Talk:Cooperative Research

From ASTPwiki

[edit] IP rights on research contracts

(Note: this thread was copied from ASTP mailing list in order to illustrate possible use of the discussion pages from ASTPWiki! Mieville 17:10, 1 May 2006 (UTC))

Colleagues,

I'm interested to know what other universities feel to be an equitable position on IP rights & ownership in fully funded research contracts with industrial sponsors. There are two reasons for this:

1. There is an EU-sponsored conference early next month (to which both UNICO and ASTP are sending representatives) - the subject of IP ownership will be high on that agenda. 2. It's currently a hot topic in the UK, one that is soon to be discussed with the CBI (Confederation of British Industry) in an attempt to come to a common understanding and (that pigs might fly) position.

I've discussed with a number of colleagues, both in the UK and Europe. A sense that a common position might be that, where a sponsor pays the full economic cost of the research, (a position to which we are quickly moving in the UK):

- The university should own the IP.

- The university should grant a free non-exclusive license to IP that results from the work.

- The sponsor should have an option to take an exclusive license (within certain fields) to IP resulting from the work - this option expiring (say) six months after the end of the contract. Clearly, the sponsor may wish to define the terms of such a license up front (at least within a band) which may or may not be realistic depending on how well the resulting IP can be envisaged.

- Should the sponsor relinquish all rights to the IP, then the university pays a proportion of any licensing income to the sponsor (at MIT it's 20%).

- No pipeline.

Background (pre-existing) IP has to be negotiated separately, but the general rule should be that, where it is free to do so, the university grants the sponsor the background rights it needs to carry out the project, but that any commercial use is subject to separate discussion. Any commercial license to background needs to be discussed on a case by case.

Shared IP also remains a problem, though the same principles can apply;

-both parties have non-exclusive rights to exploit the IP (freedom to use and to sub-license)

-If the industrial sponsor wants exclusive rights then it should have that option, but clearly at a reduced rate.

The principle behind this thinking is that:

1. Providing the sponsor is paying for the research at full cost (full indirect costs, but no profit), it's reasonable for a sponsor to have the freedom to use the results (and unreasonable to prevent them from doing so). 2. However, if the resulting IP is so important to the sponsor that it needs exclusively, then the work has clearly been of far greater value to the sponsor and some additional sum (fee or royalty) is reasonable. 3. Background IP is clear to all at the outset, and should be discussed up front if access to that IP is important to the sponsor.

It seems a good time to open the debate. I'm therefore interested to know whether this seem a reasonable basis on which to do business? Is it too soft a position, or too hard?

You will notice that I'm sending to two email groups, UNICO (in the UK) and ASTP (Europe). It might be interesting to share feedback and discussion across the two groups, thereby widening discussion. This can be achieved by hitting the 'reply to all' button.

Jeff

Dear Jeff,

You really touch all the relevant topics in IP rights derived from industry-university research relationships.

Ours is a research institute of the Catalan Government in Spain, operating under the laws of the private sector, with some 50% of its budget coming from institutional competitive projects and research contracts with companies (for details see www.irta.es).

In case it may help you, we use, as a first criterion to start conversations with companies in the subject of reference, the following question: Whose was the idea of the project?

1. If the answer is: the company's, and if it totally funds the research (something that seldom, or never, happens), then we recognize the IP rights to the company.

2. If the answer is : us, then we keep the rights and start negotiating all the rest with the company. In this second case, all the points you mention have to be considered, almost case by case. In our experience, we have to arrive to tailor-made agreements.

Best regards,

Juan 84.227.29.241 14:51, 1 May 2006 (UTC)

Colleagues,

In order to answer Jeff's subtle question, I think that a number of distinctions are to be made: "Is the University privat or public?"; "Is the company-funded research project carried out by University personnel during extra hours?"

Let us assume that the University is public and the research project is not carried out during extra hours.

In this case, while it may be feasible to calculate the full financial cost of the research, it would be a formidable task to calculate its full economic cost, that is the opportunity cost of carrying out the research (the latter should be taken into account in the case of public Universities!). This opportuinity cost equals the socio-economic value of the would-be research, should the funded research project have not been undertaken by the University. The economic cost is always above the financial cost.

Generally, Universities tend to calculate financial costs of funded research projects. Moreover, there is no reason why public Universities should favour one company vis-a-vis other domestic competitors (taxpayers as well and, consequently, stakeholders of public Universities).

The brief foregoing analysis leads to the following suggestions, which resemble most of Jeff's positions:

• The university should own (or co-own) the IP.
• The sponsor should have an option to take an exclusive license to IP resulting from the work.
• The sponsor should pay for the license (the negotiating price should take into account the sponsor's initial investment).

Consider that the option to take exclusive license embeds strategic and financial value for the sponsor!! 84.227.29.241 14:58, 1 May 2006 (UTC)

Jeff:

What is the meaning of "fully funded research contracts"? If we have a team of researchers who have used our publicly funded research facilities, worth many millions of dollars, for several decades to develop their skills to an extremely high level, do you consider a contract to be "fully funding" if it simply pays for the salaries plus some overhead component for use of the facilities during the life of the contract? As someone else has noted, what about the opportunity cost of foregone research?

