rather than normative economics.l/ The consequences of this approach for the
relations between economics and law will become apparent later (Section 6).
2. Interpretation of "Security" of Water Rights
To the economist, "security" of water rights means something different and
much broader than their "protection" means to the student of law. The latter
concept merely means protection against unlawful acts by others--as such acts
are construed by the law. Such protection is always subject to the two major
categories of "legal uncertainty," that is, to "rule uncertainty" and to "fact
uncertainty."2/ Legal uncertainty, in this sense, is a characteristic of judi-
cial decisions. Like other types of uncertainty, it also affects economic
Economists are inclined to disregard or underestimate the significance of
legal uncertainty. For them, security of a water right connotes (1) protection
against what I propose to call "physical uncertainty," that is, against vari-
ability over time of the quantity of water usable under the right due to seasonal
or annual variability of "natural" run-off and ground-water recharge, and (2) pro-
tection against what I shall call "uncertainty of water tenure"'/ or, for short,
"tenure uncertainty," that is, protection against variability over time of the
quantity of water usable under the right due to lawful acts of others--who may
be individuals or groups, private or public.
i/ For a discussion of this differentiation see: Friedman, Milton, Essays in
Positive Economics (Chicago: University of Chicago Press, 1953).
2/ This terminology has been popularized by Frank. See: Frank, Jerome, The
Law and the Modern Mind (6th ed.; New York: Coward McCann, 1948).
3/ Tenure uncertainty is not confined to water--and other "fugitive" resources--
but is one of the most important economic forces affecting resource use. See:
Ciriacy-Wantrup, S. V., "Capital Returns from Soil-Conservation Practices," Journal
of Farm Economics, vol. XXIX, no. 4, pt. 2, November, 1947, pp. 1181-1196.
_II__ _~___1~1~ I
We are concerned here with the relative degree of security resulting from dif-
ferent types of rights. Absolute security, that is, transformation of probabilities
into single-valued expectations, cannot be obtained through water law. Furthermore,
we are not concerned with the many other types of uncertainty which affect economic
decisions--for example, uncertainties connected with variability over time of wants,
technology, prices, and incomes. A few examples may illustrate differences between
water rights if "security" is interpreted in the way just indicated.-
Under natural conditions, a senior appropriative right is more secure than a
junior appropriative right against physical uncertainty but not necessarily against
tenure uncertainty. A water right in a high preference class is more secure than
one in a low preference class against tenure uncertainty but not necessarily against
physical uncertainty. A water right restricted by reservations in favor of other
users is less secure than these other rights against tenure uncertainty but not
against physical uncertainty. An appropriative right is more secure against tenure
uncertainty than a riparian right or a ground-water right under the correlative
rights doctrine--assuming that only one system of rights applies to a water resource.
Against physical uncertainty, however, an appropriative right is not necessarily
more secure than a riparian or correlative right; this holds especially (but not
solely) for a junior appropriative right.
In all these examples, the two main categories of legal uncertainty are also
present. The degree of legal uncertainty may be quite different for different water
right systems and individual water rights. For purposes of this paper, however, it
is meaningful to proceed "as if" the degree of legal uncertainty were approximately
I/ For these examples, one assumes, of course, "other things being equal," that
means in this case physical conditions and legal features other than those under
consideration being the same.
If one speaks of "quantity of water usable" under a water right, one implies
certain characteristics with respect to water quality. Over time, water quality
is also subject to legal, physical and tenure uncertainties. When water rights
are defined in quantitative terms--as under appropriation and through adjudica-
tion under the riparian and correlative rights doctrines--reference should always
be made to water quality.
Security of water rights in terms of water quality is no less significant
economically than security in terms of water quantity. Frequently, the higher
water uses--in terms of value product--require a higher quality of water than
the lower uses. Furthermore, quality and quantity are directly interrelated:
water of high quality--in terms of low mineral content, especially sodium and
sulphates--can be reused, sometimes for more than one cycle. Such water can also
be used to upgrade surface and ground water of lower quality which could other-
wise be used only for lower uses or not at all. These interrelations of quality
and quantity are especially significant when water must be transported over
great distances. Such transport may be economically feasible only for high
quality water. It follows, from the interrelations of quantity and quality just
explained, that in setting minimum standards for water quality, problems of reuse
of the same water and upgrading of other waters need to be considered. Minimum
standards based on suitability for one cycle of direct ("unmixed") use alone may
be, economically speaking, too lenient.
