There is a lot of field-dependence here. In my area most people do 2-4 years of postdoc at most, but thats because industry is so lucrative that many leave academia voluntarily. Salaries as a postdoc are low but hardly minimum wage, £30k is a typical number. Permanent staff salaries arent that bad either, a lecturer in their early 30s will make about £40-50k depending on their consultancy opportunities, which obviously isnt great compared to milk-round private sector jobs but is enough to survive on as long as you dont have kids, and would be enough to afford kids if you lived up north. There used to be a good pension on top of that, but there isnt anymore. Publishing is difficult but you get a lot of research time at Russell Groups, its a lot harder at teaching-orientated universities and the post 1992s do have fairly unrealistic requirements about what their staff should be producing giving the high teaching loads there. Administrative crap and red tape is a pain but there is still a lot less of it than in most industry jobs, with the exception of things like tiny tech startups (I sometimes wonder whether the people who complain about red tape have ever worked in industry, because while its certainly got worse in UK academia over the last 10 years, its nowhere near the sort of crap you have to put up with in most large corporate jobs. Big organisations are highly bureaucratic, thats just sadly how the modern world works, and universities are no different).
On the other hand, most of this is field dependent and I would advise people against doing a PhD in areas like the life science (or lab science in general) where hours are very long, and autonomy is very limited due to the need to work on grant-funded projects, or the humanities where there are just no jobs. The general rule is that as long as the field has good industry prospects (economics, business, finance, computer science, applied mathematics, engineering, statistics, etc), academia probably isnt going to be that bad since there is less competition for jobs, and more opportunities for consulting work to top up the low salaries.
This is worth reading (I'm not in economics and he exaggerates how bad some other fields are, but his general argument is correct).