Content uploaded by John R. Baldwin
Author content
All content in this area was uploaded by John R. Baldwin on Jan 21, 2015
Content may be subject to copyright.
Innovation and Training in New Firms
by
John Baldwin
No. 123
11F0019MPE No. 123
ISSN: 1200-5223
ISBN: 0-660-18314-5
Micro-Economic Analysis Division
Statistics Canada
24th floor, R.H. Coats Bldg.
Ottawa, K1A 0T6
(613) 951-8588
baldjoh@statcan.ca
October, 2000
This paper was presented to a conference devoted to “Assessing the Impact of Training on the
Performance of Small and Medium Sized Enterprises” held jointly by the Centre for Business
Research, University of Cambridge and the Centre for Small and Medium Sized Enterprises,
University of Warwick, March 1998.
This paper represents the views of the author and does not necessarily reflect the opinions of
Statistics Canada.
Aussi disponible en français
Analytical Studies Branch – Research Paper Series - iii - Statistics Canada No. 11F0019MPE No. 123
Table of Contents
ABSTRACT................................................................................................................................................................ V
ACKNOWLEDGEMENTS.................................................................................................................................... VII
1. INTRODUCTION.................................................................................................................................................. 1
2. THE SURVEY OF ENTRANTS........................................................................................................................... 3
3. A PROFILE OF ENTRANTS............................................................................................................................... 5
3.1 THE COMPETITIVE ENVIRONMENT ...................................................................................................................... 5
3.2 THE FOCUS OF ENTRANTS................................................................................................................................... 8
4. ENTRANTS AND HUMAN RESOURCE STRATEGIES...............................................................................11
5. INNOVATION AND ENTRANTS..................................................................................................................... 13
5.1 INNOVATIVE COMPETENCIES.............................................................................................................................14
5.2 TECHNOLOGICAL COMPETENCIES .....................................................................................................................16
5.3 COMPREHENSIVENESS OF INNOVATION IN THE ENTRANT POPULATION............................................................. 16
6. INNOVATION AND HUMAN CAPITAL ........................................................................................................19
6.1 INDUSTRY DIFFERENCES IN INNOVATION, TECHNOLOGY AND FIRMS’ EMPHASES ON SKILLED LABOUR .......... 19
7. GROWTH OF ENTRANTS AND THE IMPORTANCE OF INNOVATION...............................................23
7.1 GROWTH DIFFERENCES ACROSS ALL MARKETS ............................................................................................... 23
7.2 GROWTH DIFFERENCES IN NEW AND MATURE MARKETS ................................................................................. 27
8. THE JOINT EFFECT OF INNOVATION, TECHNOLOGICAL AND SKILL COMPETENCIES ON
GROWTH.................................................................................................................................................................. 32
9. CONCLUSION......................................................................................................................................................36
REFERENCES.......................................................................................................................................................... 37
Analytical Studies Branch – Research Paper Series - v - Statistics Canada No. 11F0019MPE No. 123
Abstract
Recent studies have demonstrated the quantitative importance of entry, exit, growth and decline
in the industrial population. It is this turnover that rewards innovative activity and contributes to
productivity growth.
While the size of the entry population is impressive—especially when cumulated over time—the
importance of entry is ultimately due to its impact on innovation in the economy.
Experimentation is important in a dynamic, market-based economy. A key part of the
experimentation comes from entrants. New entrepreneurs constantly offer consumers new
products—both in terms of the basic good and the level of service that accompanies it.
This experimentation is associated with significant costs since many entrants fail. Young firms
are most at risk of failure; data drawn from a longitudinal file of Canadian entrants in both the
goods and service sectors show that over half the new firms that fail do so in the first two years
of life. Life is short for the majority of entrants. Only 1 in 5 new firms survive to their tenth
birthday.
Since so many entrants fall by the wayside, it is of inherent interest to understand the conditions
that are associated with success, the conditions that allow the potential in new entrepreneurs to
come to fruition. The success of an entrant is due to its choosing the correct combination of
strategies and activities. To understand how these capabilities contribute to growth, it is
necessary to study how the performance of entrants relates to differences in strategies and
pursued activities.
This paper describes the environment and the characteristics of entrants that manage to survive
and grow. In doing so, it focuses on two issues. The first is the innovativeness of entrants and the
extent to which their growth depends on their innovativeness. The second is to outline how the
stress on worker skills, which is partially related to training, complements innovation and
contributes to growth.
Keywords: innovation, training
Analytical Studies Branch – Research Paper Series - vii - Statistics Canada No. 11F0019MPE No. 123
Acknowledgements
I am indebted to Joanne Johnson who assisted me with the entry survey that formed the basis of
the data used in this paper and to Daniel Stripinis and Bob Gibson who assisted with the
tabulations used herein.
Analytical Studies Branch – Research Paper Series - 1 - Statistics Canada No. 11F0019MPE No. 123
1. Introduction
Recent studies have demonstrated the quantitative importance of entry, exit, growth and decline
in the industrial population.1 It is this turnover that rewards innovative activity and contributes to
productivity growth.
The importance of entry can be gauged in the first instance by its size. The annual entry rates for
the manufacturing sector (defined as the employment in entrants divided by total employment)
from 1972 to 1986 averaged about 1.9% for the United States and 2.4% for Canada (Baldwin,
Dunne, and Haltiwanger, 1995, p. 127). When calculated over five-year periods (1972-77, and
1977-82), the US average across two-digit industries was 11.3%, while the Canadian was 10.9%
(Baldwin, Dunne, and Haltiwanger, 1995, p. 137).
These rates need to be set in context. Entry is not the only process that causes turnover in the firm
population. Growth and decline also occur in the incumbent population. The amount of market
share transferred from all declining to all growing firms (including both exits and entry) in a
typical Canadian manufacturing industry over a decade ranges between 30% and 40% (Baldwin,
1995).
A significant part of this turnover comes from the entry and exit of firms. At any point in time,
the population can be divided into those that grow and those that decline. Job turnover is the sum
of all employment increases in those growing (job growth) and the sum of the change in
employment in those declining (job loss). The importance of entry and exit can be calculated as
the share of job growth that is accounted for by entry, or the share of job decline that is accounted
for by exit. When this is done using comparable manufacturing databases for Canada and the
United States, the results are quite similar. For year to year changes from 1970 to 1985, job
creation due to entry in the United States averaged 21% of the total job increase; in Canada, it
averaged 27%; (Baldwin, Dunne and Haltiwanger, 1995, p. 126). Calculated over two
comparable five-year periods (1972-77, 1977-82), entry accounted for 44% in the United States
and 45% in Canada. As the length of period over which the importance of entry is calculated,
entry becomes relatively more important.
While the size of the entry population is impressive—especially when cumulated over time—the
importance of entry is ultimately due to its impact on innovation in the economy.
Experimentation is important in a dynamic, market-based economy. A key part of the
experimentation comes from entrants. New entrepreneurs constantly offer consumers new
products—both in terms of the basic good and the level of service that accompanies it.
By doing so, new firms provide an important stimulus to the industrial population. A few small
entrants grow to become the new dynamos of the industrial system. Others remain relatively
small but provide an important source of innovation in the small-firm sector—especially when it
comes to quality differentiation. Smaller firms excel in their ability to provide quality and
1 See Baldwin (1995).
Analytical Studies Branch – Research Paper Series - 2 - Statistics Canada No. 11F0019MPE No. 123
flexibility of service (Baldwin et al., 1994). Small firms are constantly changing their product
offerings—with respect to both types of products and services offered. Small firms are adept at
ascertaining changing consumer tastes with regards to the amount of services that are bundled
with a product, or being flexible with regards to other aspects of the product offering. New small
firms that are better able to sense consumer requirements are constantly replacing other small
firms that are less able to do so (Baldwin, 1995). One manifestation of the success of small
entrants is their tendency to pay higher wages and to be more productive than those firms that
they force out of the market (Baldwin, 1995; 1996). It is the process of entry and exit that
generates information on which combinations of products and services best satisfy consumer
tastes.
This experimentation is associated with significant costs since many entrants fail. Young firms
are most at risk of failure; data drawn from a longitudinal file of Canadian entrants in both the
goods and service sectors show that over half the new firms that fail do so in the first two years
of life.2 Life is short for the majority of entrants. Only 1 in 5 new firms survive to their tenth
birthday.3
These failures involve a cost—both in human and financial terms. Failure is accompanied by the
expenditure of both dollars and time on the part of new entrepreneurs. This is the investment that
the market economy makes while experimenting in finding the new and improved goods and
services that consumers want. It can also be regarded as an investment in managerial experience,
because some entrepreneurs who fail will learn from their experiences and go on to found other
new businesses that eventually succeed. These resources should not be regarded as wasted—
anymore than the resources that are expended on obtaining information in a world of imperfect
information are wasted. Failures are an investment that society makes in the dynamic competitive
process.
Since so many entrants fall by the wayside, it is of inherent interest to understand the conditions
that are associated with success, the conditions that allow the potential in new entrepreneurs to
come to fruition. The success of an entrant is due to its choosing the correct combination of
strategies and activities. To understand how these capabilities contribute to growth, it is
necessary to study how the performance of entrants relates to differences in strategies and
pursued activities.
The objective of this paper is to describe the environment and the characteristics of entrants that
manage to survive and grow.4 In doing so, it focuses on two issues. The first is the
innovativeness of entrants and the extent to which their growth depends on their innovativeness.
