October 27, 2005
Here's some background on the problems in the California electricity market over the last 10 years and why I believe that Californians should vote no on Proposition 80.
Proposition 80 is based on the premise that the public at large should write into the California Constitution many pages of highly specific details about how electricity production and sale should be regulated. If you're someone who likes to spend weeks researching how to cast your vote on each question, this proposition is your dream. I recommend that you start your reading about California's electricity market with excellent treatments by Stanford Economics Professor Frank Wolak and the Congressional Budget Office. No time for that? Then maybe you'll settle for this quick summary.
In 1996, California residents were paying 35% more for electricity than the U.S. average. In the hopes of changing that, California law AB 1890 forced the major California utilities to divest their fossil-fuel generating capacity, created the California Power Exchange to run wholesale auctions of electricity, and mandated a reduction in retail electricity prices.
By 2000, California had reached a situation in which electrical generating capacity was close to demand on peak usage days. Because of the way in which wholesale prices were determined by the bids from power producers, in periods when there was the least spare capacity, it turned out to be in the financial interests of producers to have as much of their production off-line as possible in order to be able to sell at a nearly unlimited price. One of the key things that made this market manipulation possible was the fact that retail demand was by construction completely unresponsive to daily market conditions.
Proposition 80 would repeal sections of AB 1890, and replace them with an outright ban on users making new arrangements to purchase electricity directly from producers. The proposition would further prohibit time-differentiated rate schedules for small users, unless the users give affirmative written consent.
That second item is particularly troubling to me. The core policy objective is to make sure that there is sufficient capacity for the most extreme conditions without burdening users with unnecessary costs. The obvious way to do that is to try to discourage use for those particular times when the system is most heavily taxed. Thus Proposition 80 would prohibit, as an article of the California Constitution, the one tool that is most promising for dealing with the fundamental problem.
The thinking of proponents of both Proposition 80 and the earlier AB 1890 is evidently that consumers wouldn't respond to price incentives. However, technological advances may shortly make it very easy for consumers to make these adjustments automatically, and it strikes me as quite bizarre that California regulators believe that consumers will respond in significant numbers to public requests for conservation from the California Energy Commission while completely ignoring any signals in the form of the price they actually pay.
Even if I thought that a ban on peak-load pricing was appropriate, I'd still be opposed to the strategy of implementing that ban in the form of an amendment to the California Constitution, which is what Proposition 80 would entail. The Constitution should be reserved for establishing broad principles of the highest law of the state, not for details of the latest fashionable thought among the tinkerers. The reason is that, in a few years when it becomes clear that the tinkerers have missed with Proposition 80 as badly as they missed with AB 1890, it will be that much harder for anybody to change it the next time.
I recommend voting "no" on Proposition 80.
Posted by James Hamilton at October 27, 2005 07:24 AMdigg this | reddit
The proposition would further prohibit time-differentiated rate schedules for small users, unless the users give affirmative written consent.
That doesn't seem that unreasonable. It doesn't say that the utility can't do time-of-day billing on small users; just that the users need to consent to it.
The typical way a user would consent is because the utility is offering a discount for time-of-day billing.
I'm a big fan of time-of-day billing and think that in the long-term we'll all be doing it (unless we finally get "too cheap to meter" fusion or Bill Gross's Sunflowers http://www.wired.com/wired/archive/13.07/solar.html
alleviate the noon-day rush on electricity), so the only way I'd modify the proposition is to put a time limit of say 10 years before the utilities can look back into forcing time-of-day billing, when those appliances are more common.
(I agree with the general concept that a Constitution is supposed to be general, and not this specific.)
Posted by: Dan at October 27, 2005 08:45 AM
The reason is that, in a few years when it becomes clear that the tinkerers have missed with Proposition 80 as badly as they missed with AB 1890, it will be that much harder for anybody to change it the next time.
This is THE most important reason one should vote no.
Posted by: DH at October 27, 2005 11:32 AM
The deal killer for me is Prop 80's requirement that we have 33% "renewable" electricity. This illustrates the biggest downside of classic electric regulation - political interference. This problem was a huge motivator during the reform in 1996 as the big consumers (ELCON) hoped to "bypass" it. Does anyone remember Willie Brown extorting cash from the utilities in the early '90s? I knew the delivery men.
Also, from the inside of the state's biggest utility, the amount of interference in our operations caused by us kowtowing to Sacramento was extremely discouraging (and expensive). Our "charitable" contributions were distorted towards payoffs to special interest groups, affirmative action was a political make-work program paid by the ratepayers, and political correctness overruled technical judgments.
