The pattern of innovation in Singapore's manufacturing sector.

Abstract

This paper discusses the pattern of innovation in Singapore's manufacturing sector, using the results of the country's first national innovation system (NIS) survey. The results of the survey show that Singapore has a long way to go in its development of an innovation-based economy, especially in comparison to the OECD countries. However, progress has been made, with companies making visible efforts to engage in innovation activities. As has been found in other countries, the proportion of companies engaging in innovative activity in Singapore increases with firm size. We also found that innovation activity is positively related to sales volume, sales growth, employment growth, and internationalisation. The survey also reveals some weaknesses in the Singapore innovation system, such as insufficient collaboration between firms and the public R&D sector, deficiencies in the availability of scientific and technical manpower and innovation-supporting services, and adverse societal attitudes to failure.

Introduction

Innovation has been identified as a major driver of economic growth. A widely used model for studying innovation and its effect on the economy is the national innovation systems (NIS) model, which adopts a systemic approach to the way in which knowledge is produced and diffused throughout the economy (OECD, 1997). Attention has shifted from focusing solely on research and development as a source to innovation, to the interaction of those involved in all aspects of technological development, transfer, diffusion, and application. This includes companies, PRICs (public R&D institutions and centres), education and training institutions, and the government.

Many innovation surveys have attempted to capture the interactions between the various actors in the NIS on an industrial, regional, national, and even international scale, mostly in the Organisation for Economic Cooperation and Development (OECD) countries. Perhaps the best known of these are the Community Innovation Surveys, launched by the European Union, allowing MS comparisons between most of the western European countries.

However, few innovation surveys have been conducted in Asia, and prior to our study, none in Singapore. This is a deficiency that needs to be redressed as Singapore seeks to position itself as a hub in the global knowledge economy.

Since 1998, when the Committee on Singapore's Competitiveness released its report on the directions and strategies that should be taken to steer Singapore through the Asian financial crisis, government policy has increasingly focused on fostering economic growth through high-technology entrepreneurship and innovation (Wong, 2002).

Time has further refined this vision. The bursting of the Internet bubble in 2000 showed that a mere increase in the numbers of high-tech startups is not enough; genuine innovation in these companies is also needed (Wong, 2002). Moreover, since the MS model shows that it is the flow of knowledge that drives technological improvement, linkages between the various innovation actors (for example, collaboration between companies, or between companies and PRICs) are extremely important. From this viewpoint, the ultimate aim of policy is to remove hindrances to these flows, increasing the innovative capacity and therefore the economic well-being of the nation.

This is where the MS survey can play a role. It allows us to measure the performance of Singapore's national innovation system, providing a framework to assess the deployment and development of innovation in Singapore (Wong, 2002). By mapping the system, we can identify the sectors that require attention and the type of action required. This is the first step in correcting the bottlenecks that hinder the smooth flow of innovation and information through the economy.

To this end, Singapore's first national innovation system survey was conducted in 1999 by the National University of Singapore's Centre for Management of Innovation and Technopreneurship, with the support of the Economic Development Board. The objectives of this survey were twofold: (a) to assess the current status of innovation among manufacturing companies in Singapore and benchmark it against those in OECD countries, and (b) to identify key problems and challenges faced by these companies in innovation.

The next section explains our survey methodology. The rest of the paper presents the results of the survey.

Methodology

The survey was confined to companies in the manufacturing sector, a key sector within Singapore's NIS. Manufacturing is one of the "twin engines" of the Singapore economy, and its firms account for the majority of private sector R&D expenditure in Singapore (Wong, 2002).

Data were collected by postal survey, with questionnaires being sent out to CEOs of 1,876 companies. Our final sample consisted of 371 companies in the electronics, chemicals, precision and process engineering, and transport engineering sectors. It was decided that only these four industrial clusters would be surveyed, as they represent the most important industries in the Singapore manufacturing sector. Response rates differed only slightly between clusters, ranging from 22 per cent for the chemicals cluster to 19 per cent for electronics.

Before commencing the analysis, data entry was tested for integrity using random checks as well as explorative statistical procedures (frequencies, ranges, cross tabulations, etc). Missing data as well as doubtful or contradictory responses were clarified by telephone calls or in some cases removed from the sample.

The realised sample obtained from this survey represents 97,000 employees. Total annual sales amount to S$35 billion, while total fixed assets add up to S$22 billion.

Over half the responses were from the precision and process engineering companies. Chemicals companies contributed 19 per cent, transport engineering 16 per cent and electronics companies 11 per cent. Almost two-thirds of the responses were from local companies.

To reduce sample bias, the data were weighted throughout the analysis to represent the population in terms of industry structure and employment size class. Unless otherwise stated, all totals given in this paper have been grossed to represent the total population, that is, the four manufacturing clusters under study.

International Comparison

In order to benchmark innovation in the Singapore manufacturing sector against those of other countries, the results from this survey were compared with those from the two Community Innovation Surveys.

The first Community Innovation Survey (CIS) was a project organised by DGXIII/SPRINT/EIMS (the European Innovations Monitoring System) and Eurostat. Developed over 1991 to 1993, the survey collected firm-level data on innovation inputs and outputs from approximately 40,000 firms. Respondents were asked for information about their enterprise, sources of information for innovation, objectives of innovation, acquisition and transfer of technology, R&D activity, factors hampering innovation, costs of innovation, and impact of innovation activities. The countries involved in the CIS included all the European Union (EU) member states, Norway, and Iceland, making it the first internationally comparable innovation survey.

