barriers to water conservation in the rio grande basin1 · barriers to water conservation in the...

BARRIERS TO WATER CONSERVATION IN THE RIO GRANDE BASIN1

Frank A. Ward, Ari M. Michelsen, and Leeann DeMouche2

ABSTRACT: The Rio Grande basin shares problems faced by many arid regions of the world: growing and com-peting demands for water and river flows and uses that are vulnerable to drought and climate change. In recentyears legislation, administrative action, and other measures have emerged to encourage private investment inefficient agricultural water use. Nevertheless, several institutional barriers discourage irrigators from investingin water conservation measures. This article examines barriers to agricultural water conservation in the RioGrande basin and identifies challenges and opportunities for promoting it. Several barriers to water conserva-tion are identified: clouded titles, water transfer restrictions, illusory water savings, insecure rights to conservedwater, shared carry-over storage, interstate compacts, conservation attitudes, land tenure arrangements, and anuncertain duty of water. Based on data on water use and crop production costs, price is found to be a major fac-tor influencing water conservation. A low water price discourages water conservation even if other institutionspromote it. A high price of water encourages conservation even in the presence of other discouraging factors. Inconclusion, water-conserving policies can be more effectively implemented where water institutions and pro-grams are designed to be compatible with water’s underlying economic scarcity.

(KEY TERMS: water economics; irrigation; institutions; water policy; water conservation; sustainability.)

Ward, Frank A., Ari M. Michelsen, and Leeann DeMouche, 2007. Barriers to Water Conservation in the RioGrande Basin. Journal of the American Water Resources Association (JAWRA) 43(1):237-253. DOI: 10.1111 ⁄ j.1752-1688.2007.00019.x

INTRODUCTION

In most of the western United States, existingwater supplies are claimed and diverted for irrigationand growing municipal and industrial demands.Remaining flows are increasingly protected forinstream flows and environmental purposes. Mosteasily accessible ground water is developed or is depl-etable. Throughout the west, drought, climatechange, and emerging environmental laws and regu-

lations intensify the competition for water. The U.S.federal government recently identified the Upper RioGrande (see Figure 1) as among river basins havingthe highest potential for conflict and crisis, especiallyin drought conditions (U.S. Department of Interior,2003). The Rio Grande exemplifies the problems facedby many arid regions of the world (e.g., Colorado,USA; Yellow, China; Jordan, Middle-East; Murray-Darling, Australia; and Nile, Africa) in which wateris overallocated and there are growing and competingdemands and river flows and uses that are vulnerable

1Paper No. J05092 of the Journal of the American Water Resources Association (JAWRA). Received July 6, 2005; accepted March 21,2006. ª 2007 American Water Resources Association.

2Respectively, Professor, Department of Agricultural Economics and Agricultural Business, New Mexico State University, Las Cruces,New Mexico 88003; Professor and Resident Director, Texas A&M University, El Paso Agricultural Research Center, El Paso, Texas 79927-5020; and College Assistant Professor, Extension Plant Sciences Department, New Mexico State University, Las Cruces, New Mexico 88003(E-Mail ⁄ Ward: [email protected]).

JOURNAL OF THE AMERICAN WATER RESOURCES ASSOCIATION 237 JAWRA

JOURNAL OF THE AMERICAN WATER RESOURCES ASSOCIATION

Vol. 43, No. 1 AMERICAN WATER RESOURCES ASSOCIATION February 2007

fward

Typewritten Text

For non-commercial and educational use only. Please ask publisher for use permission

fward

Typewritten Text

fward

Typewritten Text

to drought and climate change. These factors alongwith the need for water policies that are sustainablehave highlighted the interest by policymakers, scien-tists, and water managers to examine carefully watermanagement alternatives that encourage, promote,and reward water conservation.

Originating in the southern Colorado Rocky Moun-tains, the Upper Rio Grande extends 600 miles(960 km) from its headwaters and flows through NewMexico to the border cities of El Paso, Texas, USA,and Ciudad Juarez, Chihuahua, Mexico (see Figure1). Downstream of El Paso, the river forms the inter-national border between the United States and Mex-ico on its way to the Gulf of Mexico. The Upper Basin(hereafter referred to as the basin) supports a grow-ing population of more than 3 million. For example,in 1900, the population of El Paso, Las Cruces andCiudad Juarez was 44,000. By 1950, it had grown to357,000 and in 2000 the population was over 2 mil-lion, almost a 50-fold increase over the century. Thepopulation is projected to nearly double again by2020. The basin also supports extensive irrigatedagriculture, and fish and wildlife habitat in Colorado,New Mexico, Texas, and the Mexican state of Chihua-hua. Some 80% to 90% of the water in the basin is

used for irrigated agriculture. Yet, the basin’s popula-tion is expected by some experts to double in the next50 years, potentially doubling urban water demands.This rapid population growth, in conjunction withincreased demands by all users, will further intensifythe competition for limited water resources.

Recent years have witnessed the emergence oflegislation, administrative action, and other measuresto encourage private investment in increased effi-ciency with the intent of promoting agriculturalwater conservation. Nevertheless, several institu-tional barriers may discourage irrigators from invest-ing in measures or otherwise taking actions designedto conserve water. In the basin, more than a centuryof federal water development programs and policieshas attached great importance to providing plentifuland reliable supplies at a low price. Because of thewidespread successes of these programs, a large num-ber of institutions have arisen that promote, reward,and support water use, especially in agriculture.

Factors that influence water conservation, bothinside and outside agriculture, have seen recentattention in the literature. Schaible (1997) found thatmajor water price reforms are required to compensatefor institutional barriers to conservation for irrigatorsin the U.S. Pacific Northwest. Moore and Negri(1992) found that reducing the supply of water towestern irrigators by 10% would increase thenational price of three of ten major crops produced byfarms using Bureau of Reclamation water. Yanget al. (2003) found that rapid increases in irrigationcosts in northern China since 1993 have failed to gen-erate a sufficient force for water conservation andthat water pricing reform by itself is not an effectivemeasure for promoting water conservation. In ananalysis of municipal water use in Ontario, Canada,De Loe et al. (2001) found that limited finances, lackof political will, and public resistance all constrainthe effectiveness of municipal conservation programs.Jenkins and Lund (2000) found that the high econo-mic cost of dealing with water shortages can bereduced by jointly expanding infrastructure as wellas eliminating institutional constraints.

Winter-Nelson and Amegbeto (1998) developed aneconomic model of optimal investment under uncer-tainty to analyze the effects of both the level andvariability of water’s price on the decision to con-serve. Loaiciga and Renehan (1997) examined theeffects of pricing and drought on water conservationin Santa Barbara, California during the 1986-96 per-iod when per capita supplies fluctuated considerably.Do Monte et al. (1996) analyzed how water use guide-lines could be designed to deal with similar problemsin Europe’s Mediterranean region.

Michelsen et al. (1999a, b) found that nonprice pro-grams could be an effective instrument for achieving

FIGURE 1. Upper Rio Grande Basin Map.

WARD, MICHELSEN, AND DEMOUCHE

JAWRA 238 JOURNAL OF THE AMERICAN WATER RESOURCES ASSOCIATION

water conservation for agriculture and seven cities inthe southwestern U.S. Huffaker and Whittlesey(2000, 2003), Stonehouse (1996), and Huffaker et al.(1998) found that increasing the price of water, poss-ibly through greater water marketing opportunities,is more effective than subsidizing the cost ofimproved on-farm irrigation efficiency at promotingwater conservation in irrigated agriculture. Anton(1995) summarized the Seattle Water Department’sexperience with water use curtailment measures forpromoting water conservation during the 1987 and1992 droughts. Mulwafu et al. (2003) found that irri-gators in Malawi, Africa made the fewest water con-servation investments when the price of water waslowest. Michelsen et al. (1999a, b) confirmed a longhistory of research findings showing that a low priceof water charged by the U.S. Bureau of Reclamationfor irrigation water strongly discourages water con-servation in agriculture. Peterson and Ding (2005), intheir analysis of the U.S. High Plains area, foundthat the presence of water-saving irrigation systemsdoes not guarantee water conservation in irrigatedagriculture, as long as water’s price remains low.

Pender and Kerr (1998) found that water conserva-tion investments by irrigators in the semi-arid landsof India is significantly lower on leased land and onlands subject to sales restrictions than on deededlands. Their results suggest considerable potential forland market reforms as a way to increase water con-servation investments. Sokolov (1999) found that var-ious water-saving methods in Uzbekistan, whencombined with price incentives, could secure a pathof sustainable development for agriculture. Zougmoreet al. (2004) found that efficient combinations oforganic resources and fertilizers will improve wateruse efficiency and productivity of agriculture in Bur-kina Faso, Africa. Cuthbert and Lemoine (1996)found that increasing numbers of U.S. water utilitiesare implementing seasonal rates, inverted blockrates, and excess use rates to provide pricing signalsthat promote water conservation.

