Software is expensive to produce, bootleggers or not.
And the subscription model used for the Creative Cloud system isn't about preventing piracy, but rather it is about cash flow and a stable development environment that enables them to rapidly deploy new features and shorten development cycles.
The traditional model of software development is that a developer, publisher, or some other backer with money takes a risk and invests their money up front to fund development for upwards of several years with the hopes that the product can be finished and released to a market that will buy enough copies to not only recoup the investment but also earn a decent profit.
The new model however flips things. We are the investors. We put our money up now with the hopes of strong improvements in in the software down the road. Getting access to their existing software is a rather obvious incentive to encourage this investment and clearly limits the risks involved rather nicely. (After all I'm getting my money's worth in the mean time even if the improvements don't pan out as well as I could hope.)
By collecting fees from a subscription service the company can better manage risk. They have X cash coming in every month, which means they can scale their development cycle to what their actual income is now, not having to try and guess what their income is going to be two or three years down the road. They're freed from having to try and come up with a spiffy feature list that is going to convince non-technical people that paying for an upgrade is worth it, and can focus more on the less 'shiny' details of software development that frequently get left by the wayside in favour of 'features'. (Such as lower level optimization, advanced refactoring of code that makes future development easier/more reliable, etc.)
Shiny new features can be rolled out 'when they're done', meaning something cool that an exceptionally clever intern codes up for the company on his lunch break the week of a "Gold Launch" (final product getting shipped to customers, no new features added within that version) can be rolled into the product within the next month or so, rather than a year and a half later.
We as the end users get cool new tools to work with sooner, and a gradual change. We only have to learn one or two new things every few months to keep up with new development, rather than face radical changes from a 'whole new version', and exceptionally radical changes to the software if we had decided to skip several versions because our current software was 'good enough'.
Canon EOS 7D | EF 28 f/1.8 | EF 85 f/1.8 | EF 70-200 f/4L | EF-S 17-55 | Sigma 150-500