Top Banner
Business Strategy, Human Resource Management, Labour Market Flexibility, and Competitive Advantage By Jonathan Michie, Birkbeck, University of London, and Maura Sheehan, Graduate School of Management, University of Dallas * Address for correspondence Professor Jonathan Michie School of Management and Organizational Psychology Birkbeck, Malet Street, London WC1E 7HX [email protected] Tel: +44 (0)20 7631 6761 (direct + voice mail), +44 (0)20 7631 6836/6767/6772 (office) Mobile: +44 (0)7889 274951 (mobile) Fax: + 44 (0)20 7631 6769 (work), +44 (0)1630 685154 (home) Keywords: strategic human resource management, competitiveness, flexibility. Abstract We examine HRM and corporate performance in UK companies. Using original data collected from manufacturing and service sector companies, we find that the relationship between HRM and performance is dependent upon business strategy; and second, that companies pursuing an HRM approach coupled with a quality-enhancer or innovation focus within their business strategy perform best. We also find that HRM is more likely to 1
55
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Michie

Business Strategy, Human Resource Management,

Labour Market Flexibility, and Competitive Advantage

By Jonathan Michie, Birkbeck, University of London, and

Maura Sheehan, Graduate School of Management, University of Dallas*

Address for correspondenceProfessor Jonathan MichieSchool of Management and Organizational Psychology

Birkbeck, Malet Street, London WC1E [email protected]: +44 (0)20 7631 6761 (direct + voice mail), +44 (0)20 7631 6836/6767/6772 (office)Mobile: +44 (0)7889 274951 (mobile)Fax: + 44 (0)20 7631 6769 (work), +44 (0)1630 685154 (home)

Keywords: strategic human resource management, competitiveness, flexibility.

Abstract

We examine HRM and corporate performance in UK companies. Using original data collected from manufacturing and service sector companies, we find that the relationship between HRM and performance is dependent upon business strategy; and second, that companies pursuing an HRM approach coupled with a quality-enhancer or innovation focus within their business strategy perform best. We also find that HRM is more likely to contribute to competitive success when introduced as an integrated package of practices.

* This work was funded by The Leverhulme Trust (grant F/112/AL) and the University of London Central Research Fund and the University of Dallas (Provost’s summer stipend award) to whom we are grateful . We are indebted to Dr Sharon Milner and Elaine McDonald for research assistance. We have benefited from discussing these issues with a number of colleagues, including Professor David Guest, Dr Neil Conway, Dr Linda Trenberth and Professor Paul Teague. A number of colleagues were generous in supplying copies of their own questionnaires and details of their results, in particular Dr Peter Berg and Dr Eileen Appelbaum, Dr Sandra Black and Professor Lisa Lynch, Professor Ichniowski, Professor Steve Nickell, and Professor Paul Osterman. Patrick Burns, Director of Advocacy for The Work Foundation provided invaluable advice on the use of HRM practices in UK firms.

INTRODUCTION

1

Page 2: Michie

A consensus has emerged among scholars and practitioners alike that there is a positive

relationship between the use of, on the one hand, ‘high commitment [Human Resource]

management’ ‘(HCM)’ (Wood, 1999; Wood and Albanese, 1995; Wood and de Menezes, 1998)

or ‘high-commitment practices’ (Huselid, 1995; Huselid and Becker, 1996) and, on the other

hand, corporate performance and organisational outcomes (see Appelbaum, et al., 2000 and

Osterman, 1999 for discussions of high-performance work organizations). Considerable debate

remains, however, over the nature and causes of these outcomes. It is generally agreed that a

better understanding of the interaction between business strategy and HRM will be key to

explaining these empirical outcomes. In particular, it is argued that HR practices that are

consistent with or support organizational strategy are more effective than those that are not

(Miles and Snow, 1984; Schuler and Jackson, 1987; Truss and Gratton, 1994). The literature,

however, on the link between strategy and HR practices, while widely developed theoretically,

remains underdeveloped empirically.

Labour market deregulation and flexibility are also regarded as key determinants of national

competitiveness and successful corporate performance. A growing body of research has

examined the relationships between firms’ use of flexible work practices and human resource

systems, on the one hand, and corporate performance on the other hand (Michie and Sheehan-

Quinn, 2001; Arulampalam and Booth, 1998; Kleinknecht, 1998). This area, however, also

remains underdeveloped both theoretically and empirically.

This paper contributes to these gaps in the human resource literature by examining (1) the

link between strategy and HR practices; (2) the link between strategy and labour market

2

Page 3: Michie

flexibility and (3) the moderating effect of strategy on the link between HR practices, the use

of flexible labour and corporate performance.

The structure of the paper is as follows. The following section briefly surveys the literature;

we then outline the hypotheses to be tested; we describe our data; we report the method of

analysis and variables used; and finally we present our results and conclude.

PREVIOUS RESEARCH

Links between Strategy and HR Management

Models of situational contingency or ‘micro’ models of HRM and strategy suggest that the

appropriateness, or effectiveness, of HRM will vary depending on organizational lifecycle or

the product market within which the organization is operating (Kochan and Barocci, 1985;

Miles and Snow, 1984; Schuler and Jackson, 1987; Tichey et al., 1982). Schuler and Jackson

(1987) and Schuler (1989) provide a detailed treatment of three competitive strategies

(innovation, quality-enhancement and cost-reduction) and the associated behaviours of

employees and HR practices that are considered to be instrumental in the implementation of

the strategies. The authors show that HRM will prove effective only where the firm

emphasizes the importance of either quality enhancement or innovation within its business

strategy. In organizations perusing a cost based strategy, the logical approach to HR strategy

would be to emphasize numerical flexibility and wage cost minimization. And as noted by

Hoque, “In such a situation, the values and goals imbued within HRM would be inconsistent

with the organization’s primary cost reduction goals” (1999: 421). Consistent with other

authors (Becker and Gerhart, 1996; Khatri, 2000; MacDuffie, 1995; Hoque, 1999; Huselid,

1995) we use the term external fit to refer to the ‘organizational logic’ argument that HR

3

Page 4: Michie

strategy should be combined with business strategy, such that there should be a consistency

between the values and aims within each.