Sometimes, the only security of water rights in terms of quality is their
protection against unlawful acts of water pollution (including for purposes of this
paper, contamination and nuisance). In many states, antipollution laws are being
strengthened. More federal action is also being considered.-/ Dealing specifically
with these attempts would require a separate paper.
I/ The federal government is no newcomer in this field. See, for example, the
legislation known as the Caminetti Act f7 U. S. Stat. at L. (1893), 507, as amended
by 30 V. S. Stat. at L. (1898), 631; 30 U. S. Stat. at L. (1899), 1148; 31 U. S. Stat
at L. (1900), 631; 34 v. 8. Stt. at L. (1907), 1001i 48 U. S. Stat. at L. (193a),
.118; 52 U. S. Stat. at L. (938) 10/.
Physical uncertainty can be significantly reduced through physical means,
that is, mainly through storage of water--above and below ground--from season
to season, from year to year, and over a period of years. A reduction of
physical uncertainty in this sense produces benefits for water users and others.
Whether these benefits are "net" depends on the costs of constructing, maintain-
ing, and managing storage.
A system of water rights may impede construction of storage facilities by
limiting development to the "safe yield" of a water resource and by making cycli-
cal management of storage capacity and coordination of surface and subsurface
storage more difficult.
The appropriation doctrine can easily be used to limit water development.
In several western states, this doctrine has actually been used with this objec-
tive and result in the case of ground water. With respect to surface water, the
actual facts are somewhat different. Most western surface waters are "overappro-
priated." Sometimes, appropriation covers a multiple of the average flow during
the season of use.
After storage capacity has been provided and is managed with a view to re-
ducing physical uncertainty, the relative economic status of appropriative rights
changes without changes in their relative legal status. Priority in time, in
conjunction with the quantitative definition of appropriative rights, limits the
number of rights that can be served with the regulated flow. Rights exceeding
this flow become, economically speaking, less meaningful the better the flow is
regulated. For rights that can be served with the regulated flow, the new situa-
tion, in terms of economics, is not greatly different from that prevailing under
a water delivery contract where a limited number of users are equal in right--
although the quantities to which their contract entitles them may differ.
Thus, the differentiation of water rights, so characteristic for the appro-
priation doctrine, has undergone a shift in its economic implications: within
limits, a junior right is no longer less secure against physical uncertainty
than a senior right. Increasingly, storage is provided by large public or
semipublic projects and managed by government agencies--public districts, state
water departments and boards, and federal bureaus.
In contrast to appropriation, riparian rights are coequal in law.J/ An
economic implication of this equality is that their legal status is not a cause
for differentiation with respect to their security against physical uncertainty.
This holds both for the riparian right to surface waters, as it is generally
applied in the eastern states, and for the application of the riparian right to
ground water through the correlative rights doctrine as, for example, in Cali-
This lack of differentiation on the basis of legal status does not neces-
sarily mean that there are no differences between individual riparian or corre-
lative rights in security against physical uncertainty. Applying legal equality
through adjudication to a given stream system or ground-water basin is a time-
consuming and costly process. For a while, at least, individual riparian and
correlative rights may differ in security against physical uncertainty. Such
differences are caused not by legal but physical facts such as relative location
of diversions along a surface stream and the location of wells with respect to
the source of recharge and the geologic characteristics of a ground-water basin.
------------------------------------- ----- --- -
1/ They are coequal within the two traditional preference classes of "natural"
and "articifical" uses--except that an upstream user need not share with down-
stream users if the flow of the stream is only sufficient to satisfy his own
Differences between individual water rights (and between water-right
systems where they exist side by side in the same general area) in security
against physical uncertainty may obstruct or retard construction and coordinated
cyclical management of storage. Individuals and groups who "have" greater
security may be reluctant to join those who "have not" if a portion of the costs
of reducing physical uncertainty is incident on the former.
There are a number of institutional approaches to this problem. One
approach is based on a type of public district through which water rights are
acquired, or which acts as agent in utilizing them, for coordinated cyclical
management of surface and ground water. This approach is now being investi-
gated in our ground-water studies/ for Santa Clara County, California. In
many parts of the West, water demand is increasing so rapidly that the group
of "haves" (in terms of security of water rights against physical uncertainty)
is becoming smaller absolutely and, of course, even more so relative to the
number of "have nots." The formation of public districts and of other insti-
tutional arrangements for coordinated water management is thereby facilitated.
1/ Smith, Stephen C., "Problems in Using the Public District for Ground-
Water Management," Journal of Land Economics. (Accepted for publication.)
