miscellaneous factZ

Caught an interesting presentation at DRUID by Scott Stern of an interesting paper he’s been working on with Fiona Murray entitled: Does Formal Intellectual Property Impact The Market For Scientific Collaboration? Evicence From Patent-Paper Pairs.

Take-away: 1 in 9 research projects considered is not taken forward because of patent issues (remember this doesn’t mean a reduction in the number of research projects but just that there is substitution)

Full Paper

Abstract

Cumulative scientific research depends on access to inputs associated with prior discoveries, such as databases, research tools and models, and organic materials. While follow-on researchers may be able to limit the price of access to these inputs to providing an appropriate citation, input developers may seek to use their control rights over such inputs to extract additional concessions from follow-on researchers, including co-authorship of followon research articles. By obtaining formal intellectual property rights over nowledge disclosed in scientific publications, input developers may seek to enhance their bargaining power in the market for scientific collaboration. One implication of such rent-seeking is the creation of “patent-paper pairs,” in which a given discovery is instantiated as both a scientific research article and in the form of a patent disclosure. This paper investigates the impact of these IP rights on the market for scientific collaboration. To do so, we exploit the fact that patents are granted with a substantial lag, often many years after the knowledge is initially disclosed through paper publication. Since patent applicants cannot enforce rights over knowledge until their patents are actually granted, authors of patent-paper pairs experience an enhancement in their bargaining power with potential collaborators after a patent is granted. To evaluate the equilibrium implications of these shifts in bargaining power, we examine how the pattern of scientific citations to a research article changes after patent rights over the knowledge in that article are granted. Employing a differences-in-differences estimator for XXX patentpaper pairs (and including a control group of publications from the same journal for which no patent is granted), we find evidence that patent grant results in a sharp reduction in the number of citations from “independent” research teams and an increase in citations associated with “collaborative” research teams. These shifts are particularly important for follow-on articles published by public sector authors, etc. Overall, the evidence suggests that scientists use intellectual property to enhance their market power in the market for scientific collaboration.

A few months ago I came across a fascinating short paper authored by David Goodman (Polytechnic University, Brooklyn, NY)and Robert Myers (Fairfield Resource International) entitled 3G Cellular Standars and Patents (IEEE WirelessCom 2005, June 13, 2005).

The authors had participated in a project which had analyzed the 7,796 patents deemed ‘essential’_[1] to the 483 technical specifications promulgated under one or other of the two competing 3G cellular standards: CDMA2000 under the auspices of 3GPP2 and WCDMA under 3GPP. The paper provides a powerful example of the problems that multiple overlapping monopoly rights can engender in practice, particularly in relation to the promulgation of standards. As they state at the outset:

Governments issue patents to reward innovation and stimulate technology creation. However, distortions in the patent system can stifle creativity and block deployment of the best technology [2], [3]. The problem is especially acute when … a user needs access to multiple patented inputs to create a single userful product. In these circumstances the patent system can retard, rather than encourage innovation [4].

It also gives some indication of the transaction costs involved. For example this limited investigation involved:

• resolving 7,796 patents into 887 patent families (372 patent familes were common)
• employing an expert to evaluate each claim against the standard at an average of 1 hour per claim

If we assume between 15 claims per patent and rate of 100 pounds per hour for the relevant expert this gives a total of 1.33 Million pounds simply to form a preliminary judgement as to what patents were or were not essential to the standard.

Summary of Results

Patents deemed essential by their owners for WCDMA (up to January 1 2004) and CDMA2000 (up to February 5 2004)

	WCDMA	CDMA2000
Patents	6872	924
Patent Families_[2]	732	527
Actually Essential_[3]	157	108
Ratio	21.4%	20.5%

Thus the great majority of patents declared essential turned out not to be upon a preliminary evaluation. However as the authors point out: Nevertheless, a company that creates equipment or services for third generation cellular systems still faces a formidable task obtaining rights to patented technology. Even with the narrow definition of essential … it may be necessary to acquire rights to several dozens of patents, depending on the equipment or service to be produced. In addition to the patents that are technically essential, there are probably other patents that are commercially essential because they contain the best (albeit not the only) possible implementation of the standard. They also note that some companies may have decided not to declare their patents to standards organizations at all (e.g. Lucent).