Here at TRIUMF, Canada's national facility for research into sub-atomic physics, we are quite ready to assist SME's, but we do not give away IP to anyone who finances a short term research contract, which proposes to pay for a few weeks work that, in essence, downloads the accumulated background knowledge that has been established through decades of learning followed by decades of research, and will constitute part of any IP. While one component of our mandate is to assist in the commercialization of our research results, it is not appropriate to assign IP on an exclusive basis in return for a short-term contract.

As you are obviously aware, the inventor(s) own the IP and it has to be assigned to another party. Most universities now claim ownership of any IP developed with their facilities, and/or based on research that is founded on background IP that has been developed with their facilities, so the issue is clearly the assignment of university IP to the company or institution funding the research.

As you have noted, a royalty free non-exclusive licence is one answer, provided there are field-of-use restrictions and there is no right to sub-license. However, this is very similar to publishing the IP, since it may be difficult to negotiate a profitable licence to a third party competitor of the sponsor if the sponsor already has a free run (we have experienced this). Also the university must have the right to reclaim the IP if there is not a real attempt to put it into commercial application within a specified period of time. There is little point in having a sunset clause on the option for an exclusive licence if the industrial sponsor maintains a non-exclusive, since under those circumstances you cannot give an exclusive licence to anyone else.

If someone at the university can accurately forecast the commercial potential (even within a band) of IP that has yet to be discovered, then perhaps they should forget commercializing IP at the university and concentrate on picking stocks in the stock market.

I am not clear on what is meant by "shared IP". This term usually refers to IP with two or more inventors, one or more of whom come from each institution. If this means that their names will appear on any patent, (we have had experience with this too) each inventor named on a patent has the right to assign the IP to anyone they designate, and the IP essentially becomes the non-exclusive property of the parties to whom it is assigned. If it is simply held as some form of confidential information, then, of course there has to be an agreement, with fixed term, between the parties covering disclosure and licensing, or again, both parties will be free to do with their IP as they see fit.

If the sponsor relinquishes all rights to the IP, then I could only see paying a percentage of the royalties (after deduction of all of the university's expenses) up to the amount of the original sponsorship plus a reasonable return on the initial funding. Alternatively, the university could simply refund the full amount of the sponsorship up front, and regain all rights to the IP, if they perceive some commercial potential that the sponsor is not willing to pursue. These options should be included in the research agreement.

Jeff, I see the position you describe as fairly soft, in that the university gains nothing more than some research funding, at best, while the sponsor of the research has little to lose and much to gain by offering to pay the marginal, or possibly the variable, cost of research while obtaining strong rights to the IP. The sponsor wants to use the background knowledge and IP of the investigators plus the university facilities, thereby forcing the university to incur an opportunity cost, and yet only pay for the variable economic costs of the research. At the same time, there is presumption that the sponsor has the right to the IP, even though they have avoided the investment in any facilities. While this may appear appropriate from the perspective of a cash strapped publicly funded university, it would not be viewed quite as favourably by an industrial competitor that has made a significant investment of funds in comparable research facilities and full-time staff of its own. It is important to ensure that the university is fair both to the sponsor of research plus the general marketplace. This is most obvious, for example, in, say, engineering, where it is quite inappropriate for the university to offer the service of its professional engineers at their marginal cost, in competition with local industry. Sponsored research is simply a variation on this theme.

I hope this helps.

Philip 84.227.29.241 14:58, 1 May 2006 (UTC)

Philip,

I agree with your point that new knowledge builds on accumulated background knowledge. Still, I would consider the cost of building it as a sunk cost in measuring the cost of sponsored research at public Universities. The opportunity cost of forgone research is the "right cost" to me, although it may be a formidable measurement. The fact that it is higher than the financial marginal cost of the research (paid by the sponsor) legitimates the Universities' position to keep the IP ownership, and license it to the sponsor not for free.

Giovanni 84.227.29.241 15:07, 1 May 2006 (UTC)

Dear Giovanni:

Thank you for your kind reply to my comments. While I can agree that the building costs etc can be considered "sunk costs" to the university, they represent a significant competitive advantage to the company that does not have to include them in their amortized variable costs for the research, when compared to a company that has invested in those costs. Also, of course, there is an element of lost opportunity cost with those university facilities that are dedicated to contract research and not available for other research, either university or other contract.

I agree that opportunity cost provides the best proxy for full cost allocation to a research contract, but unfortunately it is extremely difficult, if not impossible to calculate.

Again, my thanks for your thoughtful reply.

Phil 84.227.29.241 15:07, 1 May 2006 (UTC)

I'm not a University representative, but I act for some Universities.

The position Jeff describes is pretty much the same as our standard model research agreement, but I disagree with:

Should the sponsor relinquish all rights to the IP, then the university pays a proportion of any licensing income to the sponsor (at MIT it's 20%).Why should the sponsor share in the rewards if they have declined to take up their rights?

