I think that maintaining an strong incentive to participate in the labor force is crucial to pursuing a unified political system. This may be somewhat tangential, but one of the main factors economists and migration scholars have identified in determining the success of integration of immigrant populations has been labor force participation. Here is one free article on the subject, and here is one paywalled article--you can read the abstract I think it is important to consider how government services can increase the incentives to work. Harris Rosen's philanthropic project in Orlando, FL is a great example. All he did was promise to pay college tuition for any high school graduates of Tangelo Park, and he offered to cover child care costs for young, single mothers who were in school. That's a very simple program, and it has had astounding effects: nearly 100% high school graduation, most go on to college, quadrupled property values, and plummeting crime rates, all for only $11 million. That's really cheap for such a turn around. For me, ideas like this lead less to focusing effort on welfare, and more to effort on providing and incentivizing education and making sure people get paid well. Both of these are why I tend to more support the Bernie Sanders line of reforms (free education and guaranteeing decent wages) rather than welfare. I know this doesn't really address your prompt that well, but I just thought I'd share the bit of thinking I've done on the topic. EDIT: I screwed up the links.
One of the primary issues that I relate to a minimum income is the rise in automated jobs. When employers find it cheaper to automate most labor, which will happen, productivity will continue to rise while available jobs will continue to fall. This brings up a real issue, in that some people will not just be unemployed but unemployable. Do you think that a basic income will eventually become necessary, due to a lack of available work?
I actually ask the question with that in mind. I have an interest in economics and see employment ending up as mostly automated. Technology is just moving faster than ever, and we've automated so many of the jobs that are out there already. In fact, I'm typing this to you while I'm at work since I automated my job and have time now. I did not tell my boss. Even if we were to automate everything in the world, there will still be a few jobs left. I don't know if we would be able to automate judges for example. But maybe we will. Either way, by the end of the transition there won't be enough jobs to sustain everyone. And in actuality, that will happen sometime during the transition. It would even make sense to wonder if we would recognize the point if we saw it. Perhaps it is next year? I doubt it, but I don't know. We also have to decide what sustaining someone means. In the case of a mincome, there will still be wealth inequality. If we all started making the same amount of money tomorrow, Bill Gates is still rich. Most home owners would be the wealthiest as they would actually have a piece of land. Governments would essentially have to be paid by the federal government as they won't be generating any revenue from pay checks, and sales taxes will be less due to lowered sales. That would hold them hostage to a higher powers' whims (see Greece) regarding spending. To me, it would feel like going back to live with your parents if they were paying all of the bills. So if the powerful have more power because they are also now feeding you, there is a dystopian figure hiding in the shadows of mincome. Could be nothing in practice, but I don't know. However, at this point I think that there are still significant opportunities to offer employment to those who will work to attain it. With the rapid onset of new industries and not just the mechanization of old ones, there is a lot of transitional unemployment as laborers move from one industry to the next. We see that as workers unemployed by the auto industry have moved into renewable energy (just an example) and unemployment bottomed out again. As there is sufficient employment at the time, it would not make fiscal sense to offer someone more money to not work. This would discourage workers, discourage gains in efficiency for the capital owner for a period while inflation crept up, cost a lot, cause businesses to relocate, etc. But we have to start talking about this now before we need to make decisions about it. It's a very interesting topic.
I think we tend to overestimate the capacity of people to retrain for new jobs. Sure, in the example you brought up (auto to renewable energy), people moved from a manual labor job to a manual labor job, but what happens when we run out of manual labor jobs? What happens when we automate white-collar work? Will accountants, for instance, need to retrain to become programmers? Will manual laborers? I think that we'll eventually reach a point where we've made entire types of jobs obsolete, and thus made people who are only capable of being competitive in that field obsolete as well. Transitional labor, I think, is a deferment of the problem, not a solution.
Really a lot of white collar jobs are already automated. The example you use with Accounting is mostly capable of being run by much fewer people than in the past. As an example, many businesses used to keep a payroll department as a separate functional department within accounting. Now, if a businesses even handles their own payroll instead of outsourcing it to ADP or a similar company, they have many fewer people doing it. And ADP is able to handle with their small staff what could only be handled by a dozen people for a single company before. So you could argue that we're already well into the phase of automating white collar jobs as well.