Go, Arnold!

The Other Arnold, The Economator Arnold Kling, gets a nice plug from David Brooks:

Over the past decades, many economists have sought to define the differences between the physical goods economy and the modern protocol economy. In 2000, Larry Summers, then the Treasury secretary, gave a speech called “The New Wealth of Nations,” laying out some principles. Leading work has been done by Douglass North of Washington University, Robert Fogel of the University of Chicago, Joel Mokyr of Northwestern and Paul Romer of Stanford.

Their research is the subject of an important new book called “From Poverty to Prosperity,” by Arnold Kling and Nick Schulz.

Kling and Schulz start off entertainingly by describing a food court. There are protocols everywhere, not only for how to make the food, but how to greet the customers, how to share common equipment like trays and tables, how to settle disputes between the stalls and enforce contracts with the management.

The success of an economy depends on its ability to invent and embrace new protocols. Kling and Schulz use North’s phrase “adaptive efficiency,” but they are really talking about how quickly a society can be infected by new ideas.

Brooks makes an assertion about the economics of promoting and protecting intellectual property that I bitterly dispute, with emphasis added:

For example, if you are making steel, it costs a medium amount to make your first piece of steel and then a significant amount for each additional piece. If, on the other hand, you are making a new drug, it costs an incredible amount to invent your first pill. But then it’s nearly free to copy it millions of times. You’re only going to invest the money to make that first pill if you can have a temporary monopoly to sell the copies. So a nation has to find a way to protect intellectual property while still encouraging the flow of ideas.

Ahhh!  Back in the day when my internet connection was working I would include some links, but - way back when, the British Navy was desperately interested in developing a method of accurately determining longitude (latitude, i.e. North/South, is pretty easy using the stars or the sun; the rotation of the earth makes East/West tricky).  The Royal Society spurred innovation by offering a prize for a workable technique, rather than providing patent or copyright protection and allowing the inventor a monopoly.

Even today examples of science competitions and prizes abound.  A similar model could be applied to drug development - the US could agree to pay some suitable amount (Half a billion?  Yikes!) for an Alzheimer's drug that met certain measures of efficacy.  Specifying performance is a lot trickier, but, as I belabored in this old post, we already employ a similar business model with the defense industry (and we know libs who hate Big Pharma love Big Defense!).

Anyway - I don't know what Kling said, but I am hoping Brooks over-simplified it [Apparently he did].

PILING ON:  TAP offers the same objection, with more links.  Hmm...

Comments

[clarice]

We could wait for Chinese and Japanese and Indian inventors to create things and then steal from them or change one little part of the process and patent/copyright that as the Japanese now do to us.

But--heck--we're all emitting CO2 like crazy and have to get back to mud and wattle huts and eating cold veggies or something so this is just theoretical anyway.

[boris]

You’re only going to invest the money to make that first pill if you can have a temporary monopoly to sell the copies.

Why would that be a problem? Offering a prize for an idea might work on occasion but investing millions of RnD won't fly. The lottery prize idea is only feasible if the lottery ticket is much less than the prize.

Writing songs is a low cost example of IP. A small number of people can make millions doing it and they don't work for prize money. The music industry was famous for exploiting something like the prize idea with the general result more like "ripoff" than effeciency.

[Extraneus]

For example, if you are making steel, it costs a medium amount to make your first piece of steel and then a significant amount for each additional piece.

Uh, how does that make sense? Do we make the first piece in our garages and then build a factory when we want to make more? There's no efficiency of scale in the steel business?

It's a mystery why TM reads Brooks.

[Barry Dauphin]

Maybe prizes are good for some things and patents/copyrights are good for others. There are literary prizes offered, but the writers do not give up all rights if they win the prize.

Determining if a drug is effective requires any number of double blind clinical trials, which take time and money, no matter how inventive the idea is. R & D costs a lot for pharmaceuticals.

Even if the Royal Society sponsored a successful prize-inspired innovation, that kind of thing is a one-off. Would we want the entire pharmaceutical industry to be based on a prize model?

[fdcol63]

Providers of treatment don't want cures.

They want to treat.

