instance, guarantees of reduced lifetime maintenance requirements for the machines and/or an enhanced ability for the machine end-users to make their own products with lighter, thinner materials.

In chemicals as well there was evidence of a growing pressure to innovate to meet more elaborate performance requirements specified by customers or to anticipate improvements in performance which might be necessary to survive in the marketplace in the future. Competitive pressure of this kind was clearest in the specialised intermediates sector whose prime function is to develop new kinds of ingredients or 'building blocks' for use by other chemical producers and, in particular, to devise cost-effective and reliable ways of achieving the frequently difficult chemical transformations required.

However, even in the ostensibly more straightforward paints and surface coatings sector, several producers in both countries reported facing growing complexity in product specifications, for example, lacquers which needed to be able to adhere adequately to different materials under very demanding conditions. In the bulk consumer goods branch of this industry, there is strong competitive pressure on manufacturers to seek an edge in terms of product durability, coverage and other attributes while usually seeking to compete on price as well, and (as throughout the chemicals industry) a continuing need to conform with stricter environmental requirements (for example, through development of solvent-free products or reduction of noxious emissions from production processes).

In general terms our observations suggested that the ability to respond successfully to competitive pressures to innovate depends on a wide range of skills and knowledge from craft- and technician-level upwards. These skill requirements include the need for all staff who come into contact with customers to have sufficient technical understanding to properly define (and anticipate) customer needs, and to communicate this information to their own colleagues. In both industries high levels of engineering and supervisory knowhow are also needed to achieve a smooth flow of production under difficult conditions, for instance, when the product mix in engineering is subject to rapid change or, in chemicals, when production of a new and unpredictable substance is being 'scaled-up' following its initial laboratory development.

Another core area of skill needs relates to market pressure to reduce the average time taken between receipt of new orders and the delivery of final products, even when the goods in question are highly customer-specific. One key to this in precision engineering, for example, is a highly flexible ('modular') design of products which maximises the use of 'standard' components or sub-assemblies and thus enables the actual point of customisation to be delayed till relatively late in the production process. Similarly high levels of design skills (and effective management of the links between design and production departments) are needed to ensure that even complicated products (or their components) are no more difficult to manufacture than they need to be.

In the German engineering plants, in particular, such links between the shopfloor and technical support departments were greatly facilitated by the combination of technically-trained Meister and large numbers of Fachhochschule-qualified production and development engineers with visible consequences in a steady flow of incremental product and process innovations. Several British engineering plants also reported recent improvements in shopfloor efficiency, for example, relocation of machines in order to help speed up production and reduce the movement of materials and components about the factory. However, in general such improvements were either belated as compared to German counterparts or were more limited in scope, with less attention given to in-house adaptation of machinery to cut down on bottle-necks or economise on maintenance costs. The apparent lesser ability of British plants to focus on such incremental innovations was associated with both lower levels of technical qualification among supervisors and technical support staff and with greater involvement of highly-qualified personnel in day-to-day production problems which in Germany would tend to be dealt with by shopfloor supervisors and workers.(5)

For reasons which may reflect the particular nature of chemical production processes but in any event deserve further investigation, Anglo-German disparities in shopfloor skills in our samples of chemicals plants did not appear to affect relative performance to anything like the same extent as in engineering. However, some British chemicals plants reported recent difficulties in implementing flexible work practices involving significant cross-skilling due to the limited educational backgrounds and abilities of large proportions of their shopfloor workforces.

As might be expected, the links between formal workforce qualifications and innovative performance were most apparent in respect of high-level skills. In engineering the bulk of our discussions and observations in the course of establishment visits pointed to strong links between post-graduate expertise and the ability of firms to take quick advantage of new developments in scientific knowledge. Thus, in one comparison of matched machinery plants, technical advances in complex areas such as the use of lasers were specifically identified by British research staff as a key element in their German rival's competitive advantage. Another of the plants in our German sample is regarded as a world leader in the development of electronic systems in its field of vehicle components; its success in this area has stemmed largely from high-level in-house research in close collaboration with university-based specialists. In a different branch of vehicle components, the German plant visited engaged in similar collaboration with--and recruitment of post-graduates from--a university at the forefront of research into acoustics and had achieved such breakthroughs in noise reduction that it could now offer specialist services in this field to firms in many other sectors of engineering.

