IP, BTE and funding startups

In running the @KeckGrad business plan competition this week, I was struck by how different our students’ life science startups were from the retail or IT startups that are common at other colleges. In most cases, our students needed $5 to $50 million in outside funding to jump through regulatory hoops and generate initial revenues.

It seems like this also reflects a fundamental difference in IP strategy — how a successful startup discourages entry or imitation, and how that ties back both to their funding needs and their IP strategies.

The retail startups have a brand, and locations, and perhaps a little bit of internal process trade secrets. Seed funding is available, but (shy of an IPO or acquisition) any subsequent growth tends to be organic and self-funded. Whether restaurants or clothing, these sorts of startups take years to build up, and the margins are generally thin.

The IT companies rely on copyright and trade secrets to protect their implementations, and hope to build network effects or switching costs to discourage entry. A hot property attracts plenty of money, because the scale is small and (if successful) the TTM and thus the payback time is quick. Still, a well-funded, well-run competitor could catch them. Many startups hope that the differentiator is the vision and positioning: for example, MySpace had years to respond to Facebook, but somehow never did.

Then there’s the life science companies: they need spend (and thus raise) huge amounts before ether generate revenues and — given mandatory regulatory disclosures — give rivals plenty of time to see what they are doing. The only way this works is if you have a patent, that gives investors an ironclad assurance of exclusivity for some period of time.

I'm not exactly sure what the model is for cleantech — but maybe there isn’t one. Certainly renewable energy — such as solar or biofuels — the hope is to leverage economies of scale to attain cost advantages in producing commodity energy. Given the hope of scale as a BTE (or BTI), many companies bulked up quickly, leaving a lot of dead companies strewn along the way. China’s Suntech was once world’s largest solar company — the first to sell 2 gigawatts of solar panels in one year — but is now shrinking and bankrupt.

If we look at older, mature industries, scale is never enough. Scale has reduced (but not eliminated) competition in electronics, steel and autos, but has done nothing to provide barriers to imitation to protect HP, US Steel or GM from subsequent entrants. If anything, the race for scale has led to overcapacity and thus price wars — in steel, DRAM, LCD panels, solar panels.

Intractable online IP policy controversies

Hat tip: Matt Asay (@mjasay)

Like anyone who follows the tech industry, I’ve been seeing apocalyptic stories, emails and seminar announcements about how SOPA (the Stop Online Piracy Act) will destroy the Internet, Internet companies and technological innovation as we know it.

With my new job, I haven’t been paying as much attention as I’d like to, but from what I’ve read, it’s another law in the general direction of the (much loathed) Clinton-era Digital Millennium Copyright Act. Certainly the motivation and opposition sound nearly identical to DMCA.

What also sounds similar is the oddfellow coalition: I mean, how often do Republicans try to make Hollywood studio execs richer? And how often do Democrats have to choose between the supporting the entertainers (and entertainment execs) who help them raise millions against the grass roots consumers and the EFF/civil liberties types?

I spent a fair amount of time studying DMCA. After briefly working as a consultant to one of the earliest Internet music streaming sites (Live365), I wrote two teaching cases about the challenges facing the entertainment industry in the Napster generation, and the trade-offs of Federal imposition of terms for content producer-consumer negotiations (particularly the Online Royalty Tribunal).

My favorite part of that era was when I asked my MBA students to do a five forces analysis of the record industry ca. 1995, when there were 6 major labels instead of the soon-to-be 3. After walking through the buyer power, supplier power, substitutes, rivals and entry barriers, I led them to the inescapable conclusion: this is as close to a cartel as you’re ever going to see. (Hint to students: professors love examples of industries with very high or very low forces.)

There are no white hats here. The DMCA was a sledgehammer to kill a mosquito, but that didn’t make the mosquito any less deadly. My sense is the SOPA is roughly the same.

So I was thrilled to read a blog entry (tweeted by Matt Asay) by John Lilly:

What’s bothering me about the SOPA “discussion”

There are 3 things that have really been bothering me about how the SOPA/PIPA discussion has been going so far.

