Talking ‘Data’ With IP Australia’s Chief Economist

It has been an unduly long time – just over a year, in fact – since I published the first part of a conversation I had with IP Australia’s Chief Economist, Benjamin Mitra-Kahn.  But the two of us (mainly me, if I am honest) have finally got our act together to edit most of the transcript into a readable form.  I am therefore very pleased to be able to start publishing the remainder of our discussion.  Despite the passage of time, the content is still highly relevant, indeed in some ways even more so, considering the increasing importance of economic analysis and the role of data science in driving government policy in relation to intellectual property.

Late last year, for example, the Australian Productivity Commission (PC) released its final report on its Inquiry Into Australia’s Intellectual Property Arrangements.  The report reviewed the Australian IP system in its entirety, and made a number of significant recommendations to the Government, which are currently under consideration.  While prior reviews and inquiries into components of the IP system had been conducted by panels that included economists, the PC’s review was the widest, and the first to be conducted entirely by economists.  Naturally, the PC sought to draw conclusions and make recommendations based on evidence, and a key theme of the final report is the need for accountability in the IP system, including by developing and maintaining a sound evidence base to inform policy decisions.

It is in this context that economic research, such as the 2012 study by Boston University academics James Bessen and Michael Meurer which concluded that ‘patent trolls’ cost the US economy $29 billion in 2011, can have a huge impact.  Some people (including me) questioned the reliability of the source data and methods used in that study, but a far larger number – including some widely-read media outlets – simply took the Bessen and Meurer results at face value.

I was therefore very interested to get the views of IP Australia’s Chief Economist on that particular study, and to talk about the work that is being done in Australia to make better-quality data available to researchers, and other interested stakeholders, through the IP Government Open Data (IPGOD) initiative.

The $29 Billion Question

Mark Summerfield: I guess part of the reason for there being a perception that economists tend to be ‘anti-patent’, or at least wish to weaken the patent system, is that there have been some very prominent economists – people who clearly have a gift for getting exposure in the media for the work that they do – who do appear to have an agenda. One of the greatest examples of that in recent years has been the work of James Bessen and Michael Meurer in the US, who famously concluded that the patent system was costing the US economy $29 billion every year.
 
Ben Mitra-Kahn: But it wasn’t the patent system. They were making the argument that, in the US they have a problem with non-practising entities. They are causing lawsuits and delays to people, and legal cost to people. All of that looks like $29 billion. Now, that’s a big number and it’s been taken apart and put back together by people but it’s a particular subset of a problem. The thing that’s really problematic, I think, is that all this stuff’s done in the US. If we had this debate in Australia we’d go, ‘ah there’s probably these three cases we could look at – and that’s it.’ There were 7800 cases in the US last year within ICT, software, and telecoms where non-practising entities are suing people. So that’s the problem in the US. So it’s a problem in the IP system? My personal view on that is it’s not an issue with the IP system, it’s mainly a court issue in the US.
 
MS: Certainly ‘litigation reform’ has become more of a focus in the US now. But I think some of the major criticisms of Bessen and Meurer’s work was their failure to distinguish amongst non-practising entities. To distinguish between true patent trolls and, say, universities. And that their $29 billion appeared to include transfer costs and not just actual dead-weight costs.
 
BMK: There’s a lot of interesting things under there.
 
MS: But why did they not recognise that they were doing that? How could somebody whose entire research career is devoted to this kind of economic analysis fail to make those important distinctions, and then say, oh well, it’s the data? Because those are the things that the people who are at the coalface want to see, and it generates a lot of mistrust and criticism.
 
BMK: I can’t speak for Bessen and Meurer, but I think they were trying to give you a ballpark idea of the costs they thought were implicit in the activities of non-practising entities in the US. And they were trying to give a picture to policy-makers saying that this is an issue, and it’s a big issue. Compared to the size of the US economy, $29 billion is not a lot of money – but it sounds like a lot of money. It’s a US$18 trillion economy. In relation to that, $29 billion is not a terrifying number [0.16%]. But if you look at people who have patents, people who go to court, that’s actually quite a large sum in that context.
 
Now, are they in the order of magnitude that’s correct? I’m sympathetic to their number. They do another paper and they have $42 billion, someone else does a paper and you end up with tens of billions. The magnitude of the issue is this big. Does that warrant further investigation? I think the answer is yes. I wouldn’t take these as, ‘ah ha! this is the answer’, or ‘this is the number’. It’s very much, this is the ballpark area we’re in and then you’ve got to go and pick apart the assumptions.
 
Following Bessen and Meurer, when I was still working in the UK, we looked at non-practising entities. We thought, ‘that’s a big number’. Can’t see something that indicates the maths is terrible, but there are some funny assumptions in here. I mean if we play around we still get a big number.
 
When we looked in the UK we had two academics, Christian Helmers and Luke McDonagh, go through the Patent Country Court cases, and the High Court cases that involved patents, looked at the applicants, and tried to determine whether or not they were non-practising entities. [Link to paper Trolls at the High Court on SSRN.] And in a 12-year history of cases there are 15 cases. And nine of those cases were brought by manufacturers against non-practising entities. You go, oh, this is probably not a big problem in the UK. And then we went off to actual lawyers. And then we had the Small Claims Court, we had the reforms. Through all of that, had we found similar things to what was found in the US, we would have had a very different policy conversation.
 
