Human Capital and Sustainable Competitive Advantage: An Analysis of The Relationship Between Training and Performance

Download as pdf or txt
Download as pdf or txt
You are on page 1of 25

Int Entrep Manag J (2009) 5:139–163

DOI 10.1007/s11365-008-0090-3

Human capital and sustainable competitive advantage:


an analysis of the relationship between training
and performance

Ignacio Danvila del Valle &


Miguel Angel Sastre Castillo

Published online: 22 April 2008


# Springer Science + Business Media, LLC 2008

Abstract This article analyses the importance of training as a creator of human


capital, which enables a company to obtain competitive advantages that are
sustainable in the long-term that result in greater profitability. The study is based
on the general theoretical framework of resource and capacity theory. The study not
only analyses the impact of the influence of training on performance; it also attempts
to analyse the nature of such a relationship in greater depth. This being the case, an
attempt has been made to measure explanatory capacity from two different
perspectives: the universalistic approach and the contingent approach. At the outset,
two hypotheses are formulated that attempt to quantify the relationship from a
universalistic perspective to later, in two more hypotheses, incorporate the potential
moderating effect of the strategy into the model, in order to verify whether or not
this strategy improves the explanatory power of our model of analysis.

Keywords Training . Human capital . Competitive advantage . Intangibles .


Strategic human resources management

I. Danvila del Valle (*)


Centro Universitario Villanueva, Universidad Complutense de Madrid, Madrid, Spain
e-mail: idanvila@villanueva.edu

M. A. Sastre Castillo
Universidad Complutense de Madrid, Madrid, Spain
e-mail: masastre@emp.ucm.es

M. A. Sastre Castillo
Instituto Tecnológico Autónomo de México, Mexico City, Mexico
e-mail: msastre@itam.mx
140 Int Entrep Manag J (2009) 5:139–163

Introduction

A number of studies conducted in recent decades indicate that at present, the main
elements that distinguish organisations from each other are not tangible; rather, it is
intangible resources that are best able to meet the conditions necessary1 to become
competitive advantages that are sustainable over time.
According to these studies, the strategy of organisations should be based on
intangible resources, since it is these resources that provide greater differentiation
and due to the fact that they are not available in the marketplace; rather, they are
generally inherent to, and must be developed within, companies via a complex
process over a period of time.
With regards to intangible resources, consensus seems to have emerged in recent
years that people-related resources (human capital) can be considered one element
that explains the generation of larger economic rents. This is due to the fact that the
employees’ skills, abilities, knowledge and ability to adapt are scarce, valuable
intangibles that are hard to imitate and very difficult to replace. Furthermore, human
capital is an asset difficult to transfer from one organisation to another (Hall 1992)
and can be used in different areas of the business in a number of ways
simultaneously, without losing value.
As a consequence of current interest in this area of research, some research has
attempted to address this problem, suggesting that training is indeed an instrument
that allows the generation and accumulation of human capital. Most research
approaches this issue from a purely theoretical perspective and, to a lesser extent,
using empirical studies. Nevertheless, there are a number of reasons we believe this
area is a candidate for greater research.
Firstly, very few organisations (3%) measure the direct influence of training on
performance, due to the technical difficulty associated with its measurement (Sastre
Castillo and Aguilar Pastor 2003). If the academic world is able to demonstrate the
existence of such a relationship, managers will have a greater incentive to increase
investment in this area.
Furthermore, there is little research that focuses on the influence of training given
to personnel on results. Most research measures the effect of a set of human
resources practices; as a result, the training variable tends to be measured in a
simplistic manner and with a single indicator. Our research makes a special effort to
measure training using a multidimensional indicator of the same.
Similarly, there is little research that combines universalistic and contingent
approaches to explain such a relationship; most research favours one and emphasises
the negative aspects of the other. It is our understanding that our research constitutes
a relevant contribution inasmuch as it aims to compare the explanatory power of
both approaches, seeking to determine whether or not the strategy implemented by
organisations reduces the strength of said relationship. Therefore, our objective is not

1
For information on the criteria to be met by resources to be considered a sustainable source of competitive
advantage that generate rent, refer to research by Cool and Dierickx (1989), Barney (1991), Grant (1991), Amit
and Schoemaker (1993), Peteraf (1993), Black and Boal (1994) or Collis and Montgomery (1995).
Int Entrep Manag J (2009) 5:139–163 141

merely to explain the existence of the relationship between training and


performance, but also to find out more about the nature of this relationship.

Review of the literature and formulation of hypotheses

The universalistic approach says that there is a direct relationship between training
and performance. This perspective defends the existence of universally recognised
training practices, i.e. training practices that, introduced by any organisation
regardless of external or internal circumstances, will always have a positive effect
on performance (Rodríguez Pérez 2001).
As pointed out by Lee et al. (2005) to highlight that training has a greater
universalistic effect than other human resource practices, of the 16 best practices
studied by Pfeffer (1994), training is one of the few practices which was found to
have a consistently positive impact on results.
This result is also apparent in relation to both individual performance indicators (such
as sales volumes and productivity per employee) and organisation performance indicators
in research such as that conducted by Arthur (1994), Huselid (1995), Black and Lynch
(2001), Delaney and Huselid (1996), Koch and McGrath (1996), D’Arcimocles (1997),
Huselid et al. (1997), Kidder and Rouiller (1997), Barrett and O’Connell (2001), Bassi
et al. (2002), Aragón et al. (2003) or Tzafrir (2005). Nevertheless, the literature is
inconclusive: benchmark research such as that by Delery and Doty (1996) does not
detect a universalistic effect of training on the ROA and ROE of businesses in his
sample, while Úbeda García (2005) does not find a significant link with productivity
(which he attributes to an average period of investment in higher training, as a result of
which the likely effect on productivity did not emerge in his study). Even Hitt et al.
(2001) find that investment in training first has a negative effect on results then a
positive effect, as knowledge and skills acquired are applied to the job.
Thus, advocates of this approach consider that all organisations can and should
adopt a set of training policies. The logic underlying this idea is based on the notion
that the return on training will be greater than the investment, since it will increase
productivity (theory of human capital).
This effect is explained by the fact that if employees sense an interest on the part
of the organisation in offering them training, demonstrating trust in them and an
intention to count on them in the long-term, they will respond by making a greater
effort and being more efficient in the course of their duties. In resource and capacity
theory terms, this could constitute a barrier to mobility and lead to an improvement
in the performance of the organisation inasmuch as the company can appropriate this
increase in rent (Lee and Miller 1999; Úbeda García 2005).
Under these circumstances, training would be an important element in the
creation of human capital. This argument, based on equilibrium theory, is also
supported by Tzafrir (2005), who says that investment in training can create a moral
debt among employees who receive said training.
Nevertheless, Tzafrir (2005) also believes that investment in training can have
negative performance. For one, it is clearly a cost for the business, especially if the
business is not able to secure reciprocal commitment from the employee. Two, it
represents a risk for the company: if the employee decides to leave the organisation,
142 Int Entrep Manag J (2009) 5:139–163

it would represent a loss of human capital. This is the problem of transferability


