Does the cream always rise to the top?

Or do we misallocate talent when it comes to innovation?  Here is a not so famous but very interesting paper by Murat Alp Celik:

The misallocation of talent between routine production versus innovation activities has a fi rst-order impact on the welfare and growth prospects of an economy. Surname level empirical analysis employing micro-data on patents and inventors in the U.S. between 1975-2008 combined with census data from 1930 reveals new stylized facts: (i) people with “richer” surnames have a higher probability of becoming an inventor, however (ii) people with more “educated” surnames become more proli fic inventors. Motivated by this discrepancy, a heterogeneous agents model with production and innovation sectors is developed, where individuals can become inventors even if they are of mediocre talent by excessive spending on credentialing. This is individually rational but socially inefficient. The model is calibrated to match the new stylized facts and data moments from the U.S. economy, and is then used to measure the magnitude of the misallocation of talent in innovation. A thought experiment in which the credentialing spending channel is shut down reveals that the aggregate growth rate of the economy can be increased by 10% of its value through a reduction of the misallocation. Socially optimal progressive bequest taxes that alleviate the misallocation are calculated, which serve to increase the growth rate of the economy to 2.05% while increasing social welfare by 6.20% in consumption equivalent terms.

I am not so persuaded by the idea of buying your way into innovative circles with credentials, or the analysis of the inheritance tax, but nonetheless this should stimulate thought.


So, Mark Zuckerberg wasted his time going to Harvard?

'[Facebook] was founded by Mark Zuckerberg with his college roommate and fellow Harvard University student Eduardo Saverin. The website's membership was initially limited by the founders to Harvard students, but was expanded to other colleges in the Boston area, the Ivy League, and gradually most universities in the United States and Canada, corporations, and by September 2006, to everyone with a valid email address along with an age requirement of being 13 and older.'

Of course, if one wishes to argue that the only innovation Facebook represents is getting people to freely participate in their own surveillance for Facebook's profit, that is certainly a valid perspective.

"[Facebook] was founded by Mark Zuckerberg with his college roommate and fellow Harvard University student Eduardo Saverin."

Yet another Brazilian invention perfidous Yankees stole from idealist Brazilians, who just wanted to make the world better. Like the radio, the typewriter, the pion, the caller identifier, the airplane, the discman, etc.

Eduardo Saverin just happened to be in the same room when Mark Zuckerberg began speaking about his idea out loud.

And Santos Dumont just happened to be in the same banyard when the Wright Brothers hicks were speaking about the invention they refuse to demonstrate to their prospective customers aloud...

Saverin came with the whole Facebook idea and was backstabbed by his best "friend" Zuckerberg, and such unprovoked and dastardly betrayal will live in infamy. Have you no sense of decency, sir, at long last? Have you left no sense of decency?

"Vox clamantis in deserto"

So, that's what Brazil has become, a wilderness.

There is nothing innovative about facebook. It's the same as thousands of other social networks. It's only difference is it's popularity. Why it is so popular? I conjecture that it is popular because it started out as a network restricted to Harvard and other Ivy League student's. Since these are the world's coolest people everybody wanted to be in the same social network as they were, who are at the top of the world's pecking order. It them tricked down over the entire population of the western world and absorbed all other major social networks such as Orkut and MySpace.

So he's saying that
(1) there's an eternally limited number of seats in college engineering classes,
(2) 10-20% of those seats are taken up by rich kids rather than distributed meritocratically,
(3) if we had wealth-destroying taxes, there would be no rich people, and then
(4) all the seats in the class would be distributed meritocratically.

Except (5), the weath-destroying taxes would kill all initiative to innovate, and people would stop studying engineering in the first place.

Which we're already seeing with all the people going to into gender-studies and african-american studies and gay-studies rather than engineering. Perhaps instead of raising taxes to higher more engineering professors to teach innovation, we could simply replace the existing sociology professors with engineering professors? Or just put the courses online for free and see if people can learn them without a professor?

Does doing away with the inheritance tax kill all initiative to innovate? Really???

Doing away with inheritance kills all initiative to work hard and innovate, yes.

I anything, I suspect the type "reclusive genius without family or children" is overrepresented in technical/scientifical innovators.

Are childless inventors less innovative than those with kids?

Yeah, I was going to invent a teletransportation device or a perpetual motion machine, but taxes were too high, so I went to study African American gay lesbian transgender sedevacantist women immigrant Jews Muslim Democratic Brazilian Kenyan studies instead. It is boring, but the money is good.

So I guess I'm sorta a sedevacantist regarding his Phoniness Francis I although I came to that from an orthogonal angle. I didn't even know what sedevacantism was.

*PS I would prefer a Time Machine, maybe with a perpetual motion machine to keep in charged up thru all time.