The value of post-graduate expertise in key specialist areas is recognised to some extent by leading British engineering companies which employ some higher degree graduates in their R&D centres and/or have developed links with university-based researchers. However, these companies are hardly representative of the mainstream British engineering industry and, as said to us by one British research director, very few even of the biggest British companies have assembled the 'critical mass' of highly-qualified personnel which is necessary if sustained technical advances are to be made.

As noted above, the relative deficiencies of British chemicals plants at graduate level and above were much less severe than in the engineering sample. In both the British and German chemicals industries, there is a very clear role for PhD chemists trained in the disciplines of experimental science who work in combination with First degree/Fachhochschule engineers and others with good practical skills and experience with process equipment. Although the proportion of post-graduate engineers and scientists in British chemicals plants is typically smaller than in Germany, in general the above-average productivity, trade and innovation performance of the British chemicals industry--compared to British manufacturing as a whole--is closely associated with its relative adequacy of supplies of highly-qualified technical staff.

4. High-level skills and 'information-intensive' manufacturing

Recognition of the contribution to industrial performance made by post-graduates in German manufacturing industry is implied by the changes in the mix of high-level qualifications which have occurred in recent decades.

As shown in Table 4, the number of German home students gaining University (MSc-equivalent) diplomas in engineering and technology in recent years has grown more than twice as fast as the number gaining Fachhochschule diplomas. This is reflected within the German mechanical engineering industry where total employment of University engineers grew by 183 per cent between 1968 and 1987 while, over the same period, employment of Fachhochschule graduates increased by 55 per cent. Thus by 1987 the ratio of FH to University engineers in employment stood at 2.4:1, down from 4.4:1 in 1968 (VDMA, 1988). As reported to us in the majority of plants in our German engineering sample, this development reflects a growing preference for the more highly-qualified type of engineer, not just a response to their increased availability. Indeed, since University graduates command higher salaries than Fachhochschule-leavers, German employers were very concerned to secure a 'cost-effective' mix of each type of graduate (Mason and Wagner, 1994, Section 8).

In German chemicals manufacturing, the change in mix of high-level qualifications has been less dramatic. Combined employment of University-educated scientists and engineers increased by 59 per cent between 1976 and 1991 compared to 45 per cent growth in employment over the same period of Fachhochschule technical graduates (predominantly engineers). In 1991 the ratio of University to FH graduates was approximately 1.7 compared to 1.56 in 1976 (Bundesarbeitgeberverband Chemie, 1991).

Table 4. Rate of growth in numbers of home students gaining Bachelor degree
and higher degree awards in science and engineering subjects in Britain and
(West) Germany, 1980 to 1990

(percentage change, rounded to nearest five percentage points)

Britain

 First Degree Masters(a) Doctorate

Chemistry +25 +5 +20
Physics(b) +20 +40 +35
Mathematical sciences(c) +60 +140 +85
Engineering and technology(d) +40 +50 +35

Germany

 Diplom (FH) Diplom (Uni) Doctorate

Chemistry +20 +170 +70
Physics(b) -20 +190 +70
Mathematical sciences(c) +260 +110 +80
Engineering and technology(d) +50 +115 +45

Sources:
USR, University Statistics, Volume 2 and private communication; CNAA, Annual
Reports; SB, Statistisches Jahrbuch; NIESR estimates; population data from
OECD, Labour Force Statistics, 1992.

Notes:

(a) May include a small number of other sub-doctoral post-graduate awards such
as Postgraduate Diplomas.

(b) Includes astronomy.

(c) Includes computer science (Informatik).

(d) Excludes architecture.

In the British engineering industry total employment of professional engineers and scientists holding at least a First degree rose by some 55 per cent between 1978 and 1988 while, over the same period, employment of technician-level engineers declined by 19 per cent and total employment fell by 36 per cent (EITB, 1989). No detailed information on trends in the industry's employment mix of First and higher degree graduates in technical subjects is available. However, some tentative conclusions may be drawn from the fact that growth rates in the numbers of home students qualifying at First, Masters and PhD levels in engineering and technology subjects were all within a range of 35-50 per cent, as shown in Table 4.

