Who pays for water scarcity? Evaluating the welfare implications of water infrastructure investments for cities

Abstract

Continued provision of low-cost municipal and industrial water is anticipated to be a challenge for cities in the coming decades. To address this, many are considering large-scale infrastructure projects to expand their water supply. In this article, we develop a general equilibrium model to evaluate the economy-wide distributional impacts of water infrastructure projects. The model framework includes a regulated water utility with a cost-recovery mandate and captures the trade-off between the immediate costs of financing infrastructure projects and the long-term costs that water scarcity imposes on the regional economy. We apply the model to an on-going water infrastructure project in Las Vegas, Nevada.

This is a preview of subscription content, log in to check access.

Fig. 1

Notes

  1. 1.

    An important motivation for previous general equilibrium models of public infrastructure projects is that these projects often have limited alternative uses and therefore involve sunk costs, so that ex ante analysis is particularly necessary to ensure that expected benefits exceed these costs (Rioja 1999; Seung and Kraybill 2001; Haughwout 2002; Rioja 2003; Giesecke et al. 2008; Brueckner and Picard 2015).

  2. 2.

    Previous studies using GE models to analyze the role of water in the economy have been conducted at a variety of geographic scales, including international (Berrittella et al. 2007; Calzadilla et al. 2011), national (Diao and Roe 2003; Hassan and Thurlow 2011), interregional (Berck et al. 1991; Goodman 2000; Gomez et al. 2004; Watson and Davies 2011), single-region rural (Seung et al. 1998, 2000), and single-region urban (Dixon 1990; Rose and Liao 2005; Rose et al. 2011).

  3. 3.

    Water utilities typically act as wholesalers of water rights, buying and selling water rights on an ongoing basis to provide liquidity to the water rights market and to reduce transaction costs for developers. In Las Vegas, the SNWA’s water rights are managed by an independent nonprofit corporation (Southern Nevada Water Authority 2018). Profits from water rights sales do not appear on SNWA’s balance sheet, which suggests that this nonprofit corporation does not seek to exploit arbitrage opportunities in the water rights market.

  4. 4.

    Existing infrastructure is based on delivery of Nevada’s share of Lower Colorado River Compact water. Other compact members (California, Arizona, and Mexico) are unlikely to be in a position to permanently transfer portions of their annual allotments to Nevada given that water demand is projected to exceed supply for the Colorado River Basin as a whole by a median 3.2 million acre-feet annually by 2060 (U.S. Bureau of Reclamation 2012). Further, Las Vegas has nearly exhausted its ability to transfer water from nearby agriculture to M&I use.

  5. 5.

    While our assumption that GWD is financed by customers through higher water rates accords with reality in Las Vegas and other cities serviced by a regulated water utility, previous GE models have assumed that water infrastructure investments are financed by exogenous government surplus or outside investors (e.g., Seung and Kraybill 2001; Rioja 2003; Strzepek et al. 2008; Bom and Ligthart 2014).

  6. 6.

    In IMPLAN, municipal water appears in two sectors: private water utility sector (sector 51) and public utilities (sector 526). After reconstructing the water sector for Las Vegas using both sectors 51 and 526, we find the off-the-shelf data from IMPLAN understates the size of the municipal water sector by a factor of more than two and also inaccurately represents the relative water-intensities of industrial sectors.

  7. 7.

    Las Vegas currently recycles almost 100% of indoor water and has already implemented one of the most aggressive voluntary conservation programs in the USA, suggesting likely decreasing returns from future conservation efforts. Between 2002 and 2016, the region reduced its net gallons per capita per day by 38% (Southern Nevada Water Authority 2017). Further, there is almost no scope for expanding Las Vegas’ M&I water portfolio by transferring water out of nearby agriculture or by developing alternative water resources such as rainwater recycling (Southern Nevada Water Authority 2017).

  8. 8.

    While several previous studies have assumed that raw water is substitutable with other inputs in the production of treated water (Goodman 2000; Diao and Roe 2003; Rose and Liao 2005; Watson and Davies 2011), we believe that the Leontief assumption with a provision for water recycling is a more accurate description of production by a water utility.

  9. 9.

    Our assumption of sector-specific capital implies that the regulated utility’s production and pricing decisions are influenced by the debt it assumes to finance sector-specific capital rather than by the rental rate of capital in the broader economy. In contrast, previous studies have assumed one type of physical capital that can be used by all sectors in the economy with equilibrium-determined prices (e.g., Goodman 2000; Rose and Liao 2005; Watson and Davies 2011).

