Software Piracy Stats: Not Entitled to Their Own Facts

0 Comment

A new report on software piracy commissioned by the State Intellectual Property Office (SIPO) was recently released. I’m not sure why SIPO is involved in this area — despite its name, SIPO mostly just oversees the Patent Office and related enforcement — but that’s another issue.

The report, which was researched and drafted by a consulting firm, came out with the following general results for 2009:

About 45 percent of the software installed on computers in China last year was pirated, down by 2 percentage points from the previous year.

“China’s software piracy rate has been continuously declining for four years. It is approaching that of the developed countries,” Ye Xiumin, a senior consultant with the company who worked on the project, said at Monday’s press conference.

According to Ye, the rates have been dropping over the past four years: 66 percent in 2005, 63 percent in 2006, 56 percent in 2007 and 47 percent in 2008. (China Daily)

That’s a damn fine success rate, yes? From 66 percent in 2005 to 47 percent in 2009, wow — and ten percent of that came in one year.

Well, it turns out that there is some disagreement over these statistics. The Business Software Alliance, whose recently-issued global piracy report also has China numbers for the past few years, sees the same trend but at a much-higher piracy level.

By the way, I recommend reading the BSA report (free on their site), which has some very interesting information on industry trends (e.g. cloud computing) and how they might effect piracy rates in the future.

BSA’s report includes the following software piracy rates for China:

  • 2005 — 86%
  • 2006 — 82%
  • 2007 — 82%
  • 2008 — 80%
  • 2009 — 79%

Again we see a positive trend over the past few years. However, the BSA numbers start with a much higher rate (86 compared with 66). Additionally, the drop in the SIPO report of roughly 20 points is much higher than the fairly small seven point drop in the BSA numbers.

As to the different data sets, there is of course disagreement on proper methodology and definitions. The authors of the SIPO report said this of BSA’s numbers:

Their report did not provide the details on raw data and the collecting methods. Plus, based on their descriptions, the data only applies to software used on PCs, but not the general ‘software piracy rates[.]

I’m not sure how transparent either side was with their methodology, but I don’t really trust anyone. I’ve been vocal in my criticism of such industry group studies in the past. Some groups routinely skew their methodology to come up with the highest numbers possible (e.g. by using unreasonably-high retail prices to calculate damages), which are then used to fund raise and prod home country governments to act on their behalf.

I don’t really know anything about this consulting group hired by SIPO. Let’s just say that they also have a vested interest in showing that things are improving.

Here is what China Daily reported about the SIPO report’s methodology:

The research, which was conducted through sample surveys, collected 2,803 samples of individual users and 1,834 samples of business users through computer-assisted telephone interviews.

Already there is some questionable stuff here. What we’re talking about is in-person and phone surveys. At least one of the questions must be something like “Are you using illegal software?” I don’t know about you, but if I received an automated phone call asking me that question, I would either immediately hang up or answer “No.”

If the survey was conducted in person, I would be even less likely to admit wrongdoing.

In other words, I’m kind of wondering about the accuracy of these surveys.

So I’m not going to trust these numbers. Seems like there is way too much leeway for either side to put their thumb on the scale. No one is entitled to their own facts (thank you, Daniel Patrick Moynihan), and the only area of agreement seems to be that things are getting better. Let’s just go with that, then.

One response on “Software Piracy Stats: Not Entitled to Their Own Facts