Empirically, there has been little evidence that the impact of HRM is contingent upon

the approach taken to business strategy (Huselid, 1995; Becker and Gerhart, 1996). Two recent

studies, however, find support for the contingency hypothesis. In his study of the UK hotel

industry Hoque finds a relationship between HRM and performance only among hotels

emphasizing quality enhancement and among hotels in the ‘other category’ and that HRM is

ineffective where strategy is characterized by cost control.1 He concludes that the analysis, “…

provides support for the contingency hypothesis that the effectiveness of HRM depends upon its

fit with strategy” (1999: 437). Khatri finds evidence of a strong link between the impact of

overall strategy on HR practices and of the moderating influence of overall strategy on the

relationship between HR practices and performance.2 The author emphasizes that, “The role of

strategy as a contingency factor is strengthened even more by the evidence that HR practices

unmatched with the appropriate organizational strategy could have a negative influence on

performance… In sum, we need to consider the overall organizational strategy before we

implement HR practices” (2000: 358).

Universal Relevancy of HRM

In contrast to the contingency approach outlined above which suggests that the effectiveness

of HRM is dependent upon its coupling with a quality-enhancer or innovator approach to

strategy, the universal relevance of HRM suggests that the coupling of the ‘HRM quality-

enhancer’ and the ‘HRM innovator’ approaches should have the most positive correlations

with performance.3 As suggested by Guest (1987), Hoque (1999) and Walton (1985), the

4

Page 5: Michie

universal relevance of HRM will depend to a large part upon the nature of the industry

product market. In very diverse product markets, where for example some firms will need to

emphasize product quality or customer satisfaction and other firms may need to focus on cost

control or price competition, there is no a priori reason why HRM should always prove

effective. Focussing his analysis on one industry – UK hotels – Hoque finds strong evidence

that HRM is universally relevant; among the hotels with an identifiable strategy, those

adopting an approach of service quality coupled with a high number of HRM practices

performed best.4

Internal Fit

Internal fit refers to the synergistic benefits resulting from the introduction of HRM as an

institutionally supported package of practices that are internally consistent and mutually

reinforce one another. To date there are varying degrees of empirical support for a

relationship between internal fit and performance (see e.g., Guest and Hoque, 1994; Huselid,

1995; Ichniowski et al., 1994; MacDuffie, 1995; Hoque, 1999; and Khatri, 2000).

Links between Labour Market Flexibility and Corporate Performance

Labour market deregulation has been regarded as playing a key role in the drive for a

competitive, flexible economy. The use of flexible work practices5 can result in savings on

wage costs (see Michie and Sheehan-Quinn, 2001 for further discussion). In addition, as

argued by Kleinknecht et al. ‘the decision to hire new workers is taken more easily if workers

can be fired more flexibly under adverse circumstances. In this way, part of the

entrepreneurial risk is shifted to employees, making job creation easier’ (1997: 2).

5

Page 6: Michie

On the other hand it has been suggested that the sort of labour market deregulation

pursued in Britain over the 1980s and 1990s may risk being detrimental to long-run

economic performance by leading to a neglect or undervaluing of assets and processes such

as training and innovative activity, which are vital to long-term development and economic

progress (Michie and Wilkinson, 1995; Kitson and Michie, 1996). Research using the British

Household Panel Survey 1991-95 that investigated the link between skills acquisition and

labour market flexibility (proxied by employment status, contract type, and lack of union

coverage) found that workers:

...on short-term employment contracts, who are working part-time, or are not covered

by a union collective agreement, are significantly less likely to be involved in any

work-related training to improve or increase their skills. These findings suggest that

there is a trade-off between expanding the more marginal forms of employment, and

expanding the proportion of the workforce getting work-related training.

(Arulampalam and Booth, 1998, p. 521)

In addition, if the time horizons of firms become shortened, the pursuit of what

economists would characterise as 'efficiency gains' may come to dominate other sorts of

gains to be had from innovation and technological progress. This becomes problematic if the

pursuit of short-term efficiency gains reduces the potential of the system for economic

progress (Kleinknecht, 1998; Michie and Prendergast, 1998). The analysis presented below

is unique in that it not only examines the links between the use of flexible labour and

corporate performance but also examines its link to the strategy pursued by the firm.

6

Page 7: Michie

THE DATA

We surveyed a stratified sample of publicly quoted UK manufacturing and service sector firms

with more than 50 employees. Two dimensions were used to stratify the sample: organisation

size and the primary sector of business activity. Five categories of organisation size were used –

those employing 50-99, 100-499, 500–999, 1000–9999, and 10,000 or more employees.

Thirteen sectors were identified (9 in manufacturing and 4 in services), using the 1992 UK

Standard Industrial Classification (SIC) codes. The EXTEL database was used to identify all

organisations with the above characteristics, which totalled 1280.6 Holding companies were

removed which reduced the ‘population’ to around 1000. The aim was to obtain a total quota of

33% of the ‘population’ and to achieve representation amongst the stratified sample cells.

A market research company was commissioned to conduct a telephone survey of the selected

companies. The following protocol was followed:

The name and fax number of the senior manager with HRM responsibilities was confirmed

by telephone.

This person was faxed a one page briefing on the objectives of the survey and the expected

length of the interview (around 30 minutes). Confidentiality of survey participants was

emphasised.

The day after faxing, the person was contacted to see if they would agree to the telephone

interview and if agreeable, a date and time for the interview was set. These individuals were

then faxed a ‘glossary’ of HR terminology, with alternative or variant names of many of the

practices, with examples relevant to both the manufacturing and service sectors.

7

Page 8: Michie

Interviews were conducted by telephone with the Director of Human

Resources/Personnel/Employee Relations of each company in 78% of cases. Where this person

was not available, an alternative senior person was interviewed.7 Most of the organisations in

the sample were large and had more than one establishment/site, with activities often varying

significantly across these sites. The interviewee was therefore asked to select the organisation’s

most typical establishment/site – that is, the site that most typified the organisation’s activities.

In total 934 individuals were asked to complete the survey. Of these, 559 declined, 19 agreed

but subsequently failed to complete the interview, and 369 interviews were completed

successfully. As a result of missing data, 365 of the total number of responses were usable - a

response rate of 39%. Sixty-six percent (242) of the sample is manufacturing establishments

while 34% (125) is service sector establishments.

