PERFORMANCE, HR PRACTICES AND THE HR MANAGER IN SMALL, ENTREPRENEURIAL FIRMS
Gerry Kerr
OdetteSchool of Business
University of Windsor
Sean A. Way
Faculty of Business Administration
The ChineseUniversity of Hong Kong
James Thacker
OdetteSchool of Business
University of Windsor
Proceedings of the Administrative Science Association of Canada, June 5-8; Quebec City, Quebec. Currently under review at the JSB&E
PERFORMANCE, HR PRACTICES AND THE HR MANAGER IN SMALL, ENTREPRENEURIAL FIRMS
Research indicating a positive relationship between effective HR activities (a high performance work system (HPWS)) and performance has focused primarily on large organizations. The present study provides evidence that the relationship also holds for small firms. As well, the presence of an HR manager is related to having a HPWS. Differences between our findings and similar research are also discussed.
Introduction and Review of the Literature
There is considerable evidence that effective human resources (HR) practices improve performance and provide organizations with a competitive advantage (e.g., Becker & Gerhart, 1996). The extant research, however, was almost always conducted on large organizations, leaving suspect the generalizability of the findings to small firms. Although these organizations are vital to an economy, small-firm research is limited, in spite of the many calls for such a focus (Heneman & Berkley, 1999; Good, 1998; Hornsby & Kuratko, 1990). More attention needs to be focused on the HR activities and outcomes of small firms, with resulting information provided to those involved in these organizations (Heneman, Tansky & Camp, 2000).
The first issue for consideration is defining the key focus, the small firm. Historically, when research on the small firm has been conducted, the definition of the entity has varied markedly. For example, McEvor (1984) defined small business as those with 250 employees or fewer, Hornsby & Kuratko, (1990) as those with 150 employees or fewer, and Flannagan & Deshpande, (1996) as organizations with 500 employees or fewer. But, defining "small" as 500 or fewer could result in as much variance in HR practices within the sample of small organizations as between large and small. For example, Hornsby and Kuratko (1990) examined three levels of small business (1) 50 or fewer employees, (2) 51- 100 employees and (3) 101 - 150 employees. The authors found important differences between the three classifications they examined; furthermore, the study did not even consider organizations with 150 - 500 employees.
For Canadian organizations, the definition for small business is generally accepted as fewer than 100 employees. Statistics Canada indicates that small firms (defined as 100 or fewer employees) make up about 98 percent of the Canadian economy (Industry Canada, 2003). Therefore, increasing the effectiveness of small, entrepreneurial firms through relevant research is important.
There has been some help in that regard. Wager & Ross (1998) examined recruitment, performance appraisal and workplace practices in small businesses in Eastern Canada. The authors noted a tendency for firms to have more effective HR practices as size increased, especially if there was an HR expert in the organization. In other work, using the organizations from the same region, Wager and Hammock (2000b) reported only 28 percent of small businesses had formal training in place. The authors also noted formal training was related to the size of the organization: almost 57 percent of organizations with 25 or more employees had instituted formal training. Finally, Wager & Hammock (2000a) indicated that 43 percent of surveyed firms had formal performance appraisal systems, 61 percent had some sort of incentive program tied to productivity, and 56 percent used a bonus system for high performers. In a study of small United States businesses, (between 20 and 100 employees) Way (2002) noted that high performance work systems were correlated with turnover and perceived productivity. As well, the author noted the measure of perceived productivity was problematic given it was a single item regarding performance of the workforce, a complex outcome.
We focus our study on three areas. First, the relationship between HR practices and performance will be examined. We have a more representative sample than the Way (2002) study, with a number of firms containing fewer than 20 employees. We also have a more comprehensive measure of firm performance. So, this study offers the promise of more detailed and reliable evidence of a relationship between HR practices and firm performance, if one exists. Second, we will examine the role of the HR manager in the development of effective HR practices. Understanding the relationship between having an HR manager and firm performance and the reasons for deciding to have an HR manager are potentially important issues related to firm effectiveness. Finally, we will examine our sample as it relates to the findings of Wager and others noted above. The use in the studies of a single-location sample and the large number of very small organizations (with fewer than 10 employees) casts doubt on the generalizability of the findings.
