Monday, August 16, 2010

Competition Series Part III: History

Often forgotten in the competition debate are the reasons for a monopoly’s existence in the first place? A few are:

1. Mergers and economies of scale of private enterprise
2. Government development for its own needs (including defence)
3. Government intervention due to natural monopoly features (either too high risk for private enterprise or too much scope for price gouging)
4. Government intervention due to positive externalities (for example, cleanliness and health benefits of sewerage)
5. Government intervention due to fairness and equitable access – once a technology becomes a necessity it is politically expedient to promote fair access (including regional development)

While many may disagree that government involvement in some infrastructure networks was necessary from the start, citing the textbook benefits of the profit maximising natural monopolist, the onus should be on those promoting change to demonstrate that the world has changed sufficiently for competition and/or private firms to now deliver these services.

Historically, with the advent of new technology, government will typically step in if it sees benefits to centralisation - creating an entity tasked with equitable provision of the new service. Prior to centralised water and sewerage in cities, each property owner would have had a rainwater tank, bore or well to supply water, and a thunderbox for waste. Health benefits of newly designed reticulated sewerage systems were overwhelming (although it took some time before waste was treated in any fashion before being dumped into waterways). Private investment in sewerage reticulation could only recover cost from those who accessed the system, yet the social benefits were much broader. A government established (and subsidised) monopoly was the only way to go. A similar story can be told for water reticulation.

These water and sewerage examples confirm points 2-4 above. While a private enterprise could have undertaken sewerage works, there were positive externalities, and issues of equitable access. Furthermore, a private company would expose themselves to competition with existing methods of treating waste and capturing water supply.

On a more technical note, electricity generation was originally the domain of manufacturers who generated and used their own power onsite. With the wide acceptance of Tesla’s AC power, which could be sent vast distances with greatly reduced losses, the centralisation of electricity began. In this industry economies of scale were so great that governments typically took control of all generation and distribution to reduce costs. Ironically, we now believe that competition will in fact reduce costs and subsequently consumer electricity prices – a situation yet to transpire is State’s where this has happened.

The history of water supply in Britain follows a similar pattern, with government intervening in the 1860s to take ownership of the private supply market due to massive inefficiencies from infrastructure duplication required for competition (water in 1851 was about 60% privately supplied). The costs from these inefficiencies were hindering uptake of reticulated water which proved to have external benefits of improved sanitation and health in the cities. Much of the water industry was then reprivatised in 1989.

The unfortunate part of these stories is that once an industry is in the hand of a government entity, there is no risk, and therefore no incentive to innovate. Rather than government showing the willpower to provide incentives to reduce price and improve service provision of its own monopoly, it instead decides that competitive pressure is the best way to go. To maximise social benefits, government is now required to properly regulate a market of smart, influential, profit seeking firms, instead of properly regulating or managing its own corporation properly. While I am generally pro-competition, one wonders if those in charge of competition reform appreciate this irony.

In somewhat of a return to the past, the federal government is now keen on subsidising the rollout of a fibre broadband network. The justification for government provision is the same –characteristics of natural monopoly, positive externalities of information access, and fairness of access. This is perfectly in keeping with the theory and evidence discussed above, apart from the intention to vertically separate at the wholesale level (a rather grey distinction if you ask me). It seems we have come full circle.

The turning of the tide against privatization predated the financial crisis. Internationally, a number of major privatizations have been reversed. The UK government was forced to renationalize its rail network after the failure of the privately owned operator. In Australia, dissatisfaction with the privatized telecommunications monopoly has led the government to announce that it will get back into the telecommunications business by constructing a publicly-owned national broadband network. New Zealand, where market liberalism was implemented in a radical form in the 1980s and 1990s, renationalized its national airline in 2001 and its railways a couple of years later. And even relabeled as “choice”, Social Security privatization proved so politically unsaleable that it was abandoned early in Bush’s second term.

While the economic evidence suggests it is socially optimal for government to proceed with its rollout of the NBN, I have reservations. New fibre internet connections will have to compete with existing broadband on the copper network, and with developing wireless technologies. I expect there to be a significant cost to go from existing 20Mbps connections to 100Mbps, yet most users will not notice the difference, and may not be willing to pay. If the government goes ahead, it needs be aware that cost recovery may be very low for some time and the justification for its involvement are the positive externalities.

In the long run I still believe the NBN will be a socially beneficial investment. As technology develops on fibre networks, who knows what speeds will become standard and the opportunities this technology will provide.

An understanding of the history of competition and privatisation is necessary to better critique arguments from both sides.


  1. To Cameron: from the link you provided: Wikipedia - Natural Monopoly
    Competition amongst the companies in larger industrial towns lowered profit margins, as companies were less able to charge a sufficient price for installation of networks in new areas. In areas with direct competition (with two sets of mains), usually at the edge of companies' territories, profit margins were lowest of all. Such situations resulted in higher costs and lower efficiency, as two networks, neither used to capacity, were used.

    A common argument for government monopoly is to eliminate wasteful duplication of infrastructure. This is proclaimed as an obvious market failure.

    But, note in the above example that the water companies were willing to build two sets of mains to acquire customers, at a lower profit margin in those areas. Possibly, this was also because they charged less due to the competition in these areas.