So, I think in the US case it’s a good thing for them to highlight it. I think the problem comes when you get pieces like, you referred to the piece in The Economist when we spoke on the email, and they had, you know, ‘we must fix the patent system’ and all of that is the US patent system. They’re concerned about submarine patents, patents that aren’t published. And every country in the world goes, ‘oh no, we publish all the applications’. But the US has a different practice. A number of domestic-only applications aren’t published, or will never be published, almost.
 
And so, there’s all this evidence on the US patent system, because they’ve got good data, and so we all sit around and go, ‘ah ha! this is happening in the US, it must apply here’. It’s a terrifying thought if you think of Australia compared to the US – technology importing, specialised areas, smaller, open market. There’s all sorts of things beyond that, such as 90 percent of patent applications are from abroad here.
 
The evidence they have in the US probably doesn’t apply here every time, and possibly not the majority of the time. Do we then think it’s important enough for us to develop our own evidence-base for this in Australia? I think IP Australia answered that a few years ago saying, yes we think that’s valuable to do. So, I have issues with the US system in that sense.
 
MS: I think a lot of people have issues with the US system in that respect. What the correct answer is, I think, is obviously up to the Americans.
 
BMK: Yes it is very much up to them to determine what they want to do. I mean, I think again they can learn from particularly what Australia has done in some places, they can learn from other places where they can get that analysis, and get the legal analysis, the economic analysis, oh this is why this isn’t a problem in UK and Australia. Maybe we should think about emulating that or see how they could fit into our practice.

IPGOD: Genesis

MS: So we are talking about getting the data which brings us to the IPGOD (IP Government Open Data). Now do you actually call it the ‘IP God’ where you are, or is it not politically correct to use that terminology?
 
BMK: The only argument was over whether to call it the ‘I.P. God’ or ‘Ip God’!
 
MS: And the argument worked out as …?
 
BMK: Neither. I stay with the IP God.
 
MS: So this is like patent versus pa(y)tent?
 
BMK: Pretty much.
 
MS: We can use either one in the same sentence.
 
BMK: Pretty much. It’s a really good acronym.
 
MS: So, how long has that actually existed for now?
 
BMK: We launched it in October 2014; that was the first version of it.
 
MS: And what was in there?
 
BMK: So there was basically a hundred years of patent, trade mark, design and plant breeder rights data, or data as far back as we had computers. And then some of the back-capture things where we scanned things and put it back in. And so we took all this data which we have in our backend systems and created a large data dictionary and said, ‘this field means that and this field means that’ and you are able to then look at applications and all the information around those applications in one place.
 
Then there was a lot of tidying up. And then we did even more work, we cleaned up the applicant names so that you did not have 47 instances of Nokia because for some reason there are 47 ways to spell Nokia – which is impressive! And then we created identifiers that allowed us to say ‘these are all Nokia’s patents filed in Australia’. And then we linked that to company numbers and other data so that we can use it across systems.
 
So we publish that once a year, and then this year we moved on to publishing a weekly update. So that we’ve got one IPGOD data set that’s an annual snapshot, and then every week with updated tables separately so people who want to make API searches or API services can use the live data as it comes out. So I think that’s really neat. Again we’re expanding that to include more data. Mainly a demand-driven thing and also things that we think would be interesting.
 
MS: And what is it being used for so far, other than, of course, the innovation patent review. Have there been other instances?
 
BMK: There’s been a couple of other research projects that have used the data. And all the patent analytics work that we do uses the data set, such as that food report [PDF 4.3MB] I mentioned earlier. We’ve also done a number of research projects looking at examination times, technology groups and so forth. So research in general can use it and the Department of Industry has used it for the National Innovation Map. The ABS (Australian Bureau of Statistics) have used it to link in with their survey data, and again the Department of Industry has it for their surveys, which means they hopefully do not have to survey companies so much because we just tell them ‘there were this many trade mark applications last year, for this applicant. You don’t have to waste their time asking them.’
 
Then on Twitter a couple of people who messaged me and said ‘oh we use the data for this thing. Does it work this way?’ One gentleman, I think he’s built a particular way of searching the trade mark data. So, this is customised Australian Trade Mark Search if you will. Because he’s interested in particular subsets of the data and he could do that with IPGOD – he builds an API. Another gentleman this week who is mailing us, and he is interested in the opposition case outcomes and tracking that. And opposition data’s in there, and so it becomes a way for him to do that.
 
So it’s a range of stuff and what we are trying to do in the next couple of months is build ways for people to easily access the data as a visualiser [recently released as IP NOVA], and ask questions without needing specialist statistical software. Because it’s big – millions and millions of rows, hundreds of columns, and so you end up in this space where Excel won’t open some of the tables, and that’s obviously a problem for people wanting to check the data. So with the IP report we’re hoping to put out some dynamic graphs, so you can look at applications by origin and places. So it’s the base for doing a lot of interesting things.

To Be Continued...

In the next part of my conversation with Benjamin Mitra-Kahn we talk about possible future enhancements to the IPGOD data set, perpetual motion machines, and the reasons why economists/theorists and attorneys/practitioners do not always see eye-to-eye.  And I promise that it will not take another year to get there, because the transcript is already set to go!
 
Finally, if you are on Twitter you can follow Ben @BenMitraKahn and the Office of the Chief Economist @IPAustralia_OCE (and, of course, me @patentology).

Tags: Benjamin Mitra-Kahn, Economics, Innovation policy, Interview, IPGOD, Patent Office