PPPs take a long time to die

In the immediate aftermath of the GFC, I argued that the Public Private Partnerships model then in vogue was broken beyond repair, and that, even after the crisis we would be unlikely to see many more, though we might see deals wrapped up to look like PPPs, but with the public bearing most of the risk.

The recent failure of BrisConnections, the owners of the Airport Link Tunnel in Brisbane was no surprise. The AirportLink deal was one of the last pre-crisis PPPs, when investors were still optimistic enough to buy shares based on absurdly overstated traffic flows. In fact, they bought “partly paid” shares, a startling piece of financial engineering under which shares, trading at virtually zero, carried with them the obligation to come up with a payment of $2/share later on.

Because of the eagerness of private investors, the Queensland government bore little risk in the project. Former Treasurer Andrew Fraser’s description of the project as “a zinger of a deal for the public” displayed his customary tact, but was accurate enough considering the deal in isolation. The problem is that, after this and similar failures, it is highly unlikely that the pre-2008 PPP model can be revived in Australia.

For roads at least, there’s a simple alternative – road pricing based on congestion. I and other economists have been banging this drum for years, but politicians are terrified of even mentioning the idea. At one level, I can understand this – it’s tricky and likely to be controversial – but Queensland governments of both parties have adopted politically suicidal policies, largely motivated by the perception that they will free funds for capital investment.

First, Anna Bligh, along with Andrew Fraser reduced Labor to a seven-member rump with her pursuit of economically disastrous asset sales, exactly the opposite of what she had promised. Now, Campbell Newman has dissipated a huge volume of goodwill with savage cuts to public services, again a wholesale breach of promise, and is now pursuing privatisation. Compared to these electorally suicidal policies, road pricing ought to be a doddle.

27 thoughts on “PPPs take a long time to die

  1. Why the need for PPS solution, besides ideological confusion. If the society needs and it would benefit it, government can always find finaces to make what society needs. Government has power of alchemy behind it, it can create wealth for privates when needed, ahem, it can print money for what society needs. Whatever is built with new printed money is backing that new money, no need for gold backing in fiat system.

    Budget deficits are dangerous, and unnecessary. Deficits expressed in ammount of Treasuries are additional income on savings that are parked and unused for productive purposes. By growing ammount of Treasuries, FED grows and attracts dead savings while productive capacities strugle for investments and with that FED is incentivising speculative investment.

    There is no difference in inflationary pressure from deficit spending if spending is financed by printing money or selling Treasuries. But large public debt can cause confused panic and prevent needed deficit spending when private sector needs it. This applies only on sovereign currencies, while currencies pegged to other one or to gold, or local government spending follows different rules.

  2. Planet Money did a good episode on policies backed by economists across the political spectrum which were nevertheless political poison. (Mortgage interest tax deduction, corporation tax, employer’s health insurance tax deduction, drug legalisation). I don’t remember if road pricing was mentioned.

    What’s a mechanism for getting these sorts of changes through our political system?

    (I wonder if the reason there was little anti GST backlash was that everyone realised that it had bipartisan support in reality, if not in the rhetoric of the moment?)

  3. It would be easy enough to remove all fossil fuel subsidies, annually running at $7.7 billion in Australia. Similarly, it would easy enough administratively to remove all taxes and excises on fuels and replace them with a single properly priced carbon tax. This pigovian carbon tax would effectively act as a congestion tax (among other things). After all, it will cost much more to drive a big SUV slowly through one or two hours of congested traffic than to make alternative transport arrangements if such arrangements are made progressively possible by provision of efficient and affordable mass transit.

    Administratively, logically and even “infrastructurally”, the solutions are startlingly simple. Which begs the question of why we don’t implement them. Clearly, our entire political economy system (corporate capitalism) is inefficient, illogical and maladaptive. QED.

  4. isn’t Public Private Partnerships a flash name for project finance that is common for mega-projects in the private sector? many private sector mega-projects fail. others overrun on costs and time.

  5. Could you go into more detail on your model of congestion based road pricing? Would the cost to use the road go up as congestion increases (cue, “YOU WANT ME TO PAY HOW MUCH TO SIT IN BUMPER TO BUMPER TRAFFIC?!”) or as congestion decreases (cue, “You’re gonna charge me more when the road would otherwise be empty? Are you crazy?”)?

    Either way, it seems like – as you point out – an electorally touchy issue…

  6. After Oct 17, a commuter who uses the airport link to help reduce congestion will be penalized $2254/yr for being a responsible citizen. (46 weeks x 5 days) The value of fuel savings will be negligible. For some strange reason all our tolls are on roads, tunnels and bridges that bypass congestion. This means that the tolls help encourage drivers to go through the middle of congested areas.
    Cities like London and Stockholm have congestion charges that have helped make these cities better places to live and do business. If we must have tolls, peak hour congestion charges make more sense than putting a price on helping to reduce congestion.

  7. @John D

    But the value of the time savings is big. In my experience the airport link takes 20 minutes off the trip from the inner city bypass to the airport, for a toll of $2.50. Most people’s time is worth a lot more than $7.50 per hour.

  8. Prof Q,

    Are you aware of any work (especially by economists) on totally free public transport to achieve the same, or I would argue – far better, results?

    It would be loathed by neo-cons (ie: ‘political poison’ as our democracy currently is run) but as pointed out above, we could take that $5billion+ annual fossil fuel subsidy and spend it on free public transport and probably have some left over for schools and hospitals.

    Advantages include:
    * no administrative costs for ticketing, policing, money handling etc..
    * faster passenger movement (no paying, swiping cards, turnstiles or gates),
    * massive reductions in road congestion, repair/building costs,
    * ditto pollution and GHG emissions,
    * boost for businesses as staff and customers find it easier/cheaper to get there and have spare change when they arrive.

    About 1999-2000 BCC did a trial where during school holidays (all of January from memory) all bus trips were $1. As a bus commuter I remember the buses were jam full of people who wouldn’t otherwise catch the bus. I’m pretty sure the Council/Government never released publicly their findings from those trials and they were dumped after one or two years.

    It would take “courage” but be massively popular to trial a totally free public transport system.

    The Murdoch press would have a collective seizure, which is simply an added bonus!

  9. “For roads at least, there’s a simple alternative – road pricing based on congestion. I and other economists have been banging this drum for years”

    Yes. Economists of all stripes, at that.

  10. @Uncle Milton The toll is due to go up to $4.90 on Oct 17. Tunnel useage dropped considerably as soon as the free drive period ended. I imagine that it will drop even further after Oct 17. You are right of course some people are willing and able to pay $2254/yr for a bit of conveneience and time saving. Many will not.
    Megan: A commuter travelling 3 zones will have a Gocard bill of $1888/yr. A small car travelling a similar distance will consume less than $400/yr with fuel at $1.50/litre. If we want to drive down congestion there is a good case for free or almost free public transport. (Seniors in Perth get free public transport.)

  11. @John D

    I think to work it has to be totally free for everyone who wants to use it (see advantage 1 above). Otherwise all the administrative costs remain while ‘revenue’ falls. This way we forgoe ‘revenue’ completely as well as any costs not solely related to moving people around.

    Obviously, non-users benefit as well (eg: compare congestion driving around Brisbane during school holidays with during term time).

  12. @dz
    Let us hope Newman is not using the same engineers that did the forecasts on the toll roads to forecast the number of kids that will attend each school.

  13. @Megan

    Absolutely, Megan. Free, or very cheap, public transport would go a long way to solving our transport and related crises. It would be more efficient than our current private vehicles “solution”. I mean that we would achieve more passenger miles for lower fuel costs, lower pollution and lower total costs to society. Proving once again that capitalism is highly inefficient at solving the type of problems we now need to solve. The problems we now need to solve are solvable by using natural monopolies in public ownership and by democratic dirigist action. They are not solvable by unregulated capitalism.

    Ultimately, orthodox and semi-orthodox economists are going to have to admit that a lot more is broken beyond repair than PPPs. The capitalist system itself is broken beyond repair and we have to jettison that system before it breaks the entire biosphere.

  14. @Megan,

    Last time i was in Munich I caught the underground train and buses all for free. They have tickets and passes, but no turnstiles. Only occasional policing of trains by fare collectors who will simply make you pay on the spot if you don’t have a ticket.

    If I recall, there are other cities in Europe where public transit is totally funded through explicit city property taxes, so everyone rides for free.

    The goCard in Brisbane is quite good, but I do wonder about the net revenues from fares after the cost of collecting them is considered. My personal guess is that it costs you 30c to collect $1 in public transport fares.

    I would support a congestion charge only if the revenue was used to invest in alternative urban transport – free bus routes, more frequent trains, connected bike lane system etc. Otherwise, we are getting squeezed for money to drive because we aren’t offering alternatives.

