Readers are beginning to unpack the ‘Hargreaves Review‘ — the UK government-commissioned study on the relationship between intellectual property and growth.  For good summaries, see Jamie Boyle (Boyle was one of the advisors for the study) or Mike Masnick.

I’ll skip over the modest patent and design recommendations and make a couple points about copyright and evidence–topics the report deals with at length.  The report takes a number of genuinely constructive positions on UK copyright law, including recommending broader exceptions for parody, private copies, and archival practices; ensuring that those exceptions can’t be trumped by contract law; opposing copyright term extensions (which are presently lower than in the US for recorded music); and consolidating copyright licensing practices to cut through barriers to the legal exploitation of copyrighted works.  The report is also the strongest official pronouncement yet on the low quality of the research that informs the copyright and enforcement debate–a record it calls “threadbare” and dominated by “lobbynomics.” 

On the other hand, the report also hedges some of its conclusions.  I understand why the authors do this: the report is plenty ambitious without courting unnecessary controversy. But given the analytical goals of the report, these moments are a little… annoying.

This hedging includes little things like the argument that in “middle income and emerging economies such as China, improved enforcement regimes may yield better rewards both for domestic innovation and returns to foreign firms through foreign direct investment and technical cooperation.”  Well, yes it may do that, but the Chinese have made a very effective development strategy out of rampant copying, and US and European companies are still rushing to invest.  Historically, countries raise IP enforcement when domestic companies begin to make more use of IP protection.

And it hedges on some big things, like enforcement.  Here the study begins well:

The question is: in these circumstances, what does an effective enforcement regime look like? In order to provide a realistic answer to that question, we need to understand the nature, prevalence and dynamics of rights infringement. Only on a base of carefully evaluated evidence can a successful and sustainable enforcement approach be constructed.

It goes on to note the lack of such evidence in this area:

in the Review’s four months of evidence gathering, we have failed to find a single UK survey that is demonstrably statistically robust….

The uncertain and disputed nature of the prevalence data makes it difficult to reach confident conclusions about the impact of copyright piracy on growth.

And then it punts:

At this moment, given our state of knowledge, no-one in the UK could make an informed assessment of what is the right level of resource for online enforcement in the UK. We can only guess and get on with it, using rigorous evaluation to develop the kind of cost-benefit framework described by WIPO.

This ‘get on with it’ approach would be ok if the status quo wasn’t the recently passed Digital Economy Act, which includes criminal prosecution of file sharers and plans to cut off internet access for repeat infringers.  The UK is only one of a handful of countries to undertake the former (though it is on the books in the US, Brazil, Russia, and other countries).  And the latter is increasingly seen in Europe as a violation of a “fundamental right” to expression that relies increasingly on the Internet.  The report doesn’t venture into this hugely contested territory.

In other respects, the report comes very close to our view of the broader evidentiary discourse on piracy, and in fact cites or paraphrases MPEE at several points.  Here I’ll just add a couple points that haven’t been widely picked up already:

In the catchily-named ‘Supporting Document CC‘, which deals with data quality issues, the report authors offer a very detailed dissection of the 2010 TERA report on economic losses to piracy in Europe, which got a lot of attention for claiming that Europe would lose 1.2 million jobs to piracy between 2008 and 2015.  The Hargreaves report raises many more issues than our own brief research note from last year (though doesn’t go as far in its conclusions*).  Like us, they give TERA points for actually documenting their assumptions and methods–a standard that few other industry studies meet.  But like us, they find that “there are methodological limitations and omissions at each stage of calculation which means it is problematic as evidence for policymaking.”

I won’t bore you with the details of this walkthrough.  There is a reason that this is in Supporting Document CC.  And here too, after undermining the TERA study on point after point, the report defaults toward understatement with the observation that “industry associations are not always prepared to share their proprietary data sources and methods, making it difficult to verify the data they produce.”  Not always indeed.

I don’t follow UK politics closely enough to place a bet on how this turns out, though Masnick’s scepticism seems all too realistic.  But between the Hargreaves report, last year’s GAO report, and our own beloved MPEE, perhaps we can now treat a few basic issues in this debate as settled.  Quoting the Hargreaves report:

  • not all illegal downloads are lost sales – the user may not have paid a higher price for a legalcopy absent cheap or free illegal versions;
  • money not spent on legal copies is not lost to the economy – it may be spent on other purchases. This is of no comfort to the sector suffering losses, but the effects across the economy will not necessarily be problematic;
  • even within the industry affected, purchases prompted by experience from an illegal copy (forexample, concert tickets or other merchandise) can offset losses;
  • in business software, piracy has promoted the lock-in effect for the legal provider’s software and helped to make that software the global standard.”

The first issue is basically granted by all parties at this point, though there are disagreements about what the “substitution effects” are between piracy and lost sales (which makes the disclosure of assumptions and methods critical).  The third is also acknowledged, but usually dismissed as insignificant.  I’ll have more to say on this in a future post.  But no industry-funded report, to my knowledge, has ever acknowledged the second or the fourth.   And I’ll throw in one more:

  • Citing MPEE, they note that “Original sources of data on links with organised crime are not easy to find.”

Perhaps we can vow, once and for all, that scattered anecdotes regarding these links will no longer be a driver of the enforcement debate.  That would be three clear wins for research over lobbynomics.  And maybe the Hargreaves Review’s good works will finally drive a stake through the heart of the TERA report, which unfortunately became a reference point in key EU debates on enforcement in 2010.  Perhaps in the future we can look to the example of the Australians, who were better prepared when the TERA method came to town.  What was that Men at Work lyric about tragedy becoming farce the second time around?


*We see reason for doubting, for example, that there are any real losses to software piracy–a number that represents about half the total losses in the report.  We also suggest that in the software and audiovisual sectors, Europe is a net importer rather than an exporter, and so likely to be on the receiving end of a consumer surplus from the piracy of those goods rather than a loss.  The Hargreaves study posits that the UK is an IP exporter with respect to Europe (and would therefore benefit from a unified market for audiovisual goods in Europe).  But is it an overall exporter?