Merges (1999) - As Many As Six Impossible Patents Before Breakfast

Jump to navigation Jump to search
Has bibtex key
Has article title As Many As Six Impossible Patents Before Breakfast
Has author Merges
Has year 1999
In journal
In volume
In number
Has pages
Has publisher
©, 2016


Merges, R. P. (1999), "As Many as Six Impossible Patents before Breakfast: Property Rights for Business Concepts and Patent System Reform", Berkeley Technology Law Journal. pdf


No abstract available at this time

Patents on Software and Business Methods

The case State Street Bank vs. Signature Financial (1998) overrulled cases holding or suggesting that claims to "methods of doing business" were not patentable.

There are two attributes to a business concept patent:

  1. It describes an essentially commercial, rather than technological, activity
  2. The hardware or software are claimed at a high level of generality (so that for practical purposes they are nominal)

A bad result would be seeing good, solid businesses abandoned in the face of predatory patent extortionists. The patent system was designed to protect technology. Software removed the distinction between machines and concepts. The foundation for the patent system was laid in the Intellectual Property Clause in the US Constitution. It embraced blind technology optimism, ignoring the costs. By comparison, note that the British Statute of Monopolies prohibited monopolies in business lines and drew the line that "only new and useful inventions could recieve patents", because they could increase social welfare. Now only Congress or the courts can private subject matter limits or otherwise overhaul patenting. The courts have adopted a broad interpretation of "process" and "machine", and Congress has not intervened.

The basic question is: Is it worth the social cost of granting exclusive property rights so that those things will be created. The average number of citations in a business process patent is 5, 3 of which are to other patents (despite that business process patents are so new), suggesting that they are not citing all of the prior art, and so that they quality of the patent examination is poor. On the bright side, problems with business method patents might provide the impetus to reform the entire patent office.

Costs and Benefits in the Patent System

The PTO has experienced a rapid grow in patenting in the last few years (said in 1997). The number of issued patents has increased dramatically

  • Less time to assess a patent
  • Incenentive to award

Four important issues:

  1. How much time and effort should be spent examining each patent?
  2. Is there a way to get the knowledge of competitors into the evaluation process?
  3. What is the proper division of labour between the public and private sectors?
  4. What are the optimal remedies and punishments for acquiring an invalid patent?

Function of a patent office:

  • A black box!
  • Could possibly auction off research projects (not taken seriously in the literature)
  • Goals:
    • Follow statute closely
    • Issuing patents only legislation deems desireable

Issuance of an invalid patent results in social costs that could have been avoided - by adequately searching the prior arts (all patents have social costs).

Indirect costs include:

  • Unnecessary licensing fees
  • Foregone research opportunities
  • Activities of rent seekers

The goal is not necessarily zero invalid patents - rather an acceptable error rate based on the costs of assessment. Note that some prior art takes time to materialize and so is not available to rule a patent invalid, even though it will later turn out to be.

Public vs. Private System

This section considers the benefits of having a private or public patent system. It concludes that a public-private hybrid model is probably the best.

Private System

The simplest and most used private system is central registration system and then private action through the courts.

Pros of simple registration system:

  • Similar to current copyright system
  • Patents with low intrinsic value or many close substitutes will not be defended
  • All evaluation done by the private sector - private sorting can be efficient

Cons of a simple registration system:

  • Low barrier to entry
  • Litigation is complex and risks at trial significant
  • Threat value of an invalid patent is high
  • Greater potential for hold up and rent seeking
  • Government should intervene to overcome externalities

Public System

Ideally marginal cost equals marginal benefit:

  • Need cost of each unit of validity info to the patent office
  • Need the estimated social cost of the patent

Sorting would be good - that is examine more problematic patents more, but this might be politically difficult (there may be a need to treat all patents as equal). But still rationing of a scarce resource (examination) to where it makes the most sense is a good idea. Absent this there are second best ideas:

  • Raise the standard of patentability and/or the filing fees
  • Make an estimate of average expenditure and act accordingly

Easiest way to raise the standard - tighten the non-obviousness clause of section 103 of the Patent Act. Other than that, all inventors would benefit from a rationally derived PTO budget - set the budget equal to the total social cost of patents, and spend equally. Some invalid patents would still survive as the cost to rule them invalid is beyond the cutoff. One alternative is to penalize the holders of bad patents. Tort law provides for damages to be paid by the tortfeasor. But this might deter application for even valid patents - because prior art can surface after application.

Arguements for a patent office:

  • The PTO has the lowest costs of assessment:
    • There are economies of scale
  • Public examination acts to certify the patents
  • Shift costs from small inventors to the PTO

Redesigning the PTO

The PTOs budget is raise by application fees. It is grown very large ($674m per year in 1997, based on about 230,000 applications), but a large portion is appropriated by Congress. The PTO spends this on:

  • Policy determination
  • International coordination
  • Patent Examination
    • Including interference proceedings
    • Reexaminations
    • Reissues

If it has its entire budget it could spend about $3000 per patent examination. But it is hard to determine in advance whether this is enough (or too much). Problems include:

  • Lack of information when patents are filed
  • Most technologies will not be economically viable
  • High cost in seperating out valuable innovations

However, as the average private value has risen, so has the average social cost to a patent. Ideally the PTO would know:

  • The future rent stream of a patent
  • The number and value of future innovation a patent will spawn
  • Competitive conditions in the market in which it will be sold

But this information is in the hands of competitors to the patent, not the PTO (if at all).

The PTO has problems with job design:

  • Too few senior examiners
  • Inadequate training

These could be solved by rewarding training and paying senior examiners competitive wages. Current renumeration is base + bonus, where the bonus is calculated on the number of dispositions (i.e. final allowances or rejections). However, final rejections are not final - post final amendments andthe like are permitted. Therefore the only way to gaurantee a bonus is through allowances. This gives perverse incentives.

An alternative bonus scheme might:

  • Institute a tracking system to determine the error rate for individual examiners
  • Have a private sector third party review a random sample.

Opposition Filings

Oppositions are currently allowed but seldom used (in the US - they are used more in Europe). One reason why they aren't used is they only overturn patents in a minority of cases. Also they cost $2520 and require detailed statements of prior art, specifically prior patents or publications. Other categories of prior art are allowed in Europe but not the US. A good opposition proceedure could extract vital information from the private sector, particularly competitors. It should have:

  • Better access by third parties - they should be allowed at PTO interviews
  • Possibility of appeals by third parties during reexaminations
  • Reform of 'staying' a district court litigation, which can be used by patentees for strategic delay