1. it’s not a discussion at all — it’s people calling each other names.
2. it’s highly likely to have a result that is unhelpful at best, and insanely destructive at worst
3. we’re building a completely worthless/bad roadmap for how to deal with technology policy going forward, and it’s going to get worse

Let me be very clear: SOPA is a terrible law that should not be enacted under any circumstances. It’s broken technically and misguided from a policy point of view. It not only won’t accomplish what advocates want it to accomplish, but it also will create backbreaking burdens and barriers to entry for some of our most promising technology companies and cultural movements of the coming decade.

But also: content creators & owners have a legitimate beef with how their content can be appropriated and distributed so easily by rogue actors.

I don’t share John’s interpretation of the law. Part of this is because I’ve not read the law, nor heard more than a smidgen of the articles regarding it. But part of it is also because entertainment companies are getting killer, the other side (consumers, Silicon Valley, cyberlibertarians) are at best indifferent to their situation and it’s only with the threat of bad legislation will they make even the most modest concessions.

John was particularly funny as he caricatured the name-calling on both sides:

[Here] I use the term “conversation” here very loosely, since it has characteristics more like a bunch of schoolyard name calling. The conversation that’s happening is going more like this:

- content: “you people are stealing our stuff. you’re thieves”

- techies: “we’re not stealing it. we’re just building great apps for users.”

- content: “you’re ignoring the problem and helping the thieves. you’re effectively pirates, so we’re going to shut everyone down.”

- techies: “you’re acting like jackbooted fascists, embracing censorship and your’e going to end everything that’s good about culture today.”

- content: “we’re trying to protect our content — you guys are pretending like there’s no problem, then getting rich off platforms that pillage our content.”

- techies: “you don’t understand how the Internet works — how do you even live life in the 21st century? dinosaurs.”

So that’s awesome. Then you throw Congress into the mix and hilarity ensues. Because if you’re looking for folks who really do not act like they want to understand the Internet, Capitol Hill is a pretty good place to start.

John advises that “we need to be thinking about copyright law — in an age where copies are the natural order of things, as opposed to previously, when it was harder to make copies.“

What I find particularly depressing here is that fixing entertainment copyright law is trivial compared to fixing patent law. Here we have a pretty simple mix of stakeholders: content producers, content consumers, interested bystanders and pirates.

When you’re trying to fix patent law, you a particularly rich mix of people in an industry. For example, for the patent term on a therapeutic compound you’d have big pharma, startup biotech, insurance companies, the Feds (Medicare/Medicaid/VA), consumes, and senior citizen lobbies. But when the same patent system also impacts IT — with components, systems, software, open source, patent trolls — and dozens of other industries, then the prospect of finding a patent reform compromise that a majority will endorse becomes virtually impossible.

Antitrust won't help Google's codec fight

Nowadays, Google is normally considered the next monopolist on the losing end of antitrust scrutiny. This month, Google is getting help from antitrust authorities in its uphill efforts to get VP8 video codec (and its WebM project) established against the dominant H.264 codec.

In January, Google announced plans to drop support for H.264 in its Chrome browser, nominally because of the H.264 patent royalties. Meanwhile, Google licenses VP8 and its associated patents with the claim that it’s royalty free:

Please explain how WebM is "royalty-free."

Some video codecs require content distributors and manufacturers to pay patent royalties to use the intellectual property within the codec. WebM and the codecs it supports (VP8 video and Vorbis audio) require no royalty payments of any kind. You can do whatever you want with the WebM code without owing money to anybody. For more information, see the License page.

However, all Google can promise is that it won’t charge royalties on VP8. It can’t promise that its source codec doesn’t infringe the patents of others: that’s up to the patent holders to allege and (ultimately) for a court to decide. (Its policy is that people can’t use WebM without licensing their own patents to WebM users.) The only way around third party patents would be for Google to indemnify users against these claims.