    But my personal preference is for public transport, especially the bus and passenger rail system, to be funded through land taxes with no fares for passengers.

  15. People are expected to pay in Munich and almost everyone does. If you get caught without ticket, there will be a major charge, 40€ at least. Thank your tourist bonus if that was not the case. This is certainly nothing to brag about. No turnstiles are no invitation to cheat, rather a sign of a relativly low crime rate society.

  16. PPP. Yes, Prof Q, you have repeatedly and publicly presented the argument against PPP transport infrastructure and you have done so in a most elegant fashion, going straight for the jugular of the financial body of the model. You reported repeatedly on the empirical evidence – with one or two exceptions, PPPs are economically inefficient (misallocation of resources).

    A few words to the heading of this thread: “PPPs take a long time to die”

    Dr John Goldberg worked, in a sense, the other way around. He researched and examined the elements of individual PPPs in detail. As for road transport, he repeatedly found traffic volume to be overestimated at the planning stage, using objective data. He found a case of asset revaluations to increase borrowing capacity (Gillard and Swan’s problem regarding the so-called mining tax). He analysed the financial models (from first principle, applying math statistical methods, which, IMO, are quite compatible with the theory of Finance, excluding Fama) and concluded that not only tax payers will be called upon but that investors (providers of equity) will lose their dough and taxpaying workers are included in the set of ‘investors’ via their agents, the superannuation funds. The crucial papers can be accessed on http://wwwfaculty.arch.usyd.edu.au/web/staff/homepages/johngoldberg.html#top.

    His specific findings were apparently not appreciated by ‘industry’ (MacBank) and the then Vice-Chancellor of Sydney University did apparently not appreciate having a staff member whose work is not appreciated by ‘industry’. Details are in an article by Michael West, “Academic knew best all along”, SMH Weekend Business, 23/24 Feb 2013.

    I know John Goldberg (having been on the same side in an Environmental Court case in 1995 regarding noise created by the M2; ie giving evidence for the residents’ side resulting in an out of court action by the RTA to provide at least some noise amelioration to residents). John Goldberg confirmed to me the content of West’s report, except that West erroneously wrote 4 doctorates instead of 4 degrees including a doctorate.

    John Goldberg’s experience is not an isolated case of an academic who applies his or her knowledge ‘in the public interest’. There is apparently no longer a demand for this quality of academic work. As we are told, universities are now ‘demand driven’.

  17. “isn’t Public Private Partnerships a flash name for project finance that is common for mega-projects in the private sector?”

    This question deserves an answer. The short answer is NO. The longer answer brings out the difference:

    Project finance (as distinct for financing a project) is the name given to financing relatively large and capital intensive investment projects with a high debt component. Woodside Petroleum is still a classic case locally. On the face of it, a road project is ‘large and capital intensive’. Empirically, many PPP road projects also have a high debt component. On the face of it, I can see why the question arises. But, words, which paint ‘the face of it’, aren’t good enough. The success of high leverage project financing is conditional on there being a positive net cashflow with a small variance and the time profile of the cash in and outflows being such that the net present value being non-negative. For large investment projects (eg the initial Woodside Petroleum project), the variance of the revenue stream could be reduced in a credible way by selling forward (not futures) the output (quantities) (LPG) to financially credible buyers (eg Japan) and linking the unit price to a commodity with a deep market (some further risk reduction is possible through hedging).

    A profit maximising supplier of physical output has no incentive to lie to himself about the costs. Hence costs can be assumed to be under the supplier’s control at the planning stage. This condition can be assumed to be met by toll road construction companies as well as the suppliers of LPG.

    Forward contracts of physical output, signed between the supplier and a financially credible buyers is a very different thing to the supplier of the road (private) specifying the demand (traffic volume) for his project! There is no need to go into further detail, I believe.

  18. Congestion pricing. A congested road entails a non-monetary ‘price’ for every individual on the congested road in the form of time spent in an activity (waiting or sub-optimal speed) which is not available for alternative market or non-market activities. There is a welfare loss for an individual if this individual would strictly prefer to do something else during this time. So the personalised price (rate of substitution defined on preferences) may be zero or strictly positive and, a priori, strictly positive personalised prices differ among individuals.

    The problem with congestion pricing is that unless individuals have alternative means of transport, they cannot express their individual budget constrained preferences. Congestion is a quantity rationing mechanism which is not always worse than a price rationing mechanism.