The second is to outline how the stress on worker skills, which is partially related to training,
complements innovation and contributes to growth.
2 See Baldwin and Johnson (1999).
3 See Baldwin, Bian, Dupuy and Gellatly (2000).
4 See Baldwin, T. Gray et al. (1997) for a study of the characteristics of new firms that are associated with exit.
Analytical Studies Branch – Research Paper Series - 3 - Statistics Canada No. 11F0019MPE No. 123
2. The Survey of Entrants
The development of longitudinal databases on exit and entry have allowed us to better estimate
the amount of entry taking place and to determine that this aspect of turnover shifts significant
amounts of market share and changes the identity of market participants. But the longitudinal
databases derived from administrative records do not contain very rich descriptions of the
participants. Survey data has to be developed if we are to better understand the underlying
characteristics of entrants—their competencies not only with regard to innovation but also in
such areas as human resource development, marketing, and management; their financial
structure, their training activities and how this relates to innovation. When survey data are
developed and linked to longitudinal data on firm performance, such as sales or profitability,
differences in competencies that are related to performance differences can be outlined. Here we
use a survey of entrants to outline the type of environment that entrants face and the connection
between growth, innovation and associated competencies that innovative firms develop.
We make use of firm-level data on entrants to examine the nature of their innovative capabilities.
The firm-level data are taken from The Survey of Operating and Financing Practices of entrants
performed by Statistics Canada. The focus on new firms permits us to examine the innovative
capabilities of entrants as opposed to small firms in general. The richness of the survey database
allows both output and input-based measures of innovation, technological and skill-based
competencies of smaller firms to be developed.
The survey focused on new entrants that emerged from their early childhood and survived to their
early teen years. In light of the high death rate of new firms, these are the more successful
entrants. The frame consisted of all entrants to the commercial sector (both goods and services)
in the period 1983-86 that survived to 1993; the survey was conducted in 1996. The sample
included 3,991 firms from both the goods and services sector. The response rate to the survey
was 80%.
The survey provides a broad overview of the financing and operating practices of entrants. The
questionnaire contains information on 1) management—the extent of managerial and industry
experience and the degree of ownership in the firm 2) the nature of the competitive
environment—whether products quickly become obsolete, whether production technology
changes rapidly, whether the threat of new entry is high 3) the firm’s competencies in the area of
management, technology, human resources, financing, marketing and production, with special
attention paid to various facets of competencies in each area 4) the extent of financial planning 5)
the importance of investment in R&D, technology, and training 6) whether the firm engaged in
formal training and 7) the manner by which various activities were financed.
While the entrant survey offers a wide and varied set of questions that can be used to gauge the
technological prowess of entrants, there are potential problems that should be addressed. First,
many of the questions are subjective. The potential problems with subjective questions are well
known. The most important is that subjective questions dealing with subjects that are inherently
difficult to evaluate or out of the range of a respondent’s experience provide information that is
Analytical Studies Branch – Research Paper Series - 4 - Statistics Canada No. 11F0019MPE No. 123
unreliable. The classic example is a question that asks subjects to rank various shades of blue by
their ‘blueness’ or asks someone to rank their abilities to perform a task with which they are
unfamiliar.
Although many of the questions of this survey are subjective, they fall within the range of
experience of the managers of new firms. The questions that deal with the magnitude of
competitive forces and importance of competencies in various areas are all questions that
business managers ask themselves on an ongoing basis. Competitive forces require firms to
compare themselves against their competitors. The practice of benchmarking, for instance, has
led many firms to assess themselves against industry leaders on a continuous basis.
In order to minimize ambiguity, the questions in the survey were all tested extensively with
survey managers. Some questions were taken from previous surveys. The survey also addressed
the subjectivity problem by including objective measures of activity in areas where it was
feasible. The inclusion of parallel questions on activity provides an independent check on the
validity of the answers to the subjective questions. Answers to the question about the emphasis
given to R&D can be compared to answers regarding the percentage of investment devoted to
R&D. Answers to the question about the emphasis given to training can be compared to answers
regarding whether training is done. When these comparisons are made, the subjective valuation
given by a firm to an activity and the probability that an action is undertaken are found to be
closely related.
Another potential problem arises in business surveys when respondents taken from different
areas in a company have different views of the importance being given to different strategies.
This is a particular problem when the views being solicited deal with prescriptive issues—what
should be done (as opposed to a recollection of actual events), or what emphasis has been given
in the past to these issues. Since The Survey of Operating and Financial Practices of Entrants
focuses on the latter set of questions, the potential for this problem with the data is lessened.
There is a second reason for arguing that this is a relatively unimportant problem with the survey.
Although information loss in a large company may result in some senior personnel not fully
appreciating the directions that the firm is taking, the survey being used here focuses on smaller
firms, where all senior personnel are close to ongoing events. Since the survey was usually filled
in by senior management, variability of responses across individuals is likely to be less of a
problem.
Finally, there are other issues related to the nature of measurement that derives from an ordinal
Likert scale. The measurement issue is handled here by using the extreme score measure (the
percentage of firms scoring 4 and 5 on the five-point scale) to capture all firms that assessed
themselves as higher than the median category. This gives a robust measure of the percentage of
firms that felt they were above the midpoint of the distribution of scores that was given to
them—but does not worry about distinctions above this point.
The survey data, by itself, provides a profile of the broad strategies of entrants—the marketing
and production strategies that they pursue, the innovations that they introduce and the training
activities of the firm. It is linked to administrative data on sales, assets, and the wage bill of each
Analytical Studies Branch – Research Paper Series - 5 - Statistics Canada No. 11F0019MPE No. 123
firm since birth. In doing so, we develop objective measures of performance by which the sample
can be divided so as to investigate the strategies that are associated with success. For the purpose
of this paper, performance is measured in terms of growth of output. The growth rate is measured
as the compound rate of growth in sales from the first year after birth to 1993.
3. A Profile of Entrants
This section of the paper begins the process of describing the strategies that the select group of
entrants who survive to their early teen years pursue. It will outline the product, marketing, and
production strategies that are given the most emphasis. In later sections, we turn to the emphasis
that is placed on worker skills, the importance of innovation and technology and finally how both
affect the growth prospects of entrants. But first we describe the competitive environment that
conditions the activities that entrants pursue.
3.1 The Competitive Environment
The competitive environment that new entrants face affects the skills required for survival and
growth. Competition has many dimensions: it depends on the type of rivals an entrant faces, the
pressures placed upon it by buyers and suppliers, and the rapidity of changes in products and
technology.
The type of competition in an industry is partially determined by the maturity of the market.
Industries vary by the stage of development of the market for their primary product. The
introductory stage consists of those markets where the product demand is just starting to grow,
but the product is unknown to many potential users; the growth stage occurs when product
demand is growing and the product is becoming familiar to many potential users; the maturity
stage is when product demand growth is slowing and the product is familiar to most potential
users; finally post-maturity occurs when no growth in product demand occurs and there are few
potential new users. The stage of the product is expected to influence the firm because previous
work (Gort and Klepper, 1982) suggests that early stages in the product lifecycle involve a high
degree of uncertainty. Product and technological innovations follow one another in quick
succession. In later phases, the types of problems that a firm faces change. Reducing production
costs via technological change becomes more important.
Despite the fact that successful entrants are new, they generally serve mature markets. While
29% of successful entrants are in a growing product market, 50% are in mature markets.
Moreover, a greater percentage is in the post-maturity phase (18%) than the introductory phase
(3%). As the lifecycle model would have predicted, successful entrants as a whole reported more
rapid technological than product obsolescence. When asked to indicate if they felt that products
quickly become obsolete and production technology changes rapidly in their industry, only 24%
Analytical Studies Branch – Research Paper Series - 6 - Statistics Canada No. 11F0019MPE No. 123
of entrants felt that product obsolescence was rapid in their industry5 (Figure 1). Yet, 45% said
production technology changed rapidly.
The number of competitors also serves as a measure of the amount of competition that entrants
face. About 40% face between 5 and 19 competitors. Another 38% compete with over 20 firms.
However, the number of competitors is only a rough proxy for competitiveness; firms face
competition from potential as well as existing competitors. Even when the number of
competitors is small, rivalry can be intense.6 In order to gauge the intensity of competition,
entrants were asked if they disagreed or agreed with two propositions 1) that the threat of entry
was high, and 2) that their competitors’ actions were predictable. Most entrants (41%) felt their
competitors’ actions were easy to predict. However, some 61% of entrants felt that threats from
entry were high (Figure 1).
Customer relations also affect the nature of the competitive environment. Firms with only one
customer face uncertainty due to bilateral bargaining and the loss of the customer. Firms with
few repeat customers cannot build customer loyalty. Neither factor is very important for
successful entrants. Over half obtain less than 10% of their revenue from one customer and over
two-thirds of their customers are repeat customers.
Uncertainty also arises when consumer demand is difficult to predict and/or consumers can easily
substitute among competing products. The ease of substitutability represents the largest source of
uncertainty for entrants, as almost 60% of these firms felt consumers could easily substitute
competing products (Figure 1)7. Unpredictability of consumer demand was less of a problem;
just 40% of successful entrants rated this hard to predict.8
The competitive environment that successful entrants face is also affected by the nature of
product competition. Entrants ranked competition in their industry on a scale of 1 (low) to 5
(high) in seven areas—price, customer service, quality, and flexibility in responding to
customers, product range, product customization, and the frequency of introducing
new/improved products. The percentage of successful entrants that ranked each area as highly
competitive (4 or 5) is plotted in Figure 2. In keeping with their occupying mainly mature
markets, successful entrants report that competition in their industry is greatest with respect to
price, customer service and quality. In contrast, factors that mark growth industries—
customization or introducing new products—are less important.