contemplated in resource and capacity theory.
With this first approach, an attempt will be made to find out if the positive effects
outweigh the cost of the investment in training and therefore, if the investment has a
positive effect on the company’s performance.
Some of the literature has attempted to study the effect of training on performance
within a more or less broad set of human resources policies.
On this point, Barney (1991) argues that sets of complex practices that are hard to
replace, require constant investment in terms of funds and time with uncertain
methods for implementation, rather than policies implemented in isolation that are
easier to imitate, are the main sources of sustainable competitive advantage.
Nevertheless, we believe that it is not so apparent that the effects of training can
be easily imitated. This advantage obtained via the generation of human capital has
ambiguous causality, since each company has different social complexity and
training will always have some degree of specificity (Carpenter et al. 2001).
Indeed, Hatch and Dyer (2004) find that developing human capital through
training has an effect that cannot be compared with the acquisition of human capital
via the recruitment of employees with experience, and that companies that used this
strategy incurred significant losses in terms of yield. And even in the event that rival
companies are able to imitate training programs, they will only be able to achieve
similar performance in the long-term and the advantage obtained can be maintained
for a period of time.
Furthermore, the simultaneous study of different policies normally performance in
a degree of oversimplification in the identification of specific indicators of each of
these policies, using one-dimensional measures that, in our opinion, do not reflect
the full extent of the complexity of the effort made by the company in terms of
training. On the other hand, considering sets of policies would lead us to identify and
measure complementarities that can give rise to specific configurations, which is far
from being resolved in the literature.
In keeping with a universalistic approach, we have posed the starting hypothesis
of our work, trying to isolate the training variable from the effect of other human
resource policies.
H.1. The effort made by companies in terms of training has a positive effect on
the generation of income.
Resource and capacity theory emphasises that competitive advantage must be
sustainable over time. Therefore, the contribution made by training to the
performance of the company should not be temporary if an effort is being made to
justify the role of training as an important factor in the creation of competitive
advantage. In other words, the human capital generated should be long-term.
Some research supports this idea by drawing a link between training and long-
term results, whether on an individual productivity level (Leonard 1990) or in terms
of the company’s financial results (Pfeffer 1994, 1998), market results (Bassi et al.
2002), or both simultaneously (Goval and Welch 2004).
Another interesting approach to the sustainability of income is to use the effect of
training on customer (Delaney and Huselid 1996) and employee (Huselid 1995; or
Kochan and Osterman 1994) satisfaction and loyalty, since it is a form of social
Int Entrep Manag J (2009) 5:139–163 143

capital generated in conditions with a high degree of causal ambiguity and a long-
term intangible assets, inasmuch as it is not transferable or replaceable with another
asset and is difficult to replicate.
Although we relieve that these studies partially measure training efforts, it is our
view that there are sufficient grounds for formulating our second working
hypothesis.
H.2. The effort made by companies in terms of training has a positive impact on
the sustainability of income over time.
Unlike the universalistic approach, the contingent approach takes into account the
influence of certain variables that moderate the relationship between training of
personnel and performance.
In the literature on human resources strategic management, strategy is the main
contingency variable taken into consideration, given that conceptually the
relationship between strategy and human resources policy is one of the most
attractive approaches, giving rise to the so-called external adjustment or alignment
approach (Sastre Castillo and Aguilar Pastor 2000).
Some works argue that there is a positive relationship between the training of
personnel and performance, but that the intensity of said relationship depends on the
strategy implemented by the company (Pfeffer and Baron 1988; Peck 1994; Osterman
1995; D’Arcimocles 1997; Bartel 2000; Hitt et al. 2001; or Kruse et al. 2004).
Some authors, using the competitive strategies of Porter, emphasise that
companies that adopt a differentiation strategy based on quality or innovation invest
more in training (Schuler and Jackson 1987; or Kydd and Oppenheim 1990) or
obtain better performance from the same.
In this aspect, some works have focussed on the effect on employees on an
individual level through job satisfaction and the attitude, commitment and
productivity of employees (Pfeffer and Baron 1988; Arthur 1992; Procopio and
Fairfeld-Sonn 1996; Murray and Raffaele 1997; Black and Lynch 2001; or Bayo
Moriones and Merino Díaz de Cerio 2002). In this research, companies that adopt
differentiation strategies obtain better performance from training, given that it is
strategies that require more involvement from personnel in order that their attitude
and initiative constitute an important source of competitive advantage.
Better performance on an individual level should translate into an increase in
profitability for the company, as we saw in work by Peck (1994). In the work of
Martell et al. (1996) we saw a contingent effect, despite our expectation of a
universalistic relationship with performance, since training had a positive effect on
business units that have a low-cost strategy and a negative effect on those with a
strategy based on differentiation.
On the basis of a similar justification, other authors use Miles and Snow competitive
strategies typology to highlight that extraction companies obtain better performance
from investment in training (Raghuram 1994). Nevertheless, nor do Delery and Doty
(1996) observe the contingent effect of training on ROA or ROE that is apparent in
other human resources policies taken into consideration.
Thus, we set out the hypotheses on the effect of training on income and its
sustainability, but this time as part of a contingent approach:
H.3. The positive relationship that exists between training and income is
moderated by the business strategy adopted by the firm.
144 Int Entrep Manag J (2009) 5:139–163

H.4. The positive relationship that exists between training and sustainability of
income generated over time is moderated by the business strategy adopted.

Approach of the empirical study

In order to explain the hypothesis put forward, the study has focussed on a single
sector so as to avoid problems of underlying heterogeneity when comparing the
effects of a human resources practice on productivity when activities are diverse
(Black and Lynch 2001).
The sector chosen was Spain’s private security services sector. This decision was
based on a number of factors. Firstly, this sector is in services, with its intensive use
of labour, where the client’s perception of quality basically depends on the person
providing the service, since there are not many alternatives for differentiation that do
not involve the service directly provided by the person providing said service.
Secondly, it is a sector with characteristics ideal for isolating the effect of training
from the effects of other human resources practices that can affect performance.
Finally, as a result of regulations governing the sector, training is security is
highly standardised in terms of the types of courses offered to security guards. This
enabled us to achieve a level of accuracy in the measurement of training unusual in
this type of study.
The study covered a five-year period, from 1996 to 2004.
Teniendo en cuenta los objetivos y el enfoque de nuestra investigación, hemos
pretendido obtener una muestra representativa que finalmente está compuesta
por 40 empresas, lo que supone el 44,94% de la población objeto de estudio. En
general no se trata de empresas de gran tamaño, que por otro lado son las más
representativas en la Unión Europea (Ribeiro 2005), pero tampoco deseabamos
tener empresas demasiado pequeñas, y exigimos un mínimo de 50 empleados,
con el fin de incrementar la probabilidad de que las organizaciones
pertenecientes a la muestra desarrollasen políticas de formación mínimamente
formalizadas e integradas en las estrategias empresariales.
Taking into account the objectives and approach of our research, we have
sought to obtain a representative sample made up of 40 companies, or 44.94% of
the population. In general, they are not very large enterprises, that, on the other
hand, they are the most representative in the European Union (Ribeiro 2005), but
we did not desire to work with too small, and we require a minimum of 50
employees, in order to increasing the probability that the organizations belonging
to the sample, developed formal (and integrated in the business strategies) politics
of training.
The indicators used (geographical location, size, age in years, number of
employees) appear to indicate that the sample used is representative of the
population the object of research, accounting for 88.74% of the sector’s turnover,
89.15% of its assets and 88.54% of its workforce.
For each variable, we have nine values per company in the sample (corresponding
to the years of the study). Thus, we will have 360 observations per variable.
Int Entrep Manag J (2009) 5:139–163 145

To obtain the required information and increase the validity of the research, we
used a triangulation process that combined information from additional sources, such
as public bodies, internal documents from the companies involved, institutional
records (FORCEM), sector associations and specialist publications and magazines.

Measures

Our study takes into account four types of variable: independent variables, moderator
variables (for contingent hypotheses), dependent variables and control variables.
Moderator variables are considered control variables for universalistic hypotheses.

