Geography Reference
In-Depth Information
knowledge inflows to all the firms in the group. As such, the individual
firm's view of the net benefits of these unintentional knowledge outflows
will therefore depend on its assessment of the relative costs and benefits of
these two opposing effects. These costs and benefits will depend on both
the industry structure and also the technology and knowledge characteris-
tics of the industry. For the moment we will just comment on the issue of
industry structure, and in the next section we will deal with the effects of
technology and knowledge characteristics.
In the case of a competitive market structure characterized by a large
number of firms, each with a relatively small market share and profits,
such firms probably have little to lose from unintentional knowledge
outflows and more to gain from inflows stemming from a strong clustered
location. As such, in the case of a broadly competitive market structure
the public good aspect of knowledge would appear to dominate, with
the local knowledge outflows being viewed as generally positive both
for the firms themselves and for the local region (e.g., Jaffe et al. 1993;
Saxenian 1994). On the other hand, an oligopolistic industrial structure
is characterized by a few large firms, each with a large market share and
considerable strategic interdependence. Often, oligopolistic firms realize
that unintentional knowledge outflows to industry rivals can be extremely
costly in terms of lost competitive advantage, because the private good
aspect of knowledge is one of their dominant considerations. In these situ-
ations where any unintentional knowledge outflows from a firm are more
valuable to a firm's competitors than any potential knowledge outflows
from these competitors to the firm, the overall effect of such outflows is
perceived to be negative. From the perspective of firm location strate-
gies, the clustering of oligopolistic firms would appear to jeopardize their
proprietary knowledge assets by exposing themselves to the possibility
of unintentional outward knowledge spillovers. If this is so, this will lead
such firms to decide not to locate in clusters, unless they can find a way
of avoiding such knowledge leakages. We can therefore use this argument
concerning the avoidance of unintentional knowledge outflows to recon-
sider the attractiveness of a cluster for MPDEs or MNEs, given that most
of them are oligopolistic firms.
In terms of our spatial typologies above, the possibility of unintentional
knowledge outflows is associated most obviously with the model of pure
agglomeration. Knowledge can be shared between two parties, but if there
is little or no inter-firm loyalty within the system, this knowledge will also
be passed on to third parties who are beyond the control of the originator
of the information. As such, a pure agglomeration-type of location will
create concerns for an oligopolistic MNE establishment. Similarly, in the
case of a social network type of environment, where the non-opportunistic
