Economist polls.

[agamemnus]

I had a number of (in retrospect) dumb ideas today, but here's one that I could not prove as dumb as of yet:

Why isn't there a (US) organization doing a regular survey/polls of people with advanced degrees (MA or above) in economics of top accredited institutions? Surely this would be far more democratic than relying on some arbitrary poll of a few dozen handful of so-called economists to deliver economic expectations to the nation that the stock market seems to LIVE on, and more reliable than polling the rabble that the media and B.L.S. lives on.

The data would be public and free, and IMO help inform policy much more than these private polls and BLS data. The assured boost in the efficiency to the economy could be re-invested in a yearly stipend for the poll-takers, with a small core staff on employ to make sure poll-takers are taking some thought into their replies. (as an accuracy mechanism, the core staff can randomly audit replies -- no need to go through them all)

[RandBlade]

There are various types of surveys out there already.

Polls of business leaders are probably more relevant normally.

[agamemnus]

Just added a bit. I think there should be a stipend.

Yeah, I know there are a number of "business leader" polls, but that is different from polls of people who both have a natural inclination and have been encouraged (via their courses) to use critical thinking in the analysis of economic concepts. The pool is much, much, larger, as well.

[GGT]

First, we'd have to find "independent economists". No such animal exists.

[wiggin]

http://www.economist.com/node/21564175

[agamemnus]

I don't think "independent economists" are necessary, but I am very sure that a lot of economists -- especially those who got their degrees in the last 5 years -- are not working for some sort of think-tank. Probably a lot of them are unemployed, working in different fields, or serving up coffee.

[agamemnus]

http://www.economist.com/node/21564175

Nice article, despite the above objections:

But whom would the experts pick? To find out, The Economist polled hundreds of professional academic and business economists.

Cherry-picking. Agendas. Etc.

The academics gave Mr Obama much higher marks than Mr Romney, which may in part reflect partisan preference: fully 45% of them identified themselves as Democrats, and just 7% as Republicans.

Such public employees/university professors (esp. pure "economics" profs in public unis and permanent tenure) are much more likely to be liberal. How do they get there in the first place? It all starts with the rabidly liberal unions.

[GGT]

Aggie, how do you define an "Economist" in the first place?

[wiggin]

Ag, they went to two large organizations and asked all of the members. It wasn't so much cherry picking as selection error.

[GGT]

Also, Economics is part of Social Science, just like Political Science, with a huge dose of Human Behavior and Philosophy. ie, it's not really a Science in the true sense of the word.

[agamemnus]

Ag, they went to two large organizations and asked all of the members. It wasn't so much cherry picking as selection error.

Easiest way to get the data, I guess... at least it wasn't all government employees...

Aggie, how do you define an "Economist" in the first place?

"people with advanced degrees (MA or above) in economics of top accredited institutions"

Also, Economics is part of Social Science, just like Political Science, with a huge dose of Human Behavior and Philosophy. ie, it's not really a Science in the true sense of the word.

The main part, I think, is that they are better critical thinkers than the majority. They wouldn't think that relocating a deer crossing sign to less-traveled roads would result in fewer accidents, for example.

[Loki]

What exactly do you want to survey them about? If it's about the state of the economy - which has a tiny bit of merit - then you'd be asking lots of people who'd be speaking outside of their area of expertise. If it's about the merit of some policy, then you're assuming that the validity of a policy can be determined predominantly by its economic aspects.

A minor nitpick: I'd be very hesitant to call someone with only a masters in economics to be an economist, unless they have extensive experience. I should also note that some of the top universities use masters programs in economics to make money, and the abilities of their graduates don't necessarily correspond to the prestige of the universities.

[agamemnus]

On policy merits: how can we rely on these small "real economist" polls everyone in the media always talks about, if they are inexorably prone to bias? You have your economists on corporate payrolls, then you have your economists in liberal madhouses. They will say things that are not necessarily their own opinions on the subject; they will be biased because of the work that they do or their employers' various powers over them.

On the state of the economy: everyone can read numbers, and everyone has an opinion, but I don't think it's a matter of "expertise". I just happen to think that people who have taken the time to go through and understand even basic classes where two people trade two goods for their mutual benefit even though one of them can make both goods faster (ie: absolute vs. comparative advantage) will give a more thoughtful and insightful answer than those who have not. It's an untapped market of ideas, if you will.

[Loki]

You just have to hope the bias is averaged out.

Everyone already knows what economists believe about trade. Doesn't keep them ignoring the economists. Can you give some examples of the questions you'd want asked?

[agamemnus]

Do they? Trade policy has several dimensions, I find, and there are many viewpoints, even amongst economists. But, I think, perhaps, just perhaps, with a big group of say ~50,000 economists* (perhaps monthly surveys and polls on various topics), we can help the public understand trade policy (and other things) beyond slogans and fear.


* Or at least, people with MA economic degrees; possibly BAs as a lower, less-human sub-class. There must be at least several hundred thousand in the US alone.

EDIT:
"Can you give some examples of the questions you'd want asked?"
1) I think I'd start with their opinion on tax policy in terms of repatriation to parent companies / and/or wealth inflow or outflow taxes (from/to the "borders"). There are not a lot of things than Romney and Obama have a clear/precise difference on -- they ambiguate their differences on purpose. That's one that we can readily discuss the merits of without them being able to back out.


2) Another one that I think would interest many people is the merits of unions versus their cost. Republicans and especially Democrats (and especially here in Massachusetts) have been working that one for ages. I think people need to periodically think about and re-asses their beliefs on these things; that's harder when general discourse is separated into either wild confrontations on TV or (on the other end) tea-time with Mary Poppins.