Cures put them out of business, deprive them of their livelihoods, and weaken their continuing influence on the patient.

The same is true with Big Government.

[anduril]

Two news items:

Q3 GDP Revised All The Way Down To 2.2%, Much Lower Than Expectations
from Clusterstock by Joe Weisenthal

Revised Q3 GDP came in a 2.2%, which was below the 2.8% analysts had been expecting. It's also well lower than the 3.5% that was initially reported.

And in line with what IRA is predicting about welfare:

Unemployment funds going ‘absolutely broke’:

40 state programs to be emptied by the jobless tsunami within two yearsunami within two years

http://www.msnbc.msn.com/id/34519544/ns/business-washington_post/?ns=business-washington_post

[narciso]

Are we sure he went to the U. Chicago, because
I'm beginning to have doubts, then again I'm not confident about the Harvard educated fact
checker either

[hit and run]

TM:
way back when, the British Navy was desperately interested in developing a method of accurately determining longitude (latitude, i.e., North/South, is pretty easy using the stars or the sun; the rotation of the earth makes East/West tricky).

Nancy http://www.politico.com/news/stories/0708/12122.html>loves longitude

“I have always loved longitude,” Nancy Pelosi says before breaking into laughter. “I love latitude; it’s in the stars. But longitude, it’s about time. ... Time and clocks and all the rest of that have always been a fascination for me.”

And back off because she has a very important role to play in history...

“I’m trying to save the planet; I’m trying to save the planet,” she says impatiently when questioned.

Hey,give her credit, the world hasn't ended in her reign,so she must be doing something right.

[Jane]

Her reign isn't over yet.

[Steve C.]

Dava Sobel's Longitude is the best little book ever written. Well worth the few hours it takes to read it from cover to cover.

1. Mr. Brooks is not familiar with manufacturing or cost accounting. The first item is always the costliest. Amortization of those costs over many future units is how you make a profit.

2. What's interesting about the Longitude Prize competition is that John Harrison, a craftsman who was derided as a "mechanic", chose to pursue the development of a clock while the conventional wisdom was that the "best" solution had to be based on solar/lunar observation. Yes, the best and the brightest were not willing to accept that a yeoman of no pedigree could propose and develop a practical solution that yeoman without pedigree could employ. It takes no special training to read a clock.

3. These intellectuals not only competed for the same prize, they were also the judges who determined if the prize would be awarded.

4. What's also overlooked is that the development of the chronometer gave the RN a substantial military advantage. That is why they needed to "own" the result.

Prizes have their place, but they are not some sort of silver bullet solution. As to be expected, the Founders realized that awarding patents (exclusive monopoly) were fundamental to encouraging progress in the arts and sciences.

[Stuart]

The problem with prizes versus patents is that with a prize, it depends on the imagination of the person or group offering the prize. In other words, they have to have the ability to imagine that something could be done or would be useful. But a lot of invention doesn't work that way. Plenty of things we take for granted now and can't live without, we didn't even know about years ago, much less think we had to have.

Patents promote the innovation you don't know you need. Prizes are good for solving the more obvious problems, but patents would work here, too.

It's not so easy to know which kind of problem is which, right?

[cathyf]

The most interesting part of the longitude story is that they cheated John Harrison out of the prize. Thereby teaching future inventors that they were schmucks to compete and should stick with selling things which you could enforce your rights to get paid.

A little piece of trivia about American government. The Bill of Rights is chock-a-bock full of various and sundry rights, and there are even more in the subsequent amendments. But the main body of the constitution contains only one right -- the right of inventors to a temporary monopoly on the copying of their inventions, and the obligation of government to enforce those rights.

[narciso]

That Nancy P quote, was just her scrambling for a thought wasn't it. Is it a Baltimore thing?

[Just Visiting]

>> Even today examples of science competitions and prizes abound. A similar model could be applied to drug development - the US could agree to pay some suitable amount (Half a billion? Yikes!) for an Alzheimer's drug that met certain measures of efficacy.

In the interest of full disclosure, I'm a patent attorney.

