I have posted a number of times in the past about why I think that widening roads is often a pretty pointless exercise – due to induced demand. This is the process by which expanding the supply of road capacity encourages more people to drive at peak times, thereby actually encouraging more use of the road and negating much of the supposed benefit that widening the road should have created. So what can be done to reduce congestion and actually squeeze more out of our transport system?
The obvious answer is a greater focus on public transport, but as I spend most of this blog banging home that message I will take this opportunity to look at ways we could squeeze more out of our roading system. On arterial roads there are plenty of opportunities to expand bus lanes, transit (carpool & bus) lanes and so forth, but often it is considered that motorways are simply the realm of people driving their own cars with no passengers, oh and trucks of course. However, it doesn’t have to be this way – and many overseas cities have looked at ways of shifting more people rather than just more cars along their motorways. Another idea has been to create toll lanes on motorways that would have people paying a variable toll on certain lanes of the motorway to effectively make it possible to pay one’s way out of congestion.
There seems to be two main approaches to “squeezing more out of motorways” that can be seen across the world – most particularly in the USA. These are high-occupancy vehicle lanes, and toll lanes. The image below actually shows both of these measures operating side by side, in Denver:
Starting with high-occupancy vehicle lanes (HOV lanes, often known in New Zealand as T2 or T3 lanes), these are lanes that you are only allowed to drive along if you have two, three or more people in the car (depending on the type of HOV lane). There are already a few HOV lanes in Auckland, with Onewa Road (T3) being probably the best known. This lane has proven to be quite effective in offering people who use Onewa Road on Auckland’s North Shore with a quicker option to the highly congested main lane if they are carpooling. The carpool lane is shared with a great number of buses, which helps contribute to Onewa Road being one of the busier bus routes in all of Auckland. There are a number of other T3 lanes around North Shore City, while quite a few motorway onramps with ramp signals have T2 lanes that bypass the signals.
Toll lanes obviously work a bit differently, in that you effectively pay to drive in a particular lane of the motorway – with the toll often varying according to the level of use in that lane so that it does not become congested. There are no toll lanes within New Zealand at the moment, but quite a few operate within the USA – with perhaps the “91 Express Lanes” in Los Angeles being the most well known. These lanes (two in each direction) operate down the middle of the Riverside Freeway in Orange County, California and are controlled by a varying toll of up to $1 per mile – or around $9.90 for the entire length of the lanes during the time of highest demand (4pm-5pm in the eastbound direction). The lanes have operated since 1995, and seem to have been reasonable successful – at least to the extent that we’re now seeing many other US cities establish these lanes, particularly around the San Francisco Bay Area.
I’m a big fan of expanding HOV lanes, particularly if they are established by turning current general traffic lanes into HOV lanes (and not turning bus lanes into HOV lanes as Auckland City Council have proposed for Tamaki Drive). Carpooling can be a very simple and effective way of increasing vehicle occupancy rates and therefore reducing the number of cars that are required to shift people along a transport corridor. Carpooling is also apparently very efficient and potentially environmentally friendly – as you don’t end up with empty runs as can happen with buses and trains during lower demand periods.
So one way that I think Auckland could improve the efficiency of its motorway system could be by turning one of the lanes on each of the main motorways into an HOV lane during peak hour in the peak direction. This would only really be feasible on stretches of the motorway with at least three lanes in the direction being travelled (so you would be left with at least two general lanes plus the HOV lane) – but that would actually encompass most of Auckland’s more congested sections of motorway (another sign of induced demand that our widest bits of motorway are the most congested?) The map below shows the stretches of SH1 and SH16 that currently have more than two lanes in each direction, and therefore are candidates in my opinion for the outside/fast lane being turned into an HOV lane:
The lengths of motorway coloured green are those parts wide enough for one of the lanes to be turned into an HOV lane. Now I realise that this would be a pretty dramatic step for NZTA to take – but really when you consider that it is going to cost over $800 million to simply add one lane each way to the Northwest Motorway between St Lukes and Westgate, perhaps it is time we looked at getting more out of our existing motorway lanes?
In terms of toll lanes, I will take a bit more convincing when it comes to the merits of these. The Wikipedia page for toll lanes (sometimes called HOT, high-occupancy-toll, lanes) identifies some criticisms that I think are pretty valid:
Because HOT lanes and ETLs are often constructed within the existing road space, they are criticized as being an environmental tax or perk for the rich (“Lexus lanes”). Those who criticize the concepts claim that the lanes provide congestion relief to the motorists of a higher socioeconomic class. With HOT and hybrid lanes, the attempt to address this criticism typically consists of special treatment for HOVs. Personal vehicles carrying more than a specified amount of passengers (typically two or three) are permitted to use the HOV lanes at a reduced toll (hybrid lanes) or for free (HOT lanes). Additionally, public transit vehicles are typically exempted from the toll. A counter-argument is that the rich often already have ways to ease their commute that are not available to the poor, such as buying a home closer to where they work.
