Good buses, good politics and social media

Members of parliament measure community sentiment by several means. It may be what they get asked about when in public, the volume of phone calls their office gets or the number of emails. Their staff may also monitor social media to see which posts get high engagement and which do not. 

Where do buses fit in on the latter? 

To get a rough idea I looked at the Facebook page for two MPs who represent diverse working class areas in which living costs is a major issue. The seats concerned are Mulgrave (Eden Foster - 1.3k followers) and Melton (Steve McGhie - 8.4k followers).

Both seats had recent bus service upgrades (upgraded 7 day 800 in Mulgrave, new Route 452 in Melton) that both MPs posted about on their Facebook pages. 

I counted the number of reactions they got and compared them to the four immediately previous posts. 

On Eden Foster's page the 800 bus item was the most reacted to post of the five surveyed. This had about 20 per cent more reactions more than the next most popular post. I had to go back 15 posts to find one with a higher number of reactions so the Route 800 item was likely in the top 10%. 

The Route 452 bus item on Steve McGhie's page was also a good performing post. By a very high margin compared to the previous four. I had to scroll back over 30 posts before I found one that had higher engagement (238 reactions). That one was also transport themed; in this case the West Gate Tunnel. 

It's only two MPs sampled but this reaction test shows that if politicians want online engagement they could do much worse than to talk about improved buses and other transport projects. This also ties in with housing and cost of living narratives. 

On the latter, when it comes to transport, by far the most expensive trip is that which cannot be made by existing public transport services. This makes public transport coverage and service improvements that expand the range of trips practical by public transport throughout the week so worthwhile.  

Ten top transport tips for Treasurer Symes

The Age is reporting that Jaclyn Symes will be the next Victorian state treasurer. Symes, the current Attorney General, will be the state's first female treasurer. She represents Northern Victoria in the Legislative Council. Previous portfolios held include Agriculture, Regional Development, Resources and Emergency Services. 

Ms Symes faces significant challenges. With rising interest repayments, the spectre of a debt overhang, a population averse to paying more tax yet the need to deliver services a growing state needs, being state treasurer right now will be a thankless job.

This is even more the case if you come to the job after your government's been in a decade and even the disunited opposition is ahead in the polls. 

Being Treasurer is a high risk/high stakes game. They have about an even chance of becoming leaders (Thompson, Keating, Howard, Brumby, Napthine) or being passed over as their party modernises (eg Jolly, Roper, Sheehan). Some, like Stockdale, get out to earn big bucks in the private sector. 

But for now what is undeniable is that the role gives seniority, status and licence to dabble in every portfolio. The treasurer has a massive department that can monitor the spending departments. That gives them a massive reach across government that no one but the premier has. 

No doubt hundreds of people will be offering their ideas to the incoming Treasurer. It's irresistible since, as someone wise said, the real priorities are not in a government's plans but in their budget.

With all that in mind, here's ten tips for the transport portfolio, one of the biggest-spending areas the treasurer will need to have at least a nodding familiarity with:  

1. Opex is super important in transport 

It's a message treasurers probably hate as recurring expenditure is harder to arrange than a one-off capex spend for a major project. But in transport it's opex that affects whether the train comes every 10 minutes or every 40 minutes. That has a huge bearing on asset utilisation, and for the punters, the usefulness of the network including access to jobs. 

2. A small amount of opex can go a long way 

Having got the bad financial news out of the way, the silver lining with opex is you don't need a lot to start the path to worthwhile improvement and reverse the per capita service decline for trains and trams. Fortunately most of what we pay train and tram operators are fixed costs. Even a 1% increase in Metro train service kilometres can cut waits at key times, as explained here. Buses are also cheap to boost; $300 - 500k  pa may be enough to get a local route up to 7 day service, while a $10-20m per year package could target upgrades on popular routes in our highest needs areas.   

