A simple point on railway nationalisation
One point people bring up when they advocate renationalising railways (or renationalising stuff in general) is that when private companies run something they take a chunk of the total surplus in profit, but if the government were in control, that could go to them. But there's a very basic reason why this isn't the case: opportunity cost. A firm, in doing business, puts capital to use. It uses a mix of physical and human capital and devotes it toward achieving tasks in order, usually, to turn a profit.
The best way to measure the amount of capital tied up in a project is the market's assessment thereof—the firm's market capitalisation—although of course we know that market prices are never perfectly accurate, since they are only on their way to an ever-changing equilibrium, and they may not have got there yet. And what's more, not all the relevant information will always be in the public domain.
For rail franchises—or TOCs (Train Operating Companies), as they seem to call themselves—it is relatively hard to pinpoint the exact amount of capital they are using, as they are usually subdivisions of a larger structure. But suffice to say, running trains involves tying up money on the order of billions, whoever does it (i.e. it includes Directly Operated Railways, the state body that is currently running the East Coast Mainline pretty well). You have to rent the rolling stock (trains), pay the staff, buy the fuel, pay to use the track and so on.
From this capital you get a return. TOC margins average about 4% over the last ten years. The average company got more like 10%. FTSE100 companies seem to enjoy higher returns. Of course, operating profits are not share returns, but they tell more or less the same story. The extra couple dozen billion the government would need to spend on trains could equally be spent on equities or anywhere else for more or less the same risk-adjusted return. The return they got here could be put into trains.
But even doing this makes no sense. If the government returns that couple dozen billion to the population at large, the government can tax the income that the private citizens make on the wealth, at a glance dealing with the problems of governments holding wealth—principally: they are not very good at picking winners. Or they could pay off debt and reduce their repayment costs—since the risk-adjusted return of gilts is priced in just the same way as other assets.
Either way, and whether or not rail re-nationalisation makes sense from any other perspective, it is simply not the case that government, by nationalising rail, could get a bit of extra cash to put into our network.