Despite these above contributions, little researchhas examined institutional barriers to water conser-vation and identified institutional innovations thatcould circumvent those barriers. The goal of this art-icle is to help fill those gaps by examining institu-tional barriers to agricultural water conservation andidentifying challenges and opportunities for emergingefforts to promote water conservation in the basin’sagriculture. It accomplishes these aims by brieflyexamining selected organizations, laws, and systemoperating procedures that act as institutional barriersto agricultural water conservation. It also examineshow those barriers have influenced irrigation wateruse in the basin. Finally it identifies some challengesand opportunities presented by attempts to deal with

these barriers in the search for increased water useefficiency in the basin’s irrigated agriculture.

WHAT IS WATER CONSERVATION?

This is an old question. Yet, it continues to beasked today partly because so many modern publicpolicy sentiments favor water conservation. ‘‘Indeed,part of the allure of water conservation is that it canbe molded into a concept that is attractive to anyone(Griffin, 2006).’’ Many conservation concepts presumethat all programs that reduce water use are desirable,independent of cost or benefit. While this may beapproximately true when all saved water will beused for meeting the drinking needs of the thirsty,it is unlikely to be true in most practical conditions.Economic principles can be used to fashion a defini-tion of water conservation in which economicallysound decisions are made in connection with attemptsto save water in a wide range of hydrologic, economic,and institutional conditions.

One definition of water conservation is simple tech-nical efficiency in consumptive water use, hereafterreferred to simply as ‘‘use.’’ This technical definitionexpresses the desire to secure the most physical out-put per unit of water used. Examples includeattempts to transport water to end users that minim-ize nonrecoverable leakage and evaporation, encour-aging the adoption of low-flow showerheads, low-flushtoilets, or advanced irrigation technologies. The prob-lem with this definition of conservation is it is basedon a water theory of value (Baumann et al., 1984;Griffin, 2006). Squeezing as much as is technicallypossible out of every drop of water used is good eco-nomics only when all other resources have a zero costand when only water is scarce. Actions based on thisdefinition of conservation assume other resourceshave zero value and that the only resource with anyeconomic value is the water itself. For example, canalwater lost in conveyance because of leaks and evapor-ation is costly to prevent in the sense that otherresource costs must be incurred to save that lostwater. Do those resources have a lower economicvalue than the value of the water saved or do theyhave a greater value? From the technical efficiencyview, it is water-conserving to undertake any expen-ditures to prevent any water losses. From the econo-mic view, those water losses are not wasted unlessthe cost of preventing the loss is less than the valueof the water saved (Griffin, 2006).

Several U.S. states, in attempts to save water,have recently enacted legislation to this effect. In theRio Grande Basin, various actions could be taken

BARRIERS TO WATER CONSERVATION IN THE RIO GRANDE BASIN


that would encourage or require irrigators to substi-tute additional land, labor, capital, or money forreduced use of water. Under that view, all water usereductions are conservation. While this may be aclear view, it is economically naive and politicallyimpractical: water is only one of many scarceresources required for agricultural production. Forexample, regulations could be enacted requiring allirrigators to reduce their water use by changing fromflood irrigation to drip irrigation if drip irrigation isnot already being practiced. Drip irrigation canreduce water used in agriculture, and it typicallyincreases crop yields as well. However, that partic-ular reduced use of water used in agriculture willusually be accompanied by increased use of otherscarce resources, most notably labor, capital, andmoney. Unless crop yields increase considerablyunder drip irrigation and unless water is priced muchhigher than normally seen in the irrigated west,changing from flood to drip irrigation on a given acre-age typically reduces net farm income. Irrigators aretypically in business to increase their income or tomeet other personal or farming goals, not to savewater.

Only actions that reduce the use of water withoutdisproportionately increasing the use of other resour-ces can be labeled as water-conserving in the econo-mic sense. So an acceptable definition of waterconservation requires that the beneficial effects of thereduced water use must be greater than the adverseeffects associated with the use of other resourcesrequired to support conservation. Where all beneficialand adverse effects are measurable in monetaryunits, this test amounts to the requirement that thebenefits of reduced water use exceed its costs. Whatthis means is that the essence of conservation isreduced use, but more must occur. A water manage-ment practice constitutes conservation in the econo-mic sense when it meets two tests: (1) it saves agiven supply of water through reduction in water con-sumed, and (2) measures taken to reduce water con-sumed produce a net increase in society’s economicwelfare. For example, increased labor and greatercapital to support drip irrigation have a lower econo-mic cost than the value of the water saved and madeavailable for other uses.

The first test insures that the conservation practiceresults in a reduction in consumptive use, while thesecond establishes that overall benefits exceed costs.This definition of water conservation is simply a spe-cified subset of those practices that comprise econom-ically efficient management of water resources. Whenconservation is thus differentiated from other desir-able water management measures, it becomes poss-ible to formulate policies and propose practices thatare directed to promote conservation; and it becomes

possible to evaluate their success on economicgrounds. It should be noted that this article’s defini-tion raises the bar that must be vaulted before a pol-icy is said to promote water conservation. It requiresthat a water-conserving policy produces a reductionin use for which its benefits exceed its costs. Whileconventional welfare economics would likely use theterm ‘‘economically efficient water conservation,’’ thisarticle opts for the simpler term ‘‘water conserva-tion.’’ This definition is chosen because so many mod-ern water conservation policies have had limitedsuccess in promoting water use reductions becausethey ignored the all-important economic incentivesaffecting water use.

Both New Mexico and Texas water laws considermeasures to save water. While neither states’ lawsdefine water conservation directly, they illustrate thekinds of actions that could be considered water-con-serving. New Mexico law states that the office of thestate engineer (OSE) must consider water conserva-tion when reviewing an application for water rights.Issued water rights permits include a water conserva-tion condition that states that the permittee shallutilize the highest and best technology available toensure conservation of water to the maximum extentpractical.

In 1985, the Texas legislature recognized that con-serving water may be a cheaper method for avoid-ing shortages than developing new supplies. So itrequired that conservation be a factor in granting ordenying a permit. Specifically, conservation consider-ations in Texas include ‘‘those practices, techniques,and technologies that will reduce the consumption ofwater, reduce the loss or waste of water, improve theefficiency in the use of water, or increase the recyc-ling and reuse of water so that a water supply ismade available for future or alternative uses’’ (TexasWater Code, 1985a). ‘‘Applicants for new or amendedpermits must formulate a conservation plan, whichdemonstrates that reasonable diligence will be usedto avoid waste’’ (Texas Water Code, 1985b).

Both states’ laws suggest that water conservationis defined as a simple reduction in use. New Mexicoemphasizes best technology for reducing use, whileTexas place greater emphasis on reduced use througha variety of measures. Despite the good intentions ofboth states’ laws, neither emphasizes a comparison ofthe costs of resources required to achieve the reduceduse compared with the value of the water for whichuse is reduced. Water conservation laws will have agreater likelihood of being accepted, supported, andenforced where those laws take explicit account ofthe economics of water conservation. To summarizethis section, the definition of water conservation usedfor this article is any reduction in water use that pro-motes economic efficiency or for which benefits exceed



costs. Defining conservation in this way leads to eco-nomically sound water-conserving programs. It is alsodesigned to avoid programs that save low-valuedwater at the cost of using more expensive resourcesto save that water.

BARRIERS TO CONSERVATION

This section identifies several institutional barriersthat can limit the effectiveness of future water con-servation programs in the Rio Grande Basin. Remov-ing or alleviating any of these barriers could promotereduced use economically efficiently, thus freeing upwater for alternative uses or providing more waterfor future use in agriculture, cities, or the environ-ment.

In this article, the various institutional barriersdescribed below are first enumerated then described.It is anticipated that future research will measureand rank the importance of each institution. If thereis a way to objectively measure the intensity of eachinstitutional barrier to conservation, it would be quiterevealing to perform a quantitative analysis. Forexample, an econometric analysis could be performedwith the level of reduced use as the dependent vari-able and the various measured levels of each of thevarious institutional barriers as independent varia-bles. If such a model could be estimated, then thecomparative size of the coefficients on each barrierwould tell something about the comparative import-ance of each institution.

Water Transfer Restrictions

Short-term water transfers through mechanismslike water banks could provide an economic incentivefor agriculture to save water, especially in periods ofdrought. Temporary water transfers, such as a one-season water rental or leasing arrangement, orthrough an arrangement similar to banking, couldprovide agricultural producers an incentive to reducewater use. Permanent transfers are even moreattractive to municipal and industrial users; utilitiesresponsible for municipal supplies and operation ofwater treatment plants need a predictable and con-tinuous water supply. The advantages of short-termtransfers include the immediate infusion of cash intoagriculture as well as benefits to cities and the envi-ronment, both of which could include potentiallylarge elements of consumer surplus.

