Arnold Kling  


Employer Tyranny vs. Guarantee... Barro on Marginal Utility and ...

My latest essay is on the economics of Pepco, the apparently under-performing electric utility.

I am concerned by two factors that insulate Pepco from facing market discipline concerning reliability. The first is that Pepco is a regulated monopoly. The second is that there is no price indicating the benefits of reliability.

Right now, the only way that consumers can register a preference for reliability of electricity is by complaining. In the essay, I describe a crude mechanism that might be used to put a price on reliability. A major point of the essay is to illustrate the importance of a price system in order to arrive at the proper trade-off between the cost of investment in reliability and cost of outages.

In your comments, please think in terms of realistic alternatives that recognize trade-offs. If you claim to be able to do away with trade-offs altogether, then spell out how your nirvana scheme would work.

Comments and Sharing

CATEGORIES: Microeconomics

COMMENTS (14 to date)
BZ writes:

I will start off with a stupid observation: doesn't PEPCO lose money when people don't get power, since power consumption is typically metered?

DWAnderson writes:

Sure BZ, but my guess is that the loss to Pepco doesn't begin to match the loss to consumers of being without power. Thus Pepco is not facing the right incentives to invest in a more robust system. A more robust system might well be worth higher prices to consumers (because fewer would need generators, for instance), but it is hard to tell absent reliable price signals.

Curt writes:

A few comments here...

1. The article indicates that Pepco had poor restoration performance ("slowest rate of power restoration of all the area's electricity suppliers") - is this compared to other monopoly distribution companies in the service area? If so, is it worth investigating what makes some of these providers better at restoration than others?

2. I believe there is an angle here sort of similar to Arnold's point about the short-term/long-term tradeoffs involved with banking. Reliability and restoration can be increased by spending more today and potentially over-engineering the system with more redundancy and fail-over. But most people aren't terribly excited about paying more today for potential reliability benefits in the future.

3. Given the unpredictable nature of extreme weather events, the reliability insurance concept may be hard to model very accurately.

Bryan Willman writes:

Offer a kind of service-level-agreement contract that would go as below. (This is a mechanism for price discovery.)

A customer can pay an SLA fee to force reimbursement for loss of service and indeed collateral damage. For example, customers might pay an extra 1% of their normal bills in normal months. In exchange for that, when the power is out, they (a) get paid a power-is-out free, (b) get paid collateral damages for things like gensets and ruined food and lost business.

Higher SLA fees lead to higher SLA reimbursements. So, for example, a fee of 50% might say that if the power is out to my house for more than 24 hours, I get all of the money I paid in power bills in the last year refunded.

With a catch. EVERYBODY gets SLA reimbursement from the monopoly equal to the coverage of, say, the 20% bid for SLA.

Obviously a lot of details to be worked out...

Seth writes:

One way to estimate the price of reliability is to compare the percent of Pepco's customers who have made the investment in backup generators and auto transfers switches to that of more reliable utility companies.

Johnson85 writes:

A couple of points. First, there's two ways to spend money on reliability. One is to beef up generation, transmission, and distribution infrastructure. This can take the form of increased redundancy, or more diligent maintenance of equipment and transmission and distribution lines, including periodically checking and replacing poles and ensuring that trees are cut back so they cannot fall onto distribution lines. This spending takes place before an outage.

The second way is just to spend more money after the outage occurs (i.e., more crews and equipment and more overtime and rental fees to get people back on line quickly).

I point this out because for either type of spending, people are pretty much going to get the same benefits of such spending as their neighbors, so it's hard to allow the customers to bid on increased reliability. Whatever level of reliability you decide on, generally everybody gets the same benefit, although you could beef up distribution by neighborhoods.

It's very difficult to introduce significant market mechanisms at the retail level of electric markets. There are regulatory bodies with performance based metrics that help (but can also be screwed up), but I'm not sure any jurisdiction has tried to set up an "artificial" market and done it well enough that the end result was as good as the traditional regulation employed prior to reform.

Johnson85 writes:

One more comment. You may be interested in looking at jurisdictions that use demand side management techniques, i.e. incentives that are aimed at reducing peak usage of electricity, allowing utilities to reduce spending on capacity.

A particular relevant one is interruptable service. With interruptable service, the utility has the right to shut off service or a portion of service to the customer when electricity usage peaks. In exchange, they get a lower rate. This doesn't price what customers are willing to pay for reliability in the face of storms and outages (which can last for days, obviously), but it does price what customers are willing to pay to avoid short term interruptions in power at predictable times of day.

Joe Cushing writes:

Think about this: Many people think reliability is so important that they will spend a couple thousand dollars on a natural gas generator that automatically comes on when the power goes out. They sell these generators at every big box home improvement store. They are important enough to store sales to be sometimes featured at the front of the store. So there is a price right there.

