STATEMENT OF

RONALD J. STERN, PRESIDENT

PATENT OFFICE PROFESSIONAL ASSOCIATION
BEFORE THE
SUBCOMMITTEE ON COURTS AND INTELLECTUAL PROPERTY

COMMITTEE ON THE JUDICIARY

UNITED STATES HOUSE OF REPRESENTATIVES

JUNE 7, 2001

ON THE SUBJECT OF

OVERSIGHT HEARING ON

THE UNITED STATES PATENT AND TRADEMARK OFFICE

Mr. Chairman and Members of the Subcommittee:

Thank you for the opportunity to provide the views of the Patent Office

Professional Association. Our organization is the exclusive bargaining agent for the approximately 3300 patent professionals at the U.S. Patent and Trademark Office. The vast majority of the employees we represent are engineers, scientists, and lawyers who work as patent examiners and patent classifiers.

There are three bargaining units at the USPTO. POPA represents more than one and a half times the number of employees represented by the other two bargaining units put together. The percentage of organized employees that we represent in the USPTO is continuously growing.

As everyone is aware, patents play a vital role in our high technology economy. They serve as a incentive for companies to innovate, and as an engine for efficiency in all spheres of economic activity. While many people would like to "work smarter, not harder" to achieve their goals, patents provide the know-how to enable people to actually "work smarter". Patents are the intellectual gemstones of commerce, recognized by our founding fathers.

The biggest challenge that the USPTO faces right now is that it needs more examiners. It needs them for two reasons. First, there has been a huge increase in workload. Second, more time needs to be allocated per case so that we can do a better quality job, and especially provide a better search of the prior art. We believe that the USPTO needs to put the hiring of examiners at the top of its list of priorities and leave it there for about five years.

The workload

To understand the dramatic increase in the USPTO's workload, consider that in this year FY 2001 the agency expects approximately 330,000 applications to be filed, a 21% increase over the 278,268 filed in Fiscal Year 1999. The average amount of time allocated for examination of an application is roughly 20 hours. If the 330,000 applications the agency expects to have filed this year are multiplied by 20, you will get the total number of hours necessary for examination. If you then divide the total by the number of hours the average examiner spends examining per year, usually 1664 hours, you can see that the agency will need about 4000 examiners just to handle this year's filings. Since it has only about 3000 examiners, that means that the agency is short by approximately 1000 examiners, if it just wants to stay current.

Of course, the agency also has a backlog. At the end of FY 2000, there were approximately 408,000 patent applications either awaiting action or in the process of being examined. We cringe at the thought of the FY 2001 backlog. All this means is that we are behind on our workload, and losing ground. We are losing ground fast.

We wish to state this unequivocally: more computers will not solve these problems. There simply aren't any computers on this planet that can read about inventions, understand them, and make meaningful comparisons with prior inventions to determine novelty and the obviousness of any differences. You need a person to actually read these 330,000 patent applications page by page, to research the prior inventions, to determine whether the invention is clearly expressed in the claims and has utility, and whether the invention is novel and non-obvious. Computers cannot do this.

What this means is that no matter how much money is spent on automation, the only way in which the throughput of the agency can be increased is by hiring more examiners.

This is an important point with respect to Congressional oversight. Last year Fiscal Year 2000, when record numbers of applications were being filed, the agency allowed examiner staffing levels to actually decline by about 2%.⁽¹⁾ This decline was due to the fact that the budget for the USPTO passed by the House Appropriations Committee was so small that the agency stopped hiring because it was afraid that it might otherwise run short of money. The agency didn't receive its FY 2001 budget until December, so it wasn't clear for many months how much money the agency would have to spend on new hires. Since then, the agency has been faced with a short, temporary hiring freeze as part of the transition to a new administration. Thus, even though the FY 2001 appropriation is approximately 17.5% higher than the funds available for obligation in FY 2000, hiring has so far only covered losses due to attrition.

Such disruptions in hiring are almost impossible to overcome because there is a limit on how many employees can be hired and trained in any one year. Hiring close to that limit would be necessary every year to just meet workload increases. Consequently, losing almost an entire year is serious. The agency will be feeling the effects of the disruption in hiring for years to come. If Congress provides for the agency to keep all of its fees and does not subject it to the vagaries of the appropriation process, then the agency will be able to maintain a uniform hiring pattern that will allow for proper training and an adequate increase in staff.