Independent variable

We have one independent variable: the training of personnel. We should point out
that given the importance of this variable in our study, we have made a greater effort
in how this variable is measured.
Most works use a single criterion to measure training. For the most part, time
invested in training per employee is used (Walton 1985; Schuler and Jackson 1987;
Pfeffer and Baron 1988; Arthur 1992, 1994; Kochan and Osterman 1994; Peck
1994; Holzer 1995; Delery and Doty 1996; Martell et al. 1996; Huselid et al. 1997;
Ichniowski et al. 1997; Murray and Raffaele 1997; Aragón et al. 2003; McGahan
and Porter 2003; Goval and Welch 2004; or Kruse et al. 2004).
Another common indicator tries to measure the scope of training through the
percentage of employees that receive training (Pfeffer 1994; Raghuram 1994;
McDuffie 1995; Delaney and Huselid 1996; Bukowitz and Petrash 1997; Aragón et
al. 2003; Black and Lynch 2001; Bayo Moriones and Merino Díaz de Cerio 2002;
Ordiz Fuertes 2002; Kruse et al. 2004; or Tzafrir 2005).
Another criterion used is the type of courses offered (Schuler and Jackson 1987;
Raghuram 1994; Delery and Doty 1996; Koch and McGrath 1996; Huselid et al.
1997; Kidder and Rouiller 1997; McEvoy 1997; or Goval and Welch 2004).
Some distance behind in terms of training measures are indicators such as the cost
of training courses (Delaney and Huselid 1996), expenditure on training as a
percentage of salary costs (D’Arcimocles 1997) and the duration of training
programmes (Barrett and O’Connell 2001; or Aragón et al. 2003).
We believe that in order to measure investment in training, a multidimensional
measure should be used. Some works, such as that of Delaney and Huselid (1996),
create a training index from three items. Nevertheless, one or more of these is
subjective (the perception of the effectiveness of training); furthermore, the index is
obtained from the median of the values obtained in the items used.
Using the methodology proposed by Danvila del Valle and Sastre Castillo (2005),
for our study we have chosen three different dimensions to determine this variable:
different training courses given, the number of hours of attendance per year per
employee at training courses given and the annual investment in training by the
company per employee.
We believe that these dimensions represent the training courses given (variety of
training courses given), the frequency with which a company offers courses (the
146 Int Entrep Manag J (2009) 5:139–163

Table 1 Results of the analysis of main components

Indicators Component 1

1. Number of different courses offered 0.897


2. Hours per employee/year 0.835
3. Investment per employee/year 0.692
KMO measurement 0.635
Significance, Bartlett test 0.000
Percentage of variance explained 73.317%

Source: Original table

number of hours of attendance per year per employee at training courses given) and the
cost for the company of this providing this training (annual investment per employee).
Using the data on the three indicators, we have conducted a factorial analysis to
make the variable operational when it comes to explaining the hypotheses. The
method use to conduct the factorial analysis is the principal components analysis
(PCA). The performance obtained (Table 1) are satisfactory, since a single factor
explains more than 70% of the variance in investment in training.

Dependent variables

The dependent variables used in our empirical work are as follows:

Income This is the dependent variable of hypotheses one and three in our model.
Some works, such as that by Tzafrir (2005), use subjective variability indicators of
performance in different aspects.
Others use objective economic indicators, either measurement or accounting
indicators (Huselid 1995; Huselid and Becker 1996; or Huselid et al. 1997). Rogers
and Wright (1998) point out that universalistic yield measurements have focussed
almost exclusively on financial and accounting indices.
Nevertheless, Martín-Alcázar et al. (2005) proposal to combine a broader set of
indicators with different origins and relative weightings depending on the particular
characteristics of the organisation seems difficult to implement in a quantitative study.
In view of this, we have chosen to use an objective indicator based on accounting
information, in order to avoid errors in method, cited as common by Becker and
Gerhart (1996), that are made when human resources policy and performance are
valued in a subjective manner by the same person in the company’s management.
Of the accounting measures used, EBITDA is probably the most widely used in
works in this area (Balkin and Gómez-Mejía 1990; Terpstra and Rozell 1993;
Kochan and Osterman 1994; Huselid 1995; Wood and Albanese 1995; Delery and
Doty 1996; Huselid and Becker 1996; Rajagopalan 1996; White and Miles 1996;
Youndt et al. 1996; Bacidore et al. 1997; Bukowitz and Petrash 1997; Huselid,
Jackson and Schuler 1997; Lee and Miller 1999; Purcell 1999; Hunter 2000; Richard
and Johnson 2001 and Saá Pérez and García Falcón 2002).
Specifically, the measure used in our study was EBITDA/number of employees2;
therefore, it is expressed in terms of euros per employee.
2
EBITDA: Earnings Before Interests, Tax, Depreciation and Amortization.
Int Entrep Manag J (2009) 5:139–163 147

This measure is not an absolute figure, but rather a ratio that divides an
accounting measure into the size of the company during the same period, to prevent
the size of the company from being able to distort what we are trying to verify. The
size of a company can influence performance; this has been confirmed in a number
of studies. In this way, we can compare companies of different sizes.

Sustainability of income This is the dependent variable of hypotheses two and four in
our model. The measure that represents this variable is customer loyalty. Values of this
measure are expressed in terms of months.
Numerous empirical studies (Arthur 1994; Huselid and Becker 1996; Bukowitz and
Petrash 1997; Huselid et al. 1997 and Richard and Johnson 2001) have linked the
training of personnel and performance using social indicators, aiming to find out
whether a greater effort in terms of training results in a long-term competitive
advantage over time, enabling us to produce better performance. The main measures
are: customer loyalty to a company, the work environment, and staff turnover. More
specifically, some authors approximate the sustainability of income, measuring the
effect on customer (Delaney and Huselid 1996) and employee satisfaction and loyalty
(Huselid 1995; or Kochan and Osterman 1994).
Using this measure, we can determine whether the training given to personnel
retains and strengthens loyalty among customers, creating a sustainable competitive
advantage over time that it will result in greater revenues. In this way, a high level of
customer loyalty will reveal a company that meets the expectations of its clients. A
low level of customer loyalty, on the other hand, would indicate a company that
initially offers attractive services but which, with time, fails to meet its obligations or
is unable to address the new needs of its clients.

Moderator variables

In our view, competitive strategies implemented by organisations are moderator


variables. In the analysis of the first two hypotheses, these variables are considered
control variables. In our empirical work, the moderator variables used are as follows:

Product differentiation strategy This type of strategy will use the possession of quality
certificates as a measure. Thus, we have a dichotomous variable (YES/NO) that will divide
companies into two groups, depending on whether or not they possess a quality certificate.
Many works have measured the quality of the service provided using the client’s
perception of the quality of service: questionnaires, percentage of complaints and claims,
surveys on the level of satisfaction, expectation compliance indices, etc. Nevertheless, in
recent years these criteria have been criticised on a number of occasions due to the high
degree of subjectivity inherent in these measures and the ease with which companies can
manipulate data obtained from the same.
In view of this, we have sought objective variables that are contrastable and not
subject to manipulation; certificates issued by external bodies and the introduction of
quality assurance systems meet these criteria. A number of authors have conducted
their research using these measures, among others: Terpstra and Rozell (1993);
Lloréns (1996); Kidder and Rouiller (1997); Kroll et al. (1999) and Bayo Moriones
and Merino Díaz de Cerio (2002).
148 Int Entrep Manag J (2009) 5:139–163

Cost leadership strategy This strategy will use the price of services provided as a
measure. The average hourly cost of security services, both armed and unarmed, has
been used as an indicator. It is a continuous variable, expressed in euros.
Prices have also been used in a number of empirical studies: Huselid and Becker
(1996); Koch and Mc Grath (1996); Lloréns (1996) and Bassi et al. (2002).
Nevertheless, the price level normally used is valued subjectively by the executives to
whom the survey is addressed, due to the impossibility of knowing the prices of
products of companies in the sector, especially in the case of multi-product companies.
It is our understanding that the average hourly price used for security services,
both armed and unarmed, is objective, free of measurement error and well-suited to
the purposes of our study.

Control variables

The last type of variable is the control variable. The objective of these variables is to
find out about different aspects that can influence the dependent variables. By
including them in the model, we aim to find out with greater precision the effect of
the independent variable on the dependent variables.
The control variables initially used in our empirical work are: whether or not the
company is part of a multinational group, the size and age of the company, its market
share and the year to which the data used relate. After conducting an exploratory
analysis of the variables, we decided not to use the following variables: whether or
not the company is part of a multinational group and market share.
In the case of the variable whether or not the company is part of a multinational group,
the sources of information used in our research only tell us whether a company has offices
overseas or only in Spain. Thus, we have a dichotomous variable (YES/NO) that will
divide companies into two groups, depending on whether or not they have offices
overseas. In other words, a multinational is considered to be any company with an office
outside Spain, regardless of the number of offices overseas or how long they have been in
operation. This method of measurement leads to companies with characteristics similar to
local companies but which have one or more offices overseas being classified as
multinationals. For this reason, this variable distorted performance and, although this
variable was originally used, it was later decided to remove it from the analysis.
The market share variable, meanwhile, was seen to be highly multicollinear to the
size variable. For this reason, it was decided to leave the size variable on its own and
remove this variable from our analysis.