Since MPEG LA is in the business of managing and licensing patent pools, including a pool of some 800 patents that read on H.264 (including 273 US patents and 469 Japanese patents) it has long expected that many of these patents could also be essential for implementation of VP8. So last month it asked all parties (both current and other licensors) to submit their patents for review. Needless to say, this brought squawking by Google and the usual anti-software patent crowd.

Now, the Wall Street Journal reported that the US Justice Department is looking to help Google in its efforts to resist MPEG LA patents:

Antitrust enforcers are investigating whether MPEG LA, or its members, are trying to cripple an alternative format called VP8 that Google released last year--by creating legal uncertainty over whether users might violate patents by employing that technology, these people added.

The Justice Department and Federal Trade Commission start many more antitrust investigations than they bring to court. There is no per se legal problem if patent holders want to charge royalties to VP8 users, as long as they are on comparable terms to those charged to H.264 users. As the FTC antitrust website states:

Restraints in the supply chain are tested for their reasonableness, by analyzing the market in detail and balancing any harmful competitive effects against offsetting benefits. In general, the law views most vertical arrangements as beneficial overall because they reduce costs and promote efficient distribution of products. A vertical arrangement may violate the antitrust laws, however, if it reduces competition among firms at the same level (say among retailers or among wholesalers) or prevents new firms from entering the market.

So the only problem is that MPEG LA wants to run the licensing business for both H.264 and VP8. Normally in competing standards (think Blu-ray vs. HD DVD) competing organizations run the patent pools. So while the DoJ will not find anything wrong with charging patent royalties for VP8, it might force MPEG LA to exercise greater transparency in royalties or even to spin off the VP8 pool efforts.

The research has clearly shown that patent pools increase the efficiency of patent collection for both licensors and licensees. (The only downside is that licensees might hope that without pools that would-be licensors might not bother to enforce royalties.)

If MPEG LA comes under pressure, I think it could easily solve the transparency problem by separating royalties into three piles: patents that apply both H.264 and VP8 (or WebM), those that apply to H.264 and those specific to VP8. If the terms for the first pile are the same for either standard, VP8 supporters would be hard-pressed to show any anti-competitive effects of the patent policy — other than their fantasy of releasing a patent-free codec.

All this kerfuffle over MPEG LA royalties ignores three other factors:

• the effect of the annual royalty cap (as with H.264) in making MPEG LA royalties negligible for large firms like Apple, Google or Microsoft
• as with any other standard, the existence of other patents that are not “essential” to the standard but may be commercially necessary
• Google’s ability to use its own patent leverage (including those it acquired when it bought the WebM developer) to force a cross-license or non-assertion agreement by key patent holders

Google's war on semi-open standards

Google stirred up a controversy this week with its decision dumping H.264 video codec support from its Chrome browser in favor of Flash and its own WebM (VP8).

This clearly is good for Adobe’s Flash, and bad for efforts to build a Flash-free HTML5 Internet that was (until this week) a joint effort of Microsoft, Apple and Google.

The claim that Google is motivated by openness is quite hollow. While technically a Windows browser doesn’t need Google’s help to distribute a free Flash player, Google has been very pro-Flash in its efforts to help Android overtake the iPhone.

Also, even though royalty bearing, H.264 is an open industry standard, whereas Flash and VP8 are not. Flash has only one proprietary implementation.

Still, some speculate that argument one reason is the H.264 business model, specifically that Google doesn’t like the H.264 royalties charged by MPEG LA. Here is what Google’s revised justification said Friday:

We acknowledge that H.264 has broader support in the publisher, developer, and hardware community today (though support across the ecosystem for WebM is growing rapidly). However, as stated above, there will not be agreement to make it the baseline in the HTML video standard due to its licensing requirements. To use and distribute H.264, browser and OS vendors, hardware manufacturers, and publishers who charge for content must pay significant royalties—with no guarantee the fees won’t increase in the future. To companies like Google, the license fees may not be material, but to the next great video startup and those in emerging markets these fees stifle innovation.

