Good Sentences

Anyone who believes that "sensible, centrist technocrats" have any claim to expertise in macroeconomic forecasting should read and ponder the following article by Scott Armstrong, marketing prof at Wharton:

http://marketing.wharton.upenn.edu/documents/research/Warmaudit31%205.pdf

While the paper deals with global warming, the central point is quite general: when it comes to issues on which there is no firm scientific truth established, guesses by so-called "experts" are no better than guesses by non-experts. Obviously, if there is no empirical truth already established, all you have is opinion.

So beware Trojan horses that claim to be centrist experts. Most likely, they have an agenda and are trying to sell you something.

Firstly, only one of Delong's proposals have anything to do with Keynesian economics - namely his proposal for fiscal stimulus. But even that is misleading since you can get multipliers larger than 1 within a neoclassical growth model with perfectly flexible wages and prices, so long as public investment is sufficiently productive.

So, there is actually nothing nessecarily Keynesian about his proposal.

In terms of whether or not to recommend political feasible policy - it depends. It depends on how substantial an improvement on the status quo a political feasible compromise would entail - even if it's not Nirvana. Take, for example, financial reform. It would be nice if government could credibly commit to not bailing out large financial institutions, but it's unlikely to happen. Thus, a resolution authority would seem to be a dramatic improvement even if it's not ideal.

You ask "whether an economist should propose the ideal policy, knowing that it will be rejected, or meet the political situation half-way." Why focus on just these two alternatives? There's the ideal policy, which has (say) a 0.000000001% chance of being adopted (thus you "know" it will be rejected); there's the best policy among those that have at least a 10% chance of being adopted (that's probably about what you mean by "meeting the political situation half-way"); and there's the best policy among those that have at least an x% chance of being adopted, for each x between 0.000000001% and 10%. There's also "meeting the political situation *all the way*," which would be merely specifying what was the policy most likely to be adopted. (Suppose it has a 25% chance of being adopted; then it is *the best policy among those that have at least a 25% chance of being adopted*.)

But you seem to overlook the possibility that the economist not be limited to "proposing" a single policy; he might be able to specify as many of these policies as he liked, assigning each to its proper place in the evaluative hierarchy. True, there are probably too many for him to mention all of them: he'd have to be somewhat selective. But you are supposing, implicitly, that he is able to mention *only one* policy; that is a possible scenario, but often he will have room for a more nuanced exposition. Then he need not make the choice that troubles you.

@Paya Renat - if you are quoting a marketing prof on global warming, it's no wonder you find experts unreliable - you're really, really bad at identifying experts whose expertise is relevant to the subject.

"There is a deeper question that Steve and Tyler raise, which is whether an economist should propose the ideal policy, knowing that it will be rejected, or meet the political situation half-way."

Because of the Overton window, its always better to propose ideal policy that will be rejected. By doing so, you move the mean of public discourse.

David,

Scary thing is that the marketing guy probably has a better handle on statistics than the climate scientists.

Some of the statistical work they have done would have them fail out of any modest stats program.