At the heart of the argument about having a government at all is the idea that there are some public goods that would be underprovided in a pure market system. We thus need some communal (and at times, involuntarily communal) organisation to tax so as to provide these public goods to us. At every stage closer to sanity than the anarcho-capitalists we agree with one or other of the classic examples. National defence is better organised through taxation: we tried the market system once and ended up calling itn the Wars of the Roses. Public health, in the true sense of public health like vaccination, is a public good and there's excellent reason to think that taxing everyone to make sure that everyone is vaccinated is a good idea. The reason being that herd immunity that we get from a lare enough portion of the population being vaccinated.
However, that there are good arguments in favour of the existence of government at all does not mean that those same arguments support having any level of government at all. Which brings us to a fascinating paper looking at the relationship between big government and the provision of those public goods:
Theories explaining government size and its consequences are of two varieties. The first portrays government as a provider of public goods and a corrector of externalities. The second associates larger governments with bureaucratic inefficiency and special-interest-group influence. What distinguishes these alternatives is that only in the former is governmental expansion generally associated with an increase in social welfare. In the latter, the link between government size and public goods provision (or social welfare) is negative. We study the empirical significance of these competing claims by examining the relationship between government size and a particular public good, namely environmental quality (notably, air quality measured by SO2 concentrations), for 42 countries over the period 1971–1996. We find that the relationship is negative, even after accounting for the quality of government (quality of bureaucracy and the level of corruption). This result may not prove conclusively that the growth of government has been driven by factors other than concern for the public good, but it creates a presumption against the theory of government size that emphasizes public good provision.
I'm still perfectly willing to agree that there are public goods. And that there are public goods that only government can provide. But as above, this does not mean that this justifies any amount of government. For past a certain size the government turns out to be not very good at provision of those very public goods that are the justification for it in the first place. Quite why we can argue about: my theory would be that when government tries to micromanage inequality say, or concerns itself with the voluntary activities of consenting adults, then it's taking its attention away from what it actually exists to do, provide us communally with those things that cannot be, or only will badly be, supplied by individual action.
I'd thus suggest the Worstall Measurement of government. There are things that must be done, there are things that must be done that only government can do. So let's limit government to only those things that must be done by government: where it is both necessary that they be done and also that the coercion of either the law or taxation is necessary for them to be done. Everything else we'll get on with ourselves: you know, as those free people we are?