FORTUNE — Deciding how best to combat infringement of intellectual property — or even, perhaps, whether we should combat it at all — is made difficult by the fact that nobody really knows the extent of the harms it causes.
The Rand Corporation believes it might have the beginnings of a reliable methodology in the works, but even its creators are highly uncertain about its prospects, mainly because success would depend on the willingness of companies to share proprietary sales data and forecasts.
One side of the digital-piracy debate claims that the problem is existential. The other side claims it’s non-existent. The media industry issues “studies” that are often laughably skewed to back up claims that piracy is killing their businesses. The industry’s critics claim (based on not much real data) that piracy not only doesn’t do much if any harm, but that it even sometimes helps to sell legitimate copies of movies, music, and other digital products.
Logic tells us that if a person decides to download an illicit copy of a movie when he or she would otherwise have paid for it, the movie business loses money. But how many people download illicit copies only because they’re free, and would never have paid for it? We have no way of knowing.
Rand, at the behest of the European Union, has created a basic framework for estimating losses that would take such variables into account. Viewing the market from the supply side, it would compare sales projections to actual sales, controlling for all the various non-piracy-related variables — the economy, bad marketing, supply-chain disruptions and the like — that could cause a difference between the two. Theoretically, the result would indicate the actual amount of sales lost to piracy, which could then be extrapolated to determine, for example, how many jobs are lost.
Theoretically. The best Rand can say is that the difference would be “due at least in part to [intellectual property rights] infringements.” In how big a part? Still unknown, and probably unknowable with certainty. Rand says “the statistical model then attempts to identify the portion of unexplained unfulfilled demand that is highly correlated with factors that drive … infringements of a particular product in a particular country. These factors may include: the rule of law, control of corruption, level of tourism, access to broadband Internet or government effectiveness.”
All very difficult to quantify. The approach, however, is certainly much better than most existing examinations of the problem. Media-financed studies often take a ridiculously blunt approach, essentially counting up the number of illicit downloads, multiplying that by the prices of songs, movie tickets, or rentals, and saying “here’s how much we’ve lost.” A model like Rand’s almost couldn’t help but be better than that.
But as Rand notes, large swaths of the industry would have to open its books to researchers, and it’s far from certain that they would. The model is also meant to measure counterfeit physical goods, so makers of fashion, luggage, etc., would also have to take part. The model was tested using data from a single “multinational technology firm producing consumer goods targeted by counterfeiters,” and Rand says the results were promising.
But other firms are “extremely reluctant” to share data. Partly, that’s because they understandably don’t want their proprietary data to get into the hands of competitors. But another concern is that “firms may try to manipulate their forecasting error data before submitting them to be included in our model so as to influence estimates of the size of the market.”
In other words, we would still, to some degree, be relying on the media industry for estimates of how badly the media industry is hurt by privacy. And so far, that hasn’t worked out so well.