3. Exalt the off-peaks 

The pandemic accelerated existing travel trends, that is a shift from weekday peak to more off-peak, night and weekend trips. That's good news for the bottom line as extra peak capacity is very expensive to add. At the least you're buying more trains and at worst you need more tracks. In contrast adding off-peak trips needs little more than some extra driver and maintenance crew hours.

Melbourne has especially big opportunities on evenings and Sunday mornings where gaps between services are typically only 30 to 60 minutes versus every 15 minutes. Better frequencies support major events and the diverse workforces needed to make them happen. There do need to be more driver hours but the incremental increase doesn't have to be large to make a start; fixing our notoriously poor Sunday morning train frequencies requires just six extra trips per week on our worst served lines.   

4. Make service better and watch usage grow

Whether it's roads or PT, if you make something better more people will use it. For PT that includes new rail lines, reformed bus routes or, for the simplest upgrades of the lot, higher frequency on already popular services. Service/patronage elasticity is generally higher during off-peak times, which bodes well for cost-effectiveness, as mentioned above. The Route 800 upgrade funded in the 2024 state budget is an excellent example of how weekend usage on a main road bus route can grow when timetables are upgraded. 

5. Value public transport as a great enabler / cost of living reliever

This includes matters such as expanded housing choices, access to education and jobs, being able to get to services such as health and wider social participation. It is highly space efficient at moving people and brings economic benefits to town centres. Melbourne is seen as a major event capital and improved public transport, especially evenings and early weekend mornings where services are both (a) sparse and (b) far less frequent than Sydney's would strengthen our competitiveness. 

Due to the cost of private motoring the most expensive trip is that which is either not possible or not reasonably practical by public transport. Boosting services and reforming networks can reduce the number of such trips and thus relieve peoples' cost of living pressures. 

6. Make bus reform your friend 

Poorly used bus routes may be 'hollow logs' in which savings may be found. An examination of bus route maps, timetables and productivity (ie boardings / km) stats may reveal cases where resources from low usage routes (especially in high income/low social needs areas) can be redeployed to routes with higher usage potential and/or which benefit areas with higher social needs. This can lead to 'greater good' benefits without a call on additional budget resources. Some roads have multiple overlapping bus routes that can be rationalised into fewer but more frequent routes for an overall simpler and more useful service. 

Some network reform has been done but not nearly enough. In 2021 the department released a bus plan. However they were unable to make the case for funding in either the 2023 or 2024 state budgets. Other cities such as Perth and Auckland, also with constrained opex budgets, reform bus network at much faster rates. If you want the most upgrades for the least cost bus network reform will need to be on the table, with what was done in 2014 being a demonstration of what's possible. 

7. Yes, we have a fare revenue problem

When budgets are tight not letting revenue you could have slip away is important. And DTP isn't necessarily the best at that nor even fully acknowledges the problem. For example their statistics on bus fare evasion are at odds with on-the-network observations. Certain decisions made fares easy to avoid and hard to pay, especially on buses. That may need a rethink despite the difficulties. 

The fare system needs renewed legitimacy and be generally regarded as fair to maximise compliance. Currently some short CBD or very long country trips are too cheap while some shortish trips are arguably too dear, especially for families. Some of this stems from political gimmicks like the Free Tram Zone. A strong treasurer may have to argue internally for reform that restores the progressivity of our fare system.  

In addition DTP management need to see themselves as the network leaders with a growth mindset for patronage. Likely tempered by the pandemic this has been lacking in the period since. This is mentioned here due to the revenue benefits of increased patronage. 

8. Think about how things could be done better with process reform

You're at a train station construction and each occupation has its own toilets (while the public may have none). At night there's PSOs milling around and station staff behind the counter while the nearby bus interchange remains strewn with litter and its seats are sticky. Have we gone too far with the division of labour and outsourcing? Why are internal DTP processes currently so slow that a level crossing can be removed faster than bus routes can be reformed? And why, relative to CPI, are we paying our private train and tram operators much more despite the number of trips operated not rising by anywhere near as much?

