More on benefit-cost analysis: a simple example

Here is a recent example of the classic jobs versus environmental quality tradeoff that environmental economists are very familiar with. It also allows an example of a benefit-cost analysis.

A utility would like to site a coal-fired power plant in a depressed area in Florida:

The plant, which is being proposed by several public utilities, would bring 150 permanent jobs to the area and 1,500 jobs during construction. At an estimated cost of $1.4 billion, county officials are looking at the potential tax revenue from the plant that would be used to fund improvements to roads and schools. Local businesses also would benefit from the influx of money and people.

But the promise of jobs to an employment-starved county has not clouded everyone's vision on the other potential deposit of the coal plant - pollution.

Some longtime residents, both the poor and the well-off, still have strong feelings about preserving Taylor's environment and are not willing to sacrifice health concerns for economic prosperity. Others are skeptical about the promised jobs being filled by Taylor residents.

How should the community decide whether to allow construction? The role for an economist is to provide a benefit cost analysis.

-- Warning: Back-of-the-envelope analysis ahead --

The primary costs of the plant are the "external costs" of pollution (e.g., negative health effects that occur outside a market transaction, negative impacts to agriculture, tourism, etc). The benefits of the plant include the profit earned by the plant (i.e., "producer surplus" to an economist), tax revenue, etc.

Both of these numbers can be monetized using various economic methods. I'm sure more on these methods will be posted on env-econ in the future.

An estimate of the benefits of the policy is $3.75 million annually. An estimate of the total cost is about $1.33 million annually. The annual net benefits are about $2.42 million ($2005). Therefore, the environmental economist would conclude that siting the plant is a good idea (i.e., it improves economic efficiency). This information is advisory. Politicians and other decision makers are free to make whatever decision they'd like.

Follow the link below for the gory numerical details.

Back-of-the-envelope benefit cost analysis

Here is a very simple benefit cost analysis (i.e., "back of the envelope"). I'm ignoring construction and numerous other costs (such as agricultural and other pollution costs) and the benefits to consumers from lower energy prices, tax revenues and other benefits.

The profit can be reasonably proxied by the increased income from new jobs. Given the following information:

• The plant would bring about 150 jobs with an average salary of $25,000.
• (Taylor, Fl has about 20,000 residents with per capital personal income < $20,000.)

The increased income (i.e., benefits) is $3.75 million annually.

The health costs are estimated by piecing information together from various sources. From a another siting case study in Michigan (Coal Burning Plant Fires Up Hot Dispute in Manistee):

According to the Union of Concerned Scientists in Washington, D.C., a similar-size, 500 megawatt plant typically pours into the atmosphere 10,000 tons of sulfur dioxide, which causes acid rain, 10,200 tons of nitrogen oxide that leads to smog, 700 tons of carbon monoxide, which aggravates heart disease, ...

(Annually, right?) More semi-educated guesswork, I assume that the quantity of fine particulates generated is about 5% of the tons of SO2 and NOX (readers: please offer better suggestions).

In a study of the external health costs of coal-fired electricity generation, Bill Desvousges, Reed Johnson and Spencer Banzhaf (Amazon.com link) estimate that the costs per ton of coal-fired power plant pollution in a rural area are ($2005):

• $902/ton for fine particulates
• $59/ton for nitrogen oxide
• $28/ton for sulfur dioxide
• $0.39/ton for carbon monoxide

Next, I multiply the costs per ton by the number of tons generated annually. The total cost is about $1.33 million annually.

Subtracting costs from benefits the annual net benefits to the town of Taylor are $3.75 - $1.33 = $2.42 million.

Comments, criticism, corrections and character assasination are very welcome.

Comments

[allen claxton]

I'm not someone who's done a lot of cost-benefit analyses, and am mostly familiar with them from the classroom, but it seems like the thing that's missing to me is a sense of what the alternatives are, and what their net benefit is likely to be. Could the same thing be accomplished using some form of renewable energy? (RE isn't quite cost competitive in most places yet, but surely a large part of that is because of externalities, which you're including here.) Or maybe they could push electricity generation out entirely, onto someone else, and make public investments into something else with greater benefits.

There's also no sense of who it is that benefits (who gets these jobs? who will be using the electricity?) versus who will be bearing the costs (wouldn't particulates be more localized than SOx and NOx?).

It would also seem useful to make mention of what future regulatory environments are going to be like. Particularly with regard to greenhouse gas emissions. If anything comes down the line in the next twenty years, this coal-fired power plant, or its owner, is likely to be hit with a massive financial burden.

Obviously, for a quick calculation, you can't go into all this. But as an example, this seems like stuff you should mention, as you mentioned (briefly) construction costs, agricultural impacts, tax revenue impacts, and lower energy costs. I think that presenting cost-benefit analysis as something that arrives at a number, at which point the analyst says "Yea!" or "Nay!" does a real disservice to a tool that can provide a good deal of information on complex decisions.

[allen claxton]

That came off a little stronger than I intended it. I know none of this is new to you, I just mean to say that, presentation-wise, that seems like useful stuff to include as part of the C-B analyst's job. It's part of what makes it a great tool, that you really can grapple with large parts of the complexity of these issues, in a way that you lay everything out and expose how you arrive at your answers.

[the oil drum (ianqui)]

Disclaimer: I have no background in economics.