5 That is, they scored 3, 4 or 5 on a scale of 1 to 5.
6 See Baldwin (1995) for data that show that the intensity of competition, as measured by market share turnover, is
not closely related to concentration.
7 That is, they ranked this possibility with a 3, 4 or 5.
8 Ibid.
Analytical Studies Branch – Research Paper Series - 7 - Statistics Canada No. 11F0019MPE No. 123
Figure 1. Entrants’ Perceptions About Their Industry Environment
Competitors are predictable
010203040506070
percent of firms
Threat of entry high
Product changes rapidly
Technology changes rapidly
Demand is unpredictable
Consumer s can easily subst itute
Agree
Disagree
Figure 2. Percentage of Entrants Reporting Intense Industry Competition
01020304050607080
p
ercent of firms
Price
Customer Service
Quality
Flexibi lity in r esponding to
customers
Product range
Product customizatio
n
Introducing new/improved
products
Figure 3. Innovators' and Noninnovators' Perceptions of Industry Risk
N
ew entrants are a constant threat
Liquidation value below purchase
Technology changes rapidly
Demand is unpredictable
Competitors can easily substitute
among suppliers
Competitors are unpredictable
Products quickly become obsolete
Consumers cannot easily substitute
0 10203040506070
% of firms that agree
Innovators
N
on-innovators
Analytical Studies Branch – Research Paper Series - 8 - Statistics Canada No. 11F0019MPE No. 123
Do innovators face a quieter environment than non-innovators? Is innovation in the small-firm
segment encouraged by concentrated market structures? The lifecycle model of entry offered by
Abernathy and Utterbach (1978) or by Gort and Klepper (1982) would suggest that innovative
entrants generally should be found in highly fluid, highly competitive situations.
In order to examine this issue, entrants are defined to be innovative if they are introducing new
products or processes; then differences between entrants that are introducing new products and
processes and those that are not doing so are examined.
How does the competitive environment of innovators differ from non-innovators? First,
innovating successful entrants face more competitors; only three-quarters of non-innovative
entrants face more than four competitors, compared to 87% of innovators.
Second, innovators are generally found in segments of industries where certain key aspects of
competition focus on innovation. When the perceptions of innovators and non-innovators about
the nature of competition that they face are compared (Figure 3), it is clear that changes that are
related to innovation are far more intense in the innovators’ industries. Technology is more likely
to be changing rapidly. Products are more likely to face rapid obsolescence. Demand is
unpredictable, probably because competitor actions are also more difficult to forecast.
It should be pointed out that, in other environmental areas, competition is just as intense for both
innovators and non-innovators alike. The threat of entry is high everywhere. Firms are just as
capable of switching suppliers. Both groups face the same threat that consumers can easily
substitute products if they should so choose.
While small firms may have to find a niche strategy in order to survive, the innovators among
them do not lead a protected existence. In general, all entrants continuously face the threat of
entry. Innovators are also faced with continuous changes as a result of new product introduction
and technological change that are related to the innovative nature of their industry.
3.2 The Focus of Entrants
What are the product and production strategies adopted by new firms in the face of the
environment that they face? A firm’s product-based strategies are directed at making their
existing products as attractive as possible to consumers. There are several ways in which firms
can do this: they can offer an attractive price, focus on quality, and strive to provide superior
customer service, or offer flexibility in meeting their customers’ needs. Alternatively, firms can
try to alter their product line. In doing so, they might choose to customize their products, develop
a product line that carries a wide range of related products, or continually expand and update
their product line by frequently introducing new/improved products.
Of these strategies, successful entrants give the highest priority (scored on a scale of 1 to 5) to
strategies related to quality and service. Each of the strategies here—quality, customer service,
and flexibility in responding to customers and price—are deemed to be important (Figure 4).
Analytical Studies Branch – Research Paper Series - 9 - Statistics Canada No. 11F0019MPE No. 123
Figure 4. Importance of Product-Based Strategies
012345
average score
Introducing new/improved
products
Customization of products
Offering a wide range of
related products
Price
Flexibility in responding
to customers
Customer service
Quality
Figure 5. Importance of Market-Based Strategies
012345
average score
Targeting new foreign
markets
Targeting new domestic
markets
Improving position in
existing markets
Promoting company/
product reputation
Satisfying existing
customers
Figure 6. Importance of Production Strategies
012345
average score
Using computer
controlled processes
Reducing production
times
Improving efficienc
y
of input use
Using high qualit
y
suppliers
Analytical Studies Branch – Research Paper Series - 10 - Statistics Canada No. 11F0019MPE No. 123
Alternate strategies that involve updating, expanding or enhancing their product line are
perceived to be less important by successful entrants.
The quality-oriented niche strategies are aimed at maintaining existing rather than attracting new
customers. Successful entrants concentrate their marketing strategies on their existing market.
This broad strategy includes specific strategies such as: “satisfying its existing customers”, or
slightly more aggressive strategies directed at “promoting the reputation of the company and its
products” and “improving position in existing markets”. Successful entrants, on average, place
less value on capturing new markets, be they domestic or foreign (Figure 5).
The third component examined here is the production strategy of entrants. They may seek to
improve their production by doing it better, doing it faster, doing it more efficiently, or using
better inputs. To do so, they may aim to reduce their use of material inputs; they may strive to
reduce their production times; they may focus on the functioning of their production processes by
introducing integrated computer controlled processes; or they may stress the importance of using
high quality suppliers.
Corresponding to the high score that successful entrants give to quality as part of their product
strategy, “using high quality suppliers” is rated the most important production strategy (Figure 6).
Improving efficiency of input use is next in importance, followed by reducing production times
and using computer controlled processes.
This picture of emerging successful entrants confirms the finding of other studies on small firms
(Baldwin, et al. 1994, and D’Amboise, 1991). The success of small firms depends on their ability
to produce a high quality output; their comparative advantage is the flexibility that allows them
to provide quick and efficient service. Successful entrants develop a customer-oriented business
focus. Their product strategies are aimed at enhancing the attractiveness of their current products
in their existing market: they focus on quality and responsiveness to customer needs; and their
process strategies are concentrated on improving the efficiency and quality of the production
process.
Analytical Studies Branch – Research Paper Series - 11 - Statistics Canada No. 11F0019MPE No. 123
4. Entrants and Human Resource Strategies
Managers of surviving entrants consider human resources to be critical to their success. Firms’
product, marketing and production strategies are more likely to be implemented successfully by a
workforce that is skilled and committed. Firms have to worry about obtaining the right quality of
workers. They have to make choices between hiring skilled workers, investing in their training
once they have been hired, and providing the correct motivation that will bring out the best in the
workforce.
A human resource strategy for new firms then involves three steps—finding skilled workers,
developing and nurturing the skills of their workers and keeping the best workers. Managers of
new firms rated each of these on a score of 0—no importance or not applicable—to 5—high
importance. The percent who felt strongly (a score of 4 or 5) that their ongoing success was the
result of their emphasis on recruiting skilled employees or training their existing employees was
about 55% (Figure 7). Some 32% stressed the importance of incentive compensation plans to
their ongoing success.
Data on training activities confirm the subjective emphasis that the firms give to their human
resource strategy. Just over half (52%) of successful entrants provided formal on- or off-the-job
training to some of their employees. Moreover some 32% reported investment expenditures for
training.9 Money spent on training accounted for an average of 22% of investment expenditures
across respondents who trained and also reported investment expenditures.
The value that a firm places on the importance of training is strongly related to the actual
undertaking of training. The higher the importance that a firm attributes to training, the more
likely it is to train (Figure 8).
9 Investment figures usually underestimate the extent to which small firms engage in training since firms report
having great difficulty in calculating their training expenditures. The firm typically has a record of its other
investment expenditures, as it would receive an invoice for any such expenditure. The problem with training
expenditures essentially lies in the fact that firms do not always know what to include in the estimates or they do not
separate out such expenses.
Analytical Studies Branch – Research Paper Series - 12 - Statistics Canada No. 11F0019MPE No. 123
Figure 7. The Importance of Human Resource Strategies
0 102030405060
percent
Recruiting Skilled Employees
Training
Incentive Compensation Plans
Figure 8. Percentage of Firms Training versus Score Attached to the Importance of
Training
0
10
20
30
40
50
60
70
80
%
N o ne(0) Low (1) M edium
(3) H igh (5)
score
Figure 9. Measures of Innovativeness
01020304050
%
Intrad
Infreq
Ingen
Inimp2
Inprod
Inproc
Inrd
Incomp
Inimp1
Inimp
Analytical Studies Branch – Research Paper Series - 13 - Statistics Canada No. 11F0019MPE No. 123
5. Innovation and Entrants
The sheer size of the amount of entry and exit that is taking place is a testament to its importance.
Competition is intense when the identity of the players change. Despite this, it is important to
have supportive evidence of a concrete effect of entry on the performance of an industry.
Entrants are often seen as providing the dynamic new force in an industry that leads to change.
Productivity growth is one way in which this is manifested. Geroski (1991) relates technical
progress, efficiency and productivity growth at the industry level in the U.K. to entry rates and
finds that roughly 30% of productivity growth is related to entry. Entry is seen to have a
significant impact on productivity growth in the Canadian manufacturing sector when micro
plant level data are used (Baldwin, 1995, p. 234). Somewhere between 20% and 25% of
productivity growth in a manufacturing industry in the 1970s came from entry and exit.