These are just a few questions one could ask about current processes and arrangements. By rights these are DTP's responsibility and there may be some very good answers. The incoming secretary Jeroen Weimar may be able to introduce some fresh thinking. However if value for money is suboptimal it becomes a financial matter, especially if the department hasn't been as active as it could be in maximising service per dollar on its own volition.  

9. Continue being aware of unrealised costs

Neither roads nor public transport recover enough direct user fees to pay for their construction and running. The lowest private and public cost modes per trip are active transport but state support overwhelmingly favours the motorised modes. 

Transport is full of unpaid for costs that get shifted to others. Much is associated with car use and parking, including uncharged use of public land for private benefit when there are better uses for it. Ecars address only one externality of private motoring and may make others worse. On this it is noted that the state government is winding back registration discounts.   

10. Tackle the seven ticking transport timebombs

Some more transport matters that were and remain urgent. I described them here. These may give an overview of some of the more challenging topics. 

That's my ten. No doubt there's more one could give. Anyway best of luck to Treasurer Symes - you'll have an interesting and challenging time ahead trying to juggle the many often competing requests. 

TT 197: Manningham's four big weekend bus upgrades and the slow dismantling of "austerity scheduling"

A modestly-titled item on the PTV website, titled Doncaster Park and Ride timetable changes, is proposing substantial (by Melbourne standards) bus timetable reform in the City of Manningham.

Not only that but it starts to dismantle an undeclared but pervasive timetabling doctrine that makes public transport less useful than it should be. Our practice of "austerity scheduling" holds that not even Melbourne's busiest trains, trams and buses deserve a frequent service for more than nine hours per day (typically 10am to 7pm) on weekends. This approach, based on official deprecation of non-peak travel ("nobody uses it") and the votes of people who do ("they don't count"), is so hard-coded in timetables that it deserves a name (the "nine hour rule").  

More on that later but let's first look at what's changing. 

280/282 ('Manningham loop')

Here's the bad news. These changes need to be paid for by a cut. The very long but poorly used Route 280/282 loop (known, when it started, as the Manningham Mover) will have its weekday frequency reduced from 30 to 60 minutes as per the advice below: 

What do I mean by poorly used? 2022 data I obtained from DTP shows that both 280 and 282 attracted around 6 boardings per service hour on weekdays. At around one-third the average for Melbourne buses this puts these routes in the quietest 10% of metropolitan bus routes. Part of this low usage is because parts of the 280 and 282 overlap other routes. This change reduces timetables but does not go as far as to reform the route to remove inefficient overlaps.  

The new timetables, which start on 12 January, are already on the PTV website. When you look up those for the 280 and 282 you will see that it's a Monday to Friday only timetable. Which is a problem given that 280 / 282 currently run seven days. One hopes this is a data oversight otherwise the 280 / 282 cut may be more severe than advised.

905 Pines to CBD

The biggest winner is the 905 SmartBus between The Pines and the CBD. This will have its weekend frequency doubled from 30 to 15 minutes over most of both days. This includes a span of over 13 hours on Saturdays and about 12 hours on Sundays.  

The upgrade makes the 905 bus nearly as useful on weekends as it is during the week. This is because so many more return trips won't involve one or both legs that require detailed planning (or a lot of luck with connections) to avoid long waits.  

907 Mitcham to CBD

The even more popular Route 907 also gains with a weekend timetable that's even better than the improved 905's. 

Since the timetable changes of late 2021, Route 907 has operated every 15 minutes on weekends. That got funded by cutting service on quieter routes (for which there didn't seem to be many complaints).

The catch is that the 2021 upgrade delivered frequent weekend service only over a 9 or 10 hour span, with a 30 minute service before and after. Unless you were making a short trip in the middle of the day there was a fair chance that either your away or return trip was going to be made when the timetable was a less attractive half-hourly service. In Melbourne that's not just a bus thing; the same approach is used when scheduling weekend trains and trams, particularly on Sundays. 

