COMPARING WATER ACCESS REGIMES UNDER CONDITIONS OF SCARCITY 219
are something to be minimized so that trading can ensue and allocative
efficiency can be maximized (Colby 1990). This exact argument has been
applied in the case studies I will describe with respect to water and water
markets, where high transaction costs are seen as a barrier to transfer-
ring water from supposedly “low-value” agricultural uses to “high-value”
urban uses.
I do not find the standard argument convincing, because I do not
find the notion of allocative efficiency to be an unproblematic or “scien-
tifically neutral” policy goal, given that it strongly weighs the preferences
of those with the needed political clout and financial resources to express
their interests in well-functioning markets. I have no problem with the
goal as such, but I do not see it as value neutral, which the language of
efficiency strongly implies, nor do I see it as one that should be preferred
over many others that we might specify. In any case, I do not think that
path dependence is something that can be avoided. If we like the path we
are on, then we presumably want to encourage at least some measure of
path dependence.
There is another way in which the emphasis on path dependence
departs from standard economic analysis. Within the discipline of environ-
mental economics (and applied/welfare economics generally), a diagnosis
begins by imagining an ideal situation, usually the perfectly competitive
market. The diagnosis then proceeds by analyzing the extent to which
a particular situation departs from this ideal, and these departures are
seen as evidence that one type of policy intervention or another might be
appropriate for approximating the ideal.
In contrast, a diagnostic that recognizes path dependence turns this
approach on its head. It proceeds in an inductive fashion from a particu-
lar social and ecological context and considers how it might be adapted
to achieve a particular goal, keeping in mind the constraints inherent in
the existing path. This approach is analogous to differences between “hard
energy paths” and “soft energy paths” or a chemical analysis that starts off
with observing the constraint of the “adjacent possible,” which in a chemi-
cal context is all the possible products that a set of reactants could produce
(Kauffman 2002). Simply put, recognizing path dependence means that
we cannot start with an idealized abstraction, because we may not be able
to obtain something that could meaningfully approximate such an ideal.
To a large extent we have to deal with what we have and make marginal
steps from there. With this discussion in mind we now turn to the case
two studies.