Discounts is a very misleading word in this context. The core issues relate to the fundamental legal doctrines surrounding water rights, particularly the Prior Appropriation Doctrine. See, e.g., https://en.wikipedia.org/wiki/Prior-appropriation_water_righ... and https://opensiuc.lib.siu.edu/cgi/viewcontent.cgi?referer=&ht...

People with prior appropriation rights to water don't need to pay for the water at all, except perhaps incidental fees and taxes. The right to use the water, contingent on availability, is a property right.

The prior appropriation doctrine threads through all water issues in the Western U.S. States also have appropriation rights as between each other. See, e.g., https://en.wikipedia.org/wiki/Colorado_River_Compact And then the Federal government introduces another dimension of complexity as they can pick and choose flow rates and distribution.

Theoretically, and AFAIU, there's nothing fundamentally anti-market with the prior appropriation doctrine. See https://en.wikipedia.org/wiki/Coase_theorem and also the above paper by Micha Gisser. And there are markets in prior appropriation rights--farmers buy and sell their prior appropriation rights. But apparently these markets aren't very extensive, presumably hindered by a century of complicating legislation and political bickering layered atop the underlying legal doctrines.

The costly legal process is no problem. The gov can get that back by raising the price of water afterward.

The water that farmers use in not treated so there is no treatment cost.

Agricultural water varies in price greatly but it is often like 1000x cheaper than residential water in CA.

Its not the treatment process - I think a substantial portion of actual costs of water is in delivery. This source estimates 2GW needed to move water around in California on peak days: https://energy.lbl.gov/publications/water-supply-related-ele... That's about 5-10% of total electric use in the entire state.