A water bank is a special form of a spot marketorganized and operated by a central banker, such as

the state, a state-appointed water broker, irrigationdistricts, or private companies. The bank, if estab-lished, is a mechanism for water right owners tolease water to the bank or renters, such as cities oran environmental group. A bank can typically acquirewater in at least three ways: by paying farmers forwater they would have used to irrigate their fields,by purchasing surplus water from local irrigationdistricts, or by paying farmers to use ground waterinstead of surface water where the two sources arehydrologically independent. Where the two sourcesare interdependent, conjunctive use management canhelp implement a water bank for which the benefitsexceed the costs. There are several excellent sourcessummarizing the economic theory and applications ofconjunctive use management (Burt, 1964; Provencherand Burt, 1994; Knapp (1995); Schuck and Green,1998, 1999, 2002, 2003).

A successful water bank experiment in Californiain the early 1990s taught several lessons: Water mar-kets, even when they are severely constrained andcontrolled, can work. Water can have a very highvalue for city and environmental buyers, and at asuitably high price, there are likely to be many sel-lers. Very large amounts of water can be found forthe bank if money is put on the table; and third-partyinterests in water market transactions can be protec-ted (Dziegielewski et al., 1993; California Departmentof Water Resources, 1992; Pratt, 1994). This success-ful experiment suggests that this market approachhas a considerable potential for dealing with theinstitutional barrier of water transfer restrictions.This market approach appears to be a measure withgreat potential to produce benefits exceeding costs.

Still, it is important to guard against the view thatwater marketing provides costless incentives toreduce water use. Farmers who switch from surfacewater to ground water may be pursuing a usefulstrategy for short-term management during dryyears, but this switch is not always sustainable inthe long run. While the California Water Bank wassuccessful as a whole, there is evidence that at leastsome of the farmers who sold their water to theCalifornia Bank substituted the use of ground waterfor surface water and continued to grow low-valuecrops such as alfalfa. Thanks to an anonymousreferee for this insight.

Some Rio Grande Basin irrigators have expressedconcern that these short-term transfers may be inter-preted by water administrators or by the public asevidence of a nonbeneficial use of water. Moreover,some producers may fear that their water rights willbe lost through a temporary transfer into a bank. Thiscould occur because of unclear or un-enforced legisla-tion or poor communication by water administratorsto producers. As a result, this important potential



source of saved water risks being unavailable unlessspecial steps are taken. This potential barrier to con-servation has been recognized by legislatures in anumber of states, including Washington, Oregon,Colorado, and California. These states have enactedlegislation protecting water right owners againstforfeiture of the right by stating that temporary trans-fers do not violate the principle of beneficial use.As such legislation can be accomplished at a com-paratively low cost, it has the potential to be econom-ically attractive for overcoming water conservationbarriers.

Illusory On-Farm Water Savings

Agricultural producers continue to adopt moretechnically efficient irrigation methods to producehigher net incomes through increased crop yields,increased efficiency in nutrient and chemical use,reduced labor costs, and more efficient water use.One definition of on-farm irrigation efficiency is theratio of water stored and depleted in the crop rootzone for crop consumption to the total water divertedfrom the stream for irrigation. One method toincrease on-farm efficiency, defined in this way,would be to encourage producers to apply water moreconsistently across fields, which enables crops tomaintain their consumptive water needs even thoughstream diversions are reduced.

Many policy makers believe that reduced diver-sions resulting from increased on-farm efficiency pro-duce water savings that become available to meetother growing demands. Some states across the Westare passing or are considering passing legislationthat encourages producers to invest in improvedon-farm irrigation technologies. However, this kind oflegislation should be approached carefully, becausemany of these on-farm investments in greater irriga-tion efficiency can reduce the available water thatwould have been otherwise supplied through returnflows to downstream appropriators (Huffaker et al.,1998; Huffaker and Whittlesey, 2000).

An on-farm investment that reduces the producer’sapplied water by X acre-feet can reduce downstreamsupplies by as much or more than X. This counterin-tuitive result occurs because of the potential forincreased water consumption (depletion) caused by achange to an irrigation technology that reduces waterapplied. For example, a producer who switches fromflood to drip irrigation diverts and applies to her fieldX acre-feet less than before. However, this action,while reducing water applied from the adopter’s view,can reduce return flows by more than X because dripirrigation typically increases crop yields throughhigher evapo-transpiration (ET) or greater consumpt-

ive water use. Consider the use of these concretenumbers in which the change in irrigation technologyreduces applied water from 4 to 2 acre-feet per acre(1 acre = 4047 m2). Crop water use (ET) couldincrease from 1 acre-foot per acre under flood to 2under drip. With no other losses, the overall effectcould be reduced return flows to the stream from 3under flood to zero under drip. The result is reducedwater application leading to increased water deple-tions.

Changes in irrigation efficiency make no newwater. They only change the flow patterns betweenriver, storage, and aquifer. In the long run, onlyreduced consumptive use can deliver more waterdownstream. A policy measure that guards againstfalse water savings can increase the benefits ofreduced use by more than the costs by encouragingthe right on-farm investments. Those investmentswill encourage on-farm applications that reduce netdepletion to the basin (Huffaker and Whittlesey,2000).

Insecure Rights to Conserved Water

One way to recognize water conservation is whenan irrigator reduces the use of water that would beotherwise irretrievably lost to the system. Somestates refer to this as ‘‘salvaged’’ water, i.e., watersaved that takes no wet water (as opposed to waterrights or ‘paper water’) from anybody else either cur-rently or in the foreseeable future. Water is irretriev-ably lost to the system when it is depleted throughuptake by plants, evaporation, runoff to salineground-water basins or to aquifers too deep for econo-mic use. Water conservation, under these conditions,occurs if the use of a particular stream or otherwatercourse or supply source is saved from loss andmade available for current or future beneficial use.One way to demonstrate that water is conserved is toshow that the reduction does not damage otherappropriators on the same watercourse (Huffaker andWhittlesey, 2003). For example, a producer who inter-cepts his return flows from a field that would other-wise flow unimpeded to a downstream user fails toconserve water according to the definition used inthis article. This water saved would not otherwise belost, as it takes water from another downstreamappropriator.

In both New Mexico and Texas, a person or organ-ization must apply to a state administrative agencyto obtain a valid water right. In both states, there isa constitutional requirement that the appropriatormust show to the state’s satisfaction that the waterwill be applied to beneficial use. In both states,within the Rio Grande Project, downstream of Ele-



phant Butte Reservoir (see Figure 1), the water rightpermit is granted in perpetuity or until the landand ⁄ or water has been transferred to another user.After the permit is granted, the appropriator isrequired to put the water to beneficial use. Neverthe-less, producers often express the fear that invest-ments made in reduced water use, such as changingirrigation technologies that reduce water applied oradopting new management techniques like irrigationscheduling, will result in the saved water being lostto the state or to the irrigation district, because ofthe presumption that the saved water was not usedbeneficially.

There are many ways to reduce water use. Con-crete-lined canals or ditches, for example, preventwater from seeping to uneconomical depths or to sal-ine aquifers. Other ways include removing water-using weeds (phreatophyes) to decrease water lost tononbeneficial uses or substituting water stored insurface reservoirs to shallow ground-water basins.Institutions that block producers from securing aclear title to a water right for the reduced water usein this way discourage investments in water sav-ings.

The doctrine of prior appropriation, typical in thewestern states, blocks incentives to reduce water usebecause water rights protected under the law are lim-ited to the amount of water that is diverted and putto beneficial use. The water right owner has noincentive to limit use through water-saving measures.For example in the case of Southeastern ColoradoWater Conservancy District v. Shelton Farms, theColorado Supreme Court found that appropriatorscannot sell salvaged water (Colorado Supreme Court,1974). This legal decision damages economic incen-tives to make investments that save water. In recog-nizing this problem of weak conservation incentives,several western states have taken steps to remedythe situation. California and Oregon have assignedthe right to use conserved water to the person whoimplements the conservation measure. Court deci-sions in Utah and Colorado have produced the sameresult. In Texas, as stated above, water conservationmust be considered when granting permits for pro-posed new water uses. Beneficial uses in Texasinclude the following water uses: domestic and muni-cipal uses, agricultural uses and industrial uses,mining, hydroelectric power, navigation, recreation,public parks, and game preserves (Texas Water Code,1997).

Shared Carry-over Storage

Irrigators sometimes express support for a policythat permits or encourages them to carry over and

keep track of this year’s unused water, so it can bereleased from a reservoir and used in a subsequentyear. Under the Rio Grande Project in southern NewMexico and West Texas, water users are discouragedfrom saving water in any given year and storing it atElephant Butte Reservoir for later use. High evapor-ation, which causes considerable losses to water car-ried over, and limited reservoir storage space atElephant Butte (stored water may displace futureinflows to the reservoir) are two reasons why littlecarry over storage is seen. Several preventable lossesoccur when water is released from a reservoir andused for irrigation: part of the water is consumed byevaporation, a portion percolates to the aquifer andthe drainage water quality is sometimes impaired bysalts or chemicals. If a system of carry-over storagecredits could be enacted with property rightsassigned to those who reduce their current water useby fallowing land, adopting water-saving irrigationtechnology or shifting to lower water-using crops,these losses could be reduced. In drought years,actions such as these that saved water would be espe-cially valuable in the future, thus contributing to theregion’s water conservation.

