- The government can almost always raise large amounts of money more cheaply than the private sector (because the government is in most cases lower risk than private firms, it pays a lower interest rate on borrowings);
- The government is almost always worse at running things than the private sector (think about the difference in service quality in particular); and
- Whenever the good has a large externality, or is a public or merit good, the government is almost always going to have to be involved anyway.
I like to add to that last point that, with quasi-rational decision makers who heavily discount the future (present bias), then the government is likely to need to be involved in provision of goods and services that have long-term payoffs (e.g. education).
The elephant in the room in this discussion of course is privatisation, which we didn't touch on (surprisingly - in most semesters at least one student would bring it up). Coyle's three principles would seem to argue for privatisation. In the case of infrastructure for example, the principles suggest that the government can fund the development of the infrastructure (which can be paid for with lower interest costs), then privatise to avoid much of the problems of the second principle (except in cases where there isn't a strong externality or public or merit good argument for continued government provision).
Which is why I really liked this Christopher Niesche piece in the New Zealand Herald earlier this month, which makes a strong case against privatisation based on comments from Rod Sims (chair of the Australian Competition and Consumer Commission):
For many years politicians and business people have told us privatisations are good.
Selling government-owned assets to the private sector makes them more efficient, to the benefit of us all, they have said.
It was a view shared by Rod Sims, chair of the Australian Competition & Consumer Commission, but he's now on the verge of changing his mind, saying privatisations are pushing up prices for consumers and damaging the economy...
He now argues that, instead of privatising assets such as ports, airports and power infrastructure to boost economic efficiency, Governments are just trying to maximise their profits. They do this by selling monopolies to the private sector without enough regulation to rein in excessive price hikes, thereby making the sale price higher...
But left-wing fogeys shouldn't get too carried away by Sims' comments. He is still in favour of the "theory" of privatisations: that they generally increase economic efficiency and bring down prices for consumers.
He points to the privatisations of Telecom (now Telstra) and Qantas. We do indeed have cheaper phone services and air travel but this is largely because Telstra and Qantas were privatised about the same time their industries were deregulated, introducing the disciplines of competition into the equation as well.
Privatising a monopoly is different altogether.Indeed. If the argument for privatisation is based (in large part) on the private sector being better at running things than the government, it is hard to make a case where creating a monopoly is necessarily better. For instance, we know that monopolies are X-inefficient - relative to firms in more competitive markets, monopolies have less incentive to innovate or improve customer service, leading to relatively worse quality of products and service over time (compared with a firm in a more competitive environment). Government-owned monopolies are X-inefficient as well of course, but you don't solve that problem by privatisation, you solve it by increasing competition.
Privatising a monopoly is very attractive to investors - monopolies can make big profits, which can mean good returns on investment. But simply being able to extract money from privatisation doesn't mean that privatisation is always the right approach for government to take.