Prof. James Gibson, University of Richmond School of Law
April 12, 2011
In copyright class, professors usually spend most of their time explaining the “public law” aspects of copyright – the exclusive rights that the law gives copyright holders (e.g., reproduction and public performance) and the privileges that the law gives to those who use copyrighted goods (e.g., fair use and first sale). But as they and their students know, many everyday encounters with copyrighted goods are governed not by this public law, but by the “private law” that sellers and buyers create through contracts.
Software provides the best example. If you somehow managed to legally purchase and install a computer program without agreeing to any contract, you would have certain fair use and first sale privileges with regard to that program. But in reality, that almost never happens. Instead, any program that you install on your computer these days requires agreement to a long list of contractual terms, all offered on a take-it-or-leave-it basis. And those private terms often eliminate the user privileges that copyright’s public law provides.
In theory, there is nothing wrong with this dynamic. If copyright holders and copyright consumers can strike a better private balance between rights and privileges than copyright’s public balance, then more power to ’em. Why would we assume that Congress or the courts have a better idea of what matters and what doesn’t than consumers themselves do? We don’t do that when it comes to setting the price of a product, or deciding what features a product should have. We leave that to the market.
In short, why treat contracts any differently from product price and product features? If you think the computer’s price is too high, don’t buy it. If you don’t like the number of USB ports a computer has, don’t buy it. And if you don’t like the fact that a computer’s software contract requires you to surrender your first sale privilege, well, don’t buy it. The market will punish those sellers that charge exorbitant prices, fail to supply need features, or impose onerous terms.
Indeed, seen from this perspective, take-it-or-leave-it contracts make copyrighted software even more feature-rich. Otherwise-identical programs can distinguish themselves from one another by offering different contract terms, giving consumers more purchasing options. Providing more options to consumers is generally considered a good thing.
Why is it, then, that consumers universally regard these contracts with such distaste? Perhaps it’s because as consumer options increase, so does the cost to consumers of evaluating the products they want to buy. (Property scholars call this a “numerus clausus” problem, although in this case “numerus clickus” might be more accurate.) And contracts may impose particularly high evaluation costs because of how long they are and how hard they are to understand.
To get a rough sense of exactly how hard, I decided to purchase a computer and keep track of the volume of contractual terms to which I became bound in the course of the transaction. I counted only those contracts that a court would readily enforce, because those are the only contracts that can reliably be considered equivalent to price and other product features.
Here are the preliminary results. Between ordering the computer online, starting it up when it arrived at my door, and opening the programs that came with the computer, I became bound to 29 contracts totaling 78,203 words. To put that in perspective, that’s slightly more than the number of words in the first Harry Potter book. All but four of the contracts were presented to me only after the computer was delivered; in terms of word count, this meant that at the time I tendered payment, I had had no opportunity to express my preferences regarding seven out of every eight words to which I became contractually bound.
One way of looking at this data is to estimate how long it would take a consumer to read all the contracts. Studies of contract reading rates have shown that a fair estimate is 177.5 words per minute (and comprehension was sketchy even at that rate). That would mean approximately 7 hours and 20 minutes to read all 29 contracts.
Another way to look at the data is by dollar amount. The computer cost $922.94. That’s 84 words per dollar spent. Imagine having to read an 84-word contract each time you buy a soft drink or a third of a gallon of gas. (And for what it’s worth, my impression is that a much cheaper computer would come with exactly the same contracts; none of them seemed related to the fact that I bought a slightly more expensive model.)
Of course, words-per-dollar is a valid comparison only if contract terms are not like price terms and other product features. Otherwise, the comparison is like dividing the number of USB ports by the size of the hard drive; it does not compute.
But to come full circle, this distinction is exactly what I’m trying to get at. Contract terms may be different in kind from price and other product features. Price is almost always immediately accessible and easily understood. Contract terms, on the other hand, take a long time to read and may be very difficult for a consumer to understand even if he or she takes the time to read them. So how confident can we be that the invisible hand of the marketplace is adequately regulating the contract terms?
As I said, these results are preliminary, and are based only on a single computer purchase. The computer was a Dell, and it might be the case that Dell’s computers come with more contracts than, say, Gateway’s or HP’s or Apple’s. But if Dell is in fact representative, one must wonder whether these kinds of private copyright arrangements are in fact superior to the public law that copyright would otherwise provide. If not, then what does that tell us about enforcement of software contracts?