The Second Community Innovation Survey (CIS II) was launched in 1997/1998, and aimed for a higher comparability between countries than was achieved by the CIS. Data were collected for the year 1996 (1) and the results of the two surveys are generally not comparable.

Compatibility of Singapore and European Surveys

Industries

The CIS surveys included the whole range of manufacturing industries, classified according to the European Communities 'Statistical Classification of Economic Activities' (NACE Revision 1).

The Singapore survey on the other hand, excluded traditional industries such as textiles, clothing and leather; food; beverages and tobacco, and wood, paper and printing. These industries are generally less innovative. Consequently, the innovation figures for the Singapore manufacturing sector as a whole may be higher in comparison to the European manufacturing sector than they would otherwise be.

Innovation

Since all the three surveys based their definition of innovation on that outlined in the Oslo Manual, similar definitions of product and process innovation were used. That is, a product innovation is a product new to the business or a substantially improved product, while a process innovation is a new or substantially improved production process through new equipment or re-engineering.

The European surveys, however, provided more details regarding what did and did not constitute a technological innovation. This may have resulted in the European respondents having a narrower definition of innovation than the Singaporean respondents.

The definition of an innovator as an enterprise who had introduced any product or process innovations within the last three years was common to all three surveys. (2)

Similar definitions of innovation activities were also used in all three surveys. In particular, the definitions in the CIS II and the Singapore survey are virtually identical (see Table 1).

Innovation in the Manufacturing Sector

In total, about one-third (32 per cent) of manufacturing companies in Singapore are involved in innovation activities. However, since this does not given any indication of the commercial success of an innovation, a second distinction between innovative and non-innovative companies was made. A company is classified as innovative if it meets at least one of the following criteria: 25 per cent of sales accounted for by new or improved products introduced over the last three years; or 25 per cent of production volume accounted for by new or improved processes introduced over the last three years. About half of the Singapore innovating companies (15 per cent of manufacturing companies) were also classified as innovative.

The European countries generally have a higher percentage of manufacturing innovators than Singapore does (see Table 2). In 1996, only Spain and Belgium had less than one-third of their manufacturing companies as innovating enterprises.

Singapore has approximately equal proportions of companies involved in product and process innovation while most of the European countries have a higher proportion of product innovators over process innovators, with Norway and Spain being two exceptions. The higher incidence of process innovation in Singapore reflects the fact that Singapore's manufacturing is still dominated by foreign MNCs (multinational corporations), which tend to concentrate their product innovation activities in their home countries rather than in Singapore (Figure 1). Only 47 per cent of Singapore companies conducted both product and process innovation in 1999, compared to 63 per cent of European companies in 1996.

Characteristics of Innovators in the Manufacturing Sector

Size Class

Innovation is commonly thought to be related to firm size, and the results from the three surveys bear this out. The proportion of innovators increases with employment in both Singaporean and European companies (Table 3).

Singapore generally has a lower proportion of innovators in each size class, as compared with most of the Europeans countries in 1996. The proportion of innovators in each size class in Singapore is approximately equivalent to that of some of the smaller European countries, such as Belgium and Finland, as well as to Spain in 1996.

Cluster/Industry

In the Singapore NIS, the electronics cluster has the highest proportion of product and process innovators, with more than half of its companies being innovators (56 per cent for product innovation and 55 per cent for process innovation). This is to be expected, as the electronics industry also accounts most of manufacturing's R&D spending (over 60 per cent in 1999) (Wong, 2002). At the other end of the spectrum, the transport engineering cluster has the lowest proportion of innovators. Only 15 per cent of the companies within this cluster are product innovators and 13 per cent are process innovators.

When the classification of industries in the Singapore survey is modified to be compatible with the European surveys, it is found that, as with size classes, Europe in 1996 had a much higher proportion of innovators in each manufacturing industry than Singapore does. Singapore's results are closest to the small European countries of Belgium and Spain.

However, both Singapore and Europe have the machinery and electrical/electronics industry as the most innovative industry in their manufacturing sectors, as measured by the proportion of innovators within that industry. The proportion of innovators in this sector is 41 per cent and 71 per cent for Singapore and Europe respectively.

Manpower Development

On average, companies in our sample spend around 3 per cent of their payroll on training, with values ranging from nil to 29 per cent (Table 5). Almost a quarter of firms do not engage in training at all. The majority of companies spend up to 5 per cent of their payroll on training, while only 10 per cent reported training intensities of more than 5 per cent. Training intensity is highest in the electronics and chemicals clusters and below average for transport engineering firms.

Our data suggest a clear link between innovation and training, especially between successful innovation and training. Training intensity is much higher for innovating, innovative and R&D performing companies, the distinction being particularly pronounced and statistically significant for innovative as opposed to non-innovative firms.

Spin-offs

One of the ways that innovative capacity flows through an economy is by the creation of spin-offs. As personnel from an established company form their own enterprise, they bring with them the knowledge and skills they have learnt, allowing innovations and information to be passed on to their buyers, suppliers, and collaborators.

The spin-off rate in the Singapore manufacturing sector is quite low. Only 11.6 per cent of companies reported knowing of at least one firm started up by ex-employees during the last three years, and the majority of these (89 per cent) reported one or two spin-offs only (Table 6).

In relative terms, the precision and process engineering as well as the chemicals cluster are the most conducive breeding grounds for new ventures started by ex-employees. Also, foreign companies have a slightly higher spin-off rate than local firms, while companies engaging in innovation activity are more likely to spin off new ventures on average.

Performance of Innovators in Manufacturing Sector

One question of interest is how innovation impacts firm performance. Although more rigorous analysis than that presented here is required to draw more positive conclusions about causal factors, we can at least investigate the associations between innovation and sales volume, sales growth, employment growth and internationalisation. We also consider the success of manufacturing innovations in terms of the proportion of turnover coming from new/improved products, and the commercial success of new products.