A few comments:
1. While half a billion sounds like a lot of money, it routinely costs over a billion dollars to bring a successful drug to market.
2. Thus, your half-billion dollar "prize" is an enticement to lose a half-billion dollars to bring the drug to market. In the real world there will be precisely zero takers.
3. Further, the profits earned by successful products have to cover the losses incurred in unsuccessful R&D. So not only does the "winner" lose $500 million in bringing the product to market, there are no funds to cover other R&D, much less manufacturing and distribution.
4. Oops. We forgot about profits, didn't we. Damn those pesky investors always demanding a return on their investment.

Your $500 million prize is laughable. To make a prize scheme a viable replacement for the patent system the prize would need to be on the order of 10 billion dollars. And we would need to fund a continuous stream of these prizes to keep R&D incentivized. We're talking hundreds of billions of dollars per year doled out by the government. Yeah, that's a recipe for success.

Our patent system, while imperfect, has done an extremely good job of providing incentives to advance science and engineering, and in a quite economically efficient manner.

[DrJ]

Visiting,

You have it exactly right, and I was going to post much the same thing.

[Ignatz]

--To make a prize scheme a viable replacement for the patent system the prize would need to be on the order of 10 billion dollars. And we would need to fund a continuous stream of these prizes to keep R&D incentivized. We're talking hundreds of billions of dollars per year doled out by the government.--

Hope Harry and Nancy aren't reading that.

[Captain Hate]

Is it a Baltimore thing?

How dare you defame the city of my birth by associating it with that Mafia whore! There are few things that are as pleasing to me as a Ballmer accent, crabcakes made the right way (I've given up on finding any restaurant outside of the Old Line State that makes any that compare to Mother Hate's) and waitresses that call you "hon".

[MjM]

RE: For example, if you are making steel, it costs a medium amount to make your first piece of steel and then a significant amount for each additional piece.

Others have already highlighted the above silly comment.

Just to reiterate: David Brooks is a dunce.

And a question...

If the g'vment pays a prize, who owns the liability?

[narciso]

Sorry about that Capt, I'm from the part of Jersey which inspires Richard Price, and Law
& Order SVU episodes, so I should talk. I thought you were a native Buckeye.

[cathyf]

This is a classic example of the difference between the fascist/communist central planning model and the free market invisible hand model. The patent/copyright system works because it establishes property rights to things of value which are created, and allows people to earn back some, but by no means all, of the value that they create. (Patents run out, and after that you are getting the value of someone's intellectual product for free.)

The main problem with patents is that in some ways they are too short and in other ways they are too long. The problem with prizes is that they require bureaucrats (central planners) to decide what are the important problems, and what are the adequate winners. Even under the best of circumstances you get soviet-style grim poverty. In the kafke-esque circumstances you get the nobel peace prize committee awarding the prizes.

[Jim Glass]

Brooks is talking about marginal cost of production here -- and, contrary to the apparent consensus, he is entirely correct:

"if you are making steel, it costs a medium amount to make your first piece of steel and then a significant amount for each additional piece".

Uh, how does that make sense? Do we make the first piece in our garages and then build a factory when we want to make more?

Well, yes, we do. When society can't make enough in its hearths, it has to build factories.

Then as you increase production in your factory: you have to hire more workers, which increases demand for them and thus their wage rate and your cost; and you have to pay them overtime, which increases your cost; and you increase demand for inputs, which increases their price and your cost; and you increase wear and tear and accelerate depreciation of the plant, which increases your cost.

Then you reach your physical limit, and to produce more you have to pay for another factory.

Thus marginal cost of production increases as production increases for traditional manufactured goods, such as steel and autos, in the standard case.

This is entirely different from software and pharma and the like, where to increase production you can just hit the "copy" button with the marginal cost of additional unlimited production being approximatedly zip nada (often literally with software).

The near zero marginal cost of production leads to declining average cost with increasing production, and thus to "natural monopoly" and "monopolitsic competition" in such industries as software, as the biggest volume producer obtains a big cost advantage over rivals.

It also means a business that invested a fortune to develop the software or pharma, and has to recover that cost from sales, can be quickly wiped out by a competitor that copies it a near zero marginal cost and sells it profitably at a just-above-zero price.

