I have written in the past that I have no idea how large the state should be, so this issue has no bearing on my views about austerity or fiscal stimulus. In this post I explain why I have no idea. My argument is that the optimal private/public split will depend on a number of particular and highly contextual issues, about which economics will have a lot to say but where it is unlikely to generally point in one direction. It seems worth making this - I hope uncontroversial - point when the current UK government seems keen to privatise or outsource by one means or another so much of the public sector.
There are two claims about government often associated with those who argue in favour of privatisation. One is that markets provide a better allocation system (e.g. here, and follow-ups here and here). For many activities this is undoubtedly true, particularly where markets involve a large number of buyers and sellers, and information problems are small. However much public sector activity is in areas where market imperfections and informational problems of various kinds are endemic. In that situation, market based systems may perform worse than alternatives: a comparison of health systems in the UK and US is an obvious example (e.g. here).
The simple fact that large corporations exist could illustrate potential problems with markets as an allocation mechanism. Large corporations may exist because they find it more efficient to organise activity in a non-market (often hierarchical) manner, or they may exist because markets can be circumvented by rent seeking, or they may exist because of increasing returns. This means markets can be fragile or inefficient. Economics is not a discipline that tells us market allocation is always best, but one that tells us when it may work well and when it may not.
In reality privatisation or contracting out often involves relocating some activity from government bureaucracies to corporate bureaucracies. George Monbiot argues that we are seeing power gradually shift from the former to the latter. The second general argument in favour of this shift is that the profit motive provides an effective incentive system for ensuring efficiency. Yet this argument alone is not enough. It is perfectly possible to run parts of government like a company, where the explicit aim is to maximise profits. Take the East Coast mainline rail company in the UK, for example. It has been running as a publically owned company since 2009, after the private company running the franchise got into difficulties. It appears to have been run very successfully under public ownership, but the government wants to return it to the private sector. The motive does not appear to be pressure from the public or customers.
The complete argument for preferring private sector rather than public sector bureaucracies is that shareholders are better at ensuring managers maximise profits (and therefore efficiency) than politicians. In fact the argument has to be stronger than this: the gain in profitability has to exceed the cost of diverting some of those profits from the public to shareholders. Again I think in many cases this will be true, particularly if checks within government are poor and corruption widespread.
However, control by shareholders may be far from ideal, as the recent debate in the UK about the proposed takeover of AstraZeneca by Pfizer has illustrated: see this article by Martin Wolf for example. One argument that has been made in this debate is that shareholders may be unable to prevent managers focusing too much on the short term, because that enables them to expropriate a share of the surplus in excess of their marginal product. Shareholders’ ability to directly control managers appears weak, and the takeover mechanism itself seems highly ineffective at punishing inefficiency. Martin Wolf argues that the real purpose of shareholders is to provide insurance against unexpected shocks that could otherwise lead to bankruptcy.
The process of tendering for the outsourcing of services paid for by the public sector allows periodic pressures for efficiency. However often the contracts involved are very complex, and so they also provide an opportunity for firms to exploit the inexperience of the civil servants who represent the public. Finally if we want those undertaking public activities to be accountable as well as efficient, then it is not obvious that outsourcing is helpful.
What this all suggests to me is that the costs and benefits of privatisation will vary from case to case, and that this is an area where microeconomic analysis will be central. (For example, see this book by Massimo Florio which looked at the results of the Thatcher privatisations.) In such cases, an ideology that says that the private sector is always better (or worse) is not only unhelpful but dangerous. For example, an ideology that says that private sector provision is always better can be exploited by rent seeking firms. It can lead governments to privatise on unfavourable (to the public) terms, or with inadequate mechanisms in place to ensure value for money and prevent exploitation. At worst, rent seeking firms may be able to exert sufficient control over the political process to make this happen.
Given that my area of expertise lies elsewhere, you might have expected me to consider a macro argument that is sometimes made for privatisation, which is that it can help bring down public debt. This is either a terrible argument all the time, or just most of the time, but which will have to wait for a later post.