Sales Volume

Our survey finds that sales volume is positively related to innovation. The share of both innovating and innovative companies in the Singapore manufacturing sector increases with sales volume, being above average for companies with sales of more than S$lO million (Figure 2). Only about a third of innovating and innovative companies, and about a quarter of R&D performers, reported sales below S$5 million, compared to well above half of non-innovating companies.

Sales Growth

Despite the Asian economic crisis, companies' sales grew by 9.1 per cent on average over 1996 to 1999 (Table 7). Our data suggest that sales growth was fastest in the electronics cluster, averaging 21 per cent per year, followed by the chemicals sector at 17 per cent per year. The engineering clusters reported average sales growth between 6 per cent and 8 per cent.

Innovating companies grew more strongly than non-innovating companies, though this difference is not significant statistically due to high data variance. In contrast, innovative companies representing the successful innovators grew significantly faster than non-innovative companies, confirming the link between successful innovation and firm growth.

Employment Growth

As with sales, employment also grew from 1996 to 1999, at an annual rate of 5 per cent (Table 8). Our data suggests that employment has in-creased the most in the electronics sector (by 6 per cent per year), while the other clusters have all grown by 5 per cent per year. These increases in employment are largely attributed to local companies which have expanded their staff by 6 per cent annually, compared by 3 per cent for the foreign companies.

Innovative and R&D performing companies have grown faster in terms of employment than companies not engaging in such activities. However, innovating companies have not. This confirms that new products and processes actually need to penetrate the market to benefit the labour market.

Internationalisation

Internationalisation is particularly important for Singapore companies, given the limited size of the domestic market. One measure of internationalisation is the proportion of sales that are sold in other countries.

Our survey found that 63 per cent of all sales are accounted for by the domestic market, while 37 per cent of sales are for export. The electronics cluster is the most internationalised sector, with an export intensity of 62 per cent, followed by the transport engineering and chemicals sector at 44 per cent.

Foreign-owned companies are naturally more internationalised than local firms. Companies engaging in innovation are also more oriented towards international markets than non-innovating companies, the mean export shares for innovating, innovative and R&D performing firms being significantly higher than for non-innovating companies.

Proportion of Turnover Due to New or Improved Products for Product Innovators

Successful innovation does not end with the creation of a new product or process; it must also be deployed. Commercialisation is one means of deploying a product innovation. One measure, then, of the success of an innovating company is the proportion of turnover that comes from innovative products. In total, more than half of the innovating companies (61 per cent) obtained less than one-quarter of their sales from new/improved products. This is also true for each of the individual clusters with the exception of the electronics cluster. Only 11 per cent of the companies derive at least 75 per cent of their sales from new/improved products.

By contrast, close to half of the innovating companies (46 per cent) produced less than 25 per cent of their production volume using new/improved products. 24 per cent of the companies produced at least three-quarters of their production volume using new/improved processes. The transport engineering sector stands out in this regard, with a corresponding figure of 43 per cent.

Using mid-point estimates, it was calculated that 29 per cent of sales turnover in Singaporean innovating companies was due to new or improved products in 1999. The corresponding figure for the European countries in 1996 ranged from 43 per cent for Germany to 14 per cent for Belgium (Table 10). Singapore's figure is similar to that of Austria and Sweden (31 per cent) and Spain (27 per cent).

Commercial Success of New Products

Another way to gauge the success of a company's innovation activities is to look at the volume of its new products which experience commercial success. In our survey, 'almost one-third of the companies (32 per cent) had at least three-quarters of their new products introduced over the last three years experience commercial success (Figure 3). The transport engineering and electronics sectors have been particularly successful, with a near majority of their companies having at least half of their new products achieving commercial success.

Innovators in the precision and process engineering sector, on the other hand, have been relatively unsuccessful, with 39 per cent having a commercial success rate of less than 25 per cent for their new products. However, over one-quarter of the innovators in this sector also had at least three-quarters of their new products achieving commercial success.

Innovation Activities and Intensities

Innovation Activities

One of the advantages of the NIS model is that it recognises non-R&D forms of innovation and technological development. This is particularly applicable for Singapore, where much innovation arises from the application of technologies that were developed and are already in use elsewhere.

Our survey highlights the importance of non-R&D innovation activities in Singapore. The most common innovation activity in the Singapore manufacturing sector is the acquisition of machinery, equipment and software linked to product and process innovation (80 per cent of innovating manufacturers), followed by R&D (66 per cent) (Figure 4). R&D is particularly common in the electronics and chemicals sectors (85 per cent and 79 per cent of innovators in each sector respectively), while in the engineering sectors, acquisition of machinery, equipment and software linked to product and process innovation is more common (100 per cent for transport engineering and 81 per cent for precision and process engineering).

The least common innovating activity is acquisition of R&D services, which was conducted by only 18 per cent of the innovating companies.

Innovation and R&D Intensities among Innovating Companies

Innovation and R&D intensities are generally low in Singapore manufacturing firms. The majority of innovators (75.5 per cent) spend less than 10 per cent of their total sales revenue on innovation (Figure 5). Conversely, only 1.8 per cent of the innovators had an innovation intensity of 40 per cent or more, and all of these companies were found in the precision and process engineering sector. (3)

Similarly, more than 80 per cent of the innovators spent less than 5 per cent of their sales revenue on R&D activities (Figure 6). The 4.2 per cent of innovating companies who spent at least one-fifth of their revenue on R&D were in the electronics and precision and process engineering sectors.