The common property nature of an irrigator’ssaved water in Rio Grande Project area of southernNew Mexico and West Texas, combined with the his-torical 57% water allocation to New Mexico usersand 43% to Texas users, means that any water thatone state carries over into the region’s storage thisyear is shared by both states the next year. Forexample, suppose a Texas user reduces current useby 1,000 acre-feet, the irrigation district can thensave it in Elephant Butte Reservoir to deliver in thefollowing year. The un-evaporated part of the 1,000acre-feet saved by the Texas user will accrue as 43%to the Texas irrigation district and 57% to the NewMexico irrigation district. Subsequently, the watersaved will then be allocated, not to the individualwho saved the water but, pro rata to all users ineach district. The fact that a well-defined, transfer-able and enforceable private property right fails tobe earned in water carried over discourages irriga-tors from conserving water. Assigning Texas prop-erty rights to the irrigation district and ⁄ or toindividuals in Texas who save water and New Mex-ico property rights to the New Mexico irrigation dis-trict and ⁄ or to individual water savers could be alow-cost measure for dealing with this issue, thuspromoting water conservation through increasedsecurity of tenure.

There may be other cost-effective measures fordealing with shared carry-over storage. For example,the Snake River water bank (Idaho Water SupplyBank, 2005) has established a successful policy forpromoting shared carry-over capacity such that irri-



gators can deposit water now and later withdraw it.In 1980, the State Legislature of Idaho established awater bank that authorized water rental agreementsthat had been in place informally for many years.The legislation assured the owners of stored waterthat seasonal deposits into the water bank would notjeopardize their rights to later withdraw the water.In this bank, federally supplied water is obtainableby holders of contracts for water from the Bureau ofReclamation reservoirs on the Upper Snake River.Many water right holders keep large water holdingsunused as insurance against drought. The Bureau ofReclamation offers US$2.50 ⁄ AF for water to be leasedfor 1 year and in turn rents this water to others inthe basin who are willing to pay to put it to beneficialuse, typically other irrigators and the Idaho PowerCompany. Thus, by being paid to deposit unusedwater into the bank, the market institution has copedwell with the potential for wasted water in the face ofshared carry-over storage.

Interstate Compacts

The Rio Grande Compact and the 1906 U.S.-Mex-ico Treaty are the overriding mechanisms for alloca-ting water to Colorado, New Mexico, Texas, and theRepublic of Mexico in the upper part of the basinabove Fort Quitman, Texas, about 100 miles south-east of El Paso, Texas. The quantity of water alloca-ted to each is set out clearly within the compact andtreaty allocations with little opportunity to tradewater surpluses or shortfalls for cash or other con-siderations. The compact is the most important insti-tution governing the allocation of water among thethree basin states. In basin above Fort Quitman,Texas, water is managed to comply with the com-pact. Colorado’s water deliveries to New Mexico atthe Colorado-New Mexico state line are a function ofheadwater flows produced by Colorado’s snowpackrunoff. All water not delivered to New Mexico isavailable for use by Colorado. Water that New Mex-ico delivers to Texas at Elephant Butte Reservoir,measured at the gauging station just above Ele-phant Butte, is a function of annual flows at theOtowi gauge above Santa Fe, excluding interbasintransfer San Juan-Chama River flows. So flows inNew Mexico delivered to the Elephant Butte gaugeunder the Compact are based on native flows at theOtowi gauge. In very wet years, when New Mexicodoes not have the capacity to use its full compactallocation, New Mexico may receive an annual creditof up to 200,000 acre-feet for its over delivery toTexas. In dry years, New Mexico may underdeliverto Texas by an amount not to exceed 150,000 acre-feet, and an annual debit is incurred in such cases.

New Mexico, under the compact, may accrue totaldebits, offset by wet year credits, of up to a total of200,000 acre-feet.

While the compact has proven to be a lasting andrespected institution for the basin’s water allocation,it currently provides for no institution that wouldpermit water users in Colorado or New Mexico to sellor rent surplus water to users below Elephant ButteReservoir or to buy surplus water from these users.If, for example, the compact were amended to allowColorado or New Mexico users to under-deliver toElephant Butte in exchange for cash (buy water) orover deliver to Elephant Butte Reservoir in exchangefor cash (sell water), agricultural users in all threestates might be encouraged by cash incentives to con-serve water. Mexico is allocated 60,000 acre-feet peryear under the 1906 U.S.-Mexico treaty, an amountof water that is not normally subject to negotiation. Ifirrigators in southern New Mexico or West Texascould sell or rent some of their unused water to Mex-ico in exchange for cash, the associated financialincentive might encourage all users to invest in wateruse reduction measures, thus providing incentives forwater conservation where the total benefits of thereduced use exceeds total costs.

Conservation Attitudes

Negative attitudes toward reduced water use canrepresent a major barrier to water conservation evenwhen the benefits of reduced use exceed its costs. Tounderstand more about these attitudes and the barri-ers they form to water conservation, a survey ofwater management practices was administered in2002 and again in 2003 to members of the ElephantButte Irrigation District (EBID), New Mexico. Thesurvey was designed to identify attitudes that dis-courage irrigators from reducing water use. Table 1shows irrigated acreage for the producers sampled bythe survey for both 2002 and 2003 by crop and byirrigation technology used.

Table 2 shows that the top reason for not redu-cing water use is a lack of buyers for saved water.The lack of buyers means that irrigators perceiveweak incentives for reducing water use. Theirresponse is telling policymakers: ‘‘Why should weconserve water if there is nobody to buy ourwater?’’ Furthermore, if there are no buyers, a mes-sage is signaled that there is little social value inreducing water use. So they receive a message thatthey might as well continue to use water-intensiveirrigation methods. Certainly, the development of awell-functioning water trading system depends inpart on willing sellers and buyers, otherwise notrades will occur.



Table 2 also shows that a strong majority of pro-ducers identified several barriers to water conserva-tion, defined here as a simple reduction in use:conservation is too expensive, the basin’s streamadjudications are not yet complete, additional labor isrequired to implement conservation, there are inad-equate financial incentives for reducing water use,there is an inadequate water distribution system, andthere are increased soil salts resulting from reducedwater use.

Land Tenure

Most farm rentals in the area are based on a crop-sharing arrangement, but renters typically pay a flatfee for the right to farm a set acreage. Under non-drought conditions, the first 2 acre-feet per acre aretypically a part of the lease, essentially a fixed cost.The fee is priced to pay for the District’s operationscosts, including the water that comes with it.Recently, EBID implemented a declining block ratewith additional amounts of project water available ata lower price, with the intent of providing incentives

to avoid pumping ground water. Additional surfacewater, if available at the reservoir, can be bought forUS$18 per acre-foot, so that water is a variableproduction cost. During the survey, several rentersdescribed the lack of financial incentives for water-conserving measures. For example, if a land tenantonly has a year-to-year agreement with the landow-ner, that tenant is unlikely to incur large costs inexpensive water use reduction measures, such as dripirrigation, which require several years to pay back.One measure for dealing with the problem might befor owners to offer land tenants a greater stake inwater use reduction, thus tenants would share thenet benefits produced by water use-reducing meas-ures.

Uncertain Duty of Water

Many Western states, including New Mexico, arein the process of adjudicating their streams by defi-ning clear titles to water rights. A completely adjudi-cated stream system clearly defines all owners’ rightsto use water under all possible future hydrological

TABLE 1. Irrigated Acreage by Crop and Irrigation Method, Lower Rio Grande Basin, New Mexico.1

Year 2002 2003

Crop ⁄ Method

Flood Drip Flood Drip

Acres % Total Acres % Total Acres % Total Acres % Total

Alfalfa 6,302 16.02 220 0.56 5,683 15.53 270 0.74Cabbage 3,110 7.90 310 0.79 2,670 7.30 360 0.98Chile 6,855 17.42 690 1.75 5,841 15.96 690 1.89Corn 2,729 6.94 70 0.18 2,544 6.95 120 0.33Cotton 5,979 15.20 480 1.22 5,206 14.23 530 1.45Lettuce 3,007 7.64 310 0.79 2,582 7.06 360 0.98Onion 4,755 12.09 310 0.79 5,250 14.35 620 1.69Pecans 3,077 7.82 210 0.53 2,699 7.38 262 0.72Wheat 880 2.24 50 0.13 800 2.19 100 0.27Total acres 36,694 93.26 2,650 6.74 33,275 90.95 3,312 9.05

1Source: Ward et al. (2004).