Footnotes

[1] The official definition of ‘essential’ from ETSI is: “ESSENTIAL” as applied to IPR means that it is not possible on technical (but not commercial) grounds, taking into account normal technical practice and the state of the art generally available at the time of standardization, to make, sell, lease, otherwise dispose of, repair, use or operate EQUIPMENT of METHODS which comply with a STANDARD without infringing that IPR

[2] The members of a [patent] family are patents and patent applications produced in different countries for a single invention. Note that 372 inventions were found to be common to both standards.

[3] This is only the result of a preliminary evaluation: Each patent was assigned to one panelist [an expert] according to the technical area of the patent. The panelists examined the independent claims of each patent and spent on average one hour comparing the independent claims with the standard to which the patent was declared. Based on this evaluation, the panelist formed a preliminary judgement as to whether the technology in at least one independent claim is necessary to implement the standard.

Appendix

Top Patent Owners for WCDMA

	Claimed Essential	Judged Essential
Qualcomm	279	30
Ericsson	129	34
Nokia	94	40
Motorola	38	11
Philips	22	4

Top Patent Owners for CDMA2000

	Claimed Essential	Judged Essential
Qualcomm	340	54
Ericsson	16	3
Nokia	45	14
Motorola	37	14
NTT DoCoMo	18	7

Pankaj Mishra, in a review of Suketu Mehta’s Maximum City: Bombay Lost and Found repeats the following anecdote from that book (NYRB, 2004-11-18, p.18):

In another scene, Chopra, the film director Mehta has worked with, stands before a row of cable operators arrested for screening illegally his new film and orders a pliant police officer to break their legs.

It took 12 years to produce (1988-2000) and cost 4.5 million dollars (according to its editor Richard Talbert). It has a whole page dedicated to listing donors and supporters of the project. It recruited seventy-three compilers, with ten regional editors with ninety-five reviewers and twenty-two cartographers.

It is 148 pp. long and with companion gazetteer comes in at $350.00 (if you take the gazetteer on paper — 1,383 pp. — it comes down to $150.00). This implies a unit price to fixed cost ratio of 1 to 10,000.

Reference: NYRB 60 ff. 2001-04-26

Someone mentioned to me the other day that Microsoft had spent a $1 billion on implementing DRM. This is a huge sum and should be counted as pure transaction costs (and just shows how dumb, compared to the alternatives, the whole idea of DRM is).

You often hear the following argument for retrospective term extensions put forward_[1]:

‘Retrospective term extensions will allows us to make more money on old artists we can then use to fund new talent’

This is a fallacy though all the more dangerous for being superficially plausible (for reasons we detail below). It is a fallacy for a simple reason:

1. A retrospective term extension increases the return to past projects but makes no difference to the expected payoff of new projects.
2. Investment in new projects is determined by their expected payoff alone_[2].

Point 1 is self-evident but point 2 perhaps deserves some elaboration.

In the (popular) music industry, just like many others, investors (labels) invest a variety of risky projects (bands). Some projects will be successful and some will not. Like any normal business investors will, in general and on average, invest in a project if its expected profit is positive.

This does not mean a given project will make money, in fact quite the contrary. Many projects, perhaps the majority, will lose money but the losses on these will be made up for by the successful ones so that on on average an investor makes money (averaged over its portfolio). In the music business, which is relatively risky, this is exactly the case. Most bands fail but a few make it big and return a large amount to the label.

Thus while successes do, in some sense, help to pay for failures this is only true prospectively — if one knew in advance which projects would be failures one would not invest in them. Making more money (or less money) on existing projects makes no difference_[3].

Footnotes

[1] See James Purnell, Minister for the Creative Industries as quoted in the Scotsman (excerpted on http://www.freeculture.org.uk/campaigns/14plus14/CopyrightTermExtension) [2] This assumes that the investor is not credit constrained so that she or he can make the investment. This is a reasonable assumption for most investors in music (especially for the major labels) and in any case should credit-constraints be an issue they should be addressed directly (for example by providing special lending facilities) rather than by retrospective copyright term extensions. [3] To illustrate take the extreme case an investor all of whose potential new projects are terrible (it’s clear none of them will make money). Suppose suddenly this investor receives £100 million extra for a past project. What wil the investor do? Any sane investor just put the money in the bank or return it to its shareholders. It certainly won’t use the money to invest in the guaranteed loss making projects.

Appendix

Email to Creative Friends list:

On Wed, 8 Jun 2005, Rufus Pollock wrote:

The minister’s economic analysis in this latest announcement is very flawed. Any business supports projects on the basis of expected profits. It may use past profits to finance the initial outlay but only if it expects the project to (on average) pay that back. Thus the idea that money made on successful bands helps unsuccessful bands is fallacious. The fallacy can most easily be seen by taking the argument to its logical conclusion which is that we should simply hand over a £100 million to the record companies — which strangely no-one is suggesting.