More time for improving quality

As I have testified in years past, there is a critical need for improvement in the quality of our searches and for more time to handle the increased complexity of applications. The agency's Quality Review statistics continue to show a need for improvement in our ability to find the relevant prior art. More and more reporters and writers are criticizing the agency for issuing overbroad or otherwise potentially invalid patents. One Internet news provider continues to make a business out of criticizing the quality of the USPTO's work products, primarily because he believes that the most relevant prior art is not identified during examination. In addition, USPTO examiners report much more frequently than in years past that European searches identify references that are relevant but not found in our own searches.

USPTO examiners, who get about half the time allocated per application compared to EPO examiners, are convinced that there is a dire need to spend more time per application. Individuals outside the agency are also beginning to recognize the connection between the time we spend on an application and the quality of examination. As stated in a letter to the Editor of The Wall Street Journal published on March 21 of this year, "the probable cause of this situation [the issuance of invalid patents] is that patent examiners are not given enough time or resources to conduct adequate searches of the existing technology."

The time factors we currently work under were last revised in 1976. Since that date, patent applications have become significantly more complex on average, searches have become significantly longer and more time consuming, and the number of claims per application has increased. In spite of this relentless inflation in the need for more time, the agency continues to pressure employees to work faster. Quotas, which have not changed for years in spite of this growing complexity, are such that examiners frequently volunteer time to meet them. Supervisors still analogize the work we do to an assembly line for producing good middle of the road products and admit that they do not expect top of the line work in the time allotted per case. As some have said in the past, we produce Chevys, not Cadillacs.

The agency recognizes that when there is a need for a better search, it has to give more time. An example is what has been done in the business methods area. The agency became acutely aware of serious problems in this area through scathing articles in the press about alleged poor quality patents. This area is not distinguished on the basis of technological complexity, since it is no more complex than many other areas. What is different compared to other areas is that business methods do not have a classified set of patents that comprehensively set forth all the prior advancements in the field. As a consequence, searching must be done in the non-patent literature mostly found in databases located outside the agency. This process is extremely time consuming. Because the agency has recognized this, it provides on the order of at least 50% more time per case.

The expanded search program in the business methods area has been successful and has dramatically lowered allowance rates due to the additional time provided. While some have credited the lower allowance rates to a "second pair of eyes" procedure, we understand the additional review has only changed decisions in about 6% of the cases. We do not consider this aspect of the program to be cost effective.

Just as it is necessary in the business methods area to spend significantly more time to increase the scope of the search beyond classified documents, it will also take more time in other areas to increase searching beyond the classified art.

In order for us to maintain pendency levels and increase quality, it is essential that the agency hire more examiners. To hire more examiners, it needs to keep its fees. The need for more examiners will not be solved with automation or with contracting out. There is no automation technique on the horizon that will save examination time. The automation that has been provided until now, requires extra time, if anything, to search the additional sources of prior art technology to which it provides access. European patent examiners, who have approximately twice the time to search as examiners in the United States, use automation extensively for searching. To the best of our knowledge, their searching has not gone faster because of it.

As an aside, we can report that the European examiners have the same problems as the examiners in this country and are currently on strike for more pay and for maintaining the time in which to do their search and examination.

PTO management recognizes that our quality needs improvement, but, due to funding diversions, has prioritized computerization and other projects over increasing the time allocated to examination. As a consequence, USPTO management has committed to us that they would join us in trying to get funding for more examination time per case so as to increase the quality of issued patents.

Overtime as a substitute for additional hires

A cost effective alternative to new hires is the encouragement of overtime. As we have suggested in past years, overtime can be encouraged by changing the existing statutory pay caps.

By statute the maximum hourly overtime pay is time and a half of grade 10, step 1. What this means is that our senior people, who produce at a rate approximately twice that of the junior people, get paid only about 60% of their regular pay when they are working overtime. Most folks are just not willing to work for a mere 60% of their normal pay.

Another statute limits total pay including overtime for a biweek to the amount payable to a GS-15, step 10 employee with no overtime. This cap denies, or limits, our most experienced primary examiners the opportunity to work paid overtime. Since these people are the most productive examiners in the agency, providing them an incentive to put in extra time will have a valuable payoff.