JDH mentions that in 1996, our electric rates were 35% over national average. According to the EIA, in July '05, we're now 50% over national. Northern California is much worst than state average - we just got nicked $1.2 billion in legal fees for the PG&E re-organization from bankrupcy!
The optimism about time-of-day (TOD) rates is largely unjustified in light of actual consumer behavior. For large customers, perhaps some further load leveling is possible but not much. For residential customers, the potential savings to them will just not be worth the hassles and "home management" attention required. Would you rearrange your day to save a nickel from staying up late to do your laundry? A dollar? The average bill is just not that sensitive to time-of-day adjustments, no matter how much cheaper TOD meters become. Plus, TOD meters poise privacy issues.
While I'd welcome a return to classic, well-run, electric regulation, Prop 80 ain't it. The theoritical advantages of electric markets are nowhere to be found and the downsides are profound. A bottom-up analysis of electric technology can show that.
Take it from someone who knows the California electricity business inside out - Vote NO on Prop 80.
Posted by: Joseph Somsel at October 27, 2005 11:35 AM
I'm voting no on this and all of the Propositions. Sorry ARNOLD.
Posted by: pgl at October 27, 2005 03:45 PM
You left out the part about the $27 billion bribe extorted from the citizenry--the only part of "deregulation" that actually happened as promised. Let's get the California National Guard back from Iraq before we turn these jackals loose again.
Posted by: none of the above at October 27, 2005 04:55 PM
Deregulation in California is an experiment conducted by economists that has gone tragically awry. Globalization is another experiment championed by economists that has had tragic consequences in the US – an offshoring of jobs and a huge accumulation of household debt. The victims of these experiments feel that they have no choice but to let the economists continue their experiments in hopes that they will be rescued from their tragic plight. Shakespeare couldn’t have written a more hopeless tragedy.
Posted by: touche at October 27, 2005 06:22 PM
Why should electricity cost more in California? Because everything costs more in California.
Posted by: Bill Ellis at October 28, 2005 06:30 AM
Well reasoned. I totally agree. And commenter DH pegs it.
Posted by: laxpat at October 28, 2005 07:21 AM
Part of what went wrong in the electricity market is that there was a price freeze in most of the state. As we hit a demand/supply crunch and wholesale rates shot through the roof, utilities were forbidden to raise retail prices (except in San Diego for some complicated reason). The result was that people consumed along merrily while utilities lost billions. If high prices could have been passed through, consumers might have revolted (as they did in San Diego) but at least action would have been taken before things got so out of hand. The moral of the story is that if you want to have a market, you need to have it everywhere. Keeping a price freeze in place at one level in the marketplace will make everything else go bad.
The big problem with time of day pricing is that people need time of day electric meters. These would record usage not just on a cumulative basis like meters do now, but rather they use a much more complicated system. I think they are hooked up electronically to the central billing service so that they can update usage information on an hourly basis. There is considerable cost in upgrading the entire state to using these expensive new meters, as well as adding the communications infrastructure necessary for the real-time communications these meters need.
While continually variable pricing makes economic sense, experience in many fields shows that consumers don't like this kind of pricing model. They would rather pay a fixed price that varies only occasionally, than to have to decide every single minute whether to turn the light on or leave it off because the rates just changed. It's like how we abandoned the old Internet access model where people paid by the minute in favor of flat rate pricing. A similar transition is underway in telephone rates.
Switching electricity to variable, minute by minute rates is going in exactly the wrong direction in terms of consumer preferences. This may work for big businesses, where the savings can pay for a staff dedicated just to calculating when to turn the machines on and off. But it doesn't work for consumers.
At best we could have a very simple time of day plan, with just a few time periods and prices higher at certain hours than others. That way consumers can make a one-time decision to run the washer at night, and it's not too difficult. Prices would need to be stable and not change more than a couple of times a year. That is the only model for time of day pricing that makes sense for consumers.
Posted by: Hal at October 28, 2005 10:47 AM
I live in the fog belt and do not need aircon. I paid for a timing meter-about
$300- now I save about $20 a month without changing any of my usage habits.The math shows a three year payback. If you have a choice,what are you waiting for?
Posted by: hw at October 28, 2005 12:02 PM
If you save $20 a month for NOT making a change in usage patterns then the pricing tariffs for TOD are WRONG!
You deserve savings only if you change your behavior. Why are we, the other ratepayers, rewarding you?