  10. 10.

    The applied GE model assumes, as we do here, that the debt payments for water infrastructure are paid to creditors outside of Clark county. This assumption is in keeping with our approach of underestimating the welfare benefits of GWD.

  11. 11.

    While our assumption that water scarcity rents are capitalized in the housing sector is realistic for Las Vegas, this assumption may not be appropriate in jurisdictions where the water utility does not operate under a cost-recovery mandate. Without a cost-recovery mandate, a utility would be free to set water prices to maximize profit given the constraint that water demand and supply are balanced and, hence, capture water scarcity rents. We do not consider this counterfactual scenario in this article because there is no evidence that an institutional change that would allow utilities in Las Vegas to set water prices above long-run average cost is being contemplated.

  12. 12.

    The welfare implications for this first case also apply to the case when the system is supply-constrained without new infrastructure but the additional debt causes the utility to reduce water supply, i.e., \(Y_{u,1}^{*} < Y_{u,o}^{*} = \bar{Y}_{u,0}\).

  13. 13.

    DETR data identify each firm by six-digit NAICS code. We merge county tax assessor commercial parcel data with DETR employment data using street addresses and geographic information system information to generate money and physical flows of land and capital for each productive sector in Las Vegas. Similarly, we aggregate county tax assessor residential parcel data into six groups based on property value and lot size, and distribute six housing service sectors across the nine household groups using PUMS household records. Assessed value is converted to annualized rental flows using a midterm discount rate of 11%.

  14. 14.

    One potential method to construct the water sector would be to separate the water utility from IMPLAN’s “other local government enterprises” sector based on employment numbers. However, our initial work revealed that this method using IMPLAN data would not work for Las Vegas because approximated total revenue for the water utility would be less than half of the total revenue calculated using LVVWD billing data.

  15. 15.

    Starting in 2013, a water infrastructure charge has been included with all SNWA customer water bills to fund necessary improvements to facilities at Lake Mead, the reservoir by which Nevada receives its share of Colorado River water. Our method of using tax assessor records does not capture the capital cost of this infrastructure, since it is on federal land. Thus, we use the surplus to represent the current sector-specific infrastructure costs.

  16. 16.

    Approximately 40% of water sold in Las Vegas is recycled (Southern Nevada Water Authority 2017). Of this 40%, 90% is treated to a potable standard and returned to the Colorado River (i.e., Lake Mead) for reuse by M&I customers in Las Vegas. The remaining 10% is not treated to a potable standard and is used directly for outdoor watering, primarily on golf courses. While we do not consider non-potable recycled water as a separate category in our analysis, we do not believe that this simplification influences our conclusions given that non-potable reuse accounts for less than 4% of total water used in Las Vegas.

  17. 17.

    The approval of the inter-basin transfers of groundwater for the GWD involved several lengthy hearing processes where the Nevada State Engineer’s Office had to establish that the groundwater rights in all affected basins be less than or equal to the average perennial yield, which is the amount of water that can be withdrawn without exceeding the natural recharge in the basin (Welsh and Endter-Wada 2017). Given the imperative that all basins supplying water for the GWD be in hydrologic balance, it is expected that water from the GWD will be have limited inter-annual variation in supply.

  18. 18.

    For example, the Clark County Comprehensive Planning report predicts that employment in the health, construction, transportation and warehousing, accommodation, and food services sectors will increase by 60%, 45%, 40%, 25%, and 25% by 2030, and 160%, 150%, 100%, 40%, and 40% by 2050 (Clark County 2018). We cut the growth projections in the Clark County Comprehensive Planning report in half because the high employment growth rates in the report cannot be rationalized under standard assumptions about the growth rates for total factor productivity and capital stock. Further, given that higher export growth will increase the cost of water scarcity for Las Vegas, cutting these growth rates in half is consistent with our approach of selecting model parameters so as to not overstate the potential future benefits of the GWD.

  19. 19.

    We performed sensitivity analysis of model results assuming faster TFP growth (not reported). As expected, the sensitivity analysis indicated that faster TFP growth increases the negative impact of water scarcity in 2050 and, as a result, causes the benefit of GWD project become larger.

  20. 20.

    In our baseline parameterization, “2030” or “2050” refers to when the population is increased by 25% or 50%, respectively.

  21. 21.

    Given the assumption that water ratepayers finance the GWD, the additional tax revenue in the GWD scenario is net of GWD costs.