Although these engineering graduates of course enter employment in many other sectors of the economy apart from engineering, the rough parity in growth of output at each level suggests that the mix of Bachelor and higher degree graduates employed in engineering is unlikely to have changed as sharply in Britain as in Germany in recent years. Indeed, there was a sharp split in our British sample between plants belonging to leading companies with a long tradition of graduate recruitment and other, smaller plants for whom it has been a big step even to start employing more First degree graduates recently, let alone post-graduates.

In the British chemicals industry, the proportion of all employees holding at least a First degree rose from 9 per cent in the late-1970s to 15 per cent in 1990. This includes non-technical graduates as well but recent data indicate that roughly 80 per cent of all First degree graduates recruited by the industry and 90 per cent of those with higher degrees have qualified in science or engineering subjects. At present approximately one higher degree (predominantly PhD) chemist is recruited for every two First degree graduates in chemistry; for chemical engineers the ratio of higher to First degree graduates is roughly 1:9 (Chemical Industries Association, 1990-93). As in the case of engineering, trends in the numbers of home students obtaining First degrees and PhDs in chemistry between 1980-90 suggest that the mix of First and higher degrees may not have changed greatly over this period.

In the course of our establishment visits, this overall pattern of increased employer demand for technical graduates as a whole in both countries, and for holders of post-graduate (particularly MSc-level) awards in Germany, was repeatedly explained to us in terms of a greater need for staff who could tackle new and complex technical problems in an analytical way (based on theoretical understanding) rather than simply rely on past experience and trial-and-error methods of problem-solving.

This trend is perhaps best understood in the context of recent empirical work by Bartel and Lichtenberg (1987), using data for US manufacturing industry, which supports a hypothesis that demand for highly-educated workers relative to the demand for less-educated workers is inversely related to firms' experience with technologies currently in use (as proxied by the mean age of capital equipment). One of the key premises underpinning this hypothesis is that the productivity of highly-educated workers relative to low-educated workers is greater, the more uncertainty there is in the production environment.

As emerged from our interviews with technical managers in both countries, one of the most striking features of the modern industrial innovation process is precisely the high and growing level of uncertainty which prevails. As new problems and difficulties present themselves, manufacturers need to be able to assess whether they can be resolved by incremental improvements in existing products and/or processes, or whether some kind of radical change or breakthrough innovation will be necessary. For instance, producers of industrial coatings may be able to solve new problems simply by fine-tuning existing recipes or they may be better advised to investigate new ways of cross-linking existing polymers or even seek to develop brand new kind of polymers. Similarly, in the vehicle components sector, a company seeking to reduce the weight of or the friction undergone by a particular product (thus increasing its expected life) needs to know whether to continue working with its existing materials or whether to explore the use of some completely different material.

A degree of technical uncertainty is of course intrinsic to the innovative process. However, such uncertainty now appears to be intensifying within many branches of manufacturing, partly because of the growing pressure (as previously noted) to solve problems of improving product performance while simultaneously reducing production costs and product lead-times, and also because of the remarkable speed of developments in science and technology in recent years. In particular, the growing use of programmable automation equipment to meet a wide variety of different customer requirements has been characterised as an 'information-intensive' form of production with increasing resources needing to be devoted to gathering information on both the specialist needs of customers and the technical possibilities of meeting such needs (Willinger and Zuscovitch, 1988). The pace of change in this respect has increased the risk for many firms of being overtaken by competitors who employ more highly-qualified technical personnel and/or have better access to the skills and knowledge, and specialised equipment, of the external science base (universities, research institutes and consultancies, etc).

In addition to apparent growth in direct use of scientific research results by manufacturers (Narin and Olivastro, 1992), there has also been an increasing tendency in recent decades for technological knowledge (traditionally based on practical, largely unwritten experience and knowhow specific to firms and individuals) to enter the public domain through specialist books, journals and conference papers. Hence even those firms who have no aspirations to do more than adopt innovations developed elsewhere now increasingly require the services of highly-qualified engineers and scientists in order to identify and make use of relevant information if they are to have any hope of staying in touch with more advanced competitors.(6)

In short, the increased uncertainty associated with the modern innovative process points to a continued shift in the mix of manufacturing skills towards highly-qualified scientists and engineers. This shift is already well advanced in both German and British manufacturing industry but, for a variety of reasons outlined briefly below, German manufacturing industry at present makes far greater use of the specialist knowledge and contacts possessed by post-graduates, both as direct employees of manufacturing firms and also as external specialists employed indirectly on a contract basis.