Posted by: Joseph Somsel at October 28, 2005 01:29 PM
I pay less because I don't draw more watts during peak usage time. Draw the watts, pay more.
Posted by: hw at October 28, 2005 05:42 PM
Please explain how offshoring jobs causes americans to spend (on average) 101% of their patcheck and rack up huge debts by spending like the government?
Posted by: Huh? at October 28, 2005 06:35 PM
"I pay less because I don't draw more watts during peak usage time. Draw the watts, pay more."
My point was the system is charging you less and the rest of us more and there has been NO change in the system load profile.
Mission NOT accomplished.
Posted by: Joseph Somsel at October 29, 2005 08:50 AM
Change YOUR behavior, move to the fog belt. Help the load profile.
Posted by: hw at October 29, 2005 06:53 PM
The Belgian paper i read on TOD showed only a very small difference in consumer electricity behaviour. And they have had TOD for years. So why would that be different in California?
Posted by: c at October 29, 2005 08:12 PM
Joseph, in HWs case I would argued that he already had a "good" behavior pattern, and that the TOD system simple rewarded that. The overall system load won't change until the majority of people with "bad" patterns change their behavior, but to encourage that don't you need to reward good behavior too?
Posted by: chris at October 30, 2005 05:58 AM
Chris and HW,
As Hal and I have both pointed out, the reality of consumer behavior is that residential TOD metering has little effect on usage and does little to change the system load profile.
HW's experience completely supports that assertion.
So, what REALLY happened? The liberal coastal belt of California voted itself a cut in rates that the Central Valley customers have had to subsidize. Where did HW's $20/month rate cut come from?
Why do people in high-priced real estate areas deserve an electric rate subsidy from the po' folk in Fresno?
Posted by: Joseph Somsel at October 31, 2005 08:09 AM
HW, Joseph, and Hal,
A study by Charles River Associates suggested peak-load pricing would work in California. My vision is not a complicated, variable rate, but something that might supplement the PR campaign on the few days when there really might be a crisis. And I repeat that, even if you're not in favor of trying peak-load pricing, prohibiting it in the form of an amendment to the state's constitution is not the right way to do things.
Posted by: JDH at October 31, 2005 08:20 AM
California legislators and regulators have shown quite dramatically that they do not know how to manage anything. The last thing we need is bigger government, more levels of bureaucracy and incompetence.
Most of the problems that remain in the California electricity market go back to deregulation years ago. It is part of the deregulation provisions that the utilities would reinvest X% each year back into existing physical facilities. The goal was to avoid an underinvetment problem in exisitng plant and equipment as it was anticipated the utilities would cut back on expenditures for existing plant as they focused on new customers in growth areas of the state; which would result in higher earnings and share prices as management could show real earnings growth from these new customers and this would lead to higher managerial compensation (you would be shocked to see how much the CEO of So. Cal Edison earns each year). Unfortunately, most of the utilities have failed to live up to these deregulation reinvestment requirements and we are now stuck with obsolete and worn out facilities that cannot handle to demands being placed upon them.
The solution - the state and FERC need to enforce the provisions of deregulation to the letter of the law. No more long term waivers and other tricks that allow the utilities to fail to live up to their responsibilities. If this done, the state will operate in a whole different light and we may even have electric utilities in which we can have confidence they are looking out for the public good once again.
Posted by: Roger Thomas at October 31, 2005 08:59 AM
my (admittedly imperfect) understanding of the California energy market deregulation is that there was actually very little involvement by economists. I'm sure many of my colleagues would say 'far too little.' hence the sort of incentives and opportunity for market manipulation mentioned in Jim's post.
Posted by: Peter Summers at October 31, 2005 09:04 AM
On time of day metering.
In the UK, we have had time-per-use of local phone calls for decades.
Unsurprisingly, British people use the phone a lot less than my North American relatives. And they use it a *lot* less during the day (when cost per call was as much as 10p/ minute) than at night (as little as 1p/ minute).
Something similar pertains in long distance calling (33p a minute to North America during the day, half that at night).
Now telephone calls are now so cheap that no one pays much attention.
But it *is* an example of how time of day metring works. People *do* adjust their behaviour, if the penalty is large enough.