The reason people go into gender studies and so forth is not because taxes have destroyed the economic incentives to innovate, but rather because the student loan system obscures price signals by providing loans on an equal basis regardless of the field of study. You get the same loan, on the same terms, regardless of whether you major in engineering or in gender studies.

+1 to Hazel.

"Which we’re already seeing with all the people going to into gender-studies and african-american studies and gay-studies rather than engineering."

You think these folks would last a week in an engineering program? This is how we got from 20% to 50% participation levels for college. Got to let in a bunch who were formerly not eligible.

It's a good thing there's been no significant shifts in the USA's demographics since 1930 since otherwise the analysis would fall apart. I wonder if it is valid to assume patents correlate to innovation? I am dubious, but admit my ignorance. I do know that getting a patent for a certain type of person is more about signaling than about invention. I didn't know that Al Einstein had 50 patents - were these 'truly' his - or did he leverage his position in the patent office and steal (imitate) applications he saw? Also, I wonder what the total income those patents generated amounted to for the patent assignee. At best, and this paper doesn't seem to be in the "best" category, it is nearly impossible to model social hypotheticals - such as what if education were more "targeted". Who would decide that? Perhaps a star chamber? And here I thought the US did a pretty good job of wealth turn-over and upward mobility. Huh. I guess I need to read up about the countries that do a better job and are the engines of world innovation. Which countries are they? It is probably true that having a patent is an indication of novelty, but in my experience with the US Patent System most of the patents I've seen are either obvious to those skilled in the art, hum-drum extensions of what is already known, or just plain wrong. (Of course, there are some really earth shaking exceptions, but exceptions they are.)

Einstein had the advantage of having first hand experience in a patent office. That the did what he knew and that he was successful where he had inside knowledge is typical human behavior and outcome.

Not that things might not have been different in the early 1900s, but if you have had any interaction with patent examiners you would doubt whether any patent examiner has ever invented anything.

So he's saying that if we taxed everyone's wealth away, then instead of rich people spending their money on their children's education, the state could spend it on someone's else education that they deem to be smarter. (This is assuming that there's a limited amount of educational resources---except the internet was supposed to allow unlimited educational resources to everyone! Why can't everyone earn a ph.d for free online?)

Except that if people couldn't earn wealth and spend it on their children, they wouldn't earn it--they'd all take the easiest jobs they could get, because why take a harder job with more responsibility if everyone takes home the same paycheck everyday?

Is the author saying that we should tax everyone's wealth away? or that they couldn't earn wealth and spend it on their children? I don't think that is what he is suggesting. He is talking about a progressive tax on inherited income, which leaves plenty of room for people to earn wealth and spend it on their children. Just not to the extent that it gives their children an unfair advantage.

wow that last sentence is a kicker.

Define 'unfair'.

Is it unfair that I married someone with a masters in math who is going on to get a PhD in statistics and that I'm an engineer with a masters? Our child has an advantage over the kid of a single mother with a drug problem.
Is it unfair that I put effort into ensuring my kid learns and does her schoolwork? Or maybe it is unfair that I make sure she has a proper diet.

Or what is 'unfair' is that some people have things you want.

"Except that if people couldn’t earn wealth and spend it on their children, they wouldn’t earn it..." Nah, pretty sure people will continue to love buying shit for themselves with their money.

Though I do worry about the certain decline in the Great Inheritance Innovators of tomorrow. If we tax inheritance, we may miss out on the next Wyatt:

"Nah, pretty sure people will continue to love buying shit for themselves with their money."

Well, then. No issues. People obviously don't leave their kids anything so what's all the fuss about inheritance taxes?

So, you're agreeing with our commenter above: "Except that if people couldn’t earn wealth and spend it on their children, they wouldn’t earn it"? I look forward to the boost in wages when you and your buddies quit your jobs because you can't give everything to your kids 100% tax free.

I'm saying that if people were interesting in just consuming, they'd be doing it already. Nothing stopping them.

It's mostly high income people who are in this situation, so most likely talented people. Why would you want them to stop producing?

"Nah, pretty sure people will continue to love buying shit for themselves with their money."

Correct, but of course people will tend to spend more of their money on immediate consumption goods instead of investments. After all they'll have a disincentive to create investments with payoff's longer than their life span.

So, I suppose if you think the US should promote consumption over investment, then you would be for higher inheritance taxes. Because inheritances are essentially long term investments.

If governments can't provide free college, how about they make universal free online middle and high schools, shut down all the brick-and-mortar middle and high schools, and use the money to fully fund colleges the way they currently fully fund middle and high schools?

Well if the kids have to take the online classes from child jail, then it probably won't lower costs any. If they take the classes from home then either parents riot and kill everyone, or social norms completely change to allow for unsupervised middle schoolers at home alone.

