In the US, does an employer own an invention made by an employee?

A very bright European patent practitioner asked this in the EFS-Web listserv:

From I know and have been taught, there are no provisions in the US law stipulating that an employer automatically becomes the owner of an invention made by an employee (which is why we have to make sure that assignments are signed and recorded).

Could anyone direct me to the relevant statute in the US law?

His question inspired the blog article that follows.

There are at least two ways that this is the wrong question.

One way that this is the wrong question is that for most things that matter legally, there is no “US law”.  There is the law of Colorado and the law of New Mexico and the law of Louisiana.  If you want to know what elements would have to be proven by the prosecutor for you to be guilty of murder, the list of elements and their definitions would be non-identical from one state to the next.  (Yes, yes, I know that there is a federal definition of “murder” in the special case of crimes committed on military bases.)

There is a clause of the US Constitution called the “enumerated powers” clause.  It says that once you get past the relatively short list of powers that are enumerated in the Constitution as being held by the federal government, all other powers are reserved to the individual states.

In the US we have the National System of Interstate and Defense Highways, commonly known as the Interstate Highway System.  The federal government paid about 90% of the cost of the construction of those highways.  Suppose you hop into a car in San Francisco and drive east on Interstate 80 from there to Teaneck, New Jersey.  You would cross ten state boundaries along the way, and the speed limit would change at each boundary.  Why would it change ten times?  Because each state decides how fast you can drive.  (Yes there was a stretch of time in the 1970’s when all of the speed limits in the US were 55 mph, but that happened only because the federal government browbeat each individual state into passing that speed limit into law by threatening to deny federal highway funds to any state that failed to do so.  The US government brought about a uniform speed limit across the US not because it had the legal power to make it happen but because its threat to withhold money was a potent enough threat to make it happen.)

Just like speed limits and the definition of murder, so the question in the US of who owns property and under what circumstances is a matter of state law, not federal law.  Suppose you want to know who exactly will inherit the property of a person who has died (in the absence of a will) and in what proportions.  The answer will be different depending on whether you are having this discussion in Colorado or New Mexico or Louisiana.

(Fun fact:  Louisiana is unique among the fifty US states in having its laws descend from the Napoleonic Code.  It is for this reason that Louisiana has the property-law concept of usufruct.  In Louisiana it might turn out that for a US patent, one person has the usus of the patent, and some other person has the fructus of the patent.  I am not making this up.)

And if you want to know whether an employer in the US automatically becomes the owner of an invention made by an employee, the way that you would arrive at an answer is stepwise.  First, you figure out which state you are talking about.  Is it Colorado?  New Mexico?  Louisiana?  Then having figured out which state you are talking about, you would consult competent counsel in that state, who could tell you what the answer is in that state.

As I mentioned above, there are two reasons why the question as stated is the wrong question.  The first reason, then, why there is no “relevant statute in the US law” because this is not a matter of US law.  It is a matter of the law of each of the fifty states in the US.

The second reason the question as stated is the wrong question is that what made it so that employers do not automatically own inventions is not the existence of a law.  What made it so is the absence of a law.  The state of nature, in the US, is that in the absence of a law or contract or other agreement to the contrary, if you want to know who owns an invention, the answer is the inventor.  The point of this paragraph of the discussion is that if you wanted to know why this is so, you do not look to find some “relevant statute” that says this is so.  You look for such a statute and you can’t find it, because it does not exist, and that is what makes it so.

I imagine some alert readers of this blog who are just this instant getting positioned to slap the buzzer and who are getting ready to cite 35 USC § 261 as the US statute that they will say does indeed establish a uniform US law of ownership of patents.  Nope.  There is such a statute and it contains words that sort of dance around this notion, but the statute does not actually do that.

35 USC § 261 explicitly ducks this question.  It says:

Subject to the provisions of this title, patents shall have the attributes of personal property.

This remarkable sentence is actually extremely close to being content-free.  Congress could have either enacted this sentence or not enacted it and the outcome of any question as to who owns an invention would be identical, due to the enumerated powers clause of the Constitution.  The question of who owns any particular piece of “personal property” is a matter of state law, and it would be so regardless of whether or not Congress had enacted 35 USC § 261.

You can try to draw a distinction between “who owns an invention” and “who owns the patent” on an invention.  You could try to say “well, if you want to know who owns an invention, I will grant you that the answer is found by consulting state law, but if you want to know who owns a patent, well, that’s different, patents come from the federal government and so if you want to know who owns a patent you get to rely on US law.”  And to the extent you try to say this, you won’t get anywhere because the Congressional drafters of 35 USC § 261 expressly threw up their hands and left you in the patchwork of state laws to try to get the answer to the question of who owns a patent, namely that you have to go find out who owns the invention itself and this will determine who owns the patent.

35 USC § 261 does try as best it can to impose some small bits of federal uniformity about ownership of patents onto this patchwork of state laws.  A first bit is the three-month “quiet title period” that says you really ought to get any patent assignment recorded within three months of its execution date.  A second bit is the motivation to try to get such things notarized, because if an assignment is notarized, this statute establishes a prima facie presumption that the assignment is what it purports to be.

There is language in the Manual of Patent Examining Procedure which on a quick reading might seem to settle the question of who owns inventions.  But the MPEP is not law, and just because the MPEP says something does not actually mean that it is so.  The MPEP says this at MPEP § 301:

For applications filed before September 16, 2012, the ownership of the patent (or the application for the patent) initially vests in the named inventors of the invention of the patent.

