Patent Prosecution Malpractice

In the December 10, 2007 edition of the National Law Journal, an article discussed the rising rate of patent malpractice claims being filed against firms for prosecution mistakes (Sorry, no link - paid site). The article asserted that the risks of being sued are not keeping up with the rewards, noting that “run-of-the mill prosecutions can net law firms as little as $800 and even more specialized versions can net them only $5,000.” With the number of malpractice claims going up, more firms were likely to end their prosecution practice.

What is the cause of all of this malpractice going on? Well, the article opines, it may be that in 2005 we had 390,733 new utility applications filed, while in 1996, we had 212,377 applications, an 84% increase in work load. You would certainly expect that patent prosecutors under extreme price pressure would file an increasing number of patent applications that give rise to more malpractice claims. If we reduce the number of applications being filed by increasing the fees, the pressure to file will be reduced and clients will be less likely to force their patent counsel to accept such small margins. Fewer patent applications will mean fewer deadlines missed resulting in less malpractice.

The New Rules - Patent Demand By Proxy

The USPTO has recently proposed significant modifications to the prosecution and appeal rules under which patents are granted (here for example). In my estimation (and with some confirmation by the GC of the office - see here), the rules are an attempt by the USPTO to reduce the number of patent applications that it has to examine, or Patent Demand By Proxy.

Of course, the proposed rule-making has created a backlash in the patent holder and prosecution communities because of the damage that the new rules would do to patents and the patent process. The secondary effect of the proposed rules is to drive up the expense of filing and prosecuting patent applications through increased patent applications filed and increased legal bills for complying with the rules. The first set of proposed rules were postponed by a lawsuit filed by GSK, and I would expect a similar result with the second and third sets. However, the USPTO could avoid all of this unpleasantness by simply increasing the filing fees it charges for large entities to a point where the supply of patent applications is reduced to a number that feasible to examine adequately.

What about Trolls?

How would raising the filing fee impact patent trolls? If you are not familiar with patent trolls, check out this excellent blog Troll Tracker. Patent trolls are companies that are set up with the sole purpose of suing other companies for patent infringement on patents that tend to be acquired from individual inventors and companies that have gone out of business.

In a recent post, Troll Tracker reviewed patent lawsuits filed against the Fortune 100 based upon the industry of the defendants in the lawsuits. The most interesting distinction was between the Pharmaceutical Industry and the High Tech Industry. Pharmaceutical had 28 total lawsuits, with 24 coming from competitors and 4 coming from Universities (etc.) but none from patent trolls. High Tech, on the other hand, had 192 lawsuits, with 35 from competitors, 24 from Universities (etc.) and 132 from patent trolls.

What is the distinction between the Pharmaceutical and High Tech industries around patents? Well, first the Pharmaceutical Industry depends on patents to survive because they tend to have a single innovation (the chemical compound used to treat disease x) the price of making a pill is next to nothing and anyone can follow a formula to make a chemical compound. High Tech companies tend to have differentiated products composed of lots of improvements that are very complicated to put together. Second, Pharmaceutical companies spend $800,000,000 to get one new product to market so the consequence of having a weak patent is huge. High Tech also spends a lot of money on new products, but they tend to grow incrementally from one product version to the next.

Pharmaceutical companies tend to have a relatively few number of extremely valuable patents that, were a company go out of business, would quickly be snapped up by a competitor, investor or other interested party, who would produce the compound. High tech companies have lots of patents on small features and if the business should cease, no one really knows the value of the patents. These patents fall through the cracks and end up with the trolls.

If the patent filing fees were raised, then High Tech companies would only file patents on their most important innovations and fewer patents would be more valuable. Thus, they would be less likely to fall through the cracks, and the patent troll market would dry up.

So, What Happens to the Patent Lawyers?

This is an interesting question. When we reduce the number of patent applications by three quarters, where do all the patent lawyers go? The answer for the best and the brightest is nowhere. There will always be a very competitive market for high-end patent drafting and litigation services. Perhaps even more so, with fewer patent applications that are far more valuable than the current crop. I think that raising the application fee to $50,000 will allow the legal bills for obtaining those patents to rise as well. I believe that this will be the end of the $5,000 flat-rate patent application drafting services.