Anglo-German disparities in the level of industrial demand for post-graduate engineers and scientists--as opposed to Bachelor degree graduates--partly reflect the high value placed by British employers on First degree graduates and the several years' industrial training and work experience they receive while post-graduates of the same age are still involved in full-time education. As is well known, one of the main strengths of the British higher education system as compared to Germany is its relatively 'efficient' production of young First degree graduates whose intellectual abilities (though not the academic standards reached) are in many cases the equal of much older German University graduates holding qualifications equivalent to an MSc.

However, the comparatively limited use made of post-graduate engineers and scientists by British manufacturing industry also reflects two negative factors: firstly, widespread concern that post-graduates are 'too specialised' and less likely than First degree graduates to acquire commercial and practical skills; and secondly, the relatively slow growth by international standards in British manufacturing investment in research and development which is the predominant area of employment for post-graduate engineers and scientists (Mason and Wagner, 1994).

The prevalence of complaints about 'over-specialisation' in Britain is hardly surprising since the further academic study signified by a post-graduate qualification comes on top of a very narrow and intensive period of education which, in England and Wales at least, first starts with 'A' level courses at the age of 16. By contrast, in Germany post-graduates are seen as broadly-educated people who also have a proven area of specialist expertise; to this is added in many cases the valuable experience of part-time industrial contract work for research institutes and universities during their studies.

The deep-seated reasons for the slower growth of R&D spending in Britain as compared to Germany cannot be addressed within the scope of this note. However, in the light of our investigations in both countries, we conclude that the competitiveness of many British manufacturing firms (particularly in engineering) is severely undermined by comparative under-investment in R&D and, as one consequence of this, the limited use made of post-graduate skills and knowledge. In particular, there seems little doubt that, as acknowledged in the recent government White Paper on science, engineering and technology (HMSO, 1993), existing schemes for 'knowledge transfer' between the science base and industry in Britain are 'piecemeal' in nature: taken as a whole they simply do not have the industrial 'reach' of German institutional arrangements for knowledge transfer (as exemplified by the high-profile, easily accessible networks of intermediate research institutes run by the Fraunhofer and Max-Planck Societies) nor do they facilitate the transfer of highly-qualified people between higher education and industry on anything like the same scale as in Germany.(7)

6. Summary

Our comparison of matched samples of chemicals and engineering establishments in Britain and Germany has highlighted several ways in which the different mix of workforce skills delivered by each country's education and training system affects relative innovative performance, with attendant consequences for relative industrial competitiveness.

At intermediate skill level, long-established institutions in Germany for the training of craft apprentices, supervisors and technicians contribute substantially to German plants' ability to maintain a steady flow of incremental process innovations and to deal efficiently with initial production problems when new products are introduced. The disparity with British skill levels and industrial performance was most marked in comparisons of matched engineering plants. In chemicals comparisons, Anglo-German disparities in shopfloor skills did not affect relative performance to anything like the same extent.

At higher levels of qualification, the German engineering plants benefit from the widespread availability of highly practical Fachhochschule engineers and also from greater access to the specialist knowledge and contacts possessed by post-graduate engineers and scientists. In chemicals the mix of high-level qualifications is also more oriented towards post-graduates in Germany than in Britain but the relative adequacy of supplies of highly-qualified technical staff as a whole in British chemicals plants contributes to above-average innovation and trade performance compared to other branches of British manufacturing.

In both countries the increased technical uncertainty associated with the modern innovative process has contributed to a substantial shift in the mix of manufacturing skills towards highly-qualified personnel. This shift is, however, less well advanced in Britain partly because of British employers' continued preference for relatively young First degree graduates as compared to post-graduates who are widely (and justifiably) viewed as 'over-specialised', and also because of comparatively slow growth by international standards in British manufacturing investment in research and development. At the same time institutional arrangements for knowledge transfer between the science base and industry in Britain are less comprehensive and well-organised than in Germany.