So when an economist does a benefit-cost analysis, it's possible to figure out, say, how much carbon monoxide will go into the environment, and how much that costs. I guess I don't understand what that cost refers to--how much it would be to clean up that pollution?

But when people get in an uproar about these sorts of things, there's so much more that goes into it. Let's say some of these toxins get into the water supply and someone gets cancer. The cancer treatment costs some amount of money, and we could even calculate that person's lost wages. But the person had cancer, and that's not something that can be assigned a cold, unemotional value. Of course if I were living in that Florida town I'd oppose the power plant, because even if my cancer could be cured at some cost to my insurance company, I wouldn't want to get it in the first place.

[John Whitehead]

Yes, it is possible to get a good idea of the cost of the pollution to individual households. Economists variously call this cost: consumer surplus, willingness to pay. What would the household be willing to pay to avoid the pollution? One approach to measuring the cost is the impact on housing prices. If people move away from areas with low air quality then house prices will fall. The differential is one measure of the cost of pollution.

In fact, policy analysts do ascribe cold, unemotional values to cancer cases. It is not something that economists enjoy doing. But it is a necessary evil in order to balance the benefits and costs of a policy. For example, analyses of actual policy decisions involving cancer cases have found that the implicit value of a "statistical life" (the willingness to pay to avoid cancer/death) ranges from the $100,000s to $100,000,000s. Policy decisions should be more rational than that.

We'll discuss this issue and other environmental valuation methods on ENV-ECON very soon.

Thanks for your comment.

[codeburner]

I actually do work at a Power Plant in Florida and the estimate of the average salary for Power Plant workers ($25000/yr) is low by at least 100%. No plant attempting to get by at that level would be able to hire or hold onto trained employees.

At this plant, not too far from Taylor County, I'd estimate the average pay as being closer to $55-60,000. And the utility would have to hire at least a core cadre of trained workers at industry rates to get the plant up and running.

It is unlikely that these workers already reside in that very rural county so the workers would have to be brought into the county.

[John Whitehead]

Thanks for your comment! I thought the $25k sounded low but that is what the paper said. With the additional salary, the net benefits of the power plant are even higher.

Of course, I've omitted many factors ...

[Chris Bruce]

I have two concerns with the calculations that John Whitehead made in his July 15 posting on the costs and benefits of a power plant to the town of Taylor, Florida.

First, it is only appropriate to conduct a cost-benefit analysis for a single town if the project has no effects beyond the town's borders. Otherwise, it would make sense to argue that all planes flying from LA to NY should stop for five hours at Topeka. From the point of view of Topekans, the benefits would dramatically exceed the costs; but that wouldn't make the project "socially" desirable. Similarly, in John's power plant example, he ignores both the benefits of the additional power to Floridians outside of Taylor and the costs to Florida of losing valuable labor to Taylor.

Second, the wages paid to workers in any project are rarely a "benefit" - they are a cost. Why are power plant employees being paid $50,000 or $60,000? Its because that's what they are worth somewhere else in Florida (or elsewhere in the country). So moving them to Taylor "costs" some other part of Florida. (Wages are only a benefit if the workers involved were previously unemployable.) The benefit of the plant is the value of the extra power, MINUS the wages paid to the workers (and minus the costs of construction, operation, etc.) My bet is that if John had calculated the latter value, it would have been significantly less than the $1 billion he calculated as the cost of pollution.

[John Whitehead]

Chris is absolutely right, of course. In my defense, remember that I began the exercise with "Here is a very simple benefit cost analysis (i.e., "back of the envelope"). I'm ignoring [a bunch of things]."

I stretched it a bit by saying that "profit can be reasonably proxied by the increased income from new jobs", but, from the perspective of the town that is making the decision, the income becomes spending which is a good thing for the community (I know, I know; impacts aren't benefits but I wanted to make a little comparison), and other benefits don't matter (a jurisdictional BCA?).

The purpose of the example was to show that BCAs DO include environmental benefits -- a point that I felt was being overlooked in the BCA discussion. I should have started from the beginning and defined environmental benefits. Come to think of it, I'll do just that. Very soon.

[Joy Towles Ezell]

I live in Taylor County, Florida and very much oppose this proposed coal fired power plant. For your information, since the four power entities are actually municipalities:
Jacksonville Electric Authority - JEA (what do we get from JEA?: Just Emplysema and Asthma), Reedy Creek/Disney World - Disney gets the power, our kids get asthma, The Florida Municipal Power Agency, and the City of Tallahassee.......they would pay no taxes. The Taylor County Commission is "negotiating" with the power consortium re the consortium's $170 M offer, spread over the first 30 years of operation...that comes to $230 per person in our rural, already-chlorine-bleahcing-pulp mill county. The county gets nothing the second 30 years, when the hospital and health department will need even more funding than the first 30 years. JEA has already admitted that the cfpp will emit at least, probably more than, 300 lbs of mercury per year. JEA won't say how much carbon dioxide. They've recently changed the name of the proposed cfpp from the North Florida Pollution Project to the Taylor Emphysema Center.
They would not commit to giving any percentage of the jobs to Taylor County residents, the majority of whom are educationally-challenged, and not experienced in coal fired power plant jobs.
Our county commission is afraid to let us vote on this boondoggle, as they know that at least 75% of the voters here would vote against such a polluter in our county.
Oh, and by the way, Taylor County residents get no power from this proposed power plant, only the pollution.