It’s become an article of faith among those complaining that patent reform has gone too far that the 2006 eBay case must be overturned—so much so that Sen. Coons has included it in both editions of his STRONGER Patents Act (one of a multitude of reasons that bill is bad policy.)
But, as with so many articles of faith on the anti-reform side, the data simply doesn’t back up these claims. [1][2][3][4][5]
Permanent Injunction Grant Rates
Using Lex Machina’s data, I examined pre- and post-eBay permanent injunction grant rates. And what that revealed is that the complaint about injunction rates is baseless.
In order to avoid distortion from non-contested cases, injunctions issued as part of a default or consent judgment are removed. Looking only at cases resolved in favor of the plaintiff based on judgment on the pleadings, summary judgment, a trial verdict, or a post-trial JMOL motion, permanent injunction requests were granted in 89.1% of pre-eBay cases.
Examining the same class of post-eBay cases, permanent injunction requests were granted in 85.2% of plaintiff merits wins.
In other words, plaintiffs who win their case post-eBay are almost as likely to receive a permanent injunction as they would have been pre-eBay.
But if nothing had changed, eBay wouldn’t have been viewed as an important case. There’s a simple explanation for why it is.
NPE vs. Non-NPE
The eBay case required that a plaintiff show that “remedies at law, such as monetary damages, are inadequate to compensate for that injury.” In other words, you have to show that the harm you suffered from the infringement can’t be recompensed with money. For competitors, that’s relatively easy—the harm to your market share and competitive position is often enough. But if your entire business model is licensing your patents for money, it can be difficult to show that money is insufficient to compensate your injury.[1. While more difficult, it is not impossible to make this showing. For example, a licensor that licenses its patents in partnership with a startup—the typical model for a tech transfer office—can suffer an equitable harm even though the patent owner is primarily receiving monetary rewards.] Because of this, NPEsNon-Practicing Entity. A broad term associated with trolls but now disfavored because it includes universities and legitimate technology developers that seek to license technology in advance rather than after a producing company has independently developed it. More would presumably have a harder time securing injunctions post-eBay and might well choose not to incur the expense of trying to secure an injunction when they’re likely to fail.
To test this theory, I split the data using Lex Machina’s “high volume plaintiff” (HVP) definition as a proxy for NPEsNon-Practicing Entity. A broad term associated with trolls but now disfavored because it includes universities and legitimate technology developers that seek to license technology in advance rather than after a producing company has independently developed it. More. This is an imperfect proxy, as an HVP is simply any entity that files more than 10 patent lawsuits within a single calendar year. Reviewing a list of HVP cases, the majority are well-known NPEsNon-Practicing Entity. A broad term associated with trolls but now disfavored because it includes universities and legitimate technology developers that seek to license technology in advance rather than after a producing company has independently developed it. More such as Neodron, Blackbird Tech, Intellectual Ventures, and Uniloc. However, a few operating companies do fall within the HVP definition. At the same time, a number of NPEsNon-Practicing Entity. A broad term associated with trolls but now disfavored because it includes universities and legitimate technology developers that seek to license technology in advance rather than after a producing company has independently developed it. More don’t meet the quantity threshold and thus fall outside of the HVP definition. In one particularly ironic example, the “Patent Asset Licensing, LLC” cases appear in the non-HVP list—they just didn’t file enough cases to make the cut. But on balance, few operating companies appear to be classified as HVPs, while a number of NPEsNon-Practicing Entity. A broad term associated with trolls but now disfavored because it includes universities and legitimate technology developers that seek to license technology in advance rather than after a producing company has independently developed it. More are classed as non-HVP, meaning that if there’s a differential impact on NPEsNon-Practicing Entity. A broad term associated with trolls but now disfavored because it includes universities and legitimate technology developers that seek to license technology in advance rather than after a producing company has independently developed it. More and non-NPEs, we would expect to see it reflected strongly in the HVP data and weakly in the non-HVP data.