What happens to all of those folks who are currently writing the thousands of patent applications for a single company? Well, they are already highly-trained engineers and lawyers....They should return to the workforce and do something that is productive. Like invent something!!! Think about it, we have thousands of highly-trained professionals working on patents that don't include any innovation. And then we waste millions fighting over patents that should not have been issued in the first place. Increase the fees, reduce the number of patents filed and we will see an end (or at least a massive reduction) in the waste.

Two Sides To Patent Quality

Most people think that poor patent quality refers only to patents that are improvidently granted. For example, the patent examiner did not adequately search for the prior art, or failed to appreciate a reference that was before them, or did not apply a reference correctly, or was outgunned by the applicant, or ... Improvidently granted patents are the more important type of poor patent quality, however, I believe that a second and important type of of poor patent quality is patents that should be immediately granted, but are delayed or damaged by inappropriate office actions (I refer to this second type of poor patent quality as Damaged Patents).

Damaged Patents are the product of the USPTO being overwhelmed by the nearly half-million patent applications it is tasked with reviewing each year. How does the Patent Office deal with examining this number of applications? The office has two basic choices: (1) let everything through with minimal examination (characterized by the era when examiners were told it was their job to get patents for their "customers") or (2) let nothing through (unless the examiner is "embarrassed" into issuing the patent - it seems the office is in this mode now). This second approach by the USPTO exacerbates an already huge backlog, creates a much longer patent pendency and Damages patents that should immediately be issued.

Let's assume that an inventor has created a new way to produce hydrogen from water by using RF energy to break down the molecular bonds of the water and a patent application is filed. First, of course, the application is not even substantively reviewed by an examiner for at least two years. Next the examiner, who has just graduated from the 8-month training course, misunderstands the invention and does a search for electrolysis and finds a huge amount of "prior art" to cite in the first office action (3 months later). The applicant requests an interview wherein the invention is made clear to the examiner, who signals that some dependent claims might be allowable. The applicant files amends the main claims to include the dependent limitations (3 months later). The examiner does a new search and finds a reference on a RF generator and cites that plus the electrolysis reference as 103 and goes final (3 months later). The applicant files a RCE and argues over the reference and the examiner issues the patent -- now Damaged (6 months later).

The applicant finally has a Damaged patent that should have been allowed by the USPTO on the first office action based on the first search that the examiner did. Of course, the applicant may file a continuation case to attempt to get the proper scope of claims, but that will delay the process for years to come.

Let's look at a similar case under a scenario where the number of patent applications has been reduced because of an increase of the filing fee. Obviously, in this hypothetical, the applicant would choose to file the application on a technology this important. The examiner, with 5 years of experience in the current Art Unit, having received the application 6 months after filing reviews the application carefully for technical issues -- recognizes the importance of the application and does a careful prior art search. Finding nothing relevant to the claims of the application, the examiner allows the case in three months. The applicant has the patent in under one year.

Again, the fault here is not entirely, nor even primarily the USPTO's, rather the fault is shared with Congress and large corporations filing thousands of unimportant patent applications because the fees for filing allow this abuse of the patent system.

A Two Filter System

When most people consider patent quality, they focus on examination “filter” that the USPTO provides. This misses what I like to call the “First Filter”, that is the work that the applicant does (or should have done) prior to filing the patent application. The first filter should begin when an invention is made – the inventor is responsible to determine whether the invention is novel. And, of course, the inventor is in the best possible position to determine exactly that because presumably they are one skilled in the art and they have some knowledge of the prior art. The first filter continues through the patent review committee where senior R&D folks review the proposed submission to determine whether it merits a patent application. Then, it’s on to the lawyers to do a patent search and a final determination of whether to file.

Of course, this first filter will only perform its function if there are appropriate economic incentives in place to motivate the individuals to create a quality patent application. If the cost for filing a marginal patent application is so low as to be irrelevant to the filing decision, then the first filter is removed for practical purposes. The inventor is motivated to file every patent application by his bonus structure; the senior R&D folks don’t have a cost to balance the benefit of filing the application; the patent attorneys are motivated to file another case because that’s how they make money. The result is the first, and I believe the most important filter leaks like a sieve, or it seems in recent years, it is entirely missing.

A Trip to “Homestead”

Last week I made the 4.5 hour sojourn to Hot Springs, VA to speak at and attend the IP Law Summit. I was surprised by the number of IP attorneys who turned up at the event and am still a bit puzzled by the business model. Most of the sessions were well attended and well done. They covered topics like Patent Portfolio Management, How to Create Value from your Patents, Examining the New PTO Rules and Reviewing the Pending Patent Reform Bills. Of course, there was hearty debate about the final two topics.

