Determinants of Household Payments for Drinking Water in Mexico, with Equity Considerations

doi:10.21203/rs.3.rs-2740492/v1

The situation with non-payment of the drinking water services that households receive is an increasingly common problem with significant financial implications. The principal determinants of this lack of payment was estimated using a probabilistic model for the case of Mexico, with consideration of equity among household deciles. The results show that while this problem can be associated with households belonging to lower deciles, those in upper deciles are responsible for nearly the same share of the financial impact on water management organizations, and consequently, on the wellbeing of society. Nevertheless, households in the first deciles have more reasonable motives for deciding not to pay. Therefore, when wanting to develop policy incentives for households to pay for drinking water services, the incentives should be differentiated according to socioeconomic strata.

Payment

drinking water

deciles

equity

responsibility

Interest in improving the management of water services that are provided to society has grown in recent years (García 2002), and is a result of multiple factors, including environmental, natural, socioeconomic and cultural. Among these, the most important factors are the lack of the resources, increased user demand for quality and quantity, heightened environmental awareness and financial sustainability. In this sense, the role of water authorities and water service managers has changed from merely acting as managers of water infrastructure to their actions having greater social implications (Winpenny 1994). This new orientation in the management of services has affected different water-related fields, such as socio-hydrological planning, the creation of legal structures and institutions, the development of new economic instruments and tools, and new management models, among others. Specifically in the area of economics, interest in establishing prices for water as a management tool arose decades ago (Cook et al. 2016). And in recent years there has been a focus on the socio-economic impact of fees and whether they can recuperate the cost of water services, for example, whether or not households that receive the service are willing to pay for it. That is, the complexity of setting prices now has more implications on multiple objectives, including efficiency, equity, rationale, access to services, affordability, human rights, financial factors, public acceptability and transparency (Silva-Ruiz 2010).

In terms of efficiency, fees are aimed at generating enough income to cover both the costs associated with providing sufficient amounts of quality services as well as future investments (Revollo-Fernández et al. 2019a). While the principle of equity is more closely related with the human right to drinking water and sanitation services, which was explicitly recognized in July 2010 by U.N. General Assembly resolution 64/292 (Kooy et al. 2018), policies involving the right to water and sanitation need to be determined based on criteria related to the availability, quality, accessibility, affordability and sustainability of water services (Cook et.al. 2016).

In Mexico, the geographic distribution of water does not coincide with the geographic distribution of the population. While the average volume of renewable water in the country is 4,028 cubic meters per capita annually, substantial differences exist between southeastern and northern Mexico. Some regions have a great deal of water scarcity and others experience frequent hydro-meteorological events resulting in costly floods and impacts on human settlements and infrastructure (Revollo-Fernández et al. 2019b). Furthermore, while coverage is 97% in urban areas, it is 84.8% in rural areas (less than 2,500 inhabitants) due to the population being dispersed across regions with complex physiographic conditions, and the technical and financial difficulty of developing drinking water systems in those regions. While 94% of the population officially has access to public drinking water services, that percentage decreases when considering the quality of the resource and the frequency with which it reaches homes, situations that primarily and drastically affect lower-income households or those with greater social deficits (Revollo-Fernández et al. 2019a). Given all of this context, living in remote regions or those with greater social deficits can affect decisions about whether or not to pay for drinking water services, and can thereby financially impact water management organizations (WMO).

In light of this context, the work herein was aimed at measuring the determinants and impacts of households in Mexico not paying for public drinking water services, while also taking into account equity considerations. The goal of this was to propose public policy options that benefit society. This document is structured as follows: the second section presents the situation with drinking water services in Mexico, followed by the methods and results, and lastly some conclusions and discussion.