  22. 22.

    Water intensity here refers to share of water in total sectoral production costs.

  23. 23.

    Previous studies have suggested that turf and trees moderate heat island effects and improve urban quality-of-life in cities in the southwest USA (e.g., Klaiber et al. 2017).

References

  1. Berck P, Robinson S, Goldman G (1991) The use of computable general equilibrium models to Assess Water Policies. In: Berck P, Robinson S, Goldman G (eds) The economics and management of water and drainage in agriculture. Kluwer Academic Publishing, Norwell, pp 489–509

    Google Scholar 

  2. Berck P, Golan E, Smith B, Barnhart J, Dabalen A (1996) Dynamic revenue analysis for California. California Department of Finance, Sacramento

    Google Scholar 

  3. Berrittella M, Hoekstra AY, Rehdanz K, Roson R, Tol RSJ (2007) The economic impact of restricted water supply: a computable general equilibrium analysis. Water Res 41(8):1799–1813

    Article  Google Scholar 

  4. Blanciforti LA, Green RD, King GA (1986) US consumer behavior over the postwar period: an Almost Ideal Demand System analysis. Department of Agricultural and Resource Economics, University of California, Davis. Giannini Foundation Monograph (40). http://ageconsearch.umn.edu/bitstream/11939/1/ms860040.pdf

  5. Bom PRD, Ligthart JE (2014) Public infrastructure investment, output dynamics, and balanced budget fiscal rules. J Econ Dyn Control 40:334–354

    Article  Google Scholar 

  6. Brueckner JK, Picard PM (2015) Where and when to invest in infrastructure. Reg Sci Urban Econ 53:123–134

    Article  Google Scholar 

  7. Burnett P, Cutler H, Davies S (2012) Understanding the unique impacts of economic growth variables. J Reg Sci 52(3):451–468

    Article  Google Scholar 

  8. Calzadilla A, Rehdanz K, Tol RSJ (2011) Water scarcity and the impact of improved irrigation management: a computable general equilibrium analysis. Agric Econ 42(3):305–323

    Article  Google Scholar 

  9. Calzadilla A, Rehdanz K, Roson R, Sartori M, Martina, Tol RSJ (2017) Review of CGE models of water issues. In: Bryant T (ed) Computable general equilibrium models. The WSPC reference on natural resources and environmental policy in the era of global change. World Scientific, Singapore, pp 101–124

    Google Scholar 

  10. Center for Business and Economic Research (CBER), University of Nevada, Las Vegas (2016) Population forecasts: long-term projections for Clark County, Nevada 2016–2050. http://www.clarkcountynv.gov/comprehensive-planning/demographics/Documents/2016%20Clark%20County%2C%20NV%20Population%20Forecasts.pdf

  11. Chemingui MA, Thabet C (2016) Economy-wide analysis of alternative water management policies: a comparative analysis for Morocco and Tunisia. Water Econ Policy 2(4):1–27

    Article  Google Scholar 

  12. Clark County (2018) Clark County comprehensive master plan. http://www.clarkcountynv.gov/comprehensive-planning/advanced-planning/Pages/ComprehensivePlan.aspx. Verified 30 June 2019

  13. Cutler H, Davies S (2007) The impact of specific-sector changes in employment on economic growth, labor market performance and migration. J Reg Sci 47(5):935–963

    Article  Google Scholar 

  14. Diao X, Roe T (2003) Can a water market avert the double-whammy of trade reform and lead to a win–win outcome? J Environ Econ Manag 45(3):708–723

    Article  Google Scholar 

  15. Dixon PB (1990) A general equilibrium approach to public utility pricing: determining prices for a water authority. J Policy Model 12(4):745–767

    Article  Google Scholar 

  16. Fan Q, Fisher-Vanden K, Klaiber HA (2018) Climate change, migration, and regional economic impacts in the United States. J Assoc Environ Resour Econ 5(3):643–671

    Google Scholar 

  17. Fang GH, Wang T, Si XY, Wen X, Liu Y (2016) Discharge fee policy analysis: a Computable General Equilibrium (CGE) model of water resources and water environments. Water 8(9):413

    Article  Google Scholar 

  18. Faust AK, Gonseth C, Vielle M (2015) The economic impact of climate-driven changes in water availability in Switzerland. Water Policy 17(5):848–864

    Article  Google Scholar 

  19. Giesecke J, Dixon PB, Rimmer MT (2008) Regional macroeconomic outcomes under alternative arrangements for the financing of public infrastructure. Pap Reg Sci 87(1):3–31

