Rice and water
A couple of people have suggested that I should comment on some stories in the Australian about water and the rice industry. Since I’m busy, along with the RSMG team, working on models of this very topic (for all irrigated industries, not just rice), I’m happy to oblige.
The first, by Amanda Hodge and Matthew Denholm, presents a fairly negative view of the rice industry as a profligate user of water, and the second, by Laurie Arthur, is a response from the industry. A lot of interest focuses on the amount of water used to produce a kilogram of rice. In the original article, this was erroneously reported as 21 000 litres: the correct figure is about 2000 litres/kg, or 2Ml/tonne which is still a lot of water.
How useful is this kind of figure? In general, economists like to look at all the inputs to production, but there are some occasions on which it’s useful to look at individual factors like water. The general premise of the first Australian article is that as water becomes scarcer we’ll have to stop growing crops like rice. It’s certainly true that, whereas water costs were once negligible in relation to the total costs of production when water cost a few dollars per Ml, they are now a major part of the whole. If we suppose that rice sells for $300/tonne and water for $50/Ml, water costs are one third of the total. This is still a long way below the price paid by many urban users, which is between $500/Ml and $1000/Ml, net of treatment and reticulation.
However, it’s noticeable that, as water prices have risen, with further increases expected, growers have economised on water use, exactly as economists would expect. Arthur claims water use/unit of output has declined by 60 per cent, and it’s easy to see that the industry could not have remained viable without improvements in efficiency. It seems certain that further improvements will be needed as water prices rise over time.
One thing that’s often done is to compare average output per megalitre between different industries with the presumption that, as water becomes tradeable, it will flow, literally and metaphorically, from low-value to high-value uses. I’ve presented this kind of comparison myself. It’s not strictly accurate, though. Standard microeconomic analysis doesn’t have much to say about average partial productivity measures. The crucial prediction is that the marginal value product of a factor of production, such as water, will be equalised between industries. Unfortunately, marginal product is much harder to measure than average product.
To sum up, I think it’s a mistake to adopt a broad-brush condemnation of an entire activity like rice-growing. On the other hand, I expect that, as the true social value of water is reflected in increasing prices, we’ll see some contraction in water-intensive industries like rice, and also continuous pressure for improvements in the efficiency with which water is used.
fn1. Using the standard, but bogus illustration that 1 Megalitre is enough to fill an Olympic swimming pool, that makes 500kg per swimming pool.