Haltiwanger (1998) reports the U.S share as 18%, using a similar micro-economic database.
Changes in productivity are the result of either product or process innovation. While there are
few studies of the effect of new firms on productivity growth, much more attention has been paid
to the effect of small firms on innovation. Audretsch (1995), for example, uses data from the
SBA innovation database to argue that the US small-firm sector is more innovative than the
large-firm sector. Audretsch defines efficiency as the number of innovations in small firms
divided by the number of employees in small firms and reports that in 14 of 18 industries, this
measure was larger for small than large firms.
The empirical work on small-firm innovation capability has had to wrestle with a paucity of
usable data on innovations. Exercises to measure innovations using experts or trade journals have
provided data on innovative tendencies; however, in doing so, they have tended to reduce the
world of innovation to a single dimension.
Innovations are not easily pigeonholed in one compartment. Some innovations involve new
products; others consist of new processes. Some firms will focus on the use of advanced
technologies. Others will incorporate new ideas by embodying skills in their employees who will,
in turn, devise and design new products. Technology and skills will be combined in different
proportions across firms.
The variety of innovations that are being continuously introduced means that there is no single
measure of innovativeness that innovation surveys or trade-journal searches should be expected
to yield. It is not sufficient to argue that this could be resolved by specifying that it is only major
innovations that should be included in the definition of an innovation since it is only major
breakthroughs that should interest us. First, this still leaves the problem of defining ‘major’.
Second, this position ignores the important contribution that marginal innovations make.
Hollander (1965), for example, studied the course of productivity growth in Dupont’s rayon
division and found that much of its productivity growth arose from marginal innovations.
Freeman (1982) notes that substantial progress is made in the way of incremental process
innovation and these types of improvements tend not to be captured in major event studies that
focus on ‘innovations’. Incremental innovations are not measured by patent statistics since they
Analytical Studies Branch – Research Paper Series - 14 - Statistics Canada No. 11F0019MPE No. 123
are not patented, or by historical studies that focus on major breakthroughs.
If we are to understand the importance of new firms in the innovation process, it is essential to
move beyond the very simple, partial analysis based on a single definition of innovation and
develop a classification scheme based on a number of innovative competencies. Only by doing so
can we fully appreciate the complexity of an innovation process that is multidimensional rather
than unidimensional.
In order to study the innovative capability of successful entrants, we make use of questions on the
innovative capabilities of firms and on their technological capacities. Each of these will be used
to define the extent to which new firms followed an innovation or an advanced high-tech input
strategy.
There are two types of questions that are used to provide information on the competencies of
entrants. First, there are questions that characterize an entrant as following certain strategies—as
producing a product innovation, as having formal training, or the percentage of investment in
R&D or in training. These take on 0-1 values or are expressed as percentages. Second, there are
answers to questions about the emphasis that entrants placed on factors such as R&D capabilities
that contributed to the ongoing success of the firm. These questions were scored on a five-point
Likert scale of 1(low importance) to 5(high importance). These are used here to gauge the
entrant’s competencies.10 In what follows, entrants are deemed to possess a particular
competency or to be stressing a particular strategy if they score that strategy as 4 or 5 on the five-
point Likert scale.
5.1 Innovative Competencies
In order to capture the diversity of the innovative activity in entrants, the stress that new, small
firms place on innovation is measured with 10 different variables. Each captures a different,
though related concept of innovation.
Several of the measures make use of a question that asked whether an entrant had introduced an
innovation. An innovation was defined as the introduction of a new or improved product or
process, but not the introduction of aesthetic changes that did not affect the technical
construction or performance of the product. These innovations were, in turn, characterized as
being either entirely new products, modifications of existing products, entirely new processes, or
modifications of existing processes. Finally, entrants indicated whether these innovations were
protected with intellectual property rights. Combinations of these measures are then used to
define whether the entrants were introducing innovations of varying degrees of novelty and
importance.
10 Where possible the two sets of questions were cross-tabulated in order to validate the nature of the Likert scale.
For example, there is a high correlation between the importance given to R&D and the % of investment devoted to
R&D—or the score given to training and whether a firm had implemented a formal training program (See Baldwin
and Gellatly, 1999). This has also been found in previous surveys that ask firms to score their competencies and to
indicate the intensity of their activities (Baldwin et al., 1994).
Analytical Studies Branch – Research Paper Series - 15 - Statistics Canada No. 11F0019MPE No. 123
These variables are:
INGEN–whether an entrant reports any innovation.
INIMP1–whether an entrant reports an innovation and it is protected by an intellectual property
right such as a patent or the firm reports that a strategy of protecting its innovations with
intellectual property is important or very important (a score of 4 or 5). Baldwin (1997) finds that
the most novel innovations are those availing themselves of intellectual property protection.
INIMP2–whether an entrant reports an innovation that is either a completely new product or a
completely new process.
INPROD–whether an entrant is a product innovator.
INPROC–whether an entrant is a process innovator.
INCOMP–whether an entrant is both a product and a process innovator.
Innovation is also measured by the emphasis that is given to a prime (though not the only) input
to innovation—research and development. The variables are:
INRD–whether an entrant scored 4 or 5 on the importance given to R&D capabilities.
INIMP–whether an entrant’s percent of investment devoted to R&D was above the median of all
other entrants.
Finally, the innovative tendencies are measured by the scores that are given to the importance
that an entrant attributes to different competitive strategies that involve a broader concept of
innovation. The first of these variables complements the other innovation measures by focusing
on the frequency of new-product introduction. Since a new product (as opposed to an innovation)
can involve product differentiation changes, the coverage of this variable is broader than the
variable that captures whether a firm has had a product or process innovation. The first variable
used is:
INFREQ–whether an entrant scores 4 or 5 on the extent to which it frequently introduces new or
improved products.
The second broader variable captures the extent to which entrants focus on providing a different
type of innovation—customization and quality variation. The majority of small firms cannot
compete directly on prices with larger firms because of the cost disadvantages associated with
size. Instead, they stress a niche strategy by varying the quality of product, offering slightly better
levels of customer service, showing flexibility to customer needs or customizing their product for
individual customer requirements. The value of a product to consumers is a function of its
quality. All firms, but small firms in particular, are constantly experimenting with variations in
quality to attract customers. These experiments all involve innovations—though, in most cases,
Analytical Studies Branch – Research Paper Series - 16 - Statistics Canada No. 11F0019MPE No. 123
they are incremental in nature. While incremental, their importance should not be discounted.
The cumulative effect of the sum of many changes at the margin can be large. The variable used
to catch this aspect of innovation is:
INTRAD–whether an entrant scores at least 18 out of a possible 20 points on the importance
attributed to quality, customer service, flexibility in responding to customer needs and
customization.
5.2 Technological Competencies
The second advanced competency focuses specifically on whether technological innovation
(technovation) in an entrant is important. Technological innovation involves a different though
related dimension of innovation—the extent to which an entrant focuses on advanced technology,
increases its efficiency of input use, and introduces new production processes. The variables used
here are:
INTECH1–whether an entrant scores 4 or 5 on the importance attached to developing new and
refining existing technology.
INTECH2–whether an entrant scores 4 or 5 on the importance attached to purchasing technology
from others.
TEDEV–whether an entrant both develops/refines new technology and purchases it.
TECOMP–whether an entrant scores 4 or 5 on the importance given to using computer-
controlled processes in production.
TEINFO–whether an entrant scores 4 or 5 on using information technology for management
purposes.
TEINP–whether an entrant is in the top half of all entrants when it comes to the percent of
investment that is devoted to technology acquisition and licensing.
PROD1–whether an entrant scores 4 or 5 on improving efficiency of input use in the production
process or reducing production times.
5.3 Comprehensiveness of Innovation in the Entrant Population
The percentage of the population of entrants that are innovative differs considerably depending
upon which of the summary measures is used (Figure 9). Judged by the traditional measure
(INTRAD), entrants are quite innovative. Many firms are experimenting with the type of
innovation that requires bundling service or quality with a good. About 42% of the population
place a heavy emphasis on varying quality to provide a unique product to the consumer (INTRAD).
Somewhat fewer are introducing new products. Some 32% place more than average importance
Analytical Studies Branch – Research Paper Series - 17 - Statistics Canada No. 11F0019MPE No. 123
on frequently introducing new products (INFREQ). When ‘new’ is interpreted to mean an
‘innovation’, fewer firms fall into this category. Some 22% have introduced an innovation over
the 1992-94 period (INGEN). Some 14% emphasized an R&D strategy (INRD), but some 29%
either report an innovation or that they place above average importance on R&D—about the
same percentage that emphasize the frequent introduction of new products.
When the constraint of novelty is imposed on the innovation, the percentage declines by amounts
that vary depending on the definition of novelty that is imposed. Only 16% introduced what they
consider to be an entirely new product or process (INIMP2). Even fewer (8%) introduced an
innovation where intellectual property rights were perceived to be important (INIMP1). Despite
these differences, it should be noted that almost all firms are innovative by one or other of these
standards: roughly 70% of all firms fall into one of the categories defined here. Innovation is an
activity that is widely pursued.