The idea that Google’s latest push will cause VP8 to pass H.264 is fanciful at best: it will take more than support from the #3 browser to cause the rest of the industry to shift from H.264. If anything, Google’s efforts fragment and thus undercut any efforts to establish an open alternative to Flash.

One theory is that Google wants to ditch H.264 support from YouTube (which, if true, would send iPhone users away from YouTube — good for Android, bad for YouTube.) The theory that Google hates H.264 royalties doesn’t hold water according to an analysis by Ed Bott of ZDNet, because even the worst case cost is not material for a $29 billion/year company.

Clearly there is more to this strategy than meets the eye. A company that aspires to be the (unregulated) benevolent dictator of the Internet would be more transparent about its motivations — perhaps something the next CEO will be better at.

But for now, the only good explanation I’ve found is at the comic strip “Joy of Tech,” which argues that it’s part of a cynical Machiavellian strategy by the “do no evil” crowd to retaliate against Apple and generate controversy.

Productive publishing period

[HornTooting]
In terms of quantity, 2009 was my most productive year ever for academic publishing. After publishing three journal articles in 2008, in 2009 I published five: one about telecommunications, one about standards, two about telecommunications standards and one about open innovation.

The first two articles were based on a four-year collaboration with my now-friend Rudi Bekkers, looking at patents in W-CDMA. One paper focused more on the case study of standardization, while the second paper looked at the quality and timing of essential patents as laid against the standardization process. The latter paper was published in Telecommunications Policy, the leading academic journal on, well, telecommunications policy.

Building on Rudi’s earlier pathbreaking papers on GSM patenting, we noted several shifts from the 2G to 3G era in European mobile phone standards. The number of essential patents increased eightfold and the number of claimants increased threefold. Equipment makers retained about the same proportion of overall patents, but the network operators virtually disappeared, replaced by component suppliers (notably Qualcomm) and technology licensing firms (notably InterDigital).

These were our conclusions:

The sources of UMTS patent proliferation have often been ascribed to IPR-focused companies outside the ETSI process, particularly Qualcomm and InterDigital. However, this study shows that the largest numbers of patents are held by two firms (Nokia and Ericsson) centrally involved in the UMTS standardization, and the timing of their patenting suggests that they used their knowledge of the standard’s development for anticipatory patenting—further contributing to patent proliferation.

Still, a cozy oligopoly of four main UMTS patent holders might have produced a manageable IPR regime comparable that to the five major holders of GSM patents. However, the number of firms claiming at least a one patent has grown threefold, increasing the risk of holdup, transaction costs and royalty stacking for firms implementing the newer standard. This uncertainty is magnified by the self-determination of essentiality: while it is virtually impossible to determine how many of the 1227 patents are actually necessary to implement UMTS, at the same time other parties may fail to provide an itemized list of essential patents.

This was not even the longest collaboration of the papers. One paper was based on an eight-year collaboration with my friend Scott Gallagher, which began when we met during the bubble era conference of the Strategic Management Society (2001) in San Francisco.

Then as now, the goal was to re-examine, critique and extend the traditional view of positive feedback in the adoption of standardized goods. We brought together a number of observations that (when we started) were somewhat novel, although the field has not stood still during that period. After many delays (including other projects, work and life), the paper was published in September in the Journal of Engineering and Technology Management, a respectable journal that has attracted papers from some of the top names in the field

I’ve already mentioned on my other blogs the two other papers published in 2009.

In April, I published the cover article in the Journal of San Diego History, based on my research into the origins of the San Diego telecom industry. The paper was entitled “Before Qualcomm” to make it more relevant to the general readership, and traced the early round of spinoffs of Linkabit, the region’s seminal company. It also included discussions of the role of Qualcomm co-founders Andy Viterbi and Irwin Jacobs in applying Claude Shannon’s to space communications, drawing on my 2008 article in the Journal of Management Studies.

