Follow us on Twitter

Car vs Train Off Peak

Earlier this year I did a comparison between catching a train from my local station and driving to work at peak time. This is a similar look at the journey off peak. I had arranged with some friends to meet them for lunch today in town and the experience highlights perfectly just how far we have yet to go to truly have an alternative to getting around by car.

I was meeting my friends at 12:00 near the bottom of Albert St which is just a short walk away from Britomart. As I have mentioned before I live near a train station and only have a 5-10 minute walk to get there which is great. The first thing I did was to have a quick look at the timetable shows that I was in luck as despite having a pretty crappy off peak frequency of only one train every half hour, one was due to arrive at Britomart at 12:00. Being only a couple of minutes late by the time I walked to my destination would have been ok but the real problem is at the other end where I catch the train, Sturges Rd. You see that train leaves my station at 11:13 so combined with the walk to the station, I’m looking at close to an hour all up to get into town. On top of that I also have the train fares to consider which with an AT HOP card come in at around $6 each way. I also knew I would be there for about an hour or so had a look at the first train to leave after 1pm, it was due to depart Britomart at 1:23. Add on another hour and a further $6 to get home and I’m looking at almost 2½ hours travelling and $12 all up.

By comparison my other option was to drive.  The presence of the North Western motorway means an off peak trips only takes 20-25 minutes to get to Auckland Transports downtown carpark. The new parking system introduced late last month for the city centre means that the AT controlled car parks only cost $3 per hour off peak and so knowing I would be just a little bit over I budgeted for $6 in parking costs. There would also be roughly $7 in petrol and other associated costs.

At a most simple level the decision came down to the difference between spending 2½ hours travelling and paying $12 to do so or hoping in my car spending an hour (or less) travelling and paying ~$13 in parking and fuel costs. So the costs are similar but there is a huge amount of difference in the amount of time needed for each mode. Even as someone who is very enthusiastic about public transport its very hard to justify using it in this situation (at the peak is a different story).

I like to maintain that most people are pretty rational and will make generally logical decisions so taking taking 1½ times longer to make a journey via PT is never going to be attractive. This is why it is so important that Auckland Transport works to find ways to improve PT infrastructure as a key way to make PT a more attractive option, both on and off peak. By my estimation, with the CRL my trip would have been around 35-40 minutes including the walking, combining that with the higher frequencies proposed in the RPTP and it would have made a huge difference in my decision because there is always that risk of random congestion or an accident that holds things up.

On the issue of parking though, it would have been so much easier and more useful if I could have paid for my parking with my AT HOP card. Lets hope AT can get on to that once they have got the system rolled out to buses next year and it would also be a good way to get the card into the hands of drivers making it easier for them to convert to using PT in the future.

80 comments to Car vs Train Off Peak

  • pete

    Just a thought, could you do the “paper” based exercise pre CRL but post RPTP and see what the interim timings might be?

  • Dan

    $7 for that journey (36km) allows about $0.19/km for costs of driving. Most economists would say that’s a serious underestimate.

    It still doesn’t detract from your main point, though, that the journey is way more convenient in the car than by transit.

    • Yeah doesn’t the AA recommend around $1 per km?

      • Dan

        From the IRD: The mileage rate for motor vehicles is 77 cents per kilometre.

        By which measure Matt’s trip cost him about $34! (still worth it in time savings…)

        • Isn’t that taking into account things like the cost of vehicle ownership, servicing etc. Those are things I am paying for already regardless of if I had driven today or not.

          My car is fairly standard and does about 550km on a tank of gas which is ~55L, thats 0.1L per km. 36km would then use 3.6L of fuel at a price of something like 2.07 per L last time I filled up means ~$7.5 worth of fuel.

          • Robert

            Depends how you view it. You’re adding wear and tear to your car, adding km’s, that’s what the 77c takes into account.

          • Dan

            Right, but the costs of owning the car need to be taken into account somewhere. Also, just owning the thing isn’t that expensive — there is some depreciation and a few fixed costs (insurance, reg) — but most of the costs come from using it.

          • No Dan the biggest cost is the capital tied up in it, unless it is very old and cheap in which case the operating costs are likely to be much higher…. It’ll get you one way or the other.

          • Yes I was initially thinking roughly $6 of fuel and another $1 for wear and tear. A quick calculation, if you did say 10k kms per year and had an average of $500 per year (my car is about 5 years old and has never had a bill that much yet) for wear and tear/maintenance you would be paying ~5c per km. That would work out to roughly $1.80 for this 36km journey. Perhaps instead I should say it cost $9 in fuel and other costs, $7.5 for fuel and another $1.5 for maintenance.