The new weekend timetables for both routes 905 and 907 largely fix that, unless  you are returning late. In both cases these add shoulder service so that the 15 minute frequency operates over about 13 to 14 hours, substantially equalising maximum waits with weekdays. As well it can be claimed that both these routes have better service levels than some train lines (eg Upfield and Glen Waverley) at pretty much all times. That's rare for buses in Melbourne.   

The all-pervasive nine hour rule 

You need to have reviewed all weekend train, tram and bus timetables to appreciate how rare having high weekend frequency over a 14 (or even 12) hour span is in Melbourne. That is even if you draw the line at every 20 rather than at every 15 minutes (which not even US transit planners working in their service-starved cities consider frequent). 

Across all modes weekend timetables on the best served routes build up to reach their maximum frequency around 10 am. This typically continues until 7pm with a sharp fall-off after. That's a span of nine hours, hence me calling it the "nine hour rule". It is rarely challenged since inertia is a more powerful force in service planning than patronage trends or community needs in Melbourne. 

Depending on when you want to start your history you could say it dates back to postwar evening and weekend tram frequency cuts, the 1978 evening rail frequency reduction or when decisions were made on the time-span of off-peak and Sunday frequency boosts in the 1990s. 

The original 1996 Kennett government promise was for Sunday train and tram timetables to be the same as Saturdays, all day. This was partly delivered in 1999. Then Saturday frequencies were applied to the Sunday timetables on all metropolitan train and tram lines for the busiest nine hours of the day (10am - 7pm). 

Outside those times service remained at the then existing 30 to 40 minute frequencies. One cannot be too harsh on this as that doctrine was an improvement in its time; on trains and trams it replaced another to the effect that "no one uses public transport on a Sunday and thus minimal service shall be provided". For buses then it was often still a case of no service. That became 'minimum service standards' from 2006 when the MOTC plan funded many (but not all) bus routes to operate at least hourly until 9 pm, 7 days a week.  

The nine hour rule persisted when weekend train services to Ringwood, Dandenong and Frankston were improved from every 20 to every 10 minutes in 2012. The previous 20 minute frequency applied over a 9 hour span, as did the 10 minute frequency that replaced it. The lower frequencies outside that 9 hour window remained unchanged. Thus if you look at a Dandenong line weekend timetable (the network's busiest) there is a sudden cliff from 10 to 30 minute frequency around 7pm. Ditto for Ringwood. Only the politically marginal Frankston line (the quietest of the three) got maximum wait improvements, about nine years later. 

This pattern has persisted in the 2020s with the evening and Sunday morning upgrades (to 20 min) on the Werribee, Williamstown and Frankston lines in 2021 being partial exceptions. They reduced maximum waits over a wide span but the Frankston line's 10 minute frequency still obeyed the nine hour rule. Every crowded after 7pm tram, train or rail replacement bus is testament to the influential falsehood that "no one uses public transport at night and thus minimal service shall be provided".  

The 10am to 7pm thing even carried across to buses in the occasional cases they got upgraded frequency, such as with the Route 234 and 907 improvements in 2021. Weekend service outside those times generally remained every 30 minutes. The result was that while people making short midday trips enjoyed the frequent service there and back, trips with longer time away (including work commutes and many day trips) often had 30 minute gaps apply for one and sometimes both legs. 

The nine hour rule is unique to Melbourne. It lags Sydney, whose population we have just exceeded. Sydney trains operates a twice as generous nineteen hour rule at most stations, with their 15 minute frequency applying from before 5 am to midnight, even on Sundays. Almost no matter what trip you make you will have a frequent service. Their 'All Day Frequent Network' buses vary more (10 min day, 20 min night) but the latter is still better than the 30, 40 and even 60 minute gaps that Melbourne's key routes have at certain times.