METHODS OF ANALYSIS AND VARIABLES ANALYSED

Measurement of Performance

In contrast to many other studies, we use objective, rather than subjective measures of

performance. In addition, the data used in this analysis were collected at the establishment,

rather than at the company level. The performance data are averaged over a three year

period, which helps to reduce problems of simultaneity of one period explanatory variable

measures. We use three measures of performance: percentage change in total sales (sales

growth), percentage change in labour productivity, and percentage change in pre-tax

profitability. Logged values of each of the variables are used in the estimations.

8

Page 9: Michie

To examine the representativeness of the establishment level data, confirmatory

performance measures at the organizational level were obtained from the EXTEL database

for the sample companies. For percentage change in total sales, the correlation between the

two sources was 0.77 (p < .05); for percentage change in labour productivity, the correlation

was 0.84 (p < .01); and for percentage change in pre-tax profitability, the correlation was

0.68 (p < .05). A duplication of the analysis for a subset of 76 single site organizations (48 in

manufacturing and twenty eight in services) yielded even higher correlation coefficients

(0.94 (p < .001); 0.96 (p < .001) and 0.89 (p < .01), respectively). These results indicate that

the data obtained at the establishment levels closely reflects performance at the

organizational level, which is important, given our analysis of strategy which, of course, is

normally formulated at the organizational, or corporate headquarters level.

Control Variables

(a) Business strategy. We build on Schuler and Jackson’s (1987) business strategy

typologies. The classifications are based on both subjective responses to questions on strategy

and objective data. Companies were classified as having a cost based strategy if they

reported that their strategy focussed on cost minimization/reduction or price

competition/reduction and if they had a lower than average remuneration rate or if the

proportion of their wage bill paid to permanent, full-time employees had declined over the

past five years (16.8% of establishments). Firms were classified as having a quality based

strategy if they reported that their strategy focussed on quality enhancement and if they had

achieved a recognized quality award (48.2% of establishments). Firms were classified as

having an innovator based strategy if they reported being innovative as the primary focus of

9

Page 10: Michie

their strategy and if they had introduced either a product or process innovation over the past

three years (17.4% of establishments). Seventeen percent of establishments did not meet any

of these category specification and fall into the ‘other’ category.

(b) Background variables. The following standard control variables were included in the

analysis:

establishment size (see Appendix for details on the distribution), with dummies for

establishments with up to 49 employees; 50–99 employees; 100–199 employees; 200–

499 employees; and 500 or more employees (the omitted category is establishments with

less than 49 employees);

the age of the establishment (Age);

whether the establishment is UK or foreign owned (dForeign);

two sector dummies: manufacturing (dManuf) and services (dServ) (see Appendix); and

trade union recognition (20.1% of establishments).

(c) Labour market flexibility. In their analysis of the effect of labour market flexibility on firm

growth and innovation, Kleinknecht et al. (1997) categorise flexibility into two categories:

internal and external (these definitions are similar to Beatson’s (1995) definitions of flexibility

at the intensive and extensive margins, see above). The questions used in our survey on labour

market flexibility adopt a similar approach, as indicated in Table 1. However, the variables

associated with internal flexibility are grouped with HR practices. To make the analysis more

manageable, only the aggregate indicator of external flexibility – that is the number of

employees that are externally flexible as a percentage of total employees is used.

10

Page 11: Michie

Insert Table 1 here

(d) HRM practices. The measures of HRM used to test the relationships outlined above are

based upon the HRM practices listed in Table 2. The practices cover the areas of recruitment

and selection, training, job design, communication techniques and consultation, quality issues,

pay systems and terms and conditions of employment. The mean number of practices used

within the sample are 15.6. The nature in which the independent HRM variables are constructed

is discussed below.

Insert Table 2 here

HYPOTHESES TESTED

Testing the Impact of External Fit

As suggested by Schuler and Jackson (1987) and Hoque (1999), HRM should prove effective

only within firms emphasizing a quality-enhancer or innovator approach to business strategy,

and should prove ineffective where the primary emphasis of strategy is on cost-cutting, or price

based competition. The sample establishments are split into the four business strategy typologies

outlined above. Following the approach adopted by Hoque (1999), the measure of HRM used

within this part of the analysis is cumulative, with each establishment being ranked according to

the extent to which they have adopted the 22 HRM practices discussed earlier. This variable at-

tempts to examine the relationship between the extent to which HRM practices have been adop-

ted and performance. By splitting the sample by strategy typologies, and then regressing the ag-

gregate HRM variable on each of the dependent outcome variables, we assess the effectiveness

11

Page 12: Michie

of HRM in the context of ‘cost-reducer’, ‘quality-enhancer’, ‘innovator’ and ‘other’ business

strategies.

Hypotheses 1: The effectiveness of HRM is contingent upon the approach to business

strategy that has been adopted.

Testing the Universal Relevance of HRM

To examine the universal relevance hypothesis, we investigate whether establishments that

adopt HRM coupled with a focus on quality and/or innovation have performance levels

superior to those achieved by other establishments.

This issue is tested as follows: having split the sample four ways to perform the

external fit tests described above, the sample is reclassified here to enable comparisons

between business categories. These categories are as follows:

1. ‘low-HRM cost reducer’, using 10 or fewer HR practices: 28 establishments fall into

this category;

2. ‘medium-HRM cost reducer’, using more than 10 but fewer than 16 HR practices: 22

establishments fall into this category;

3. ‘high-HRM cost-reducers’, using 16 or more HR practices: 11 establishments fall into

this category;

4. ‘low-HRM quality-enhancers’, using 10 or fewer HR practices: 25 establishments fall

into this category;

12

Page 13: Michie

5. ‘medium-HRM quality-enhancers’, using more than 10 but fewer than 16 HR practices:

68 establishments fall into this category;

6. ‘high-HRM quality-enhancers’, using 16 or more HR practices: 83 establishments fall

into this category;

7. ‘low-HRM innovators’, using 10 or fewer HR practices: 10 establishments fall into this

category;

8. ‘medium-HRM innovators’, using more than 10 but fewer than 16 HR practices: 13

establishments fall into this category;

9. ‘high-HRM innovators’, using 16 or more HR practices: 41 establishments fall into this

category;

10. ‘low-HRM other’, using 10 or fewer HR practices: 31 establishments fall into this

category;

11. ‘medium-HRM other’, using more than 10 but fewer than 16 HR practices: 21

establishments fall into this category;

12. ‘high-HRM other’, using 16 or more HR practices: 12 establishments fall into this

category.