Nevertheless, toll lanes are really taking off overseas and if social equity issues can be overcome, they could be an effective way to reduce congestion for time-sensitive transport requirements at a relatively low cost.
I think another issue is the practicality and efficiency of developing toll lanes (although this also applies to HOV lanes). Are the advantages worth the cost of the physical works to actually build these lanes? Do the lanes have their own dedicated ramps? How do you manage things to ensure people only drive in the HOV lanes when they have two, three or more people in the car and so forth? These are some interesting questions to answer, but with the cost of building additional general lanes becoming higher and higher, I think that we will have to look at these options sooner rather than later.
Processing...
I think a T2 lane during morning and evening peak in the direction of peak flow isn’t a drastic step and would be a positive thing…
It’s a bit weird seeing a motorway post on this blog but good, Auckland is going to have to squeeze everything we can get out of our transport system by 2041 and given the amount of motorways and arterial roads we must start doing the roads we have better…
I’m not opposed to using existing roads smarter, just generally opposed to building more roads unless there’s a compelling argument to do so. I’m reasonably supportive of the Victoria Park Tunnel project, for example.
T2 lanes in Mana just north of Wellington have failed to work, drawn high costs and confusion. I’m worried that this fall back will have major implementations on setting up more in Wellington and be a bad look on the rest of the country as they were operated by NZTA. If there were more T2 lanes then people would be more aware, so it would be less confusing.
I’m interested in how they check each car for a 2nd occupant. Because in Wellington they have two people standing on the side of the road, ready to give send out warning notices to offenders! (Warning notice? – but no fine!)
People are used to the Onewa Rd T3 lane. It has been operational for many many years and people do get fined.
The major problem with the critisim opposed to toll lanes is that it is focusing on individuals traveling to and from work, what about buisnesses who need to get around faster to increase productivity, these lanes would be great in boosting the economy as these are the people who really need to get around by private vehicle use. Although I do agree that a toll lane would work more effectively if combined with a T2 or T3 on the same route eg. your example above.
I tihnk that having a T2 lane during peak time that turns into a truck only or toll lane off peak (9am-4pm Monday to Friday) could work. It would prioritise economically important trips during off peak times (when some motorways can still be very busy).
I wonder how many people would carpool to save 10 or so minutes of their trip? It seems like often there is a belief that carpooling is not possible for many.
Writing from Seattle, where as an Auckland based transport person I am viewed as something of an expert (definition of expert: someone who is some distance from home). I am a carpooling advocate, or as the NZ Herald once put it, a carpooling activist. I am here to speak at a conference on transforming transportation systems. The agenda is at:
This post is of great interest to me because Trip Convergence is about squeezing the maximum out of the transport system. Our approach is a concept that we call ‘flexible carpooling’, that makes carpooling a convenient choice by taking away the need for pre-arrangement. Check out our for details.
The technology we have developed to record participation in flexible carpooling also enables recognition of carpoolers by whatever means they formed, and provides the possibility of rewarding people for sharing rides even when there is not an HOV lane. If the reward is large enough we believe larger numbers of people would share rides, and so contribute to a)reducing the traffic and b)increasing ‘people throughput’. Imagine an ‘inverted toll’ point on the Harbour Bridge where carpoolers get a bonus for sharing the ride.
Our goal is to make it easier and more rewarding for people to share rides. We have no argument with public transport, just that we cannot see why the first solution chosen to solve congestion is to add more vehicles, especially when there are usually nearly three empty seats in each of the existing vehicles. Our objective here in the US is to secure funding to test some of our ideas (and some ideas of other people) for the purpose of increasing the effectiveness of the transport systems. It is a global problem.
We are making some progress. We are carrying out a feasibility study funded by the US Transportation Research Board to use flexible carpooling to deliver more riders to transit stations. And the Washington State Legislature has committed funds to a ‘flexible carpooling pilot project’ to help reduce traffic on the SR520, a particularly congested transport corridor.
What we fervently wish is that we could also get support for testing this in the city where we developed it: Auckland. We estimate that we could remove 5,000 vehicles from the busiest routes, and 11,000 if there was a network of HOV lanes such as your post suggested. In fact, without adding flexible carpooling meeting places to make it easy for people to form carpools, I suggest that your idea of converting general purpose lanes to HOV lanes would be a failure. In many parts of the world the trend is to convert HOV lanes into GP lanes because the HOV lanes do not get enough traffic. Also, any HOV lanes should be HOV3 (or T3 in NZTA’s parlance), simply because if you make it easy to form carpools, you want to take as much advantage of the capacity increase as possible.