The fifth paper is the first in what I hope will be a series of papers that contrast open innovation with user innovation and related theories. Published in the Washington University Journal of Law and Policy, it (not surprisingly) focuses on policy issues related to open, user (and cumulative) innovation.

I’ll be glad to send out a PDF of the published version of any paper to anyone who’s interested.

This morning I got email notice of acceptance of my first paper for 2010, a paper on the success of the iPhone that will be published by Telecommunications Policy. Michael Mace have been working on this paper since 2007 — actually before the iPhone shipped — although our understanding of the phenomenon has shifted significantly since then. Additional details as they become available.

[/HornTooting]

References

Rudi Bekkers and Joel West, “Standards, Patents and Mobile Phones: Lessons from ETSI’s Handling of UMTS,” International Journal of IT Standards & Standardization Research, 7, 1 (January 2009), 13-34.

Rudi Bekkers and Joel West, “The Limits to IPR Standardization Policies as Evidenced by Strategic Patenting in UMTS,” Telecommunications Policy, 33, 1-2 (Feb.-March 2009): 80-97. DOI: 10.1016/j.telpol.2008.11.003

Scott Gallagher and Joel West, “Reconceptualizing and expanding the positive feedback network effects model: A case study,” Journal of Engineering and Technology Management 26, 3 (Sept. 2009): 131-147. DOI: 10.1016/j.jengtecman.2009.06.007

Joel West, “Before Qualcomm: Linkabit and the Origins of the San Diego Telecom Industry,” Journal of San Diego History, 55, 1-2 (Winter/Spring 2009): 1-20.

Joel West, “Policy Challenges of Open, Cumulative, and User Innovation,” Washington University Journal of Law & Policy 30 (2009): 17-41.

Health reform vs. biotech innovation

A major reason for the high cost of US healthcare is that Americans pay full prices for brand new drugs that are either not available in other countries or (due to monopsony buying power) sold at a vastly reduced price. We have an American pharmaceutical (and now biotech) industry because of this policy, and we get solutions before anyone else, but effectively American consumers subsidize the rest of the world.

Without healthcare reform, one of the biggest healthcare policy questions before Congress would be establishing the policy for generic biotech drugs. Unlike small molecule (chemically produced) drugs, it is much harder to establish the biological equivalence of large molecule biologics (which tend to be protein-based created using recombinant DNA) without conducting new clinical trials.

Of course, generic biotech drug makers do not want to be required to conduct expensive and time-consuming clinical trials. They would rather sell knock-off drugs after patents expire, just as small molecule generic makers did after Hatch-Waxman came into effect.

The government has two hats here. As the safety regulator, it must reduce the chances of unsafe drugs being sold. As the nation’s largest drug purchaser (and antitrust regulator) it wants lots of competition for drugs to push down prices, as Hatch-Watchman has done for small-molecule drugs.

Alas, there is also the little inconvenient problem that if the biotech companies never get monopoly rents, then they won’t get VC investment and won’t develop drugs and probably won’t even come into existence. Large risky investments in innovation (new compound investments are among the riskiest) don’t get made without financial incentives and a supply of capital. (There are academic criticisms of using patents to incentivize innovation, but patents are a much cleaner incentive for pharma innovation than for say software or electronics).

Blogger Gene Quinn of IP Watchdog notes an interesting policy discussion last week on CNBC with two former government officials and the head of BIO (Biotech Industry Organization), the industry trade association. He summarizes the issues succinctly:

[T]he segment is well worth watching and will no doubt dispel the myths and lack of understanding by open-minded individuals who question why the biotechnology industry wanted 12 to 14 years of exclusivity for biologics, when the FTC said zero years of exclusivity would be sufficient, President Obama wanted no more than 7 years of exclusivity and Congress opted for 12 years of protection. Co-anchor Joe Kernan started off the questioning by saying: “How did you get 12 years? Why exclude biotechnology from the cost pressures that everyone else is going to have to live under?”