            What this really highlights is the massive hidden costs associated with car ownership. Imagine how people would change their travel patterns if they actually knew how much each trip cost them?

          • Yes and the savings really accrue to a place that supports less can use and lower rates of car ownership. Because if households are able to say go down from two cars to one, or three to two, they will have considerably more disposable even after paying our very expensive Transit rates. And that money is likely to be spent locally, and not go to Japanese car firms or the back pockets of Arabian princes.

            Example from US: from 2005-2009 Washington DC (which has a good subway system) population grew by 15, 862 but car registrations fell by 15, 000. This reduction in car has been calculated as as much as 127 million being retained in the local economy each year.

            Page 31 Jeff Speck Walkable Cities 2012

          • Usually the biggest car cost is depreciation.
            My rule of thumb is generally to double petrol costs to cover wear and tear, registration & WoF costs and depreciation etc.
            But the AA or IRD’s cost estimate is probably better if you’re doing it properly…

      • pete

        AA quotes total all up cost of around $0.50 per Km for average small family car. So for return journey it would be around $20 in car costs. Either way with average NZ wage at $20 hour, and city workers generally earning more than that, saving $20 but consuming more than a hour on slow PT is a silly waste of a life

  • pete

    This is starting to sound like all the network planning should pay off but the small things aka bike parking at each station help build the case to improve travel times too.

    • Bike parking would be very easy to implement at the station too. There is a 170 space park n ride and the few times I have been there during the day there tends to be ~50 carparks empty. Taking a few away and putting in a protected bike storage area would be great (I don’t have a bike so am limited to walking or driving there).

  • Dan

    It hasn’t had a big bill yet, but it will one day! But yes, there are huge upfront costs in car ownership that people never take into account. (even the cost of having to make room to store one on your property is massive — but you get that whether or not you own one).

    And Patrick, it is the depreciation that costs, not the capital. Annual cost of capital on a $15k car is not much at the moment, certainly less than $1000. But every km driven makes it worth less.

    • Dan

      Gah! was supposed to be in the thread above…

    • Oh yeah, and the opportunity cost of, say, not being able to say have that figure off your mortgage.

      • Greg N

        Patrick,
        But did you consider that by having a car and living further out your mortgage is way lower (lower than the car cost by many a factor usually even when factoring in the owning and driving costs of said car),

        So in the past owning a car and living in the ‘burbs was seen a (very) cost effective way to lower your mortgage.

        Perhaps this is why we have such a track record of urban sprawl – unless you price the time spent in a car travelling to and from your lower cost house in the ‘burbs, you will seem to be better off by doing so.

        • bbc

          A house out in the suburbs and spending an hour driving each way each day only appears cheaper if you ignore the waste of 2 hours of your time a day, not to mention the elevated stress levels which can have all sorts of negative health outcomes.

          • Greg N

            BBC, Don’t disagree.
            It more pointing out that Patrick said not owning a car is a good way to cut your mortgage, which is probably true to a point.

            but just remember, when you bought your house out in the ‘burbs, it wasn’t a two hour each way commute was it? more like 30-45 minutes tops and petrol was way cheaper then too, and your mortgage interest rate probably wasn’t much below 10% – so it was by all measures cost effective to do this. Now that your have a two hour commute and no decent transport options you have to keep using your car regardless of petrol prices as you can’t (even now) afford a place closer to town. And wiser folks won’t buy your place as they see your folly for what it is.
            – so you have now been “hoist by your own petard” and are now a victim of the stampede to the ‘burbs you and your neighbours have caused by (short term gain, but longer term pain) using your car to lower your mortgage.

        • I think what Patrick is getting at is that money not spent on cars is money that can be spent on paying off your housing sooner, or conversely buying a nicer house.

          My mortgage is way lower purely because I don’t live in the suburbs. I could never afford the standard of place I live in if I was forced to pay for a whole plot of land for it to sit on. I share the cost (and value) of the land with 39 other owners and mostly just pay for the dwelling itself.

          I can also walk to work, the supermarket and most other things within ten minutes so I really don’t need a car. So that has saved me about five grand a year in the first instance, but it also allows me to rent out my parking space for $60 a week. The way I figure it going carless in the city has saved me $8,000 a year. That’s about a third of my mortgage payments, or in other words I could afford to service a place 50% more expensive by doing away with the car. Or another way to think of it, it’s the same as a twelve grand pay rise less tax.