Hypothesis 2: The level of performance is dependent on the approach taken to HRM

and to business strategy; ‘high-HRM quality-enhancer’ and ‘high-HRM innovators’

outperform the other categories within the sample?8

13

Page 14: Michie

Given the diverse product markets in which the firms in our sample operate, evidence of the

universal relevance of HRM would be very significant.

Testing the Importance of Internal Fit

The final hypothesis examines the importance of introducing HRM as a synergistic package

of mutually supporting practices. According to bundling theory, establishments that adopt a

wide range of HRM practices, and those introducing their HRM practices as a coherent,

institutionally supported synergistic package should outperform establishments within which

HRM has been introduced in an ad hoc manner. To test this, a trichotomous variable was

constructed as follows:

(a) ‘strategic HRM’ establishments: above-average (16 or more) usage of HRM

strategically integrated with each other: 141 (46.3%) fall into this category;

(b) ‘non-strategic’ HRM establishments (16 or more) usage of HRM practices, which are

not strategically integrated: 53 (17.3%) fall into this category;

(c) ‘low-HRM’ establishments (fewer than 16) usage of HRM practices: 110 (36.5%) fall

into this category.

If a senior HR representative takes part in the company’s overall strategic decision

making and if the respondent reported that HR policies are deliberately integrated with each

other, the establishment is categorized as having ‘strategically integrated’ its HRM practices

in the typologies above.

14

Page 15: Michie

Hypothesis 3: Firms that introduce HRM techniques within an institutionally

supported, coherent package outperform those that introduce similar numbers of HRM

practices but in an ad hoc manner and not as part of an overall policy or strategy.

If internal fit is important, the ‘strategic HRM’ establishments should outperform the other

establishments within the sample.

1 Hoque (1999) closely follows Schuler and Jackson’s (1987) strategy typologies.

2 Khatri (2000) uses Miles and Snows (1984) strategy typologies. Khatri’s

analysis is carried out on a sample of firms based in Singapore.

3 It is important to differentiate between the universal effects that HRM might

have, and the universal relevance of HRM as an approach. Universal effects of HRM

suggest that HRM effects performance irrespective of the business strategy adopted,

which contradicts the external fit hypothesis. In contrast, the universal relevance of

HRM suggests that the effectiveness of HRM may be contingent upon the type of

business strategy adopted. In other words, HRM may be most effective when coupled

with a quality-enhancer or innovator strategy and may be least effective when coupled

with a cost-based or non-coherent (‘other’) strategy. Thus, tests for the universal

relevance of HRM do not contradict the contingency arguments (see Hoque, 1999:

422).

4 Hoque argues, therefore, that, “…a ‘high-HRM quality-enhancer’ strategy

would be the key to competitive success within hotels of the nature under investigation

here, with there being little or no scope for a strategy based on cost reduction or price

15

Page 16: Michie

RESULTS

Hypothesis 1: Effects of HRM on Performance is Contingent upon Strategy

Table 3 examines the relationship between HRM and performance outcomes and provides

strong evidence for Hypothesis 1, that the effectiveness of HRM is dependent upon the

achievement of external fit.

competition to achieve comparable results” (1999: 439).

55. For a detailed discussion and typology of flexible work practices see Beatson, 1995.

6 The EXTEL database is a company/organisation level database whereas our study is of

establishments. Large organisations can have many quite small establishments. The

EXTEL database is biased towards larger, publicly quoted organisations and those in the

manufacturing sector. For example, almost 50% of the sample organisation’s primary

activity was in manufacturing whereas only 18% of private sector establishments in the

WERS sample were engaged in manufacturing.

7. We spent a day during the first week of the interviews with the staff at the survey company

answering queries, listening to interviews and making suggestions in relation to prompting

and explanations of terminology, etc.

8. If in doubt of a term during the interview, the individual was asked to refer to this glossary.

7 For 73 companies, the HR person was unable to answer parts of the questionnaire (e.g.,

some of the questions about performance and innovation). In such cases, the name of the

most appropriate person in the company was obtained from the HR person and this person

was contacted. Completed questionnaires were obtained for 61 of such companies.

16

Page 17: Michie

Insert Table 3 here

It can be seen that across the sample as a whole, there is a strong positive relationship

between the extent to which HRM is used and all three of the performance measures. However,

for the cost-reducer firms, this positive relationship disappears for sales and labour productivity

and is negative (although insignificantly so) for financial performance. For quality enhancer and

10. Namely, the Company Chairperson, Managing Director, Chief Executive, Manufacturing

Director, or Production Director. Since the data were collected primarily from a single

source that the potential for common-method variance must be taken into consideration. In

the context of HRM and performance, common-method variance problems are likely to

arise because of the phenomenon of higher performance ratings being reported by

respondents, especially when the respondent is a HR Director, who claim to have adopted a

wide range of HRM practices. This problem is reduced to some extent in this analysis in

terms of the performance variables since these are drawn the EXTEL database and

therefore objective rather than subjective evaluations of performance.

11. The definition of ‘establishment’/‘site’ in this study corresponds broadly with that of

‘workplace’ used in the Workplace and Employment Relations Survey (WERS):

‘workplaces are sub-sets of organisations, except where the workplace is the sole one in an

organisation and it is only here that the term ‘workplace’ and ‘organisation’ are

interchangeable’ (Cully, et al. 1999: 3-4). For detail of WERS see the December 2000

special issue of the British Journal of Industrial Relations (Volume 38, number 4).

12. Labour productivity is computed as the log of total sales/number of total employees.

17

Page 18: Michie

innovator establishments the HRM measure correlates positively and significantly with all

performance measures. For establishments in the ‘other’ category, while the correlations

between the HRM measure and performance measures are positive, none of the relationships are

significant. It is likely that this reflects the ambiguities within such establishments. The use of

externally flexible labour (interacted with HRM usage) reduces the positive correlation found

between HRM and the performance measures.

