31 March 2009

Fixed and Variable Costs

I was explaining "cost" to a reporter last week, and we got onto fixed and variable costs (and revenues).

A fixed cost does not change with your behavior. My bimonthly water bill from EBMUD has a "water service charge" of $19.06 and "seismic improvement program surcharge" of $3.52. I pay these charges even if I use NO water.

A variable cost changes with your behavior. EBMUD has a "water flow charge" of $2/unit (one hundred cubic feet -- ccf -- or 748 gallons) and "drought surcharge" of $2/unit for "excessive use" (see Stupid Water Bills).

So EBMUD is collecting money from me for having a connection and for the amount of water I use. That seems to be a no-brainer, but here's the twist: This billing structure probably does NOT match EBMUD's costs.

According EBMUD's Budget Fact Sheet [PDF], 73 percent of its costs (debt service + capital appropriation) are fixed and 27 percent (operating expenses) are variable. 27 percent is WAY too high, of course, since salaries are included in operating expenses. Let's just say that 80 percent of EBMUD's expenses are fixed and 20 percent are variable.

Now get back to EBMUD revenues. If we take my water bills as typical, then 47 percent of EBMUD's revenues are fixed and 53 percent are variable.

Why does this matter? If everyone reduces their water use by 20 percent, EBMUD revenue will drop by 10.6 percent but its costs will only drop by 4 percent. Given that EBMUD -- a public utility -- runs on a break-even basis, such a change in costs and revenues will result in a "revenue shortfall," and EBMUD will have to raise rates to cover the losses.

(Note that Metropolitan Water District of SoCal has revenue that's 80 percent variable and costs that are 80 percent fixed; see p15 in my dissertation. In fact, most water utilities -- both public and investor-owned -- have this "upside down" structure.)

Those of you who have noticed the "use less, pay more" pattern at utilities experiencing water "shortages" will now understand three things:
  1. Utilities hit shortfalls because mostly variable revenues are dropping faster than mostly fixed costs.
  2. The resulting rate increases are unpopular because they appear to punish conservation.
  3. Utilities do NOT LIKE conservation because they will have to increase rates, a process that forces water managers to appear in public and be questioned/insulted by politicians and irate ratepayers.
So, how does all this matter?
  • The "right" way to bill customers for water is to charge a fixed price in proportion to fixed costs and variable price in proportion to variable costs.
  • This is not done, IMO, because the People want water bills to reflect water use -- so "water wasters" cover most of the costs of water service.
  • My own policy prescription of "some water for free, pay for more" is based on variable prices being a BIG part of people's water bill.
We can reconcile revenue-cost matching, wasters pay most, and "some for free" in this way: Water customers pay a fixed price in proportion to fixed costs, but a variable price that is FAR greater than variable costs. The surplus (variable) revenue is then refunded to ALL customers (per capita) so that "water wasters" still pay a BIG share of total costs, water misers get "some for free," and the utility breaks even.

And that's today's lesson :)

Bottom Line: Water bills do not reflect the cost of delivering water. If people vary their behavior, financial imbalances result and things get messy. If we fixed water bills (cover fixed costs and feebate variable costs), we would see the end of "I conserved water, and now I pay more!" headlines.


BobS said...

"The "right" way to bill customers for water is to charge a fixed price in proportion to fixed costs and variable price in proportion to variable costs."

I agree that this is generally "just"; I think though that there are some important nuances.

It makes sense that an individual who has a water hook up should pay whether or not he uses water. Take for example a family's vacation home: they only use a small quantity of water annually, but are willing to pay for the infrastucture (including sinks, toilets, etc) so that it is there when they want it.

This simple scenario is insufficient though when you consider that larger users benefit more from large-scale infrastructure such as dams, reservoirs, aqueducts, etc.: a larger portion of the value of these high-cost features can be assigned to the larger users. This means that the cost per gallon needs to be higher for larger users. This is also generally agreed to and shown in tiered rate structures.

Moving toward conservation, we need to move away from a simple rate based system. The whole story isn't told by number of gallons per hookup. If 10 people live in a single home, their overall "footprint" is generally going to be significantly lower than that of 1 person living in a similarly sized house; however, their water usage may be much greater. Why should they pay at a higher rate? Per capita, the 10-person household has low infrastructure demand than the 1-person household.

Conservation needs to be driven by regulatory programs, not rate-incentives. This is already happening. Marin Municipal already has a program wherein homeowners without low flush toilets are assessed fines. Someone using a 3.5 or 5-gal toilet is a bona-fide waster (not a larger user or a mistakenly classified larger user, as in the case of larger households) and should be incentivized to reform.

We have the technology to easily and cost-effectively track use and process data. The challenge in my view is finding ways to demonstrate transparency at utilities, build public confidence in regulation, and implement programs in a way that does not overly burden water users or jeopardize privacy.

David Zetland said...

@Bob -- Looks like you are in favor of my "some for free, pay for more" suggestion wrt the per capita base allocation (http://tinyurl.com/6ah7dl).

As for regulations vs. rates, I disagree. I think that we need BOTH, but -- more importantly -- we need way more emphasis on rates. I'd say that the rate/reg ratio is about 20/80, and I'd be in favor of 75/25...

(Keep watching this space -- I'll post Goermans et al.'s paper soon. In it, they show that regs are good for drought, and rates are better for other times...)

wrt your last point, note that we need to know the number of people in the HH to do per capita block rates. I agree that such a count "violates" privacy, but by far less than some people claim.

Anonymous said...

Great article.