    Like toll roads, the administered price (congestion pricing) is potentially a regressive tax. This becomes a binding problem when town planning and commercial interests (profit maximisation) results in metropolitan areas like Sydney. Lower income or wealth people often live very far way from their place of work because house prices are lower but there is insufficient public transport to give these people a transport mode choice. Congestion pricing potentially can lead to these people being priced out of the ‘labour market’.

    I imagine it is possible to deal with the regressive tax issue by, for example, linking the congestion price expenses of individuals to their income tax according a range of conceivable criteria (tax bracket, tax deductibility to treat this expense as it is treated for commercial enterprises, …). But then the question arises why not change the tax system such that there is enough revenue to built public transport systems/

    So far only the micro-economics for various individuals are considered. There is the bigger picture, the society and the economic theory of incomplete markets.

    Cost-benefit analysis of road projects has as its major ‘benefit’ component ‘travel time saved’. The estimated number of hours is valued in monetary terms (non-cash of course); an administered price. This is an attempt to monetise positive externalities. But the negative externalities (air pollution locally, ghg emssions, acoustic pollution locally) are not monetised. Roads (and air transport) is mispriced as a consequence of the method of project evaluation.

    Carbon pricing has given some financial institutions an idea on how they can benefit from this (trading in certificates, market making) but this does not ‘complete the market’ and therefore can make everybody worse off. (I am not arguing here against carbon pricing, I am merely saying that existing economic methods seem to stem from an era when the implications of the theory of incomplete markets for practical problems hasn’t been incorporated.)

    Sydney has the additional problem of its topography and harbour(s) and major rivers flowing into the sea. This excludes a grid-like transport system for a start.

    In conclusion, congestion pricing is not an unconditional solution. The conditions in London are very different from the conditions in Sydney and these are different from the conditions in Canberra, and elsewhere.

  19. @Ernestine Gross

    I imagine it is possible to deal with the regressive tax issue by, for example, linking the congestion price expenses of individuals to their income tax according a range of conceivable criteria (tax bracket, tax deductibility to treat this expense as it is treated for commercial enterprises, …). But then the question arises why not change the tax system such that there is enough revenue to build public transport systems

    I daresay that it would be administratively complex, given that road pricing would be a state matter and most taxes are collected by the Commonwealth. Likewise, public transport is a state matter. Assuming no change in Federal-State arrangements, a state government would probably need to impose the charges and then build the requisite public transport with the funds. Given the long lead times in the latter task, they’d need to start building before the charges were imposed, but debt is a dirty word these days and there would also be an element of ‘picking winners’.

    The other option might be to collect the money and then use it to build better housing (i.e. close to the city and services and on established PT routes) targeted at low income earners (i.e means and asset tested qualified lease access). Lead times here are much shorter and you could take a lot more vehicles off the road and deal with some of the housing problem. Since housing is close to a non-discretionary spend, and even wealthier people could occupy the housing (at market rates) the collections should be non-regressive.

    Then again, we could simply abolish state and local governement in favour of regional government, avoid the patchwork and do the whole thing properly. These days, PMs and LotOs seem to want to run for Premier anyway.

  20. In general, congestion pricing applied to the CBD is likely to be progressive.

    Toll roads vary a lot – in Sydney at least the first ones were mostly in the Western suburbs, since the North Shore and Eastern suburbs already had well-developed road and public transport networks – their turn came with the failed tunnel projects.

    But I think all of these are likely to be less regressive than the standard practice of using registration charges (fixed per vehicle) to finance roads

  21. There was a tollway from Berowra to Gosford but this was dropped after the opening of the F3. Only state funded roads can be tolled and the F3 was funded by the Commonwealth.

  22. @John Quiggin

    How is it progressive??? there are two meanings –

    meaning 1: the more you travel the more you pay, or

    meaning 2: allocating the service of travel equitably amongst society based on need not purchasing power.

    I suppose progressive in one sense, reactionary in another.

  23. sweden introduced road tolls by having them for 7 months, lifting the tolls for the next 6 months and then having a local vote on whether to bring them back.

  24. @Chris Warren
    Meaning 2 – travel by private car into or through the CBD is more likely to be undertaken by high-income earners than is use of roads in outer suburbs

  25. @John D

    As a senior in Perth, only weekends and public holidays are free. On weekdays the free public transport is from 09:00 to 15:30 only.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s