    Article  Google Scholar 

  20. Gomez C, Tirado D, Rey-Maquieira J (2004) Water exchanges versus water works: insights from a computable general equilibrium model for the Balearic Islands. Water Resourc Res 40(10):W10502

    Article  Google Scholar 

  21. Goodman DJ (2000) More reservoirs or transfers? A computable general equilibrium analysis of projected water shortages in the Arkansas River basin. J Agric Resour Econ 25:698–713

    Google Scholar 

  22. Hannum C, Cutler H, Iverson T, Keyser D (2017) Estimating the implied cost of carbon in future scenarios using a CGE model: the case of Colorado. Energy Policy 102:500–511

    Article  Google Scholar 

  23. Hassan R, Thurlow J (2011) Macro-micro Feedback links of water management in South Africa: CGE analyses of selected policy regimes. Agric Econ 42:235–247

    Article  Google Scholar 

  24. Haughwout AF (2002) Public infrastructure investments, productivity and welfare in fixed geographic areas. J Public Econ 83(3):405–428

    Article  Google Scholar 

  25. Klaiber HA, Abbott JK, Smith K (2017) Some like it (less) hot: extracting trade-off measures for physically coupled amenities. J Assoc Environ Resour Econ 4(4):1053–1079

    Google Scholar 

  26. Lofgren H, Harris RL, Robinson S (2002) A standard computable general equilibrium (CGE) model in GAMS, vol 5. International Food Policy Research Institute, Washington

    Google Scholar 

  27. Luckmann J, Grethe H, McDonald S, Orlov A, Siddig K (2014) An integrated economic model of multiple types and uses of water. Water Resour Res 50(5):3875–3892

    Article  Google Scholar 

  28. Partridge MD, Rickman DS (2010) Computable General Equilibrium (CGE) modelling for regional economic development analysis. Reg Stud 44(10):1311–1328

    Article  Google Scholar 

  29. Rioja FK (1999) Productiveness and welfare implications of public infrastructure: a dynamic two-sector general equilibrium analysis. J Dev Econ 58(2):387–404

    Article  Google Scholar 

  30. Rioja FK (2003) Filling potholes: macroeconomic effects of maintenance versus new investments in public infrastructure. J Public Econ 87(9):2281–2304

    Article  Google Scholar 

  31. Rose A, Liao S-Y (2005) Modeling regional economic resilience to disasters: a computable general equilibrium analysis of water service disruptions. J Reg Sci 45(1):75–112

    Article  Google Scholar 

  32. Rose A, Liao SY, Bonneau A (2011) Regional economic impacts of a Verdugo scenario earthquake disruption of Los Angeles water supplies: a computable general equilibrium analysis. Earthq Spectra 27(3):881–906

    Article  Google Scholar 

  33. Schwarm W, Cutler H (2003) Building small city and town SAMs and CGE models. Rev Urban Reg Dev Stud 15(2):132–147

    Article  Google Scholar 

  34. Seung CK, Kraybill D (2001) The effects of infrastructure investment: a two-sector dynamic computable general equilibrium analysis for Ohio. Int Reg Sci Rev 24(2):261–281

    Article  Google Scholar 

  35. Seung CK, Harris TR, MacDiarmid TR, Shaw WD et al (1998) Economic impacts of water reallocation: a CGE analysis for walker river basin of Nevada and California. J Reg Anal Policy 28:13–34

    Google Scholar 

  36. Seung CK, Harris TR, Englin JE, Netusil NR (2000) Impacts of water reallocation: a combined computable general equilibrium and recreation demand model approach. Ann Reg Sci 34(4):473–487

    Article  Google Scholar 

  37. Southern Nevada Water Authority (2017) Water resource plan. https://www.snwa.com/where-southern-nevada-gets-its-water/water-resource-plan/index.html. Verified 30 June 2019

  38. Southern Nevada Water Authority (2018) Comprehensive annual financial report. https://www.snwa.com/assets/pdf/reports-cafr.pdf. Verified 30 June 2019

  39. Strzepek KM, Yohe GW, Tol RS, Rosegrant MW (2008) The value of the high Aswan dam to the Egyptian economy. Ecol Econ 66(1):117–126

    Article  Google Scholar 

  40. U.S. Department of the Interior, Bureau of Reclamation (2012) Colorado River basin water supply and demand study. https://www.usbr.gov/watersmart//bsp/docs/finalreport/ColoradoRiver/CRBS_Executive_Summary_FINAL.pdf