Surrounded by Republicans, apparently the CNBC host interjected himself to speak for the administration’s position:

Specifically, Kernan said: “it just seems like you are a poster child for the exorbitant costs of treatment and maybe some of the costs we shouldn’t be undertaking and we have to judge where to spend the money.” And people wonder why throughout the health care debate the public has been fighting so hard, despite the erroneous and scandalous labels hurled at ordinary citizens who simply want answers and know the government is lying about so much.

Here, Kernan defines the problem as should we be undertaking the cost of exorbitant treatments? Despite what President Obama and his team say, and despite what Democrats in Congress say, the truth is that the overwhelming majority of health care costs come at the end of life, and the only way to lower costs is to ration care at the end of life, as was suggested by Kernan.

Or, as Harvard economist Marvin Feldstein put it Wednesday: “The Obama strategy is to reduce health costs by rationing the services that we and future generations of patients will receive.”

Patent term is the key policy lever for increasing or decreasing the incentives for new drug discovery: it can make or break companies in this industry (while at the same time, excessive term delays competition that makes readily available).

I hope that the patent issue can be debated outside the context of healthcare reform, because it’s too important an issue to get buried among 1,011 unread pages of the healthcare reform bill.

Business method patents going away?

In the case of In re Bilski, the US Court of Appeals on Thursday rejected a business method patent, in a decision many interpret as restricting or possibly eliminating such patents.

There is a lot of debate about what it means. Legal Times concluded that the appeals court (or SCOTUS) may come out with an entirely new standard; Ars Technica called it “a severe blow” to such patents.

Erick Schonfeld of TechCrunch (posting via the Washington Post) was even more expansive in his interpretation:

If you are one of the recipients of the 1,300 business method patents issued in the U.S. last year, or the thousands more that have been issued rampantly and indiscriminately over the past decade, you are probably out of luck.

Some even speculate that this marks the end of software patents, but that seems a bit expansive.

Both business model and software patents are controversial, but it seems like we’ll have several years of uncertainty before this is resolved.

SCOTUS trims patent excesses

The Supreme Court of the United States issued an opinion eliminating one of the more obvious excesses of patent royalty-seekers. Basically it says that if I license a patent to Intel to make a chip, then I shouldn’t be surprised (or seeks additional compensation) when Intel sells that chip to someone to use in a phone or a PC or a router. This is an extension of the existing doctrine of “patent exhaustion,” which in layman’s term means that once you pay for use of a patent, your customers don’t have to pay again.

The decision by Justice Clarence Thomas in Quanta v. LG Electronics was unanimous. The ruling was covered by the WSJ, the WSJ law blog, the SCOTUSblog, and Patently-O, among others.

While this is seemingly a clearcut victory for sanity, there are two dissents from the peanut gallery. An EFF attorney argues that the limited scope of the ruling may encourage further litigation. And Patently-O blogger Scott Crouch argues that the SCOTUS ruling seems to allow contractual restrictions on the principle of patent exhaustion (i.e. to require additional downstream payment; as he quotes from the ruling:

“LGE points out that the License Agreement specifically disclaimed any license to third parties to practice the patents by combining licensed products with other components. But the question whether third parties received implied licenses is irrelevant because Quanta asserts its right to practice the patents based not on implied license but on exhaustion. And exhaustion turns only on Intel’s own license to sell products practicing the LGE Patents.”

“No conditions limited Intel’s authority to sell products substantially embodying the patents. Because Intel was authorized to sell its products to Quanta, the doctrine of patent exhaustion prevents LGE from further asserting its patent rights with respect to the patents substantially embodied by those products”

Certainly it would be bigger leap for the court to hold that “you can’t grant restricted rights via contract.” Of course, the licensees would find a ready loophole if the licenser's actions implied that full rights were being granted.

So is this a significant patent reform that produces further clarity in patent rights, or merely generate more tightly written contracts by licensors?

Rambus wins -- who loses?

The Court of Appeals overturned the Federal Trade Commission antitrust sanctions against Rambus, which in turn overturned an earlier FTC administrative law judge who ruled in favor of Rambus. The whole issue was whether Rambus was honest about its patents in a timely fashion while JEDEC was making a RDRAM standard.