The various measures of the firm’s technological capabilities also indicate a diversity of
technological competencies (Figure 10). In accord with the finding of Baldwin et al. (1997b) that
communications technologies have been expanding fastest, the largest group of firms (47%)
emphasize the importance of computer-based information technologies (TEINFO). The next
largest percentage of entrants (34%) focus on methods to reduce the cost of inputs and to reduce
production times (PROD1). About 25% place heavy emphasis on developing new technologies
(INTECH1) or use computer-controlled processes in production (TECOMP). There are about 20%
who stress the purchase of new technologies (INTECH2). About 16% of firms both develop new
technologies and purchase new technologies from others (TEDEV). Once again, a clear majority
of entrants are engaged in some form of technovation; here too approximately 70% fall into at
least one of the technovation categories.
These data show that a substantial proportion of new entrants consider themselves to be either
innovative or technologically advanced. Moreover, when we consider the characteristics jointly,
the percentage of entrants that fall into at least one category increases. For example, some 22%
report an innovation, some 14% place a very high importance on R&D and 29% report at least
one of these two characteristics. Similarly, 25% of entrants report that they develop new or refine
existing technology, while 20% bring in outside technology; but some 30% perform at least one
of these two. Some 47% use information technology in management and 25% use computers for
process control; however 53% of entrants exhibit at least one of these characteristics. When we
expand our definitions of innovation to encompass a characteristic from more than just the
innovative group, the percentage of entrants that can be said to be innovative is quite large. Some
39% report an innovation, or perform R&D, or emphasize either the development or purchase of
technology. Some 65% do one of the above or emphasize computer controlled processes or stress
the use of information technologies.
Analytical Studies Branch – Research Paper Series - 18 - Statistics Canada No. 11F0019MPE No. 123
Figure 10. The Importance of Technology Types
Figure 11. Innovation Scores Ranked by Industry
Figure 12. Technology Scores Ranked by Industry
0
0.1
0.2
0.3
0.4
0.5
0.6
1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49
Services Goods
Index
0
0.2
0.4
0.6
0.8
1 6 11 16 21 26 31 36 41 46
Services Goods
Index
0
5
10
15
20
25
30
35
40
45
50
Teinfo
Prod1
Tecomp
Intech1
Intech2
Teinp
Tedev
% of firms
Analytical Studies Branch – Research Paper Series - 19 - Statistics Canada No. 11F0019MPE No. 123
6. Innovation and Human Capital
The previous section examines the extent to which different types of innovation are prevalent in
new young firms. Innovation, however, is not produced in a vacuum. Skilled workers are key to
innovative firms. Research on small and medium-sized firms in Canada has demonstrated that a
stress on human capital is an important part of an innovation strategy (Baldwin and Johnson,
1996a). This section asks whether this relationship extends to new firms. It does so by asking
whether industries that tend to be more innovative or more technologically advanced than others
also tend to place greater stress on skilled workers and to do more training.
6.1 Industry Differences in Innovation, Technology and Firms’ Emphases on
Skilled Labour
In order to examine how industries differ across the dimensions that our survey measures, the
respondents’ scores on each of the measured characteristics were averaged for a set of 48
industries in order to produce two indices—one for innovation (INAV), one for technology
(TEAV).
Not all firms stress each of the aspects of innovation and technology that are measured here.
Some focus on new products. Others develop new processes. But most of the innovation and
technology variables are correlated at the industry level11. Therefore, they are combined into two
separate indices that will be used to rank industries by their degree of innovativeness or
technological competencies.
The innovation index is generated in the first instance by taking the average of the variables that
are generally regarded as being closely associated with innovation—the existence of an
innovation (INGEN, INPROC, INPROD), an important innovation (INIMP1, INIMP2), with the
importance of investment in R&D (INIMP), and the importance attributed to R&D (INRD). In
addition, the frequency with which new products are introduced (INFREQ) was included in the
innovation index.12 The technology index, in the first instance, consists of INTECH1, INTECH2,
TECOMP, TEDEV, TEINFO, and TEINP. The more extensive definition of a cost-cutter (PROD1) was
also included.
These indices measure the percentage of firms in an industry that may be said to be innovative or
technologically advanced using the definitions outlined above. Industries are ranked by the value
of the innovation index from lowest to highest and the values of the index are plotted in Figure
11. In the goods industry the value ranges from a low of 3% to a high of 60%. The same is done
for the technology index in Figure 12. Here the goods industries range from around 0 to about
50%.
11 For a discussion of these correlations, see Baldwin and Gellatly (1999).
12 Interindustry rankings are not sensitive to the inclusion or exclusion of this more expansive concept of innovation.
Analytical Studies Branch – Research Paper Series - 20 - Statistics Canada No. 11F0019MPE No. 123
Figure 13. Skill Intensity Ranked by Innovation Intensity
Figure 14. Skill Intensity Ranked by Technology Intensity
0
0.2
0.4
0.6
0.8
1
1 5 9 13172125293337414549
Services Goods
Skill Index
0
0.2
0.4
0.6
0.8
1
1 5 9 13 17 21 25 29 33 37 41 45 49 53
Services Goods
Skill Index
Analytical Studies Branch – Research Paper Series - 21 - Statistics Canada No. 11F0019MPE No. 123
Finally, an overall index was created that captured the emphasis that is placed on human capital
by new firms. The emphasis on human capital is measured here by the value that an entrant
attaches to recruiting skilled labour, on the emphasis it gives to training, and finally on the extent
to which it implements a formal training program and invests in training. The variables that are
employed are:
LABSKL–whether an entrant scores 4 or 5 on the importance given to recruiting skilled
employees.
LABSCOR–whether an entrant scores 4 or 5 on the importance attached to training.
LABFOR–whether an entrant does formal training.
LABTRAIN–whether an entrant’s share of investment devoted to training is positive.
LABINT–whether the percentage of investment that is devoted to training is above the median
for all entrants that have positive levels of investment in training.
The skill-related index (LABAV) consists of the average value of LABFOR, LABINT, LABSCOR,
LABSKL, and LABTRAIN. Each industry is ranked by the value of its innovation index (INAV) and
then LABAV plotted in Figure 13. The upward slope from left to right of the skill-related index
shows that entrants place greater emphasis on training and hiring skilled workers in industries
that are more innovative.
When industries are ranked by the technology index (TEAV) and the skill-related index is plotted
(Figure 14), the same results are found. New firms in industries that are more innovative or that
are more technologically advanced place more emphasis on human capital.
A more detailed examination of the relationship between the industry values of the underlying
variables is provided in Table 1, where the skill variables are correlated with the innovation and
technology variables. All of the variables have positive correlations. However, it is noteworthy
that the stress that is placed by entrants on obtaining skilled workers (LABSKL) always has a lower
correlation than the stress that is placed on training (LABSCOR). Innovation is sufficiently firm-
specific that entrants have to develop their own human capital skills via training programs. It is
also the case that the correlations with the two variables that capture whether a training program
is implemented (LABFOR and LABTRAIN) are higher than the variable capturing the importance of
labour skills. Training then is the key to both an innovation and a high-tech strategy.
Analytical Studies Branch – Research Paper Series - 22 - Statistics Canada No. 11F0019MPE No. 123
Previous research (Baldwin and Johnson, 1996a) has stressed that the connection between the
human resource strategy and the innovation strategy of a firm differs across industrial sectors. In
the goods sector, higher skills are associated with accompanying technical emphases. In the
services sector, this connection is somewhat more tenuous—because the human capital strategy
is the innovation strategy. To investigate the differences in the importance of training in the two
sectors, the correlations between the variables measuring the human-resource strategy and the
average innovation and technological competency indices are correlated (Table 2). In the goods
sector, all of the human capital variables are more closely related to a technology strategy than to
the innovation strategy. This is not the case in the service sector, where there is no clear
relationship. While entrants in services are more likely to be stressing the importance of skilled
labour (LABSKL) where they stress their technological capabilities, they are more likely to be
implementing a training program (LABFOR, LABTRAIN) where they stress that they are introducing
new products and processes.
Table 1. Correlation between Skill Characteristics and Innovation/Technology Variables
LABSKL LABFOR LABSCOR LABTRAIN
INGEN 0.27 0.54 0.41 0.52
INIMP1 0.28 0.42 0.43 0.37
INIMP2 0.26 0.46 0.3 0.44
INTRAD 0.34 0.48 0.34 0.48
INPROC 0.3 0.57 0.43 0.5
INCOMP 0.32 0.47 0.45 0.39
INRD 0.39 0.49 0.49 0.4
INIMP 0.24 0.34 0.32 0.19
INFREQ 0.34 0.56 0.45 0.4
INPROD 0.26 0.42 0.38 0.4
INTECH1 0.42 0.46 0.58 0.47
INTECH2 0.29 0.32 0.53 0.29
TEINP 0.18 0.2 0.32 0.3
TEDEV 0.3 0.31 0.48 0.31
TECOMP 0.37 0.19 0.52 0.06
TEINFO 0.48 0.26 0.54 0.2
PROD1 0.28 0.32 0.33 0.18
Table 2. Correlation between Skill Characteristics and Innovation/Technology Variables
LABSKL LABFOR LABSCOR LABTRAIN
Goods
TEAV .54 .52 .72 .58
INAV .33 .52 .55 .48
Services
TEAV .45 .34 .68 .21
INAV .36 .61 .44 .54
Analytical Studies Branch – Research Paper Series - 23 - Statistics Canada No. 11F0019MPE No. 123
7. Growth of Entrants and the Importance of Innovation
7.1 Growth Differences Across All Markets
We have already demonstrated that despite the emphasis that the economics literature has placed
on the connection between innovation and firm size, small firms and entrants in particular
demonstrate many different types of innovative behaviour. Nevertheless, small firms do face a
highly competitive environment as demonstrated above and there are those who argue that
competition does not encourage innovation. We investigate this here by asking whether
innovation is rewarded by faster growth. In order to investigate the differences between faster
and slower growing firms, we separate firms into two groups based on the annual average growth
in real revenue from their first full year to 1993.13
Faster growing entrants are more innovative in a number of associated ways. They are more
likely to introduce new or improved products and to seek out new markets, while at the same
time striving for efficiency gains through process innovation.