Another example of a non-price regulatory mechanism. My parents are part of a pilot scheme in Ontario where the utility can shut off their heating/ air conditioning at peak periods (typically 4-6.30pm M-Friday). Their electric water heater has long been on this basis as well. This hasn't impacted them much, but it can massively reduce the cost of producing power, given that at peak power:
- a utility is using its most expensive and dirtiest power (typically coal-fired, or gas rather than nuclear/ hydro)
- the electric distribution network struggles to carry the max load at peak
Toronto has migrated from a winter peak to a summer peak, with the change in weather the last decade or two and the spread of air conditioning.
Both these examples show that economic incentives (in my parents case, lower electricity rates overall) can be made to work, and yield significant economic savings.
Proposition 80, from the sound of it, is nuts. But the whole proposition system in California is 'democracy' run amok. No framers of a democratic system ever intended there to be such an ill-informed and corruptible way of entrenching constitutional change. The rule of the legislature was always intended to be sovereign.
Posted by: John at October 31, 2005 09:23 AM
Because electricity cannot be 'stored' it is always far more vulnerable to market manipulation than, say, a stock market (where you can always refuse to sell if the price is too low, or flood the market if it spikes too high).
At times of peak demand, withdraw capacity and watch prices *soar*.
This was something the advocates of electricity deregulation never understood, because they didn't understand basic physics and were in many cases blinded by ideology.
In the UK we deregulated electricity production but we have always had a very active and powerful regulator, who monitors the pool price like a hawk and frequently issues and enforces corrections on the generators.
Posted by: John at October 31, 2005 09:27 AM
The issue here isn't one of physics -- it's market structure (or "mechanism design" if you're a game theorist). This isn't something I'm expert in and I certainly don't know all the details of the California situation.
My point is that economists have learned quite a lot about how to design things like the wholesale electricity auctions JDH refers to, and design them in such a way that precludes the sort of market manipulation that you and he mention.
My original comment was in response to touche's statement that (paraphrasing now) economists screwed this whole thing up. Ideology could certainly be one way for that to happen, as is ignorance (wilful or otherwise) about economic principles.
I don't know any economists who would disagree that there are right ways and wrong ways to deregulate a particular market. I also don't think it's fair to blame economists for a situation that arose without much input from economists. As I said originally, that last statement reflects my impression of the CA situation, not detailed knowledge. I'm pretty sure, though, that economists were far less involved in the CA energy auctions than, eg, that of the 3rd generation broadband spectrum. I'd argue that there's a correlation here between the level of economic reasoning (ie, input from economists) and the outcomes of these two auctions.
Posted by: Peter Summers at October 31, 2005 01:23 PM
From long experience in the California electric utility business, I'm continually amazed at the fallacies and misunderstandings that economists broadcast about electricity markets (sorry, JDH!)
They just don't seem to be able to come down out off their Ivory Towers and see the wheels turning, the politicians dealing, the businessmen scheming. The "animal spirits" of all involved are nowhere captured in the differential equations I see in the energy econ textbooks.
When I was a kid, I wanted to me an "Intellectual" when I grew up. After a lifetime of watching them from my vantage point in the real world, I agree with Paul Johnson who tagged them as the instigators of the greatest human calamities of the 20th century.
In any case, we've shown numerous reasons to vote against Prop 80. Neither hard-nosed engineers like myself, who would prefer a sound regulatory scheme, well and honestly run, nor the theoritical economists, who dream of "free markets" want it or see merit in it.
Who would vote for it? It will probably go down 60/40 or worse. Has anyone heard anyone say they like it?
Posted by: Joseph Somsel at October 31, 2005 04:18 PM
Your points are fair.
I think it would tax any economist to ex ante, design a system for 'free market' trading of electric power that is proof against manipulation.
Further, unless we wish to put a power price meter on every desktop, I don't see the consumer being short run very responsive to the pool price. (consumer and commercial will not be: if you have to run an air conditioner in your shopping mall during store hours, you will. Industrial is very sensitive to market pricing, but something like an aluminium smelter *cannot* be shut down).
Electricity is just inherently difficult to leave to the 'free market'-- because of its unique physical characteristics (I cannot think of an analagous market that works-- the stock market has the advantage that only a tiny fraction of the total value of companies is traded on any given day, there are vast and deep pools of capital, etc.: even then we still had Long Term Capital Management).
The spectrum auctions were different in that it was a one off sale of a license. It was a lot easier for a bunch of game theorists to get it 'right'. Even there, you could argue the market 'spiked' which was great for Chancellor Brown, but not for the long run consumers of mobile telephony services.
Economists who understand electricity probably understand these things, but you wind up with a solution like the UK has: privatise with massive excess capacity, and a very aggressive regulator.