That's not true with manufacturers of traditional goods -- with their high marginal costs, neither can wipe out the other by charging near zero without being willing to take a huge loss in the process.

It costs near zero to copy and spread an idea. To copy and spread a physical product typically costs an increasing marginal amount. So the ratio of intellectual-to-physical content in a product makes a big difference in its economics. As Brooks says.

[jimmyk]

What Jim Glass said, though I think what others were saying is that there may be some element of fixed costs in a lot of manufactured goods, it's a matter of degree.

But these are not new ideas. Economists have been talking about them for decades. It's closely related to the economics of "superstars." The difference between software and steel is a bit like the difference between live theater and movies, or between live music and recorded music.

[Jim Glass]

As others have noted: A prize if great if you know in advance what you want to invent, how much it will cost, and what it is worth -- and you can afford to pony up the prize.

A prize for a Longitudenator, yes. X Prize, great. Alzheimer's remedy, wonderful, let me try to remember that.

But OTOH, e.g.: back in the 1970s all the big computer companies (IBM, DEC, Digital, etc.) surveyed consumers to see if there was any market for a "personal computer". The unanimous response was "No -- I have a secretary who types my letters, and hardly need an electronic calculator to balance my checkbook. What else would it be good for?" Thus all the big computer companies passed.

So back then, absent patents, "Do we establishe all the prizes needed to bring about a personal computer industry and Internet?" "Hell no, and what's an 'Internet'?"

With no prizes and no IP protection for investments in that unknown, highly speculative technology, how long would it have been delayed? Maybe the Soviets would have developed it first?

It'll be difficult to set up and allocate a trillion-dollar prize fund "for all unknown good ideas that prove to deserve to be rewarded, whatever they might be."

Beyond that, with the amount of money we are talking about as necessary for prizes (see comments above) the only party that could pony it up would be the government.

That means, (1) the reward to the innovator comes from taxpayers who mostly don't use or benefit from the innovated product, instead of from its users who do; and much worse (2) the course of innovation is directed by politicians: "UNIVAC, Honeywell and NCR are already rich enough, there's no need to subsidize more technological innovation by them. Let's put up a prize for figuring the best way to finance no-downpayment mortgages for voters with no income!"

Besides, what's so terrible about the patent system the way it works? The term is only 20 years, hardly a lifetime, and after counting the period after the patent is issued but before the patented product gets to market the effective patent term often is a lot of years shorter than that (see "drugs"). Then the product goes into public domain -- and at the very start, when the patent is published, the new knowledge behind it goes public right away, which wouldn't happen without the patent if the innovator protected its investment using the "trade secret" alternative route.

The optimal patent term is an empirical issue that can be argued, and sure, as with all areas of the law, there are plenty of "the law is an ass" stories in patent law.

But if one looks at the rate of technical innovation over the last 30 years and its benefit to society, the bottom line looks pretty healthy. And what's the better alternative as a replacement (not a supplement, as prizes may be)?

Now, as to copyright, if one wants to argue against the 6,000-year term of protection it is up to, thanks to Disney, that'd be another story.

[AL]

“It'll be difficult to set up and allocate a trillion-dollar prize fund "for all unknown good ideas that prove to deserve to be rewarded, whatever they might be."
Like DARPA?

“Besides, what's so terrible about the patent system the way it works?”

Only couple of percent of patents contains claims which are novel, realizable, and usable. The rest is just the right bought by deep pockets to sue out of the market newcomers over patent infringement.

“And what's the better alternative…?”

Grant patents only to real inventions, not just innovative products.

BTW, as you know, IP rights for software are covered by copyright, not patent, and hence lasts virtually forever. The side effect is that almost any software (aside from rarely patented algorithms) could be re-written from scratch and sold without infringement of copyright of original product. See, for example, how internet search engines and browsers are multiplying.

