A Detailed Explanation Of How The BSA Misleads With Piracy Stats

from the and-on-and-on-it-goes dept

A couple months ago, when the Business Software Alliance (BSA) released its latest stats on "piracy," it's VP of anti-piracy, Neil MacBride, gave me a call to discuss my earlier complaints about the organizations methodology. Needless to say, we did not see eye-to-eye, and the phone call did little to resolve our differences. I'm still hopeful that eventually the BSA will recognize that it's doing more damage to its own position by publishing obviously bogus numbers. So, with the organization releasing another bogus stat today, it's time to explain why it's wrong and misleading.

Today's report is an attempt to get the government involved in protecting BSA member companies' business model, by claiming that the US is losing out on $1.7 billion in tax revenue due to "pirated" software. And, of course, it comes with a lovely quote from Mr. MacBride: "The most tragic aspect is that the lost revenues to tech companies and local governments could be supporting thousands of good jobs and much-needed social services in our communities." And the BSA is even so kind as to quantify what that (not really) lost tax revenue could do: "For example, the lost tax revenues to state and local governments -- an estimated $1.7 billion -- would have been enough to build 100 middle schools or 10,831 affordable housing units; hire 24,395 experienced police officers; or purchase 6,335 propane-powered transit buses to reduce greenhouse gas emissions."

Except that this is almost entirely incorrect and it's relatively easy to show why:

1. The report counts every unauthorized piece of software as a lost sale. You have to dig through separate PDFs to find this info, but when you finally get to the methodology it states:

   The software losses are based on the piracy rate and equal the value of software installed not paid for.

   That's a huge, and obviously incorrect assumption. Many of the folks using the software likely would not have paid for it otherwise, or would have used cheaper or open source options instead.
2. The report makes no effort to count the positive impact of unauthorized use of software in leading to future software sales. This is something that even Microsoft has admitted has helped the company grow over time. But according to the BSA's report, this doesn't matter.
3. The report also proudly notes: "Software piracy also has ripple effects in local communities." However, "ripple effects" are easily disproved as double or triple counting the same dollar. Using ripple effects like that inflates the final number by two or three times. In the link here, Tim Lee explains this (in reference to an MPAA study done by IPI, but it applies here to the BSA study done by IDC as well):

   If a foreigner gives me $1, and I turn around and buy an apple from you for a dollar, and then you turn around and buy an orange from another friend for a dollar, we haven't thereby increased our national wealth by $3. At the beginning of the sequence, we have an apple and an orange. At the end, we have an apple, an orange, and a dollar. Difference: one dollar. No matter how many times that dollar changes hands, there's still only one dollar that wasn't there before.
   
   Yet in IPI-land, when a movie studio makes $10 selling a DVD to a Canadian, and then gives $7 to the company that manufactured the DVD and $2 to the guy who shipped it to Canada, society has benefited by $10+$7+$2=$19. Yet some simple math shows that this is nonsense: the studio is $1 richer, the trucker is $2, and the manufacturer is $7. Shockingly enough, that adds up to $10. What each participant cares about is his profits, not his revenues.

   This is a huge fallacy that the BSA an IDC refuse to acknowledge. When I discussed it with them in May, they insisted that they only wanted to talk about piracy rates, not the loss number. I wonder why...
4. Next, if they're going to count ripple effects in one direction, it's only fair to also count them in the other direction. That is, they complain that:

   Lost revenue to technology companies also puts a strain on their ability to invest in new jobs and new technologies. For example, the $11.4 billion in piracy losses to software vendors and service providers in the eight states would have been enough to fund more than 54,000 tech industry jobs.