Size We have used the number of employees as the unit of measurement for this
variable. According to the main private security sector associations, size is a determining
factor for the smallest companies in the sector, for whom it is very expensive to invest in
training for their employees. For this reason, we have used companies with more than 50
employees in our empirical study, since companies with fewer than 50 employees tend
to limit their training to that which they are obliged to provide under the law.

Age of the company This variable tells us the number of years each company in the
sector has been in operation, i.e. the age of each company in years. It should be
remembered that companies in our sample must have been operating in the sector for
Int Entrep Manag J (2009) 5:139–163 149

at least 10 years. On the other hand, private security companies were authorised by
law in Spain in 1974. This would suggest to us that all of the companies in our study
were created within a 20-year period.

Year of data used This variable selects the data used according to the year to which
they relate. We will use this control variable exclusively in hypotheses 1 and 3 of the
model. The reason we have removed this variable from hypotheses 2 and 4 is the
existence of a clear relationship between the loyalty variable (measured in months)
and the year of the data used. This is due to the fact that for each year that passes, all
companies that retain their clients automatically increase loyalty by 12 months.
Therefore, it does not make sense to include this variable in hypotheses 2 and 4.

Analysis of the results

In light of the objectives of our empirical study, we have conducted the calculation
of the main descriptive statistics of each variable in the model: average, median,
standard deviation, minimum and maximum; as well as the bivariate correlations of
variables in the model (Tables 2 and 3).
In view of the correlations found, a multicollinearity analysis was conducted
between variables observing the condition index (CI) and the variance inflation
factor (VIF). The VIF was low in all cases; nevertheless, we obtained a high CI,
which would indicate the presence of multicollinearity. The problem this could
create is that one or more of the essential variables of the model may not appear as
significant, even if they are, concealing its relationship due to the effect of the
existence of multicollinearity. However, this did not occur in our case, as in

Table 2 Main descriptive statistics

Variables Mean Median Standard deviation Minimum Maximum

1. Training 0.0433 −0.1525 0.9778 −1.4089 2.2223


2. Price 19.5479 19.46 1.43 10.13 23.01
3. Quality 0.3333 0 0.4714 0 1
4. Training×price 1.8375 −3.064 19.944 −27.4031 50.9764
5. Training×quality 0.3348 0 0.71197 0 2.2223
6. Size 2,303.21 2,255 1,210 57 5,180
7. Age 17.975 18.00 5 5.00 29
8. 1997 0.20 0 0.40 0 1
9. 1998 0.20 0 0.40 0 1
10. 1999 0.20 0 0.40 0 1
11. 2000 0.20 0 0.40 0 1
12. 2001 0.20 0 0.40 0 1
13. 2002 0.20 0 0.40 0 1
14. 2003 0.20 0 0.40 0 1
14. 2004 0.20 0 0.40 0 1
15. EBITDA/no. of employees 27.7877 2 65.263 0.0351 334.931
16. Customer loyalty 15.783 15.80 5.71 6.00 30.80

Source: Original table


150

Table 3 Bivariate correlations of variables in the model

Correlations

1. Training 1
2. Price 0.708** 1
3. Quality 0.695** 0.539** 1
4. Training×price 0.998** 0.701** 0.695** 1
5. Training×quality 0.868** 0.593** 0.665** 0.886** 1
6. Size 0.244** 0.136** 0.156** 0.257** 0.321** 1
7. Age 0.094 0.189** 0.282** 0.104* 0.156** −0.119* 1
8. Year 1997 0.000 −0.008 0.000 0.001 0.000 −0.013 −0.103 1
9. Year 1998 0.000 −0.024 0.000 0.002 0.000 −0.017 0.000 −0.250** 1
10. Year 1999 0.000 −0.029 0.000 0.000 0.000 0.009 0.103 −0.250** −0.250** 1
11. Year 2000 0.000 0.020 0.000 −0.001 0.000 0.038 0.206** −0.250** −0.250** −0.250** 1
12. Year 2001 0.000 0.018 0.004 −0.003 0.003 0.051 0.221** −0.250** −0.253** −0.258** −0.252** 1
13. Year 2002 0.000 0.025 0.008 −0.007 0.006 0.062 0.229** −0.250** −0.261** −0.263** −0.265** −0.272** 1
14. Year 2003 0.000 0.021 −0.002 −0.008 −0.002 0.069 0.234** −0.250** −0.267** −0.271** −0.265** −0.274** −0.279** 1
15. Year 2004 0.000 −0.009 0.003 −0.009 −0.004 0.079 0.240** −0.250** −0.271** −0.284** −0.285** −0.288** −0.292** −0.289** 1
16. EBITDA/N of 0.535** 0.351** 0.390** 0.537** 0.511** 0.158* 0.247** 0.191** 0.174* 0.112 0.005 0.008 0.115 0.129 0.008 1
employees
17. Customer loyalty 0.690** 0.476** 0.556** 0.683** 0.579** 0.089 0.188** 0.172 0.061 0.052 0.186** 0.375** 0.395** 0.397** 0.397** 0.399**

Source: Original table


**p<0.01; *p<0.05; N=360
Int Entrep Manag J (2009) 5:139–163
Int Entrep Manag J (2009) 5:139–163 151

universalistic hypotheses training always appeared to be highly significant, with a p-


value of 0.000. in contingent effect hypotheses, the interaction of training with the
moderator variable also appeared to be significant. Thus, the presence of multi-
collinearity in our case did not pose a serious problem.
We also tested heteroscedasticity using the White method, obtaining standard
errors and robust covariance to heteroscedasticity. The results obtained demonstrate
that we obtain new significant variables for no hypotheses; rather, said variables
remain the same, confirming the validity of results obtained.
This existence of high correlations between some variables made it advisable to
contrast hypotheses via a hierarchical regression analysis (Cohen and Cohen 1983
and Sánchez Marín and Aragón Sánchez 2002).
Before analysing the results, we should remember that often the efforts made by a
company in terms of training do not have immediate results; a period of time may be
required in order for the benefits of training to be seen in performance, for personnel
to be able to put knowledge acquired through training into practice (Wright and
McMahan 1992; Broadbent 1993; Bartel 1994, 2000; White and Miles 1996;
D’Arcimocles 1997; Murray and Raffaele 1997; Black and Lynch 2001 and Aragón
et al. 2003).
In order to mitigate this possible limitation in our empirical study, we have
replicated the study introducing a time delay of one year, thus also taking into
account the causal relationship between the one-year training variable and the other
variables of the model of the following year. In the latter, we have used the values of
the training variable for the first four years of our study and the values of the other
variables for the last 4 years, resulting in a total of 320 observations, rather than the
original 360 observations. Given that we have not found significant differences in
the results obtained, in this work we only reflected the study with variables from the
same year, so as to prevent the unnecessary loss of observations, since for each year
of delay an important number of observations would be lost.
H.1. Efforts made by companies in training have a positive impact on income.
To explain this hypothesis, a hierarchical regression analysis has been conducted
(Table 4). The results obtained indicate that training has a highly significant positive
effect on variable income. There is a moderate increase in the explanatory power of
the model when the remaining variables are introduced.
Neither the quality variable neither the companies’ price level are significant in
any case. Nevertheless, the age variable has a positive influence on the variable
income. In other words, older companies obtain more income, due to the experience
effect of companies that arrived in the market first.
H.2. The effort made by companies in terms of training has a positive influence
on the sustainability of income obtained over time.
The regression analysis conducted (Table 5) tells us that the training and quality
variables have a highly significant and positive effect on the sustainability of
income.
The training given to the security guards that provide the service is not only
important for the generation of income, but also for the sustainability of this
income, since we can see that it has a direct and unequivocal impact on customer
loyalty.
152 Int Entrep Manag J (2009) 5:139–163