The case of Rambus and its IP strategies are among the most controversial issue in IP and standards. I would say “among” only because as with certain other topics, only one side of the story tends to get reported and many would just as soon see the company lynched. Rambus did win one trial alleging patent infringement by big vendors, who never really wanted to pay royalties.

Even for those (few) who sympathize with Rambus, the ruling has the unfortunate effect of introducing turmoil into many aspects of IP and standardization. The Rambus decision was previously a bright line that SDO managers (and participants) could count on. Now, we need to see whether Rambus’ escape is merely a process issue (JEDEC left a loophole that meant Rambus couldn’t be punished) or if it sets a more general precedent — theoretical or practical — that makes it impossible for SDOs to prevent gaming of the system.

Perhaps the 1995 case that Dell settled would be the precedent for now. My sense is that absent FTC v. Rambus we will need a new legal decision (or administrative act upheld by a court) to set the boundaries of gaming the system. There have been very few cases on this issue that have gone to trial in the past 15 years, so it might be another 5 years before we get a controlling precedent. That uncertainty would be good for law firm billable hours but bad for everyone else.

IP haves and have nots

The question of nullifying or not enforcing patents for foreign drug manufacturers has been a brewing controversy over the last few years. This year’s earlier dust up was between Abbott Labs (of the US) and Thailand, and this month it is Novartis (from Switzerland) and India. There is also the issue of whether these drug companies will be supported by their home governments, or attacked for political purposes.

Having a consistent enforcement of intellectual property was supposed to be solved by having emerging countries like India and China join the WTO, and then enforcement of the TRIPS (Trade-Related Aspects of Intellectual Property Rights) agreement that was part of founding the WTO. Still, enforcement of IP rights is up to national governments, legislatures and even local magistrates.

Supporters of the Indian (or Thai or other) actions argue that human health is more important than drug company profits. (Presumably not economists let alone Friedmanites making this argument). Or that rich markets should cross-subsidize poor markets. Ironically, the same activists who argue that developing countries should import the labor and environmental laws of Western countries are the most vocal in trying to keep out the IP laws.

Such pro-LDC analyses tend to downplay the benefits to local pharma companies that will make money from the compulsory licensing (or nullified patent). But then, before TRIPS, it was quite commonplace for countries to ignore foreign IP to help develop their local industry. Legally, countries are going to enforce the IP laws they feel like enforcing. No foreign countries are going to go to war over copyright or patent theft.

However, as Manuel Davis observed 40 years ago, TANSTAAFL (i.e., no free lunches). Ranjit Shahani made exactly this point in an interview Friday. Certainly as head of Novartis India, he’s hardly a dispassionate observer — in fact, he was very passionate:

First of all, the Madras High Court’s decision is not a setback for Novartis. It is a setback for innovation, for public health. …

On the issue of data protection and patents, the “copycat Indian generics” companies and MNCs are split. But there are no differences [between the MNCs and] the Indian research oriented companies, which are putting quite a bit into research themselves. The average spend of the top ten drugmakers stands at 6.6% of revenues and it compares to 18-20% spent by global drug companies. But this 6.6% expenditure will become infructuose if we don’t have a strong patent law, an environment which protects data and IP. So, it is not the Indian companies but the “copycat generic companies” versus the MNCs.

Shahani makes the argument if you want a domestic industry, you need good IP enforcement. Based on political reality, I argued exactly the inverse in 1995 paper on Japanese copyright law: you get good IP enforcement only when you have a strong domestic industry that needs it.

Meanwhile, as the FT reported earlier this week, Novartis has decided to cancel plans to increase R&D spending in India. As others have noted, research in developing countries is normally a prereq for researching health problems specific to such countries, so India takes a double loss.

Speculation is that the €90 million investment will go to China, consistent with earlier comments by Shahani that China has a more favorable IP climate. Foreign MNC money has created the high-tech industries in many counties — with talent and knowledge that spilled over to local startups - so the Novartis decision could hurt the Indian pharma industry more than they gain from cloning one drug.