   But what they don't acknowledge is the ripple effects in the other direction. That is, if (going by their assumption, remember) every company that uses an unauthorized copy of software had to pay for it, that would represent $11.4 billion in money that all of those other companies could not use to fund jobs at those companies. What about all of those jobs?
5. The BSA/IDC stat on lost tax revenue also miscounts on the point above, since it includes the lost income tax revenue from those 54,000 lost jobs, but does not count the equivalent income tax revenue from those other jobs. In fact, in the fine print, the report notes:

   "Employment losses are calculated from revenue losses, and only apply to employment in the IT industry, not IT professionals in end-user organizations. Tax revenue losses are calculated from revenue losses (VAT and corporate income tax) and employment losses (income and social taxes)."

   In other words, the income tax losses only count one side of the equation and totally ignore the lost income tax revenue from the lost jobs on the other side of the equation. Oops.
6. It seems likely that the eventual tax benefits of the unauthorized use of software is most likely to greatly outweigh the lost tax revenue elsewhere. That's because the use of software within industries is a productivity tool that increases overall productivity and output, which would increase taxes beyond just the income taxes of the employees. The study, of course, ignores this point.
7. Worst of all, the report seems to assume that direct software sales are the only business model for the software industry, ignoring plenty of evidence from companies that have adopted business models that embrace free software -- generating billions of dollars for the economy (and in taxes). And that's what this really comes down to. It's a business model issue. If others started adopting these business models as well, there wouldn't be any "losses" at all.

Oh, and just for good measure, the report also falsely claims that: "What many don't realize or don't think about is that when you purchase software, you are actually purchasing a license to use it, not the actual software." That's not exactly true and goes directly against a recent court ruling that said the opposite and goes through a detailed explanation for why a piece of sold software is a sale with restrictions, rather than a license, using previous court precedents.

Most of these points have been made to the BSA and IDC in the past, and both organizations chose not to address them. The fact that they're continuing to use these obviously false numbers and methodology to now push for the government to prop up an obsolete business model should be seen as troubling not just for the dishonesty of it, but for the negative impact it will have on the software industry and our economy as a whole.

144 Comments | Leave a Comment..

Media Uncritically Cites Flawed Piracy Studies, Again

from the balance-needed dept

The Washington Post has a story about the never-ending lobbying campaign for more draconian penalties for copyright infringement. It's extremely one-sided, quoting several industry executives but not a single person on the other side of the copyright debate, who could have pointed out that Congress has passed "tougher penalties" repeatedly in recent years and the results haven't exactly been good. The report also cites the widely discredited series of piracy reports that, as I explained here and here, dramatically overstate the effects of piracy on the copyright holders by double- and triple-counting uncaptured revenues by adding in "ripple effects."

Those previous studies covered the movie and music industries, and both made the same basic error: after they counted some amount of CDs or movie tickets unsold as "lost revenue," they then counted the same lost revenue again by calculating how much less the movie and music industry is paying their suppliers due to those lower revenues. In some cases, they count the same dollar a third or fourth time when those suppliers pay their suppliers less than they would have without piracy. But that's obviously silly. What matters is the total number of dollars earned, not how many times each dollar changes hands. When you strip out these spurious double- and triple-counted revenues, and correct some of the other questionable assumptions I detail in the posts linked to above, you find out that uncaptured revenues from music piracy to the recording industry are around $3.2 billion, not $12.5 billion as the report claims. Similarly, IPI claims the uncaptured revenue for the movie industry is $20.5 billion, but if you strip out the double-counting, you find it's more like $6.1 billion. I pointed these problems out to them after they released the movie study, yet they still made the same flawed assumptions for the music study, and presumably their latest study has the same problems. The Post says the latest study finds that the combined cost of movie, music, and software piracy is $58 billion. I think it's a safe bet that this number, too, is wildly overstated, and that the actual figure is closer to $15 billion. Unfortunately, this reporter apparently didn't have the time to dig into the report in detail, or to pick up the phone and call someone who could offer a critical perspective. Once again, readers will see these inflated numbers reported as fact and assume there must be some rigorous methodology behind it.

Timothy Lee is an expert at the Insight Community. To get insight and analysis from Timothy Lee and other experts on challenges your company faces, click here.

27 Comments | Leave a Comment..