Accompanying these differences in investment activities associated with innovation and
innovation itself comes a considerable difference in the emphasis that is given to technological
competencies and R&D capabilities (Figure 16). The largest difference (over 35%) can be found
in the extent to which the firm introduces new technologies by purchasing them from others. But
differences in R&D capabilities are a close second though they are less statistically significant.
Differences between faster and slower growing firms are somewhat smaller for developing new
technologies and protecting products using intellectual property rights.14
The greater emphasis on innovative inputs, whether it be R&D or technology, is also
accompanied by a greater emphasis on other aspects of the production process that are associated
with improving production efficiencies. Faster growing entrants also rate each of the production-
related strategies more highly than slower growing entrants, with the largest differential being
associated with the use of computer-controlled processes (Figure 17).15
13 This allows us to be sure that the size for the starting point is a full-year measure rather than a part-year measure of
sales.
14 Only the difference “purchasing others technology” is statistically significant at the 5% level using a one-tailed t-
test. R&D capabilities is significant at the 10% level.
15 Differences for computer controlled processes and high quality suppliers are statistically significant at the 5% level
using a one-tailed t-test; the others are statistically significant at the 10% level using a one-tailed t-test.
Analytical Studies Branch – Research Paper Series - 24 - Statistics Canada No. 11F0019MPE No. 123
Figure 15. Differences in the Percentage of Entrants Investing and Innovating Between
Faster and Slower Growing Entrants
Figure 16. Differences in the Importance of Technological Strategies Between Faster and
Slower Growing Entrants
Figure 17. Differences in the Importance of Production Strategies Between Faster and
Slower Growing Entrants
0 20 40 60 80 100 120
Investing in R&D/
innovation
Innovating
Investing in technology
% difference
0 5 10 15 20 25 30 35 40
Purchasing others' technology
R&D capabilities
Developing new technology
Protecting products with intellectual property
% difference
0102030405060
Using computer controlled process
Reducing production times
Improving efficiency of input use
Using high quality suppliers
% difference
Analytical Studies Branch – Research Paper Series - 25 - Statistics Canada No. 11F0019MPE No. 123
The purchase of advanced technologies enhances the physical capital that is critical to
innovation. The development of R&D contributes to one type of intangible knowledge capital.
The other type of intangible knowledge capital is embedded in the human capital of the firm.
Since both of the former receive greater emphasis from faster growing entrants, it is not
surprising to find that faster growing entrants also give greater emphasis to each of the strategies
associated with creating and maintaining high levels of human capital (Figure 18).
Faster growing entrants give a greater emphasis to incentive compensation plans. Plans such as
these provide the incentive to be inventive where risk and rewards need to be shared if innovative
projects are to be brought to market. But faster growing entrants also place more emphasis on
recruiting skilled employees and on training.16
Faster growing entrants rate almost all of the product-specific strategies higher than slower
growing entrants. They are most distinguished from slower growing firms in the value they place
on product innovation-related strategies such as customizing their products, frequently
introducing new/improved products, and offering a wide range of products (Figure 19).
Innovative strategies in faster growing firms do not replace the attention that these firms give to
enhancing existing products. Growing entrants also place more value on flexibility in responding
to customer needs, customer service, and quality; moreover, the former two differences are
statistically significant. Growing entrants do not maintain unchanged product lines; rather they
focus on introducing new products that are of a higher quality and on improving the delivery of
the product.17
The innovative stance of growing entrants is evident in their marketing strategy as well. Growing
entrants place significantly greater emphasis on expanding their market reach (Figure 20). The
greatest differential occurs in the emphasis that is placed on using third-party distributors. The
next most important differentials exist in the emphasis that is placed on targeting new foreign
markets and improving their position in existing markets.18
16 All these differences are statistically significant at the 5% level using a one-tailed t-test.
17 Except for customer service and range of products, all the positive differences are statistically significant at the 5%
level for a one-tailed t-test. The former are significant at the 10% level
18 Promoting reputation, improving position in existing markets, and using third-party distributors are statistically
significant at the 2.5% level; the other differences are statistically significant at the 10% level using a one-tailed t-
test.
Analytical Studies Branch – Research Paper Series - 26 - Statistics Canada No. 11F0019MPE No. 123
Figure 18. Differences in the Importance of Human Resource Strategies Between Faster
and Slower Growing Entrants
Figure 19. Differences in the Importance of Competitive Strategies Between Faster and
Slower Growing Entrants
Figure 20. Differences in the Importance of Marketing Strategies Between Faster and
Slower Growing Entrants
0 5 10 15 20 25 30
Providing incentive compensation plans
Recruiting skilled employees
Training
% difference
-10-5 0 5 10152025303540
Customization of products
Frequently introducing new/
im
p
roved
p
roducts
Offering a wide range of products
Flexibi lity in r esponding to
customer needs
Customer
Q
ualit
y
Price
% difference
010203040
Using third-party distributers
Targeting new foreign markets
Improving position in existing markets
Promoting reputation
Targeting new domestic markets
Satisfying existing customers
% difference
Analytical Studies Branch – Research Paper Series - 27 - Statistics Canada No. 11F0019MPE No. 123
7.2 Growth Differences in New and Mature Markets
Examining differences between faster and slower growing firms for the entire population of
entrants establishes which competencies are associated with growth, irrespective of the specific
environmental factors affecting the entrants. Yet, there is reason to believe that the importance of
these factors may vary across industries. In particular, the stage of the industry is likely to affect
the type of innovative activity and the complementary strategies that are adopted by faster
growing entrants.
In new product markets that are in their early growth phase, the characteristics of the product are
continually changing. In these volatile markets, the entrants that grow should be those that keep
pace with or lead product changes. Entrants that grow in these markets are those that anticipate
and stimulate demand for new product features, and focus on product development. At this stage
of development, product change is so rapid that firms have little time to focus on process
innovation, and there is relatively less to gain by focusing on improving the efficiency with
which existing products are produced. The emphasis that firms give to improving the way
existing products are produced, or to extending their market reach, is likely to become more
important as markets mature.
In order to examine the extent to which there are differences in the innovation profile of entrants
and differences in the types of competencies that are associated with growth, entrants are
separated into two groups—those in markets that are in their introductory or growth stage (new
markets) and those in markets that are in their mature or post-mature stage (mature markets).
Differences in the competencies of faster and slower growers are then examined in order to
determine the extent to which the stage of the market changes the conclusions previously derived
as to the nature of competencies associated with growth.
When this is done, innovation is still found to be strongly associated with growth, regardless of
the market’s maturity. Innovation is more prevalent in faster growing firms than in slower
growing firms in both new and mature markets (Figure 21). In new markets, an emphasis on
product innovation serves to distinguish faster from slower growers; this is not the case for
process innovation. In mature markets, faster growers are more likely to be doing more of both
process and product innovation. Our hypotheses about the differences in the importance of
innovation across the product lifecycle are only partially confirmed. It is true that process
innovation matters more in mature markets than it does in new markets. But product innovation
is equally important in terms of growth in both markets.
In keeping with the importance of human capital to the growth process, growing firms are more
likely to train in both new and mature markets (Figure 21). We know that innovation brings
about greater skill requirements (Baldwin and Johnson, 1996a), and firms that innovate tend to
train more. Consequently, it is not surprising to find that faster growing firms, in each of the
groups, are more likely to train than are slower growing firms.
Analytical Studies Branch – Research Paper Series - 28 - Statistics Canada No. 11F0019MPE No. 123
Figure 21. Differences in Innovative Activities Between Faster and Slower Growing Firms
Figure 22. Differences in the Perceived Importance of Customer Strategies Between Faster
and Slower Growing Entrants
0
20
40
60
80
100
120
140
160
180
200
% difference
New Markets
Mature Markets
Process
Innovators Product
Innovators Training Investing
in R&D Investing
-0,3
-0,2
-0,1
0
0,1
0,2
0,3
0,4
0,5
% difference
New Markets
Mature Markets
Customization of Products
Introducing New/Improved
Products
Offering a Wide Range o
f
Related Products
Flexibility in Responding to
Customer Needs
Quality
Customer Service
Price
Analytical Studies Branch – Research Paper Series - 29 - Statistics Canada No. 11F0019MPE No. 123
However, it is noteworthy that the training differences between faster and slower growing firms
are greater in new than in mature markets. Mature markets are more likely to require skills that
can be obtained by hiring workers with existing skill sets. In new markets, new products are
being introduced so rapidly that new skills are constantly required. The very novelty of the
products often requires skills that are so new or firm-specific that companies have to train
workers to match the desired skill level. Therefore, training becomes more critical to the growth
process in markets where the product is in the earlier stages of the product lifecycle.
R&D also serves to distinguish faster from slower-growing firms in both stages of product
development (Figure 21). However, the difference is greater for mature markets. At first glance,
this is counterintuitive since R&D is associated with new product development, which in turn is
linked with new markets. But, as we have seen, product development is equally important in both
market stages, while process innovation is more important in mature markets. If R&D is a critical
component required for both types of innovation, then R&D differences should be greater for
mature markets—as they are.