Sorry if I confused you, by "child jail" I was referring to normal public schools

I was disappointed that the paper doesn't cite any examples of "richer" surnames or of "more educated" surnames. How big of a difference is there in the two lists? And if there was a big difference in 1930, wasn't it more due to more recent immigrants being longer on education than income?

I know that in Britain among the 25 most common surnames, "Hamilton" is by far the most likely to be a Oxford or Cambridge graduate, with "Freeman" second, and Smith and Jones lagging. I wouldn't be surprised if in the U.S. "Hamilton" weren't also above average in intelligence.

A less common surname with a remarkable number of distinguished individuals is "Huntington," almost all of whom appear to be descendants of one Puritan settler from 385 years ago. It's not surprising that the late Samuel Huntington was just about the last American intellectual to speak up for Anglo-Americans in his 2004 book "Who Are We? The Challenges to America's National Identity" (2004).

Well said by SS, the paper seems bogus. It could well be that the paper is simply back fitting data known to produce winners. For example, "Armstrong" would be a "rich" name, as in Edwin Armstrong, one inventor of TV. "Tesla" is a "rich" name now, only because the original Tesla was a prolific inventor. Another reason might be the fact that rich people have more leisure time and are more likely to invent (for example a Rockefeller woman became a world class scientist in some obscure field, ditto a Rothschild became an expert in wine and so on). Rich people can afford to dabble in science, which frankly does not pay well, and only the super rich and super desperate end up inventing. Against that, what do you make of "Smith", and the numerous US slaves named after their master?

Bonus trivia: the original name "Smith" was given to those craftsmen who kept trade secrets related to metallurgy, aka smiths of the forge. How many secrets did they possess? Lots of annealing in the metal arts is trade secret, like baking a cake, probably even today in very sophisticated steel mills.

Good research on this from Sweden.

' “Hamilton” is by far the most likely to be a Oxford or Cambridge graduate': which is deeply odd given that it is a Scots surname.

Maybe they all go to Balliol College.


Also +1 to Ray’s comment.

The eye rolls that this type of paper incite are dangerous to one’s health.

Is it really all that bad that wealthy Americans tend to be more inventive than the wealthy in, say, Mexico? It seems like America having a culture where rich people are inclined to try to invent things is less a crisis of "misallocation" than a symptom of cultural health.

'the training necessary to become a worker in the innovation sector is scarce. Agents compete against each other in order to acquire this scarce training so that they can get innovation sector jobs that pay more.'

Firstly, no. The training, as in raw knowledge, is widely available. Scarcity does not apply to the training.

Second, consider day one of econ 101 lesson of 'correlation isn't causation'. I didn't find evidence in the paper to reject the obvious rebuttals. The network effects of attending MIT and then working for Lincoln Labs is why some people are positioned to become serial inventors. This is not the education they received, although it does matter, but it's a sorting process that specifically looks for exceptionally smart people, mixes them with other really smart people, has some classes, throws money at them, and patents the results. This works explicitly because it is limited. One cannot just draw a line between training and successful invention, it is also the intentional removal the less talented.

Third, it takes money to bring an product to market. It's easier to get money from one's network than a bank. This is a network effect that both children of wealthy parents and graduates of top tier universities have access to that the average engineering student does not.

Fourth, having more students graduate from engineering schools with a PhD will not turn them into serial inventors. More graduates won't equal more innovation. The process as implemented works, not because the student gets a PhD, but because others don't. That the vast majority of people do not graduate with an advanced degree from a top-5 engineering school is the core reason for the value. This education/signal is why the investment risk is taken.

That money is necessary to bring a product to market is obviously true.
That networks, whether familial or otherwise, give access to funds is obviously true.
That having a doctorate from a well recognized university is helpful in convincing people to give you money is obviously true.

That this results in a misallocation of funds is not defended by the evidence provided in the paper.

As an aside, I think it is a fairly good thing that being an inventor is a 'cool' profession. So much so that the idle rich are interested in pursuing it.

"As an aside, I think it is a fairly good thing that being an inventor is a ‘cool’ profession. So much so that the idle rich are interested in pursuing it."


Here's a page of Celebrity Patent Holders. It would seem to betoken a culture in which even the most privileged sometimes feel inclined to be inventive, which is a good thing:

2. James Cameron
5. Bill Nye
7. George Lucas
9. Francis Ford Coppola
12. Paula Abdul
14. Andy Warhol

You forgot: Hedy Lamarr was an Austrian-born American film actress and inventor. According to the Electronic Frontier Foundation, at the beginning of World War II, Lamarr and composer George Antheil developed a radio guidance system for Allied torpedoes, which used spread spectrum and frequency hopping technology ... used in all mobile phones today, that's how you can get a million to one signal to noise ratio and still get "one bar" on your smart phone, enabling you to at least send a text message to somebody while stuck in an elevator.

