Arnold Kling  

Bryan's Challenge

PRINT
Greg Clark at the Movies... John Papola on Teaching Econom...

He wrote,


Name the most credible measure of idea production that isn't at least moderately positively correlated with population.

I think this is a somewhat imprecise formulation. First of all, "moderately positive" is vague. It could be a very low bar.

But what is the relevant boundary for measuring population? The nation-state? So if 10 million people are spread over a large area that means the same thing as if they are concentrated on an island? Or if 10 million people speak 8 different languages that is the same thing as if they all speak the same language?

If you look at population at a nation-state level, then any simple linear correlation is bound to be dominated by India and China. If you instead choose a more robust measure, such as rank correlation, you may see different results.

The classic example suggesting less than perfect correlation between population and ideas would be Israel vs. the Arab countries in Nobel Prizes in science. I think it is a striking example, even though it might not be enough to offset a "modestly positive correlation" overall.

I suspect that communication is a huge factor in idea generation and propagation. I would rather have a small population of people who travel and communicate broadly than a large population of people who stay in their villages their whole lives and rarely communicate with outsiders.

But I go back to my original thought, which is that I would like to see some more thought put into coming up with something more precise than "moderately positive correlation" and also a more precise definition of what constitutes high and low population.



COMMENTS (16 to date)
Evan writes:
I suspect that communication is a huge factor in idea generation and propagation. I would rather have a small population of people who travel and communicate broadly than a large population of people who stay in their villages their whole lives and rarely communicate with outsiders.
That is an excellent point, having a huge nation full of towns that rarely communicate is essentially the same as having a bunch of tiny nations. No doubt one of the reasons for Israel's superiority in Nobel Prizes is its intellectuals' willingness to communicate with intellectuals from other countries, and share ideas with them.

I think the main problem with Bryan's question is that, unless you live in North Korea, it's impossible to measure intellectual productivity by nation, because national borders are porous to ideas. Ideas and culture generally have the least trade restrictions of any good, I think probably because they're (wrongly) seen as less important than physical goods. You hear lots of people complain about Americans buying Japanese cars, but very few complain about Americans watching Japanese movies.

David Friedman writes:

For one sort of idea creation, I suggest something along the lines of "number of pages written in that society and currently in print in at least one major language other than the one they were written in, divided by the period over which all the works were written, with some positive weighting for how long ago they were written." The latter is intended to reflect the fact that it takes more to be of interest to people after a thousand years than after fifty years.

Three societies that I would expect to have very high per capita scores in that measure are Periclean Athens, Elizabethan England, and saga period Iceland. All relatively small populations. I'm not sure that China and India would outrank them even if we made it absolute rather than per capita.

SB7 writes:

I think the Nobel prize measure in Israel vs Arab countries (or St Lucia vs Caribbean ones, for that matter) is incomplete. This is more similar to the mention in Bryan's post of Iceland as having high production of films. You're picking one small country as an outlier: I don't think Nobel production is uncorrelated with population if we look across all nations.

I like David Friedman's idea, but I think it might be better to weight not by population, but by share of the society's world population at the time the work was produced.

Finally, my guess for an intellectual output which is less correlated with population is cuisine. I have no idea how to quantify it. Perhaps that is a question for Tyler Cowen. The only thing that comes to mind is to count up all the restaurants in a selection of global cities which serve an identifiably national or ethnic cuisine, and divide by the number of people in that national/ethnic group.

stephen writes:

[Comment removed pending confirmation of email address. Email the webmaster@econlib.org to request restoring this comment. A valid email address is required to post comments on EconLog and EconTalk.--Econlib Ed.]

Nathan Smith writes:

See Geoffrey West's research on "superlinear scaling" as reported in Steven Johnson's *Where Good Ideas Come From.*


A city that was ten times larger than its neighbor wasn’t ten times more innovative; it was seventeen times more innovative. A metropolis fifty times bigger than a town was 130 times more innovative. Kleiber’s law proved that as life gets bigger, it slows down. But West’s model demonstrated one crucial way in which human-built cities broke from the patterns of biological life: as cities get bigger, they generate ideas at a faster clip. This is what we call ‘superlinear scaling’: if creativity scaled with size in a straight, linear fashion, you would of course find more patents and inventions in a larger city, but the number of patents and inventions per capita would be stable.

How to measure the size of the network is a very interesting question. My dissertation research seeks to vindicate Adam Smith's claim that "the division of labor is limited by the extent of the market." To test this empirically, we need more sophisticated concepts of network size. The theorem certainly does *not* predict merely that larger countries will grow faster, since "the market" is by no means coextensive with national boundaries. Markets might be larger (oil, computers) or smaller (haircuts); they will rarely be the same size.