John Miller from Rockwell Automation and I did a joint presentation about how to “deal with” Patent Trolls (namely, don’t feed them). Surprisingly, I got the largest reaction of the event from the crowd when I proposed raise the patent filing fee to $50,000. There was literally a “gasp” followed by several vocal protests that finally resolved to some nervous laughter. In any event, there were several folks lined up outside the room to discuss the idea with me and more than a few that thought it was a novel approach that should be explored. The GC of the USPTO was asked about raising the filing fee during his presentation later in the conference and he responded that they floated that idea several years back and it was shot down by practicioners. So, he continued the new rules are the approach the USPTO has decided to take, and you will have to adjust.

Hire, Train, Repeat . . . Again

The Government Accountability Office (GAO) recently published a study where the title says it all: “Hiring Efforts Are Not Sufficient to Reduce the Patent Application Backlog.” The full report is here. First, the report found that over the past 5 years, the PTO has based its hiring upon its funding rather than on the backlog and expected workload. Second, 70% of attrition was from examiners who have less than 5 years of experience. Finally, the report notes that 67% of examiners identified production goals as the primary reason that they consider leaving.

The most illuminating paragraph of the report is on page 12:

“Even with its increased hiring estimates of 1,200 patent examiners each year for the next 5 years, USPTO’s patent application backlog will continue to grow, and is expected to increase to over 1.3 million at the end of fiscal year 2011. According to USPTO estimates, even if the agency were able to hire 2,000 patent examiners per year in fiscal year 2007 and each of the next 5 years, the backlog would continue to increase by about 260,000 applications to 953,643 at the end of fiscal year 2011. The agency has acknowledged that it cannot hire its way out of the backlog despite its recent increases in hiring, and is not focused on slowing the growth of the backlog instead of reducing it.”

So, the best we can expect by 2011 is a backlog of 953,643 applications. I don’t believe that the PTO can even do that well. How does the PTO slow the growth of the backlog? New continuation rules, limitation on claim numbers, making the applicant do the examiner’s job etc. Basically, the PTO is going to limit the growth of the backlog by limiting the value of the patents that work through the system.

PTO Hiring

During the last 5 years the PTO has based its hiring levels on funding levels and the capacity to train and supervise new examiners. The GAO report looks at this practice negatively and insinuates that the PTO should set hiring levels based upon the backlog and expected workload. So, the GAO would like the PTO to ignore its funding in setting hiring goals? Then, the PTO should exceed the capacity of its training and supervisory (note the 12:1 ratio of examiners to SPE’s on page 10) functions in the effort to deal with the burgeoning application flow. Apparently, the PTO should just hire 5,000 new examiners a year and somehow the backlog will be addressed. This makes no sense. The flow of applications is the problem.

One interesting note here: on page 14, the GAO states that it takes 4-6 years of on-the-job experience to become proficient as an examiner and 70% of those who leave have 5 years of experience or less…not surprising, but you do the math.

PTO Attrition

It can not be a shock to anyone that PTO examiners have a high attrition rate. Recent engineering grads doing a few years at the PTO before going on to law school and high-paying patent jobs. Not so, says the GAO. In its survey of PTO examiners, the GAO found that 67% of examiners believe that production goals are the primary reason they would consider leaving the PTO. So, examiners want more time to do their jobs – then they will stay put. The PTO should believe them. How do they get more time? Reduce the number of patent applications coming through the system.

It was interesting to me that 89% of examiners believe that their work is important. That is, in my opinion, an indicator that they like their jobs and would be willing to stay. Fewer cases per examiner would be the key, and more money in DC couldn’t hurt.

Patent Fees Rise!!!!

Well, its only by $30 per application, but its a start....Let's see blogging for under a month and progress at the PTO already. This must be some type of record.

PTO Fee Schedule

Patent Demand Impact

The theory behind this blog is simple: when fees for patent applications are raised, companies will respond by filing fewer, higher-quality patent applications. Of course, that would mean that we should be able to show a negative relationship between filing fees and patent applications filed (stay tuned for the actual data). However, filing fees have been and remain so low (currently $1,000) that they are below the demand curve so the impact of raising the fees will only be felt if the fees are increased on a massive scale, say to $50,000 per application.