Objectives for Innovation Projects

Companies' objectives in undertaking innovation give an indication of factors that should be considered when designing incentives to encourage them to engage in, or extend, innovation activities.

Not surprisingly, the main motivations for undertaking innovation projects are market-related. Among all other factors, opening up new markets or increasing market share, improving product quality and reducing production cost were ranked the most important factors (Figure 7).

By contrast, environmental concerns are not major drivers for innovation, with reduction of environmental effects and energy consumption being rated as the least important factors.

External Collaboration in Innovation

Intensity of Collaboration

Recent theoretical models suggest that innovation is a highly interactive process. Linkages or interactions between and within sectors in a national innovation system are key determinants of its performance, for only when these linkages are adequately developed will innovations be effectively developed and deployed throughout the system (Wong, 2002).

This appears to be widely supported by our data. All innovating companies have reported some kind of external collaboration, 89 per cent even cited intense collaboration with at least one external party.

Overall collaboration intensity tends to increase with firm size, pointing to the transaction costs involved (see Table 11). Specifically, large companies collaborate more intensely with external parties, or cooperate with a wider range of partners, although small and medium sized firms may still have intense collaboration with a narrower set of partners.

Partners for Innovation Collaboration

Respondents to our survey cited customers and buyers as the single most relevant collaboration partner, highlighting the importance of the market to innovation (Figure 8). It also reflects the strong relationship that has developed in the enterprise sector between MNC buyers and local suppliers. This has been both a natural outcome of long-term buyer-supplier relationships, and a result of government policies designed to strengthen them, such as EDB's Local Industry Upgrading Programme and the Industry Cluster Development strategy (Wong, 2002).

The next most relevant collaboration partners are suppliers, parent or associate companies overseas, and technical service providers. Interaction with parent/associate companies overseas appears to be more important to product innovators than to process innovators who emphasise suppliers and technical service providers instead.

The importance of parent and associate companies is to be expected given the dominance of MNCs in Singapore. MNCs tend to look to their headquarters and associate companies for their technological needs, such as R&D, because in this way they can protect the intellectual property and technology generated. The government also promoted these linkages in order to develop innovation activities in Singapore, for example, by offering incentives to induce MNCs to send local employees to headquarters for training, or to locate higher value-added operations in Singapore (Wong, 2002).

By contrast, interactions between universities/PRICs and the enterprise sector are less developed, as reflected in our data. Overall, R&D institutes and universities both local and foreign are of lesser importance to Singapore innovators, as are business service providers, competitors, and other companies. The lack of collaboration between manufacturers and universities/PRICs is a cause of concern, because the public R&D and tertiary education sectors are a main source of basic research. Although basic research is usually too expensive for the private sector to engage in, adequate linkages would ensure they have access to the research conducted in universities/PRICs so they can apply it in developing their own innovations. A *Star (Agency for Science, Technology and Research) has already implemented measures to increase collaborations between the sectors, such as setting targets for PRICs to obtain a percentage of cost recovery from funding by industry, and monitoring their technology licensing to, and joint R&D with, industry (Wong, 2002). However, as our survey shows, there is still some way to go.

Reasons for Cooperation

Factors such as knowhow transfer, establishment of long-term strategic partnerships, entry to new technology fields, and faster time to market were rated as important reasons for cooperation by at least two-thirds of respondents (Figure 9).

By contrast, the sharing of risk and pooling of resources have been rated clearly as less important.

Problems in Cooperation

The main problem reported by companies collaborating in innovation relates to confidentiality and secrecy, followed by the inability to keep to the original schedule. Less dominant but still relevant are co-ordination problems, partners' different capabilities as well as unintended leakage of knowledge (Figure 10).

Successful innovators are less concerned about confidentiality, while non-innovative firms reported more problems arising from mismatches in capability.

Innovation Environment for Singapore's Manufacturing Companies

In this section, we turn to the innovation environment facing Singapore's manufacturing companies. This includes the environment internal to the company, and the wider environment, including aspects such as infrastructure, financial support and societal attitudes.

Companies' Internal Environment for Innovation

Asked to comment on nine statements relating to their company's internal environment for innovation, respondents were most supportive of the statement "Our top management is highly supportive of innovation", followed by "Our management tolerates failure and encourages our staff to learn from mistakes" (Figure 11).

Generally, statements on individuals' attitudes achieved higher levels of agreement while more 'systemic' or organisational aspects such as compensation systems, process management, cross-hierarchical information flows and corporate performance measurement systems rank at the bottom.

Hindrances to Innovation Projects

Generally, the four most significant factors hampering innovation projects are excessive costs and risk, inadequate information on information technology, and a lack of qualified manpower.

For projects that either did not start, or were terminated, excessive cost was reported as the most significant factor. By contrast, a lack of qualified personnel was the biggest factor in delayed projects.

Excessive government regulations and organisational rigidities were less significant factors.

National Innovation Environment of Singapore

Respondents were invited to assess a set of 16 aspects of Singapore's innovation environment. Outstanding at the top of the table is the positive assessment of Singapore's telecommunication and information technology infrastructure, outcome of the Intelligent Island's 1T2000 and related strategies to raise infrastructure levels to world class (Figure 12).

At the other end of the table, the low scores awarded to 'tolerance of failure' are just as outstanding, as was highlighted by Singapore's Prime Minister and the Senior Minister in their respective Chinese New Year's addresses in 2000. The low score for this aspect was confirmed by a number of text comments such as:

- "Loss of face."

- "One mistake society."

- "Expectation is to succeed, not 'fail to learn'."

