In the days of vinyl records, the record companies both recorded and manufactured the records. Both these steps were difficult, and there were no IP problems, as anyone attempting to set up a record manufacturing plant to make "pirate" records would be easily found.
That security was weakened by cassette tapes and destroyed by MP3 players. Now anyone can "manufacture" copies of music recordings, on a large scale or a small scale.
The pharmaceutical industry resembles vinyl records. The largest costs for the drug companies are in design and testing, but the manufacturing costs are high enough to protect their patents.
Imagine that a "generic synthesiser" were developed. I have in mind a general-purpose programmable chemical plant. If you want to produce, say, asprin, you put in some basic feedstocks, feed it a program, it churns away like a bread machine, and out comes your asprin. (or cocaine, or whatever...)
Is this feasable? I would say it's inevitable, though I couldn't say whether it would be closer to 2015 or 2100. Twenty years after appearing in the laboratories, it will be in your kitchen.
Once that process gets going, pharmaceuticals becomes a "pure IP" business like software or music. Development and testing (and marketing) of new drugs will still be expensive, but once a drug is on the market, it only takes one "hacker" to write the program and I can download it and synthesise the drug in my kitchen. (Or, in earlier stages, in my University Chem lab).
Think about the effects of this device: It will revolutionise medicine. It will improve many other areas of life, opening up possibilities that are hard to imagine. But it will make drug development very difficult to fund, but technically easier to do, and it will make narcotics prohibition impossible.
Attempts will be made to restrict the distribution or capabilities of the devices, but without a mass market, it will not be developed quickly enough. It is likely to be built of components that will be used throughout manufacturing industry, and the only way to restrict capabilities will be at the "programmer" component, which can be replicated with general-purpose computing equipment.
Of course by that time, the software/music/film issues will be worked out — either with a police state like this, or with a new model of development — so we will have more clues to the solution of the new problem (which, it is to be remembered is not primarily a problem, but a new world of opportunities).