I would ask: in what situations do economists agree that unions are a good thing? Perhaps we could also construct some multiple-choice or scale questions that seek to find out "why" they think what they do.

[Loki]

The few polls I've seen show that well over 95% of economists support free trade. They do not support a vast majority of the restrictions that are part of the political discourse in the US. No one gives a damn.

[RandBlade]

Do you have any idea how much it'd cost to poll 50,000 in one field scientifically? And how redundant* it'd be. Who'd pay for it and why?

* There's a reason it's rare to find polls significantly over 1000 people. Above that level random sampling errors become exponentially harder to remove and residual errors are more due to methodological rather than random errors.

[Loki]

Assuming the "population" is 50k, I doubt you'd need a sample of more than a few hundred to get a fairly accurate portrayal of their views. And it's not like having a moderate margin of error would matter, since no one would care if 48% or 51% of economists believe x.

[agamemnus]

Loki: I edited my reply above, with some sample questions I would ask.

Assuming the "population" is 50k, I doubt you'd need a sample of more than a few hundred to get a fairly accurate portrayal of their views. And it's not like having a moderate margin of error would matter, since no one would care if 48% or 51% of economists believe x.

Do you have any idea how much it'd cost to poll 50,000 in one field scientifically? And how redundant* it'd be. Who'd pay for it and why?

The US government, or Bill Gates, for the lulz. As I said, I think the benefit of informing both policy and the public would be great. There is no significant politician who will say that 50,000 educated people all are out to get him and no politician can brush off this poll. Right?!?

Sure, statistically speaking, you could do it with 2,000. It will give a significant strength as well as geographical and occupational granularity to the poll, though -- remember: these people don't all work as economists; most of them wouldn't.

Edit: re: random sampling errors-- I think the poll would not be quite like your regular statistical sample poll, but more of an "ask everyone and their dog with an MA in economics" poll. Kind of like voting...

[Loki]

As long as the sample is randomly selected, it will already reflect geographical and occupational differences. Polling 2k people provides us with a pretty accurate view of the beliefs of 300 million Americans. Why would you think the same wouldn't apply to a far smaller number of economists.

[GGT]

...
"people with advanced degrees (MA or above) in economics of top accredited institutions"

The main part, I think, is that they are better critical thinkers than the majority. They wouldn't think that relocating a deer crossing sign to less-traveled roads would result in fewer accidents, for example.

Are these 'critical thinkers' and professional geniuses the same group that couldn't advise our legislators in advance of three decades of deficit spending?

On policy merits: how can we rely on these small "real economist" polls everyone in the media always talks about, if they are inexorably prone to bias? You have your economists on corporate payrolls, then you have your economists in liberal madhouses. They will say things that are not necessarily their own opinions on the subject; they will be biased because of the work that they do or their employers' various powers over them.

Why would you think more (or different) polls or surveys would be any better?

All Economists have a bias, whether it's from their academic education or professional employment. Hell, I'm a Home Economist...and I know that running my domestic budget isn't the same as running a political budget, and that operating a household isn't the same as running a national economy. But money in and money out is a fairly simple concept. Borrowing against future growth is also pretty simple. But it comes with certain attitudes (and trust) regarding fundamentals. The fundamentals that ten "economists" have twenty prescriptions for.

On the state of the economy: everyone can read numbers, and everyone has an opinion, but I don't think it's a matter of "expertise". I just happen to think that people who have taken the time to go through and understand even basic classes where two people trade two goods for their mutual benefit even though one of them can make both goods faster (ie: absolute vs. comparative advantage) will give a more thoughtful and insightful answer than those who have not. It's an untapped market of ideas, if you will.

In that case, you don't want profession economists weighing in, because they're definitely not the "experts".

[Loki]

Yeah, because our legislators were listening to the economists.

[GGT]

Legislators haven't proven themselves to be very "expert", either.

[agamemnus]

As long as the sample is randomly selected, it will already reflect geographical and occupational differences. Polling 2k people provides us with a pretty accurate view of the beliefs of 300 million Americans. Why would you think the same wouldn't apply to a far smaller number of economists.

But again, if you think about it in terms of geographic granularity (divide 2K by 50, that's 40 per state), that's not sufficient. I guess if you were just going for a general national feeling, it would be. However, I don't think it will carry enough weight with the public. Further, 50K would get the respondents themselves talking, and a bigger spotlight on the national stage.

...excuse me, I'm late for not watching the debate.

[Loki]

Why would anyone care about the differences between the economists in Iowa and Missouri?

[GGT]

Because those economists are hired by Big Businesses, and can impact consumer prices from bacon to pharmaceuticals beer?

[Loki]

Weren't you just saying how we should trust government economists by the way?

[GGT]

Weren't you just saying how we should trust government economists by the way?

Nope. But I am saying academic "Economists", in general, are a disparate group of professionals in the first place. There's the London school vs the Chicago school, various Liberal Arts Colleges, etc.

But it's an intriguing proposal, aggie. If you propose "polls and surveys" does that mean you're a Demographic Economist, a Behavioral Economist, or an Actuarial?

[Loki]

A vast majority of economists are quants who don't care much for any particular economic philosophy. Theory is rapidly disappearing from social sciences.

[GGT]

A vast majority of economists are quants who don't care much for any particular economic philosophy. Theory is rapidly disappearing from social sciences.

Maybe those quants are influencing economic "science" the same way they influenced financial market "science"....and are selling theory as facts, to the highest bidder? Sophisticated snake oil strategy, exploiting human irrational expectations, ie greed?