13. The definition of a product innovation was as follows: ‘a new or significantly improved

manufactured or service product which is introduced to the market and requires

changes in knowledge or skills, routines, competence, equipment, or engineering

practices to make the new product. Changes which are purely aesthetic (such as

changes in colour or decoration) or which simply involve product differentiation

(minor design or presentation changes which leave the product technically unchanged)

are NOT to be classified as product innovations.’ The definition of process innovation

was as follows: ‘a new or significantly improved production, delivery or distribution

method and which requires changes in knowledge or skills, routines, competence,

equipment, or engineering practices to introduce the new process’.

14. Seventy-nine percent of the sample establishments were multi-site. Thus, in almost all

cases establishment size will vary from organisation size.

15. Trade union recognition is used rather than trade union presence, since presence does

not necessarily imply recognition. Recognition is a key indicator of a formal role for a

trade union (see Guest, et al., 2000 for further discussion).

18

Page 19: Michie

These results give support to the external fit hypothesis, that the effectiveness of HRM is

strongly dependent upon the business strategy pursued. As hypothesized, there is no evidence

that the adoption of HRM leads to improved performance where establishments emphasize cost

control within their business strategies. In contrast, the evidence presented shows that where

HRM is coupled with strategies that emphasize quality-enhancement and innovation,

performance is positively affected. Reliance on externally flexible labour offsets, to some

extent, the positive association between HRM and performance. This is not surprising. While

‘external flexibility’ may include some examples of what might be termed ‘high road’ practices,

consistent with investing in progressive HR practices – for example where skilled employees

16. Future research will compare performance outcomes between ‘high-RM quality-

enhancers’ and ‘high-HRM innovators’.

17. Hypotheses concerning the relationship between the adoption of HRM and performance

is difficult for establishments in the ‘other’ category. The ambiguity of such a strategy

may be what Porter (1985: 16-17) describes as ‘stuck in the middle’ (see Hoque, 1999:

427, for further discussion).

18. This approach follows closely that of Hoque (1999), with cost-reducer establishments

being dropped from this section, as there is little evidence of an HRM-performance

relationship within these establishments.

8 This series of dummies outlined above enables this hypothesis to be tested. By holding

6 and 9 constant we can test whether the ‘high-HRM quality-enhancer’ and ‘high-HRM

innovators’ outperform the other categories.

19

Page 20: Michie

choose to work on fixed-term contracts – much of this type of ‘labour flexibility’ will be of the

‘low road’, hire-and-fire type. Reliance on this type of externally flexible labour is usually

associated with the sort of cost control that we found led to HRM practices having no positive

effect on performance.

Hypothesis 2: Universal Relevance of HRM

This section examines whether establishments that adopt ‘high-HRM quality-enhancer and high

HRM innovator’ approaches are the best performing establishments within the sample. Such a

finding would suggest that HRM coupled with an emphasis on either quality or innovativeness

holds universal relevance for firms. In contrast, if ‘low-HRM cost-reducer’ establishments are

performing equally effectively, the implication will be that a high-HRM approach is not

necessarily universally relevant, and that there is sufficient diversity within product markets for

alternative approaches to business strategy and HRM to prove equally effective.

Insert Table 4

The results in Tables 4 suggest that the ‘high-HRM quality enhancers and innovators’ are the

highest performing establishments in terms of labour productivity and financial performance

within the sample (Hypothesis 2). The exception to this universalism however appears in

relation to sales growth, where ‘high-HRM cost reducers’ and ‘high-HRM other’ (where the

signs are positive but not significant). This may reflect the fact that it is relatively easy for firms

to increase sales growth compared to the other performance measures. Moreover, increased

sales growth will not necessarily improve financial performance, especially if the price of the

product or service is reduced in order to secure the increased volume. It is also important to

recognize that where universalism does not hold, it is in the context of ‘high’ HRM usage which

20

Page 21: Michie

may thereby offset, at least in part, the potential negative effects of cost based and ambiguous

strategies.

Hypothesis 3: The Importance of Internal Fit

This analysis examines whether establishments that appear to have introduced their HRM

practices as a strategically integrated package of mutually supporting practices outperform

establishments that have introduced their practices in a more ad hoc manner.

Insert Table 5 here

The results in Table 5 show that ‘strategic HRM’ establishments outperform the ‘low-

HRM’ establishments across all of the performance measures (Hypothesis 3). In contrast, the

‘non-strategic HRM’ firms do not outperform the ‘low-HRM’ establishments significantly

for any of the measures. A higher dependence on externally flexible labour weakens the

positive correlations between strategic HRM and all of the performance measures.

Interestingly, however, where non-strategic HRM is present, the use of externally flexible

labour actually strengthens the correlation (although the relationship is not significant)

between the performance measures. In other words, external flexibility may help to offset the

lack of strategic HRM in these establishments. Overall, these results suggest that, when firms

do introduce HRM practices, the returns on HRM are much greater where they are

introduced as part of an institutionally supported and coherent package.

CONCLUSION

Investing in ‘progressive’ HR practices can clearly pay dividends in terms of corporate

performance. However, the results reported in this paper suggest that the degree to which this is

21

Page 22: Michie

true – in statistical terms, the size and significance of the effect – will vary according to a range

of factors. One of these factors is the strategy that the firm adopts. Broadly, it may be pursuing a

‘high road’ strategy of investing in progressive HR practices that will be expected to lead to a

greater degree of commitment and motivation amongst the workforce, as well as to both an

increased ability and greater opportunities to work more productively. Hence the HR investment

will lead to improved productivity and profitability. Alternatively, the firm may choose a ‘low

road’, cost-cutting strategy. This may include putting employees onto short-term contracts

and/or part-time working, and accepting a greater degree of labour turnover. In this case we

would not expect to find a high level of investment in HR. And if such HR practices were to be

pursued, they would be unlikely to lead to improved outcomes. There is thus no automatic or

general link from HR practices to outcomes; it will depend, in part at least, on the strategy being

pursued.

If a ‘high road’ strategy is consciously chosen, then the costs of investing in HR practices can

be expected to be recouped through improved performance. However, for this to happen

requires the HR practices to lead not just to higher levels of commitment and motivation

amongst staff, it is also necessary for this to be matched firstly by the skills to work more

productively, and also by the opportunities to put those skills and motivation to good effect. For

these three factors to be present – motivation, skills and opportunities – requires HR practices to

be pursued as coherent ‘packages’, and to be combined with appropriate organizational design.