Table 4 Hierarchical regression analysis for income variable

Variables Step 1 Step 2 Step 3

Training 0.535*** 0.550*** 0.504***


Price −0.061 −0.032
Quality 0.041 0.079
Size 0.001
Age 0.154***
1997 −0.019
1998 −0.034
1999 −0.063
2000 −0.090
2001 −0.087
2002 0.023
2003 0.046
2004 0.067
R2 0.286 0.289 0.437
Adjusted R2 0.284 0.285 0.428
F-value 143.725*** 48.221*** 45.756***
Durbin–Watson 0.672 0.671 0.689

Source: Original
***p<0.01. N=360

Quality certificates are also an element that differentiates the service provided,
and have a positive effect on customer loyalty. Unlike price, which is insignificant in
our model although the coefficient is negative like was expected, this criterion is
important in ensuring customer loyalty.
Furthermore, size has a negative effect on the dependent variable. In small
companies, accessibility and face-to-face contact with the client is more fluid and the
relationship between client and supplier is essential in companies that provide
intangible services, as is the case of security; as a result, small companies are better
able to ensure the loyalty of their customers. es razonable pensar que las empresas
más asentadas en el sector han conseguido obtener una cartera de clientes
fidelizados. Likewise, age presents a significant effect, in this case positive, on the

Table 5 Hierarchical regression analysis for sustainability of income variable

Variables Step 1 Step 2 Step 3

Training 0.690*** 0.612*** 0.657***


Price −0.039 −0.062
Quality 0.152*** 0.111**
Size −0.039**
Age 0.100**
R2 0.476 0.488 0.499
Adjusted R2 0.475 0.484 0.492
F-value 325.240*** 113.186*** 70.626***
Durbin–Watson 0.604 0.627 0.619

Source: Original
***p<0.01; **p<0.05. N=360
Int Entrep Manag J (2009) 5:139–163 153

loyalty, is reasonable to think that the oldest firms of the sector have managed to
obtain greater loyalty of their clients.
H.3. The positive relationship that exists between training and income is curbed
by the business strategy adopted.
This third hypothesis seeks to explain whether the strategy adopted by companies
shapes the relationship between training and income. Thus, in this hypothesis we
contend that strategy is a moderator variable. In our empirical study, we will analyse
product differentiation strategy and cost leadership strategy. As with the first
hypothesis, we will use training as the independent variable and income as the
dependent variable.

Product differentiation strategy We will seek to explain whether companies that adopt
a quality differentiation strategy seek a competitive advantage that generates greater
income through a greater focus on training.
We begin the study by conducting an exploratory analysis. First, we divide the
companies in the sample into two groups, as a function of the dichotomous variable of
quality orientation; in other words, according to whether the company has or does not
have a quality certificate, to later determine via a variance analysis whether there are
significant differences with regards to efforts made in terms of training between the
companies in both groups, given that companies with certification invest more in training.
We then conducted a regression analysis (Table 6). The results of this analysis indicate
that the training variable and the interaction of training by product differentiation
strategy (training×quality) have a highly significant positive effect on variable income,
both in isolation (steps one and two) and as a whole (steps three and four).
By simultaneously considering efforts made in terms of training and the interaction
of training and quality, the model moderately increases its explanatory power.

Table 6 Hierarchical regression analysis for income variable

Variables Step 1 Step 2 Step 3 Step 4

Training 0.535*** 0.371*** 0.500***


Training×quality 0.511*** 0.189*** 0.042***
Size −0.005
Age 0.160***
1997 −0.016
1998 −0.027
1999 −0.053
2000 −0.075
2001 −0.081
2002 −0.087
2003 −0.092
2004 −0.097
R2 0.286 0.261 0.295 0.435
Adjusted R2 0.284 0.259 0.291 0.427
F-value 143.725*** 126.675*** 74.780*** 54.406***
Durbin–Watson 0.672 0.668 0.682 0.690

Source: Original
***p<0.01. N=360
154 Int Entrep Manag J (2009) 5:139–163

On comparing the result obtained with the model of hypothesis one (universalistic
approach), we see that R2 is practically the same one. The moderating effect of the
quality orientation only explains a small proportion of the result obtained from the
universalistic effect of training, such that the companies most focussed on quality
obtain better performance from investment in training that are seen in larger
increases in profits than those from investment in training by non-certified
companies.
Furthermore, the age variable also has a positive influence on income, due to the
experience effect of more established companies. This relationship is similar to that
found in hypothesis one.

Cost leadership strategy As in the preceding strategy, the starting point was an
exploratory analysis in which companies in the sample were divided into two
groups, depending on whether they were above or below the value of the median. By
grouping the companies in this manner, we sought to determine whether there are
significant differences in investment in training between companies in one group and
those in the other. For this reason, we conducted a variance analysis in which
training was used as the dependent variable and the factor that divides the companies
into two groups as a function of price was used as the independent variable.
The results obtained show that the average values of the price variable differ
significantly between the two groups in question. It was also shown that companies
that have higher prices offer more training.
Later, we conducted a regression analysis (Table 7). The results indicated that
taken separately, the training variable (step 1) and the interaction of training with
cost leadership strategy (step 2) have a highly significant positive effect on the
dependent variable. Nevertheless, if we analyse the combined effect of both
variables (steps 3 and 4), we can see that the interaction of training with cost

Table 7 Hierarchical regression analysis for income variable

Variables Step 1 Step 2 Step 3 Step 4

Training 0.535*** −0.169*** −0.151***


Training×price 0.537*** 0.706*** 0.386***
Size −0.005
Age 0.160***
1997 −0.007
1998 −0.011
1999 −0.024
2000 −0.032
2001 −0.046
2002 −0.068
2003 −0.065
2004 −0.075
R2 0.286 0.289 0.289 0.435
Adjusted R2 0.284 0.287 0.285 0.427
F-value 143.725*** 145.229*** 72.458*** 54.461***
Durbin–Watson 0.672 0.681 0.683 0.694

Source: Original
***p<0.01. N=360
Int Entrep Manag J (2009) 5:139–163 155

leadership strategy continues to have a positive influence on the dependent variable,


while training now has a negative effect.
This is an interesting result. In hypothesis one (put forward as part of a
universalistic approach), we saw that training has a positive effect on income, while
prices were insignificant.
Now we see that the model does not improve its explanatory power, but in
qualitative terms, it demonstrates that no type of training has the same impact on
variable income. Training in itself represents a cost for the company, but when it
results in an improvement in the way the service offered is perceived and the
customer is prepared to pay more for said service, the result obtained by the
company improves significantly.
Thus, the effect of training on results not only depends on training expenditure,
but also on the results the company is able to obtain from training itself. For this
reason, a multidimensional measure was used to measure training in our study, a
measure that goes beyond expenditure.
Just as in the product differentiation strategy, the age variable has a positive effect
on income due to the experience effect.
H.4. The positive relationship between training and the sustainability of income
obtained over time is tempered by the corporate strategy adopted.
This fourth hypothesis seeks to explain whether the effect of training on the
sustainability of income is influenced by the business strategy adopted by
companies. Thus, in this hypothesis, we also consider the strategy a moderator
variable.
On the basis of identical reasoning to that set out in the previous hypothesis, we
will analyse product differentiation strategy and cost leadership strategy. As in the
second hypothesis, we will use training as the independent variable training and the
sustainability of income as the dependent variable.