TABLE 2. Water Conservation in Agriculture by Land Tenure Arrangement, Rio Grande Basin, New Mexico, 2004.1

Land Tenure Arrangement Own Land (N = 67) Rent Land (N = 20)

Reasons for not reducing water useWater conservation costs too much 692 60Incomplete stream adjudications 88 75Water conservation requires additional labor 76 70No buyers for saved water 96 100No financial incentive to reduce water use 85 65Water distribution system restricts conserving 90 93Reduced water use builds up salts in soil 79 67

1Source: Ward et al. (2004).2Entry reflects percentage of all respondents indicating agreement with statement.



conditions. General stream adjudications fulfill twofunctions: (1) public recording and validation of allwater claims and rights, (2) facilitating a clear set ofrules for allocating water in periods of shortfall.

Stream adjudications give certainty to waterrights, provide the basis for water right administra-tion, reduce conflict over water allocation and waterusage, and facilitate important market transfers forwater rights. Most of the stream segments in the RioGrande basin are still not adjudicated, which meansthere is considerable uncertainty over who currentlyhas the right to use how much water in what watersupply conditions. One problem presented by thislegal uncertainty over who owns what water rights isthat water authorities have a difficult time adminis-tering water rights (e.g., locking gates of junior users)to guarantee sufficient downstream flow to meetinterstate compact obligations. It is unclear who thejunior users are.

Adjudication began in earnest in New Mexico’sLower Rio Grande in the late 1990s, with the firstoffers of adjudicated water rights sent to landownersin 2000. Despite the considerable progress made onthese adjudications, the ‘‘duty of water,’’ or theamount of water right assigned per acre, has yet tobe established. The duty of water has little to do witha moral or legal duty. The concept was established tolimit the amount of water that may be diverted undera priority system, and was designed to prevent waste.It is the relation between the area to be irrigated andthe quantity of water required to irrigate it for thepurpose of maturing its crop. Of course, what isrequired depends heavily on the price of water, thecrop grown, soils, and on the price of various water-using technologies. Where water has a low price andwhere water-conserving technologies are expensive,the amount of water required to irrigate an area eco-nomically will be quite high indeed.

There is considerable uncertainty over what theduty of water will be or how it will be established.Will all irrigators receive an equal amount of adjudi-cated water per acre, for example, 3 acre-feet acre for

every irrigator? Or will the offer vary with type ofcrop? For example, pecan growers could receive morewater rights per acre than cotton growers because ofthe greater water applied historically per acre topecan trees. Tied to this is uncertainty of groundwater adjudication. The question centers on whetheror not water rights offers will be defined on combinedrights to surface and ground water use. For NewMexico producers, ground water is an importantsource of water during drought, but it is also usedwidely in normal years. Water allocations per acrecould be made based on an efficiency criterion definedby the relative economic value of water. Using theefficiency criterion for defining a water right per acrecould be accomplished by defining duties of water insuch a way that the marginal value of the last acre-foot consumed is equal for all crops (1 acre-foot = 1233 m3).

This uncertain duty of water prior to the completedadjudications may establish perverse incentives forwater conservation: If there is widespread belief thatproducers who plant more water-using crops willsecure a larger adjudicated offer per acre, growerswill have an incentive to plant crops or trees that uselarger amounts of water solely to receive a favorablehigher future assigned property right.

Despite the long run increase in certainty targetedby the ongoing adjudications in the basin, their currentincomplete status creates a significant institutionalbarrier to incentives for conserving water, a hypothesisborne out by Table 3. The table presents results of thesurvey organized by major crop. Crops listed are onion,pecans, alfalfa and cotton. Onion farmers were especi-ally sensitive to the incomplete status of the ongoingadjudication process; 67% stated that the adjudicationprocess discourages them from conserving water. Pro-ducers of other crops were somewhat less sensitive touncertainties caused by the ongoing adjudications.Despite the greater water-conserving incentives, theseadjudications are typically expensive, both in terms oflegal as well as technical expertise. As a measure forpromoting increased water conservation, stream adju-

TABLE 3. Water Conservation in Agriculture by Major Crop, Rio Grande Basin, New Mexico, 2004.1

Most Important Crop Onions (N = 3) Pecans (N = 35) Alfalfa (N = 38) Cotton (N = 19)

Reasons for not reducing water useWater conservation costs too much 672 69 74 63Incomplete stream adjudications 33 91 89 68Water conservation requires additional labor 67 83 76 68No buyers for saved water 67 100 97 95No financial incentive to reduce water use 67 86 87 63Water distribution system restricts conserving 100 82 94 94Reduced water use builds up salts in soil 67 71 81 82

1Source: Ward et al. (2004).2Entry reflects percentage of all respondents indicating agreement with statement.



dications are unlikely to be the most cost-effectiveinstitution for finishing the job, especially in the two-to-four year time horizons that influence most electedrepresentatives.

ECONOMIC FACTORS AFFECTING WATER USE

The price of water, as well as the price of meas-ures that promote reduced water use, are importantindicators of water’s economic scarcity. Economic the-ory suggests that producers will pay close attentionto both sets of prices. Several studies have found thatthe price of water as well as the price of water-savingirrigation technology have a significant major influ-ence on the adoption of water-saving measures(Caswell and Zilberman, 1985, 1986; Caswell et al.,1990; Moore and Negri, 1992; Green et al., 1996;Green and Sunding, 1997).

Effective Price

Various institutions set both the price of water andestablish the rules governing how water can be tra-ded both inside and outside agriculture. These insti-tutions potentially have an important influence onwater-use patterns. For example, each EBID memberis charged US$50 per acre per year for the right touse up to 2 acre-feet per acre, if there is enough res-ervoir water available. Any water user who does notuse the 2 acre-foot allotment still pays the US$50even if the land is left fallow. For this reason, theUS$50 charge is a district membership charge andnot a price for incremental quantities of water used.Because any increase or decrease in water usebetween 0 and 2 acre-feet per acre results in the pro-ducers paying the same US$50 price, the first 2 acre-feet per acre are effectively priced at zero.

EBID members also have the right to purchaseadditional water at US$18 per acre-foot up to theamount of water that is available, so the incrementalprice after two acre-feet is US$18. If policies wereinstituted that allowed members to buy each acre-footafter two at US$18, then rent any unused portion ofit out at US$100 or even US$200 per acre-foot to acity or recreational or industrial buyer, there wouldbe considerable financial incentive to invest inon-farm water-saving measures. However, currentwater transfer practices do not permit trading ofwater outside agriculture, so the effective marketprice of water received from non-agricultural buyersis at or near zero. Thus, water is effectively lockedinto agriculture, which discourages investments in

water-saving measures and raises the price of waterto city or environmental users since they will have tosearch for new supplies.

The Special Role of Price

Section 3 described several institutional barriers toreduced water use, all of which have economic impli-cations. Yet, with increased population, growing envi-ronmental demands, and other water demandsoutside the agricultural sector, there is continuedinterest in and debate surrounding the design of cost-effective measures for promoting increased water-useefficiency in agriculture. Answers to the followingthree questions could provide useful insights toinform this debate: (1) what are the economic costsresulting from institutional barriers to the adoptionof water-saving measures? (2) what kinds of actionscan producers be expected to take to deal with watershortages when there are no barriers to adoptingwater-saving measures? (3) what are the economiccosts of subsidy programs designed to maintain farmincome levels while reducing agricultural water use?

The question of what changes it would take to pro-mote investments in reduced water use, defined hereas allocating some acreage into drip irrigation, takeson considerable economic and political importance.

Water price reforms are increasingly used toencourage improvements in irrigation efficiencythrough technology adoption. However, because ofthe political sensitivity of price reform, debates centeron the effectiveness on irrigation technology choicesresulting from economic factors like price v. those ofenvironmental characteristics and institutionaladjustments (Green et al., 1996).

As is typical throughout the world, drip irrigationin the Lower Rio Grande Basin is considerably moreexpensive than flood irrigation. It also requires lesswater applied per acre and produces greater cropyields. The answer to the question of whether or notdrip irrigation is economically attractive to irrigatorsturns on what combination of economic and watersupply conditions make it profitable to favor drip overflood irrigation. As of 2005, virtually all EBID pro-ducers use flood irrigation. Because Tables 1-3 pre-sent few conclusions with rigorous analytical content,the analysis turns to a linear programming approachdesigned to provide insight into the economic andinstitutional forces promoting or discouraging water-use reductions.

Table 4 shows economic conditions for a represen-tative irrigated onion farm in the lower Rio GrandeBasin. A comprehensive treatment of agriculturalimpacts in Table 5 would require accounting for amajority of the region’s acreage: pecans, several kinds



of vegetables, alfalfa, cotton, and grains. The analysisshown in the table includes only a single crop, onions.While this is a simplification of reality, it has theadvantage of offering clear insight into irrigatorbehavior. It also provides an understanding ofimpacts on farm behavior resulting from input andoutput prices, yields, production costs, and crop waterrequirements.

Table 5 shows the impact of changes in water scar-city as well as impacts of a range of adjustmentsunder four possible sets of production cost and cropprice conditions. Rows 1-4 show current conditions,consisting of high production costs (US$4120 ⁄ acre forflood irrigation) and low crop prices (US$6.38); Rows5-8 show conditions of high production costs and highcrop prices (US$7.38); Rows 9-12 show low productioncosts (US$3820 ⁄ acre for flood irrigation) and low cropprices (US$6.38). Rows 13-16 show low productioncosts and high crop prices. Drip irrigation is priced ata constant US$5320 per acre except where it isreduced through public subsidy.

For each set of these three future economic condi-tions, the table shows three possible irrigatorresponses to a water supply reduction from 300 to200 acre-feet for the 100 acres. In all cases, the pro-ducer is presumed to maximize net income subject toconstraints of water supply and available land. Thefirst scenario is one of the continued income maxim-ization in which the producer is constrained frominvesting in any additional drip irrigation acreagebeyond that, which occurred under the unconstrainedincome maximization produced by a water supply of300 acre-feet. The second scenario is continuedincome maximization with no institutional barriersthat constrain the irrigator from investing in dripirrigation.

The third scenario assumes the presence of a gov-ernment subsidy program designed to compensate forthe effect of a reduced water supply. In this scenario,the program’s aim is to design a water use reduction

subsidy that saves a known amount of water at mini-mum financial cost to the taxpayer while maintainingfarm income at least as high as income earned underfull water supply. To simulate this program, a modelis developed in which the producer’s net cost per acreof drip irrigation is reduced by the amount of the sub-sidy. Because the taxpayer-cost-minimizing subsidy isnot known in advance, the model’s objective minim-izes the financial cost of a drip irrigation subsidy thatsaves 100 acre-feet of water. It also produces a farmincome equal to or larger than that earned under thebaseline 300 acre-foot supply. Results of incomemaximization provide insights into factors that mostinfluence irrigators’ investments in water-use reduc-tion.

The decision to invest in drip irrigation raisesimportant issues in connection with capital invest-ment and with uncertainty. The producers decision toinvest in drip irrigation or the taxpayer’s decision tosubsidize it is made for many years into the future.The decision involves not only current output pricesand costs, but an irrigator’s expectation about futureoutput prices and costs. For example, a subsidy onthe adoption of drip irrigation might have be paidevery year, not just the year of adoption, to cover theadditional costs to the farmers. If these subsidies arepaid for a longer term, the costs will continue to behigher than those of flood irrigation. Research hasshown that uncertainty about future relative pricesposes a major impediment to the adoption of conser-vation irrigation technology. To deal with these com-plexities, Tables 4 and 5 treat all costs of dripirrigation to both producers and to taxpayer-subsidiz-ers as added costs in annual equivalent terms com-pared to those of flood irrigation. The authors thankan anonymous referee for this observation.

Table 5 shows that net farm income is lowestunder current conditions where production costs arehigh and crop prices are low (Row 1). Under these2005 conditions, abundant water (300 acre-feet) andlow crop prices make it economically attractive tosubstitute cheap water for land and for other inputs.Some drip irrigation could be put onto the 25 acres ofidled land, shown in Row 1. But at US$6.38 the cropprice is too low for the value of the additional yield topay for the additional annual equivalent costs of driptechnology. Row 1 shows that income maximizationunder initial conditions requires idling 25 of the 100available acres. All 75 of the acres under productionare produced under flood irrigation, a fact that waslargely observed in 2005. The value of one additionalacre-foot of water is US$47, and the farm produces atotal net income just under US$14,000. When watersupply falls to 200 acre-feet, irrigators simply reducethe scale of farming from 75 to 50 acres. No invest-ments in drip irrigation are made because the added

TABLE 4. Economic Conditions for an Irrigation Farm, Lower RioGrande Basin, New Mexico, 2005.1

Water supply 300 acre-feetCrop OnionsFarm size 100 acresCrop price US$6.38 per sackFlood irrigation

Production cost US$4,120 per acreCrop yield 675 sacks ⁄ acreWater use 4 acre-feet ⁄ acre

Drip irrigationProduction cost US$5,320 per acreCrop yield 845 sacks ⁄ acreWater use 2 acre-feet ⁄ acre

1Source: Libbin and Hawkes (2005).



TA

BL

E5.

Fact

ors

Aff

ecti

ng

Wate

rU

sein

Irri

gate

dA

gri

cult

ure

,L

ower

Rio

Gra

nd

eB

asi

n,

New

Mex

ico,

2005.

Su

pp

ly(a

-f)

Inst

itu

tio

ns

Inp

ut

Co

sts

Pric

eL

an

dA

llo

ca

tio

nW

ate

rA

llo

ca

tio

nE

co

no

mic

Imp

act

Su

bsi

dy

Ba

rrie

r1

Flo

od

2

(US

$⁄a

c)

Drip

3

(US

$⁄a

c)

On

ion

s(U

S$

⁄sa

ck

)F

loo

d(a

c)

Drip

(ac)

Idle

d(a

c)

Av

era

ge

Use

(a-f

⁄ac)

Idle

d(a

-f)

Wa

ter

Va

lue

(US

$⁄a

-f)

Net

Inco

me

(US

$)

Pro

gra

mC

ost

(US

$)

300

4N

oN

o4,1

20

5,3

20

6.3

875

025

4.0

00

47

13,9

88

0200

No

Yes

4,1

20

5,3

20

6.3

850

050

4.0

00

47

9,3

25

0200

No

No

5,3

20

50

050

4.0

00

47

9,3

25

0200

Yes

No

5,2

05

25

50

25

2.6

70

47

13,9

88

5,7

70

300

No

No

4,1

20

5,3

20

7.3

80

100

02.0

0100

091,6

10

0200

No

Yes

5,3

20

0100

02.0

00

091,6

10

0200

No

No

5,3

20

0100

02.0

00

091,6

10

0200

Yes

No

5,3

20

0100

02.0

00

091,6

10

0300

No

No

3,8

20

5,3

20

6.3

875

025

4.0

00

122

36,4

87

0200

No

Yes

5,3

20

50

050

4.0

00

122

24,3

25

0200

No

No

5,3

20

50

050

4.0

00

122

24,3

25

0200

Yes

No

4,9

05

25

50

25

2.6

70

122

36,4

87

20,7

70

300

No

No

3,8

20

5,3

20

7.3

850

50

03.0

00

123

103,8

80

0200

No

Yes

5,3

20

25

50

25

2.6

70

290

74,8

42

0200

No

No

5,3

20

0100

02.0

00

123

91,6

10

0200

Yes

No

5,1

97

0100

02.0

00

61

103,8

80

12,2

70

ac:

acr

e;a-f

⁄ac:

acr

e-fe

et⁄a

cre;

a-f

:acr

e-fe

et.

1In

stit

uti

onal

barr

ier

pre

ven

tsp

rod

uce

rfr

omin

ves

tin

gin

dri

pir

rigati

onev

enif

itin

crea

ses

net

inco

me

orre

du

ces

loss

es.

2E

qu

als

cost

per

acr

eof

crop

pro

du

ctio

nu

nd

erfl

ood

irri

gati

onte

chn

olog

y;

crop

yie

ldis

675

sack

sp

eracr

e;w

ate

rco

nsu

mp

tive

use

is4

a-f

⁄ac.

Data

sou

rce

isL

ibbin

an

dH

aw

-k

es(2

005).

3E

qu

als

cost

per

acr

eof

crop

pro

du

ctio

nu

nd

erd

rip

irri

gati

onte

chn

olog

y;

crop

yie

ldis

845

sack

sp

eracr

e;w

ate

rco

nsu

mp

tive

use

is2

a-f

⁄ac.

Th

isla

rge

am

oun

tof

wate

rsa

v-

ings

pro

du

ced

by

shif

tin

gfr

omfl

ood

tod

rip

irri

gati

onm

ay

be

opti

mis

tic.

Th

eu

seof

twic

eas

mu

chw

ate

rfo

rfl

ood

irri

gati

onas

for

dri

pir

rigati

onis

un

reali

stic

for

man

yp

art

sof

the

cou

ntr

y.

Itis

pos

sible

for

dif

fere

nce

sin

ap

pli

edw

ate

r,bu

tas

des

crib

edin

the

art

icle

,is

less

lik

ely

for

wate

rco

nsu

med

from

the

basi

n-w

ide

vie

w.

Itis

imp

orta

nt

toco

nsi

der

an

dacc

oun

tfo

rre

charg

eth

at

occu

rsaft

erw

ate

ris

ap

pli

edto

afi

eld

,as

that

rech

arg

eco

ntr

ibu

tes

tod

own

stre

am

wate

rsu

pp

lies

an

du

se.