[Just Visiting]

AL,

I'm the patent attorney who posted above. A few comments in response to your post:

1. "Like DARPA"
I though about mentioning DARPA in my previous post. DARPA works for defense contracting where the entire production run is frequently one, or maybe a few hundred and there is only one customer. It is a command model. A centralized board picks the winner and the solution is forced onto the user. Is this really what you want in our consumer economy? A centralized board designating the products which shall be sold in the marketplace? Very Soviet.

2. You are correct that the vast majority of patents never get used. This happens for a variety of reasons. In the end this is fine. They still make a contribution to the public domain of knowledge.

3. The inventive standard for patentability is under much debate right now in the patent community. It is important to keep in mind that science and technology move in incremental steps, baby steps if you will, always building on the work of others. It always has and it always will. Examine any invention that society considered groundbreaking (penicillin, light bulb, airplane, transistor, etc.) and you'll find that it was a minor step building on the work of others in the field that just happened to work. If the patent system is to remain effective, the standard for patentability must reflect this reality. Our society will be much better off with too many patents than too few.

4. Your statement about software is overly broad. There is a case pending at the Supreme Court right now (Bilsky) that may clarify the status of software patents.

[Ignatz]

I may be dumber than I think; probable in fact.
But I have a hard time seeing the difference between steel, pharma and software, unless we're talking about software or other products transeferrable electronically.
Any physical product, whether a steel stamping, a pill or a software dvd is produced in a factory. Pill and software factories are not magically and infinitely scaleable for free and they are subject to the same depreciation and employee factors as a steel one.
The engineering costs of a steel stamping may be quite low but they are not zero. The same is even more true of a pill.
Now the costs of developing the drug in the pill may be monumental, but they are no more a part of producing the pill than the costs of enginieering a new tank or automobile are part of the costs of producing the steel stamping that goes into them.
The marginal costs of essentially every physical product go down the more you produce. Different products have different marginal rates of course but that is a difference of quantity not kind.
It's only in the electronically producable and transferrable realm that new production reaches toward zero marginal rates.

[cathyf]

The marginal costs of essentially every physical product go down the more you produce.

Wrong. When you see those pictures of supply curves crossing demand curves, the demand curves slope down and the supply curves go up. Yes, at the very beginning supply curves slope down, but in that region producing more and more is free money. So, yes, technically the supply and demand curves meet twice -- once when they are both going down, and then again when they are going in opposite directions, but it is trivially obvious that the market-clearing price is the second intersection, where the supply curve is going up.

The "natural monopoly" argument is precisely those situations where the supply curve doesn't turn around, so it is parato optimal to have only one supplier.

Now the costs of developing the drug in the pill may be monumental, but they are no more a part of producing the pill than the costs of enginieering a new tank or automobile are part of the costs of producing the steel stamping that goes into them.

Wrong again (or vacuous.) Without the R&D, there would be no working pill. The value to the customer of a pill is its therapeutic effects -- that it works. That is what the customer is buying, and the price that the customer pays is the value of the therapeutic effects -- so the cost of the R&D is precisely what is being bought and sold with the purchase of the pill.

[Jim Glass]

The marginal costs of essentially every physical product go down the more you produce.

So you are saying that when you increase the production runs at ye olde steel plant, the time-and-a-half overtime you have to pay workers doesn't increase your wage cost? And the increasing demand for inputs such as iron ore, coke, leased railroad cars, etc., doesn't increase their cost? The law of supply and demand doesn't apply to these things?

OTOH, When you hit the "copy" button for software, how much does that increase your labor costs and demand for raw materials?

Pill and software factories are not magically and infinitely scaleable for free and they are subject to the same depreciation and employee factors as a steel one.

Depreciation is part of fixed cost, not marginal cost.

As to employee factors, do you really imagine that increasing a production run for software or pills increases demand for labor as much as increasing production of steel? Do those extra pills require more teams loading and unloading railroad cars, moving all the ore and coal and final product around, stoking the ovens?

BTW, the use of "infinite" in an economic argument is always a tipoff of dubious thinking ("the obvious problem with capitalism is the world cannot sustain infinite growth" ... "a software factory is not infinitely scalable") whether someone else is using it or one is oneself.

There is nothing anything like "infinite" in any human economic activity. All relevant scales are entirely finite.

