Wednesday, March 25, 2020

Bad Move: Infrastructure Victoria's scheme to disintegrate public transport fares

With non-essential trips discouraged, many businesses shut, school holidays extended and more working from home due to the Corona Virus, transport congestion couldn't be further from most peoples minds right now. 

However this time will pass. We'll likely be back to being a growing city growing by more than two thousand people a week. And pressure will return to our roads and railways. Infrastructure Victoria are the boffins charged with thinking about how we should cater to and respond to this growth. 

Anyway today (of all days!) they released their Good Move: Fixing Transport Congestion report. It's all about changing how we pay for transport. Changing this is cheaper than building infrastructure (which might only be fully used for a few hours a day) and improves the utilisation of existing assets. 

For example if we have to pay for things at the point of consumption that can cause us to think twice about whether we want to use it or not compared to if it was laid on without direct charge (like most roads and parking). In the latter case the roads and parking still needs to be paid for through higher charges and taxes but more will be built than is necessary due to the demand induced as a result of it being 'free'. That's detrimental for other modes since the increased spacing out that catering for cars causes makes active and public transport less economical, accessible and effective. 

That's all good stuff. Although it's incredibly hard to implement, whether at the local, state or national level. Parking, for example, is considered a necessity. And indeed driving is the most popular mode for many trips. However people make the leap from something being a necessity to it being free (and often untimed) like air. Imagine how fat we'd all be if Kingston biscuits were free?

It's interesting to compare parking with other things we'd consider necessities. For example food is even more necessary than parking but we expect people to pay for what they eat. We don't make it free. And the millionaire pays exactly the same for two minute noodles as does the struggling pensioner behind. If society wishes to help the less well off we do it not by setting the price of food according to ability to pay but in providing welfare payments and allowances. After all the farmer does the same amount of work to produce a potato regardless of who consumes it. 

Health care is also necessary. But we don't pay for it like we do food. Medicare or private insurance tips in most. Bulk billing and Pharmaceutical Benefits limit some charges. A single payer national health system has wider social and economic benefits with outcomes better, coverage higher and total costs lower than alternatives such as the US system. And even if philosophically you do not care for collective medicine, if you live in a city with communicable diseases then it is in your own interest that others are not sick.  

Free health is considered much more of an entitlement than 'free food' or 'free housing'. Food is most 'market-oriented' followed by housing, health and education less so. 

'Free storage' though is more nuanced. Everyone accepts paying to store stuff in an industrial area storage unit if they have more posessions than room to house them. Car parking is another form of storage (often in a high-value CBD or suburban centre) but acceptance of paying for space is much lower. Instead it's considered a necessary entitlement like health care. This is something that IV gallantly seeks to challenge. In compensation they seek to remove other charges like car registration which you can read about in the report. 

I'll soon discuss what the report says about public transport fares. However first off it's worth reading their executive summary. They don't get off to a good start. Take below for instance. 

It's not believable, at least as far as public transport services go. We've added a city half (or more) size of Adelaide in the last 10 years but have we added anything like their service quantity, especially for buses where the most new housing is? 

They need to talk to Monash's Prof Graham Currie whose graph shows that Melbourne has been in a per capita service decline for the better part of a decade. At least once and sometimes several times per week this blog brings up examples of bus routes and networks whose service levels and destinations have been stagnant for years if not decades. Tram timetables have hardly changed. And the last comprehensive greenfields Metro train timetable rewrite is a distant memory despite services on much of the network (and the City Loop!) remaining complex and sometimes infrequent. 

While we have been busily building infrastructure we have hardly been adding service. It's mistaken to  imply that it's time to forget service and shift to pricing, at least in the public transport arena. I can only assume that this view has come about since IV's boss, Michel Masson, is a tram man and may not have seen main road/major destination bus timetables like this. 

If he had then he'd realise that pricing, especially public transport fare policies, is the least of the network's problems. A bad service with lower fares is still a bad service. 

Speaking of fares it's pretty clear our system's not perfect. For example our fares are quite flat. Short trips in Zone 1 cost as much as a trip to Melbourne's edge. That can dissuade people from off-peak travel where the system has cheap capacity to carry them. You could argue that some longer distance peak trips are a bit too cheap, especially for high-income myki pass holders. There are some odd bumps in peri-urban area fares. And families (who probably already own a car) may not find some trips good value if they travel together. 