Auckland and Perth, both smaller than Melbourne but with conspicuously active and competent service planning cultures, have cultivated their frequent networks such that they operate at least 12 and sometimes 14 hours on all days of the week. Even US cities like Atlanta and Los Angeles, not know for their transit-mindedness, have better evening rail frequencies than Melbourne. All this adds up to Melbourne being an outlier with regards to its restrictive nine hour rule though change may now be in the air. 

The first welcome sign that, after a quarter century, Melbourne is taking steps to progressively ending its prohibition of frequent weekend service outside the narrow 10am to 7pm band came from an unexpected place. To be exact the revised December 1 Geelong V/Line timetable that I discussed here. This is based on its weekend 20 minute frequency operating between 7am and 9pm, ie about 14 hours. This is better than both the Geelong line weekday pattern and even most Metro train lines. 

As noted before, route 907 already operates every 15 minutes on weekends, though again only over a 9 or 10 hour span, with a 30 minute service before and after. The new weekend timetable adds shoulders so that the 15 minute frequency operates over as long as 14 hours, substantially equalising frequencies with weekdays.  

Thus it, and the 905 timetable, have given sustenance to what could be a new rule; one that allows frequent service to operate for around 14 hours on Saturdays and 12 hours on Sundays.

All eyes will be on future timetable releases (including those related to the Metro Tunnel) to see if the concept of the 14 hour (or better) rule takes hold beyond these few examples. But for now they have set a welcome precedent that frequent service should operate from 7am to 9pm on all main routes across all modes all week. All that remains is for timetables for other routes to follow suit, a process that will take decades at the current rate of timetable reform. 

284 Box Hill - Doncaster Park & Ride

This Monday to Saturday routes gains a new Sunday service.

Route 284 is one of a growing number of bus routes that is more productive on Saturdays than it is on weekdays. This bodes well for strong usage on Sunday as routes with high Saturday usage are typically also popular on Sundays. Box Hill is a particularly strong destination on Sundays. 

Like with the recently upgraded Route 612 (which also serves Box Hill) the new Sunday timetable is just a copy+paste of the Saturday timetable. This means an hourly 7am to 6pm service. Thus it's 7 days but doesn't meet the 9pm minimum service standards finish usual for local bus routes.

Route 284 isn't the busiest Melbourne bus route that lacks 7 day service, though it is in the top 50 per cent. It has the benefit of being operated by Kinetic who run all the routes whose timetables are being reformed here.

More than most other bus operators in Melbourne Kinetic's network had some 'fat' of well serviced but poorly used routes from which service kilometres could be pulled to boost service on more popular routes. This is partly a result of its government bus heritage (timetables on its routes were cut less during the big timetable purges of the early 1990s than on private operator routes) and subsequent service uplifts as a result of a political wish to 'do something' if governments weren't willing to fund rail to Doncaster.

Reallocating bus service kilometres within the one bus company (and especially the one depot) is far easier than transferring resources from one bus operator to another that true needs-based bus planning would entail. Hence though the 284 upgrade is still good, the confining of resources within artificial bus company boundaries presents an opportunity cost. Here it is that routes in less well-off areas, like Glenroy's 536 or Dandenong North's 802 and 804 continue to lack 7 day service despite having a stronger equity and patronage case to upgrade sooner than the 284. 

285 Doncaster Park & Ride - Camberwell

Many similar comments to 284, though usage isn't quite as strong, particularly on Saturdays. Thus of the two routes the 284 is likely to be the busier one on Sundays. 

The timetables show that weekend service patterns are similar to the 284, ie hourly until 6pm with the same times applying both days.  

Other timetable and infrastructure changes

Route 207 gets a small cut. Currently Night Network trips extend to Box Hill. This is different to the Doncaster terminus used on all other trips. This change standardises the pattern so that all trips finish at Doncaster. 

The consequential infrastructure change is the closure of Doncaster Park & Ride for three years and associated rerouting of buses. Those who use Park and Ride to change between routes may be walking further to make their connection. This is important as some routes, like 284 and 285 terminate at Park & Ride which is otherwise not a strong terminus. 