[AL]

Attorney:

1.By no means have I suggested that DARPA could substitute patents. However, even such rigid “Soviet” thing like DARPA could serve as extremely potent catalizator of technical and scientific progress. The key, IMO, is that decisions are made by top professionals in relevant field, not bean counters in boardroom or judges who in the courtroom.

2. “…vast majority of patents never get used.”

It is not the problem. The problem is that patent protection is granted to huge number of trivial claims. Example: Practically any patent in the field of diesel engines has as first claim basic layout of turbocharged internal combustion engine, known and widely used for almost a century. Every new engine developed by, say, Ford, is protected by portfolio of about thousand patents each containing dozens of claims, covering such trivial things as multycylinder engine cast from aluminium alloy. 99.9% of covered claims are trivial and will not stand in the court of law, but one got to go through years of litigation and millions in legal fees to clear his way from bogus infringement claims.

BTW, you can be sure than from 1000 Japanese patents cross-filed in US at least 900 contain no-nonsence original inventions. From 1000 domestic US patents barely 50 would qualify as such.

3. “Our society will be much better off with too many patents than too few.”

No. Our society will be much better off with too many INVENTIONS than too few. Millions of patents granted to trivial non-inventions only choke the system.

[Ignatz]

--The marginal costs of essentially every physical product go down the more you produce.
Wrong.--

So there are no economies of scale?
500 F-22s have the same unit cost as 100?
I would bid the same price per thousand on a 100k logging job as on a 1,000k one?
I probably used a poor term by making a point concerning amortized costs over a production run by poking my nose into a discussion of marginal costs but that was my point. And the discussion was rather muddled by all concerned as apparently marginal costs include initial costs for some research and plants but not others.

--Without the R&D, there would be no working pill. The value to the customer of a pill is its therapeutic effects -- that it works.--

But the customer buying the car or the tank is not purchasing the engineering or the R&D of it? Those aren't part of the purchaser's cost? Tank and automobile purchasers don't care if their purchases work?

--So you are saying that when you increase the production runs at ye olde steel plant, the time-and-a-half overtime you have to pay workers doesn't increase your wage cost? And the increasing demand for inputs such as iron ore, coke, leased railroad cars, etc., doesn't increase their cost? The law of supply and demand doesn't apply to these things?--

I thought we were discussing the production of single items at a plant. If industry or nationwide some product is so vastly popular that iron ore increases in price because of it or several shifts need to be added then of course costs go up. That is not the discussion that I read above and is not the ordinary course of events.
In the ordinary course of events if ye old stamping supplier gets a contract and sells a few thousand parts he will more likely see his material costs decline as he has more purchasing power and will have planned his labor needs according to his contract run.
If we're discussing 100,000 ye old steel plants in a booming economy then we are discussing macro economics not the costs of running a stamping mill. But that isn't what the original post or David Brooks was discussing.

--Depreciation is part of fixed cost, not marginal cost.--

Then you should not have used it in your description of those things which increase marginal costs of production: --and you increase wear and tear and accelerate depreciation of the plant, which increases your cost.--

--As to employee factors, do you really imagine that increasing a production run for software or pills increases demand for labor as much as increasing production of steel?--

No. Which is precisely why I said they don't. I said it was a difference in degree not kind, which it is.

--BTW, the use of "infinite" in an economic argument is always a tipoff of dubious thinking..-

Yes it is. I was applying the word to your implication that a pill factory is essentially unlimited in its productive capacity such as this statement: -- This is entirely different from software and pharma and the like, where to increase production you can just hit the "copy" button with the marginal cost of additional unlimited production being approximatedly zip nada--

I believe unlimited means without limit, right? As in infinite.

[Joseph Somsel]

As a concrete example for the argument made by Stuart above, let me offer Viagra.

The PDE-5 inhibitor was under investigation as a heart attack treatment but was discovered in clinical trials to give male heart attack victims rather satisfying woodies.

No government prize had been conceived (ha ha!) for an erection drug but some attentive Pfizers employees saw the potential.

There may be a place for government prizes but to rely on them for economic innovation is bad policy.