4T

he

top

row

refl

ects

2005

con

dit

ion

s.



revenue produced by the increased yields are toosmall to pay for the added cost of drip irrigation(Rows 2-3). Row 4 shows that a subsidy for drip irri-gation of US$115 per acre is enough to maintain ori-ginal income through the producer’s investment inthe drip technology even with 100 fewer acre-feet ofwater to apply.

Irrigators’ highest willingness to invest in watersavings occur when flood irrigation is expensive andcrop prices are high, shown in Row 5. The increasedcrop price shown by comparing Row 1 and Row 5 pro-duces a dramatic increase in the economic desirabilityof investing in drip irrigation. The higher crop priceenables the added physical yield gained through dripirrigation to pay for its higher production costs. Bycontrast, under low crop prices, the economic value ofthe added yield produced by drip irrigation is insuffi-cient to pay for its added production costs. Under con-ditions described by Row 5, drip irrigation becomes soprofitable that all 100 acres are produced under dripirrigation even without reducing the water supply.Furthermore, installing drip irrigation reduces wateruse by so much that its adoption results in 100 acre-feet of unused water. This water becomes availablefor uses outside agriculture, such as environmental orcities’ needs. For this reason, when agricultural watersupply falls from 300 to 200 acre-feet, there is zeroeconomic loss to agriculture. In these conditions, thecost of a water-saving program subsidy is zero, sincemaximum conservation already occurs. These resultsare shown in Rows 5-8.

The highest cost to the taxpayer of implementing aprogram to subsidize water savings occurs when floodirrigation is cheap and crop prices are low, as shown inRow 12’s US$20,770 total program cost. Producerresponses to these conditions are shown in Rows 9-12.In this situation, an on-farm water supply reductionfrom 300 to 200 acre-feet encourages no water conser-vation investments by irrigators, as shown by compar-ing Rows 9 and 10. A low crop price means that theadded cost of drip irrigation is larger than the econo-mic value produced by drip’s increased crop yield, sothe irrigator reduces water use by the required 100acre-feet by reducing the scale of the farm’s operationand idling 25 more acres of land (compare Rows 9 and10). These economic conditions, while weak at encour-aging drip irrigation investments, are precisely thesame conditions that cause the water-savings programsubsidy required of the taxpayer to be so high. Whenfarm income falls from US$36,487 to US$24,325 afterthe water supply reduction, the cost-minimizing dripirrigation subsidy needed to maintain farm income atthe US$36,487 base after the 100 acre-foot water lossis a high US$415 per acre.

Institutional barriers to water savings have thelargest effect on reducing net farm income in the face

of reduced water supply in conditions where floodirrigation is cheapest and crop prices are highest.Rows 13-16 show that the economic cost of the insti-tutional barrier to installing drip irrigation atUS$17,000 (US$91,610-US$74,842). That is, remov-ing the institutional constraint to drip irrigationsaves the irrigator about US$17,000. In the face ofinstitutional barriers to reduced water use (Row 14)in these most attractive economic conditions, the eco-nomic value lost by suffering a water supply shortfallof 100 acre-feet produces the highest of all losses,about US$29,000. This large economic loss producedby institutional barriers to saving water, not surpris-ingly, also produces the highest marginal economicvalue of water at US$290 per acre-foot. Remarkably,the economic cost of a subsidy designed to encouragewater savings is comparatively small (US$12,270shown in Row 16) compared to conditions describedin Rows 12 (US$20,770). It is smaller because irriga-tors can afford to invest in drip irrigation thanks to ahigh prices. Therefore, the drip irrigation subsidyrequired to restore base income (US$103,880) is onlyabout US$123 per acre, a total subsidy of US$12,270for the 100 acres. Moreover, an anonymous refereereminds us of an even potentially lower cost policyoption than the one described for Table 5. That is toprovide a lump-sum subsidy to farmers after reducingtheir water supply, instead of distorting the relativecosts of drip compared with flood irrigation. This pos-sibility would allow farmers to keep using flood irri-gation if it is more profitable, but still reduce waterconsumption from 300 to 200 AF per acre. Theauthors are indebted to an anonymous referee forthis insight.

The results above should be qualified by recogni-zing the importance of output price in the decisionabout irrigation choices. While few of the crops inthe Rio Grande Basin are government supportedcommodity crops, a considerable amount of irrigatedacreage in the west does produce these supportedcrops. The distinction between supported andunsupported crops is important for several reasons,including the establishment of expected futureprices and price volatility. In addition, the abovediscussion of commodity price, as well as the sce-narios presented in Table 5, should make it clearthat output prices as well as other costs are aresult of both markets and of policy actions thatinfluence those markets.

Price and Substitution

Ground-water substitution occurs when irrigatorsrespond to surface water price increases or shortagesby reducing surface water demand and tapping



instead into ground water. An unfortunate side effectof this is that ground-water substitution can lead toactions that reduce the use of one water resource atthe expense of another to which it is hydrologicallyconnected. As a result of the interdependencebetween ground-water and surface water use, it isdifficult to determine if a surface water pricing pro-gram promotes saved water from the view of the sys-tem. One water source is potentially saved at theexpense of the other. For this reason, the hydrologicand economic ease with which ground water is substi-tuted for surface water is important to understandand measure when debating, designing or enactingpolicies that promote water use reductions. What allthis means for policy analysis is that the net result ofsurface water pricing, including marketing and con-servation legislation or incentives, is an uncertainconservation policy tool when ground water is avail-able as a close substitute for surface water. An effect-ive conservation policy will account for theinteraction between the two water sources and willattempt to encourage irrigators to manage the twowater sources jointly (Schuck and Green, 2003).

CONCLUSIONS

The ability for an irrigator to realize an economicgain by reducing current water use in agriculture isan overarching factor that can be used to designincentives to promote water conservation. Despite theimportance of water-conserving incentives, severalbarriers to water conservation are identified. Theseinclude lack of clear titles to water rights, barriers towater transfers, on-farm water savings that fail tosave water for the basin, and barriers to securingrights to conserved water. Other barriers include theease with which greater ground water use can besubstituted for reduced surface water, water’s uncer-tain duty, the common property nature of carry-overstorage, interstate compact constraints, and water’slow price, all of which can effectively lock water intolow-valued agriculture.

Institutional, as well as hydrologic factors, affectthe potential for agricultural water conservation.While institutional barriers to water conservationexist for the entire Rio Grande Basin, the hydrologi-cal conditions affecting agricultural water use varygreatly within the basin. A significant proportion ofirrigation water rights are dependent on return flowsfrom upstream irrigation diversions in the basin. Ifwater conservation measures that decrease the mag-nitude of return flows, such as drip irrigation, becomewidely adopted, the existing hydrologic integrity of

the basin will be affected. The potential for agricul-tural water conservation is greater in basins that areless dependent by irrigation operations on returnflows.

The potential for future agricultural water conser-vation in the Rio Grande Basin varies greatly amongregions. More importantly, improved policy initiativesdesigned to implement conservation would be basedon how water is used at the basin level rather thanat the individual farm level. Consideration of theexisting structure of water use at the basin level willminimize any negative implications of conservationstrategies.

One constructive measure to promote water-con-serving decisions is to design institutions thatremove barriers to informing water users about theopportunity cost of current water uses. Another is toenact laws and policies that guarantee that reducedupstream water use does not simply come at theexpense of water taken from a downstream user.Considerable differences in the value of water usedin agriculture versus urban and environmental usecreate an opportunity for designing legal and pricinginstitutions that reduce barriers to market transfersand incentives that discourage conservation. Waterthat could be saved in agriculture is typically quitesensitive to price changes. Owners or users of agri-cultural water rights could use this price sensitivityto their advantage by renting or leasing their waterto cities or environmental users in periods ofdrought or other shortages with no change in waterrights ownership. Without legislative action, percep-tions by many farmers that all unused water maybe lost pose a barrier to water conservation. Manywater users in the basin avoid water conservationbecause incomplete stream adjudications throw intodoubt the security of their water right. Higher cur-rent use is believed by some to be an indication ofbeneficial use of a larger quantity of water than iscurrently needed, particularly when a severedrought may reduce all quantities. Where there iswater infrastructure to store and move tradedwater, legislation that defines water trading to be abeneficial water use could remove this barrier towater conservation.

ACKNOWLEDGMENTS

The authors gratefully acknowledge financial support for thisresearch by the New Mexico Agricultural Experiment Station,Texas Agricultural Experiment Station, and Rio Grande Basin Ini-tiative Project, U.S. Department of Agriculture, Agreement Num-bers 2005-34461-15661 and 2005-45049-03209. This article is anextended version of a talk given at the 50th annual New MexicoWater Conference in Las Cruces, New Mexico, in October 2005.Responsibility for errors rests with the authors.



LITERATURE CITED

Anton, W.F., 1995. Implementing ASCE Water-Conservation Pol-icy. Journal of Water Resources Planning and Management-Asce121(1):80-89.