Although the rest of your post was sound, this analysis is flawed. It is common practice in many industries to spend money on a portfolio of risky projects where it is expected that only some small proportion of the projects will actually turn a profit, but those that do will more than cover the costs of the failed projects. Examples: pharmaceutical companies; R&D departments at the likes of IBM, Intel, and Microsoft; the whole venture capital industry;, etc, etc. I doubt very much that someone signing a band can say with any reliability at all how much money they’ll make, so I’d be staggered if the same practice wasn’t common in the recording industry.

Your point about diversification over risky projects is a good one. However I think there has been a misunderstanding. Yes, revenues from successful bands cover losses on unsuccessful ones but I wouldn’t term this as ‘pay’ for or as cross-subsidization. Why?

Cross-subsidization to my mind denotes intentionally using profits from one area to fund an area that loses money, where this is /known in advance/. For example urban utility users cross-subsidize rural utility users.

But what happens in music and the other ‘risky’ ventures you describe I would term diversification. In advance it is unknown whether any given project will succeed only that it will succeed with some probability. By diversifying across projects we reduce variance while maintaining the average payout. It is of course true that successful projects cover the costs of unsuccessful projects but all projects are equally likely to succeed or fail in advance (the cross-subsidy is unintentional — and would be eliminated if it could be).

This distinction is essential for it explains the divergent effects of increasing payouts in the two cases.

1. Increasing payouts in case of cross subsidy:

Suppose initially a successful project makes £X and that this is used to fund a loss-making project (for whatever reason). Now suppose due to changes the project makes £X + £Y there is no reason at all to think that this extra £Y would be used to pay for more loss-making projects, instead it should be just put in the bank. Moreover increasing the payoff in this situation by £Y is exactly the same as handing over £Y directly.

1. Increasing payouts in case of diversification:

Suppose average payout is £X initially but will be £X + £Y after change (copyright term extension say). This means that the firm may now take on riskier or more costly projects. (now p * (X + Y) >= cost whereas before p * X >= cost)

Thus it is correct to say that money made on successful bands does not help unsuccessful bands since this would be cross-subsidy. However the prospect of making money on successful bands does result in investment in bands in general some of whom will lose money.

All this may seem academic but there is an important point. Retrospective term extensions increase income in the first manner, i.e. the increase the income to /known/ successful projects. Thus they should have no effect at all on the funding of new artists — they just increase the profits of record companies or the income of a few lucky descendants of very succesful artists.

Prospective term extensions are of the second kind and increase payout to individual projects in the /future/. Thus these can have a positive effect on funding. However since the extra revenues from term extensions are so distant their present value is very low (£1 in fifty years is worth less than 1p at a 10% discount rate). They are therefore not of much interest either to the labels or to individual artists.

My point about the minister’s statement (which hasn’t yet appeared in an official press release as far as I know) is that it conflates the two effects by suggesting that term extensions somehow relate to ‘diversification’ whereby the prospect of successes such as coldplay allow for investment in several groups many of which may fail. As we have seen this isn’t really the case since retrospective term extensions are ‘cross-subsidy-like’ and prospective term extensions increase revenue so minimally as to be practically irrelevant.

Regards,

Rufus

This speech was delivered in my capacity as Director of FFII-UK in the “IP and the Knowledge Commons - Political Parties” panel of the TACD conference on The Politics and Ideology of Intellectual Property, which took place in Brussels on the 20-21 March 2006.

Politically, IP is where the Environmental movement was 30 years ago

I always prefer discussion and questions so I’m going to keep my formal presentation very short. In keeping it short I’m also going to restrict myself to telling you one, well maybe, two things.

The first is that, at present, when it comes to intellectual property there are no political parties. That is there are no, or very few, discernible ideological differences between political grouping on intellectual property (and on innovation policy in general). If you look at other areas: labour law, monetary policy, etc you will see clear differences between political parties. In advance you can predict with a fair degree of confidence which way a party or grouping will go. But when it comes to intellectual property that really isn’t the case.

What this means is that when it comes to voting either all parties tend to go the same way or all parties split. The most extreme case I know of this second situation was the first reading of the European Parliament on software patents when practically all of the major groups split, in some cases not only at the supra-national but at the national level.