Once again we have a sentence that is quite close to being content-free.  It is not up to the federal government (or to the nameless author of this sentence in the MPEP) to say who owns an invention.  It is up to the fifty states to say this.  The author of this sentence is basically saying “you wanna know who owns a patent?  Go find out who owns the invention (by consulting state law) and once you get your answer, that will tell you who owns the patent!”

What about applications filed on or after September 16, 2012, asks the alert reader.  MPEP § 301 proceeds to not answer this question as well.  It says:

For applications filed on or after September 16, 2012, the original applicant is presumed to be the initial owner of an application for an original patent.

This sentence is also essentially content-free.  “Presumed” by whom?  By a judge and a jury in New Mexico, applying New Mexico state law?  By a judge and a jury in Colorado, applying Colorado state law?  What this sentence skips over is that “the original applicant” is not determined by any federal law, and is not determined by any state law.  It is not determined by looking to see whether there was a signed assignment.  It is not determined by looking to see whether there was some sort of employment agreement in the relationship between the employee and the employer.  No, if you pick any particular US patent application that was filed on or after September 16, 2012 and if you decide you want to know who exactly is “the original applicant” in that application, the way you find out the answer is you go and look to see which box somebody checked on Form AIA/14 and how they filled in that form.  Maybe the person who filled out the Form AIA/14 said on the day of filing the patent application that the Binford Corporation is “the applicant” or maybe they did not.  Maybe the person who filled out the form was just making stuff up out of whole cloth, or maybe the person who filled out the form was actually correctly characterizing the answer that you would get if you were to investigate the facts of the case and apply the law of the relevant state to those facts.  But anyway you have this sort of random behavior on the part of somebody who filled out Form AIA/14, perhaps in great haste a few minutes before midnight during some rushed e-filing activity, a person who might or might not know what they were talking about, and the way they filled in the form is going to lead to a legal presumption of ownership?  

It turns out that the best way to understand the scope of this sentence in the MPEP is not that it somehow brings into existence any legal presumption of ownership in a court in Colorado or New Mexico or Louisiana.  The scope of this sentence in the MPEP is nothing more than the USPTO telling you what the USPTO itself will presume for its own internal purposes.  It is telling you that when the time comes to for the USPTO to decide who gets to control the prosecution of an application, when the time comes for the USPTO to decide whether or not to recognize a Power of Attorney, then what the USPTO will do is presume that the answer can be obtained by looking to see how the signer of Form AIA/14 filled in that form.  That’s it.  You want to know what are the consequences of “who is the named applicant” in a US patent application?  The consequences are not “who will turn out to be the owner in litigation?”.  The consequences are “who will the USPTO take to be the entity whose signed Power of Attorney will get recognized?”

Oh but there is one more consequence of how some flustered practitioner did or did not accurately complete Form AIA/14 in those minutes leading up to midnight on that frantic e-filing day.  The one more consequence of a box that did or did not get checked in Form AIA/14 is that you might win or lose some billion-dollar litigation about who owns, say, CRISPR.  Yes, under the Paris Convention we have SAOSIT (Google it!) which determines whether some would-be priority claim does or does not turn out to be valid at litigation time.  And if the priority claim were to fail, then the patent whose validity depended upon that priority claim might turn out to be invalid.  So yeah the correct filling-out of Form AIA/14 does have consequences that go far beyond just whether you will or will not succeed in getting the USPTO to recognize some Power of Attorney document.

Okay so now we circle back around to the original question posed by the European practitioner.  If you wonder who owns an invention, what I have discussed above is that you cannot and will not get the answer by looking for any US statute.  This is for two reasons — first that it would not in any event be a US statute, it would be a state statute.  And second, that what you are really looking for is the absence of a statute.  That the state of nature is that if there isn’t a statute to say otherwise, the ownership of the invention vests in the particular human being who did the inventing.

Of course there are some employers in the US who plan ahead and who arrange for each newly hired employee to sign some agreement on their first day of work, an agreement that says something about who will be the owner of an invention.  Maybe the agreement will say that ownership vests in the employer if the inventing happened on an odd-numbered day on the calendar.  Maybe it will say that the ownership vests in the employer only if the invention relates to the work for which the employee was hired.  For something like this it’s like asking, if the person dies having signed a will, who will inherit the property?  And the answer is, you cannot possibly answer this question unless you take a look at the will.  And if there is an employment agreement, and you are wondering who owns an invention, you cannot possibly answer this question unless you take a look at the employment agreement.

But I must emphasize that with wills, and with employment agreements, you cannot and must not simply read the document and then presume you know the answer.  For example there are states in the US that say that you cannot disinherit a spouse.  In such states, you could try to put ink on the page of the will that says the spouse will not inherit anything, and the all of the money goes to some charity, and you would fail at your purpose.  The state law would trump the ink on the page.  There are likewise some states that have laws that make it so that some words that you might try to put into an employment agreement will not work.  I am told for example that in California and in Colorado, the words in an employment agreement that purport to impose a non-compete obligation on the employee will mean nothing, or at least will have much narrower effect than you might think.  The state law will trump the ink on the page that purports to create a non-compete obligation on the part of the (former) employee.

I am told that there is some California case in which the ownership of an invention was not completely determined by an employment agreement but was in part determined by whether the inventor was married at the time the inventing happened, in which case maybe the invention was “community property”.  

I am told that there are some states in the US that have a legal concept of “shop rights“, meaning that if an invention was made by an employee who was working within the scope of his or her employment, maybe there will be an implied license of the invention (and thus the patent) to the employer.   But a mere “shop right” is clearly a lot less than a transfer of ownership.

Which brings us back around to the real situation which is that if you ever want to know in some particular case who owns an invention, there is just no choice about it.  You have to hire competent counsel in whatever state you are talking about.