22

Page 23: Michie

Table 1. Labour Market Flexibility Variables

External Employment Flexibility Variables. Mean No. of valid cases (n = 365)

External flexibility:

Number of employees that are externally flexible as a

percentage of total employees:

a. Number of part-time employees as a percentage of total:

b. Number of employees on temporary contracts as a

percentage of total employees:

c. Number of employees on fixed-term, casual, or seasonal

contracts as a percentage of total employees:

0.22

0.21

0.19

0.28

354

364

361

354

Note: The results for external flexibility are similar to those reported by the 1998

WERS survey. See Milward, et al., 2000, pp. 43-49.

23

Page 24: Michie

Table 2. HRM Practices used in the Analysis

HRM Practice Mean No. of valid cases(n = 365)

Recruitment and Selectiona. Use of at least one of the following selection methods: psychometric testing; personality/attitudinal tests; work sample; aptitude tests.b. Trainability as a major selection criteriac. Use of realistic job previews during recruitment and selection

0.264

0.6830.552

362

357358

Incentive Pay and Appraisala. A merit element in the pay of staff at all levelsb. Formal appraisal of the majority of non-managerial staff on a regular basis (at least annually)

0.605

0.623

360

361Training a. Formal induction programme for new employeesb. The majority (> 50%) of non-managerial employees received formal off-the-job training in the past 12 monthsc. Investor in People (IIP) Status

0.73

0.4870.547

364

363364

Job Designa. Redistribution of tasks amongst non-managerial employees over the past 3 years.b. Extent to which non-managerial employees have variety in their work (‘moderate’ – ‘entirely’).c. Work organized around team-working for the majority of employees

0.489

0.5320.623

352

358360

Quality Issuesa. Production/service staff have quality responsibilitiesb. A majority of employees (> 50%) are involved in quality circles or quality/continuous improvement teams

0.734

0.482

362

361Communication and Consultationa. A system of regular, planned team briefings involving senior management during which time work stopsb. Joint Consultative Committees (JCCs) c. All staff are informed about the market position and competitive pressures faced by the establishment and/or company d. All staff are informed about the strategic objectives and targets for the establishment and/or company

0.7930.415

0.689

0.734

364363

362

362

Terms and Conditionsa. Harmonized terms and conditions between management and non-management staffb. Internal promotion the norm for appointments above the basic levelc. Written policy of guaranteed job security or non-compulsory redundancies amongst permanent staffd. Written commitment to a goal of long-term employment security amongst permanent staff.

0.6890.634

0.152

0.091

361364

363

363

24

Page 25: Michie

Table 3. The Relationship between HRM Practices, Labour Market Flexibility, and Organizational Performance

Sales Growth

All Firms Cost-reducers Quality-enhancers

Innovators Others

HRMHRM * External flexd50 – 99 employeesd100 – 199 employeesd200 – 499 employeesd500+ employeesAgedForeignUnion RecognitiondManuf

nPseudo R2

0.126 (2.825)*** 0.103 (2.265)**-0.225 (1.632)-0.221 (1.963)*-0.113 (1.234) 0.252 (2.452)*** 0.006 (1.113) 0.184 (1.989)*-0.167 (1.732)-0.056 (1.156)

3640.156

0.098 (1.138) 0.075 (1.076) 0.067 (1.113) 0.024 (0.438)-0.167 (2.665)**-0.215 (2.234)* 0.003 (0.062)-0.156 (2.076)*-0.129 (1.100) -0.069 (1.123)

600.123

0.135 (2.316)** 0.123 (2.118)* -0.058 (1.113) -0.089 (1.342) 0.213 (2.267)* 0.228 (2.327)** 0.115 (1.145) 0.197 (1.772) 0.156 (1.223)-0.067 (1.356)

1760.138

0.158 (2.567)*** 0.136 (2.135)** 0.120 (1.067)-0.096 (1.123)-0.102 (1.223) 0.233 (2.223)***-0.127 (1.976)* 0.156 (1.886)* 0.145 (1.665) 0.158 (1.823)*

630.119

0.076 (1.003) 0.052 (0.536) 0.103 (1.043) 0.145 (1.129) 0.076 (1.067) 0.192 (1.145) 0.007 (0.467) 0.192 (2.113)*-0.163 (1.654)-0.045 (1.056)

610.115

Labour Productivity

All Firms Cost-reducers Quality-enhancers

Innovators Others

HRMHRM * External flexd50 – 99 employeesd100 – 199 employeesd200 – 499 employeesd500+ employeesAgedForeignUnion RecognitiondManuf

nPseudo R2

0.189 (2.991)*** 0.154 (2.251)** 0.117 (1.227) 0.198 (1.554) 0.156 (1.146)-0.195 (2.403)*** 0.015 (1.002) 0.221 (2.113)*-0.172 (1.665) 0.078 (1.123)

3640.165

0.102 (1.100) -0.093 (1.123) 0.054 (1.117) 0.033 (0.449)-0.129 (1.332)-0.182 (2.112)* 0.007 (0.058) 0.087 (1.223)-0.189 (1.226) 0.112 (1.103)

600.130

0.126 (2.267)** 0.119 (1.932)* -0.022 (1.001) -0.066 (1.225) 0.227 (2.123)* 0.238 (2.426)** 0.005 (0.765) 0.178 (1.550) 0.112 (0.384) 0.115 (1.118)

1760.125

0.234 (2.876)*** 0.202 (2.209)**-0.118 (1.123)-0.100 (0.107)-0.042 (0.289) 0.267 (2.118)*-0.112 (1.382) 0.123 (1.956)* 0.152 (1.434) 0.116 (1.076)

640.115

0.128 (1.345) 0.112 (1.259) 0.007 (0.765) 0.106 (1.197) 0.055 (1.110)-0.206 (1.232) 0.056 (0.827) 0.177 (1.219)-0.065 (1.113) 0.089 (1.116)

