Yeah, it'd necessitate a second dev team to keep the previous versions up-to-date which would bloat the budget.
This is also running on the assumption that Mojang keeps all source code for all versions which may not be the case. If they didn't keep the source code; it'd be a major rewrite which would be a huge time sink for a team that could otherwise be doing other things... OR... they could outsource the job to another company to handle it; which would involve money trading hands for a truly negligable gain. This is for a title that has hit market saturation years ago.
There's an industry term called "ROI" or Return On Investment. It's basically a metric used to determine if the energy you put into an objective is going to yield a positive gain.
It's like a burger joint opening a soup kitchen to feed the homeless. The shop incurs a cost by operating for free; but the hope is that they'll gain publicity and positive PR for their attempt (assuming altruism is the face of ulterior motives). It's like a major publisher purchasing a well-known video game developer who's major selling game is already starting to stagnate in sales in hopes that the publicity will reinforce both brands without it horrifically backfiring.
In this case, if I was speaking for Mojang (I'm not, just being pragmatic here), the ROI would be very costly. They would likely be better off putting a stack of money on the table and lighting it on fire to keep them warm during those Stockholm winters.