- "Individual fear of failure (and associated face loss) is limiting."

- "They will tell you why you failed, but not how not to fail."

- "Left alone to swim until you sink."

- "Risk participation by government is lacking."

It is clear that this aversion to failure relates to both the individual (fear of failure and "loss of face") and the societal level. This needs to be resolved, as this fear of failure will stifle the creativity essential for innovation. It is also Singapore's main impediment to adopting the 'Silicon Valley model' of wealth creation by technopreneurship, in which failure is treated as an integral part of learning and achieving.

Conclusion

In this paper, we have discussed the pattern of innovation in Singapore's manufacturing sector, using the results of the country's first NIS survey. The results of the survey show that Singapore has a long way to go in its development of an innovation-based economy, especially in comparison to the OECD countries. However, it is also clear that some progress has been made, with companies making visible efforts to engage in innovation activities.

There are some important implications to be found in our data; for example, the finding that a higher proportion of big companies engage in innovative activity, as compared to small and medium-sized enterprises (SMEs). This is consistent with results found in other countries, and highlights the importance of providing encouragement and assistance for SMEs to conduct innovation, especially in the light of the rising interest in SMEs as vehicles for economic growth.

Secondly, we find that innovation activity is positively related to sales volume, sales growth, employment growth and internationalisation. Particularly noteworthy is the result that higher sales and employment growth is associated with innovative, not innovating, companies. Further analysis is required before we can conclusively establish whether the higher growth rates are a direct result of their successful innovation activities. If this is so, as seems likely, then this emphasises the fact that creation of innovation is not sufficient; successful deployment (for example, commercialisation) is also necessary for economic growth. The quality of innovations produced cannot be sacrificed in favour of quantity.

Lastly, the survey reveals some weaknesses in the Singapore innovation system. One is the relative lack of collaboration between the universities and PRICs. This is one bottleneck in Singapore's NIS which needs to be removed. Companies need access to the latest research from universities/ PRICs--especially basic research--in developing their own innovations, if they are to stay on the technological frontier. The availability of scientific and technical manpower and the sophistication of innovation-supporting services, including venture capital, also needs to be further enhanced.

Finally, Singapore society needs to learn to accept failure as an inherent part of the innovation and entrepreneurial process. If failure continues to be viewed as a stigma, rather than merely an opportunity to learn, Singapore's transformation into a knowledge economy will be greatly hampered.

[FIGURE 10 OMITTED]

Figure 1

International Comparison: Innovators by Type of Innovation

                                     Europe, 1997  Singapore, 1999

Process innovation only                  27%             29%
Product innovation only                  10%             24%
Both product and process innovation      63%             47%

Source for European data: Foyn, F (1999). "Community Innovation Survey
1997/1998". Statistics in Focus: Research and Development, 2/1999,
Eurostat

Note: Table made from pie chart

Figure 2

Total Annual Sales for Latest Fiscal Year (Percentage)

                         R&D performers  Innovative companies

S$500 million and above        7.8                7.0
S$100 to S$499 million        11.7               12.3
S$50 to S$99.9 million        11.7               10.5
S$10 to S$49.9 million        31.2               24.6
S$5 to S$9.9 million          11.7               10.5
Below S$5 million             26.0               35.1

                         Innovating companies  Non-innovating companies

S$500 million and above           5.2                     1.6
S$100 to S$499 million           11.2                     1.6
S$50 to S$99.9 million            8.6                     3.5
S$10 to S$49.9 million           29.3                    20.9
S$5 to S$9.9 million             11.2                    13.8
Below S$5 million                34.5                    58.7

Note: Table made from bar graph

Figure 3

Commerical Success of New Products Introduced * (Percentage)

                                   Below 25 per cent  25 to 49 per cent


Electronics                              11.8               23.5
Chemicals                                16.7               27.8
Precision and Process Engineering        39.7               19.6
Transport Engineering                    37.5                0.0
Total                                    29.2               20.2

                                   50 to 74 per cent  75 per cent and
                                                           above

Electronics                              17.6              47.1
Chemicals                                38.9              16.7
Precision and Process Engineering        13.0              28.3
Transport Engineering                    12.5              50.0
Total                                    19.1              31.5

* Percentage of new products introduced which were commercially
successful

Note: Table made from bar graph

Figure 4

Activities Engaged by Companies (Percentage of Innovating Firms)

                                   R&D   Acquisition  Acquisition
                                         of R&D svcs  of machinery




Electronics                        85.0     30.0          85.0
Chemicals                          78.9     15.8          63.2
Precision and Process Engineering  58.2     13.6          80.6
Transport Engineering              60.0     30.0         100.0
Total                              66.4     18.3          80.2

                                      Licensing of
                                   external technology
                                    linked to product
                                       and process
                                       innovation

Electronics                               42.9
Chemicals                                 22.2
Precision and Process Engineering         14.9
Transport Engineering                     30.0
Total                                     22.4

                                   Industrial design,
                                   market research and
                                   marketing expenses
                                       for product
                                       innovation

Electronics                               42.9
Chemicals                                 27.8
Precision and Process Engineering         40.3
Transport Engineering                     30.0
Total                                     37.9

                                   Training directly
                                       linked to
                                     technological
                                      innovations


Electronics                              81.0
Chemicals                                56.6
Precision and Process Engineering        56.7
Transport Engineering                    60.0
Total                                    61.2

Note: Table made from bar graph

Figure 5

Innovation Intensity *

                                   Less than   2 to 4.9  5 to 9.9
                                   2 per cent  per cent  per cent