While the connection between growth and the intensity of product innovation does not serve to
distinguish new from mature markets, other aspects of the customer product strategy do so
(Figure 22). There are positive differences between faster and slower growing firms with regards
to both customization of new products and the frequency of introducing new/improved products
for both mature and new markets. This is what one would expect since the emphasis on product
innovation is greater for faster growing than for slower growing firms in both markets. However,
the difference between faster and slower growing firms is greater in new markets where products
are changing more rapidly. By way of contrast, differences in mature markets are greater for the
strategy of responding to customer needs in a flexible manner. There are smaller differences
between faster and slower growing firms for the two conventional aspects of competition—
quality and customer service—but the differences are nonetheless positive. The difference is
statistically significant in new markets for a quality strategy; in mature markets, it is significant
for a customer-service strategy.
Associated with a greater stress by growing firms on product innovation and other new product
strategies comes a greater emphasis on a number of different marketing strategies (Figure 23).
Faster growing firms in both markets are significantly more likely to stress product reputation.
They are both more likely to stress market expansion—but the source of that expansion differs
because the markets are not at the same point in the lifecycle. Faster growing firms in mature
markets are more likely to target new foreign markets and new domestic markets than are faster
growing firms in new markets (Figure 23). Despite the fact that mature markets offer less growth
on average, faster growing firms in the mature stage of a product lifecycle expand by targeting
both new domestic and foreign markets.
Analytical Studies Branch – Research Paper Series - 30 - Statistics Canada No. 11F0019MPE No. 123
Figure 23. Differences in the Perceived Importance of Marketing Strategies Between
Faster and Slower Growing Entrants
Figure 24. Differences in the Perceived Importance of Production Strategies Between
Faster and Slower Growing Entrants
-5
5
15
25
35
45
55
Using third-
party
distributors
Targeti ng new
foreign
markets
Improving
position in
existing
markets
Targeting new
domestic
markets
Promoting
company or
product
reputations
Satisfying
existing
customers
% difference
New Mar kets
Mat ur e Ma rk et s
-0.05
0.15
0.35
0.55
0.75
0.95
1.15
Using
computer
controlled
processes
Reducing
production
times
Using high-
quality
suppliers
Improving the
efficiency of
input use
% difference
New Markets
Mature Markets
Analytical Studies Branch – Research Paper Series - 31 - Statistics Canada No. 11F0019MPE No. 123
All firms in the initial stage of the product lifecycle are producing new products for new markets.
Therefore, it is not surprising that the stress on new markets does not distinguish between faster
and slower growers here. However, it is true that the percentage difference in the stress that is
placed on improving existing market position, which for firms in new markets means expanding
market share in their new product markets, is greater between faster and slower growers in new
markets than in mature markets. Therefore, growing firms in this phase of the product lifecycle
do place a greater emphasis on the aspect of market expansion that was relevant to them.
More importantly, what serves to distinguish faster growers from slower growers in new markets
is the extent to which third-party distributors are used. The demands being made on new firms in
the early stages of the lifecycle are sufficiently large that relying on outside parties for
distribution, where they are available, husbands scarce resources.
This section demonstrates the universal importance of product innovation in markets that differ
in terms of their position in their lifecycle. While differences in emphasis exist, they involve
detail only—with more new products and more customization in the early stages of the lifecycle.
The degree of process innovation shows more market differences. Mature markets are those
where growth is more closely related to process innovation.
This finding should not be interpreted to mean that an emphasis on production strategies is
unimportant in new markets. While faster growing firms in new markets place relatively less
stress on process innovations (Figure 24), they are more likely to stress new computer-based
production strategies. On the other hand, faster growers in mature markets are more likely to
focus on reducing production times or using high-quality suppliers.
In summary, similarities in the importance of product innovation can be found across markets
that are both young and old. The generalizations that originated in comparing faster and slower
growing firms in the entire population are qualified but are not overturned when the analysis is
extended to markets that differ by degree of maturity. Adding detail by examining different
markets provides insights that bolster the original conclusion that innovation is critical for
growth. In particular, product innovation is important in both young and old product markets
alike.
Successful entrants are those that reach beyond the bounds of their established product markets.
They introduce new and/or improved products and seek out new customers. They look inside the
firm. They give greater emphasis to training. They continually strive to improve, update and
modify their operations—but in different ways. Despite the primary focus of entrants on
established markets that has been drawn in a previous section, innovation in new firms in most
markets is rewarded with growth. These results corroborate those found in the GSME survey
(Baldwin et al., 1994) where more successful19 firms outperform less-successful firms in almost
every area, but where innovation is the key factor that discriminates between more- and less-
successful firms.
19 In the 1994 study, success is defined by an index consisting of growth in market share, productivity and
profitability.
Analytical Studies Branch – Research Paper Series - 32 - Statistics Canada No. 11F0019MPE No. 123
8. The Joint Effect of Innovation, Technological and Skill
Competencies on Growth
Innovation has different dimensions. Previous sections have examined the extent to which each
of these dimensions is pursued by new firms and how some of these dimensions, such as training
and technology use, are related. However, the connection between an entrant’s emphasis on skills
and innovation was conducted primarily at a high level of aggregation. The various dimensions
of innovativeness or technological competence were combined into one index and compared to
the dimensions of a skill-based human resource strategy, either separately or together.
In order to pursue the distinctions that might exist across types of innovators in terms of their
emphasis on training, the emphasis that is given to skill development in four different groups is
investigated here. Entrants are divided into those that produced just a product innovation, just a
process innovation, both product and process innovations, and those that did not report an
innovation.
There are two reasons for choosing this classification scheme for analyzing differences in human-
resource strategies. First, a cluster analysis of all the innovation and technology characteristics in
entrants that were described earlier produced these four distinct clusters. Secondly, Baldwin and
Johnson (1998) used an associated survey on small and medium-sized firms and found that these
four types differ substantially in terms of their success and strategic emphasizes. Confirming that
these differences can also be found in the entrant population corroborates the differences that
were previously observed.
In order to assess the success of each category, the survey respondents were divided into two
groups—the faster and the slower growers.20 Then the proportion of each innovator group that
fell in the highest growth category was calculated. The combined product/process innovators are
most likely to be in the highest growth category. Some 70% of this group fall in the faster
growing group (Figure 25). Product innovators are second (59%) and process innovators are third
(49%). Less than 43% of noninnovators are found in the top half of growers.
These differences accord with previous research that also found the highest growth rates
belonged to comprehensive innovators and the lowest to non-innovators (Baldwin and Johnson,
1998). This suggests that comprehensive innovation is the key to success, or that this type of
innovator is located in the fastest stage of the lifecycle of a product market. That product
innovators grow slightly more quickly than process innovators could also be connected to
differences in the position of each innovator type in the product lifecycle. If process innovators
are more likely to be in the later mature phase of the product cycle, their lower growth rates
simple reflect the more mature stages of the market. The results of the previous section that show
a larger percentage of firms in mature markets to be process innovators suggest that this
explanation accounts for part of the difference. The innovative group with the highest growth
rates also places the greatest stress on training.
20 The sorting was done separately for each birth year in the sample.
Analytical Studies Branch – Research Paper Series - 33 - Statistics Canada No. 11F0019MPE No. 123
The average score on each of the human resource strategy variables is highest for the combined
product/process innovators and generally lowest for the noninnovators (Figure 25). This too
accords with the previous findings (Baldwin and Johnson, 1998). Comprehensive innovators tend
to emphasize a number of strategic areas more intensely than do noninnovators (Baldwin and
Johnson, 1996b) and human resource strategies is one of these areas. For three of the four
variables, product innovators place greater stress on the importance of the strategy than do
process innovators. In the previous study, differences between these two groups were not found
to be large. Our results here suggest that product innovators may indeed need to worry more
about training because the skills required of the workforce at the early growth stage have to be
developed internally rather than acquired externally.
In the previous two sections, growth was shown to be associated with a number of separate
strategies. Faster growing entrants were more likely to be innovative and more likely to pursue a
human resource strategy that focused on recruiting skilled employees, training them and retaining
them. The connection between these different strategies was not investigated at great length. Was
it innovation itself that led to growth or skill emphasis? Or was the difference in skill emphasis
and its relation to growth simply caused by the fact that innovators generally are in the faster
growth class and innovation requires a special type of human-resource strategy?
We investigate this issue in two ways. In the first instance, we divide each of our innovative
types into those that emphasize a particular human-resource strategy—for example, into those
that have or do not have a formal training program (LABFOR). This is also done for LABTRAIN,
LABSKL, and LABSCORE. Then we calculate the percentage of those firms emphasizing the skill
strategy that are fast growers and those that are slow growers within an innovation class. These
percentages are plotted for each innovator type in Figure 26 for the two variables that capture
whether a training program is in place (LABFOR, LABTRAIN). In each case, there are a higher
percentage of fast growers in the category where training is being done. These differences do not
exist for the two variables that just rank the importance of training (LABSCORE) or the importance
of recruiting skilled labour (LABSKL) except for the noninnovator category. In conclusion,
irrespective of the category, if training plans are implemented, growth is superior. Training is an
important complementary strategy for innovators but training matters, irrespective of the
innovation environment.