Baumann, D.D., J. Boland, and J. Sims, 1984. Water Conservation,The Struggle Over Definition. Water Resources Research20(4):428-437.

Burt, O.R., 1964. The Economics of Conjunctive Use of Ground andSurface Water. Hilgardia 36(2):31-111.

California Department of Water Resources, 1992. The 1992Drought Water Bank, Sacramento, California.

Caswell, M. and D. Zilberman, 1985. The Choices of IrrigationTechnologies in California. American Journal of AgriculturalEconomics 67(2):224-234.

Caswell, M.F. and D. Zilberman, 1986. The Effects of WellDepth and Land Quality on the Choice of Irrigation Technol-ogy. American Journal of Agricultural Economics 68(4):798-811.

Caswell, M., E. Lichtenberg, and D. Zilberman, 1990. The Effectsof Pricing Policies on Water Conservation and Drainage. Ameri-can Journal of Agricultural Economics 72(4):883-890.

Cuthbert, R.W. and P.R. Lemoine, 1996. Conservation-OrientedWater Rates. Journal of the American Water Works Association88(11):68-78.

De Loe, R., L. Moraru, R. Kreutzwiser, K. Schaefer, and B. Mills,2001. Demand Side Management of Water in Ontario Munici-palities: Status, Progress, and Opportunities. Journal of theAmerican Water Resources Association 37(1):57-72.

Do Monte, M.H.F.M., A.N. Angelakis, and T. Asano, 1996. Neces-sity and Basis for Establishment of European Guidelines forReclaimed Wastewater in the Mediterranean Region. Water Sci-ence and Technology 33(10–11):303-316.

Dziegielewski, B., H.P. Garbharran, and J.F. Langowski, 1993. Les-sons Learned From the California Drought (1987–1992): NationalStudy of Water Management During Drought, IWR Report 93.U.S. Army Corps of Engineers, Washington, District of Columbia.

Green, G.P. and D.L. Sunding, 1997. Land Allocation, Soil Quality,and the Demand for Irrigation Technology. Journal of Agricul-tural and Resource Economics 22(2):367-375.

Green, G., D. Sunding, D. Zilberman and D. Parker, 1996.Explaining Irrigation Technology Choices: A MicroparameterApproach. American Journal of Agricultural Economics78(4):1064-1072.

Griffin, R.C., 2006. Water Resource Economics: The Analysis ofScarcity, Policies, and Projects. MIT Press, Cambridge, Massa-chusetts.

Huffaker, R. and N. Whittlesey, 2000. The Allocative Efficiency andConservation Potential of Water Laws Encouraging Investmentsin on-Farm Irrigation Technology. Agricultural Economics24(1):47-60.

Huffaker, R. and N. Whittlesey, 2003. A Theoretical Analysis ofEconomic Incentive Policies Encouraging Agricultural WaterConservation. International Journal of Water Resources Devel-opment 19(1):37-53.

Huffaker, R., N. Whittlesey, A. Michelsen, R. Taylor, andT. McGuckin, 1998. Evaluating the Effectiveness of Conserva-tion Water-Pricing Programs. Journal of Agricultural andResource Economics 23(1):12-19.

Idaho Water Supply Bank, 2005. http://www.idwr.state.id.us/water-board/water%20bank/waterbank.htm, accessed February 13,2006).

Jenkins, M.W. and J.R. Lund, 2000. Integrating Yield and ShortageManagement Under Multiple Uncertainties. Journal of WaterResources Planning and Management-Asce 126(5):288-297.

Knapp, K.C., 1995. The Economics of Conjunctive GroundwaterManagement With Stochastic Surface Supplies. Journal of Envi-ronmental Economics and Management 28(3):340-356.

Libbin, J. and J. Hawkes, 2005. New Mexico Farms by County.http://costsandreturns.nmsu.edu/2005Projected.htm, accessedNovember 12, 2005.

Loaiciga, H.A. and S. Renehan, 1997. Municipal Water use andWater Rates Driven by Severe Drought: A Case Study. Journalof the American Water Resources Association 33(6):1313-1326.

Michelsen, A.M., J.T. McGuckin, and D. Stumpf, 1999a. NonpriceWater Conservation Programs as a Demand Management Tool.Journal of the American Water Resources Association 35(3):593-602.

Michelsen, A.M., R.G. Taylor, R.G. Huffaker, and J.T. McGuckin,1999b. Emerging Agricultural Water Conservation Price Incen-tives. Journal of Agricultural and Resource Economics24(1):222-238.

Moore, M.R. and D.H. Negri, 1992. A Multicrop Production Modelof Irrigated Agriculture Applied to Water Allocation Policy ofthe Bureau of Reclamation. Journal of Agricultural andResource Economics 17(1):29-43.

Mulwafu, W., C. Chipeta, G. Chavula, A. Ferguson, B.G. Nkhoma,and G. Chilima, 2003. Water Demand Management in Malawi:Problems and Prospects for its Promotion. Physics and Chem-istry of the Earth 28(20–27):787-796.

Pender, J.L. and J.M. Kerr, 1998. Determinants of Farmers’ Indi-genous Soil and Water Conservation Investments in Semi-AridIndia. Agricultural Economics 19(1–2):113-125.

Peterson, J.M. and Y. Ding, 2005. Economic Adjustments toGroundwater Depletion in the High Plains: Do Water-SavingIrrigation Systems Save Water American Journal of Agricul-tural Economics 87(1):147-159.

Pratt, K.B., 1994. Water Banking: A new Tool for Water Manage-ment. The Colorado Lane 23(3):595-597.

Provencher, B. and O. Burt, 1994. Approximating the OptimalGroundwater Pumping Policy in a Multiaquifer StochasticConjunctive use Setting. Water Resources Research 30:833-844.

Schaible, G.D, 1997. Water Conservation Policy Analysis: An Inter-regional, Multi-Output, Primal-Dual Optimization Approach.American Journal of Agricultural Economics 79(1):163-177.

Schuck, E.C. and G.P. Green, 1998. Drought-Contingent Water Pri-cing in a Conjunctive use System. American Journal of Agricul-tural Economics 80(5):1198.

Schuck, E. and G. Green, 1999. Water Management and Pricing ina Conjunctive use System. American Journal of AgriculturalEconomics 81(5):1320.

Schuck, E.C. and G.P. Green, 2002. Supply-Based Water Pricing ina Conjunctive use System: Implications for Resource andEnergy use. Resource and Energy Economics 24(3):175-192.

Schuck, E. and G.P. Green, 2003. Conserving one Water Source atthe Expense of Another: The Role of Surface Water Price inAdoption of Wells in a Conjunctive use System. InternationalJournal of Water Resources Development 19(1):55-66.

Sokolov, V.I., 1999. Integrated Water Resources Managementin the Republic of Uzbekistan. Water International 24(2):104-115.

Southeastern Colorado Water Conservancy District; Fort LyonCanal Company; and Holbrook Mutual Irrigating Company v.Shelton Farms, Inc.; Southeastern Colorado Water ConservancyDistrict v. Colorado-New Mexico Land Co., Inc. Nos. 26420,26432. Supreme Court of Colorado. 187 Colo. 181; 529 P.2d1321; 1974 Colo. LEXIS 669. December 16, 1974, Decided.

Stonehouse, D.P., 1996. A Targeted Policy Approach to InducingImproved Rates of Conservation Compliance in Agriculture.Canadian Journal of Agricultural Economics-Revue CanadienneD Economie Rurale 44(2):105-119.

Texas Water Code, 1985a. 63. § 11.002 (8:A,B; 9).Texas Water Code, 1985b. 63. § 11.134 (b:4).Texas Water Code, 1997. 63. § 11.023, (a:1-8)).



U.S. Department of Interior, 2003. Water 2025: Preventing Crisesand Conflict. Bureau of Reclamation, Washington, District ofColumbia.

Ward, F.A., A. Michelsen, and L. DeMouche, 2004. Telephone Sur-vey of Water Conservation Practices in Irrigated Agriculture inthe Rio Grande Basin (unpublished). New Mexico State Univer-sity, Las Cruces, New Mexico.

Winter-Nelson, A. and K. Amegbeto, 1998. Option Values to Con-servation and Agricultural Price Policy: Application to TerraceConstruction in Kenya. American Journal of Agricultural Eco-nomics 80(2):409-418.

Yang, H., X.H. Zhang, and A.J.B. Zehnder, 2003. Water Scarcity,Pricing Mechanism and Institutional Reform in Northern ChinaIrrigated Agriculture. Agricultural Water Management61(2):143-161.

Zougmore, R., A. Mando, and L. Stroosnijder, 2004. Effect of Soiland Water Conservation and Nutrient Management on the Soil-Plant Water Balance in Semi-Arid Burkina Faso. AgriculturalWater Management 65(2):103-120.



barriers to water conservation in the rio grande basin1 · barriers to water conservation in the...

Documents