Product differentiation strategy As in the previous hypothesis, we conducted a


hierarchical regression analysis (Table 8) to verify this hypothesis. The results
obtained indicate that when taken separately, the training variable (step 1) and the
interaction of training with product differentiation strategy (step 2) have a highly
significant positive effect on the dependent variable. Nevertheless, if we analyse the

Table 8 Hierarchical regression analysis for sustainability of income variable

Variables Step 1 Step 2 Step 3 Step 4

Training 0.690*** 0.559*** 0.759*** 0.778***


Training×quality −0.080* −0.109*
Size −0.021
Age 0.128***
R2 0.476 0.336 0.478 0.495
Adjusted R2 0.475 0.334 0.475 0.489
F-value 325.240*** 180.991*** 163.186*** 86.876***
Durbin–Watson 0.604 0.557 0.604 0.600

Source: Original
***p<0.01; *p<0.1. N=360
156 Int Entrep Manag J (2009) 5:139–163

combined effect of both variables (steps 3 and 4), we can see that training continues
to have a positive influence on the dependent variable, while the interaction of
training with product differentiation strategy now has a negative effect, albeit less
significant.
Comparing the results obtained in hypothesis two, put forward by the
universalistic perspective, with the results obtained on this occasion, we can see
that there is no improvement in the explanatory power of the model, since R2 is
virtually identical. What is apparent is an interesting effect explained by the
universalistic effect of training on customer loyalty.
It is the case for all companies that customer loyalty depends to a large extent on
employee training, although not to the same degree across all companies.
For companies with quality certificates, training is not such an important factor in
ensuring customer royalty, since their customers also perceive and value other
aspects of the company’s approach to quality that encourage them to remain with
that supplier of services.
On the other hand, companies that are less focussed on quality have in training
their only strong point when it comes to ensuring customer loyalty; as a result, the
effect of training on customer loyalty is even greater.
Furthermore, the age variable also has a positive influence on the sustainability of
income, due to the experience effect of more established companies.

Cost leadership strategy The regression analysis conducted (Table 9) demonstrates


that when taken separately, both the training variable (step 1) and the interaction of
training with the cost leadership strategy (step 2) have a highly significant positive
effect on the dependent variable. Nevertheless, if we analyse the overall effect of
these two variables (steps 3 and 4), we can see that training continues to have a
positive influence on the dependent variable, while the interaction of training with
the cost leadership strategy now has a negative effect, albeit only slightly significant.
The new model leads to an increase in R2 compared to the universalistic approach
of just 0.002. Nevertheless, it allows us to explain the universalistic effect of the
training variable.
If the company goes to the effort of training its personnel (which has a positive
effect on the service provided), is able to offer attractive prices, makes a serious
effort to be competitive and sustains a reduction in its margins in the short-term (as

Table 9 Hierarchical regression analysis for sustainability of income variable

Variables Step 1 Step 2 Step 3 Step 4

Training 0.690*** 1.899*** 2.172***


Training×price 0.683*** −1.212** −1.496*
Size −0.011
Age 0.137***
R2 0.476 0.466 0.483 0.501
Adjusted R2 0.475 0.465 0.480 0.496
F-value 325.240*** 312.827*** 166.553*** 89.229***
Durbin–Watson 0.604 0.593 0.628 0.631

Source: Original
***p<0.01; **p<0.05; *p<0.1. N=360
Int Entrep Manag J (2009) 5:139–163 157

we saw in the results of the third hypothesis), but is earning the loyalty of its
customers. Nevertheless, the effect is modest.
Finally, age has a positive influence on the dependent variable due to the
experience effect.

Conclusions

The results obtained in the research have interesting implications both on a practical
level and for academic discussion on the universalistic or contingent nature of the
effect of training on performance.
In the management sphere, it confirms the idea that training, despite being a direct
cost for the company, generally has a positive impact on performance. Studies that
reinforce this finding should create an awareness among top executives that training
should not be considered a cost, but rather an investment that can produce a positive
return.
This finding is consistent with resource and capacity theory, which defends the
view that training is the main way of creating human capital: a valuable, scarce
resource that is hard to imitate (due to the existence of causal ambiguity) and hard to
replace. Furthermore, it tends to have positive effects on the extent of commitment
felt by the employee to the organisation, averting the risk of transferability, the only
factor that can qualify the positive effect of training on performance.
With regards to the hypotheses of our research, the first of these is put forward as part
of a universalistic approach. The contrasting of these hypotheses shows that companies
that make a greater effort in terms of training obtain greater economic returns.
If the increase in training is seen by the customer as a source of value-added, they
will be prepared to pay more for the service. Thus, if the cost of the increase in
training feeds through to prices, company income will grow.
In the classic “make” vs. “buy” debate, our results are in line with those of Hatch
and Dyer (2004), which confirm that the development of human capital through
training has an effect that cannot be compared with the acquisition of human capital
via the recruitment of personnel with experience, and that companies that adopt this
strategy lose significant ground in terms of yield.
Some research has failed to appreciate the positive effect of training on
performance. In the case of Úbeda García (2005), the author attributes this effect
to his way of measuring productivity, which is his indicator of performance. In
research by Delery and Doty (1996), we believe that the problem lies with the
excessively simplistic method used to measure training.
Nevertheless, results are conclusive when a greater effort is made to measure
efforts made in terms of training, taking different dimensions of training into
account, as in our work or in work done by Aragón et al. (2003).
Hypothesis two, also put forward as part of a universalistic approach, comple-
ments the contributions made by the previous hypothesis; the contrast of this
hypothesis shows that income obtained is not merely temporary but rather
sustainable over time, inasmuch as we have found significant differences in the
greater loyalty of customers towards companies that are more concerned with their
personnel and as a result make a greater effort in terms of training.
158 Int Entrep Manag J (2009) 5:139–163

This has been the focus of lesser study in the literature, which has primarily
focussed in measuring the impact of training on individual or performance
indicators. In this case, however, it has focussed on short-term yields. Only Delaney
and Huselid (1996) incorporate customer loyalty into his perceived business yield
indicator, although together with six other indicators.
Moreover, we have found that smaller companies have greater customer loyalty,
due to the fact that in smaller companies accessibility and direct contact with
customers is a much more fluid affair. We also found that the relationship between
customer and supplier is essential for companies that provide intangible services, as
in the case of security services companies.
Acceptance of these two first hypotheses supposes that training results in the
creation of human capital as a source of sustainable competitive advantage and
allows a company to generate higher income in the short- and long-term.
Having verified the universalistic effect of training on performance, we thought
we should strive for a deeper understanding of the nature of such a relationship,
taking into account possible moderating effects that qualify the interpretation of
results. This would involve a significant contribution from research to the current
body of knowledge on the subject.
In order to achieve this objective, we have analysed whether greater alignment or
adaptation between training policy and the strategy adopted by the company leads to
better performance. In other words, we aim to find out whether the influence of
training on performance increases or decreases depending on the type of strategy, or
even whether the application of a particular strategy could prove counter-productive.
Thus, hypotheses three and four are put forward as part of a contingent approach.
We conducted both variance analyses, which showed that companies that have a
quality certificate and which offer services at higher prices make a greater effort in
terms of training, a position already defended by Schuler and Jackson (1987) and by
Kydd and Oppenheim (1990). Research by Kidder and Rouiller (1997), on the other
hand, found that more training helps in the better introduction of quality systems.
Nevertheless, there is little research that demonstrates that this relationship
between strategy and effort in terms of training affects the variability of results.
Thus, when verifying hypothesis three, on comparing the result obtained with the
result obtained using the model of hypothesis one (universalistic approach), we do
not observe increment of the R2 upon considering the moderating effect of the
quality neither of the prices.
In the first of these cases, both training and the interaction of quality with training
have a significant effect in the final model, whether separately or taken together. The
contingent effect only serves to confirm that, even though training is positive in all
companies, the impact of training on performance is slightly greater in those
companies that have a quality certificate. This result is in line with expectations and
contradicts the findings of Martell et al. (1996), who found that the training had a
negative impact in business units that had adopted a strategy based on
differentiation.
The result obtained is also found to contradict Martell, Gupta and Carroll’s
findings when the moderating effect of prices is analysed. These authors recorded a
positive effect for training in companies with low costs. In our research, the
explanatory power of the model does not does not improve that of the universalistic
Int Entrep Manag J (2009) 5:139–163 159

effect. Nevertheless, we noted an interesting nuance in the results obtained: training