          • But what if that car you haven’t bought is a Porsche: You’re saving hundreds of thousands! ……Another the glass of Krug?

            [another way of saying i am amazed at the sunk wealth in depreciating bits of expensive tin in Auckland]

  • Did you consider that you could do something useful on the train vs driving is essentially wasted time. Still probably doesn’t change the result though.

  • Adam W

    Your link to your other post on ‘peak’ does not work, are you are to update the post so I can have a read?

  • Matt Lee

    $6 each way fare is too high as well.

  • Rob Mayo

    Matt, you talked about how useful it would be for the HOP card to be able to be used as an additional payment method at AT-run car parking buildings in the CBD such as Victoria St and Civic etc. What’s the bet however that the HOP card can only be used with Thales equipment and not with the Skidata equipment in use at AT cark parks. It would be nice to say that the HOP card can be read by equipment other than Thales-produced devices – in particular by the Skidata barrier gates and payment machines but I wonder if this is possible in reality. Can someone out there give a definitive yay or nay on this?

    • Anyone is able to make equipment to that can read the HOP cards which are designed to a national interoperability standard that the NZTA is in charge of so it. I am aware that it is on AT’s wish list of things to do and I’m also aware that they are about to replace the on street parking meters (which are apparently obsolete and breaking down quite regularly) and the tender docs specify that they must be able to accept AT HOP cards.

      • bbc

        Would be a big saving for AT if the new machines were card only, be it HOP, credit, eftpos, or phone rather than coins which require constant checking, removal and refilling.

  • Well once we have the new capacity networks up and running; new trains, new bus network. Then it will be time to campaign for lower fares, both for improved equality of access but also in order to improve the farebox recovery. And particularly for lower off peak fares because of the low marginal cost of adding pax to those services.

    I am confident that with increased capacity and frequency a move to a lower fare structure would so substantially lift ridership to more than recover the drop in revenue received per trip. Actually I see little barrier to implementing lower fares off now, except perhaps that the services are so infrequent….?

    • Lower fares will come simply by not increasing them as network efficiencies reduce service delivery costs and increase utilisation. There would be the opportunity to actually drop them with a new integrated fare structure, but I think that will be designed to be fare neutral. We just need to get boardings up on the same amount of service working more efficiently, then there will be no need to increase fares over time.

      • Luke C

        The key should be do hold hop fares constant, but start ramping up non HOP fares over time.
        Also make it really easy to get a HOP card. I get an Oyster even when I was in London for a day or 2 as it makes things so much easier and cheaper. Also can get 5 pound deposit and all change back when returning it.

      • I think there’s two really big opportunities with fares. One; to differentiate peak and off peak, and two; to accelerate uptake and improve farebox recovery. The later is a hunch and needs analysis, but anyway, as I said above it is too soon as the services aren’t there yet.

        It is my view that ticket prices are high enough to keep significant new users from trying Transit and that along with the arrival of more and better services there should be a comprehensive change in pricing policy. My guess is we could be moving more people with at least no loss in revenue if not a substantial increase if done well and, importantly, marketed and timed well.

        Just waiting for inflation to make transit cheaper is to miss a huge marketing opportunity. Especially with off peak.

        • There are two sides to the ‘price of fares’ equation, how much a fare costs, and what you get for it. Personally I’d like to see an approach that keeps fares about the same but greatly improves what people can actually do with the fare, i.e. improving value for money rather than cutting the price. So penalty free transfer is the place to start, plus things like allowing a return on the same fare if the trip is short, or getting free travel in the evenings after paying for a commute.

          • Yes, good plan Nick, but why not look into what might happen with lower prices as well? We should not assume that there is a revenue advantage in what are, by international standards, high ticket prices; they may very well be costing us revenue. Wants a look in my view. Marketing gold, and timed with the big upgrade in service across the region and modes could lead help reinforce crazy growth in ridership and therefore greater farebox recovery, or at least have no negative revenue impact. And even if only revenue neutral this would greatly enhance the decongestion and accessibility economic benefits significantly.

          • Greg N

            Nick,
            I agree with Patrick on the off peak fares, and think we should do it sooner than later.
            If you are running trains off-peak then whatever you can do to fill them up with paying passengers is good.
            Since each fare paying passenger increases fare box recovery.

            Of course you’d need the EMUs operating fully to make this really cost effective with the lowered running costs of the EMUs over the older Diesel jobs we have now.

