Buybacks to Restore the Southern Murray-Darling Basin

Abstract

We use TERM-H2O in analysing the effects of the Australian Government buying back water from irrigators in the Southern Murray-Darling Basin (SMDB) and thereby increasing river flows. Results are explained using data from the model and simplified theory. We refer to this as the ‘back-of-the-envelope’ approach. Back-of-the-envelope calculations and regressions allow us to explain key features of the results including differences in regional outcomes. Controversially, our results suggest that buyback would increase economic activity in SMDB. Although a scheme of environmentally useful size would sharply increase the price of irrigation water, there would be little effect on aggregate SMDB farm output. Instead, farm resources would be reallocated between activities. Because farmers are owners of water rights, they would benefit from the price increase induced by buyback. Community anxiety in the basin over buybacks may have arisen because the buyback process started during a period of drought-induced stress.

This chapter reproduces with permission substantial portions of Dixon et al. (2011).

Appendix: Calculating the Price of a Permanent Right to a Unit of Irrigation Water in the Southern Murray-Darling Basin

We calculate the price [PPerR(t)] that a farmer would need to receive in year t (t  =  2009, …, 2016) to induce him/her to give up the permanent right to an annual allocation of one unit of irrigation water according to:

$$\text{PPerR}\left(\text{t}\right)={\displaystyle \sum _{\text{y}=\text{t}}^{\infty }\frac{\text{E}[\text{P}(\text{y})]*\text{E}[\text{S}(\text{y})]}{{(1+\text{d})}^{\text{y}-\text{t}}}}\text{}\text{t}=2009,\dots,2016$$

(6.14)

where

• E indicates expectation

• P(y) is the price of water in year y

• d is the discount rate (assumed to be 0.08 reflecting 3% inflation and a 5% real rate of interest)

• S(y) is the share of water rights in year y that is in fact allocated

As mentioned in Sect. 6.2, the S(y)s in 2009, 2010 and 2011 were assumed to be 0.7, 0.8 and 0.9, reflecting drought conditions that have made delivery of full water allocations impossible. For 2012–18, we set S(y) at one.

We assume that the expected values for P(y) and S(y) are given as follows:

$$\text{E}[\text{P}(\text{y})]=\text{PS}(\text{y}),\text{}\text{y}=2009,\dots,2018$$

(6.15)

$$ \text{E}[\text{S}(\text{y})]=\text{S}(\text{y}),\text{}\text{y}=2009,\dots,2018 $$

(6.16)

$$\text{E}[\text{P}(\text{y})]=\text{PS}(2018)*{1.03}^{\text{y}-2018}*\text{SF}(\text{y})\text{}\text{y}>2018 $$

(6.17)

$$\begin{array}{l}\rm{E}[\rm{S}(\rm{y})]=\rm{S}(\rm{t})\rm{y}>2018,\rm{ t}\in \left\{2009,\dots,2018\right\}\\rm{and}\\ \rm{}\rm{y}=\rm{t}+10*\rm{n for n a positive integer}\end{array}$$

(6.18)

and

$$\text{SF}\left(\text{y}\right)=\left\{\begin{array}{ll}1\hfill &\text{if}\text{E}\left[\text{S}\left(\text{y}\right)\right]=1\hfill \\ 1.4\hfill &\text{if}\text{E}\left[\text{S}\left(\text{y}\right)\right]=0.9\hfill \\ 1.84\hfill &\text{if}\text{E}\left[\text{S}\left(\text{y}\right)\right]=0.8\hfill \\ 2.4\hfill &\text{if}\text{E}\left[\text{S}\left(\text{y}\right)\right]=0.7\hfill \end{array}\right.$$

(6.19)

Via (6.15), we set expectations for water prices in 2009–18 according to the simulated values [PS(y)] obtained in our policy simulation, that is, with the buyback scheme in place. Via (6.16), we set the expected allocation shares in 2009–2018 according to the values adopted in our simulation. Via (6.17), we allow for 3% inflation in the determination of expected water prices for years beyond 2018. We also introduce a scarcity factor [SF(y)] to reflect periodic droughts. As shown in (6.19), in years in which the expected allocation share is less than one, the scarcity factor magnifies the expected price of water. The magnifications (1.4, 1.84 and 2.4) were calculated via simulations showing the effects on prices of reduced allocations. Via (6.18), we assume that the pattern of droughts (and hence allocation shares) in the decades beyond 2018 repeats the pattern assumed for the decade from 2009 to 2018.

Prices in 2009, dollars per ML calculated from (6.14) to (6.19), are shown in Table 6.2. They imply an average price between 2009 and 2016 of $2,081. The average price per ML for sales of permanent water rights in SMDB up until 31 January 2010 was $1,654 on sales of 502 GL (the average annual equivalent of 796 GL of entitlements; see PC (2010, Table 1.2)). A somewhat higher price could be expected if the Commonwealth implemented a buyback scheme totalling 1,500 GL as assumed in this paper. The Howard plan included a provision of $3 billion for buybacks (Australian Government 2007). Given the prices in Table 6.2, the total cost of the 1,500-GL program is $3.1 billion.

Table 6.2 Prices of permanent water rights ($ per ML, 2009 prices)

The expected frequency of future droughts is likely to be higher when the basin remains in drought and then fall after the drought ends. Available data indicate that the asset price of water fell in 2010.¹⁴ That the price only fell more than a year after the drought had broken indicates that market players remained cautious for a time before changing their expectations. The calculated costs to the Australian Government of the scenario analysed in this chapter would have fallen had the assumed frequency of future droughts fallen.