Instead I would use this:

If the Option lapses and the University licenses any patent applications that were the subject of the Option to a third party, the University will seek to recover any patenting costs paid to it by the Sponsor in respect of those applications from the third party and reimburse any recovered costs to the Sponsor.

The principle could be extended to other exploitation costs which the sponsor has paid during the option period, but reimbursing expenses is quite a different thing from giving over a share of the income.

On the question of joint ownership I am not quite sure how what Jeff proposes would work, since if the joint owners have the non-exclusive right to sub-license then the sponsor would not then be able to have an exclusive licence. Perhaps I have misunderstood.

Best wishes

Elizabeth 84.227.29.241 15:07, 1 May 2006 (UTC)

A subtlety that you don't mention is what the automatic non-exclusive licence to use resulting IP entitles the sponsor to do. My impression is that US universities are increasingly limiting this licence to use for internal research purposes, i.e. a non-exclusive licence for commercial purposes would have to be separately negotiated.

Mark 84.227.29.241 15:08, 1 May 2006 (UTC)

From a US perspective, I would agree with Mark and suggest that it's rare for the research sponsor to obtain any royalty-free commercial rights (whether to sponsored or background university IP). Of course, many universities will make regular IP disclosures to the research sponsor, but that's for informational purposes only, and often conveys no commercial exploitation rights.

84.227.29.241 15:08, 1 May 2006 (UTC)

Colleagues,

Thank you for all the responses to my question last week relating to indirect costs. As I expected, there was no real consensus, some being resigned to negotiate on a case by case basis.

Clearly, as long as the price is a variable, the deliverable (IP) also varies. I was trying to fix the price (at the full economic cost, no more no less), and ask whether it was then reasonable to fix the IP rights granted to the sponsor. I assumed that the work was being carried out in a university. I assumed that the full economic cost was the sum of the direct and indirect costs (no opportunity costs or profit).

As some pointed out, this is a bit of a theoretical argument since few of us achieve full costs. Companies object to paying overheads/indirect costs and we often find that the academics side with the sponsoring company, forcing us to accept lower indirect costs. However in the UK we are soon to change to a very different environment where all research activity will have to be properly and transparently costed and where it will be much harder to subsidise industry (or charity, EU etc) funded work from other public sources.

Ultimately, everything comes down to market power, unless you are in the US, where the minimum you can charge is effectively set by the federal funding agencies.

Leaving aside the problems raised by joint IP, I remain convinced that the arrangements I set out are the right ones, and would like to argue the case once more.

Free non-exclusive license.

This leaves both sides free to use the resulting IP, possibly in conjunction with their own background IP:

• The sponsor (company) has complete freedom to use the results

but does not rely upon the results of the research for any competitive advantage, this advantage presumably deriving from background IP and other assets (reputation, management, channels to market & market knowledge).

• Likewise, the university is free to use the new IP (including

know-how) in combination with its own background for whatever purpose, including commercialisation. It would be a huge shame if the university was somehow prevented from commercialising a technology because due diligence showed that one essential part of the jig-saw had unthinkingly been exclusively licensed to a sponsor.

There is a softer point here, which you here in the US more than in Europe. This has to do with the university's social mission. We are here primarily to serve our community, which includes businesses. Knowledge transfer offices are primarily sources of innovation, not revenue. Why then should it be so wrong to grant a sponsor freedom to use? Is it really right to charge a sponsor simply for the non-exclusive right to use the outcome of research that they have funded. - thereby creating obstacles to innovation? What are we here for? More to the point, if we push too hard, are we in danger of losing wider support of industry? Anyhow, how many of us derive any significant license income from non-exclusive license within research contracts? Pardon the polemic.

Option to an Exclusive License.

If the sponsor really needs an exclusive position to maintain its competitive advantage then this surely means that the IP has real value to that sponsor and that they should pay (much) more than simply the full cost (which is where a License Fee or Royalty comes in). It is also preventing others from using the same IP (and preventing the university from commercialising the IP itself - a true opportunity cost).

This charge can take many forms. We might negotiate a higher price at the outset - one that incorporates a real profit. We might seek a one-off fee. Or we might agree to risk-share with the sponsor by seeking a royalty.

I assume that we can never give exclusive rights to know-how.

Background

Most agreed that background is a separate issue, but one that should be sorted in advance. It was suggested by another that where the background knowledge is substantial then it would be appropriate to charge a higher price for the research. It amounts to the same thing.

Revenue sharing with sponsor.

This is less important, but I nevertheless like it as a mechanism for giving a company a reason for relinquishing all rights to IP, which they otherwise have no incentive to do.

There are other considerations which some mentioned, like joint IP, taxation, patent costs & publication rights. These are all important, but it seems to me that it's important to get the principles right first.

Any further thoughts welcome. As before, 'Reply to All' in order to include both ASTP & UNICO discussion groups.

Jeff