James Gannon's IP Blog

Last week, representatives from Telus Communications, Rogers Communications and Bell Canada testified in what became the penultimate hearing of the Bill C-32 Parliamentary Committee . The hearing focussed on evidence from the ISPs on the effectiveness of the notice-and-notice system (N&N), which would have been codified in Bill C-32.

Only one of the ISPs, Rogers, provided any sort of tangible numbers related to the effectiveness of their N&N system. According to the testimony, only one third of ISP subscribers who receive a first notice of infringement will receive a second in the same year. However, according to European studies, the following is currently known on how effective (measured by the rate of infringers who are not deterred by notices received) such N&N regimes have been in Europe:

• Where notices do not mention potential sanctions for repeated offenses, the effectiveness rate is around one third (i.e. two thirds of notice recipients will not be deterred).
• Where notices do mention potential sanctions for repeated offenses, the effectiveness rate is around two thirds (i.e. one-third of notice recipients will not be deterred).

Rogers’ testimony last week ran counter to this EU data, since their notices do not include potential sanctions yet claim to be effective two thirds of the time.

According to Rogers’ testimony:

Let me put it in perspective for you. On Rogers customer base of 1.5 million, for arguments sake, Internet customers, about 5% of that customer base will receive a notice. As I said, in 2010 we processed 207,000 notices. Then if I go down the pipe to the next level, of that group, of that 5% of our customer base who are getting a notice, which is in the area of about 70,000 customers, the next number of households to get a second notice will be somewhere in the area of—I don’t know—20,000 notices. So that’s about one-third of those who got the first notice who will actually get a second notice. As I get to the third level, I’m down one-third again. So if I go down the list, the number of notices drops in accordance with the times the notices are received.

Contrary to what the EU data shows, Rogers testified that its N&N system is effective two thirds of the time (only one third of noticees will repeat infringe), even though the notices do not foresee potential sanctions for repeat infringers.

However, a closer look at Rogers’ numbers show some inconsistencies with this assertion.

It was stated that they sent 207,000 notices in 2010 to 70,000 subscribers. They also state that only about one third of those 70,000 who received a first notice, around 20,000, received a second. And of those 20,000, again only a third had received a third notice, and so on.

The problem with these numbers is that if we assume this one-third re-offense rate is accurate, the number of notices sent to the 70,000 infringers doesn’t come close to adding up to 207,000 total notices:

.

.

.

.

.

.

.

.

.

.

.

It’s quite apparent that, starting at an initial “pool” of 70,000 infringers and assuming a one-third rate of repeat infringement, the total number of notices for the year comes far short of 207,000. One of these numbers must be off.

Which metric is then the most likely to be incorrect? The witness appeared quite confident with her assertion of 207,000 notices sent in 2010 (further emphasized by the relative precision of the number). The 70,000 initial infringer number was also backed up by the statement that “about 5% of the customer base [of 1,500,000] will receive a notice” (with 1,500,000 x 5% = 75,000, close to 70,000). However the witness was more uncertain when giving the repeat infringement numbers. According to her, the number of subscribers who receive a second notice is “somewhere in the area of – I don’t know – 20,000 notices”.

Let’s assume that the numbers produced from the EU studies would also apply to Rogers’ N&N system as well – that two thirds rather than one third of notice recipients will re-offend in the absence of potential sanctions. How many notices would Rogers have to send its 70,000 infringers with a two third re-offense rate?

.

.

.

.

.

.

.

.

.

.

.

.

.

.

I stopped at 15 notices, but for the mathematically inclined, the total number converges at 212,121 notices. This number comes surprisingly close to the 207,000 total notices that the ISP claimed to have sent out in 2010.  Given the shaky testimony on the re-offense rate, the statistics that were cited by the witness on the number of notices sent can only be accurate if a re-offense rate of two thirds, rather than one third, is the case. This would also be consistent with the EU studies that show that two thirds of notices that do not include the possibility of future sanctions are ignored.

Of course, the biggest unknown in these numbers is that we cannot assume that 100% of infringers are “caught” under the N&N system. So even if only one third of first-time infringers receive a second notice, it is likely that much more than one third did in fact re-offend. It is also likely that many more than the 5% of Rogers subscribers that received a notice have engaged in illegal file-sharing. The actual percentage of infringers would be hard to measure, although an (admittedly aging) OECD study placed this number in Canada closer to 24%.

What is certain is that despite the ISP witnesses’ testimony, the numbers produced at the hearing last Tuesday back up the recent conclusions from studies in Europe: N&N regimes conditioned with potential sanctions for repeat infringers are at least twice as successful in deterring copyright infringement than those without. Indeed, such regimes in Europe have yet to actually disconnect a single Internet subscriber, while ISPs such as Rogers are testifying that they sent out up to thirty-three notices (!) to certain Canadian households in 2010.