High Performance Work Systems and Firm Performance
The implementation of a number of effective human resource practices in an organization provides what is called a High Performance Work Systems (HPWS). These HPWS's are positively associated with subjective (Delaney & Huselid, 1996) and objective (Guthrie, 2001) indicators of performance. HPWS's are conceptualized as a set of distinct but interrelated HRM practices that together select, develop, retain, and motivate a workforce in a competitively superior manner. In larger organizations, HPWS's are correlated with a variety of measures of performance (Becker & Huselid, 1998; Guthrie, 2001; Huselid, 1995; Way, 2002). These findings lead us to state hypothesis one:
H1: There is a positive relationship between small-firm performance and the degree to which the firm has an HPWS.
High Performance Work Systems and the Role of the HR Manager
Within the small business sector, gaining access to a workforce that produces superior employee output is key to firm success (Deshpande & Golhar, 1994; Golhar & Deshpande, 1997; Holt, 1993; Hornsby & Kuratko, 1990; McEvoy, 1984) and offers a source of sustainable competitive advantage (Flanagan & Deshpande, 1996). Wager and Ross (1998) noted that small firms with an expert in HR resulted in more effective HR practices. Thus, having an HR manager in charge of HR issues would likely result in more effective HR practices, because he or she would have the time and understanding necessary to set up such systems. The findings lead us to state hypothesis two:
H2: Small firms that have an HR manager will have a more advanced HPWS than firms that do not have an HR manager.
Firm Size and HR Management
Generally, the size of the organization could be expected to influence the formalization of the HR function. Wager and Ross (1998) reported the significantly higher presence of HR managers in larger firms. The authors noted that having an HR expert and the institution of better HR practices was related to the size of the firm. The findings lead us to state hypothesis three:
H3: There is a positive relationship between the small firm size and the presence of an HR manager.
Finally,we will compare some of our findings with the few studies that have been conducted using Canadian small businesses to determine the level of generalizability of the findings.
Methodology
Data Collection
A survey was administered to a group of small businesses, using a list purchased from Dunn & Bradstreet. Businesses were contacted by telephone and the most knowledgeable person regarding company practices (usually a top manager/owner) was requested. The survey was utilized to conduct an interview that generally required about 20 minutes to complete. Results were obtained from 99 organizations, but one had to be removed as it had only provided demographic information. The result was a 20 percent response rate.
Measures
Firm performance. Firm performance is a multi-dimensional construct (e.g., Myer & Gupta, 1994; Rogers & Wright, 1998), which is difficult to measure accurately. Furthermore, research suggests industry-related factors influence the absolute financial outcome scores of a firm, and thus it may be misleading to compare objective financial data across diverse industry settings (Covin & Sleven, 1989; Sapienza, Smith, & Gannon, 1988). To deal with the problem, some have suggested that firm performance should be conceptualized (and measured) as the degree to which the firm is generating outcomes that are important to the firm (necessary for the firm to attain firm goals and objectives). (See Gupta & Govandarajan, 1984; Kaplan & Norton, 2000; Rogers & Wright, 1998; Steers, 1975; Way & Johnson, 2001). These "subjective" measures of firm performance are closely correlated with objective measures (Dess & Robinson, 1984; Robinson & Pearce, 1988; Venkatraman & Ramanujam, 1987). Moreover, objective measures are often more difficult to obtain for small firms for the following reasons:
- managers of small firms usually exhibit high levels of secrecy, including great reluctance in providing “hard” financial data (Litvak, 1976; Fiorito & LaForge, 1986),
- the vast majority of small firms are privately held; thus, publicly disclosed financial data is not available and it is not possible to check the accuracy of responses using filing data (Covin & Sleven, 1989),
- even assuming the accuracy of reported financial data, these data are difficult to interpret because of the wide variety of business goals and practices in the small firm (Cooper, 1979), and
- the effects of industry-related factors on absolute firm outcome scores can be such that interpretation is misleading (Covin & Sleven, 1989; Sapienza et al., 1988).
Thus, a number of factors contributed to our decision to use subjective performance measures. Their use seemed appropriate because respondents were a group of managers in small firms, whose overview of organizational operations could be reasonably anticipated as secure, including their awareness of success or goal achievement.