Based on this split, the data supports the thesis—corroborated by Prof. Seaman’s research into the same topic—that the decline in injunction rate is likely due almost entirely to reduced NPENon-Practicing Entity. A broad term associated with trolls but now disfavored because it includes universities and legitimate technology developers that seek to license technology in advance rather than after a producing company has independently developed it. More success rates. Prior to eBay, the rate at which a successful HVP plaintiff’s injunction request was granted was 88.8%, nearly identical to the 89.1% rate for non-HVP plaintiffs. However, post-eBay, that rate drops to 62.5% for HVP plaintiffs, while non-HVP plaintiffs receive a requested injunction 86.5% of the time.
Even if the grant rate for injunction requests has held more or less steady for operating companies, there’s a second possibility—maybe they’re asking for them less often.
Permanent Injunction Request Frequency
If operating company plaintiffs were requesting injunctions in a smaller percentage of cases, then the lack of a decline in grant rate could be due to plaintiffs sorting for meritorious claims, rather than any reduction in success rate for injunctions for operating companies.
And looking again at data from Lex Machina, there has been a significant decline in the number of requests for permanent injunctions. Pre-eBay, a request was made in 56.3% of contested plaintiff wins; post-eBay, that number drops to 43.0%.
But again, the HVP/non-HVP split reveals that much of this is likely due to NPENon-Practicing Entity. A broad term associated with trolls but now disfavored because it includes universities and legitimate technology developers that seek to license technology in advance rather than after a producing company has independently developed it. More behaviors. The HVP rate of requests went from 52.9% to only 29.6% of contested plaintiff wins, while the non-HVP rate showed a significantly smaller decline from 56.6% to 44.1%. Combined with the knowledge that the HVP data is mostly NPEsNon-Practicing Entity. A broad term associated with trolls but now disfavored because it includes universities and legitimate technology developers that seek to license technology in advance rather than after a producing company has independently developed it. More, while the non-HVP data is more mixed, it’s likely that much of that non-HVP decline—if not the entirety—is due to non-HVP NPEsNon-Practicing Entity. A broad term associated with trolls but now disfavored because it includes universities and legitimate technology developers that seek to license technology in advance rather than after a producing company has independently developed it. More.
The Impact of Injunctions
The data illustrates that the eBay decision has primarily impacted NPEsNon-Practicing Entity. A broad term associated with trolls but now disfavored because it includes universities and legitimate technology developers that seek to license technology in advance rather than after a producing company has independently developed it. More, with limited to no impact on operating companies in terms of success rate on injunctions. And that’s a very good thing.
The core principle of patent damages is that the patent owner should be awarded a remedy adequate to compensate for the infringement. And over a hundred years, the Supreme Court clearly stated that remedy, holding that “[w]hen a patent is for an improvement, and not for an entirely new machine or contrivance, the patentee must show in what particulars his improvement has added to the usefulness of the machine or contrivance. He must separate its results distinctly from those of the other parts, so that the benefits derived from it may be distinctly seen and appreciated.” In other words, the value of a patent is the contribution of the patent to the whole—the patentee is not entitled to recover the value contributed by other aspects of the device.
That’s particularly important for modern technology, where a device might be covered by hundreds, thousands, or even hundreds of thousands of patents. Each individual patent may contribute only a tiny portion of the value of the full device. But if the patent owner can obtain an injunction, that allows them to block distribution of the whole device. A manufacturer, faced with an injunction, risks losing the entire value of their product even though only a small sub-component of the product is relevant to the patent.
By ensuring that injunctions are unavailable when monetary damages are adequate, the eBay decision eliminates that in terrorem threat, allowing patent licenses to be made based on actual technological value of the invention rather than on the value of the product as a whole.
So when a commentator claims that eBay has harmed innovation, remember that the innovation they’re defending is the ability of NPEsNon-Practicing Entity. A broad term associated with trolls but now disfavored because it includes universities and legitimate technology developers that seek to license technology in advance rather than after a producing company has independently developed it. More to extract a royalty that’s higher than their contribution to technology.
