The bivariate analysis revealed that larger firms were
more willing to grow and had more positive growth expectations.
Two explanations for the differences are possible. The larger
firms have already grown quite considerably and have a growth
experience that may give them a more positive attitude towards
growth, whereas smaller firms may overestimate the problems of
growth. Larger firms have also passed the critical stage where
professional management is introduced which usually takes place
between 10 and 20 employees (Stanworth & Gray, 1991). Further
growth beyond this point may be perceived as less problematic.
The other explanation has the reversed causality. It may be that
larger firms had a more positive growth attitude already when
small which has facilitated their growth, i.e. a process of
self-selection. The universality of the size effect across
dimensions suggests that self-selection may be the main cause of
the differ-ences. Also, younger firms were more positive than
older firms, giving support to life-cycle models. The industry
comparisons suggested that no industry differences existed.
A central finding is that expectations of financial gain is not the outstanding determinant of growth willingness. This is clearly contrary to economic theory, but also to normative management theories were the willingness to grow is taken for granted and financial outcome the primary concern of the manager. Our findings suggest that other expected outcomes of growth that are largely non-economic in nature also influence growth willingness. This includes, e.g., beliefs concerning the effect of growth on the managers workload, his or her ability to keep full control over the operations of the firm, the ability to keep up the quality of products and services, the firms degree of independence in relation to external stakeholders, and its ability to survive crises. The effects are generally not very strong. However, while not always statistically significant according to conventional criteria, they appear very consistently with the same sign across samples, industries, size classes and age groups. What this means is probably that the effects are real, but they are small because each type of expected outcome has a substantial effect on growth willingness only for some managers, whereas it is relatively uninfluential for others. Hence the modest average effect represented by the regression coefficients. This illustrates the strength of our combining data from three studies using the same instruments. In a single and perhaps smaller study, each and every one of these effects may have been disregarded as non-significant, therefore non-existent--a very common practice among social scientists. The same would probably have happened had three different studies used different instruments to assess the same theoretical relationships. With our same instrument, multiple samples approach, we can conclude with confidence that these expectations do have an effect on growth willingness.
Our most important finding is that expectations concerning the effect of growth on employee well-being is the single most important determinant of over-all growth willingness. This result stands out with impressive consistency across samples, industries, size classes and firm age groups. The important question is, of course, exactly what this means. The very translation of the Swedish original is tricky. While no fully satisfactory translation to English presents itself, the reader might find it informative that the Swedish wording is likely to evoke associations to companionship, comfort, atmosphere, and job satisfaction.
Several interpretations are possible. Firstly, a critical mind would perhaps like to dismiss the result as not really showing that small firm owner-managers great concern for their employees affects their growth willingness. Rather, one might argue, is this variable picked as their scapegoat for less socially desirable reasons to refrain from growing their firms larger. However, results reported previously (Davidsson, 1989b) suggest that while perhaps more of a growth deterrent than a growth motivator, the variable work both ways. That is, some managers believe that growth will improve employee well-being, and this enhances their growth willingness. Another reason for not dismissing the result in this way is that there are other expected outcomes, most notably product/service quality, that would be an equally logical candidate for a social desirability effect.
However, we would agree that it is not unlikely that the
managers also have their own well-being in mind when answering
the question. A sound interpretation is probably that the result
reflects a real concern for the "soft" qualities
associated with small scale, and that this concern is
justifiable. There is research to suggest that on issues like
comradeship, involvement and job satisfaction, employees and
people in general think highly of small firms (Curran et al,
1993). Even more impressive evidence for the "soft"
advantages of small scale is presented in the classical study by
Barker & Gump (1964; cf. also their extensive references to
other studies). Therefore, the small firm owner-manager may have
a very real reason to be concerned about the
atmosphere of the small firm when faced with expansion opportunities. The generality of our finding in all sub-analyses suggests that this concern is a source of an eternal goal conflict for many small business owner-managers.
At the very least, the issue is worthy of further research efforts. One aspect of this would be to assess whether the concern for employee well-being and its effect on growth willingness is a cultural peculiarity of Sweden (or perhaps the Scandinavian countries) or if it is more universal. While both outcomes of such an assessment are conceivable, recent research suggests that a very high concern for employees is not unheard of in other cultures (cf. Kazumi et al, 1996). Future research should also try to capture in more detail what the "concern for employee well-being" really encompasses, whether it affects real growth and not only growth willingness, if managers who have successfully led their companies through a growth phase also felt such concerns before the expansion, whether their expectations were accurate, and what they might have done to counteract the possible negative effects of growth on employee well-being.
Finally, our study has been about growth willingness,
not actual growth. Whether expected consequences have an
influence on actual growth or not is an empirical question that
remains to be answered. We would hypothesise that such is the
case, and that the relative importance of different expected
outcomes also would be similar to what we obtained here. As
regards the control variables age, size and industry, these have
already been used and found to be important in several studies of
© 1997 Babson College All Rights Reserved
Last Updated 06/01/98