now appears with a negative regression coefficient, while the interaction of training
with prices is positive and equally significant with a p-value of less than 0.01. This
would imply that training in itself represents a cost for the company, but it is higher
prices resulting from a positive assessment on the part of clients of the service
offered (through more training) that allow a company to increase its income.
Hypothesis four adopts an approach similar to that of the hypothesis above. A
verification of this hypothesis shows that the effect of training on the sustainability
of income is moderated by the business strategy adopted.
If we consider the interaction of training with the focus on quality, we see that
there is no improvement in the explanatory power of the model. However, it does
reveal that customer loyalty, although it depends to a large extent on the training of
personnel, is not such an important factors in companies that have a quality
certificate, since customer satisfaction with the service received is not an exclusive
function of the training given to the employee providing said service.
Nevertheless, the absence of other factors that can influence the perceived quality
of the service means that in companies that are less focussed on quality, training is
the key element in achieving some kind of differentiation. As a result, training
multiplies the intensity of the effect on customer loyalty.
Something similar can be seen in the results obtained from the interaction of
training and price. The new model barely increases R2 compared to the universalistic
approach; however, it allows us to verify the effect of training.
The negative coefficient of the interaction of training with price leads us to the
conclusion that companies that, in addition to allocating significant resources to
training their personnel, are able to continue to charge reasonable prices for their
services, even if this means reducing their income in the short-term (as pointed out in
the results of hypothesis 3), improve their position in terms of the perceived quality-
price relationship as perceived by the customer. This has a positive impact on
customer loyalty.
Acceptance of hypothesis four led us to accept the validity of the analysis model
put forward. Nevertheless, there are a number of limitations of which we are aware.
One of these limitations arises from the nature of what is being studied. The
training of personnel, the development of skills, motivation and the ability to work
as part of a team are intangibles that are difficult to measure and difficult to assess.
For this reason, the results obtained only show a relationship between efforts made
in terms of training and the performance of these companies. However, we do not
know exactly the influence of training on performance, since it is impossible to
determine the extent to which what is learned is applied to the service provided.
Moreover, it is known that the degree to which employees take advantage of training
can vary greatly, as this will depend on interest, motivation, training and other
factors among personnel that receive said training that are difficult to measure.
On the other hand, given that research has focussed on a specific subsector, it
would appear prudent to limit the scope of the results obtained and the conclusions
reached to the population the object of research, i.e. the private security subsector in
Spain.
Although we consider the sample chosen to be valid and representative, by
limiting inclusion in said sample to companies with a workforce of at least 50 we
160 Int Entrep Manag J (2009) 5:139–163

have ruled out all small companies in the subsector, which often limit themselves to
providing only training required by law. Thus, our study does not reflect the effects
of training on these smaller companies.
Another limitation stems from the fact that our research has measured efforts
made in terms of training, and considered the diversity of courses offered, but
without differentiating between generic and specific training given. As a result, we
cannot verify whether, as found in research by Barrett and O’Connell (2001), generic
training is more productive than specific training.
Finally, we should consider the possibility of an inverse causal effect, i.e. that
companies that have made a greater effort in terms of training have made this effort
due to the fact that they have been allowed to by their results. This is a common
limitation is studies with a universalistic approach, since the application of these
practices is a direct outcome of good results. Measuring this possible effect could be
an interesting line of research in the future.

References

Amit, R., & Schoemaker, P. J. H. (1993). Strategic assets and organizational rent. Strategic Management
Journal, 14(1), 33–46.
Aragón, A., Barba, M. I., & Sanz, R. (2003). Effects of training on business results. The International
Journal of Human Resource Management, 14(6), 956–980.
Arthur, J. B. (1992). The link between business strategy and industrial relations systems in American steel
minimills. Industrial and Labor Relations Review, 45(3), 488–506.
Arthur, J. B. (1994). Effects of human resource systems on manufacturing performance and turnover.
Academy of Management Journal, 37(3), 670–687.
Bacidore, J. M., Boquist, J. A., Milbourn, T. T., & Thakor, A. V. (1997). The search for the best financial
performance measure. Financial Analysis Journal, 54(1), 11–20.
Balkin, D., & Gómez-Mejía, L. R. (1990). Matching compensation and organizational strategies. Strategic
Management Journal, 11(2), 153–169.
Barney, J. (1991). Firm resources and sustained competitive advantage. Journal of Management, 17(1),
99–120.
Barrett, A., & O’Connell, P. J. (2001). Does training generally work? The returns to in-company training.
Industrial and Labor Relations Review, 54(3), 647–662.
Bartel, A. (1994). Productivity gains from the implementation of employee training programs. Industrial
Relations, 33(4), 411–425.
Bartel, A. (2000). Measuring the employer’s return on investments in training: Evidence from the
literature. Industrial Relations, 39(3), 502–524.
Bassi, L. J., Ludwig, J., Mc Murrer, D. P., & Van Buren, M. (2002). Profiting from learning: Firm-level
effects of training investments and market implications. Singapore Management Review, 24(3), 61–
76.
Bayo Moriones, A., & Merino Díaz de Cerio, J. (2002). Las prácticas de recursos humanos de alto
compromiso: Un estudio de los factores que influyen sobre su adopción en la industria española.
Cuadernos de Economía y Dirección de la Empresa, 12, 227–246.
Becker, G. S., & Gerhart, B. (1996). The impact of human resource management on organizational
performance: Progress and prospects. Academy of Management Journal, 39(4), 779–801.
Black, J. A., & Boal, K. B. (1994). Strategic resources: Traits, configurations and paths to sustainable
competitive advantage. Strategic Management Journal, 15(2), 131–148.
Black, S. E., & Lynch, L. M. (2001). How to compete: The impact of workplace practices and information
technology on productivity. The Review of Economics and Statistics, 83(3), 434–445.
Broadbent, S. (1993). Advertising effects: More than a short term. Journal of the Market Research Society,
35(1), 37–49.
Bukowitz, W. R., & Petrash, G. P. (1997). Visualizing, measuring, and managing knowledge. Research-
Technology Management, 40(4), 24–31.
Int Entrep Manag J (2009) 5:139–163 161

Carpenter, M. A., Sanders, G., & Gregersen, H. B. (2001). Bundling human capital with organizational
context: The impact of international assignment experience on multinational firm performance and
CEO pay. Academy of Management Journal, 44(3), 493–511.
Cohen, J., & Cohen, P. (1983). Applied multiple regression/correlation analysis for the behavioural
sciences. Hillsdale: Lawrence Erlbaum.
Collis, D. J., & Montgomery, C. A. (1995). Competing on resources: Strategy in the 1990s. Harvard
Business Review, 73, 118–128.
Cool, K., & Dierickx, I. (1989). Asset stock accumulation and sustainability of competitive advantage.
Management Science, 35(12), 1504–1514.
Danvila del Valle, I., & Sastre Castillo, M. A. (2005). Problemas y propuestas de medición de la
formación en la empresa. Cuadernos de Estudios Empresariales, 15, 27–45.
D’Arcimocles, C. H. (1997). Human resources policies and company performance: A quantitative
approach using longitudinal data. Organization Studies, 18(1), 857–874.
Delaney, J., & Huselid, M. (1996). The impact of human resources management practices on perceptions
of organizational performance. Academy of Management Journal, 39(4), 949–969.
Delery, J. E., & Doty, D. H. (1996). Modes of theorizing in strategic human resource management: tests of
universalistic, contingency, and configurational performance predictions. Academy of Management
Journal, 39(4), 802–835.
Goval, A., & Welch, I. (2004). A comprehensive look at the empirical performance of equity premium
prediction. National Bureau of Economic Research, 12(3), 580–631.
Grant, R. M. (1991). The resourced-based theory of competitive advantage: Implications for strategy
formulation. California Management Review, 33(3), 114–135.
Hall, R. (1992). The strategic analysis of intangible resources. Strategic Management Journal, 13(2), 135–144.
Hatch, N. W., & Dyer, J. H. (2004). Human capital and learning as a source of sustainable competitive
advantage. Strategic Management Journal, 25(12), 1155–1178.
Hitt, M. A., Bierman, L., Shimizu, K., & Kochhar, R. (2001). Direct and moderating effects of human
capital on strategy and performance in professional service firms: A resource-based perspective.
Academy of Management Journal, 44(1), 13–28.
Holzer, M. A. (1995). The impact of human resource management practices on turnover, productivity and
corporate financial performance. Academy of Management Journal, 38(3), 635–672.
Hunter, L. W. (2000). The adoption of innovative work practices in service stablishments. International
Journal of Human Resource Management, 11(3), 477–496.
Huselid, M. A. (1995). The impact of human resource management practices on turnover, productivity,
and corporate financial performance. Academy of Management Journal, 38(3), 635–672.
Huselid, M. A., & Becker, B. (1996). Metholodogical issues in cross-sectional estimates of the human
resource-firm performance link. Industrial Relations, 35(3), 400–422.
Huselid, M. A., Jackson, S. E., & Schuler, R. S. (1997). Technical and strategic human resource management
effectiveness as determinants of firm performance. Academy of Management Journal, 40(1), 171–188.
Ichniowski, C., Shaw, K., & Prennushi, G. (1997). The effects of human resource management practices
on productivity. American Economic Review, 87(3), 291–313.
Kidder, P. J., & Rouiller, J. Z. (1997). Evaluating the success of a large-scale training effort. National
Productivity Review, 16(2), 371–380.
Koch, M. J., & Mc Grath, R. G. (1996). Improving labor productivity: Human resource management
policies do matter. Strategic Management Journal, 17(5), 335–354.
Kochan, T. A., & Osterman, P. (1994). The mutual gains enterprise: Forging and winning partnership
among labor, management and government. Boston: Harvard Business School Press.
Kroll, M., Wright, P., & Heiens, R. A. (1999). The contribution of product quality to competitive
advantage: Impact on systematic variance and unexplained variance in returns. Strategic Management
Journal, 20(4), 375–384.
Kruse, D., Freeman, R., Blasi, J., Buchele, R., & Scharf, A. (2004). Motivating employee–owners in
firms: Human resource policies and company performance. National Bureau of Economic Research, 6
(2), 170–202.
Kydd, C. T., & Oppenheim, L. (1990). Using human resource management to enhance competitiveness:
Lessons from four excellent companies. Human Resource Management, 29(2), 145–166.
Lee, J., & Miller, D. (1999). People matter: Commitment to employees, strategy and performance in
Korean firms. Strategic Management Journal, 20(6), 579–593.
Lee, S. H., Phan, P. H., & Chan, E. (2005). The impact of HR configuration on firm performance in
Singapore: A resource-based explanation. International Journal of Human Resource Management, 16
(9), 1740–1758.
162 Int Entrep Manag J (2009) 5:139–163

