policies, is the eradication of discrimination. This is not to deny the importance of
HRM contributing to performance; merely to note thatWnancial performance is
only one dimension of eVectiveness.
26.4 The Problem of the Scope
of HR Practices
.........................................................................................................................................................................................
Having dealt with problems concerning the dependent variable, performance, we
now need to consider the input to the model of HRM: the HR practices used in
analyses. Alas, many problems abound here too. There is no agreement on what
constitutes ‘HR practices’ let alone a full set of them. Boselie et al. ( 2005 ) identiWed
twenty-six general categories of practices used by researchers. Most researchers
construct a list of practices but there is no agreement on what or which practices to
include. While core practices associated with recruitment and selection, training
and development, and performance management (appraisal and variable pay) are
nearly always included, others like job design and involvement are much more
sporadic.
There is little debate on where lists of practices come from or what criteria to use
in their construction. The lists appear to emerge from sets of practices normally
associated with activities undertaken by well-staVed, sophisticated HR depart-
ments in largeWrms often linked to so-called ‘transformational’ approaches to
the management of labor. These predetermine what type of HR practices are
hypothesized to lead to outcomes of higher performance as made clear by PfeVer’s
list of seven practices ‘for building proWts by putting peopleWrst’ ( 1998 ). Behind
such choices is the long-standing debate of whether HRM requires ‘a distinctive
approach to employment management’ (Storey 1995 : 5 ), sometimes referred to as
‘developmental humanism’ or ‘soft’ HR, or a generic term covering all aspects of
the management of labor (Boxall and Purcell 2003 : 4 ). The former tends to focus
on ‘innovative’ HR policies while the latter takes a broader perspective to include,
for example, collective work relations often excluded from American studies
(Deery and Iverson 2005 ). Godard’s ( 2004 ) review of the ‘high-performance
paradigm’ (HPP)Wnds little support for links with performance but notes increases
in work stress (an outcome not usually included in studies). He observes that ‘it is
again possible that practices traditionally considered to yield positive outcomes for
workers such as traditional group work or information sharing are as eVective or
even more eVective than practices associated with HPP’ ( 2004 : 360 ). We need to
Wnd a list which does not assume positive outcomes and may reveal negative
associations with performance, or no links at all.
538 john purcell and nicholas kinnie