For the same reason researchers choose to do so now: the government pays for a lot more fundamental research than private industry does.
Why ask that question? Because, quite interestingly, without the value of basic research being captured as revenue with a market price, we don't have numbers to perform a direct comparison between research costs and research revenues to compute the profitability of basic research. We don't actually know right now whether basic research is financially self-sustaining - in which case the apparent "expense" is actually a state subsidy to technological industries - or whether basic research is a financial net-negative. We also don't actually know what lengths of time are necessary to realize the final economic result of basic research.
So we kinda sorta feel pretty sure that basic research eventually helps the economy, but nobody's accounting ledgers actually get forced to reflect that or pay for it.
But putting that aside, if the government owns the basic research that it funds, why would any (self-interested) researcher or institution choose to work for the government vs. private industry, which can offer a share of future revenue? How does the government set prices for its licensing? Don't we benefit that Jobs and Wozniak didn't have to have a large amount of capital to invest in licensing in order to take advantage of all that basic research?