APPENDIX

TABULAR DATA OMITTED

TABULAR DATA OMITTED

Table A3. Estimated annual flows of home students per million population
gaining Bachelor degree and higher degree awards in science and engineering
subjects in Britain and (West) Germany, 1990

(adjusted for double-counting; rounded to nearest five)(a)

Britain First Degree Masters(b) Doctorate
Total

Chemistry 35 5 15 55
Physics(c) 30 5 10 45
Mathematical sciences(d) 90 20 5 115
Engineering and technology(e) 190 25 15 230

Germany Diplom (FH) Diplom (Uni) Doctorate
Total

Chemistry 5 15 30 50
Physics(c) 2 35 15 52
Mathematical sciences(d) 30 45 5 80
Engineering and technology(e) 315 140 20 475

Sources:
USR, University Statistics, Volume 2 and private communications; CNAA, Annual
Reports; SB, Statistisches Jahrbuch; NIESR estimates; population data from
OECD, Labour Force Statistics.

Notes:

(a) To avoid double-counting of students gaining awards at more than one
level, the following assumptions have been made: For Britain 95 per cent of
chemistry and physics PhDs, and 85 per cent of mathematics and engineering
PhDs, are assumed to have proceeded directly from their first degrees without
obtaining a Masters degree (based on information supplied by the Universities
Statistical Record). The remaining PhDs in each subject are assumed to have
first obtained b both a First and a Masters degree. For Germany all PhDs are
assumed to have previously obtained a University rather than a Fachhochschule
diploma.

(b) May include a small number of other sub-doctoral post-graduate awards such
as Postgraduate Diplomas.

(c) Includes astronomy.

(d) Includes computer science (Informatik).

(e) Excludes architecture.

NOTES

(1) Unless otherwise stated the term 'Germany' is used throughout this note to refer to the former Federal Republic.

(2) For further details of sample selection and methodology, see Mason and Wagner, 1994, Section 4.

(3) See Appendix Table A3 for details. For the purposes of this comparison German University diplomas (typically requiring in excess of six years to complete) are regarded as broadly equivalent to a British MSc degree. The German Fachhochschule diploma course usually lasts four years and the standards reached are nearer to British First degree level than any other British award; however, the greater orientation of Fachhochschule courses towards practical applications makes them less capable than British First degree courses of serving as a basis for subsequent post-graduate study.

(4) PhD-holders were also strikingly well-represented in senior management positions in German plants in both industries. In Britain there were relatively few senior managers with higher degrees but significant proportions of managing directors and of senior production and technical managers in both industries held First degrees in technical subjects (see Mason and Wagner, 1994, Section 5 for details). Qualified technical staff appear more likely to be under-represented at company director level than at factory management level in British manufacturing.

(5) Elsewhere we have described this problem as the 'drawing-down' of graduate engineers and higher technicians in British engineering factories to deal with, for example, recurrent equipment breakdowns and protracted 'teething' difficulties with the installation of new equipment (see Steedman, Mason, Wagner, 1991).

(6) As described in Cohen and Levinthal (1989), in-house R&D activity plays a key role in developing the capacity of any firm to 'absorb' relevant knowledge generated elsewhere, whether by other firms or in the academic science base. Indeed, their empirical results for US manufacturing suggest that, in some industries such as chemicals and electrical / electronic engineering, the positive incentive for firms to invest in R&D in order to take advantage of 'spillovers' of knowledge into the public domain may be strong enough to offset the negative impact on R&D arising from the inability of innovating firms to fully restrict the appropriation by others of their R&D output.

(7) For detailed comparisons of the organisation of R&D in Britain and Germany which support these conclusions, see Atkinson, Rogers and Bond (1990), CEST (1991, Chapter 7) and Mason and Wagner (1994, Section 7.3).

REFERENCES

Acs, Z, Audretsch D and Feldman, M (1993), 'Innovation and R&D spillovers', CEPR Discussion Paper no. 865, December.

Audretsch, D and Vivarelli, M (1994), 'Small firms and R&D spillovers: evidence from Italy', CEPR Discussion Paper no. 927, March.

Atkinson, H, Rogers, P and Bond, R (1990), Research in the United Kingdom, France and West Germany: A Comparison, SERC, Swindon.

