Does it 'cost' anything to nationalise a company?
Labour's manifesto reveals that they intend to re-nationalise water companies if they are elected. How much would this 'cost'? was many people's response. For example, the BBC's Laura Kuenssberg noted that a 'costing' of this policy was omitted from the manifesto.
The nationalisation would happen by the government selling debt to the market, and using the money raised to buy shares [presumably at least a majority, perhaps all] in the company. There are questions about which internationally acredited accounting conventions this borrowing would be included under, and which it would not. But this issue is not very interesting [to me, anyway].
The interesting thing is to think through the economic consequences and 'costs' of the policy.
Supposing that the fair price for the assets were knowable, and that the cost of doing the transaction was small, in principle the company could be sold again at some point in the future, leaving public finances back to square one. The extra liability of the government is matched for a while by an asset; then the transaction is reversed. The 'ifs' there are significant.
The next question is what happens to the revenues that ownership of the firm would, under private management, generate. If the company is managed in a commercial way, with appropriate amounts of revenues ploughed back into the company, there need be no 'cost' from this source. Once again that is a substantial 'if'.
Taking ownership may inject risk into public finances, if the public sector does not already implicitly stand entirely behind the water companies. At some times, when market dysfunction means that risk is not priced properly, taking on risk like this may be a good thing; it's similar to the Bank of England buying private sector assets with QE. At other times, the drain of uncertainty on the public finances will be one that comes without a compensating social benefit.
A stark example of how to maximise the 'cost' of nationalisation is the oil industry of Venezuela, where its output was sold at subsidised or sometimes zero cost, and many jobs treated as posts of patronage, experts let go, and capital expenditure cut. This is not a serious comparison, but it does make a point. There will be political pressure on government owners NOT to manage the assets in a commercial way and to use it as a means of redistribution. As Mike Bird of the WSJ pointed out on Twitter, one has to ask: why nationalise? It's possible, but unlikely, that the reason is to run it on the same commercial terms as before.
It's worth pointing out that interfering in commerce like this could make things better, not worse. If ownership proves simpler than regulation, and rent extraction is reduced, there might be benefits in terms of lower costs that generate offsetting revenues in other parts of the tax base (that use water as an input).
Buying a company like this does not 'cost' in the same way that it costs me to borrow to buy a house. There, I am borrowing to consume the housing services that the house affords me. A better analogy is borrowing to buy a house to let it out. This would consume a lot of my personal fiscal space. But only because the credit risk associated with the capital value of the house, and my rental management skills are large relative to the other income streams that I have available to me to make good the interest on the loan and repay the principle. For the same reason, the government would find it tricky to raise the money to buy Microsoft.
Continuing the analogy, the credit history of the government is pretty good, bar a period after WW2 when there was a hefty amount of inflation and what economists came to term 'financial repression' [taking our savings and charging less than market rates for it]. But the record of managing public companies is somewhat less compelling; markets may therefore conclude that public ownership is a drain on public funds if that history was to repeat itself.
Some on Twitter commented that public ownership might make it easier to realise environmental benefits. If that can be done without affecting the revenue streams, then this also does not need to be seen as a 'cost'. However, if this means placing further restrictions or duties on the water companies that impairs revenue, then the environmental benefits that ensue have to be seen as something that the taxpayer pays for out of future taxes, and today's fiscal 'space'. That is, unless those benefits somehow undo a market failure [flooding? long-term land productivity?] whose resolution commands a price. It may be money well spent, in these terms, but it would be money spent.