Super-funds and privatised infrastructure.


The superannuation industry is getting very excited about the prospect of a ‘wave’ of privatised infrastructure creating low risk assets for fund investment. NSW looks like it will privatise some generation assets as well as Port Botany and Port Kembla.  The super industry wants more and appears to have union backing to turn state-owned enterprises (SOEs) into ‘mum and dad owned’ enterprises via super fund investments. Thus:

Signalling a shift in thinking on asset ownership, the ACTU has backed the campaign on the grounds that “social privatisations” can transfer government assets to the community using super funds that represent “mum and dad” owners.

There strike me as two issues here:

First, the issues of asset ownership and regulation are inseparable. Rohan Pitchford from ANU and I have done some research on this over the years. Regardless of ownership, key infrastructure assets often need regulation. But the underlying incentives that drive a privatised firm differ from those that drive a SOE. So privatisation is only desirable if the regulated privatised enterprise will operate in a way that raises economic surplus compared to a regulated SOE. Put simply, there is no point privatising something if overall it makes us worse off.

For example, consider emergency services provision (ambulance, fire service, etc). A for-profit service will have less incentive to keep ‘spare capacity’ than a government-owned service. Spare capacity is costly and the private owners pay that cost. But from a social perspective we may want the spare capacity as insurance against rare but calamitous events. Public servants and their political masters have strong incentives to avoid and respond to these types of events. So there is a trade off between a privatised service that may ‘under weight’ extreme events and a public service that may ‘over weight’ extreme events.

There are a number of factors that favour privatisation. For example, if there is already competition and private provision is working (not perfect, but working, as in the case of electricity generation) then privatisation is probably a good idea. If government interference is a problem then privatisation makes that interference harder (or at least improves transparency). Of course, there are factors pushing in the opposite direction particularly where cost minimising private owners create negative externalities for the rest of us.

So privatisation is not a simple solution – it just changes the problem. And having Mum and Dad owners via superannuation funds does not change the principle of profit maximisation that should drive the privatised firm.

Second, while privatised assets usually provide a stable revenue stream, privatisation crystallises some risks. This is well reflected in Kenneth Davidson’s article here. Davidson argues:

To reveal what is at stake, assume that, as state monopolies, these assets earn 5 per cent on capital for the government and would be expected to earn 10 per cent on capital for private investors.

This means that the assets worth $100 billion on the government’s books, would only be worth $50 billion to the private operators if the prices for the services remained the same. The government might get $100 billion for the assets if it allowed the new owners to double prices for the services.

This argument is really only a starting point. The real question is ‘why does the return to a privatised business have to be higher?’

Privatisation transfers risk from the tax payer to the private owners. The government can borrow at low rates because it has taxing powers. If a SOE gets into financial trouble then the government can fund it via tax income. The private owners do not have this option. So the increased return required of a privatised asset partly reflects a risk that was hidden and being borne by tax payers. Privatisation improves the transparency of the risks.

Further, a privatised asset faces sovereign risk. The SOE also faces sovereign risk but as the state owns the asset it does not need to be compensated for its own actions! Of course, sovereign risk may not be bad for society. State intervention may lower private profits but raise welfare. However, part of the increased return to a private owner reflects the sovereign risk being ‘crystallised’ by the privatisation process.

Finally, for regulated private assets, the allowed return is often set through a regulatory process. In the twenty or so years of Australian experience we have learnt a lot about regulating private utilities and the types of risk they should face and be compensated for. These are reflected in the regulated company’s allowed return. So, a simple comparison with the government borrowing rate, as Davidson does, is not very useful.

Overall, the superannuation industry needs to be careful what it wishes for. Privatisation comes with lots of regulatory strings attached. And simply privatising assets to create something for the funds to invest in would be bad public policy. But there are clearly some SOEs that should be privatised – and the NSW electricity generation assets are a good place to start.


5 Responses to "Super-funds and privatised infrastructure."
  1. It may be worse than that – the unions may have a hidden agenda here. They can see industry superannuation funds being pressured to invest in their own privatised industries rather than seeking maximum returns for members.

    IOW industry policy by (compulsory) contributor subsidy rather than by those distressingly transparent taxpayer subsidies.

