ABSTRACT. Van Dijk et al. found different results for some
variables explaining the large/small firm innovation advantage
in a number of Acs and Audretsch publications. Their own
research adds to this variety. However, the differences van
Dijk et al. found between the Acs and Audretsch studies can
be largely attributed to specification and sample differences.
The same applies to the observed differences between Acs
and Audretsch and van Dijk et al.
A considerable amount of research has been
devoted to the question, whether large or small
firms are more innovative. Schumpeter’s hypoth-
esis that large firms have an advantage in inno-
vation has sparked this debate (Schumpeter, 1942).
However, this seems to contradict his earlier con-
tention that new, entrepreneurial firms are more
vigorous innovators (Schumpeter, 1939).
The first hypothesis is based on the appropri-
ability advantages from which large firms benefit.
Obviously, an innovation will produce more
profits at larger quantities sold, if profit margins
are identical. Since new firms are on average
smaller than established firms, they have a dis-
advantage in this respect. But, small and new firms
have more to gain from innovation, since innova-
tion will boost their profits more. This applies with
the greatest force when new (and small) firms can
reach large size quickly.
The empirical research on the relationship
between firm size and innovativeness did not pro-
duce unambiguous answers to the question which
firm size is most conducive to innovation. Which
firm size has the innovative advantage varies by
industry and by the dependent variable used.
Research has shown, that large firms predomi-
nate in R&D. The largest firms (over 10,000
employees) performed 81.3% of all company
financed R&D in 1982. Among firms conducting
R&D, the broad picture is that R&D does not
increase more than proportionally with firm size
(Scherer and Ross). Research conducted at the
industry level shows that R&D increases propor-
tionally with size in most industries. The number
of industries, in which R&D spending increased
more than proportionally with size slightly out-
numbered those with the opposite pattern (Scherer
and Ross, 1990).
But, small firms are much more innovative than
large firms when an output measure of innovation
is used (direct innovation counts). Acs and
Audretsch demonstrated that small firms are much
more efficient at innovation than their larger
counterparts. Small firms produced 43% more
innovations per employee for all manufacturing
industries. Their advantage is paramount in highly
innovative industries, where they produced 6.6
times more innovations per employee (Acs and
Audretsch, 1988). Hence, small firms (< 500
employees) need much less R&D per innovation
than their larger counterparts.
Bob van Dijk et al. contribute to the ongoing
debate on relative innovative advantage and firm
size in their august 1997 contribution to this
journal. They summarize a number of the Acs and
Audretsch results in a neat table, in which they
note some inconsistencies. Subsequently they
present their own results based on Dutch data on
R&D intensity and firm size. I want to add to the
discussion by commenting on these alleged incon-
sistencies. A second objective is to discuss their
Firm Size and Efficiency in Innovation:
Comment on van Dijk et al.
Small Business Economics
11: 391–393, 1998.
1998 Kluwer Academic Publishers. Printed in the Netherlands.
Final version accepted on November 25, 1997
Department of Economics and Econometrics
University of Amsterdam
1018 WB Amsterdam