Yes, of course it would change Monsanto's incentives. In order for their crops to remain proprietary without a patent monopoly they'd have to invent new crops which don't produce seeds and are painstakingly difficult to reverse engineer and reproduce by their competitors. That way they wouldn't need patent protection to protect their proprietary crops. They would possess a natural monopoly until the competition caught up with them, which ideally would give them enough time to recoup the R&D investment and profit in the process.
I don't know if inventing such a thing as hard to copy GM crops is even possible in the GMO field (though generally speaking it is possible in other technology fields), but it seems to me that creating and monopolizing a proprietary crop should require that kind of ingenuity to offset the enormously unfair competitive advantage that possessing such a monopoly gives to the proprietor at the expense of the rest of the economy.
Now assuming for the sake of argument that inventing hard to copy GM crops is too difficult, or too risky, or even simply impossible, that doesn't mean new research on GM crops wouldn't get done in the absence of patent protection, it just means it would no longer be proprietary. There are many possible non-proprietary funding paths. Competing companies could collaborate on open source GM crops as is often done in the software world, governments could subsidize new research, private charities could fund new research, etc.
Such reform would force Monsanto and its competitors to compete on the merits of their manufacturing capabilities rather than their IP monopolies, much to the economy's benefit.
As a side benefit, if R&D of GMOs shifted more towards an open source model, I think the fact that it would be subject to the scrutiny of different contributors with different agendas would effectively end the controversies surrounding these much-maligned companies and probably do much to assuage public panic and ignorance about the science of GMOs.