The second approach to test the separate effect of training on success involved the use of
multivariate analysis. The rate of growth of the successful entrants was regressed on the three
indices—INAV, TEAV, and LABAV. This regression was run separately by three groups that were
used for stratification purposes in the original sample. This was done first for the entire sample.
Second, it was done for goods versus service industries. Third, it was done for high knowledge
and low knowledge industries. High and low knowledge industries were defined using
R&D/sales ratios, average wage and multifactor productivity growth.21
21 See Baldwin (1998) for a discussion of the creation of this industry taxonomy.
Analytical Studies Branch – Research Paper Series - 34 - Statistics Canada No. 11F0019MPE No. 123
Figure 25. Differences in Growth and Skill Emphasis Across Innovator Types
Figure 26. Differences in the Percentage of Firms that are Fast-Growers by Skill Group
0%
10%
20%
30%
40%
50%
60%
70%
80%
PROCESS
PRODUCT
COMBINED
NOT
INNOV
Labtrain=0
Labtrain=1
Labfor=0
Labfor=1
% of firms that are fast growers
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Growth Rate
Labscore
Labtrain
Labfor
Labskl
PROCESS
PRODUCT
COMBINED
NOT INNOV
Average Scor
e
Analytical Studies Branch – Research Paper Series - 35 - Statistics Canada No. 11F0019MPE No. 123
For the entire sample, the results show that innovation and labour skills are each significantly
related to the growth of entrants. The skills variable has about the same coefficient as the
innovation variable. When the sample is broken down into goods and services industries, the
innovation variable has a larger and more significant coefficient in goods than services.
Moreover, the innovation variable is significant in the former but not the latter. The reverse is
true of the skills variable. This confirms earlier results (Baldwin and Johnson, 1996a) that stress
the human-resource emphasis in services is a key factor behind growth. When the sample is
broken into ‘high’ and ‘low’ knowledge industries, innovation is seen to be more important in the
former, while labour skills are equally important in both. In conclusion, the effect of an emphasis
on skills is generally felt independently of the effect of innovation on growth, and this is
pervasive across the subsectors examined here.
Table 3. Coefficients from a regression of growth on innovation, labour and technology indices
Innovation
(INAV)Skills
(LABAV)Technology
(TEAV)
All .072 .077 -.011 F(3,2794)=9.0
(.031) (.030) (.029)
[.021] [.009] [.707]
Goods .156 .066 -.009 F(3,2794)=10.7
(.065) (.069) (.053)
[.016] [.344]4 [.858]
Services .043 .087 -.015 F(3,2794)=6.3
(.032) (.031) (.035)
[.176] [.005] [.664]
High Knowledge .084 .070 .041 F(3,2794)=8.0
(.032) (.035) (.45)
[.009] [.045] [.369]
Low Knowledge .068 .077 -.045 F(3,2794)=3.7
(.045) (.043) (.038)
[.134] [.082] [.247]
Note: Bracket figures are probability values for a two-tailed t test for the null hypothesis that the
coefficient is zero.
Analytical Studies Branch – Research Paper Series - 36 - Statistics Canada No. 11F0019MPE No. 123
9. Conclusion
Entry is important. At any point in time, a substantial amount of market share or employment is
accounted for firms that entered in the recent past.
Entry is an inherent part of the dynamic competitive process that leads some firms to grow and
others to decline. And it is within this context that it needs to be appreciated. It is not the only
process at work—but it is an important part of it.
The entry process involves trial and error. Firms have to develop basic skills before they can
survive—and large numbers appear not to have these skills at birth. Firms that fall by the
wayside generally start off smaller, pay lower wages and have lower labour productivity.
Moreover, their management lacks knowledge of basic management skills in many cases
(Baldwin et al., 1997). Nevertheless, there is a subset of new entrants that survive and grow—and
the growth of this group is substantial.
It is in the area of innovation that entrants make a widespread contribution. The contribution of
entrants to some industries is well understood. New technology-based firms play an important
role in the early stages of the lifecycle of many industries. They are the means by which the ideas
of new entrepreneurs are initially commercialized. Whether it be in electronics, instruments,
medical equipment, steel, or biotechnology, new firms have played an important role in the
innovation process.
While the profile of entrants in industries such as electronics and biotechnology is highly visible,
the role of small firms is important in many other industries. Innovation is widespread and is
generally associated with growth. Firms in many industries have developed the capacities that are
needed for innovation in their particular industries. These capabilities often involve development
of new technology or devising new products or processes that are highly novel. The range and
diversity of the innovation skills of new firms across all industries is remarkable.
The importance of the innovation process is attested to by the connection between growth and
innovation. The successful entrants that grow the most are those that develop one or other type of
innovative activity—either with respect to the introduction of new products, an emphasis on
technology, or human resources.
This innovation process, to be successful, requires complementary skills. In particular, firms that
develop new products and processes have to focus more intensely on human resource
capabilities. Key among these capabilities is the training activities of entrants. New firms develop
competencies in a number of different areas. An innovative firm may focus more intensely on a
broad range of competencies, but training is a key complementary capability. More importantly,
while training complements the innovation strategy, it has an effect on growth that is quite
separate from the innovation/technology strategy that a firm pursues. Entrants that train are more
likely to grow—irrespective of the emphasis that a firm places elsewhere on innovation and
technological capabilities.
Analytical Studies Branch – Research Paper Series - 37 - Statistics Canada No. 11F0019MPE No. 123
References
Abernathy, W.J. and J.M. Utterbach. 1978. “Patterns of Industrial Innovation,” Technology
Review June/July 80: 41-7.
Audretsch, D.B. 1995. Innovation and Industry Evolution. The MIT Press.
Baldwin, J.R. 1995. The Dynamics of Industrial Competition. Cambridge University Press.
Baldwin, J.R. 1996. “Productivity Growth, Plant Turnover and Restructuring in the Canadian
Manufacturing Sector,” in D. Mayes (ed.) Sources of Productivity Growth. Cambridge:
Cambridge University Press.
Baldwin, J.R. 1997. Innovation and Intellectual Property. Catalogue No. 88-515-XPE. Ottawa:
Statistics Canada.
Baldwin, J.R., L. Bian, R. Dupuy and G. Gellatly. 2000. Failure Rates for New Canadian Firms:
New Perspectives on Entry and Exit. Catalogue No. 61-526-XPE. Ottawa: Statistics Canada.
Baldwin, J.R., W. Chandler, C. Le and T. Papailiadis. 1994. Strategies for Success: A Profile of
Small and Medium-Sized Enterprises in Canada. Catalogue No. 61-523R. Ottawa: Statistics
Canada.
Baldwin, J.R, T. Dunne and J. Haltiwanger. 1995. “Plant Turnover in Canada and the United
States,” in J. Baldwin, The Dynamics of Industrial Competition. Cambridge University Press.
Baldwin, J.R. and G. Gellatly. 1999. “Developing High-Tech Classification Schemes: A
Competency-Based Approach in New Technology-Based Firms in the 1990s.” Edited by R.
Oakey, W. During and W. Mukhtar. Pergammon Press.
Baldwin, J.R., T. Gray and J. Johnson. 1996. “Advanced Technology Use and Training in
Canadian Manufacturing,” Canadian Business Economics 5(1): 51-70.
Baldwin, J.R., T. Gray, J. Johnson, J. Proctor, M. Rafiquzzaman and D. Sabourin. 1997. Failing
Concerns: Business Bankruptcy in Canada. Catalogue No. 61-525-XPE. Ottawa: Statistics
Canada.
Baldwin, J.R. and J. Johnson. 1996a. “Human Capital Development and Innovation: A Sectoral
Analysis,” in Peter Howitt (ed.) The Implications of Knowledge-Based Growth for Micro-
Economic Policies. Calgary: Calgary University Press.
Baldwin, J.R. and J. Johnson. 1996b. Business Strategies in More- and Less-innovative Firms in
Canada,” Research Policy 25: 785-804.
Analytical Studies Branch – Research Paper Series - 38 - Statistics Canada No. 11F0019MPE No. 123
Baldwin, J.R. and J. Johnson. 1998. “Innovator Typologies, Related Competencies, and
Performance,” In Microfoundations of Economic Growth: A Schumpeterian Perspective, edited
by C. Green and C. McCann, Ann Arbour: University of Michigan Press.
D’Amboise, G. 1991. The Canadian Small and Medium-Sized Enterprise: Situations and
Challenges. Halifax: The Institute for Research on Public Policy.
Dunne, T., M. Roberts and L. Samuelson. 1988. “Patterns of Firm Entry and Exit in U.S.
Manufacturing Industries,” Rand Journal of Economics 19(4) 495-515.
Freeman, C. 1982. The Economics of Industrial Innovation. Cambridge, Mass: MIT Press.
Geroski, P.A. 1991. Market Dynamics and Entry. Basil Blackwell.
Gort, M. and S. Klepper. 1982. “Time Paths in the Diffusion of Product Innovations,” Economic
Journal 92:630-53.
Haltiwanger, J. 1998. “Measuring and Analyzing Aggregate Fluctuations: The Importance of
Building from Micro-Economic Evidence,” St. Louis Fed. Reserve Bank Econ. Rev.
Hollander, S. 1965. The Sources of Increased Efficiency: A Study of Dupont Rayon Plants. MIT
Press.
Picot, G., J.R. Baldwin and R. Dupuy. 1994. Have Small Firms Created a Disproportionate Share
of New Jobs in Canada? A Reassessment of the Facts. Research Paper No. 71. Analytical Studies
Branch. Ottawa: Statistics Canada.
.