Floccina writes:

I expect more wealthy people to produce more ideas per person, higher IQ people to produce more ideas per person but holding those constant I would expect more people equals more ideas. The question is difficult, it is do more people impede wealth enough to at some point make more people bring down total number of new ideas. IMO we are far from that.

Also, I believe looking a countries is limited because in a world with great transportation and free trade for all practical purposes no country can be over populated unless all countries are overpopulated.

Daublin writes:

Immigration.

Hollywood is blowing the world away with its film production. I don't know that I've ever watched a film from Iceland. So what makes Hollywood tick?

I don't think it's its own population. So long as there is enough there to support a city, the bulk population doesn't matter much. More people would just mean more people tending bars and drinking at them.

Immigration matters, though. Hollywood's ideas don't come from people born in Hollywood. They're from people who immigrate there. Thus immigration law makes a huge difference on how effective Hollywood is going to be.

I also suspect the global population doesn't matter much. At the margin, there are plenty of good people in Hollywood who aren't getting work. Doubling the potential pool of people working there should not greatly increase the quality of people actively working in Hollywood.

Finch writes:

> Doubling the potential pool of people working
> there should not greatly increase the quality of
> people actively working in Hollywood.

Doubling the market for their output would probably increase its quality. Niches that were previously below minimum scale would then be large enough to be profitably served. Budgets for broadly appealing stuff could be higher.

This is analogous to what has happened with TV, cars, and pornography, to name some examples.

Tracy W writes:

Evan: In terms of ideas protectionism, I think a large part of the reason you don't hear that is that amongst the English speaking countries, America is so big and so dominant.

As a NZer I've heard protectionist arguments, eg in NZ there's a "voluntary" quota for radio stations to play 10 percent NZ music and public money spent on NZ broadcasting, books, etc (by which I mean specifically NZ, about NZ themes). Of course, there's not the same cultural protectionism, as freedom of speech is one of those values held very highly by most people regardless of political persuasion. But the impulse is there.

Mike writes:

Society is like a giant brain. Its productivity is tied to the number of synapses, not necessarily the number of neurons.

Mark Bahner writes:

It seems to me that the answer is obvious. Maybe I'm missing something, but:

1) South Korea vs North Korea,

2) West Germany vs East Germany,

3) Hong Kong vs PRC, at least in the time of Mao.

Mark Bahner writes:

"Hollywood is blowing the world away with its film production. I don't know that I've ever watched a film from Iceland. So what makes Hollywood tick?"

Are you accounting for the fact that every film that Michael Bay makes is the same film? ;-)

Oh, I guess he does two films. Mostly blood and mostly explosions.

Steve Sailer writes:

Charles Murray created a vast database in 2003 of 4002 eminent individuals in the arts and sciences from 800 BC to 1950. You could use that to check your theory. I don't see much evidence for a positive correlation.

Hugh writes:

the most credible measure of idea production?

There is none. There are some proxies like patents filed or learned articles published but these are weak, weak measures indeed.

Even if we could measure idea production we would still be left with the problem of comparing truly great ideas with the rest to come up with some measure of idea production weighted for brilliance.

Good luck with that.

Alfred Differ writes:

I'm not sure how to detect an idea let alone measure an aggregate. Some are easy enough to spot after the fact by the effects they have. My personal experience says most ideas don't cause ripples big enough to see from a distance, so unless there is something like a pricing system that compresses the information and then transmits it there is no hope. The few that do get seen get noticed BECAUSE an otherwise unrelated transmission mechanism picks them up and spreads them.

That leaves us with proxies that are likely to miss most ideas. How those proxies correlate with population addresses the nature of the proxy more than the nature of idea production I suspect. For example, we could measure journal articles in an academic field weighing each by the references within. Journal page space should correlate with the number of researchers and subscribers (I think), but that describes the nature of the publishing business and not the nature of the ideas in the articles.

Joe Cushing writes:

While you are right to point out that there are other factors that can affect idea production, this fact does nothing to disprove the idea that population is an important factor. I suppose part of your quest towards precision would have to involve holding other factors constant. The two cities comparison mentioned in a comment above illustrates this because to nearby cities are likely to be more similar than India and Luxembourg.

That said, there is a flaw in that example too. It's possible that the larger city just attracts more creative people. They go there because that's where people go to create like people go to Nashville because that's where country music is made.

Comments for this entry have been closed
Return to top