Bartel, A and Lichtenberg, F (1987), 'The comparative advantage of educated workers in implementing new technology', The Review of Economic and Statistics, February.

Bundesarbeitgeberverband Chemie, AT-Angestellte, Leitende Angestellte, Akademiker und Fachhochschulabsolventen in der Chemischen Industrie, Wiesbaden, 1991.

Centre for Exploitation of Science and Technology (CEST) (1991), Attitudes to Innovation in Germany and Britain: a Comparison, London.

Chemical Industries Association (1990-93), Graduate Recruitment Surveys, London.

Cohen, W and Levinthal, D (1989), 'Innovation and learning: the two faces of R&D', Economic Journal, September.

Daly, A, Hitchens, D and K Wagner (1985), 'Productivity, machinery and skills in a sample of British and German manufacturing plants', National Institute Economic Review, May.

Engineering Industry Training Board (EITB) (1989), British Engineering: Employment, Training and Education, Watford.

Fagerberg, J (1988), 'International competitiveness', Economic Journal, vol. 98, June.

Freeman, C (1988), 'Japan: a new national system of innovation?', in G Dosi et al (eds), Technical Change and Economic Theory, Pinter, London.

Gesellschaft Deutscher Chemiker (1992), Statistik der Cbemiestudiengange in der Bundesrepublik Deutschland 1991, Frankfurt/Main.

HMSO (1993), Realising our Potential: a Strategy for Science, Engineering and Technology, London.

Lundvall, B (1988), 'Innovation as an interactive process: from user-producer interaction to the national system of innovation', in G Dosi et al (eds), Technical Change and Economic Theory, Pinter, London.

Mason, G, van Ark, B and Wagner, K (1994), 'Productivity, product quality and workforce skills: food processing in four European countries', National Institute Economic Review, February.

Mason, G and Wagner, K (1994), High-level Skills and Industrial Competitiveness: Post-graduate Engineers and Scientists in Britain and Germany, National Institute Report no. 6, London.

Narin, F and Olivastro D (1992), 'Status report: linkage between technology and science', Research Policy, vol. 21.

O'Mahony, M (1992a), 'Productivity levels in British and German manufacturing industry', National Institute Economic Review, February.

O'Mahony, M (1992b), 'Productivity and human capital formation in UK and German manufacturing', National Institute Discussion Paper no. 28.

Oulton, N (1993), 'Workforce skills and export competitiveness: an Anglo-German comparison', National Institute Discussion Paper no. 47.

Prais, S (1981), Vocational qualifications of the labour force in Britain and Germany, National Institute Economic Review, November.

Prais, S and Wagner, K (1988), 'Productivity and management: the training of foremen in Britain and Germany', National Institute Economic Review, February.

Patel, P and Pavitt, K (1987), 'The elements of British technological competitiveness', National Institute Economic Review, November.

Patel, P and Pavitt, K (1989), 'A comparison of technological activities in West Germany and the United Kingdom', National Westminster Bank Quarterly Review, May.

Soskice, D (1993), 'Product market and innovation strategies of companies and their implications for enterprise tenure: a comparative institutional approach to some cross-country differences', Wissenschaftszentrum, Berlin, (mimeo), November.

Steedman, H, Mason, G and Wagner, K (1991), 'Intermediate skills in the workplace: deployment, standards and supply in Britain, France and Germany', National Institute Economic Review, May.

Steedman, H and Wagner, K (1987), 'A second look at productivity, machinery and skills in Britain and Germany', National Inwtitute Economic Review, November.

Steedman, H and Wagner, K (1989), 'Productivity, machinery and skills: clothing manufacture in Britain and Germany', National Institute Economic Review, November.

Verband Deutscher Maschinen- und Anlagenbau (VDMA) (1988), Ingenieur-Erhebung im Maschinen- und Anlagenbau 1988, Frankfurt/Main.

Westney, E (1993), 'Country patterns in R&D organisation: the United States and Japan', in B Kogut (ed), Country Competitiveness: Technology and the Organising of Work, Oxford.

Willinger, M and Zuscovitch, E (1988), 'Towards the economics of information-intensive production systems: the case of advanced materials', in G Dosi et al (eds), Technical Change and Economic Theory, Pinter, London.