In Mexico, water use, supply and treatment is regulated by the Political Constitution of the Mexican United States. Specifically, Article 27 establishes that national waters are the property of the nation. In addition, Article 115 indicates that municipal governments are responsible for providing drinking water, drainage and sewer services, and for the treatment and disposal of wastewater. State laws also exist that regulate the actions of the various municipalities. This set of legal instruments establishes the guidelines for the use and conservation of this resource as well as for the rights and obligations of water management organizations (WMOs). The WMO is the economic unit that manages and operates the systems and provides drinking water, sewer and sanitation services. The structure and makeup of the WMOs differ. They are generally affiliated with municipal governments and operate in departments and commissions that are involved in drinking water and sanitation services, or in decentralized water systems. They also operate as local boards, water user committees, and to a lesser degree as private concessionary companies. To provide water services, the WMO must have a decree that is issued by the municipal authority that recognizes its establishment as a WMO. This decree delegates to it the functions of managing the water resources and generating the technical infrastructure needed, and grants the WMO the authority to exploit the water and distribute it throughout a specific geographic area (municipality, urban locality or rural locality). For the case of private companies, they are granted water concessions. Both public and private organizations must comply with water regulations.

In 2018, approximately 2,826 WMOs were operating in Mexico, of which 166 (5.8%) were private and employed 4,756 people. The rest of the organizations (94.2%) belonged to the public sector (94.2%) and employed 118,445 people (CONAGUA 2018). Of all the WMOs, 1,642 (58.1%) provided water services strictly to urban populations and 1,184 (41.9%) provided them to rural populations (INEGI 2018). A greater number of the WMOs were located in urban regions since these regions have more water demand for domestic use and for economic activities. With respect to coverage and the number of inhabitants with access to drinking water services in Mexico, this has been increasing over recent decades. In 1990, the total coverage nationwide was 75.4% (61.2 million people), with 87.0% provided to the urban population (50.5 million people) and 46.5% to the rural population (10.8 million people). By 2020, this coverage reached 94.2% nationally (115 million people), with 94.2% coverage in urban regions (97.2 million people) and 86.3% in rural regions (24.9 million people).

3.1 Data

Information was used from the 2020 National Survey of Household Income and Expenses (ENIGH 2020), which was developed by the National Institute of Statistics and Geography (INEGI in Spanish). This survey was aimed at providing a statistical overview of the behavior of household income and expenses, the occupational and sociodemographic characteristics of household members and the characteristics of housing infrastructure and furnishings. It was nationally representative of Mexico and its 32 states and included 35,749,659 households.

3.2 Analysis of the Information

The information was analyzed in three stages. The first involved a statistical analysis to determine the number of households in Mexico that had and had not paid for the drinking water services that they received, taking into account income deciles. In the second stage, a probabilistic economic model was built and estimated to determine the variables that had greater and lesser effects on the likelihood of paying or not paying (Table 1). The variables for the econometric model were selected based on a prior literature review (Akinyemi et al., 2018; Revollo et al., 2019a), a correlation analysis and multicollinearity among independent variables, as well as considerations involving the country’s situation. The probit model thereby estimated the following:

Table 1

Dependent and independent variables included in the econometric model
Variable	Explanation	Type of Variable
Dependent Variable
expense	Whether or not the household pays the quarterly cost of drinking water	Dichotomous: Yes (1)/No (0)
Independent Variables
First group: Characteristics of Drinking Water Services
supply_d	The home receives water daily.	Dichotomous: Yes (1)/No (0)
supply_3d	The home receives water every other day (every third day).	Dichotomous: Yes (1)/No (0)
supply_2w	The home receives water two times per week.	Dichotomous: Yes (1)/No (0)
supply_1w	The home receives water once per week.	Dichotomous: Yes (1)/No (0)
supply_w	The home receives water once in a while.	Dichotomous: Yes (1)/No (0) Baseline variable for comparison
per_water/inc	Percentage of income spent on the cost of water	Percentage
Second Group: Socioeconomic Characteristics of the Households
loc1	Locality with less than 2,500 inhabitants (rural)	Dichotomous: Yes (1)/No (0) Baseline variable for comparison
loc2	Locality with 2,500 to 14,999 inhabitants (rura-urban)	Dichotomous: Yes (1)/No (0)
loc3	Locality with 15,000 to 99,999 inhabitants (rural-urban)	Dichotomous: Yes (1)/No (0)
loc4	Locality with 100,000 or more inhabitants (urban)	Dichotomous: Yes (1)/No (0)
decil1	First decile of total household income	Dichotomous: Yes (1)/No (0) Baseline variable for comparison
decil2	Second decile of total household income	Dichotomous: Yes (1)/No (0)
decil3	Third decile of total household income	Dichotomous: Yes (1)/No (0)
decil4	Fourth decile of total household income	Dichotomous: Yes (1)/No (0)
decil5	Fifth decile of total household income	Dichotomous: Yes (1)/No (0)
decil6	Sixth decile of total household income	Dichotomous: Yes (1)/No (0)
decil7	Seventh decile of total household income	Dichotomous: Yes (1)/No (0)
decil8	Eighth decile of total household income	Dichotomous: Yes (1)/No (0)
decil9	Ninth decile of total household income	Dichotomous: Yes (1)/No (0)
decil10	Tenth decile of total household income	Dichotomous: Yes (1)/No (0)
sex	Biological distinction that classifies the head of household as man or woman	Dichotomous: Man (1) / Woman (0)
age_head	Age of head of household, in years	Continuous
tot_mem	Number of members in this household, not including guests or domestic workers or their family members	Continuous

Source: Developed by the author, 2023.

Lastly, the third stage involved estimating the total economic loss from households not paying for drinking water services, considering income deciles, and which directly affected the operations of the different WMOs across the country, as well as their investments.

4.1 Who pays and does not pay for drinking water consumption in Mexico?

The information obtained from the 2020 ENIGH shows that 26.6% of all the households with access to drinking water in Mexico do not pay for the service, that is, roughly 8.8 million households. When analyzed according to income decile, 14% of all households that do not pay correspond to the first decile (1.2 million households) and 6.2% correspond to the tenth decile (553,000 households). The survey also reported that 44.4% of all the households belonging to the first decile do not pay for the service while only 14.0% of those belonging to the tenth decile do not pay. That is, more households in the lower deciles do not pay for the drinking water service than households in the upper deciles (Table 2). Therefore, while all household deciles were responsible for the non-payment of services, this situation is greater for households with less economic income.

Table 2

Composition of households in Mexico in those who pay and those who do not pay for the drinking water service
DECIL	ACCESS AND DO NOT PAY			ACCESS AND PAY			ACCESS
DECIL	Number of households	% access and do not pay	% of total access	Number of households	% access and pay	% of total access	Number of households	% access
All	8,877,470	100.00%	26.6%	24,523,194	100.00%	73.4%	33,400,664	100.0%
1	1,252,382	14.11%	44.4%	1,569,369	6.40%	55.6%	2,821,751	100.0%
2	1,057,631	11.91%	36.1%	1,871,310	7.63%	63.9%	2,928,941	100.0%
3	1,036,536	11.68%	33.5%	2,061,322	8.41%	66.5%	3,097,858	100.0%
4	989,043	11.14%	30.5%	2,250,310	9.18%	69.5%	3,239,353	100.0%
5	944,421	10.64%	28.9%	2,323,971	9.48%	71.1%	3,268,392	100.0%
6	899,966	10.14%	26.1%	2,542,911	10.37%	73.9%	3,442,877	100.0%
7	785,938	8.85%	22.6%	2,687,462	10.96%	77.4%	3,473,400	100.0%
8	714,904	8.05%	20.4%	2,783,413	11.35%	79.6%	3,498,317	100.0%
9	643,136	7.24%	17.5%	3,023,326	12.33%	82.5%	3,666,462	100.0%
10	553,513	6.24%	14.0%	3,409,800	13.90%	86.0%	3,963,313	100.0%

Source: Developed by the author based on information supplied by ENIGH-2020, 2023.

4.2 Determinants of household payments for drinking water consumption in Mexico

After determining the percentage of households that do not pay for drinking water consumption even though they receive this service, the next step was to identify the variables related with the service (first group of variables) as well as the socioeconomic variables of the households that could have more or less of an effect on the decision about whether or not to pay (second group of variables). For the first group, the characteristics of drinking water services, the results show that receiving the service on a more regular or frequent basis increases the likelihood of paying for drinking water. That is, for example, receiving water in the home daily increases that likelihood by 13.9%. It was also found that the higher the percentage of household income that is spent on water the less likely the households are to pay for the service. Therefore, for this first group, receiving water more frequently and spending less household income increases the likelihood of paying for the service (Table 3). In terms of inequality, it is important to mention that households in the lower income deciles receive the service more sporadically and spend more of their household income on water.

For the second group of variables, the socioeconomic characteristics of the households, the results show that the greater the number of inhabitants in the locality, that is the more urban the region, the greater the likelihood that households will pay for the service. That is, for example, a household in a locality with over 100,000 inhabitants is 7.4% more likely to pay than those in a locality with less than 2,500 inhabitants. This situation may be due to localities with more people having more established WMOs with better services. Meanwhile, as the income deciles increase, the likelihood of paying for the service also increases. For example, households in the second lowest income decile are 6.4% more likely to pay than the lowest decile, while belonging to the highest income decile (tenth) results in a 20.4% greater likelihood of paying. With regard to sex, a male head of household has a 1.6% higher likelihood of paying. And the larger the number of household members the less likely they are to pay.

Lastly, the variables with the most weight with respect to whether or not a household pays for drinking water services are the frequency of the service and the strata level, as measured by income.

Table 3

Determinants of spending or not on drinking water in Mexico
Marginal effects after probit		MODEL PROBIT
		Obs.	82,121
		LR chi²(19)	1,976,388.3
Y = Pr(expense)(predict)	0.74547839	Prob > chi²	0
Y = Pr(expense)(predict)	0.74547839	Pseudo R²	0.0511
Variable		dy/dx	P > \|z\|
First group: Characteristics of Drinking Water Services
supply_d		0.139685	0.000
supply_3d		0.1083586	0.000
supply_2w		0.088014	0.000
supply_1w		0.0805327	0.000
per_water/inc		-0.0294529	0.000
Second Group: Socioeconomic Characteristics of the Households
loc2		0.0130123	0.000
loc3		0.0281317	0.000
loc4		0.0740084	0.000
decil2		0.0644359	0.000
decil3		0.085552	0.000
decil4		0.1034398	0.000
decil5		0.1117389	0.000
decil6		0.1282033	0.000
decil7		0.1560298	0.000
decil8		0.1652177	0.000
decil9		0.1860697	0.000
decil10		0.2037653	0.000
sex		0.0159652	0.000
age_head		0.0026648	0.000
tot_mem		-0.0053484	0.000

Source: Developed by the authors, 2023.

4.3 Economic impact of households in Mexico not paying for the drinking water consumed

The fact that nearly one-third of households in Mexico do not pay for the consumption of drinking water has an impact on the finances of the water sector, and particularly on the activities performed by the WMOs, which are responsible for providing the service, and on their investments. It was also estimated that WMOs fail to receive approximately 790 million dollars per year, which corresponds to 32.3% of the total income that they could receive for providing drinking water services (Fig. 1). It is also interesting to note that while the four highest income deciles represent, on average, 18.0% of the households that do not pay for water services and the lower-income households represent 36.1%, on average, the losses from non-payment are very similar in economic terms. That is, the lower four deciles fail to pay a total of roughly 305 million dollars per year, while the highest four deciles fail to pay roughly 316 million dollars. Therefore, while there are more households in the lower four deciles that do not pay for the service than in the upper four deciles, the contribution to the total economic loss resulting from the decision not to pay is roughly equal for each group.

The three stages used in this research made it possible to show the problem with non-payment of the drinking water services received by households in Mexico, as well as its determinants in terms of both the quality of the service and the characteristics of the households. The research also found that while the problem of non-payment may be more closely associated with households in the lowest income deciles, households in the upper income deciles are responsible for nearly the same share of the monetary impact on the finances of WMOs, and thus on the wellbeing of the society. Nevertheless, the households in the lower deciles present other more reasonable motives for deciding not to pay for the service. Therefore, when wanting to implement policy incentives for households to pay for the drinking water that they receive, those incentives should be differentiated based on socioeconomic strata and the related variables.

Acknowledgments We thank the reviewers for reviewing the article.

Funding

The article was prepared with the technical and financial support of the National Council for Science and Technology (CONACYT in Spanish) through the CONACYT Chairs Program (Programa de Cátedras in Spanish) Nº 1812: “Economic and Water Model Project for the Valley of Mexico Basin”.

Ethical Approval

Not applicable due to the type of information considered.

Competing interests

The authors have no competing interests to declare that relevant to the contests of this article.

Authors’ contributions

Not applicable

Funding

The article was prepared with the technical and financial support of the National Council for Science and Technology (CONACYT in Spanish) through the CONACYT Chairs Program (Programa de Cátedras in Spanish) Nº 1812: “Economic and Water Model Project for the Valley of Mexico Basin”.

Availability of data and materials

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Akinyemi, B.E., Mushunje, A., & Fashogbon, E: Factors explaining household payment for potable water in South Africa. Cogent Sciences 4(1), 1464379 (2018).
Comisión Nacional del Agua (CONAGUA): Estadísticas del Agua en México 2018. Secretaría de Medio Ambiente y Recursos Naturales. Ciudad de México, México (2018).
Cook, J., Kimuyu, P. & Whittington, D: The cost of coping with poo water supply in rural Kenya. Water Resources Research 52: 841-859 (2016).
García, M: Tarificación óptima para el servicio de agua en las ciudades: aplicación a tres municipios españoles. En Tesis doctoral en economía. España: Departamento de Economía, Universidad de Oviedo (2002).
Instituto Nacional de Estadística y Geografía (INEGI): Panorama censal de los organismos operadores de agua en México. Censos Económicos 2018, México (2018).
Instituto Nacional de Estadística y Geografía (INEGI): Encuesta Nacional de Ingresos y Gastos de los Hogares 2020 (ENIGH 2020). Ciudad de México, México (2020).
Kooy, M., Tina, C. & Prabaharyaka, I: Inclusive development of urban water services in Jakarta: The role of groundwater. Habitat International 73: 109-118 (2018).
Revollo-Fernández, D.A., Rodríguez-Tapia, L. & Morales-Novelo, J.A: Impactos de los subsidios al agua en los hogares pobres de la Ciudad de México. Revista Gestión y Política Pública 28(1), 39-67 (2019a).
Revollo-Fernández, D.A., Rodríguez-Tapia, L. & Morales-Novelo, J.A: Economic value of water in the manufacturing industry located in the Valley of Mexico Basin, Mexico. Water Resources and Economics (2019b).
Silva-Ruiz, J: La eficiencia y la equidad en la fijación de precios de los servicios públicos: evolución de la teoría de la tarifa óptima en do spartes y el caso del servico de agua potable. Revista CIFE: Lecturas de Economía Social 12(16), 95 (2010).
Winpenny, J: Managing as an Economic Resource. Routledge. Londres (1994).

No competing interests reported.

Determinants of Household Payments for Drinking Water in Mexico, with Equity Considerations

Status:

Version 1

Abstract

Figures

1 Introduction

2 The Situation With Drinking Water Services In Mexico

3 Methods

3.1 Data

3.2 Analysis of the Information

4 Results

4.1 Who pays and does not pay for drinking water consumption in Mexico?

4.2 Determinants of household payments for drinking water consumption in Mexico

4.3 Economic impact of households in Mexico not paying for the drinking water consumed

5 Conclusions And Discussion

Declarations

References

Additional Declarations

Status:

Version 1