The construct of performance effectiveness was measured using a methodology similar to that employed by Gupta and Govindarajan (1984), Colvin and Sleven (1989), and Miles et al. (2000). The measure is obtained by first asking respondents to indicate, on a 5-point Likert-type scale, ranging from "of little importance" to "extremely important”, the degree of importance their firm’s top managers attach to each of the following firm outcomes: (1) sales level, (2) sales growth rate, (3) cash flow, (4) gross profit margin, (5) net profit from operations, (6) profit-to-sales ratio, (7) ability to fund business growth from cash flow, (8) return on investment, and (9) change in (a) the value of the firm (in private firms) or (b) share price (in public firms). Respondents are then asked to indicate on a 5-point Likert-type scale, ranging from "highly dissatisfied" to "highly satisfied”, the extent to which their firm's top managers are satisfied with the firm's performance with regard to each of the nine performance outcomes. Each of the nine "satisfaction" scores is then multiplied by its corresponding "importance" score to create nine performance scores, which were then added together to create the firm effectiveness measure. The resulting measure of firm performance defines the degree to which the firm generates outcomes necessary for the attainment of a firm's broad-based goals and objectives.
High Performance Work Systems (HPWS). The methodology for measuring a HPWS is also taken from previous research in HR (see Becker & Huselid, 1998; Combs, Hall & Liu, in press; Huselid, 1995; Way, 2002). Seven components make up the measure (see Table 1). All seven components were assessed as they relate to two categories of employees, management/professional employees and non-management employees. The process worked as follows. We first generated two additive HR systems scores (the sum of the Z-scores of our HR systems’ seven components) for the two major categories of employees. Next, the two HR scores were weighted by the proportion of employees in its corresponding category (for example, non-management weighted HR score = (non-management HR score) * (number of non-management employees / total number of employee)). Finally, the 2 weighted HR scores were then added together to generate a unitary index, the high performance work systems measure. (Refer to Becker and Huselid (1998) and Delery (1998) for comprehensive discussions regarding the strengths and weaknesses of employing an additive approach to create a unitary index in HR systems research.)
Table 1
The Seven Components of the High Performance Work System (HPWS)
Component Label
/Definition/Measure
Selection ratio / The number of qualified applicants that the firm’s recruitment process generates per opening (on average) – five-point scale: 1 = 0, 2 = 1, 3 = 2 to 3, 4 = 4 to 8, and 5 = + 8 qualified applicants per positionScreening tests / The frequency in which screening tests (aptitude, work samples, etc.) are used to assist in the selection process – six-point scale: 1 = 0, 2 = up to 10, 3 = 11 to 30, 4 = 31 to 50, 5 = 51 to 75, and 6 = + 75 percent of the time
Formal training / The average of (A) and (B): (A) the number of days of training that the firm delivers to new hires in their first year – seven-point scale: 1 = 0, 2 = 0.5 or less, 3 = +0.5 to .999, 4 = 1 to 1.999, 5 = 2 to 2.999, 6 = 3 to 4.999, 7 = + 4.999 days. (B) the number of days of training that the firm delivers to a typical employee (per year) after their first year – seven-point scale: 1 = 0, 2 = 0.5 or less, 3 = +0.5 to .999, 4 = 1 to 1.999, 5 = 2 to 2.999, 6 = 3 to 4.999, 7 = + 4.999 days
Performance appraisal / The percent of employees who are regularly (i.e., at least annually) assessed via a formal performance appraisal process – six-point scale: 1 = 0, 2 = up to 10, 3 = 11 to 30, 4 = 31 to 50, 5 = 51 to 75, and 6 = + 75 percent of employees
Performance related pay / The percent of employees who receive merit increases or incentive pay based on performance – six-point scale: 1 = 0, 2 = up to 10, 3 = 11 to 30, 4 = 31 to 50, 5 = 51 to 75, and 6 = + 75 percent of employees
Compensation level / The level of employees’ total compensation – three-point scale: 1 = below industry standard; 2 = around industry standard; 3 = above industry standard
Communication / Frequent, accurate communication of firm “values” and “beliefs”, derived from agreement with statement “Your firm frequently and accurately communicates its culture to employees” (five-point scale): 1 = strongly disagree; 2 = disagree; 3 = neither disagree nor agree; 4 = agree; 5 = strongly agree
Analysis
A correlation coefficient was computed to test hypothesis one. For hypothesis two, a correlation of two dichotomous variables was computed. The process required dichotomizing the HPWS variable by defining each data point as above and below zero. Scores above zero were defined as more advanced than those below zero. To test hypothesis three, we created categories for organizational size similar to those found in the Wager and Ross (1998) study. A correlation coefficient was then calculated.