But one of the most beautiful and intelligent women in the world wasn't Hedy Lamarr, it was Hedwig Eva Maria Kiesler, her real name. Many of the celebrity patent holders changed their names so perhaps innovation in appellation is an indicator of innovation in other things.

I'm inclined to think that there's a significant number of engineering students who could be innovators, but don't have the social training or access to those networks. There's a lot of people who are "A" students but need some guidance and management. There are people who can work independently on their own initiative, but there are a lot more people who work better when they have effective leadership.

Spengler goes into thisin Man and Technics:

Meh. I also think leadership can be learned, but that too has to be taught. It should be self-evident that most people become more competent and require less leadership over time. And eventually some of those people will be able to teach and guide others in turn. The idea that people are just born innovators or born leaders results in a "sink-or-swim" system which wastes a lot of potentially productive minds.

Vast vast majority of inventors work on teams in corporate labs

An entrenched class whether race, sex, or wealth is less inclined to question a system from which they benefit. Those inquiries are where innovation is born.

Related, employees with more diverse Twitter networks ..

I think that innovation is scarce because the engineering schools don't actually teach innovation. There is nothing in the undergraduate curriculum about how to apply for a patent, or even a grant application. If someone's good enough to get into grad school, they learn that stuff on the job in between writing technical papers and doing their research.

We could do a way better job training kids to be innovators in high school. There's no nearly enough ways in which STEM courses allow kids to be creative, which is the problem. Math and engineering courses pretty much lay out well establish theory and make you go through the process of deriving known formulas. What would be interesting would be to have open-ended lab classes where kids can build devices to do whether they want. Then at the undergraduate level, have a one credit seminar on patents and grant applications.

The amount of education that has to be crammed into a four year degree to get ABET approval is already overloaded.

We can't add more without taking something out. We are pressed to combine the hands-on with the theory, and there is always someone who is pointing out something is missing. Not that they are wrong exactly, but they seem to skip the part where they must also suggest what to remove.

I've seen that most schools manipulate the 4 year degree into 5-6 year for engineers.

I'm on the advisory board for engineering at my university. Since we have a mandate to keep our curriculum at 120 credits (a real 120), I spend a bit of time with this issue. My local state university gets around this by having a four year degree that begins after the student is accepted into the college of engineering. Preparatory to applying, students typically complete 1.5 years at the university. One of the courses is Engineering Chemistry that cannot be substituted by any other class including two semesters of gen chem. This must, along with other courses, be completed at the university prior to acceptance into the college. I don't get this luxury. I have to get them in and out within four years.

To prepare them I have to make sure that they also get internships. The likelihood of success goes down noticeably if there isn't two summers of work, which means I'm trying to get them hired based off of freshman year.

I don't agree with the point that STEM is less creative. This is just a consolation prize that the liberal arts have claimed along with 'critical thought'. I find a tremendous herd instinct among the supposed creative degrees.

You've saved my son some grief (from me). He went into college with enough AP credits to be a sophomore, but looks like he may need 4 years to graduate as an engineer. I was wondering WTH.

I wouldn't say liberal arts is very creative either. I'm talking about having students work on actually creative hands-on projects. You do get creativity in fine arts programs, but the overlap between students taking social sciences and fine arts is pretty small.

Think about the "Maker" subculture - it's mostly engineers who like to work with their hands, but it effectively integrates artists. That's the sort of innovation culture that needs to be incorporated into schools. Nobody in the makerspace is writing a thesis on cultural appropriation, they're integrating art and engineering.

I agree with the idea of having more engineers from low-income fams but not the idea of having fewer from high-income fams

Also more trade schools, which give students/grads contact with the makings of the world. Also more smart immigrants and permission for apartments and subways in LA

Note that to a 1st order approximation everyone in the US of a “high” level of talent gets to go to university for free.

They also get to go to engineering programs. No one of “high” talent is shut out of engineering (or any other major) due to legacy students.

Does this mean we shouldn’t abolish legacy programs? No. There are some reasons to eliminate such programs, but those reasons have nothing to do with stimulating innovation to such a degree that it would affect GDP growth.

Depends where they want to go. No merit scholarships at Ivies

Do they have need based scholarships? Are they very generous?

You do bring up a good point however: children of moderate affluent parents are at a serious financial disadvantage when enrolling at many brand name schools, no matter their merit. It is a tragedy but I don’t see the liberal establishment saying a word about it.

Is this due to the liberal establishment being bankrupt? Yes.

"Surname level empirical analysis ..."

So we are judging books by their covers.

The misallocation of talent between routine production versus innovation activities has a first-order impact on the welfare and growth prospects of an economy.

I don't really understand the argument but the results don't agree with what I see of the patents. In particular, 90+% of patents in the areas of my knowledge are obvious and/or meaningless.

Comments for this post are closed