For other studies in this series, see:
Bartz, Patricia McBride, with a foreword by S. V. Ciriacy-Wantrup, Ground
Water in California. The Present State of Our Knowledge (Berkeley: University
of California, College of Agriculture, Agricultural Experiment Station, Septem-
ber, 1949), 67p. (Giannini Foundation Ground Water Studies No. 1.) Processed.
Snyder, J. Herbert, with a foreword by S. V. Ciriacy-Wantrup, Ground-
Water in California. The Experience of Antelope Valley (Berkeley: University of
California, Division of Agricultural Sciences, Agricultural Experiment Station,
February, 1955), 171p. (Giannini Foundation Ground Water Studies No. 2.) Proc-
Turning now to security of water rights against tenure uncertainty, it
was mentioned earlier that appropriative rights are more secure in this respect
than riparian rights or ground-water rights under the correlative rights doc-
trine. The former are clearly defined in priority, quantity, period of use,
points of diversion, and in other ways. The latter are coequal, and quanti-
tative definition depends on adjudication which is in terms of shares and sub-
ject to the restriction that there shall be reapportionment if the conditions
upon which the original apportionment was made change sufficiently to justify
There are, however, several factors which modify such a general compari-
son of water-right systems. The first factor is prescription which operates
under all water-right systems--with some differences in economic significance
(Section 4). The second factor is differentiation of preference classes based
on purpose (kind, type) of water use. This factor also operates under all
water right systems but under appropriation only before rights are vested and
under emergencies. Customary preference classes are "natural" and "artificial"
under the riparian doctrine and "domestic," "municipal," "industrial," "agri-
cultural," and "recreational" under appropriation. The third factor is a re-
striction ("reservation") on water rights to yield to future water demands by
others. This factor operates only under appropriation and affects future
(not already vested) rights. Water reservations are frequently used in favor
of certain preference classes. But the connection with preference classes is
not a necessary one. In California, water reservations in favor of whole re-
gions are in force regardless of preference class.
There can be little argument on economic or any other grounds that domes-
tic and municipal uses deserve special consideration as far as security against
tenure uncertainty is concerned. This type of use can frequently prevail over
other uses by eminent-domain proceedings. Still, the highest preference ranking
and water reservations in favor of this use are desirable because other uses,
for example, agricultural, may also be organized as public districts. By ade-
quate requirements for the treatment of sewage, domestic and municipal use can
be made largely nonconsumptive. Agricultural use, on the other hand, is largely
An economic argument can be made in favor of abolishing the usually-/
lower preference rating of industrial uses relative to agricultural. The
average value product of consumptive use is higher in industry than in agri-
culture. Furthermore, in the western states, total consumptive use in industry--
although increasing--is still relatively small as compared with that in agricul-
In ascertaining the value product of recreational uses--which are largely
nonconsumptive--one faces the difficult problem of extramarket values. There
are cases, however, in which the economic argument suggests a change in the
present preference ranking of recreational uses. Generally, this ranking is
the lowest. Under some statutes, recreational uses are not even recognized as
"reasonable and beneficial."
1/ Significant exceptions occur, for example, in Texas.
2/ We are using the word "average" because each of the two preference classes
covers water development by different users for different products in different
time intervals. In the present context, we refer, of course, to future "addi-
tional" water development. A change in preference classification does not affect
already vested appropriative rights. Under the riparian doctrine, industrial and
agricultural uses are both "artificial" and, therefore, coequal.
In spite of the existence of a favorable economic argument for abolishing
the usual statutory ranking of industrial and agricultural uses and for other
changes in preference classifications, inferences with respect to public policy
would be premature without considering the criterion of flexibility jointly
with that of security. The economic implications of water reservations, like-
wise, cannot be fully appraised without considering the criterion of flexibility.
An interpretation of this criterion will be undertaken in a later section (Sec-
tion 4). Before this can be done, our interpretation of security of water
rights must be related to "protection of investment" in water resources develop-
3. Security of Water Rights and Protection of Investment
The relations between security of water rights and investment in water
resources development are generally the main point of emphasis when the eco-
nomic implications of differences in individual water rights and whole water-
right systems are discussed. This emphasis is justified. Most economic im-
plications of security which were mentioned in the preceding section are in
this area. The objective of the present section is to state some of these
relations more explicitly and to consider them in the light of the concept,
"protection of investment."
In economic theory, "investment" and its corollary "disinvestment" refer
to value changes in total capital of individuals or whole social groups as a