I've been complaining about the inefficiency of water conservation programs for a long time. My main target has been command-and-control measures; I've more or less ignored the city's active conservation program, through which the city offers residents free, low-flow toilets, provides rebates for certain water-conservation measures, conducts irrigation audits and encourages water reclamation. These programs do not produce the obvious dead weight losses that C&C measures do.
But I'm now quite skeptical that Austin's water conservation program is cost-effective. Heretical, I know. But bear with me.
The root of my skepticism is this 28-page, single-spaced memorandum (and 40 pages of appendices) (8.8 MB pdf), which Austin Water Utility prepared at the request of Council members Spelman and Morrison before last Thursday's city council meeting. The memorandum is chock full of interesting technical data. But this chart, on page three, immediately caught my attention. It depicts AWU's own estimate of the impact of its water conservation program:
Column three shows the total annual reduction in water use produced by Austin's conservation program. Column four shows the city's cost (including staff cost) for reducing peak day demand by one gallon. (Important note: These are the usage reductions from the city's low-flow toilet replacement, rebate, irrigation audit and reclamation programs. They do not include reductions from mandatory watering restrictions.)
The first thing that jumped out at me is that the city's conservation program (again, ignoring mandatory watering restrictions) is a horribly expensive way to reduce peak-day demand. The city spent $1.09 per gallon in FY 2008 to reduce peak day demand by one gallon. (I initially thought this was a typo -- surely, AWU meant $1.09/1,000 gallons -- but on further reflection, I think AWU meant what it said. I've e-mailed AWU for confirmation.) If you want a benchmark, AWU's peak charge for residential customers is $8.50/1,000 gallons, or 0.85 cents/gallon. The city is spending 120 times as much to reduce peak-day demand by one gallon. At that price, it would be cheaper to have Ozarka on emergency stand-by with a few hundred thousand water-cooler bottles.
The city is getting a better deal on annual water reduction. In fiscal year 2007, the city spent $5.2 million on conservation programs that reduced total water use by 363 million gallons -- a cost of 1.42 cents/gallon. In FY 2008, it spent $6.3 million to save 766 million gallons -- a cost of 0.8 cents/gallon. That's slightly less than the highest rate it charges residential users (0.85 cents/gallon).
This would be a pretty good deal if the city's marginal cost for treating and delivering one gallon were 0.8 cents/ gallon. This may be counter-intuitive, but it doesn't matter whether the city sells one gallon at cost or spends an equal amount to avoid producing that gallon. If it costs the city 0.8 cents to supply one gallon of water, then it earns zero profit by selling that gallon for 0.8 cents. If the city pays someone 0.8 cents not to buy a gallon, then the city is in exactly the same position -- it's out the 0.8 cents it paid the would-be customer, but it's saved 0.8 cents in production and distribution costs.
The problem is that AWU's marginal cost of supplying one gallon of water is nowhere near 0.8 cents. AWU hasn't published its marginal cost (at least not that I've seen). But AWU does estimate its average variable cost -- a decent proxy for marginal cost -- at $275.20/million gallons. That works out to just 0.03 cents (three one-hundredths of a cent) per gallon. This means the city is spending 26 times its (approximate) marginal cost to save each gallon of water. That's a lot better than its cost of saving a peak-day gallon, but it's still pretty bad. We'd all be better off if the city sold that water and spent the savings on sidewalks or firefighters or the hike-and-bike trail.
The active conservation program looks like a bad deal.
Again, let me emphasize, these conservation estimates do not include the savings from mandatory watering restrictions. AWU estimates the watering restrictions reduce peak-day demand by 5-9 million gallons or approximately 2-4%. Since the watering restrictions cost the city virtually nothing (unless it's hired water police), they are, literally, millions of times more cost-effective than the city's active conservation program.