Thanks for raising this interesting topic.
On the matter of encouraging business to use motorways outside peak times, and thinking out loud here, I wonder if tolling large vehicles during peak times (say 8-9am, 5-6pm) on motorways would be worth consideration.
George, I think it would perhaps be better to toll those that don’t necessarily need to be there.
You could argue that due to how badly rail has been maintained for the last 20yrs in this country that trucks need to remain, however everybody who is commuting does have the option to bus/train/walk/cycle to work, so why not actually toll those causing the bulk of the congestion, single occupant vehicles that arn’t transporting goods.
Encouraging a four day working week would in theory reduce commuter traffic by 20%. Would also save a lot of carbon (less fuel) and would allow people to be more involved in school and community events. Down side is who wants to do a ten hour day? But would definitely take the load off our road and bus system.
“Encouraging a four day working week would in theory reduce commuter traffic by 20%.”
I think this is a good example showing that you can use statistics to prove anything! 20% of the work week is one day, so of course commuter traffic will drop by 20%.
This will, of course, do nothing to reduce congestion on the remaining four days of the week – four those four days the roads, trains, buses are used just as heavily.
@Topcat, I’d do four 10 hour days if I got a three day weekend every week, that is for damn sure…
If people were split between having Mondays and Fridays off that would mean you would only get regular traffic flows 3 days a week. I would work 10 hour days to get a three day weekend for sure.
But you could easily create the situation where people are happier to drive longer distances or spend longer in traffic. If someone has to travel 20% less often each week, then they might be happy to drive 20% longer each time.
If everyone had a three day weekend every week, I bet you would see a big increase in the number of people commuting from lifestyle blocks and rural coastal areas.
I worked for a company where a ten hour day, four day week was available. Surprisingly few people took it up and of those who did a significant number opted out after a while. Also available was a four and a half day week. Again not popular. Humans are essentially social (herd) critters and this includes things as basic as, among many other things, start/finish times at work. This is something that Liberty Scott doesn’t understand when he pushes for staggered hours.
@Nick R, I dont think that is a fundamentally bad thing, people spending more time with their family is good in my opinion however as long as congestion eases and overall car use goes down, bring it on…
@Ian, the government could regulate a 4 day working week…
It is an interesting idea, although one imagines that if it was really so fantastic we would have done it already.
I think that it’s likely that generally having everyone at work on the same days at the same time has been found to be the most economically efficient way to operate – and even though it means you have to overbuild your transport networks to cope with the peaks, overall it is worth it.
Jeremy, if people spend a longer time travelling less often in place of a shorter time more often, there might not actually be any net gain in terms of spare time or congestion.
I don’t buy the argument, “if it’s a good idea someone would have had it already”… I think it would merit a study…
The Trickster “everybody who is commuting does have the option to bus/train/walk/cycle to work”, I don’t have that option and neither do many.
oh and with the 10hour a day, I already do that, however I guess i’m still working 5 – 6 days a week. Staggering working hours does work in terms of productivity, however I can’t imagine people sitting behind a desk in the city wanting to give up their accustomed timetable, so from a transport point of view would not work very effectively.
Joshua, why not?
How far is your commute?
I only commute from Epsom to mangere bridge, however from their I end up traveling either to the city or Newmarket or rema’s for sport so need to carry two sets of gear, (and yes one is cricket) so cycling is out of the option, and of course walking. Bus or Train does not service the area at the times I need. I either have to get to work the night before I start or be an hour late by bus, train does not service Mangere bridge as of yet.
Effectively I only have one option. Car.
There is little evidence that HOV lanes achieve any sizeable mode shift, what they do is avoid bus lanes being grossly underutilised. Most users of HOV lanes are couples or families that would have travelled together anyway. Unfortunately they rarely include HGVs which could really benefit from the capacity, since they have no alternatives. Although there is one in London I am aware of which is Bus/HGV (not HOV) and works well.
How would the motorway network become more efficient by essentially segregating a quarter to a third of the capacity to families, couples and the handful of those who will carpool, at a cost to other motorists, particularly freight? The only economic argument would be a proxy for willingness to pay, which would put buses, taxis, HGVs and LGVs in their own lane (all of those users have a higher value of time collectively than many car users), and a toll for cars based on capacity.
The Wikipedia page talks nonsense. The “Lexus lane” claim is false, as there is some evidence that those using toll lanes include people with two jobs, or who have appointments (medical, job interview, pick up children, flights), so value time highly. Why shouldn’t people pay a premium if they value time? The status quo is patently absurd, with Soviet style queuing on highways, generated enormous waste of time and fuel.