I think Novartis is being realistic — go where it’s wanted, avoid where it’s not, and let sovereign governments make their own decisions. If Big Pharma gives up on developing country health problems — because they can only sell the drugs in countries that don’t believe in patent monopolies — then perhaps philanthropist Bill Gates can pay for all the missing R&D. It may be harder for Mr. Gates now that (since March) he’s only the world’s 2nd richest man.

Technorati Tags: business models, intellectual property, patents

When is a patent not a patent?

A number of businesses depend on IP-based business models, none more so than the pharmaceutical companies. With high up front R&D (and low success rate), the patents support high profits (far in excess of manufacturing costs) that provide the gross margins to support that R&D.

As with any IP-based business model, the model is only as good as the enforcement of the IP right. One of the nastier fights this year has been between Thailand and Abbott Labs, a major drug company headquartered in a Chicago suburb.

Thailand doesn’t want to pay what Abbott wants to charge for its Kaletra AIDS drug, and has threatened compulsory licensing (the traditional approach of developing countries). Abbott did what most such companies do, which is to hold off on introducing new drugs into a country where IP rights are uncertain, bringing an angry response from CalPERS (my pension fund).

Not surprisingly, the Chicago Tribune has done a good job of covering its hometown company, including Abbott’s decision earlier this month to capitulate and unilaterally cut its prices. Roger Bate of the American Enterprise Institute also offered a detailed analysis on April 4, as well as an early report on compulsory licensing. The April 24 Wall Street Journal account is available free (at least for now) at PatentLens.

Monday morning, a Wall Street Journal editorial (registration required) blasted Abbott for its unwillingness to stand on principle, the lack of reaction by the Bush administration, as well as the efforts of the World Health Organization to side with Thailand against Abbott. Their conclusion:

The stakes here are far larger than Thailand's greed and WHO's political opportunism. Anti-pharmaceutical activists have looked for years for a government pliable enough to test WTO rules on compulsory licensing. They want to set a precedent that erodes property rights, with a goal of selling drugs at cut-rate prices everywhere. In the NGO nirvana, governments would share the burden of paying for drug research, and then create some kind of "reward" scheme for companies to innovate. This is socialism as alchemy, as if companies will take billion-dollar risks without an incentive to make a profit.

A few hours later, on Monday afternoon the Bush administration criticized the Thai IP policy (covered by the Chicago Tribune and Reuters). I suspect the shareholders of Abbott (and other pharma companies) wish that it had happened sooner.

I spot-checked three or four IP law blogs, and somehow they didn’t think it worth mentioning. That’s shocking. Among the few to cover it has been Patent Baristas.

The case has repercussions beyond pharma. Thailand is not among those countries best known for nationalizing IP, whether for pharma, software, or other technologies. If Thailand is joining those ranks, it would be a disquieting trend.

Meanwhile post hoc weakening of IP means that economically rational R&D managers have no idea what they’ll be able to recoup of their investments. Given how much California firms (Silicon Valley, SD/SF biotech, Hollywood) depend on IP business models, the decision of CalPERS to back those weakening IP suggests that it’s either stupid or captive to cheap political posturing. (Note: the CalPERS board consists of union-elected trustees along with a few politicians).

Technorati Tags: business models, intellectual property, patents

Qualcomm news elsewhere

I’ve decided to post all my future Qualcomm-related news at the SD Telecom blog. I hope that this isn’t a brand extension too far, and that’s certainly not the motivation.

Instead, I see the charter of this blog as offering observations about IT firm strategies — theoretical, descriptive and practical. I follow Qualcomm rather closely for my book, to a level of detail that’s probably far beyond the interests of OITS readers.

Still, I imagine the blog (or at least the Qualcomm-related postings) would be of interest to those that care about patents and patent licensing business models.

Technorati Tags: patent infringement, patents, Qualcomm