Introduction

Brazil is the country with the largest availability of renewable water resources in the world (8.65 trillion m3 per year), according to FAO data from 2017. Agriculture is the activity that uses water the most in Brazil, accounting for 68.4% of the flow consumed from water bodies (ANA 2019).

Although Brazil is well endowed with fresh water, its growing demand has caused concern. According to the survey "Atlas Irrigation—use of water in irrigated agriculture" (ANA 2017), the area equipped for irrigation soared from 455,429 to 6,954,710 hectares between 1960 and 2015, an increase of 1427%. The irrigated area is expected to reach 10.09 million hectares by 2030.

On the other hand, drought has also been a recurring problem. Drought events have been recorded since 1962 in the different regions of Brazil; however, only between 2012 and 2014 did drought events occurred simultaneously in the five geographic regions of Brazil. Droughts events from 2011, especially, were the most severe and intense of the last six decades, except in the South region. In São Francisco River Basin, particularly, those droughts led to inflows to the reservoirs at levels below the long mean term (LMT) and even below the minimum values ever recorded in the historical series, with severe consequences for water conflicts throughout the basin (Cunha et al. 2019).

In Brazil, charging for water use is a management tool from the national water resources law enacted in 1997, which has among its objectives to recognize water as an economic good, give its users an indication of its real value and encourage the rationalization of its use. However, the amounts charged are far from truly reflecting the economic value of water and inducing a rational and efficient use of water resources (Brito and Azevedo 2020). Even though water use charges for agriculture are very undervalued, as compared to the other uses, it is common for irrigators to complain about the prices imposed on them (Falcão 2019).

The objective of this work is to determine the economic value of water for irrigation, in order to reflect on the profitability and actual payment capacity of producers. The methodology adopted is the shadow price of water, which is methodologically defined and calculated as the ratio between the production net returns and the total amount of water used for irrigating the respective crops. In other words, it reflects the price that would need to be paid by farmers to veritably account for the actual value of water (Ziolkowska 2015).

The case study was conducted in the São Francisco River basin, which covers partly six Brazilian states, and has a strong agricultural vocation: the water withdrawals for irrigation granted by water authorities in the basin are 22.3 billion m3 per year, which represents 81% of the total withdrawals. São Francisco River basin has 64% of its area located in the Semiarid region, with annual average rainfall equal to or less than 800 mm.

In recent years, some studies have produced reliable estimates for the economic value of water for irrigation. Kelman and Ramos (2005) estimated the net economic return obtained per cubic meter of water applied in production of seven crops in the Brazilian Semiarid. Hellegers and Davidson (2010) determined the economic value of irrigation water for eight crops in the Musi sub-basin in India. Al-Karablieh et al. (2012) estimated the economic return of water of fourteen irrigated crops in Jordan. El-Gafy and El-Ganzori (2012) developed a decision support system to estimate the economic value of irrigation water for 45 crops in Egypt. Hosni et al. (2014) applied this system to evaluate the irrigation water value for 38 crops in the same country. Ziolkowska (2015) assessed the shadow price of water for irrigation of five crops—corn, cotton, wheat, soybeans and sorghum—in US High Plains. Ashayeri et al. (2018) studied the value of marginal product of irrigation water for rice in Guilan province, Iran. Takatsuka et al. (2018) examined the economic values of irrigation water use for South Florida cropland. Araya et al. (2019) determined the median and average return after variable cost (RAVC) for grain sorghum and maize (corn) in southwestern Kansas.

This paper seeks to contribute to the research on this topic by evaluating the economic value of irrigation water in a strategic basin for Brazil, which deals with recurrent droughts and has a high demand for water for agricultural purposes. The analysis is carried out for several crops in different periods.

Case study area—São Francisco River Basin

The case study is proposed in the São Francisco River Basin. From its source in Serra da Canastra, in the municipality of São Roque de Minas/MG, the São Francisco River runs about 2697 km and its basin, with drainage area of about 640,000 km2, covers 507 municipalities in the states of Minas Gerais (MG), Goiás (GO), Federal District (DF), Bahia (BA), Pernambuco (PE), Sergipe (SE) and Alagoas (AL), until it flows into the Atlantic Ocean, between Piaçabuçu/AL and Brejo Grande/SE (Fig. 1).

Fig. 1
figure 1

Case study area—São Francisco River Basin

Full size image

São Francisco River basin has 64% of its area located in the Semiarid region, totaling an average annual rainfall equal to or less than 800 mm, a Thornthwaite Aridity Index equal to or less than 0.50 and a daily percentage of water deficit equal to or greater than 60%, considering every day of the year.

According to ANA’s data from 2018, the water withdrawals for irrigation granted by water authorities in the basin are 22.3 billion m3 per year, which represents 81% of the total. Figure 2 shows the distribution of the extracted water among the irrigation systems.

Fig. 2
figure 2

Source: Author’s calculation based on ANA data from 2018

Water withdrawal by irrigation system in São Francisco River Basin.

Full size image

The top ten irrigated crops of the São Francisco River basin, in terms of water use, account for almost 70% of the total annual volume authorized, in a universe of more than 100 registered crops. These crops are corn, soybean, mango, beans, coffee bean, banana, cotton, sugar cane, papaya, and rice (Fig. 3). These were the crops selected for the application of the methodology of this paper. For each selected crop, a municipality and irrigation system in which it is common were defined to extract specific data for the analysis.

Fig. 3
figure 3

Source: Author’s calculation based on ANA data from 2018

Water withdrawal by crop in São Francisco River Basin.

Full size image

Methodology and data

Shadow price of water for irrigation

The shadow price of water for irrigating a given crop can be defined as the ratio between the production net return and the total amount of water used for irrigating this crop. Conceptually, the shadow price of water can also be viewed as the difference between a given water rate for irrigation and the actual economic value of water as a natural resource. In other words, the shadow price reflects the price that would need to be paid by farmers to veritably account for the actual value of water (Ziolkowska 2015).

In this work, the shadow price of water for irrigation is estimated by the Residual Value Method. The method is applied to situations where water is used as an intermediate input for production. Valuation of water in production is based on the idea that a profit-maximizing firm will use water up to the point where the net revenue gained from one additional unit of water is just equal to the marginal cost of obtaining the water (Lange and Hassan 2006). Residual Value Method assumes optimizing producers who can forecast the production function and prices of outputs and inputs other than water and who are assumed to add increments of each input up until the point where its value of marginal products (VMP’s) is equal to price or opportunity costs of the inputs (Young 2005; Young and Loomis 2014).

Residual Value Method considers a simple production process in which it is desired to assign a value for the unpriced input, water. Assume a single product denoted by \(Y\) which is produced by using a set of factors of production: purchased materials and equipment (\(M\)); human input, e.g., labor (\(H\)); equity capital (\(K\)); other natural resources, such as land (\(L\)); and the remaining factor: water (\(W\)). The production function is written as (Young and Loomis 2014):

$$ Y = f(X_{M} ,X_{H} ,X_{K} ,X_{L} ,X_{W} ) $$
(1)

Inputs and outputs are assumed to be continuously variable, and the level of technology is given and unchangeable. Production is modeled via a static, deterministic model of the profit-maximizing firm. Given competitive markets for the purchased inputs and perfect knowledge and forecast, prices for these inputs may be treated as known constants. Thus, the total value of the product is given by (Young and Loomis 2014):

$$ \left( {Y \cdot P_{Y} } \right) = \left( {P_{M} \cdot X_{M} } \right) + \left( {P_{H} \cdot X_{H} } \right) + \left( {P_{K} \cdot X_{K} } \right) + \left( {P_{L} \cdot X_{L} } \right) + \left( {P_{W} \cdot X_{W} } \right) $$
(2)

where \(Y \cdot P_{Y}\) represents the total value of product \(Y\), \(P_{i}\) represents the price of resource \(i\) and \(X_{i}\) is the quantity of the \(i^{th}\) resource. If one knows or can empirically estimate the appropriate values for all price and quantity variables, as well the quantity of water needed, one can determine the shadow price of water (\(P_{W}\)), also usually termed “economic value of water” or “net return to water” (Young and Loomis 2014):

$$ P_{W} = \frac{{\left( {Y \cdot P_{Y} } \right) - \left[ {\left( {P_{M} \cdot X_{M} } \right) + \left( {P_{H} \cdot X_{H} } \right) + \left( {P_{K} \cdot X_{K} } \right) + \left( {P_{L} \cdot X_{L} } \right)} \right]}}{{X_{W} }} $$
(3)

In this article, shadow water prices were estimated at an aggregate level. For each crop in Fig. 3, production costs, sales prices, productivity, and water requirements for a given location were obtained from average estimates, as explained below. This means that the results obtained reflect the average behavior of the agricultural sector for the selected crops in the analyzed regions.

Costs of production

The term \(\left[ {\left( {P_{M} \cdot X_{M} } \right) + \left( {P_{H} \cdot X_{H} } \right) + \left( {P_{K} \cdot X_{K} } \right) + \left( {P_{L} \cdot X_{L} } \right)} \right]\) of Eq. 3 corresponds to the value of production factors other than water, that is, it represents the costs of crop production of an amount of product \(Y\). These data were obtained from Companhia Nacional de Abastecimento (CONAB), which systematically publishes the agricultural costs of production of several crops in Brazil, expressed in reais per hectare (R$/ha).

CONAB's cost sheets are structured in such a way as to separate the components according to their accounting and economic nature. In accounting terms, variable costs are separated into farm costing expenses, other expenses and financial expenses, the latter levied on the working capital used. Fixed costs are differentiated into fixed capital depreciation and other fixed costs involved in the production and remuneration of land and fixed capital factors. In economic terms, the cost components are grouped according to their function in the production process, in the categories of variable costs, fixed costs, operating cost and total cost (CONAB 2010).

Among the variable costs, all components which participate in the process are considered as the productive activity develops, that is, those that only occur or affect if there is production. These are items of costing, post-harvest expenses and financial expenses, which constitute, in the short term, a necessary condition for the producer to continue in this activity. Fixed costs are differentiated in depreciation of fixed capital and other fixed costs involved in the production and remuneration of land and fixed capital factors (CONAB 2010).

The operating cost is composed of all variable cost items (direct expenses) and the portion of fixed costs directly associated with the implementation of the crop. It differs from the total cost only because it does not include the income from fixed factors, considered here as expected remuneration on fixed capital and land. It is a concept with greater application in studies and analyzes with medium-term horizons. The total cost of production comprises the sum of the operating cost plus the remuneration attributed to the production factors. In a long-term perspective, all these items must be considered in the formulation of policies for the sector (CONAB 2010).

Production costs are determined based on several steps which make up a rigorous methodology, detailed in CONAB (2010). The results of the estimates are organized in spreadsheets that are published each harvest. The spreadsheets are available for different cultures and different municipalities. All production factors (materials, labor, capital, land) are included in the spreadsheets, as well as the opportunity cost associated with each input (as if the money were used in another alternative investment). The data indicate the variable cost, the operating cost, and the total cost, in order to offer the conditions for studies of public policies and government programs, in addition to supporting technical discussions aimed at improving the production and commercialization process.

Table 1 shows a typical costs structure of an irrigated crop. It is observed that seeds, fertilizers and pesticides account for 56% of the total production cost.

Table 1 Corn production costs in Unaí/MG in March 2019.
Full size table

Productivity and price

In addition to estimating the cost of crop production, CONAB reports also exhibit the productivity of the crop considered, usually expressed in kilograms or tons of crop per hectare, in other words, the term \(Y\) of Eq. 3.

CONAB also systematically conducts price research for more than 100 agricultural products. The database contains thousands of series records distributed across all federation units. Market prices published, at the producer's marketing level, were adopted for the \(P_{Y}\) value in Eq. 3, expressed in R$ per unit of crop product (kilogram or ton, for instance).

Amount of water used

The volumes of water consumed by crops in the production process were also estimated. For this, the demands for irrigation were determined through agricultural water balances, based on the water needs for crops—which depends on the specific irrigation methods used—as well as the precipitation and evapotranspiration data in the region.

The water balances were based on the climatological normals of Brazil, which contain data on precipitation and evapotranspiration (ET) in several meteorological stations for the period 1961–1990. For each culture, its respective culture coefficient was identified, which means the ratio of ET observed for the crop studied over that observed for the well calibrated reference crop under the same conditions. Thus, depending on the irrigation method most practiced for each crop in the region and their respective reference efficiencies, water needs were estimated.

These agricultural water balances were calculated in previous studies for the same basin and the same crops of this paper (CBHSF 2016 and EMBRAPA 2000). In this way, the variable \(X_{W}\) was determined, expressed in cubic meters per hectare (m3/ha).

These are average estimates of water consumption, commonly used in irrigation projects for the purpose of requesting authorization to use water. Therefore, this work considers the average long-term theoretical water consumption of crops. It is important to note that above-average rainfall implies less irrigation and higher shadow prices, while in years of drought, with below-average rainfall, water abstractions are greater with a consequent reduction in the shadow price of water.

Limitations of shadow price approach

Although it is an established method and has several applications in previous studies, there are some limitations when using shadow price as a valuation method for water that deserve to be discussed.

As mentioned previously, using the shadow price methodology implies assuming that all markets (materials and equipment, labor, capital, land) are perfectly competitive except for the water market. If this assumption does not hold, part of production net returns (the residual value) will be attributed to other inputs instead of water, and the estimate of water value would be biased (most likely upwards). In other words, any input not accounted for would be considered as part of water’s economic value.

In this work, nonetheless, the data on production costs used include all possible expenses of the production process, from the initial phases of soil correction and preparation to the initial phase of selling the product. In Brazil, there are several subsidies’ policies to the agricultural sector: supply of inputs, low interest loans, technical assistance, etc. However, CONAB's production cost data include all theoretical expenses, even though, in practice, these are not fully absorbed by farmers. Therefore, it is reasonable to say that the methodology used in this work allows a reasonable estimate of the economic value of water for irrigation.

Still, some natural inputs are not directly valued, for example solar radiation and soil fertility. Even though these inputs may be implicit in the value of the land—which is accounted for—it is not appropriate to conclude that the estimated shadow price corresponds exactly to the actual and definitive economic value of water. On the other hand, this work does not intend to determine the actual economic value of water, but rather to reflect the price that could be paid by farmers for water without impairing their economic activity.

In addition, it is possible survey respondents tend to under-report profits or over reporting them depending on what is their perception about the survey.

Results and discussions

Shadow price of water in 2019

Table 2 presents the shadow price of water for irrigation of the ten major São Francisco River Basin crops in terms of water use. It also shows productivity, crop price, cost of production and water consumption, which appear on the right-hand side of Eq. 3, as well as the net return (product of productivity and crop price). For currency conversion purposes, the quotation of R$ 1 = US$ 0.26, referring to March 2019, should be considered.

Table 2 Shadow price of water for irrigation of the ten major São Francisco River Basin crops in 2019
Full size table

This methodology has some limitations, and the results must be carefully analyzed. Throughout the paper some caveats are presented. First, the basin has certain heterogeneity in its agroecosystems, and many of the production nuances are unlikely to be capture by cost-return studies on the crop products selected.

These results show that except for sugar cane, all water shadow prices were positive for the conditions analyzed. This means that the 2019/2020 sugar cane production harvest in São Miguel dos Campos/AL, with a market price of March 2019, had a profitless season. On the other hand, the other crops were profitable in the period. The production of mango in Curaçá/BA obtained the highest net return per m3 of water used (R$ 1.44/m3). This price is the maximum a farmer could pay for water and still operate at breakeven, that is, cover costs of the crop production. It is also worth mentioning the beans harvest in Unaí/MG, with R$ 1.34/m3. On the other end are the following crops: corn and soybean in Unaí/MG (R$ 0.14/m3 and R$ 0.11/m3, respectively) and rice in Igreja Nova/AL (R$ 0.06/m3), which obtained relatively low shadow prices, though positive.

For comparison, according to the São Francisco Basin Water Agency (Agência Peixe Vivo) data, irrigation users paid R$ 17,247,788 in 2019 for a total volume of extracted water equal to 8,797,302,778 m3. This means the average amount paid by farmers corresponding to the charging for water use in the São Francisco River Basin in 2019 was only R$ 0.002 per cubic meter, which is much smaller than the calculated shadow prices of water.

Table 2 shows that the highest net revenues do not necessarily correspond to the highest shadow prices of water, because of the water consumption variation. Papaya, for instance, which had the highest net return (R$ 59,382/ha), obtained the third highest shadow price of water (R$ 1.26/m3) because of its great water consumption (47,310 m3). On the other hand, beans had a relatively low net return (R$ 7972/ha) and a large shadow price (R$ 1.34/m3) due to its low water consumption (5,961 m3/ha). It is important to note that ratios between variables can be problematic and unstable measures. High water consumption can suggest a low shadow price and vice versa. Therefore, the shadow price should not be seen in isolation, but associated with the net return on production.

It was noticed that the approach challenges the assumption that under competitive equilibrium the marginal value of water is the same for all production processes in a region. Except for crop prices, which are determined by the market, farmers can influence the shadow price of water by varying crop productivity, costs of production and water consumption.

Improving crop productivity is key to increase net revenue and, consequently, the shadow price of water. In recent years, several authors have studied the most varied strategies to maximize crop productivity, among which one can cite Shen et al. (2013), Shrestha and Subedi (2019) and Bailey-Serres et al. (2019).

Production costs also have a strong influence on the shadow price of water. As shown in Table 1, fertilizers are the most expensive corn production factors, accounting for 28% of its production cost. According to Kanter et al. (2015), it is possible to reduce farmers’ fertilizer costs up to 20% by reducing the nitrogen application rate, adopting fertilizer best management practices (FBMPs) and applying enhanced efficiency fertilizers (EEFs).

Finally, the amount of water consumed by the crop directly impacts the shadow price of water. Producers should adopt irrigation management techniques to save water and increase the economic value of their production. Successful experiences of water saving in corn, banana and rice irrigation are reported in Xue et al. (2017), Panigrahi et al. (2019) and He et al. (2020), respectively.

For the sake of comparison, Table 3 presents the economic water values for irrigation of the crops studied in this work estimated by other authors, for other regions and periods. It is noteworthy that the comparison with the values of the past studies must be careful as the production processes could be quite different.

Table 3 Comparison with water values results found in other countries
Full size table

Inter-annual change in the shadow price of water between 2014 and 2019

The analysis presented in the previous section is static, valid only for the 2019 harvest. However, it is known that the sales prices of agricultural products, as well as their production costs, are variable in time, depending on supply versus demand and other factors. Thus, this article also evaluated costs and prices of previous years and the corresponding variation in the shadow price of water. The analysis was made from 2014 through 2019, because this is the time period with availability of sales prices disclosed by CONAB. The results corresponding to the crops for which a historical series is available—corn, soybeans, beans, cotton, coffee, and sugar cane—are presented graphically in Figs. 4, 5, 6, 7, 8 and 9. In these figures, blue bars represent positive shadow prices, while red bars represent negative shadow prices.

Fig. 4
figure 4

Shadow price of water for corn irrigation (2014–2019)

Full size image
Fig. 5
figure 5

Shadow price of water for soybean irrigation (2014–2019)

Full size image
Fig. 6
figure 6

Shadow price of water for beans irrigation (2014–2019)

Full size image
Fig. 7
figure 7

Shadow price of water for cotton irrigation (2014–2019)

Full size image
Fig. 8
figure 8

Shadow price of water for coffee bean irrigation (2014–2019)

Full size image
Fig. 9
figure 9

Shadow price of water for sugarcane irrigation (2014–2019)

Full size image

The results show that the shadow price of water varies widely year by year, due to the high annual change in production costs and revenues. In absolute terms, the crops with the highest variation in the shadow price of water in this six-year period was beans (R$ 0.97/m3) and coffee bean (R$ 0.89/m3). In percentage terms, corn and sugarcane were the crops whose shadow prices varied the most in relation to the six-year average Table 4).

Table 4 Variation of shadow price of water in the period between 2014 and 2019
Full size table

The results for the 6-year period suggest that the value of a given shadow price of water represents solely the economic value of water for irrigation of a specific harvest in a specific year. In other words, it is generally not possible to infer the medium–long-term behavior of a crop from the shadow price of water calculated for a single year, due to the high variability in production costs and revenues.

It is important to note that the differences in shadow prices are due solely to changes in production costs and sales prices over the years. Climatic factors such as precipitation and evapotranspiration were considered static, so there was no variation in water consumption by crops. It is known that, for instance, the effect of a drought would increase the need for irrigation and reduce the shadow price of water, but this work, when using the climatological normals in the simulations, considers the medium long-term behavior of crops from the point of view of water need.

Except sugarcane, which has obtained negative shadow prices of water in five of the six years analyzed, meaning that production was not profitable in these years, all crops have been profitable most of the time. Corn, beans, and cotton obtained negative shadow prices in a single year, while coffee production was profitable throughout the period.

To identify the reasons for the non-profitability of sugarcane in the analyzed period, a search was made for publications from the sugar and alcohol sector, as well as interviews with representatives of the sector. First, it was identified that this behavior was also observed in the Center-South Region of Brazil, where from 2014 to 2019 only the 2017 harvest was profitable, as presented in an article by Carvalho (2019). The author points out that the cost of sugarcane production in the last 11 years has increased by 177.4%, due to the intensification of mechanization, while productivity has decreased by 12.5%, due to the expansion to areas not yet appropriate for the crop, to conclude that the activity is totally "uneconomic". The increased costs of fertilizers, fossil fuels and automotive maintenance, in addition to the devaluation of the real currency, are other external factors identified.

According to a report by the Confederation of Agriculture and Livestock of Brazil (CNA 2019), the farmer's income has been restricted by the systematic increases in production costs, by the low renewal and aging of the cane fields, by the decline in productivity and, mainly, by the lag in the participation of the cost of the raw material in the formation of the price that remunerates the sugarcane. Another report (CNA 2016) highlights that the drop in productivity was due to climatic adversities, soil compaction and low renewal rate of the cane field. This combined with the drop in the quality of the raw material, decreased profitability and increased farmers' production costs. On the other hand, the sale price of products (sugar and ethanol) did not keep up with the increase in production costs, which prevented new investments in the activity.

According to a magazine article (JornalCana 2019), the low remuneration of sugarcane in the 2008 to 2018 harvests in the interior of São Paulo state is explained by the drop in the price of sugar in the international market (due to the global surplus in the production of this commodity) and in the price of petroleum, which has a direct influence in the price of ethanol. In turn, the increase in costs is explained by the unfavorable weather conditions and the increase in pests.

According to PECEGE (2020), the costs of inputs and machines underwent significant increases in the 2018/2019 harvest and, therefore, contributed to the increase in agricultural costs. The factors that led to the increase in the price of inputs are related to the restriction in the global balance of fertilizers and pesticides and to the devaluation of the Brazilian currency (real) in the period in question. The cost of machinery, in turn, was impacted by the increase in the international price of oil, which raised the price of diesel. Such rise was particularly important in the most intensive stages in machinery, namely, soil preparation and CTT (cutting, transshipment and transportation).

To illustrate, Fig. 10 shows a great increase, from 2017 to 2019, in the prices of three factors of production: operation with own machines (irrigation), fertilizers and pesticides.

Fig. 10
figure 10

Variations in costs of sugarcane production factors

Full size image

Conclusions

This paper evaluated the economic value of water for irrigating each of the ten major crops of São Francisco River Basin by using the "shadow price of water" approach, based on the Residual Value Method. The results show that water has a high economic value for irrigated agriculture in the case study area. Except for sugarcane production, which showed a negative shadow price of water in 2019 (R$ -0.13/m3), meaning that this year's crop was unprofitable, all other crops made a profit, presenting significant shadow prices, especially mango (R$ 1.44/m3), beans (R$ 1.34/m3) and papaya (R$ 1.26/m3).

Results also show that water shadow price is highly influenced by crop productivity, costs of production and water volume used. Therefore, farmers should invest in optimizing these variables to maximize the economic value of their production.

This research has also added as a novelty the evaluation of the shadow prices of water over a six-year period (2014–2019). By so doing, it was observed that corn and sugarcane were the crops whose shadow prices varied the most in relation to the six-year average: 365% and 316%, respectively. This analysis shows that water shadow prices vary widely depending on the high annual variability of production costs and revenues.

In general, the shadow prices of water determined in this work were quite significant. These values represent how much farmers could pay for water and still breakeven. On the other hand, water for irrigation in Brazil—especially in São Francisco River Basin—is underpriced. The average amount paid by farmers corresponding to water charges in the São Francisco River Basin in 2019 was only R$ 0.002 per cubic meter, regardless of its shadow price. Other water user sectors pay considerably higher amounts: public supply (R$ 0.010/m3), sewage (R$ 0.025/m3), industries (R$ 0.013/m3), thermoelectric (R$ 0.018/m3). Even so, these prices seem low if we consider the users’ payment capacity. For instance, according to de Lima et al. (2019), who analyzed water charges in the dairy industry, the amount paid per kg of mozzarella cheese produced in the São Francisco River basin is R$ 0.0038, while the sale price of 1 kg of mozzarella cheese in the state ranged from R$ 14.11 to R$ 15.82.

There is, therefore, a comfortable space for more realistic water prices for irrigating water users in Brazil respecting their payment capacity. We believe that public prices should not be determined solely based on economic models; instead, decision makers, notably the Hydrographic Basin Committees, who by the Water Law are responsible for defining water charges, should at least consider the economic value of water in the charging schemes in order to encourage an increasingly rational and efficient use of water.