Electronics                           20.0       25.0      20.0
Chemicals                             33.3       38.9      5.6
Precision and Process Engineering     29.5       24.6      23.0
Transport Engineering                 27.3       18.2      36.4
Total                                 28.2       26.4      20.9

                                   10 to 19.9  20 to 39.9  40 per cent
                                    per cent    per cent    and above

Electronics                           20.0        15.0         0.0
Chemicals                             16.7        5.6          0.0
Precision and Process Engineering     13.1        6.6          3.3
Transport Engineering                 18.2        0.0          0.0
Total                                 15.5        7.3          1.8

*Total expenditure on innovation as a percentage of total sales

Note: Table made from bar graph

Figure 6

R&D Intensity *

                                   Less then   1 to 2.9  3 to 4.9
                                   1 per cent  per cent  per cent

Electronics                           12.5       25.0      25.0
Chemicals                             20.0       46.7      20.0
Precision and Process Engineering     55.6       16.7      16.7
Transport Engineering                 20.0       20.0      40.0
Total                                 36.1       25.0      20.8

                                   5 to 9.9  10 to 19.9  20 per cent
                                   per cent   per cent    and above

Electronics                          18.8       12.5         6.3
Chemicals                            13.3        0.0         0.0
Precision and Process Engineering     2.8        2.8         5.6
Transport Engineering                20.0        0.0         0.0
Total                                 9.7        4.2         4.2

* Total expenditure on R&D as a percentage of total sales

Note: Table made from bar graph

Figure 7

Important Objectives for Undertaking Innovation Projects (Mean)

    Open up new markets or increase  4.2
                       market share
            Improve product quality  4.1
             Reduce production cost  4.0
               Extend product range  3.7
Reduce material consumption/improve  3.5
                              yield
                 Improve cycle time  3.3
  Replace products being phased out  3.1
     Improve production flexibility  3.1
    Fulfill regulations & standards  3.0
        Improve work conditions for  2.9
                          employees
         Reduce environment effects  2.7
          Reduce energy consumption  2.5

Note: Statements are ranked on a scale of 0 (not relevant) to 5
(very important)

Note: Table made from bar graph

Figure 8

External Collaboration on Innovation: Mean Relative Sorces * by Party

                Customers, buyers  191.9
Parent/associate company overseas  150.1
                        Suppliers  147.1
      Technical service providers  106.0
  R&D instituties/universities in   85.2
                        Singapore
       Business service providers   67.5
                      Other firms   59.8
      R&D institutes/universities   49.7
                         overseas
                      Competitors   42.7

*Mean score across all parties = 100

Note: Table made from bar graph

Figure 9

Reasons for External Collaboration on Innovation (Mean Sources *)

 Establish long term strategic  3.94
                   partnership
Entering new technology fields  3.94
             Know-how transfer  3.94
         Faster time to market  3.89
      Share/Reduce risk & cost  3.1
      Pool financial resources  2.61

* Scale from 1 = not important to 5 = very important

Note: Table made from bar graph

Figure 11

Internal Environment for Innovation (Mean Scores)

    Our top management is highly
        supportive of innovation  2.2

                                  2.4

     Our employees are very open
        to changes and new ideas  2.6

                                  2.6

   We encourage intrapreneurship
             among our employees  2.8

                                  2.9

Our middle management constantly
   delivers innovative proposals
                      to the top  2.9

                                  3.0

       Our corporate performance
      measurement system closely
         monitors our innovation
                     performance  3.0

Note: Statement are ranked on a scale of 1 (strong agreement) to 5
(strong diagreement

Figure 12

External Environment for Innovation (Mean Scores)

             Telecoms & IT services  3.85
   Intellectual property protection  3.33
Openness of govt depts & regulatory  3.29
                        authorities
              Government incentives  3.27
              Openness of customers  3.21
Listing requirements on local stock  3.2
                           exchange
              Openness of suppliers  3.14
                 Attitude of people  3.1
       Consultancy support services  3.08
                 Local universities  3.06
        Manpower in business sector  3.03
                   R&D institutions  3.00
Other technical supporting services  2.97
                    Venture capital  2.95
      Scientific-technical manpower  2.89
              Tolerance for failure  2.64

Scale from 1 = poor to 5 = good

Note: Table made from bar graph

Table 1

Comparison of Innovation Definitions used in the CIS, CIS II, and
Singapore NIS Surveys

CIS I                         CIS II

* R&D                         * R&D
* Acquisition of patents      * Acquisition of other
  and licences                  external technology
                                linked to product &
                                process innovation
* Product design              * Acquisition of R&D
                                services
* Market analysis             * Industrial design and
  (excluding launch costs)      other production
                                preparations for
                                technologically new or
                                improved products
* Investment in plant,        * Market introduction of
  machinery and equipment       technological
                                innovations


* Trial production, training  * Acquisition of
  and tooling-up                machinery and equipment
                                (including integrated
                                software) linked to
                                product & process
                                innovation
                              * Training directly
                                linked to technological
                                innovations

CIS I                         Singapore

* R&D                         * R&D
* Acquisition of patents      * Licensing of external
  and licences                  technology linked to
                                product & process
                                innovation
* Product design              * Acquisition of R&D
                                services
* Market analysis             * Industrial design,
  (excluding launch costs)      market research &
                                marketing expenses for
                                product innovation

* Investment in plant,        * Acquisition of
  machinery and equipment       machinery, equipment &
                                software linked to
                                product & process
                                innovation
* Trial production, training  * Training directly
  and tooling-up                linked to technological
                                innovations

Table 2

International Comparison: Innovating Enterprises in Manufacturing Sector

(Percentage of Total Number of Enterprises)

