 The firms who apply for names for pharmaceutical products, so for medicines, cough juices or painkillers or medicines against hypertension, for hypertension, tend to apply for more than one name for these, so for more than one trademark for these products. And they would do this because they might be afraid that a regulator who's trying to make sure that the name of this particular medicine doesn't confuse the patients might be strict and might say that a particular trademark that the firm wants to use to protect its pharmaceutical product is too similar to another trademark that's already in use. And so what the applicant will do is they will choose a whole set of names that say five or eight names and register all of these in the trademark office and then they will take all eight names to the medical regulator and the medical regulator will test every single one of those and then their medical regulator may just approve two or three of them. And then the pharmaceutical company at the end of all of this will choose one of these names to protect its medicine. From the perspective of the person looking at the trademark register, the problem is that there are another seven or another five or another six trademarks that that firm also applied for for that product that it's not going to use. And these trademarks to some extent are no longer available to other firms and potentially similar trademarks are not available to other firms and that makes it expensive for these other firms to look for their own names. And that's really what I'm trying to understand, this particular phenomenon is widespread and whether it's important. If you look at European enlargement as a particular experiment where the European Union was enlarged by 10 countries and as a consequence the probability of convincing a medical regulator that a particular name is unproblematic becomes much smaller. It's much more difficult to do that in the new regulatory environment. I would expect to see the pharmaceutical firms to apply for more names at the same time. And so what I'm trying to see empirically is whether this is really true and then once I see the signal which is the case in this particular piece of research I'm trying to find out whether extrapolating from the signal I'm seeing I can claim that the problem is very large and we've got a lot of clutter on the register or it's not very large. And the last piece of the work is really then to try and figure out well what are the costs associated with creating all these surplus marks. And the bottom line is that it's quite a significant sum that's spent on creating surplus marks but it's not a sum which is so huge that it's a first order problem for this particular industry. So this particular research I think is just going to help to try and understand what procedures the people who run trademark offices should adopt in order to make it most efficient for the users to use this particular type of IP protection. So the underlying question really is can we make it cheaper for applicants to find the names that they need to protect their products. This paper is obviously specific to a very clear industry context which is a pharmaceutical industry. But it has motivated me to think that it would be interesting to try and understand the filing behavior of companies that use multiple trademark jurisdictions more widely. So companies that are multinational that will be going to many, many different countries because you could argue that they face quite similar problems in the sense that they will often want to have the same name for their product across the world and the probability that that particular name is going to survive the procedures that are active in let's say Uruguay and Paraguay and Canada and the United States and Russia and Asia is not so high. So again the companies are likely to pursue a strategy where they're going to ensure themselves by going for multiple applications. And it would be very interesting to try and understand what that means for all of these trademark systems. So do we have a lot of surplus trademarks that are being created in this way?