Even then there have been constant rejiggings of the pool price mechanism, a Monopolies and Mergers Commission investigation of the generators, and now, soaring power prices.
Another unintended side effect was that as soon as they could, the distribution companies bought generation assets (vertical integration) and the generation companies either sold themselves to distribution companies, or merged with international utility companies.
I am usually pretty sceptical of the benefits of deregulating markets where there are locational advantages or increasing returns to scale. Think the UK train network.
Again and again the pattern is you get an early price war, followed by a slow reversion to oligopoly and managed pricing, and a need for an activist regulator.
Posted by: John at November 1, 2005 06:38 AM
You claim this is a Constitutional Ammendment. It isn't. As to the details, I'm still trying to educate myself, but getting that detail wrong sin't exactly giving me confidence in the rest of the analysis here.
Posted by: Larry at November 3, 2005 06:47 PM
You're right, Larry. Proposition 80 states, "This initiative measure is submitted to the people in accordance with the provisions of Section 8 of Article II of the California Constitution." Said section of the Constitution states, "The initiative is the power of the electors to propose statutes and amendments to the Constitution and to adopt or reject them." I was misreading this as an amendment rather than an initiative statute.
Section 8, Article II of the California Constitution further specifies that the Legislature "may amend or repeal an initiative statute by another statute that becomes effective only when approved by the electors unless the initiative statute permits amendment or repeal without their approval." The amendment procedure specified by Section 12 of Proposition 80 states, "The Legislature may amend this act only to achieve its purposes and intent, by legislation achieving at least a two-thirds vote of each house and signature by the Governor."
I believe the primary point I was making, that if approved, these changes would be difficult to undo, is accurate, and a correct reason to vote against this proposition. The language I used of calling it a constitutional amendment was inaccurate.
Posted by: JDH at November 3, 2005 07:19 PM
I'm with you on the evolution deregulated electricity markets. You're looking at the UK's experience - mine is from California. For the Swedish experience, read this energy economist:
As you noted, to start you have to have excess capacity so that deregulation causes a price war so that the backers (usually large electric consumers like ELCON) can see a quick win. Generators either eat their mortgage payments on their fixed investment, demand "stranded asset" payments, or, as was the case in the 90's, jump on technological improvements like CCGTs with great heat rates and low priced gas. Note that generation investments are highly illiquid so a key aspect of a free market - the ability to leave or withdrawl - is next to impossible. Turning off your electric generator during a shortage is a criminal act in my state, even if you're not covering your fuel bill.
As load growth continues, little investment takes place since there are few profit opportunities in an oversupplied market (see domestic oil refineries.) Once the demand exceeds or at least strains supply, profit opportunities open, rents are realized ("price gouging") and investment streams in to over-build the supply and the cycle begins anew.
Electric markets then resemble pork bellies - boom and bust. Meanwhile the public is wondering what the hell happened to the trusted, organized, reliable electric service they once enjoyed. Plus, the new cyclic nature means that investment horizons shorten making long-term investment decisions suicidal. Maintaining a steady capacity reserve is next to impossible - electricity is always "just-in-time" but new generation takes years.
Electric deregulation strives to turn essential infrastructure into boom-or-bust, get-rich-quick bubbles - good for a few but a disaster for the many.
Posted by: Joseph Somsel at November 4, 2005 09:55 AM
Setting the price of power by a free market may be quite sensible
*if* you can construct a system of stable power rates for small consumers.
My rationale is that by and large, large consumers of power *can* flex their consumption, but small consumers are almost entirely price inelastic (and income constraints are very real).
What you need are intervening agents who buy power forward using futures and options contracts, to lock in consumer prices.
You also need a regulator that can act against unusual movements in the pricing pool.
I agree the natural tendency of a market with long time lags and 'lumpy' capital expenditure (as is the case in electricity generation, or copper mining or whatever) is price volatility.
The defeat of this proposition (yesterday) leaves California back in a 'Solution Unsatisfactory' (one of my favourite Robert Heinlein short stories, and as good a title as any short story (other than 'The Roads Must Roll' and 'If this Goes On'). Although actually the plot of this drama is closer to The Roads Must Roll (about an attempt to disrupt the US transport infrastructure) than to Solution Unsatisfactory.
Posted by: John at November 9, 2005 03:04 AM
Prop 80 went down 34/66, even worst than I predicted. (All precincts reported.)
The only San Francisco City and County voted Yes.
Posted by: Joseph Somsel at November 9, 2005 07:51 AM