Why is this? I think the answer is fairly simple. For intellectual property there are, very, very roughly, two sets of interests. On one side you have rightsholders. On the other side you have the general public. Rights-holders see direct gains from extensions of IP be they in term or scope. The general public by contrast while they may see benefits from extensions in the form of increased innovation, also bear the costs.

At the same time rightsholders are generally very concentrated while the general public is diffuse and poorly organized — so much so that many significant changes pass virtually unnoticed.

This means that:

1. much of the time the debate is very one-sided. Because it is one-sided there is no need for parties to take ’sides’
2. the interests that do exist are very broad-based and tend to impinge equally across the political spectrum

This brings me to my second point which comes in the form of an analogy. It is the analogy between debates over innovation and intellectual property and those over environmental issues. I believe that where we stand today politically with respect to innovation policy and IP is where we stood with respect to environmental issues 30 or 40 years ago.

Just like innovation policy, with environmental issues you see concentrated interests pitted against those of the general public. There also you have growing political engagement as a result of significant external changes. And just like with the environment 40 years ago we are just beginning to build a movement to properly represent the public on the relevant issues.

Today when you look at a political party it will have a position on the environment and we even have ‘Green’ parties — albeit generally small ones — specifically focused on those issues. We’ve even got to the stage where, in the UK, when the Conservatives (which is a right wing party) took out full-page ads to announce its new policy agenda one of the five bullet points was about the environment.

Similarly I think 30 years from now innovation policy will have the same prominence. All political parties will have positions on these kinds of issues, and not buried away somewhere in their manifesto but the kind of things they mention when they take out those full-page ads.

There will also be a much, much fuller civil society engagement in these issues. Just think of Greenpeace, Friends of the Earth, Conservation International now and how they were 40 years ago (if they even existed). In 40 years I believe we’ll see organizations on the same kind of scale and with the same level of membership in the area of innovation policy.

This won’t mean no IP, quite the contrary, but I do believe it will mean a far better balance in the way we use and regulate IP, and that ladies and gentleman can only be a good thing.

Thank you.

The BSA (Business Software Alliance) has released a study (prepared by IDC) along with a heavy PR blitz claiming all kinds of absolutely astronomical benefits from reducing software piracy.

There are immediate reasons to be suspicious. While the study is touted as an ‘economic’ analysis of the benefits of combatting piracy no self-respecting economist I know of would come out with an ‘estimate’ in the way the BSA does because we just don’t have the data to do that. At the same time there are no difficulties in pointing out multiple errors in the BSA’s estimates which indicated they are almost certainly wildly too high.

Of course we should remember as we get hot under the collar about this latest piece of misinformation that that for those of us opposed to software monopolies the best thing right now for the developing world would be a really massive crackdown on piracy with draconian enforcement. That way people would stop with the free candy (MS windows + office) and get into linux before it is too late :).

Below is a quick analysis of the paper pointing out some of the bigger howlers.

1. they claim with a section title: “Lower Software Piracy Produces Higher IT Benefits” (p.4) and then produce as their evidence the fact that:

“A country’s software piracy rate is a key dif- ferentiator among countries that enjoy vast IT eco- nomic benefits and those that have yet to unlock them …

In general, there is an inverse relationship between a country’s software piracy rate and the size of its IT sector as a percentage of GDP. Thus, the lower the software piracy rate, the higher the IT related bene- fits, including IT-generated taxes.’

[+ nice accompanying graph and repetition innumerable times in the section]

This is outrageous and is just the old analytical fallacy of equating correlation with causation: just because more piracy is associated with a smaller IT industry doest not mean piracy causes a smaller IT industry and that reducing piracy will increase the size of the IT industry. In fact it seems more likely causation goes the other way (smaller IT sector -> more piracy) or is simply the result of omitted variables: developing countries have both a smaller IT sector — e.g. due to education levels, etc, and more piracy — because commercial software prices are higher, enforcement is worse etc. Putting these two facts together but leaving out conditioning on development will give you the resulting correlation

2. p.13 Summary of IDC’s methodology:

‘the theoretical losses from piracy in terms of revenue to software vendors, software- related revenues to services firms, and software- related revenues to channel players. Employment losses are calculated from revenue losses, and only apply to employment in the IT industry, not IT pro- fessionals in end-user organizations (although IDC believes there is some impact.) Tax revenue losses are calculated from revenue losses (VAT and corpo- rate income tax) and employment losses (income and social taxes). The software losses are based on the piracy rate and equal the value of software installed and not paid for, adjusted by IDC’s soft- ware analysts to account for software in a country (such as enterprise and server software, not meas- ured in the annual BSA study).’

