We're quite onside with the idea that there really are things which must be done by government. We're also entirely happy with the thought that there are things which government should encourage, incentivise, aid markets in doing and so on. We do though think that it's important to distinguish here. Between those things which must be directly done and those things which need to be encouraged, could be done better with a bit of intervention and so on.
For example, a decent vaccination campaign will produce "herd immunity." Those who cannot be vaccinated will still be protected as the diseases have no corner of the population in which to hide. Thus we're all in favour of government having a role in making sure that vaccination takes place, that herd immunity is achieved. We could, as the NHS does, simply get government to pay for vaccinating all children. We could also, as in much of the US, simply provide an incentive. Children can only go into the public school system, that one that has already been charged for through taxes, if they are vaccinated. Either works and so we're fine with either.
However, there is that set of things - a rather larger set than most suspect - where government might direct, encourage, foster, but should not be doing. Say, space rockets:
A new research paper by Edgar Zapata, who works at Kennedy Space Center, looks closely at the finances of SpaceX and NASA.
...
Zapata estimates that SpaceX launches cost NASA around $89,000 per kilogram of cargo delivered to the space station. There's no telling what precisely would have come from a cargo spacecraft developed by NASA, but Zapata estimates that it would be $272,000 per kg.
For future commercial crew missions sending astronauts into space, Zapata estimates that it will cost $405 million for a SpaceX Dragon crew deployment of 4 and $654 million for a Boeing Starliner, which is scheduled for its first flight in 2019. That sounds like a lot, and it is, but Zapata estimates that its only 37 to 39 percent of what it would have cost the government.
This is not something limited to rockets of course. We've no doubt at all that there is a government role in both health care and education. But perhaps it should be limited to making sure that it exists, providing a financing method for it, rather than actually being the producer of the services? Or train sets or generating plants or water pipes or....well you get the idea.
A currently fashionable and most pernicious idea is that if there should, righteously, be a role for government then that should be that government is the actual provider and manager of production. Something which we're sure is indeed true at times and mostly not given the areas to which people try to apply this argument.