We recommend that you provide a legislative fix for the USPTO. H.R. 1770, which was introduced in the 106th Congress, and had the blessing of the Office of Personnel Management, had the right technical language to accomplish this. H.R. 1770 would not only have increased the hourly cap, but also would have increased the biweekly cap. The increases are limited to amounts that are already available to a few other Federal employees such as air traffic controllers and law enforcement officials. Although the bill did not pass, we think it should be implemented NOW for the USPTO because we are a production oriented organization in which overtime is directly linked to the examination of extra applications.

Lifting overtime caps is a win-win situation for both patent applicants - they get experienced examiners at a bargain rate - and for employees. Even though examiners will be paid at straight time rates, the agency will save money because there will be no payments for additional employee benefits (retirement, health care and life insurance), no recruitment expenses, no payments for additional office space, no payments for additional computer workstations, etc.

Other needed quality improvements

In addition, the agency needs to keep improving its search systems and this also requires an investment. The agency desperately needs to increase the amount of reclassification in the U.S. classification system, needs foreign art to be classified into the U.S. system and needs to expand its classified non-patent literature collections. Throughout much of the '90s, the agency took the position that automation would entirely eliminate the need for reclassification. As a result, the agency has not budgeted for reclassification and has permitted the size of the classes and subclasses to grow. Now that we have areas where the subclasses are too large to properly search, we have found the agency's position that reclassification is unnecessary in the electronic age to be far from the truth. Properly classified classes and subclasses are critical to doing a quality job in an automated search environment.

Without reclassification, subclasses have grown to include thousands of documents, making it difficult or impossible to do a classified search, either in paper or electronically, in the time allotted. To save search time, examiners are forced to rely more heavily on text or key word searching. Text searching, i.e., searching the patents using the key words or phrases, is risky because concepts can be described many different ways using many different sets of words. The examiner will try to use terms that encompass all or most possible ways of representing a concept, but when doing so, will frequently end up with a result set that is either so large that it can't be reviewed in the time allotted, or one that is so small, that many of the relevant documents have been missed completely. The ability to combine a search for certain terms with the subclasses where the concept being searched should be classified permits the examiner to avoid using overly limiting search terms and aides the examiner by eliminating many irrelevant documents. We need to fund reclassification of the U.S. classification system.

For three years in a row, we have reported the need to reclassify foreign art into the U.S. system so that the foreign documents may be located through a classified search. There is still no system in place to do this. The agency still says that it is in the process of developing a system to do this.

One idea that has been discussed in the Patents Public Advisory Committee (P-PAC), which we strongly support and recommend be adopted, is to allow examiners to classify documents they receive from applicants as parts of Invention Disclosure Statements into the U.S. system for inclusion in our search files. What is relevant for one application is, more likely than not, also relevant to others. This is a good source of non-patent literature which can be implemented now in paper form easily and inexpensively. Eventually, references provided in IDS's could be captured electronically, although this might be more costly. While examiners should make the determination if a document should be retained and added to the search files, clerical support is required to copy and actually place a document into the search files. Alternatively, applicants could be asked to submit duplicate copies of documents. We are disappointed that the agency hasn't provided the resources to do this yet.

Accountability

The USPTO had a huge increase in its budget from last year to this year, i.e. FY 2000 to FY 2001. Based upon what is available for obligation, as reported in agency's FY 2002 Corporate Plan (budget), the 2001 budget is approximately 17.5% higher than the $895 million obligated in FY 2000. We believe that it is appropriate, as part of your oversight, to look carefully at what we are getting for that money. Unfortunately, up to this point, the increase has not enabled us to meet our needed increased staffing levels. If the money is not going to increased staff, what is the money being spent on? We believe that much of the increase is going into automation.

Although the automation budget grows significantly every year, we are not seeing huge improvements in this area. As a general matter, automating the internal handling of patent application files is probably going to be a good thing. But, care needs to be taken so that it is not done in a manner that is not cost effective. It is essential that the agency be able to publish a cost benefit analysis that shows precisely what the costs of automation are to the agency and what savings are attributable to each specific automation project.

Currently, the salary cost, including the most recent increase, of all the examiners and classifiers that POPA represents in the Technology Centers is about $220 million. On the other hand, the budget of the Chief Information Officer is $239 million for the current year. The trade-offs between production workers and automation programs need to be made public so that there can be an informed discussion of what is most valuable to the country.