  41. Watson PS, Davies S (2011) Modeling the effects of population growth on water resources: a CGE analysis of the South Platte river basin in Colorado. Ann Reg Sci 46(2):331–348

    Article  Google Scholar 

  42. Welsh LW, Endter-Wada J (2017) Policy debates over the Southern Nevada water authority groundwater development project: beneficial uses of water in a desert. J Southwest 59(1–2):302–337

    Article  Google Scholar 

  43. Zhong S, Shen L, Sha JH, Okiyama M, Tokunaga S, Liu LT, Yan JJ (2015) Assessing the water parallel pricing system against drought in China: a study based on a CGE model with multi-provincial irrigation water. Water 7(7):3431–3465

    Article  Google Scholar 

Download references

Acknowledgements

This research was funded by the USDA Agriculture and Food Research Initiative (AFRI) through the grant “Rural to Urban Water Transfers, Climate Change and the Future of Rural Agricultural Economics in the Semi-Arid West: A Comparative Regional Analysis” (Agreement #: 2015-67023-22987). We would like to thank Harvey Cutler, Thomas R. Harris, and Kevin H. Crofton for their invaluable guidance and assistance.

Author information

Affiliations

Authors

Corresponding author

Correspondence to Michael H. Taylor.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary material 1 (CSV 23 kb)

Appendices

Appendix A: Mathematical presentation of model

See Table 9.

Table 9 Mathematical presentation of the model

Appendix B: Elasticity parameters

Elasticities we use in our simulation are as follows:

ElasticityValueReferences
Elasticities for residential land supply2Cutler and Davies (2007)
Elasticities for commercial land supply1
Labor supply elasticity in response to the labor average wage (HH1–HH9)0.1– to 0.8Berck et al. (1996)
Labor supply elasticity in response to household taxes (HH1–HH9)− 0.55 to − 0.15
Migration elasticity in response to after tax earnings (HH1–HH9)1.5–2.3
Migration elasticity in response to unemployment (HH1–HH9)− 0.7 to − 0.2
The elasticity of substitution between primary factors0.8Watson and Davies (2011)
Income elasticities for household private consumption
Agriculture and food0.48–0.5Blanciforti et al. (1986)
Utilities0.52
Retail0.8
Services0.7
Hospital and health0.35
Durable and manufactured consumption1.5
“Miscellaneous” sector1
Housing services0.8
Own-price elasticities
Agriculture and food− 0.3Blanciforti et al. (1986)
Utilities− 0.5
Retail− 0.4
Services− 0.2
Hospital and health− 0.3
Durable and manufactured consumption− 0.42
“Miscellaneous” sector− 1
Housing services− 0.2

Note that we observe a broader range of income elasticities for housing services from previous literature, where elasticities range from 0.14 to 1.4. In our study, housing services is an agent sector including sales, rentals, and maintenance. We ran a sensitivity analysis with regard to the elasticities and our major conclusions remain. The results can be provided upon your request.

Appendix C: Model implementation

Table 10 describes scenarios implemented in this paper. Scenario a presents a baseline of our simulation assuming that Las Vegas is unable to build up the infrastructure. If the resource constraint binds, the impact of water shortage is capitalized into the housing market. \(\mu\) is a price slack variable associated with the resource constraint, and can be explained as the “profit” in housing market caused by the shortage. Our model then redistributes the “housing profit” back to the household as a lump sum income suggesting that owners of housing sectors receive all of the housing profits. Our empirical setting assumes that all of capital and land are owned by regional households and foreign investors. Hence, we first allocate the housing profit to local households and foreign investors (rest of the world) according to local and foreign ownerships of capital and land (see Eq. A5 in “Appendix A”), then we distribute the housing profit owned by local households to each household group based on households’ ownerships of capital and land (see Eq. A7 in “Appendix A”). Scenarios b simulates the economic growth with the infrastructure. We assume all water customers in the region contribute toward funding the infrastructure. The infrastructure cost is imposed in Eq. (A11) in “Appendix A”.

Table 10 Model implementation

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Zhong, H., Taylor, M.H., Rollins, K.S. et al. Who pays for water scarcity? Evaluating the welfare implications of water infrastructure investments for cities. Ann Reg Sci 63, 559–600 (2019). https://doi.org/10.1007/s00168-019-00943-w

Download citation

JEL Classification

  • R13
  • L95
  • Q25