Cost-Benefit Analysis for Scientific Research

[freyar]

Today's physics pre-print arXiv listing includes an interesting economic analysis of scientific research, specifically the Large Hadron Collider accelerator experiment at CERN. See the paper here. I am by no means an economist, but the basic idea is as follows: it's generally considered very hard to put a monetary value on scientific research, since we have no idea what kind of technology our understanding of the Higgs boson (or whatever) might lead to in hundreds of years. But it's not that unusual for economists to work out monetary values for things that don't directly produce goods or services, like parks or a clean environment. The authors take those principles and apply them to the LHC and find that it's 92% likely that the LHC will pay for itself --- even leaving out the unpredictable benefits of the actual research --- and will most likely generate a 3 billion euro excess. Food for thought.

And, in case you're wondering, one of the authors is a physicist, and the other two appear to be economists by their academic affiliations.

 

[Kramodlog]

Fundamental research is very important period. Even if it just shows that one avenue is a dead end. That is useful research because others will know not what to do. I'm tired of the ideology that everything must be economically viable. Especially if we consider that economists more often than not do not take externalities* into the equation. The cost of obesity is rarely factored in when we talk about the revenues of McDonald's.

Studying history or archeology's, just to give an example, are worth doing because we learn about ourselves. CERN lets us learn about the universe. Just learning makes those endeavors worth while.

Of course, the flip side is that science can come with its problems. Take the research that lead to drones. They are amazing feats of science, but killing people remotely and the potential lost of privacy are externalities that haven't been tackled. But maybe that is because of capitalism and not because of how science is conducted. Some people saw a profit to be made with drones and to hell with the other consequences as long as there is profit.

*A side effect or consequence of an industrial or commercial activity that affects other parties without this being reflected in the cost of the goods or services involved, such as the pollination of surrounding crops by bees kept for honey.

 

[tomBitonti]

Interesting article. I'm a big fan of the research, and my belief is that the project is worthwhile, but I'm not sure the article proves it very well. But, it's a very dense article and hard to evaluate.

At a first glance, it seems very very busy with data and charts, and too little "gentle explanations" of what the charts show.

Also, the article states the following, which I can't accept as asserted without some discussion:

"A project is socially valuable if its benefits exceed costs over time, NPV > 0. If Bti and Cti are respectively benefits and costs incurred at various times time ti,"

In part because "socially valuable" and "benefits exceed costs" seems too big of a span of meanings. It seems odd to put side by side "social value" and measures of economic value.

Thx!

TomB

 

[freyar]

Yeah, I absolutely agree with both of you regarding the value of scientific research (if I didn't, I wouldn't be in this line of work, for instance!). This was an unusual paper to come up in my morning reading, and I thought the idea was interesting enough to share...

Fundamental research is very important period. Even if it just shows that one avenue is a dead end. That is useful research because others will know not what to do. I'm tired of the ideology that everything must be economically viable. Especially if we consider that economists more often than not do not take externalities* into the equation. The cost of obesity is rarely factored in when we talk about the revenues of McDonald's.

Studying history or archeology's, just to give an example, are worth doing because we learn about ourselves. CERN lets us learn about the universe. Just learning makes those endeavors worth while.

Like I said, I agree absolutely that the gain of knowledge makes it worthwhile. I've made that argument many times myself. But a lot of people (including politicians who make a lot of decisions about big experiments) think in terms of money. The point of this article is to express the value of pure knowledge in financial terms and possibly more convincing to decision-makers. Too many times, as you say, people think about only the direct financial benefit to some specific corporation or industry (that is a concern in how scientific research funding is being directed in Canada right now, for example). This gives an argument on the pure research side to say, hey, pure knowledge is worth just as much in some quantitative way.

Also, if you look at it in more detail, the article talks about externalities, as well, so they are at least giving lip service to the idea of looking at all outcomes.

Interesting article. I'm a big fan of the research, and my belief is that the project is worthwhile, but I'm not sure the article proves it very well. But, it's a very dense article and hard to evaluate.

At a first glance, it seems very very busy with data and charts, and too little "gentle explanations" of what the charts show.

Yeah, I'm no economist, so I can't evaluate the technical merits very well. And it's certainly meant as a technical paper for at least close-to-experts, which is why they don't have a lot of the explanation, I think. Honestly, it's not the normal thing for the physics arXiv and is probably destined for an econ journal, but I but the authors decided a physics audience would also be interested.