It's interesting that you are saying that electricity isn't reliable enough for lack of a market. I tried to explain this to my dad a year or two ago and he just didn't get it. A storm blows through and 10s of thousands of people are without power for a week. He'd keep saying, "it's a lot of work to put the power back on. I think they are doing a good job. They bring in crews from other states when this happens, so they are responding with a sense of urgency"----but it happens every year.

I'm saying to myself (and to my dad) We have been using power for 100 years. You'd think that in that time, they could have buried the cables in the neighborhoods. All new neighborhoods have buried cable. Not only is it better for reliability but you don't have ugly cables and ugly cut in half trees next to them. The fact that they bury cables in new neighborhoods tells me the process is not prohibitively expensive to do.

Collin writes:

Solar Panels!!!

Joking aside...I think utilities tend to the worst mix of both worlds, rent seeking government with profit seeking private company. I would stick strong anti-trust judge in room with a handful of utilities (like Greene with AT&T) and let them has out the details. Oddly enough, I would recommend to Romney to run on creating a free market in the power companies as an issue that government backed monopolies are not good policy.


MingoV writes:

Joe Cushing said: "The fact that they bury cables in new neighborhoods tells me the process is not prohibitively expensive to do."

In new neighborhoods the costs of buried power cables are paid by the developer and/or building contractors, not the public utility companies. A study showed that buried power lines cost almost ten times as much to install as wood pole suspended lines. Maintenance costs for buried lines are almost as high as for suspended lines: the latter need regular limb lopping and ad hoc repairs after vehicles hit poles, but any problem with a buried line is very costly because many feet of line must be dug up to find the problem area, and digging costs are high because manual labor is needed.

Gerald writes:

From the substance of your article and the nature of the comments, it seems that you may not be really familiar with the regulatory process as it applies to electric utilities. A few points for your consideration:

1. The price that a utility is allowed to charge to each catergory of customers (residential, commercial and industrial) is determined by the Public Service Commision based on the investment necessary to adequately serve those customers. The operative mechanism is the allowed return on equity investment for the utility. Debt incurred and interest payments are merely a pass through at cost if the investment in approved by the regulator. Utilities are thrilled to make more investment as long as it is "allowed" in rate base by the regulators. The principal reason that utilities use above ground distribution wiring is that the regulators in response to residential customer complaints about the cost of service do not "allow" the investment in rate base and therefore it is not made as the price would not provide adequate returns. As a side point, if an investment by a utility is disallowed by the regulators, the utility and therefore its stock holders are out of luck and have to swallow the loss.

2. Some regulators use reliabilty statistics and storm recovery statistics in allowing costs to be included in rates, and provide incentives (higher returns) for above average performance and/or penalties for poor performance. This requires no significant change to current regulatory processes, only the effort by customers/citizens to force regulators (usually elected positions) to make it a priority.

Before proposing massive changes that regulators will not approve (because it threatens their power) take a look at how changes could be made within the current framework to achieve your objective.

Mr. Econotarian writes:

I happen to be in the DC area this week. I will note that the house I am staying in is in a development with buried power lines - however most of the development has not had power since the storm until tonight. I suspect an above ground feeder line or transformer got hit.

The DC suburban area has gone from largely rural/agricultural in the 1950's to largely residential today. 60 years ago, there were no trees anywhere, just open fields. Even in my youth in the 1970's, there were several large working farms around my house. During the 1980's these were developed, and trees were planted. Now large trees have grown up in residential areas. I watched two ex-Christmas trees my parents planted in the 1970's become power-line threatening huge evergreens by the late 1980's.

If we cut down every tree within 100 feet of a power line, there would not be such crazy outages with summer storms. But people are unlikely to do that. The trees are otherwise enjoyable.

Joe Cushing writes:

Did I mention that they had 100 years to come up with the money? Also, we weren't talking about it being as costly, we were talking about how much would we be willing to pay to not have the power go out for 2 weeks when a storm comes through.

Edmond writes:

Multiple tiers of power reliability already exists in the form of generators and backup power systems.

How could a power company effectively offer different tiers of service to neighbors? They use the identical infrastructure. Most issues don't even break down into neighborhoods - the infrastructure is too integrated, and on too large of a scale, for any meaningful separation.

I think the nearest power plant is about 30 miles away and overload cascades can wipe out the power grid for entire regions - even if the power lines in your neighborhood were well maintained in underground conduits, you are still vulnerable.

In any case, the cost of having high reliability to every customer, with hordes of idle maintenance crews waiting to leap into action at a moment's notice, would vastly exceed the cost of getting a backup generator or battery. If you want to pay more for better reliability, they are cheap and easy to find.

While it would be nice to have a free market determine the appropriate trade-off between costs and reliability, I doubt many people would be willing to pay the massive costs to ensure 99.99999% uptime for the region (no matter how much they complain during an outage).

Comments for this entry have been closed
Return to top