Nevertheless, despite the chequered histories of various ticketing systems (scratch tickets, Metcard and Myki), Melbourne and Victoria's integrated fare structure can be counted as a major strength of the network. At certain times (particularly in the early years of rail franchising), Metcard fares were one of the few things that held the system together. 

Most people choose to use public transport versus other modes based on what suits their trip best. In other words 'the customer is always right'. Fares is just one, and not necessarily the most important, consideration. A network that presents itself as an integrated system, including fares, routes and timetables, is useful for the most number of trips and thus has the best chance of being useful. 

IV doesn't get the integrated network concept even though it's something that thousands of passengers navigate daily. For example they seek to smash the network's multimodal identity. That's been painstakingly built up over years of integrated fares that visitors from overseas (and Sydney!) envy. 

IV's recipe is a regression to single-mode distance-based fares like that which existed in the bad old days up to the late 1970s. That is the period that saw the fastest decline in Melbourne's train and tram patronage. In contrast fare integration in the '80s revived Melbourne bus patronage which was then in free-fall.

IV approvingly quote Sydney, Australia's only capital without proper integrated fares, as an example. What they didn't mention was that Sydney has greatly improved services in the last few years. They had large network-wide train frequency upgrades in 2017 of the magnitude that Melbourne has not seen in more than 20 years (when Jeff Kennett boosted Sunday trains and trams). And there were train presentation and reliability improvements with Gladys B at the helm. 

Sydney's improvements have been such that their trains now run two to three times more frequently than do ours during evenings and Sunday mornings. And the number of extra bus services proposed (14 000 weekly) is a number that Melburnians can only dream of. Maybe it's these that have had a bigger impact on patronage? 

The table below suggests a pricing structure for IV's proposed mode-based fare system. 

The report's first principle is that all modes and routes should be priced. The theory seems to be that if modes are fully priced and charged at the point of use then people can make choices that better reflect the cost of providing that particular transport option. That user-pays approach may reduce inefficiencies and discourage over-provision.  

Here's a well-known curve used to illustrate how demand falls as parking is charged.  

The various public transport modes are considered separately, as per the table above. IV's thinking seems to be that people can make choices between modes within public transport, eg bus vs train vs tram. And that the pricing system should reflect the cost of running each mode and even route along with demand variations such as during peak times. 

It is thought that these price signals would encourage behaviour change amongst price-conscious passengers. For example using a different mode or route because it's cheaper.  Their idea seems to be that you have a sort of round-robin between modes by pricing them differently. For example you price roads and parking to encourage some drivers to take the train. But then the train gets crowded so you price that higher to encourage the price-conscious to take the bus. That frees up more room on the train for 'high value passengers'. Presumably someone then replaces the displaced car driver. 

IV seem to assume that passengers can pick and choose between public transport modes like a cyclist could choose to walk instead. If they could then differential pricing between modes may have a meaning. But usually people can't choose as outside the inner-city only a handful of routes (maybe just one) will be suitable. 

If you live on a crowded tram route like the 96 the availability of a cheaper fare on the 582 bus is no help. Trying to dodge a direct (but more expensive) route by taking a combination of other routes isn't very practical either. Public transport is slow enough without subjecting value-conscious passengers to these gymnastics. And don't forget the importance of saving money for a large passenger group; the Grattan Institute recently found that 10% of households had less than $90 in the bank

What if you wanted to explicitly design a public transport network so you could choose between whether to take the train or the bus into town? Choice is good, right? Especially if you had a fare system like what IV wants that gives incentives to ride buses instead of trains. 

As it happens we already have such public transport networks in Australian cities. Unfortunately they're not very good for diverse trips. The shape of what I'll call a duplicative network is on the left below. There are many buses and trains but they all go to the CBD. Buses may be quite good but trains may only run half-hourly and be poorly used for the size of the network. Many people can choose between bus and train for trips to the CBD. However access to places other than the CBD is poor, involving backtracking via the city. Brisbane and Adelaide are like this. Despite allowing choices between modes for many city trips they are poorly set up for the sort of patronage growth (76% by 2030) that IV is projecting for Melbourne due to their limited versatility. 

On the right is the connected network. Like with the duplicative network heavy rail serves the CBD. however unlike the duplicative network most buses, particularly in the outer suburbs, do not. Instead they provide cross-suburban travel and connect with trains. While there is a need to connect for trips from some places to the CBD, the large number of circumferential routes and interchange points make cross-suburban travel easier. This makes for a more versatile network that enables more people to get to more places. 

Because buses feed passengers to train stations rail patronage is higher than  in cities with duplicative networks. And because there are fewer buses on clogged inner-suburban streets they can instead operate as faster and more frequent feeders in the suburbs. Perth and Melbourne have this more useful style of network.  A trade-off is that far fewer can take one bus to the city. Instead they need to change but services may be more frequent to more destinations as compensation. 

While an efficient public transport network could have some built-in redundancy for robustness it should avoid wasteful duplication between services. An efficient connected network maximises the number of people near frequent service and the places they can go with a given number of buses, trains and trams. The different public transport modes should work with one another not against one other. Contrary to IV's recommendation, this needs an integrated multimode fare system otherwise there are pressures to retain or add duplicative routes or modes on the grounds that removing one would cause fares to rise.

What else does Good Move say? Again, according to IV, price is (nearly) everything. They reckon that making buses cheaper than other modes will increase bus boardings by 110 000 per day. That's about 40 million per year or an increase of about 33% on existing numbers. Bold! 

Given how static bus patronage has been for the better part of 10 years, it's hard to see where those numbers will come from. You might get something like that if you implemented network and service reforms but then the increase would be service-driven, not price-driven. Very little will come at the expense of trains or trams because whatever else you might say about our bus network, sensibly very little of it duplicates these modes. 

If the source of new passengers is intended to be motorists then service convenience (along with priced or scarcer parking) rather than cheap bus fares is likely to be the key.  This is especially when we look at other modes where an overall 76% increase by 2030 is projected. The only way you're going to get this is major investment in service levels to deliver the needed capacity. That includes for buses both as feeders to rail and suburban transport in their own right.    


IV's report says some good things about transport pricing. We need better parking and road space policies for instance. Also fares. For example our fare structure is too flat and we could do more with off-peak pricing. However I rank the fares issue as middle-order issues; certainly not top priority. 

Maybe I'm being unfair given it's a report specifically about pricing, but their obsession with pricing over service is a major blind spot. A poor service is still poor no matter how cheap. And you can't catch a service that doesn't run; a major issue in many areas that still lack route coverage or span of hours. 

Dismantling our integrated fares would be disastrous. It would confuse passengers, do little for patronage and potentially split the network into disintegrated fiefdoms like existed during the rail franchise mania of the early 2000s (which Britain is now undoing).  

Politely listen to IV by all means, but don't regard them as the experts on public transport network  connectivity. If you have any comments on the report or wish to disagree on anything then please leave them below. 

Melbourne on Transit bookshop

Favourably reviewed books about transport and cities. Purchases via these links support this blog and its independent reporting (at no extra cost to you).   


Frederick.aSA said...

"The different public transport modes should work with one another not against one other. Contrary to IV's recommendation, this needs an integrated multimode fare system otherwise there are pressures to retain or add duplicative routes or modes on the grounds that removing one would cause fares to rise."

This reminds me of the importance of making the tram network accessible for people of limited mobility. Otherwise you (quite justifiably) get pressures to retain duplicative bus routes with kneeling buses where trams are high-floor and/or have no accessible stops.

Unknown said...

A wonderful review of IV's report - thank you. Their thoughts on road pricing are interesting and have some merit, but given the political sensitivies, likely to be ignored. Their thoughts on PT pricing are frightening, and as you well put, contrary to the needs of an effective multimodal transit network. Passengers dont WANT to transfer betweeen modes, but doing so gives access to a much greater range of origins and destinations. So you dont charge them extra to do something they dont want to do in the first place. You are so correct in noting that it is service rather than pricing which is the limitation on bus ridership in Melbourne. If passengers needed to pay again to transfer to a bus, we'd likely see ever lower ridership.
Exactly what the interaction of pricing applied to road usage might be on PT ridership in general is a less explored, but interesting area.

Alex ken said...

A shareholder who is paying Higher Rate Tax will have the dividends added to their income and will have extra tax to pay.

London tax specialist