  2. “This means that the assets worth $100 billion on the government’s books, would only be worth $50 billion to the private operators if the prices for the services remained the same. The government might get $100 billion for the assets if it allowed the new owners to double prices for the services.”

    I don’t really understand the significance statement, perhaps someone could explain it to me?

    Regardless of the book value that the government carries for the asset on its balance sheet ($100m in the above example), if the value of the asset on the private market is $50b, then that is its true market value, right? All else equal, what value is there in government retaining an asset on its balance sheet at an inflated value that does not reflect its true market value? I don’t understand how they gain from this.

    Also, I was under the impression that the main driver of privatisation at the moment is the huge balance sheet constraints that governments are facing rendering them unable to adequately invest in sorely needed new infrastructure assets? If they can recycle the capital they currently have tied up in assets that could be easily owned and operated by the private sector (e.g. power generation assets) and put it into new infrastructure then wouldn’t it be a net gain to society for them to do this?

    Any help regarding these questions would be appreciated.

  3. UQ Student
    Good points.

    On the ‘value’ bit, it comes from a difference in the interest rates. Suppose an asset was going to pay $5m per year forever. Then if the interest rate is 5% the Net Present Value (NPV) of the asset is $(5/0.5)m = $100m. If the interest rate is 10% then the NPV of the asset is $(5/0.10)m = $50m. Davidson’s argument is that as private sector borrowing costs are higher than government borrowing costs, the asset is ‘worth more’ in government ownership than private ownership.

    But this raises the question of why the private sector borrowing costs are higher? If they reflect real costs that are hidden by government ownership then the government ‘value’ is wrong (your point). Davidson (I assume) believes that they reflect private sector distortions and capital market irrationalities – but he hasn’t said that. He implicitly argues the government value is the ‘right’ one. As an aside, Profs Grant and Quiggin at UQ Economics have looked at this in some detail.

    So in brief, ‘true value’ depends on whether you believe private capital markets are efficient or not.

    On the government constraint – that is self imposed at the federal level. The federal government wants to run a surplus and so is acting like Scrooge. But Davidson correctly notes that government borrowing costs in Australia are extremely low. There has never been a better time for our government to issue debt and invest in infrastructure (e.g. roads, public transport, water infrastructure, etc). It is a political constraint not an economic constraint that stops this. And it is hurting Australia (particularly the east coast which is the slow lane of the two-speed economy).

    So the federal government does not need privatisation revenue to fund infrastructure investments. It just needs to get over its non-economic surplus obsession.

  4. However, part of the increased return to a private owner reflects the sovereign risk being ‘crystallised’ by the privatisation process.

    Is it being ‘crystallised’ or is it being created by the privatisation process? Does the sovereign risk exist at all when the asset is in public hands?

    If the latter is the case then surely it follows that the creation of the sovereign risk around the asset represents the destruction of value during the privatisation process.

  5. kme. Good question. I think it depends on what you think the sovereign risk relates to.

    Suppose politicians always act in the interest of maximizing economic surplus. Then any ‘sovereign risk’ for a privatised firm is actually a social gain – the politicians may act in a way that reduces the privatised firm’s value but raises social welfare. In that situation the first best would be to compensate the privatised firm when the political interference occurs (and then there would be no increase in the cost of capital to the private firm before this). If this can’t happen for some reason then the sovereign risk will be built into the privatised firm’s cost of capital and this is inefficient (the cost is spread over the customers and distorts their behaviour in every period).

    The alternative is to assume that politicians act in ways that may maximize their own welfare but harm society’s welfare (e.g. force prices below cost to curry favour with voters). In that situation, the politicians can reduce social welfare for either a public or privatised firm. But the cost of this is hidden for a public firm while it is reflected in the cost of capital for a privatised firm. So the benefit of privatisation is that the sovereign risk becomes more transparent and the arms length ownership can make undesirable political interference harder for a privatised firm than for a state-owned firm.

    So I guess it comes down to your view of what sovereign risk relates to. As I saw it as a bad thing then you can guess my own views on the social welfare ramifications of political interference!

%d bloggers like this: