Actually, what you should do, if you want to behave ethically, is purchase legal copies of the books you've stolen.
This answer has generated a lot of controversy. Let me explain a bit of the background and thinking behind it. It will take me more than one edit to be complete, so please be patient. Most of this annex is derived from comments I've made elsewhere here.
First, I don't make any legal argument at all. Everything I've written here is that it is unethical to unilaterally break a social contract, substituting your own terms, taking something that isn't yours and benefitting from it without compensating the producers (both authors and publishers) who have expended resources in its creation. It is an insult to creators. I haven't discussed legality. Others here seem to be trying, like the OP, to find a way to make it sort of ok, when there is an obvious, clean, and simple solution. Purchase a legal copy. Other "solutions" are just self delusion.
The OP and others here seems to believe that the publishing industry is itself immoral and should be combatted. I agree in part with that, but only in part. But that doesn't change the ethics of this action (downloading without payment). But most people who think that publishers just rip them off, haven't thought about the problem very deeply. The main costs in publishing (paper or electronic) are acquisitions, reviewing, editing, layout/graphics, manufacture/hosting, and marketing. This in addition to the time and effort of the author(s) who produce the work. All of these are expensive undertakings and require skilled professionals.
Some of the work is done by volunteers (often reviewing). But copy editors, who improve the language and layout/book-designers etc. need to be paid. Many of the people in acquisitions, editing, and marketing need to be on the road visiting (and paying for) every conference they can find. And the marketers give away a lot of books, also.
Some here seem to think that the costs of books etc are just too high, but they have always been high. I just read that textbooks account for about 1% of the cost of education. A Calculus book (e.g. Stewart) now costs about 10 times what mine did in the early 1960s. But so does everything else. Food, housing, transportation, etc. The kindle edition of Stewart is only about 5-6 times what my hardcover was back then. I remember spending about $100 for most of a year's books and was horrified. Now is is said to be about $900. BTW, I still have that book, so it was a good investment.
Note also that the price charged on successful books includes the amortized cost of creating, but not manufacturing, the ones that never sell. It is hard to predict a winner so publishers create a lot of failed books; several for every successful one. If the price difference between ebooks and hardcovers is an indication, about half the cost is due to manufacturing. Which means that much of the cost of trying to develop most books is never recovered. So publishers absorb those costs initially, but include it in price of books that sell. This is the "cost" of choice that we pay. One model is to charge back the development cost of a failed book to the author. A clear disincentive to write.
Another reason for the high cost of all but elementary books is that the total market is both small and divided up by the presence of several book options. Choice again. If we all used Johnson & Kiokemeister's Calculus with Analytic Geometry from 1960 (a good book) then the cost would be very small. But there are new books to choose from, increasing the cost of all since most are unsuccessful, but still eat up development resources.
Some have stated that publishers have a monopoly and that they exploit it. But publishers don't have a legal monopoly. Anyone can attack their business model. Anyone can offer competition. If their profit margins are outrageous then someone has a lot of incentive to do it. But no one has yet been able to put all the pieces together (from acquisition to distribution) to make it any cheaper for buyers. Some models replace paid employees with volunteers and that works up to a point, but hasn't been shown to scale. Apple's profit margin is 22%. McGraw is 25% (one of the highest). Creating things is hard work from a lot of people.
Don't get the idea here that I'm against a system in which IP is free to use. But there needs to be some incentive to produce it or it won't get produced at all. I've written some ebooks, actually, and have produced software that has been downloaded (free) more than 15,000 times. But that was my choice to do, not someone else's decision who tried to override my wishes.
My preferred solution is to work toward a system in which authors are compensated separately from book sales, say via grants funded via tax revenues. This makes the creation of IP truly a social good and a shared responsibility. Books could then be distributed for free or sold for manufacturing cost, or whatever.
My biggest complaint about the publishing industry, actually, is simply that they don't spend enough effort on marketing their titles after two years. Most authors get almost all of their revenue in those first two years because of this. Always on to the new thing.
But still, my argument here is an ethical one, not a legal or economic argument. It is wrong to substitute your decision for that of the creator of something of value - especially if it is of value to you.
