Deadline Effects on Productivity: Evidence from the Patent Office

04/01/2024
Featured in print Digest
This figure is a histogram titled, First-Round Review Completed by GS-14 Patent Examiners. The y-axis represents the frequency of first-round review completions, ranging from 0.000 to approximately 0.025. The x-axis displays the date of first-round review completion, with labels for the start of each quarter from October 1, 2010, to October 1, 2011. The line graph shows the frequency of first-round review completions by GS-14 patent examiners over time, from October 1, 2010, to approximately December 31, 2011. The graph has several distinct spikes, labeled “Biweekly deadline spikes”, indicating deadline-driven surges in review completions. In the third quarter of 2011, a new docket management system was introduced, expanding the biweekly quota system with additional incentives linked to average review-completion times. Following this change, while the pattern of biweekly spikes aligned with the deadlines is still prominent, the spikes are not as big as they were prior to the introduction of the new system. The note on the figure reads, The new docket management system expanded the biweekly quota system by adding new incentives tied to average review-completion times. The source line reads, Source: Researchersʼ calculations using the USPTO’s Patent Application Information Retrieval database.

Some organizations reward employees who meet production quotas by pre-set deadlines. Such structures allow employees to manage their time between deadlines, but may also result in procrastination or rushed efforts near the deadline. In Deadlines versus Continuous Incentives: Evidence from the Patent Office (NBER Working Paper 32066), Michael D. Frakes and Melissa F. Wasserman explore whether shifting from a discrete quota structure to a continuous one reduces the bunching of work effort before deadlines. They also assess changes in outcome quality. Their analysis draws on more than 7 million patent applications filed between March 2001 and June 2022 with the US Patent and Trademark Office (USPTO).

The researchers study a 2011 reform of performance incentives for patent examiners that was designed to reduce the disproportionate share of task completions at the end of quota periods. The reform modified the USPTO’s biweekly quota system by adding incentives tied to examiners’ average time from patent application review to completion. They exploit differences in the ability of patent examiners with different degrees of seniority to control the timing of their review completion to analyze the impact of these incentives.

The researchers find that completion of patent application reviews on the last day of a quota period fell by roughly 50 percent following the USPTO reform. They also find a decrease in average examination pendency — a measure of the timeliness of reviews — and an increase in the day-to-day consistency of examiners’ approval rates. In the pre-reform period, examiners were more likely to issue easily reversed first-round rejections as the quota deadline approached than on other days. Because these rejections could be reversed in later reviews, they might not affect the ultimate disposition of the applications, but they would prolong the overall application process. This differential spike in first-round rejections fell to almost zero following the reform.

To assess whether the change in the timing of reviews affected their quality, the researchers analyze applications that were submitted to both the USPTO and the European Patent Office (EPO). The EPO maintains similar patentability standards as the USPTO, so its decisions can provide a comparison group for the USPTO outcomes. There was no change in the concordance between USPTO and EPO decisions around the time the USPTO added the continuous incentive system. This suggests that the reforms led to a reduction in application processing time without any diminution of examination quality.

— Lauri Scherer


The researchers acknowledge funding from NIA grant 1R01AG076586-01A1.