620.107

Financial Performance

All Firms Cost-reducers Quality-enhancers

Innovators Others

HRMHRM * External flexd50 – 99 employeesd100 – 199 employeesd200 – 499 employeesd500+ employeesAgedForeignUnion RecognitiondManuf

nPseudo R2

0.143 (3.221)*** 0.105 (2.551)** 0.123 (1.334) 0.205 (1.678) 0.134 (1.223) 0.204 (2.333)*** 0.123 (1.113) 0.186 (2.383)**-0.153 (1.432)-0.075 (0.887)

3590.178

-0.093 (1.003) -0.132 (1.223) 0.047 (1.062) 0.027 (0.33`) 0.104 (1.192) 0.124 (1.028) 0.013 (0.066) 0.105 (1.382)-0.165 (1.079)-0.004 (0.103)

600.128

0.176 (3.331)** 0.166 (3.115)*** -0.114 (1.123) -0.043 (1.116) 0.219 (2.256)* 0.251 (2.737)*** 0.018 (0.589) 0.228 (2.118)* 0.007 (0.067) 0.125 (1.382)

1750.154

0.134 (2.754)*** 0.113 (2.207)**-0.123 (1.289)-0.176 (1.134) 0.218 (1.876)* 0.222 (2.209)*-0.107 (1.234) 0.157 (2.103)* 0.066 (1.348) 0.111 (1.279)

630.122

0.116 (1.067) 0.109 (1.008) 0.100 (0.892) 0.056 (1.107) 0.128 (1.223) 0.256 (2.532)** 0.102 (1.003) 0.145 (1.332) 0.085 (1.009)-0.176 (1.225)

610.119

Notes: Ordered probit analysis. Coefficients given (t statistics in brackets): *** significant at 1%; ** significant at 5%; * significant at 10%.

25

Page 26: Michie

Table 4. HRM, Strategy, Flexibility, and Performance Outcomes

Sales Growth Labour Productivity

Financial Performance

‘Low-HRM cost reducers’

‘Medium-HRM cost reducers’

‘High-HRM cost reducers’

‘Low-HRM quality enhancers’

‘Medium-HRM quality enhancers’

‘Low-HRM innovators’

‘Medium-HRM innovators’

‘Low-HRM others’

‘Medium-HRM others’

‘High-HRM others’

d50 – 99 employees

d100 – 199 employees

d200 – 499 employees

d500+ employees

Age

dForeign

Union Recognition

dManuf

n

Pseudo R2

-0.315 (2.422)**

-0.223 (1.382)

0.136 (1.206)

-0.207 (1.123)

-0.119 (1.100)

-0.121 (1.007)

-0.105 (1.288)

-0.312 (2.383)**

-0.221 (1.765)

0.123 (1.064)

-0.134 (1.056)

-0.228 (1.754)

0.156 (1.661)

0.226 (1.997)*

0.014 (0.076)

0.382 (2.356)**

-0.082 (1.005)

-0.116 (1.045)

364

0.124

-0.662 (3.876)***

-0.477 (2.557)**

-0.187 (1.333)

-0.324 (1.283)

-0.229 (1.007)

-0.189 (1.082)

-0.087 (0.976)

-0.886 (3.664)***

-0.176 (1.223)

-0.107 (1.022)

0.125 (1.043)

0.138 (1.156)

0.277 (1.966)*

-0.214 (1.877)*

-0.033 (1.003)

0.447 (2.636)***

0.162 (1.543)

0.128 (1.116)

364

0.150

-0.224 (1.904)*

-0.113 (1.662)

-0.105 (1.107)

-0.457 (2.483)**

-0.336 (1.813)*

-0.526 (2.959)***

-0.332 (2.286)**

-0.256 (1.632)

-0.101 (1.013)

-0.088 (0.659)

-0.182 (2.005)*

0.126 (1.256)

0.344 (2.245)**

0.350 (2.488)**

-0.023 (1.134)

0.358 (2.307)**

0.123 (1.327)

-0.097 (1.003)

359

0.144

Note: See Table 3. Omitted Category = ‘High-HRM quality-enhancers’ and ‘High-HRM innovators’.

26

Page 27: Michie

Table 5. Internal Fit, Flexibility, and Performance Outcomes

Sales Growth

Labour Productivity

Financial Performance

‘Strategic HRM’

‘Strategic HRM’ * External flex

‘Non-strategic HRM’

‘Non-strategic HRM’ * External flex

d50 – 99 employees

d100 – 199 employees

d200 – 499 employees

d500+ employees

Age

dForeign

Union Recognition

dManuf

n

Pseudo R2

0.536 (2.887)*** 0.456 (2.554)** 0.115 (1.266) 0.134 (1.703) -0.103 (1.023) -0.172 (1.635) 0.121 (1.226) 0.205 (1.766) 0.010 (0.081) 0.323 (2.141)* -0.070 (1.012) -0.115 (1.042)364

0.104

0.775 (3.345)*** 0.663 (2.441)** 0.125 (1.456) 0.187 (1.636) 0.117 (1.132) 0.130 (1.132) 0.205 (1.803)* -0.209 (1.812)* -0.045 (0.099) 0.376 (2.291)** 0.177 (1.668) 0.119 (1.007)364

0.102

0.802 (3.554)*** 0.653 (2.346)** 0.144 (1.234) 0.163 (1.357)-0.160 (1.102) 0.105 (1.177) 0.322 (2.117)* 0.267 (2.200)*-0.033 (1.141) 0.342 (2.318)** 0.104 (1.296)-0.086 (0.953)359

0.112

Notes: See Table 3.

‘Strategic HRM’ = above average no. of HR practices used and establishment has a formal HR strategy.

‘Non-strategic HRM’ = above average no. of HR practices used but establishment does but have a formal HR strategy.

Omitted category = below-average no. of HR practices used (‘low-HRM’).