“Are the advantages worth the cost of the physical works to actually build these lanes? Do the lanes have their own dedicated ramps? How do you manage things to ensure people only drive in the HOV lanes when they have two, three or more people in the car and so forth? These are some interesting questions to answer, but with the cost of building additional general lanes becoming higher and higher, I think that we will have to look at these options sooner rather than later.”
You don’t need to build new toll lanes, you can retrofit current ones, but I support new lanes being tolled. They do not need their own ramps. Eliminating the HOV component makes enforcement easy. I strongly supported the Manukau Harbour Crossing project being tolled lanes, 2 in each direction, but Transit was not interested.
However, the bigger picture is to use Intelligent Transport Systems to give drivers of all vehicles information on routes, incidents and to plan trips, including changing mode, changing route and changing time. The technology is there to allow this, and simple things like variable speed limits so that traffic doesn’t speed into queues, ramp metering (which works well) and the like can make a big difference. I have heard several experts say that between a third and half of all congestion is related to incidents like accidents and breakdowns. Being able to respond quickly to these, and then being able to advise people as early as possible to avoid such locations can make a huge difference. 3G cellphones may well be the platform to do this.
Ian: You said “This is something that Liberty Scott doesn’t understand when he pushes for staggered hours” when what I am suggesting is that pricing effectively allows people to shift to the edges of peaks. I either leave early for work in the morning and early in evening or vice versa (or both a lot of the time). It makes sense to price both public transport and roads so that people who could travel off peak do, it just takes about 15% to move (time or mode or consolidate trips) and you can significantly relieve congestion.
Finally, it is worth noting that dual carriageways in rural areas are primarily about safety, not congestion. That’s the main benefit from the Waikato expressway projects that are worth doing (not all are).
My main doubts regarding this idea of T2 lanes on the motorway is whether there are enough people able to carpool that could take advantage of their dedicated lane. After all, we do want to increase the ‘person throughput’ of the motorway, not reduce it. Are there studies overseas showing how effective HOV lanes are in this regards?
There are such studies. And as usual with studies, they prove the point of the person doing the study, so some say that they increase the person throughput, and some say they do not.
Any lanes they change on the motorway should be T3 if they are ever changed, and provided with systems to make it easy to carpool.
I think a bus and HGV lane is a good idea…
What is HGV?
Fair enough, although I remember as a kid regularly riding with the good ole ‘cricketbat handle out the top of the bag’ trick. Cricket is one sport that does make it harder although football etc are far easier to handle, then again I’ve been car-less for about 7 years now so I’m used to it.
Thing is with everything, if there is a will there is always a way.
HGV = Heavy Goods Vehicle (at least that is my understanding)…
Haha yea those were the good days, when u were able to fit all your gear in your backpack, you should see the size of my bag now.
Liberty, I have to agree with you on the tolling road, however I do think there is a point in T3 or T4 Lanes in particular as having 3 or 4 people in a single car can actually be more efficient than a bus load, thats energy wise anyway.
That’s for sure, unless the bus is loaded to crush capacity, because the bus deadheads empty so only gets 50% impact for the fuel it uses.
I think you need to be careful about where you choose T2 lanes and where you choose T3 lanes. I think that it’s fairly likely that more people would be able to use a T2 lane, so therefore you should be wary about using T2 lanes where you’re trying to operate the lane as a bus lane. This is why Onewa Road works well as a T3 lane – because it’s primarily a bus lane but also allows a few extra cars on it.
T2 lanes are good where the main use otherwise would be general traffic – like a motorway lane for example. You don’t worry about there being plenty of users because you’re not trying to run it as a bus lane.
One thing to be wary of with carpool lanes is the “I’ll drive you to work” factor. That effectively results in an occupancy rate of 0.5 as it takes two trips to get one person to work.
What everyone always proposes is a standard motorway with specialty lanes. Whether HOV, Toll or Hybrid the idea is the same: go out of your way to meet special criteria and get a special lane, or don’t and don’t.
The problem with this equation is that the BENEFIT is invisible. Here as outside observers we see the benefit: you’ll save some time… but how do you quantify that benefit? Does it really matter to a driver? Is it really worth going out of their way for? The answer is not a clearcut “YES”.
They say “Sure I could carpool, but what would I save? One minute? Five?” They don’t know because they have never tried it, and they will not try because the perceived benefit is low. Hense no modal shift.
If you want to see a modal shift, here’s what you do:
Make all lanes toll lanes EXCEPT the T2 / T3.
Suddenly there’s an immediately quantifiable benefit for your average commuter. “If I don’t carpool it costs me money”. Loss minimisation is a powerful driver!