Consider this, the AIPLA, in its annual economic survey states that the median charges for US Utility Patents of minimal complexity is $6,500, relative complexity is $8,600 to $12,000 depending on the field of invention (should a minimally complex case be filed at all?). So, the last thing a company considers in filing a patent application is the cost of the filing fees. They should be much more interested in squeezing the outside counsel bills --- “Don’t do a prior art search, that will cost more than the filing fee.” This is a small price to pay to file a case that may someday be worth millions in reduced license fees or litigation expense, no wonder the default position is to file every application possible.

What might happen to the decision process if the fee was raised to $50,000:

Prior art may be found and considered to help define the scope of the invention;

In-house IP counsel may be called upon to make tough choices between filing cases or not (the most important cases would get filed);

Attorneys fees will go from being the largest to the smallest portion of the cost of the application (this may allow the flat fees to rise);

And, most importantly, fewer cases will be filed permitting the USPTO to actually do its job (more about this later).

By now creating a limited resource (the number of cases to be filed), market forces would naturally begin to separate wheat from chaff and improve the quality of patent applications well before the USPTO gets involved in the process.

Why Patent Demand?

I have spent my entire legal career working within the patent system. My first experience as a patent litigator taught me the incredible power that a single patent can impose on the market, and the huge expense of patent litigation. Now, as in-house IP counsel at a large software company, I see the enormous risk that all companies face from patent litigation just by being in business. Large corporations are filing huge numbers of patent applications so that, if they are sued by a competitor (or another company that actually makes a product), they will likely have a patent within their portfolio to file their own lawsuit (sort of like mutually assured destruction).

Because patent infringement lawsuits cost so much to defend ($4 million), just the filing or threat of filing a lawsuit is a very powerful weapon. I have heard IBM patent professionals publicly declare that the size and scope of IBM’s patent portfolio allows them to enter any business it chooses without fear of patent infringement suits. So, when IBM decides that it would like to enter a market, the player who created the market in the first place can’t use its patent portfolio against IBM without fear of a retaliatory strike from IBM in an area that may be completely unrelated to the technology that created the market. Thus, a real inventor’s market advantage from its patent portfolio is limited by the huge patent portfolios developed by huge companies.

IBM files more than ten patent applications each and every day of the year…..

Supply and Demand Reviewed

I like to examine markets using economic principles. In a market for commodities, the Law of Demand generally states that if all other things are equal, the price of a good has an inverse relationship with the quantity of the goods demanded. In other words, the higher the price, the lower the quantity of the goods demanded. This relationship is depicted graphically by a Demand Curve, with price on the y-axis and the quantity demanded on the x-axis as follows:

Of course, there is an equally powerful Law of Supply in most markets, which reflects that the supply of a good is directly related to its price graphically reflected in a Supply Curve. The interaction between the Demand Curve and the Supply Curve in a perfectly competitive market yields a price and quantity demanded that will efficiently allocate resources.

The Patent Application Market

What does this have to do with the patent system? Well, consider that a patent application is a “good” within its market system. The Law of Demand would require that as the price of a patent application increases, the number of patent applications demanded by the market would decrease. The Law of Supply would predict that as the price of a patent application rises, so does the quantity supplied. However, in the market for patent applications, the US Government is, in effect, a monopolist that controls both the price of patent applications and the quantity of patent applications supplied.

In a normal market controlled by a monopolist, economic theory would predict that the monopolist would set a price and supply goods that would maximize its profit. In the market for patent applications, the US Government follows a policy that the price of a patent application should not keep an inventor from accessing the patent system. Large corporations have determined that it is in their interest to file as many patent applications as possible because of the potential value of a single patent application. The interaction of the government policy and the corporate determination have lead to the 400,000+ annual patent applications and the chaos created within the USPTO simply trying to deal with the incredible volume of applications – to say nothing of providing a quality examination process.

Hire, Train, Repeat, Repeat.....

In its 2006 Annual Report (Report), the USPTO congratulated itself for exceeding its hiring target by over 200 new examiners for a total of 1,218 new examiners. That is, more than one third of the total examining corps are new examiners, which would lead to a three year average examination career. The Report predicts that the USPTO will hire 1,200 more new examiners in 2007…. Of course, for the first eight months of a new examiner’s career they are training in classroom groups prior to being new examiners actually examining cases for two years prior to moving on.

So, the United States has one of its most important resources being evaluated by recent college graduates with three years of experience. I believe that the examining corp is trying its best to do its job, but the numbers are overwhelming. How do we fix this problem? 1. reduce the number of patent applications and 2. increase the pay for the examination corp to improve retention of talent. Raising the patent fees to $50,000 would cut the number of patent applications filed by 2/3 and increase the revenue to the patent office from $1.6 billion currently to $5 billion ($50,000 * 100,000 (est) applications per year).

Two Interview Stories

One of my former and much esteemed colleagues, Ken Adamo of Jones Day, told me of a trip to the USPTO for an examiner interview. According to Ken, the examiner was a true expert of the subject matter, to the extent that when Ken would make an argument about a novel aspect of the invention, the examiner would leave the room for a few minutes only to return with an applicable reference deflating the argument. I can’t remember whether Ken was ultimately successful (knowing him, he probably was), but the point is that this is the type of individual that we all would like to have protecting the free market from impingement by improvidently granted patents.

Having been involved in numerous examiner interviews over the years, I was recently struck by a comment from an inventor about an interview: “I expected to discuss my case with someone like Thomas Jefferson . . . not a recent college graduate, who did not even read my application entirely.” Of course, I made some excuse regarding the enormous time pressure the examiners are under (which is true). The unyielding pressure of 400,000 plus patent applications has made Ken’s examiner at the USPTO extinct.

Opening Salvo

The US patent system was intended to help promote the useful arts by providing inventors of new technologies the time-limited right to exclude others from the market in exchange for disclosure of their inventions. Thus, it was thought, inventors would be fairly compensated for their contributions and society would benefit because the early disclosure of invention would encourage more rapid evolution of technology.

The current US patent system is in crisis. In my opinion, this crisis is the result of the sheer volume of patent applications that companies file to protect themselves from the threat of patent litigation and to extract as much revenue from the resulting patents as possible. As a result of the volume of patent filings, the USPTO has been overwhelmed and seems to vacillate between the policy of granting patents quickly without the necessary rigorous review and the policy of creating as many obstacles to obtaining patents as possible (the current trend). The USPTO is not primarily responsible for the problems in the US patent system, but rather the volume of patent applications has created an environment wherein no human process can adequately decide whether any single patent application should be entitled to a patent. In other words, the USPTO has been given an impossible task with no rational solution.

Here is the data:

Wow! Over the past 5 years, 1.8 million patent applications were filed in the US. It is no wonder that the USPTO has a 1 million case backlog, and it takes an Examiner nearly 3 years on average to read and reply to one of my cases in the software art. But, even scarier, is to look at a plot of the data over time:

In your mind, just visualize the extension of that line through 2010 (don’t worry, I’ll provide a model that will give accurate forecasts later). Half a million patent applications per year by 2010 is a great possibility. The patent system is collapsing under this weight of patent applications and granted patents and none of the proposed “patent reforms” wending their way through Congress address this fundamental problem.

So, how can we possibly fix this problem? Each year, we hear of some new patent reform, but the reform tends to focus on treating a symptom of the problem (litigation) rather than the actual problem (application volume). Similarly, courts have begun to provide limits on elements of the litigation process. The main problem that I have with these solutions is that they de-value all patents regardless of whether the patent should have been granted by the USPTO or not.

I like to solve problems using market-based solutions if possible. There are too many patent applications, in other words we have a classic over-supply problem. How do you fix an over-supply of goods in a government-controlled market? Raise the price for large entities. Currently, filing a patent application costs $1000 for a large entity, peanuts for any large corporation when you consider the potential benefit of a single patent. Of course, this ignores the attorney fees to prepare the application. My hypothesis is that the $1000 application fee does not provide any disincentive for filing marginal patent applications. A corporation will always file the marginal patent application and the application fee is not even a consideration in the decision process.

PROPOSED SOLUTION

I would raise patent application fees to $50,000 for large entities (no change for small entities, individuals and universities for now). I would make this fee apply to all applications currently being examined by the USPTO. My guess (to be validated later) is that this will dispose of the current backlog of applications and leave the USPTO with about 100,000 applications to examine per year.

Patent Demand is a forum to discuss the proposed solution, its benefits and drawbacks in an effort to improve the US patent system. I hope that you will contribute to the discussion.