            But with AT HOP we have the mechanism to teach people to think about using the trains off-peak as well as on-peak, both for travel demand management (which we need to have in place in some form soon in some form until CRL is operating to handle the TPH restrictions at Britomart), and for secondary journeys during the day.

  • Richard D

    Even with electrification and with CRL, the real killer on journey time is the 14 stops all-stops service. Only with some sort of inner/ outer, all stops / semi fast mix of services can there be a serious reduction in journey time.

    New Lynn to Kingsland is 17 minutes – an average of 25.8km per hour. If the train could instead be fast with an average speed of 90kph (which in the grand scheme of things isn’t that fast) it could make that journey in 5 minutes. Assume a good fast alignment into and through the CRL and call that leg to Britomart 7 minutes. Assume Sturges Rd – New Lynn journey time at 14 minutes now, becomes a couple of minutes shorter with electrification.

    So a 47 minute journey time could become maybe a 24 minute one.
    Now we are at a comparable journey time with the car.

    Given that most people ignore the fixed costs of their car, it is a much likely proposition that people will choose trains over cars for the off peak journey.

    • Skipping stops doesn’t do a hell of a lot, it saves about 60 seconds per station dropped especially as trains need to slow through stations regardless. Dropping seven stations to save seven minutes would probably lead to a big net loss in passengers. Generally you end up spending longer waiting for a train that actually stops at your station than you save once you are on it.

      • Richard D

        If the station and layout are designed correctly, why should the train have to slow through the stations regardless? The UK seems quite comfortable with trains going through platform lines at 140kph or more.
        Whilst for the stations being skipped, there is no benefit, for the stations beyond, taking even 7 minutes off a 31 minutes or more journey time is pretty significant.

        • But why would an off peak service skip stations?

          Well there is one station that if skipped would make a difference because of the unusual amount of time spent there and that’s Newmarket. But it is also the second most important destination on the line after Britomart.

          So you’d only consider ‘direct’ Western Line services to Britomart if there were also new services to Newmarket. So only when we have more trains and as a way to add more, and different services, ones that have to avoid the Britomart bottleneck. I guess it is a possibility for running west/south services post-EMU but pre-CRL.

          This has been raised before; as say a Henderson-Manukau addition, but we are moving to a transfer based model so the general rule is to increase service with higher frequencies of the same highly legible patterns which make taking connections easier and easier, rather than adding different patterns.

          Perhaps, this might be a reasonable exception, while we wait for the CRL?

          Here’s a possible pattern [map by commenter Andrew]

          • Don

            Why not try to reduce the delay at Newmarket?
            At present the delay occurs because the LE has to walk from one end of the train to the other before setting off. Why not have a ‘spare’ LE that is already waiting at the appropriate end of the platform. Then all that is needed is for the arriving LE to step off at the same time the departing LE takes over at the other end. Although this may require an extra LE the difference in time between up and down trains is fairly close which would in little LE downtime.

          • Having off peak services miss stations would require (a) a lot of trains (needing a CRL and more EMU purchases quite likely, plus higher operating costs) or (b) fewer trains for many stations. While there is a case for stop rationalisation, this would have to be taken into account.

    • New Lynn to Kingsland is currently scheduled at 14 minutes and by my calculations it would probably drop to about 12 with electrification. I always thought that it would have been a good idea to drop a station by combining Baldwin Ave and Morningside into a station at St Lukes Rd but even that would only save about a minute. The only other one I think could be considered pre CRL is Mt Eden which still languishes. I don’t think that the train has to be down to exactly the same time as a car off peak but if it can be close (say 30 mins from Sturges to Britomart) then that would be enough for most people.

        • Richard D

          That and parking availability. If you knew there was doubt that you could park at or near your desired destination, there would be a greater number that would opt for PT and not car.

          • Another dividend of the Great Ak Transit Upgrade is we can put the city on a parking diet; just keep chipping away at the supply, especially of the space wasting, place ruining, transit slowing on street bays.

          • bbc

            I have seen quite interesting photos of Swiss, Dutch and Danish cities from the 60’s and they look eerily like Auckland with on-street parking everywhere and cars eating up all the (removed) cycle lanes. Took them a few decades to recover from that, and they did it as you say Patrick by slowly reducing parking much as Auckland is doing with the shared spaces. A key point however, is that many cities made sure to not allow additional carparks and in the case of Zurich everytime someone wanted to build one they had to offset it by removing a space somewhere else in the city – we could learn something from them, especially as Zurich ranks consistently as one of the most liveable cities in the world – for good reason.

