Chris Devonshire-Ellis must be deliberately enticing me to comment on his China Briefing post about IP infringement and criminal thresholds – how can an IP nerd like myself resist such a topic? Unfortunately, after taking cold medicine this morning and then skipping lunch, I'm feeling a little loopy, so you may find this post a bit on the creative side.

I've probably written more about valuation of IP theft from the trade association angle than anything else. My usual rant is against some trade group that is inflating their China losses by using the retail sales price of a legal product, which I think is a pretty lame manipulation of statistics. Unfortunately they seem to get away with it all the time.

The criminal threshold issue is a bit different. This has been a bone of contention in bilateral trade talks between China and its trading partners for quite a few years now, with the foreign parties asking for a lowering of the thresholds. There has been some success with legal reform on this, but as Chris points out in his post, a huge problem continues to be proving damages, the amount notwithstanding.

To illustrate from the civil side first, there are only a few ways to get sales records from an infringer, assuming they exist at all. You can hire an investigator to "find" them, you can have the AIC seize them during a raid action, or you can petition a judge to discover them during an infringement action. There are many reasons why these options all generally suck. Well, to be fair, judges do consent to discovery requests more often than they used to, but it's still a rare thing.

That's for civil actions. You would think that for criminal actions, the police would have an easier time getting this stuff from the bad guys. Not so much, and infringers have gotten quite clever with JIT and online-based transactions to minimize their footprints. Very tough to track them down these days.

So given all this and assuming we somehow have sales records, what value should be used to calculate whether that criminal threshold has been met: price of the infringing or genuine goods?

Chris argues for the latter:

[T]he damage caused by counterfeiters is measurable by the loss of revenues, meaning valuation should be against the original goods and not the counterfeit product.

For sheer entertainment purposes, I'm going to think outside the box on this and not choose either option. (Note: if I had to choose for real, I would agree with Chris and go with the genuine article – but it's no fun being a conformist.)

OK, here goes. You can certainly measure the loss of revenue by multiplying the profit of the genuine article by the number of sales of the fake. Easy calculation, but you still wind up with the same problem I have with those trade associations: you now have an artificially high number of sales that would never have taken place at that price. For these sales figures to be at all realistic, there would have to be no price elasticity on these things at all. Obviously this is not the case or there would be no market for fakes in the first place.

I am not saying that use of genuine product sales prices is somehow unfair. Screw 'em, these are infringers we are talking about. No, I just think we should stop talking about profits based on genuine product sales as if this allows us to get to the "real amount" of damages. That's tough to defend, in my opinion, but it is also not necessary.

The government could just as easily take the actual (fake product) profits and apply whatever multiple they want to arrive at a higher, deemed profit amount for the specific purpose of establishing whether the cops should be involved. Maybe this gets us to the same amount as if we had used the genuine product sales price, maybe not, but at least now we do not have to pretend that the number is based in reality. This is statutory, it can be whatever the government says it is.

Third, the quote Chris gives us from MOC on this issue is indeed bizarre. Essentially MOC is saying that if calculations are based on genuine product losses, then too many cases would meet the thresholds. I would have a lot more sympathy with MOC if every case that met the threshold now was being prosecuted aggressively. This is sadly not the case.

However, if we use a statutorily-defined profit multiple, then the MOC argument would be moot. The government would then need to simply choose a number, or a set of numbers for certain sectors/products (lots of flexibility here). If that choice resulted in a flood of cases that overwhelmed the PSBs and prosecutors across the country, then it could be changed.

[Disclaimer: This idea would require other changes to the law as well. It's not as simple as I make it out to be, which is why no one in their right mind would suggest it. I, happily, do not face that mental limitation. ]