    It is purely the opinion of the Wikipedia author that the prices charged were not sufficient.
    It is quite possible that duplicate, competing infrastructure, "neither used to capacity" was a worthwhile expense overall to promote competition and lower overall prices.

    There is always some duplication in competition, but the benefits almost always outweigh these costs.

    You say [edited]: Once an industry is run by a government entity, there is no risk, and therefore no incentive to innovate. The government does not provide incentives to reduce price and improve the service of its own monopoly. Instead, it decides that competitive pressure is the best way to go.

    Government captures an industry ostensibly to avoid duplication and lower costs. Then, it stops innovating, and can only produce improvements by reinstituting some type of competition. Maybe the governemnt should have stayed out from the beginning.

  2. Andrew,

    The second point you highlight is the core of the discussion. I agree that governments are pretty bad at running anything and we should avoid government control of production unless there is some kind of social justification. if people prefer government involvement, even if it is strictly inefficient in economic terms, who is to argue.

    I don't know if there is a correct answer here, but my main concern with the pursuing competition at any costs are:

    a)governments will always control bottlenecks. For example, once existing airports are utilising runways at capacity and another airline wants to enter, the government needs to approve new airports or runways. It is back in the monopoly position and has layers of competition dependent on its timely and efficient decisions. What incentive is there for the government to approve new airport development (knowing the NIMBY culture)?

    b)not only this, but governments need to regulate the access regimes to monopoly infrastructure. Why would we think government would be better at this than running its own monopoly? Wouldn't it still take the easy way out when it came to tough decisions? Wouldn't the existing market operators put pressure on government to raise barriers to entry for some technical reason?

    c)if things were going so well under a private competitive enterprise, why would government become involved in the first place? Where is the incentive to do this? There must be some pressure on them from citizens who were not satisfied with the way things were being done (possibly due to concerns of equitable access in the case of the British water example) Whether they were correct or not is, however, up for debate.

    d) private firms who provide infrastructure can only generate revenue from direct beneficiaries. If there are positive externalities (in the case of sewer systems for example) then to maximise welfare government needs to be involved.

    I prefer private competitive markets - the success of private internet service providers and mobile phone companies, particularly the rapid adoption of new technology, has clearly demonstrated the benefits of moving from monopolist to a competitive environment.

    But the reality is that governments control public space and need to respond to concerns of the people that are outside the scope of homo economicus. Therefore we have a comparison of two second bets options - poorly run government monopolist, or poorly run government regulated market where they still hold monopoly power at some level.

  3. Andrew,

    I just wrote out a very long response but then lost it for some reason. You make good points.

    To respond to your first. If things were so great in the private water market, why would government intervene? Where is the incentive? If peopl had, for example, a concern about equitable pricing of water than, rational or not, the government is justified to intervene (maybe not by taking full control).

    Your second point is the heart of the matter, but there is more to it. I govenment stayed out there may have been no innovation due to the economies of scale required to do so.

    In any case, my point is that even with competition, governments generally still control a monopoly bottleneck due to their control of public space (they decide how many airports or rail lines can be built, where cables and pipes can or cannot go).

    Our choice is between a poorly run vertically integrated government monopolist, and a poorly regulated compeitive market where government still retains some monolpoly power.

  4. I'm not sure the choice is as stark as that - it may be that we have a choice between a well-regulated but poorly run GOC or a poorly-regulated but well run private firm.

    Governments know how to do it themselves but are notoriously bad at setting up others to achieve the same outcome

  5. The way I figure it, the government is there to build and sustain a framework of laws, services, and infrastructure within which their people can prosper.

    I think that networks are the probably the only example of a natural infrastructure monopoly. One where duplication is value destroying and regulation of standards must be high.

    Each of the examples used above are networks. Whether it's water, gas, electricity, data, cars or trains, they each travel over a network.

    This does not mean that electrical power plants should necessarily be government owned. There is plenty of room for competition on the supply side, and there is plenty on the retail side. The network side simply makes no sense duplicating.

    Japan's initially private electrical networks led to an ingrained regional split between 50hz and 60hz power which partially contributed to the Fukishima incident. A government imposed standard could potentially have avoided this.

    It wasn't that long ago that Australia finalised changing its rail networks to standard gauge. I am not sure if this was a private or state caused incompatibility, but it is another example of incompatible regional monopolies.

    Compare the Optus and Telstra HFC (cable) rollouts, and you'll see that network competition quickly exhausts the resources of the competitors. They both made net losses on their investment.

    Now compare ADSL, which uses the monopoly copper network, and you see dozens of competing providers. Some provide their own supply, others are retail level resellers.

    This market is quite functional although it has been necessary to impose heavy regulation on the network owner to ensure this is the case.

    But we're at an impass. The network is degrading in many places and must be replaced. Many are fine, but the medium itself is incapable of meeting the future requirements.

    The only option in this case realistically, is to take what worked from the copper model, and upgrade it.
    The network monopoly works. The supply and retail competition works. They're not ideal models, but overall it's functional.

    Due to the wording of the original agreement between Telstra and the ACCC stating specifically that only the copper network must be a declared service, forcing Telstra to upgrade the fibre and then declare it is at best an uphill battle

    This is why the NBN makes sense.

    We need a standardised declared service network monopoly with supplier and retail competition.

    There's really no other way.

    - myne