The agency should be held accountable achieving results. When we institute new programs, a cost benefit analysis should be performed and published so that we understand whether a program is worthwhile. In order to really understand time available for examination, the USPTO needs to disclose the total time spent on overhead activities and the time that is actually spent on examination, information that is already very carefully tracked. Conspicuously absent from the agency's annual report is the number of employees on board by category, and how those employees account for their time. We recommend that the agency publish the statistics on examining time, on all examining related activities, and on overhead and supervisory activities so that the time allotted to examining can be better understood.

Labor Management Relations

We have lots of good news to report in the area of labor relations and employee satisfaction. I can assure you that employee morale has increased enormously. Working conditions have been significantly enhanced. Employees are delighted that a substantial pay raise has been put into effect. In addition, employees have been given significant additional flexibility in when they are allowed to do their work. Employees may now choose their work hours in a band from 5:30 in the morning to 10:00 in the evening and may choose work on one weekend day per week. Compensatory time programs may be used to shift work from one pay period to another. This flexibility makes the USPTO a particularly family-friendly environment. A casual dress policy provides a more relaxed atmosphere.

Programs to enhance the automation training for employees have been implemented. More employees are now also able to visit the laboratories and factories of applicants to get first hand knowledge of new machines and processes, and to meet inventors. The agency also provides tuition payments for both law and technical courses.

I would also like to thank the senior management at the USPTO, including Todd Dickinson, Nick Godici, Esther Kepplinger, and Clarence Crawford for implementing programs to enhance employee satisfaction and improve labor relations. The policies put into place in the past year will go a long way to enhance the recruitment of new employees and induce current employees to stay with the agency in spite of the very attractive salary packages that have been offered by many in the private sector.

The agency has also agreed to pilot a work at home program one day per week for our senior patent corps employees. Employees have been requesting inclusion in such a program for years and it will be a big selling point for recruitment and retention.

I would also like to thank the Subcommittee for placing employee organizations on the Public Advisory Committees. This gives us a sense that we are a part of the process of planning for the future, an opportunity to hear the needs of different segments of the patent community and a real opportunity to understand the origin of policies that affect us very directly.

Summary Statement of Ronald J. Stern, President

Patent Office Professional Association

June 7, 2001

Regarding Oversight of the

United States Patent and Trademark Office

Based on an average examination time of roughly twenty hours, the agency is short by approximately 1000 examiners just to stay current with the 330,000 applications that are expected to be filed this year.

We wish to state this unequivocally: more computers will not solve these problems. There simply aren't any computers that can read about inventions, understand them, and make meaningful comparisons with prior inventions to determine novelty and the obviousness of any differences. No matter how much money is spent on automation, the only way to increase the throughput of the agency is by hiring more examiners.

There is a critical need for improvement in the quality of our searches and for time to handle the increased complexity of applications. The time factors we currently work under were last revised in 1976. Since that date, patent applications have become significantly more complex on average, searches have become significantly longer, and the number of claims per application has increased. In spite of this relentless inflation in the need for more time, the agency continues to pressure employees to work faster.

As an inexpensive way of increasing available examining time, we recommend, encouraging overtime, by increasing through legislative action, the overtime pay rate to a level equal to the straight time rate.

We need more reclassification efforts independent of whether we automate our search system or maintain our paper files. Foreign patents and non-patent literature, especially that submitted in an IDS, need to be classified into the U.S. Classification system and added to our files.

Currently, the salary cost, including the most recent increase, of all the examiners and classifiers that POPA represents in the Technology Centers is about $220 million. On the other hand, the budget of the Chief Information Officer is $239 million for the current year. The trade-offs between production workers and automation programs needs to be made public so that there can be an informed discussion of what is most valuable to the country.

Good news for labor relations. POPA thanks the senior management at the USPTO including Todd Dickinson, Nicholas Godici, Esther Kepplinger, and Clarence Crawford for implementing programs which enhance employee satisfaction and improve labor relations. Flexibility in choosing work hours and pay increases will go a long way to enhance recruitment of employees and induce current employees to stay with the agency despite attractive salary packages that have been offered by the private sector. Thanks also go to the Subcommittee for placing employee representatives on the Public Advisory Committees for the USPTO.

1. In FY 2000 there was a net loss of 62 examiners. United States Patent and Trademark Office, Performance and Accountability Report, Fiscal Year 2000, page 27.