    • Luke C

      Some work reducing dwell times might help too. Dwell times seem excessively long here. This requires Veolia doing their job, and bringing in worlds best practice operational efficiencies, which I thought was the point of competitive international tendering. Maybe efficiencies brought in with EMU’s will help alot here.

  • Luke C

    The real killer would be a North Western driverless metro, giving 15 min journey into town from Lincoln Road interchange. This would suck a bunch of CBD patronage from the Western line from Henderson and further West, and but train still would be busy for internal West patronage serving those going to New Lynn which would of become major regional centers.
    A NW metro would also mean that the Westgate Kumeu corridor starts to look very attractive for development, as opposed to the Drury Papakura corridor, as half the travel time to CBD.

  • I forgot to add my electrification time mentioned above includes a Parnell station which we obviously don’t have yet.

  • sting

    what would the comparison with the bus be for the same journey peak and off-peak?

    it points again to the problem that our train is trying to do two things at the same time. it’s trying to be a “suburban rail system” for the stations some distance from the CBD, but also trying to be more of a “metro” system closer in.
    the difference? you would expect a suburban rail system to stop at relatively few locations on the way in to a city, with stops maybe 5km apart. a metro system you would expect much more frequent stops, such as every kilometre.
    if we had a decent suburban rail system, a trip such as Kumeu/Huapai to the CBD should be approx. 30 mins (its only approx. 40km) with stops only at Henderson, New Lynn and Newmarket. However, because our system is trying to be both, such a journey ends up taking almost 2 hours! consequently, there is no public transport incentive to live in our outlying towns such as Kumeu, Helensville, Pukekohe etc. which there would be in comparitive cities.

    London is the classic example, with inter city and suburban trains able to operate independently due to multiple lines through stations, and certainly independent from the Underground.
    but because we only have 2 lines on our rail, we will never be able to operate such a system – so unless we sacrifice our rapid metro style service close to town, the service for stations further out will never compete with the car.

    • The bus from Kumeu/Huapai would be much faster, especially with bus priority on the Northwestern.

      Personally I think it would be foolish to sacrifice regular ‘metro’ service to the suburbs in order to give faster rail trips to the small towns on the fringe.

    • Not strictly true Sting. The Western Line follows a very rambling route, both with the crazy diversion to Newmarket and further out as it skirts around the inlet. The first problem will be sorted by the CRL but the second is a trade off, as you say, between serving the communities along the way and speed.

      The answer is is to serve the NW catchment with its own RTN, as busway on the motorway, so people there get a direct way into town and those on the western Line still get a service too.

      The Southern and Eastern Lines don’t suffer this problem nearly as much which is reflected in the boarding figures on the next post from the distant stations being very strong.

      5 km between stations? Well that wouldn’t serve the city at all, but privilege the ex-urban traveller and places over the closer in ones, which is exactly what the motorway system already does.

      The answers are to increase speed and frequency everywhere, which is coming with the new trains, and, especially for the Western Line: the CRL.

      My view is in fact the reverse of yours: Auckland needs a Metro system and not a Commuter one. And with the CRL and good frequent running patterns we can have one.

      • sting

        I agree with you Patrick. I wasn’t advocating for abandoning a more metro like system for a commuter rail one, i was just pointing out the difficulty of trying to make one system do both…and what this highlights is the difficulty of growing the city efficiently, as the out-lying towns will always be more reliant on the car to get into the city centre.
        other comparitive cities therefore have an advantage if they have a more effective PT system for such longer distances. whilst i am not suggesting we change our rail system to accommodate this, it does point to the need to think about how we service our hinterland towns, which will be essential to our growth as we push through 2 million people and so on.

    • Looking at MAXX the bus option would be slower.

      Post electrification and CRL, I don’t have too much of an issue with the current set up with perhaps the exception of one or two stations Getting a trip from Henderson to town reliably down to ~30 minutes by train is going to be pretty competitive when you take into account things like random traffic jams and the fact you don’t have to worry about things like parking etc.

  • The amount of time it takes to park and retrieve your car from a central city carpark is often overlooked I think. I remember when John Banks was championing a new eastern motorway, he claimed it would take 15 minutes to get from the central city to Pakuranga. My response to that is that was that it can take 15 minutes just to retrieve your car from a multi-storey carpark.

  • Andrew C

    You missed an option – ride there using the cycle network looks like approx 17.5K to bottom of Albert Park. 40-45mins each way zero cost and a free workout

  • Andrew C

    *Albert Street

Leave a Reply