Country           1990/92  1996

Ireland             72      73
Germany             67      69 *
Austria             na      67
Netherlands         57      62
United Kingdom      na      59
Denmark             56      na
Sweden              na      54
Norway              53      48 **
France              39      43
Luxembourg          37      42
Finland             na      36
Singapore (1999)    na      32
Spain               37      29
Belgium             61      27
Italy               34      na

* Including ex-GDR from 1991

** Figure for 1997

Source for European data: European Communities. (1997) The First
Community Innovation Survey. Eurostat database, CD-ROM. European
Communities (1999). Innovat. Eurostat database, CD-ROM

Table 3

International Comparison: Innovators by Size Class

(Percentage of Firms in Size Class)

                  20-49  50-249  250

Singapore (1999)   26      43    65
Belgium            22      29    50
Germany            63      70    85
Spain              21      43    76
France             34      48    75
Ireland            68      78    85
Luxembourg         21      52    85
Netherlands        54      71    84
Austria            59      73    88
Finland            26      40    77
Sweden             43      61    79
United Kingdom     54      59    81
Norway             39      56    77

Note: In the Netherlands, the size classes are 20 to 49, 50 to 199, and
>200 Source for European data: Foyn, F (1999). "Community Innovation
Survey 1997/1998". Statistics in Focus: Research and Development,
2/1999, Eurostat

Table 4

International Comparison: Innovators by Industry

(Percentage of Firms in Industry)

                       Coke; chemicals;     Basic metal and
                       rubber and plastic;  fabricated products
                       other non-metallic
                       mineral

Singapore (1999)               30                   25
Belgium                        34                   30
Germany                        69                   59
Spain                          40                   25
France                         55                   31
Ireland                        79                   68
Luxembourg                     52                   44
Netherlands                    73                   53
Austria                        50                   68
Finland                        49                   31
Sweden                         59                   41
United Kingdom                 62                   56
Norway                         60                   43
Total European sample          58                   47

                       Machinery and      Transport
                       equipment NEC;     equipment and
                       electrical and     manufacturing
                       optical equipment       NEC

Singapore (1999)              41                16
Belgium                       44                25
Germany                       81                70
Spain                         50                30
France                        62                43
Ireland                       88                77
Luxembourg                    61                0
Netherlands                   78                59
Austria                       83                82
Finland                       44                28
Sweden                        74                58
United Kingdom                70                52
Norway                        64                47
Total European sample         71                52

Source for European data: Foyn, F (1999).

"Community Innovation Survey 1997/1998".

Statistics in Focus: Research and Development, 2/1999, Eurostat

Table 5

Spain-Out Intensity * for Selected Categories

Sample subset                    Training intensity

Total                                   2.56

Electronics                             2.87
Chemicals                               2.82 -
Precision & process engineering         2.62
Transport engineering                   1.92

Locally owned                           2.63 -
Foreign owned                           2.42

Innovating                              2.94 -
Non-innovating                          2.38

Innovative                              3.67 +
Non-innovative                          2.36

R&D performers                          3.23 -
Non-R&D performers                      2.23

*Expenditure on training as per cent of payroll.

+ ANOVA: Difference is significant at 5 per cent level.

- ANOVA: Difference is not significant at 5 per cent level.

Table 6

Spin-Out Intensity * for Selected Categories

Sample subset                    Percentage of companies

Total                                     11.6
Cluster
Electronics                                6.5
Chemicals                                 14.0
Precision & process engineering           13.7
Transport engineering                      3.6

Ownership
Local                                      9.1
Foreign                                   12.9

Number of employees
1-9                                       11.9
10-19                                     19.0
20-49                                      9.5
50-99                                      8.1
100-499                                   10.2
500 +                                     23.1

Innovation
Innovating                                17.9
Non-Innovating                             8.7
Innovative                                16.1
Non-Innovative                            10.8

*Percentage of companies reporting start-ups by ex-employees during the
last three years.

Table 7

Average Annual Sales Growth

Sample subset                    Sales growth (per cent)

Total                                      9.1

Electronics                               20.6
Chemicals                                 16.9 +
Precision & process engineering            6.2
Transport engineering                      7.7

Locally owned                              8.7 -
Foreign owned                             10.1

Innovating                                10.7 -
Non-innovating                             8.4

Innovative                                15.0
Non-innovative                             8.0 ++

++ ANOVA: Difference is significant at 10 per cent level.

+ ANOVA: Difference is significant at 5 per cent level.

- ANOVA: Difference is not significant at 5 per cent level.

Table 8

Average Annual Employment

Sample subset                    Employment growth (per cent)

Total                                        5.2

Electronics                                  6.3
Chemicals                                    5.4 -
Precision & process engineering              5.1
Transport engineering                        5.1

Locally owned                                6.2 -
Foreign owned                                3.4

Innovating                                   4.3 -
Non-innovating                               5.7

Innovative                                   8.8 -
Non-innovative                               4.6

R&D performers                               4.8 -
Non-R&D performers                           2.7

- ANOVA: Difference is not significant at 5 per cent level.

Table 9

Export Share * at Company Level for Selected Categories

Sample subset                    Share of exports
                                    (per cent)

Total                                  36.7

Electronics                            62.1 +++
Chemicals                              43.5
Precision & process engineering        30.0
Transport engineering                  44.9

Locally owned                          24.3 +++
Foreign owned                          62.5

Innovating                             48.5 +++
Non-innovating                         31.1

Innovative                             52.3 +++
Non-innovative                         33.8

R&D performers                         54.0 +
Non-R&D performers                     38.9

*Share of sales outside Singapore

+ ANOVA: Difference is significant at 5 per cent level

++ ANOVA: Difference is significant at 1 per cent level

+++ ANOVA: Difference is significant at 0.1 per cent level

- ANOVA: Difference is not significant at 5 per cent level

Table 10

International Comparison: Proportion of Turnover Due to New or Improved
Products for Product Innovators in Manufacturing Sector

Country           1996

Germany *          43
Ireland            32
Austria            31
Sweden             31
Singapore (1999)   29
Spain              27
Netherlands        25
Finland            25
United Kingdom     23
France             21
Norway (1997)      20
Belgium            14

Note: Singapore value is derived from mid-point estimate of categorised
variable.

*Including ex-GDR from 1991.

Source: Foyn, F (1999)". Community Innovation Survey 1997/1998."

Statistics in Focus: Research and Development, 2/1999, Eurostat

Table 11

Average Collaboration Intensity *

(Mean Scores, Innovation Firms Only)

Category                         Intensity

Total                              2.36

Process innovation                 2.29
Product innovation                 2.39

Electronics                        2.38
Chemicals                          2.48
Precision & process engineering    2.29
Transport engineering              2.62

Locally owned                      2.40
Foreign owned                      2.32

1-9 employees                      2.06
10-19 employees                    2.16
20-49 employees                    2.32
50-99 employees                    2.33
100-499 employees                  2.51
500+ employees                     2.63

Innovative                         2.34
Non-innovative                     2.39

*Measured on a scale from 0 = no collaboration to 5 = Intense
collaboration

End Notes

(1.) 1997 for Norway.

(2.) When looking at CIS results from individual countries, please note that Portugal only surveyed innovating firms, while the Greek sample was biased towards innovating firms, so their results are not representative of their manufacturing sectors as a whole. In addition, the figures for Germany from 1991 include the ex-German Democratic Republic.

(3.) One chemicals company in our sample also reported an innovation intensity of 40 per cent or more, but it does not appear in the analysis because the weight ascribed to it was relatively small.

References

Archibugi D, P Cohendet, A Kristensen and KA Schaffer, Evaluation of the Community Innovation Survey (CIS)-Phase I, EIMS Publication No 11, European Innovation Monitoring System, Luxembourg, 1995.

Bosworth D, P Stoneman and U Sinha, Technology Transfer Information Flows and Collaboration: An Analysis of the CIS, ELMS Publication No 36, European Innovation Monitoring System, Luxembourg, European Commission, 1996.

Confederation of British Industry (CBI), 1999 Innovation Trends Survey, Summary downloaded from http://www.cbi.org.uk/innovation/, 15 November 1999.

European Communities, The First Community Innovation Survey. Eurostat database, CD-ROM, 1997.

_____, Innovat, Eurostat database, CD-ROM, 1999.

European Commission, European Trend Chart on Innovation: The European Innovation Scoreboard, European Commission, Directorate General Enterprises, "Innovation And SME" Programme, 2000.

Foyn F, "Community Innovation Survey 1997/1998", Statistics in Focus: Research and Development, 2/1999, Eurostat, 1999.

Galli R and M Teubal. "Paradigmatic Shifts in National Innovation Systems", in Systems of Innovation: Technologies, Institutions & Organizations, C Edquist, ed, London, Pinter, 1997.

Muzart G, Description of National Innovation Surveys Carried Out, or Foreseen, in 1997-99 in OECD Non-CIS-2 Participants and NESTI Observer Countries, STI Working Papers, DSTI/DOC(99)1, Directorate for Science, Technology and Industry, Organisation for Economic Cooperation and Development, 1999.

Nelson RR and N Rosenberg, "Technical Innovation and National Systems", in National Innovation Systems: A Comparative Analysis, R R Nelson, ed, New York, Oxford University Press, 1993.

OECD, National Innovation Systems, http://www.oecd.org/dsti/nispub.html, 1997.

_____, Oslo Manual: Proposed Guidelines for Collecting and Interpreting Technological Innovation Data, Paris, Organisation for Economic Cooperation and Development, 1997a.

Sandven T and K Smith, Innovation Activities and Industrial Structure: Industry and R&D in a Comparative Context, EIMS Publication No 1, European Innovation Monitoring System, Luxembourg, 1993.

SPRU, Innovation Outputs in European Industry: Analysis from the CIS, EIMS Publication No 34, European Innovation Monitoring System, Luxembourg, European Commission, 1996:

STEP Group, Innovation Expenditures in European Industry: Patterns of Innovation Input, Innovation Expenditures, Non-Research and Intangible Inputs. Analysis Using CIS, EIMS Publication No 48, European Innovation Monitoring System, Luxembourg, European Commission, 1997.

Twente University, Innovation Measurement and Policies: Proceedings of the International Conference, Luxembourg, 20-21 May 1996, EIMS Publication No. 50, European Innovation Monitoring System, Luxembourg, European Commission, 1997

Wong PK, National Innovation System: The Case of Singapore, Seoul, Science & Technology Policy Institute, 1995.

_____, "From Using to Creating Technology: The Evolution of Singapore's National Innovation System and the Changing Role of Public Policy", in Technology Policy in East Asia, S Lall et al, eds, World Bank and Elgar Press, 2002.

Wyckoff A, Investment, Innovation and Competitiveness: Sectoral Performance within the Triad, EIMS Publication No 2, European Innovation Monitoring System, Luxembourg, 1993.

Zentrum fur Europaische Wirtschaftsforschung (ZEW), The Impact of Innovation on Employment in Europe: An Analysis using CIS Data, EIMS Publication No 46, European Innovation Monitoring System, Luxembourg, European Commission, 1997.