27

Page 28: Michie

Appendix: Sample Details

Sample EstablishmentsCount %

Manufacturing:

Establishment Size:Under 5050 – 99100 – 199200 – 499500+

Total

Industrial Activity:Food and DrinkTextiles and ClothingWood, wood products, paper & printingChemicalsMetals and fabricationIndustrial and Commercial MachineryElectronic and Electrical EquipmentTransportation EquipmentOther Manufacturing

Total

Services:

Establishment Size:Under 5050 – 99100 – 199200 – 499500+

Total

Industrial Activity:Hotels and RestaurantsWholesale and RetailFinancial Other Business Services

Total

41 16.9 46 19.0 51 21.1 62 25.6 42 17.4

242 100.0

33 13.623 9.545 18.621 8.730 12.431 12.829 12.014 5.816 6.6

242 100.0

46 36.8 14 11.2 18 14.4 16 12.8 31 24.8

125 100.0

31 24.8 39 31.2 23 18.4 32 25.6

125 100.0

Note: To preserve the robustness of the results, only two dummy variables – dManuf and dServ – are used in the estimations. Given the relatively small cell sample sizes for each industry, when the results were run using a higher degree of industry disaggregation, these coefficients were not significant.Foreign Ownership: 293 (80.2%) of the sample firms were UK owned and 72 (19.7%) were foreign owned.

28

Page 29: Michie

References

Appelbaum, E., Bailey, T., Berg, P. and Kalleberg, A.L. (2000). Manufacturing Advantage:

Why High-Performance Work Systems Pay Off. Ithaca, NY: Cornell University Press.

Arulampalam, W. and Booth, A. (1998). ‘Training and Labour Market Flexibility: Is there a

Trade-off?’. British Journal of Industrial Relations, vol. 36, no. 4, pp. 521-536.

Beatson, M. (1995). Labour Market Flexibility, Research Series No. 48. Sheffield: Employment

Department.

Becker, B. and Gerhart, D (1996). ‘The impact of human resource management on

organizational performance: progress and prospects’. Academy of Management Journal, vol.

39: 779-801.

Cully, M., O’Reilly, A., Woodland, S., Dix, J. (1999). The 1998 Workplace Employee Relations

Survey: First Findings. London: HMSO.

Guest, D. (1987). ‘Human Resource Management and Industrial Relations’. Journal of

Management Studies, vol. 24(5): 503-21.

Guest, D. and Hoque, K. (1994). ‘The good, the bad and the ugly: employee relations in new

non-union workplaces’, Journal of Management Studies, vol. 31.

Hoque, K. (1999). ‘Human Resource Management and Performance in the UK Hotel Industry’.

British Journal of Industrial Relations, vol. 37, no. 3, pp. 419-443.

Huselid, M. (1995). ‘The Impact of Human Resource Management on Turnover, Productivity

and Corporate Financial Performance’. Academy of Management Journal, vol. 38, pp. 635 –

672.

29

Page 30: Michie

Huselid, M. and Becker, B. (1996). ‘Methodological Issues in Cross-Sectional and Panel

Estimates of the Human Resource-Firm Performance Link’. Industrial Relations, vol. 35, pp.

400–422.

Ichniowski, C., Shaw, K. and Prennushi, G. (1994). The Effects of Human Resource

Management Practices on Productivity. New York: Columbia University Press.

Khatri, N. (2000). ‘Managing Human Resource for competitive advantage: A Study of

Companies in Singapore’. International Journal of Human Resource Management, vol.

11(2): 336-365.

Kleinknecht, A. (1998). ‘Is labour market flexibility harmful to innovation?’. Cambridge

Journal of Economics, vol. 22, no. 3, pp. 387 - 396.

Kleinknecht, A., Oostendorp, R. and Pradhan, M. (1997). Flexible labour, firm growth and

employment: an exploration of micro data in the Netherlands, mimeo, 2 December 1997.

Kochan, T. and Barocci, T. (1985). Human Resource Management and Industrial Relations:

Text, Readings and Cases. Boston: Little Brown.

MacDuffie, J. (1995). ‘Human Resource Bundles and Manufacturing Performance:

Organizational Logic and Flexible Production Systems in the World Auto Industry’.

Industrial and Labour Relations Review, vol. 38: 197-221.

Michie, J. and Wilkinson, F. (1995). Wages, Government Policy and Unemployment, Review

of Political Economy, vol. 7, no. 2, pp. 133-149.

Michie, J. and Prendergast, R. (1998). Government Intervention in a Dynamic Economy,

New Political Economy, vol. 3, no. 3, pp. 391-406.

30

Page 31: Michie

Michie, J. and Sheehan-Quinn, M. (2001). ‘Labour Market Flexibility, Human Resource

Management and Corporate Performance’, British Journal of Management, vol. 12, 287-306.

Michie J. and Sheehan, M (2003). ‘Labour Market Deregulation, Flexibility and Innovation’,

Cambridge Journal of Economics, vol. 27(1), January.

Miles, R. and Snow, C. (1984). ‘Designing strategic human resource systems’.

Organizational Dynamics, Summer: 35-52.

Osterman, P. (1999). Securing Prosperity. Princeton: Princeton University Press.

Porter, M. (1985). Competitive Advantage: Creating and Sustaining Superior Performance.

New York: Free Press.

Schuler, R. (1989). ‘Strategic Human Resource Management and Industrial Relations’. Human

Relations, 42: 157-84.

Schuler, R. and Jackson, S. (1987). ‘Linking Competitive Strategies with Human Resource

Management Practices’. Academy of Management Executive, vol. 1: 207-19.

Tichy, N., Fombrun, C., and Devanna, M. (1982). ‘Strategic Human Resource Management’.

Sloan Management Review, vol. 11(3): 47-61.

Truss, C. and Gratton, L. (1994). ‘Strategic Human Resource Management: A Conceptual

Approach’, International Journal of Human Resource Management, vol. 5(3): 663-86.

Walton, R. (1985). ‘From Control to Commitment in the Workplace’, Harvard Business

Review, March-April: 76-84.

31

Page 32: Michie

Wood, S. and Albanese, M. (1995). ‘Can we Speak of High Commitment Management on the

Shop Floor?’, Journal of Management Studies, vol. 32, pp. 215 – 247.

Wood, S. and de Menezes, L. (1998). ‘High Commitment Management in the UK: Evidence

from the Workplace Industrial Relations Survey, and Employers’ Manpower and Skills

Practices Survey’, Human Relations, vol. 51, pp. 485 – 515.

Wood, S. (1999). ‘Getting the Measure of the Transformed High-Performance Organization’,

British Journal of Industrial Relations, vol. 37, no. 3, pp. 4391 - 417.

32

Page 33: Michie

Endnotes

33