So there it is baldly stated, they calculate societal loss by:

a) assuming that every pirated copy would have been purchased if piracy were prevented (blatantly false) b) take no account of general equilibrium issues: i.e. that people have a fixed budget so that if you suddenly spend a load more on ‘commercial’ software (because you pay for everything) that means less to spend on other goods (such as bespoke software, fridges, apples etc)

This suggests:

1. They have never heard of a demand curve (i.e. that people buy different amounts depending on price!)

2. Sampling effect: ‘piracy’ is one way for people to sample a product before buying. Even in the developed world we often see people using a friend’s copy at home but the purchased version at work which is a similar situation (Let’s remember that the biggest shareware company in the world is microsoft)

3. That estimates of welfare/revenue losses (at least on this scale) have to be done on a society wide basis so that you include the displacement effects from spending more on non-pirated software

Finally and most importantly, the whole study gets its analysis wrong from the very start. For information goods, once the good is made it is optimal for societal welfare to distribute at marginal cost (i.e. at 0 for software).

Thus piracy of e.g. MS windows should be welfare improving for the world as a whole and definitely welfare improving for every developing country. Of course this may be bad for MS (though that depends on sampling effect).

Now the usual trade-off (and the reason we have IP) is that if it was known in advance that everything would be sold at marginal cost there would not be the money to pay for the original development. Thus we do grant a temporary monopoly to help rather than distributing at marginal cost.

What this means is that any ‘economic’ study of the effects of piracy is really about the trade-off of lower consumer prices (due to piracy) vs. less innovation.

Given that most piracy involves: a) developing countries (so that loss of sales is low since they wouldn’t be able to afford the stuff anyway) b) big name products (who are already deep within the black in relation to paying for the innovation)

It seems on the face of it likely that software piracy is actually welfare improving. (Backing this up with proper empirical evidence studies is a mammoth task because we don’t really know the demand function for software or the supply function of innovation).

I was looking through The Political Economy of Intellectual Property Treaties. 2004. Suzanne Scotchmer. Journal of Law, Economics and Organizations 20:415-437. curtesy of the PDF available on her web page when I came to the following section which I wished to copy and paste into my notes:

The Organization of Economic Cooperation and Development (OECD) (2002) reports that in 2000, only 56% of R&D spending in the European Union was by industry. The industrial share is higher in the United States, about 68%, still a considerable departure from 100%. In Latin America, public sponsorship[ tends to be dominant. In 1996, industry funded only 40% of R&D in Brazil, 28% in Argentina, and less than 20% in Chile, Costa Rico and Mexico (National Science Foundation, 2000)

But guess what: an attempt to copy results in the following message: ‘Without the proper password, you do not have permission to copy portions of this document. Would you like to enter a password for the document?’. Here in the flesh we have an example of DRM impeding fair-use and scholarship. In all, timing myself, I estimate it took me around 30-60 secs to transcribe the quote and undoubtedly it will be of lesser quality (checking for typos/spell-checking would increase the time). Okay so this is a real annoyance but what is this costing society?

Estimating my time at around £10 per hour this transcription equates to 10-17 pence. Cost of copy and paste would have been around quarter of this (allowing for tidying of PDF text) meaning DRM has just cost me between 8 and 12 pence. This may sound trifling but suppose I read 4 papers a week and want to extract 2 quotes each from them. My quote above is probably average in length (perhaps even a little on the short side since very simple stuff I often paraphrase) so taking this cost as average we have 52 * 4 * 2 * 0.10 ~ £40. Multiply this by the number of researchers, and we are starting to talk about a serious cost if such DRM became widespread.

Unsatisfactory Nature of Traditional Growth Modelling

There is a large intellectual discrepancy between most of the formal growth models economists have devised, and descriptions of growth that take the form of economic history. Contemporary formal growth theories treat economic growth as almost all ‘quantitative’. They aim to explain why various magnitudes, like per capita income, the real wage rate, and capital intensity, rise over time. Other magnitudes like the savings rate, or the share of labor in national income, or the rate of return on capital, tend to stay constant in these models, either because they are assumed to be constants, or because of various mechanisms built into the models. In any case, in economic growth as it is thus depicted, nothing much happens qualitatively. On the other hand, in the historical accounts, lots of qualitative things are happening. New technologies are emerging, and so also are new forms of business organization, and new institutions. Put another way, development is moving forwards and not simply things getting bigger or smaller or staying the same size.

Source: Richard Nelson in dosi_ea_1998, p. 319-320