Leonard, J. S. (1990). Executive pay and firm performance. Industrial and Labor Relations Review, 43(1),
13–29.
Lloréns, F. J. (1996). Procesos, contenido y efectividad de la calidad total: Una aproximación desde la
dirección de empresas. Revista Europea de Dirección y Economía de la Empresa, 5(3), 163–180.
Martell, K., Gupta, A., & Carroll, S. J. (1996). Human resource management practices, business strategies,
and firm performance: a test of strategy implementation theory. Irish Business and Administrative
Research, 17(1), 18–35.
Martín Alcázar, F., Romero Fernández, P. M., & Sánchez Gardel, G. (2005). Strategic human resource
management: Integrating the universalistic, contingent, configurational and contextual perspective.
International Journal of Human Resource Management, 16(5), 633–659.
Mc Duffie, J. P. (1995). Human resource bundles and manufacturing performance, flexible productions
systems in the world auto industry. Industrial and Labor Relations Review, 48(2), 197–221.
McEvoy, G. M. (1997). Organizational change and outdoor management education. Human Resource
Management, 36(2), 235–250.
McGahan, A. M., & Porter, M. E. (2003). The emergence and sustainability of abnormal profits. Strategic
Organization, 1(1), 79–108.
Murray, B., & Raffaele, G. C. (1997). Single-site, results level. Evaluation of quality awareness training.
Human Resource Development Quaterly, 8(3), 229–245.
Ordiz Fuertes, M. (2002). Prácticas de alto rendimiento en recursos humanos: concepto y factores que
motivan su adopción. Cuadernos de Economía y Dirección de la Empresa, 12, 247–266.
Osterman, P. (1995). Skill, training, and work organization in American establishments. Industrial
Relations, 34(2), 125–146.
Peck, S. R. (1994). Exploring the link between organizational strategy and the employment relationship:
The role of human resources policies. Journal of Management Studies, 31(5), 715–736.
Peteraf, M. A. (1993). The cornerstones of competitive advantage. A resourced-based view. Strategic
Management Journal, 14(3), 179–191.
Pfeffer, J. (1994). Competitive advantage through people. Boston: Harvard Business School Press.
Pfeffer, J. (1998). Seven practices of successful organizations. California Management Review, 40(2), 96–124.
Pfeffer, J., & Baron, J. N. (1988). Taking the workers back out: Recent trends in the structuring of
employment. Research in Organizational Behaviour, 10(2), 257–303.
Procopio, A. J., & Fairfeld-Sonn, J. W. (1996). Changing attitudes toward quality. An exploratory study.
Group & Organization Management, 21(2), 133–145.
Purcell, J. (1999). Best practice and best fit: Chimera or cul-de-sac? Human Resource Management
Journal, 9(3), 26–41.
Raghuram, M. (1994). Linking staffing and training practices with business strategy: A theoretical
perspective. Human Resource Development Quarterly, 5(3), 237–251.
Rajagopalan, N. (1996). Strategic orientations, incentive plan adoptions, and firm performance: Evidence
from electric utility firms. Strategic Management Journal, 18(2), 761–785.
Ribeiro, D. (2005). The new role of the corporate and functional strategies in the tourism sector: Spanish
small and medium-sized hotels. Service Industries Journal, 25(4), 601–613.
Richard, O. C., & Johnson, N. B. (2001). Strategic human resource management effectiveness and firm
performance. International Journal of Human Resource Management, 12(2), 299–310.
Rodríguez Pérez, J. M. (2001). Los sistemas de gestión de recursos humanos y la efectividad de las
organizaciones. Oviedo: Documento del Departamento de Administración de Empresas y Con-
tabilidad de la Universidad de Oviedo.
Rogers, E. W., & Wright, P. M. (1998). Measuring organizational performance in strategic human resource
management: Problems, prospects and performance information markets. Human Resource
Management Review, 8(3), 311–331.
Saá Pérez, P., & García Falcón, J. M. (2002). A resource-based view of human resource management and
organizational capabilities development. International Journal of Human Resource Management, 13
(1), 123–140.
Sánchez Marín, G., & Aragón Sánchez, A. (2002). Retribución del directivo, factores contextuales y
resultados de la empresa: evidencia empírica en el sector químico-farmacéutico. Cuadernos de
Economía y Dirección de la Empresa, 12, 315–336.
Sastre Castillo, M. A., & Aguilar Pastor, E. M. (2000). El enfoque estratégico en la dirección de recursos
humanos. Revista de Estudios Financieros, 202, 103–134.
Sastre Castillo, M. A., & Aguilar Pastor, E. M. (2003). Dirección de recursos humanos: Un enfoque
estratégico. Madrid: McGraw Hill.
Int Entrep Manag J (2009) 5:139–163 163

Schuler, R. S., & Jackson, S. E. (1987). Linking competitive strategy with human resource practices.
Academy of Management Executive, 1(3), 207–219.
Terpstra, D. E., & Rozell, E. (1993). The relationship of staffing practices to organizational level measures
of performance. Personnel Psychology, 46(1), 27–48.
Tzafrir, S. S. (2005). The relationship between trust, HRM practices and firm performance. International
Journal of Human Resource Management, 16(9), 1600–1622.
Úbeda García, M. (2005). Training and business performance: The Spanish case. International Journal of
Human Resource Management, 16(9), 1691–1710.
Walton, R. A. (1985). From control to commitment in the workplace. Harvard Business Review, 63(2),
77–84.
White, J. B., & Miles, M. P. (1996). The financial implications of advertising as an investment. Journal of
Advertising Research, 36(4), 43–52.
Wood, S., & Albanese, M. (1995). Can we speak of a high commitment management of the shop floor?
Journal of Management Studies, 32(2), 215–247.
Wright, P. M., & McMahan, G. C. (1992). Theoretical perspectives for strategic human resources
management. Journal of Management, 18(2), 295–320.
Youndt, M. A., Snell, S. A., Dean, J. W., & Lepak, D. P. (1996). Human resource management,
manufacturing strategy and firm performance. Academy of Management Journal, 39(4), 836–866.

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy