Whose Values Count? Demand Management for Melbourne's Water*
Abstract
An assessment is provided of Melbourne's rising block tariff for water and its prescriptive regulation of water use through non-price restrictions. The three-step rising block tariff fares poorly on economic efficiency and on widely used notions of equity. A two-block tariff could remove much of the inefficiency in a three-block tariff, but one-price-for-all water would be the most efficient way to ration water on conventional definitions. The non-price restrictions on water use are prescriptive, denying households the opportunity to decide how to make any cuts in water use required of them. State values usurp private values.
I Introduction
Melbourne has had water restrictions since 1 November 2002. The immediate reason for their introduction was depletion by drought of water in storage. However, the Victorian government has expressed concern that the combined effects of population increase and possible effects of climate change on water supplies, along with the traditional problem of variable runoff into water storages, could mean Melbourne will be approaching its supply limits within 15 years (Victorian Government, 2004: 94). The government recently reviewed options for Melbourne's water economy, in the context of an overall assessment of Victoria's water future, with green and white papers prepared on water (Victorian Government, 2003, 2004). For Melbourne's water economy key outcomes included:
- •
A new dam would be a last resort.
- •
A target per capita reduction of 15 per cent in consumption of drinking-quality water by 2010, compared with the 1990s average, to be achieved by ‘smarter’ use of water.
- •
A target for recycling 20 per cent of Melbourne's waste water by 2010.
- •
Retention of a regulatory regime for managing water in Melbourne, with innovations being a rising block water tariff for Melbourne households and the introduction of ‘permanent’ water restrictions.
- •
Imposition of an ‘environmental levy’ on Melbourne's government-owned water retailers to fund environmental and other water-system improvements throughout the state.
- •
Continuation of the policy of not allowing trade in water between the irrigation and Melbourne water systems.
The aim of this paper is to examine the regulatory demand management regime for Melbourne's residential sector, which accounts for about 60 per cent of the city's water use (Victorian Government, 2003: 42). With one significant qualification, the supply side of the water economy is not considered. The qualification is to recognise that some approaches to containing demand for water are more conducive to efficient sourcing of new water supplies than others. The focus in the paper is on providing a critique of the main measures – the three-tier rising block tariff (RBT) and the prescriptive restrictions on water use1– used to reduce water consumption by households.
In assessing the demand management measures for water in Melbourne, two criteria are used. The main criterion is economic efficiency. Economic efficiency requires the equating of social marginal values of water for all water users and uses. The proposition is accepted that, with correspondence between marginal private and marginal social valuations of water, water use can be reduced at lower social cost through non-prescriptive measures such as price increases or household quotas than through prescriptive ones. Non-prescriptive measures, unlike prescriptive ones, give water users the freedom to decide where and how to reduce their water use. The second criterion used is equity. Views differ on the requirements for equity, including on whether equity is appropriately pursued through policy for individual commodities such as water. It is suggested that equal treatment of households using water in different ways and ‘equal’ treatment of households of different sizes should be significant considerations if equity objectives are addressed through demand management measures for water.
Alternatives to the current three-block RBT and prescriptive non-price restrictions for managing demand for water include a uniform water price, a two-block RBT and household water quotas. Each of those measures could potentially enhance the social value obtained from water. Household water quotas would involve major administrative difficulties, including moral hazard issues, and are not considered further. The reintroduction of a uniform water price or the adoption of a two-block RBT would be modest reforms in that they lie within the regulatory paradigm that characterises the demand management regime in Melbourne and other Australian cities. Consideration of more thorough-going reforms that would introduce contestability to the functions of the urban water system – water sourcing, storage, wholesale and retail water services – is beyond the scope of the paper. Because the approach in regulating Melbourne's water implies that the social marginal value of water is often very different from private marginal values, consideration is given first to the question ‘whose values count?’
The assessment of Melbourne's demand management measures for water and the consideration of alternatives provided here is an analytically simple case study, using standard microeconomic principles. Quantifying the efficiency costs of the measures is not attempted. That would be difficult because of the multiplicity of measures used, interactions between them, and the need to obtain detailed information on the number and the demand characteristics of households impacted by the different measures. As with other topics, a principles-informed case study of approaches to reducing consumption of urban water can be useful, even insightful, sans empirics – especially, perhaps, when much of the recent literature is of the ‘government promoting its smart-water initiatives’ genre!
II Whose Values Count?
Central to assessing the economic efficiency of a demand management regime for water is the judgement made on whether individuals’ values should be the basis of comparing the marginal value of water for different users and uses, or whether individual valuations should be overridden by values determined by water authorities or state governments. The values assigned to private goods by individual consumers are conventionally taken to be the economically and socially relevant values. Although there are important public goods to be considered in managing water, including health and environmental public goods, water used in urban settings exhibits the two key characteristics of private goods – competition in consumption and excludability – and economists have usually viewed the private valuations of individual water consumers as appropriate (e.g. Harberg, 1997; Agthe & Billings, 2003; Freebairn, 2003). Harberg (1997), writing in the context of a need to reduce water consumption in a dry period, said:
‘… the tastes and preferences of individual customers may vary widely … a particular residential customer might prefer never to wash a car in order to provide water for a cherished garden. Other customers might cheerfully let all landscape die in order to retain longer-than-average showers. These taste differences are generally of no concern to the water provider.’ (p. 45)
Acceptance that individual valuations of water are appropriate for allocating water has the policy consequence that the adjustment of demand to available supply in times of drought would be made by non-prescriptive measures. In Melbourne – and in other Australian cities – however, urban water policymakers have opted to ban some water uses, to restrict other uses, and to leave yet other uses of water unrestricted.
The Victorian government has not justified the pattern of water restrictions – and its implicit judgements about the social value of water in dif-ferent uses – by differences in externalities between water uses. The nature of the water restrictions does appear to have been influenced by the notion that ‘essential’ water use is more socially valuable than ‘discretionary’ water use (for example, Victorian Government, 2003; pages 90, 98; Victorian Government, 2004; p. 127). The use of water inside the house for drinking, cooking, washing, laundering and for toilets is viewed as essential, whereas the use of water outside for watering gardens and lawns and washing paths, patios and cars is seen as discretionary. Outside use of water accounted for about 30 per cent of water used in the average Melbourne household before the introduction of restrictions (Victorian Government, 2003: 38, 2004: 92).
Given the government's emphasis on reducing water use in Melbourne, and its unwillingness to rely mainly on non-prescriptive measures to do this, some form of prescriptive rationing is necessary. If rationing takes the form of bans and other restrictions on particular uses of water, it has to be possible to enforce them. Without eavesdropping by government inside private homes, the rationing needs to be confined to outside uses of water.2 This constraint in policing water restrictions may partly explain the form of the restrictions, and the government's judgements, implicit in the regulations, about the social value of water in different uses.
Notwithstanding that the use of water for washing is viewed as essential and is hidden inside the house, the government has implied that the social value of water used in the bathroom is less than its private value. A full-page advertisement in Melbourne's The Age newspaper on 28 June 2004, authorised by Premier Steve Bracks, said Melburnians needed to take fewer baths and that their average shower time, already reduced by 1 min to 6 min, needed to be reduced to 4 min3. In early 2005, an illuminated sign near the Tullamarine freeway advised that a 3-min shower saves 200 glasses of water. Thus far these exhortations are in the realm of suasion, untranslated into regulations imposing state values on private acts of consumption!
A ban on the use of water for a particular purpose can be taken to mean that, in the prevailing circumstances, the net marginal social value of that use of water is determined by the government to be zero or negative. However, the government has gone beyond this circumstances-specific prohibition on the use of water, saying: ‘Behaviour such as hosing driveways and footpaths must become actions of the past’ (Victorian Government, 2003: 43). The white paper contains the assertion ‘The introduction of permanent water savings measures recognises that there are currently unacceptable uses of our precious water resource.’ (Victorian Government, 2004: 98). Those statements suggest that, even though some Melburnians place a high value on them, and even though significant external costs have not been established, certain water uses are viewed by the government as having negative social consequences that offset the private benefits regardless of the amount of water available. That takes the usurping of private preferences by state determinations to a higher order than temporary prohibitions that end with the restoration of water supplies.
In Melbourne, there is no clear policy of varying the price of water inversely with the level of water in storage.4 Taken on its own, this suggests that the state sees the social value of water as independent of the scarcity of water. However, with the regulatory restrictions on water use being made tougher as the volume of water in storage drops, a schizophrenic syndrome is revealed: the social value of water for rationed uses is reduced (to zero or less in the case of water for prohibited uses) whereas it is unchanged in unrestricted uses. From 1 October 2004, however, Melbourne householders have had a rising block tariff. What that pricing approach means for the private and social value of water is examined in the next section.
The approach to determining the value of water in the urban sector contrasts with the approach in the irrigation sector. ‘Put simply, instead of the market determining what represents a higher value activity (as in irrigation), the government has chosen to specify what behaviours are of greater value to the community in urban settings’ (Crase & Dollery, 2005: 18–19).
III Demand Management Measures
In this section, the main demand management measures used recently for water in Melbourne – the rising block tariff (RBT) and regulations that restrict the use of water – are examined.
(i) Rising Block Tariff
In its white paper, the government said ‘a smarter pricing system in Melbourne’, a rising block price scheme, would be introduced for water in October 2004 (Victorian Government, 2004: 125). Part of the context was that many householders thought the price of water was too low to encourage its conservation (Yarra Valley Water, 2004: 91). Under the RBT, the marginal water price fell from $0.79 per kL to $0.75 per kL (down 3.8 per cent) for households using up to 40 kL of water a quarter. The choice of 40 kL for the first block ‘is based on an estimate of essential indoor use’ (Victorian Government, 2004; p. 127), although the household size was not stated. For households using 40–80 kL of water, the marginal price rose from $0.79 to $0.88 per kL (up 11.4 per cent), and for households using more than 80 kL, the marginal price rose from $0.79 to $1.30 per kL (up 64.6 per cent).5 The white paper suggested that RBTs ‘are widely regarded as the fairest and most effective way to price water for conservation. Not only do they encourage water conservation, but they also recognise the need to provide water for essential domestic use at an affordable price’ (Victorian Government, 2004: 127).
Economic Efficiency
According to the white paper, the rising block tariff would ‘reward efficient users and penalise excessive use’ (Victorian Government, 2004; p. 125). ‘Efficient users’ here means not those for whom the marginal value of water is high (economic efficiency), but those using small amounts of water. Similarly, ‘excessive use’ means not that the marginal value of water to users is low, but that more water is used than the government considers appropriate.
A three-level pricing scheme ensures that the marginal value of water cannot be equal across all households, a condition for efficient water use, unless the state denies the validity of private valuations in the case of water and imposes its own. Moreover, the necessary adjustments required to bring the three different marginal private valuations of water for consumers in each of the three blocks to a common marginal social valuation would fail any ‘reasonable assumptions’ test. To illustrate, the excess of marginal social valuation of water over marginal private valuation for a wealthy retired judge living alone and using little water (consumption within the block-one volume) would have to be larger than the excess for a struggling family of four within block two.
Regulating for a three-level price scheme in Melbourne creates incentives for trade in water between neighbours on different water price blocks – a reality not acknowledged in the green or white papers. On the usual assumption that water prices reflect marginal private, and social, valuations, interhousehold water trade would be efficiency enhancing. However, that would not be so if the price structure introduced with the RBT did in fact result in a common marginal social valuation for water across all users; interhousehold trade in water would then be efficiency reducing. The small potential gains to households from water trade will likely mean the amount of trade is small.
With the watering of gardens and lawns, many households are expected to use more than 80 kilolitres in the summer quarter, and pay the block-three tariff for usage over that volume.6 In effect, for many households, the block tariff scheme represents price discrimination against the use of water for lawns and gardens in the ‘garden state’!
Although this paper is concerned with demand management, price plays the dual role of rationing demand and eliciting new water supplies. Decisions on the sourcing of supplementary water supplies through changing catchment management, reducing system losses, recycling waste water, harvesting runoff from rooves, desalination, building new dams, intercatchment water trade and other means can only be evaluated satisfactorily from an overall economic perspective if the price that rations demand also signals the value of extra water supplies. The introduction of the block tariff, with three different marginal water prices for different groups of consumers, complicated further the task – already difficult because of the mix of price rationing and restrictions on use to reduce water consumption – of assessing options for adding to water supplies for Melbourne.
A Two-block Tariff
Although economic efficiency points to use of a uniform price for all water sales, a two-block price scheme may be worth considering if holding down household outlays for a ‘base’ volume of water is seen as important for social reasons. The efficiency costs of rising block pricing would be much reduced if the number of blocks was kept to two, with the volume of water supplied at the block-one price sufficiently low that nearly all households were buying block-two water. Then, nearly all households could be supposed to have a marginal private valuation of water equal to the block-two price, and the scope for win–win trade in water between households would be small. With almost all households paying the block-two price at the margin, it would be easier to see pricing policy as consistent with the government's view that Victorians understand that the prices they pay for water have not reflected ‘the true value of the resources they use’ (Victorian Government, 2004: 126).
If base water corresponded to the estimated essential requirements for inside use, its price elasticity of demand would be very low. ‘Extra’ water would include a substantial volume used on gardens and lawns, and the marginal minutes under the shower; these would be expected to face more price-elastic demand.
Suppose price is the sole means used to ration water. If price increases for water applied entirely or mainly to the excess water, a given reduction in water use could be obtained with a smaller average price increase and a smaller consumer welfare loss, as measured by consumer surplus. This is illustrated in Figure 1. Householders’ demand for water, Dt, is disaggregated into base water, Db, and extra water, De, with De more price-elastic than Db. To reduce aggregate consumption from Q to Q′ with uniform pricing for all water, price must increase from P to P′. Consumer surplus for water users is reduced by area PP′CD per period. If the reduction in water use is obtained by increasing price for block-two water only, the price for that water increases from P to P″, and consumer surplus falls by PP″EF, which is less than with uniform pricing. (Note that (Q − Q′) = (Qe−Qe″)).
Reducing Water Consumption Using Price, with and without Price Discrimination
Equity
Spending on water and sewerage rates and charges was just over $1 per day for the average Victorian household in 2003–2004, less than 1 per cent of total spending on goods and services (see Table 1). For households in the two lowest income quintiles, presumably the households whose water bills the government would most want to hold down, spending on water was less than $1 per day, a little over 1 per cent of total spending.
| Household gross income quintile | ||||||
|---|---|---|---|---|---|---|
| 1st | 2nd | 3rd | 4th | 5th | All | |
| a. Household gross income cut-off ($ per week) | 417 | 723 | 1144 | 1726 | n.a. | n.a. |
| b. Expenditure on water and sewerage rates and charges ($ per week) | 5.33 | 6.37 | 7.84 | 9.58 | 10.54 | 7.93 |
| c. Total expenditure on goods and services ($ per week) | 416 | 594 | 859 | 1062 | 1511 | 888 |
| d. b. as per cent of c. | 1.28 | 1.07 | 0.91 | 0.90 | 0.70 | 0.89 |
- Source: Australian Bureau of Statistics, 2003–2004 Household Expenditure Survey, ABS data available on request.
How large would the price increase for drinking-quality water need to be to reduce per capita consumption by 15 per cent against average 1990s water use, the government's target for 2010? Assuming that a uniform water price was the sole measure used to reduce water use, and a price elasticity of demand for water in the range −0.2 to −0.5,7 the price increase required to reduce consumption by 15 per cent would lie between approximately 30 per cent and approximately 75 per cent. However, because there was a downward trend in per capita water use in the 1990s, and further reductions at the beginning of the 2000s, approximately half the target reduction had been achieved before water restrictions commenced late in 2002 (Victorian Government, 2004: 96).8 Hence, had price been used instead of water restrictions and associated measures to achieve the consumption target, the estimate of the required price increase would be around 15–37.5 per cent.
Because a Melbourne household's water bill comprises service charges and user charges, the percentage increase in its water bill in response to a given percentage increase in the (uniform) price of water would be smaller than the price increase for water, even with a zero price elasticity of demand for water. With some elasticity in water demand, the percentage increase in the typical household's water bill with a given increase in the price of water would be smaller again. The reality that service charges are a significant part of the water bill for most households together with the unimportance of water in household spending provides grounds for questioning the equity case for depart-ing from an efficient, simple one-price-for-all potable water.
Putting aside the issue of whether a uniform water price would have a substantial adverse effect on low-income households, there remains a significant question about the RBT from the view of equity between households. Why should a household of two using 200 L a person per day pay a marginal water price of $0.75 per thousand litres, whereas a household of five using 200 L each a day pays a marginal price of $1.30?9 O’Brien (2005[link]: 10) calls it ‘grossly unfair’. Notwithstanding that households of five or more may be eligible for special assistance to help them reduce their water use, the question arises whether water pricing policy is seen as a way of discouraging larger families and households?
Summary
Melbourne's rising block tariff ensures (unless bizarre assumptions about adjusting private marginal values to social marginal values are accepted) that water will be used inefficiently in an economic sense; it creates incentives for trade in water between households; and it discriminates between households in a way that many will see as inequitable. Boland and Whittington (2000) and Sibly (2006) were also critical of RBTs.
(ii) Regulation of Water Use
Since the introduction of stage 1 water restrictions in November 2002, Melbourne households have been subject to regulatory restrictions on the use of water. These are examined below.
Bans and Other Restrictions
In considering the effects of Melbourne's water restrictions, summarised in Table 2, on economic efficiency several points are relevant.
| Level | Period | Main measures |
|---|---|---|
| Stage 1 | 1 November 2002–31 July 2003 | Washing paved areas – banned except in emergencies. |
| Washing cars, windows, facades – to be performed with a high-pressure hose or with buckets/cans filled directly from taps. | ||
| Rinsing with a regular hand-held hose with trigger-nozzle permitted. | ||
| Watering gardens and lawns – use of manual watering systems only between 5 a.m.−8 a.m. and 8 p.m.−11 p.m., and automatic systems between 11 p.m. and 6 a.m. | ||
| Swimming pools and spas – approval to fill pools and spas of 2000 L or more dependent on saving equivalent water elsewhere. | ||
| Stage 2 | 1 August 2003–28 February 2005 | Stage-1 restrictions plus: ban on watering lawns, ban on water toys using water from a hose. |
| ‘Permanent’ | 1 March 2005 to? | Washing paved areas and timber decking – banned ‘except in exceptional circumstances’. |
| Watering gardens and lawns – manual watering systems to operate only between 8 p.m. and 10 a.m., and automatic systems between 10 p.m. and 10 a.m. | ||
| Mandated fitting of rain or soil moisture sensor with watering systems installed from 1 September 2005. | ||
| Hand-held hoses used to water gardens and lawns and to wash cars to be fitted with trigger nozzles. | ||
| ‘Equivalent savings’ condition for approving filling pools and spas of 2000 L or more applies to new pools only. |
- Source: Melbourne Water (2006) and earlier years.
First, the restrictions do not require households to use less water than in their absence, although for households in aggregate that is the expected outcome. Second, for many people, the restrictions on hours of operating their watering systems would have no effect on their behaviour, and perhaps be seen by them as costless, even though their lawful options are reduced. For others, the restrictions would involve inconvenience costs dependent on the particular circumstances of individual households.10 Third, the greater technical efficiency of watering lawns and gardens in the early morning and at night, due to lower evaporative losses, does not, if individual values have primacy, provide an economic justification for confining watering to those times. Fourth, the water restrictions are prescriptive, requiring households to adhere to regulations that overall, although not for each household, reduce water use. This regulatory regime is inconsistent with obtaining the greatest value from the water resource if individual marginal valuations of water in different uses are accepted as the best guide to social marginal valuations. During the stage 2 restrictions, many Melbourne householders would likely have welcomed the opportunity to use a bit less water for unrestricted inside uses, or restricted outside uses, and use it instead to water their lawns. The restrictions on water use result in different marginal private valuations for water in different uses within households – as the rising block price scheme leads to differences in marginal private valuations of water between households.
Can the regulatory package be seen as equating marginal social valuations of water across uses within households? Realistically, no. The prohibition on watering lawns implied that the marginal social value of water in that use was zero. The absence of restrictions on inside use of water could be viewed as indicating that the marginal private and marginal social valuations for that water coincide. Do the ‘permanent’ restrictions on hours of watering gardens mean that marginal private and marginal social valuations of water in that use will forever coincide between 8 p.m. and 10 a.m. – or between 10 p.m. and 10 a.m. in the case of households using automatic watering systems – whereas marginal social valuation outside those hours will always be zero? The permanent prohibitions on watering outside approved hours and on washing paved areas will impose high disutility costs on some in times of water scarcity and abundance.
Swimming Pools and Spas
The regulation of water for filling swimming pools and spas takes a different form to regulation of other water uses. Filling a pool or spa requires written approval from the water retailer.11 To obtain approval, it is necessary to commit to a plan to achieve savings in water use in other areas of the household equivalent to the water needed for the pool/spa. Enforcement is by random audits by the water retailers. This ‘offsets’ arrangement represents internal trade – trade within the household. Householders can ‘buy’ an exemption from the prohibition by making water economies elsewhere if they value the pool experience highly enough.
The offsets arrangement for swimming pools and spas, considered on its own, is conducive to freedom and to a more efficient system of water rationing. Introducing an offsets mechanism is a step towards a non-prescriptive household water quota. However, the offsets system for pools generates two questions:
- •
Why should there be an offsets option for swimming pools and spas but not other prohibited uses of water, such as watering lawns under the stage 2 restrictions? Are the preferences of people who place a high value on a green lawn in summer judged more socially unacceptable than those of people who assign a high value to a full pool?
- •
Why should internal trade be viewed as an acceptable way to obtain the right to fill a swimming pool but not external trade? That is, why shouldn't a household be able to pay other households to make the required offset savings in water use if that is cheaper than doing so themselves?
These questions highlight the logical consequences of an initial step towards allowing householders to use the (internal) market to find a lower-cost solution while still achieving the water savings of a prescriptive regulation. Why should further steps towards providing options to householders not be taken and reliance be placed on non-prescriptive measures to reduce water use?
IV Conclusion
Unless assumptions about adjusting private marginal valuations of water to obtain social marginal valuations are accepted that would normally be rejected, Melbourne's three-level rising block tariff for water is inconsistent with economic efficiency in the allocation of water between households. With usual approaches to equity, the rising block tariff scores poorly on that criterion also. Similarly, with the conventional economic approach of accepting individuals’ marginal valuations of resources in different uses, the prescriptive restrictions on households’ water use in the city ensure economic inefficiency in the allocation of water within households. Melbourne's demand management regime for water therefore does not allow the maximum potential value to be obtained from the available water unless the state's implicit valuations are accepted as overriding private valuations in a way that would not generally be contemplated for private goods.
Given the very small share of water in households’ expenditure, even for low-income households, it is appropriate to ask ‘are the costs of reducing water use under the current regime excessive?’ The cost that has been the focus in this paper is the reduced value obtained from urban water on conventional, liberal approaches that give primacy to individuals’ marginal valuations of water use. The introduction of ‘permanent’ water restrictions will mean the denial of highly valued water uses for some regardless of the scarcity of water. Other costs include the costs of developing, communicating, administering and enforcing the water restrictions; the costs of administering a complex regulatory price system; the costs of education and moral suasion directed to reducing water consumption; and the costs of reduced progress on the COAG/NCP goal of reducing cross-subsidisation of water. Any required rationing of water could be achieved efficiently entirely by the first-best method of increasing a (uniform) water price. In the absence of non-price restrictions, and with a price elasticity of demand for water between −0.2 an − 0.5, a price increase between about 15 per cent and 37.5 per cent in a uniform water price might have achieved the target 15 per cent reduction in per capita water consumption when allowance is made for the target being approximately half-met when the water restrictions were introduced. Not only would administration and enforcement costs be much reduced with that approach, but the clear marker of the value of water provided by reliance on a single price to reduce demand would help greatly in the supply side objective of sourcing new water supplies in the most cost-effective way.
A compromise between relying on a simple, efficient, single price for water to ration demand and the present multipart regulatory and prescriptive regime would be use of a two-block tariff, under which almost all households faced the block-two price at the margin. That would allow the political objective of providing a base amount of ‘essential’ water at a low price to be achieved and it would have several advantages over the current demand management regime: it would be non-prescriptive; it would reduce cross-subsidisation; and it would provide incentives for economically efficient use and sourcing of water on conventional definitions.