Results
Demographics
Of the 98 respondents, 17% had fewer than 20 employees, 47% had between 20 and 39 employees, 21% had between 40 and 59 employees, 7% had between 60 and 79 employees and the remaining 7% had 80 to 99 employees. Ten percent of responding firms had revenues up to $100,000, 20% between $100 and $200 thousand, 37% between $200 and $500 thousand, and 23% between $500 thousand and 1 million. The remaining 10% had revenues above $1 million. Half of the companies returning surveys were from Ontario, 14% from Alberta, and the remainder from the other provinces across Canada. The only province not represented was New Brunswick.
The organizations represented a cross section of industries, with the most coming from sundry manufacturing (29%) and service industries (28%). Five percent came from the finance area, 9% from transportation, and 13% and 14% were focused on wholesale and retail industries, respectively. Nine percent of our respondents were unionized.
Twelve of our reporting organizations (12.2% of total respondents) indicated they employed a Human Resource Manager (HRM) to deal with HR issues for management and non- management membership. Furthermore, only 5 of the 12 HRM's had some level of HR training, such as the provincial certificate or community college/university degrees. Thus, the level of education did not seem to have an impact on the HR procedures conducted in the organization.
Hypotheses
Hypothesis one was accepted. The relationship between high performance work systems and firm effectivenesswas significant (r = .184; p<.05, one tail test). Hypotheses two was also accepted (r = .179, p < .05 one tail test)[i]. (Table 2 shows that of the 12 HR managers, 9 are in firms with HPWSs.)
Table 2
The Presence of the High Performance Work Systems (HPWS)
and Human Resources (HR) Managers
HR MANAGER
HIGH PERFORMANCEWORK SYSTEM / No / Yes / Total
No
/ 45 / 3 / 48Yes
/ 41 / 9 / 50Total
/ 86 / 12 / 98However, hypothesis three was not supported (r = -.08; p> .05). Firm size did not significantly affect the presence of an HR manager in responding firms.
The HR manager
No single issue determined a need for an HR manager. For the 12 organizations that had an HR manager, four (33%) indicated his or her presence related to the growth or size of the organization, four (33%) indicated the need to conform with various legal requirements, three required increased effectiveness, and one indicated the amount of work required. This last issue may also be related to growth or conforming to legal requirements.
Comparisons with other Canadian small business findings
Contrary to previous research, having an HR manager was not significantly related to the size of the firm. Four of the firms with HR managers had fewer than 20 employees, three had between 20 and 39 workers, two had between 40 and 59 employees, one had between 60 and 79 workers, and two had between 80 and 99 employees (see Table 3). However, there may be a trend related to type of industry (see Table 4). The group of respondents in the finance area has the highest percent or HR managers, but the total number of responding finance organizations is small. Of the two industry types with the highest number of organizations in the sample, manufacturing has more than twice as many HR managers than the firms in service industries.
Table 3
Responsibility for Human Resources Activities*
FIRM SIZE
PERSON(S) RESPONSIBLE FOR HR ACTIVITIES / Under 20 employees / 20-39 employees / 40-59 employees / 60-79 employees / Over 80 employees / TotalHR manager / 4 (4) / 3 (3) / 2 (2) / 1 (1) / 2 (2) / 12 (12)
Owner/Manager / 2 (6) / 12 (17) / 5 (4) / 1 (3) / 0 (1) / 20 (31)
One Manager / 1 (2) / 3 (9) / 1 (5) / 0 (1) / 0 (0) / 5 (17)
Respective Dept Heads / 1 (2) / 2 (10) / 2 (5) / 0 (1) / 1 (2) / 6 (20)
Team of Managers / 2 (3) / 1 (4) / 1 (4) / 1 (0) / 1 (1) / 6 (12)
Other / 1 (0) / 3 (2) / 1 (1) / 0 (0) / 0 (0) / 5 (3)
No one / 6 (0) / 23 (2) / 9 (0) / 4 (1) / 3 (1) / 44 (3)
Total / 17 (17) / 47 (47) / 21 (21) / 7 (7) / 7 (7) / 98 (98)
*The organizational count for the responsibilities for non-management employees is provided in brackets