Should CityRail Depart IPART?

IPART is the independent economic regulator for NSW. It oversees regulation and conducts pricing reviews in industries such as electricity, gas, water,taxis and public transport.

IPART recently completed a review of pricing for RailCorps rail pricing provided under the CityRail brand – essentially the rail network around Sydney, Newcastle and Wollongong. In this post I consider whether IPART is the right place to decide these rail pricing outcomes and conclude that as currently set up it is not.

The final review recommended an increase of 12.5% in real terms or 25% in absolute terms over the next 3 years. This sounds fine and is the sort of reasonable pricing change you would expect to be produced from a reasonable independent regulator.

This post does not question this increase per se, but rather whether it is appropriate for an independent regulator to be producing pricing models for a service that has an across the board 74% (p. 67) public subsidy. Yes, thats right, that weekly ticket for $30 really costs $120 to provide with the NSW State Government contributing $90, that $6 return ticket really costs $24 etc.

Even on those crowded trains from 1967 and a rail network from circa 1928.

The IPART model is commendable and does a good job where the Government has a monopoly or in markets where there are high costs of entry and few competitors to separate fair pricing from the whims of the government of the day, or the exploitative pricing of an essential service. I would argue, though, that where the across the board subsidy is so very high and the outcome is expected to be broadly inline with current pricing (nobody would expect a recommendation to quadruple fares) then the pricing conclusions will be fairly arbitrary and the model is of questionable value.

The problem is that the key question at this level of subsidy is not whether services are being provided efficiently and not overcharging consumers but what level of subsidy should be maintained.

I would argue that IPART is not the appropriate place to make this decision. IPART can consider what changes in costs and efficiencies should be reasonably reflected in the total cost of provision of rail services and allow fare revisions that reflect these factors. But this work is largely irrelevant compared to the bigger question of what level of subsidy is to be continued.

Before considering the subsidy, IPART needed to calculate what it should cost to supply train services.
In the end IPART looked at three alternative models to calculate what it would estimate as a reasonable cost to supply train services. The building block approach, an operating and maintenance cost approach, and a long run marginal cost approach. It decided on the building block approach which calculates the revenue required based on forecasts of capital and operating expenditure inter alia. The first step of the building block approach is determining the Regulatory Asset Base (RAB). There are two main schools of thought on how this should be done and this is the point (along with model selection itself) where the process gets fairly arbitrary.

Most of the actual capital costs are not included in the RAB. In the discussion paper, this is justified because The inclusion of these assets would lead to pricing outcomes that are significantly above the economically efficient level, which is likely to lead to significantly higher fares and lower patronage levels.(see page 5 draft, the equivalent in the final report is on page 63). This is a circular argument we are trying to decide what fares should be, first we must decide our inputs to calculate fare supply costs, we decide our inputs on the basis that we want fares to be low, now let us proceed to calculate fare supply costs- I wonder how they will turn out?

So we have a fairly arbitrary estimate of what it should cost supply train services designed in a way so as not to push up prices too much. To this IPART now adds a consideration of what percentage of each train fare should be funded by the taxpayer.

As to how the level of subsidy should be determined, IPART struggles, and fair enough too. IPARTs terms of reference require it to consider an appropriate range for the allocation of costs between government and users, taking into consideration the positive environmental, economic and social benefits for the community generated by CityRails services. That is not a great deal of guidance.

Even having IPART consider the question of a level of subsidy leaves out alternative questions that are available to governments. Government could reconsider the broad nature of the subsidy and whether it could be better targeted, made progressive, means and asset tested, or better dealt with in a cash-back style arrangement.

These are important considerations as the burden of the rail subsidy to NSW taxpayers is significant, costing $750 per household per year. This in turn is directed to the 20% of Sydneysiders who use rail once a week or more, equivalent to around 14% of the States population (see final report page 105). Further the typical weekday commuter has a household income $12,000 higher than the Sydney average (see page 5 of discussion paper). It is not prima facie clear that these commuters should receive a blanket subsidy at all.

In any case after considering the benefits to the commuting of decreased road congestion, lowered air pollution etc. it found the fund split should be 70%/30% between the taxpayers and commuters, from the existing 74%/26%. After hearing submissions it reduced this to 71.5%/28.5%. Mission accomplished again, almost no change with a slight win to government. Maybe that was the point of the whole process?

But where it all falls down for mine is the lack of consideration of the opportunity costs. Necessarily, the opportunity cost of the broad subsidy in terms of lower taxation, increased health and education spending is outside the scope of any IPART style review.

More significantly, for Sydney residents there is a major opportunity cost to rail commuters themselves and would-be rail commuters, in the opportunity cost to the expansion of the network. The 1920’s design network, it is universally acknowledged, falls far short of anything like meeting Sydneys needs. It is dangerously overcrowded in the city, and does not go at all to the major population growth area of the last 40 years in the northwest, along with having significant deficiencies in the southwest. These major projects have been recently canceled again by the NSW State Government on the basis of funding concerns and misplaced concerns about its credit rating (an argument for another post).

The $1.7 billion a year subsidy, and the fact that any new line will at most return only 28.5% of its costs means that rail expansion is severely constrained. As a consequence the benefits of rail expansion to the community in terms of further reductions in congestion, travel times, rail access etc are also restricted. Trading some of the subsidy to benefit rail commuters themselves and would-be rail commuters should be an option open to the government. The current regulatory set up does not allow this and I think that is a shame.

I cannot remember the politics that got rail pricing into a regulator’s hands with so little guidance for the regulator and so little flexibility for the government, but it is time it was undone.

This entry was posted in Economics and public policy, Uncategorised and tagged , . Bookmark the permalink.
Notify of
Newest Most Voted
Inline Feedbacks
View all comments
12 years ago

Yes, thats right, that weekly ticket for $30 really costs $120 to provide with the NSW State Government contributing $90, that $6 return ticket really costs $24 etc.

This is not completely true. During times of the day when the trains are packed and the prices are full (i.e. peak periods) the effective subsidy is less (probably nearly zero). During off peak times, there might be two or three people per carriage but the prices off peak are lower and all the infrastructure costs just as much to maintain. The real problem is that we have built infrastructure designed to handle massive peak loading but only for a few hours out of every day.

If more businesses were willing to move their offices out to where their employees can afford to live, then the commuter peak on the trains (and roads) would be greatly reduced and the rail network could run much leaner. From the business point of view, why care about the cost of transport? They don’t pay it.

I cannot remember the politics that got rail pricing into a regulators hands with so little guidance for the regulator and so little flexibility for the government, but it is time it was undone.

If the weekly ticket really went up to $120, that is approx $6k per year but at a marginal tax rate of around 33% you actually need to earn $9k more per year just to pay for transport. The roads would become bedlam and a lot of people would be demanding pay rises. Of course, maybe we could get some of it back in tax breaks but seems rather unlikely. Rents in all the city apartments would go up too.

Richard Tsukamasa Green
Richard Tsukamasa Green(@richard-green)
12 years ago

One of the obvious reasons why commuters are better off is that the majority of the network remains in inner suburbs that either gentrified or, due to their proximity, were always more wealthy than otuer suburbs.

But I can’t help but think that and geograpically located public good (transport, beaches, parks etc.), if subsidised, may end up passing much of its subisdy directly to landowners. If such a good is underpriced, then agents are prepared to pay more for access, whether by directly higher prices, or trading off space for access (living in a flat near a station) or time (driving/walking to the station).

Which may mean that whilst a user may have higher incomes, if they are a recent buyer or renter, their incomes after transport and land may not be much better off than if there was no subsidy, most of it dissapearing into rents in the form of….rent.

Which poses another quandary, that if a subsidy is required for other purposes (lowering road congestion/pollution etc.), they may need to increase it just to get the appropriate price signals at the level of the agents they want to influence.


and now, amusingly, I am using my travelpass to take a bus and a ferry to swim in an enclosure elsewhere rather than buy a $5 pool ticket here, because the travelpass makes the marginal cost $0 in dollars.

12 years ago

Interesting that IPART is not required to assess service efficiency in their calculations. I recently travelled on an outer extremity of City Rail. The reasonably large station I joined the train had no ticketing mechanism and there was no obligation on me to purchase on board (guard) or when I got off. Other passengers from more distant stations confirmed they too had no ticket or opportunity to purshase. Ticket machines in lieu of station staff (wages) were said to be too prone to vandalism. Result? Passengers willing to pay were effectively excused. Suggestion? The guard could easily note those boarding and do a quick walk through a la bus conductors.