Unfortunately the PTV website is vague on the new stopping locations - they should really have published a map to explain. More on how PTV could have done better in this Philip Mallis item here. 

Opportunities for more?

These are just timetable changes. The area has significant route alignment issues that these changes don't address. This affects some route that have quite high patronage potential. Examples include the following: 

* Route 279. Opportunity could have been taken to remove the Blackburn deviation in this round of changes. This would provide a consistent 15 minute weekday frequency along its trunk. Ultimately you would also want to remove the hourly Templestowe variation but this may need to be done in conjunction with reform to other routes. 

* Route 281 and 293. Scope exists to consolidate into a single route for improved legibility. This has other dependencies, eg local routes around Greensborough/Eltham and (especially) the alignment of the SmartBus 901 and 902 (both of which are suboptimal since neither, unlike the 293, offers a direct Shoppingtown - Greensborough connection). 

* Routes 280 and 282. While their timetables got cut the routes did not. This is despite significant overlaps with other routes. Only tourists go around in circles, not people getting from A to B, so there is little value in the route remaining circular. Future reform will surely have to split this into shorter routes that are easier to understand and don't duplicate others. These routes are unlikely to have high patronage potential but reform to them could facilitate the emergence and further upgrade of routes that do. 

Summary

This is a good set of changes that are not without pain but will advantage far more trips than they disadvantage.

They also offer an insight into current DTP bus planning thinking. Because there is no publicly available service planning manual nor bus reform implementation plan this information can only come from watching what they do (or don't do). 

Themes that arise include: 

* More 7 day frequent routes: The upgrade of 905 increases the number of Melbourne bus routes operating every 15 minutes or better during the day from four (234, 246, 732 (short section only) and 907) to five (In contrast Auckland, a smaller city, has 40, working at ten times our rate to roll them out). 

* A new 14 hour rule: Frequent weekend service on main routes should run over a 12 to 14 hour span, and not just 9 hours, such as has restricted service across all modes for years. 

* Saturday and Sunday timetables should be the same, at least for local routes: This has applied for new or upgraded routes 612, 452 and now also 284 and 285. The most notable effect of this is that Sunday service starts around 7am instead of the more typical 9am as is the pattern of minimum standards upgrades. 

* The 2006 minimum service standards are not sacrosanct, especially with regards evening finish times: You can see evidence of this in that the upgraded 612, 284 and 285 got Sunday daytime service but not evening service extensions. Thus they remain with finishes around 6 or 7pm. Even the upgraded Route 800 has a final departure before 9pm on Sunday from Dandenong.       

* There is greater boldness at reducing service levels on poorly used routes like 280 / 282: A less bold approach could have been to reduce service from 30 to 45 minutes. However a bolder cut to 60 minutes was made, allowing bigger upgrades on busier routes. This point would be stronger if it is true that 280 / 282 lose weekend service as the new timetables as currently available seem to indicate. 

* Service trade-offs are kept within the same operator and preferably within regions: This is apparent in that the resources freed from the 280 / 282 cuts were reinvested into Kinetic routes like 284 and 285. Though upgrades to these are welcome, routes in other areas have higher justification for 7 day service on both patronage and social needs grounds. But because they are not run by Kinetic they miss out. 

* Even small route reforms remain hard: All but one of these changes (the 207 Night Network variation) are timetable only or are necessary to reflect the closure of the Park and Ride. Even if still gradual a faster rate of network simplification would be desirable especially if they can be made cost-free. For example removing 279's Blackburn deviation. 

* One arm of DTP/PTV doesn't know what the other is doing: Text, maps and timetable data appear done by different departments. Hence website text may not include maps or videos even if essential to help passengers navigate a change to routes around a key interchange point. And even if timetable data is online there may not be reference to it in the text. Neither is there necessarily cross-checking to avoid salient points being missed (eg 280 / 282 timetables being out of kilter with the text).   

Other Timetable Tuesday items are here