Still can’t get there from here: A review of André Brett’s history of passenger rail in New Zealand since 1920
By Ross Clark
2021 saw the publication of André Brett’s history of passenger railways in New Zealand since 1920. As a start date, 1920 is a useful one to work from, as it is at a point when New Zealand was coming out of the First World War, so was returning to a “normal”, and at a time when cars and buses were becoming much more common.
Having read through the history, I am impressed by its coverage and presentation, and especially of the maps. However, there are aspects in which the history is lacking, and where inclusion of additional information and background would aid significantly in both under-standing the choices which were made at the time and making some of them defendable as well. One key weakness of the book, for example, is that nowhere is there a table which shows the components of passenger rail demand over time, although the actual numbers are touched on here and there. The changes over time in some of the time-series are quite spectacular.
First, there needs to be more discussion of the institutional environment after 1920, and especially in the post-war period
New Zealand was hardly alone after 1920 in lacking a way of planning its road and railway networks together; roads were planned by the Main Highways Board and then the National Roads Board, railways by the Railways Department and never the twain did meet. There was no opportunity for what today we would call joined-up thinking; there was no structure for it and certainly no culture of it in practice, and this worked against Railways’ interests. It also explains (at least in part) why in 1954 the Government opted for a motorway-based solution for Auckland’s issues (p160). The establishment of a unified Ministry of Transport in 1968 did not address the issue either. It did some overall planning in aviation policy and ports policy, and contributed to land transport policy, but that was all; the critical weakness was that the road management remained with the Ministry of Works & Development as-was, this partly because of some tortuous bureaucratic politics, I gather, and no love lost either between the respective agencies, as I found out years later. Essentially, there was no understanding of the need for a coherent strategy for all land transport, never mind all transport modes as a whole.
The establishment of Transit New Zealand in 1989 was meant to address the issue, by allowing comparison of road and rail projects together. That was the intention, anyway. However, a huge budget cut the following year stymied that intention from the start – meaning no money for public transport or rail investment, and precious little enough for roading, for the rest of the 1990s. There is more joined-up planning now, especially in Auckland, and there are attempts at joined-up thinking being made in Wellington, which really aren’t getting anywhere; but there is still nothing at a national level, as far as I can see. And it is not only land transport policy; there is now a need for transport as a whole to be considered together, aviation and marine as well as land transport modes.
A further feature of the institutional environment, and perhaps at the heart of the problem, is the model developed in the post-war period for funding roads. Here, there was the extensive use of “dedicated” taxes to fund road – the highways tax on petrol, and a variety of charges on truck operators. These taxes went into a fund out of which the road network was funded (100 percent for the State Highways, about 50 percent on average for local roads). After 1989, vehicle licencing fees were included as well, and the fund’s coverage was expanded to include paying for public transport operations and traffic policing. As a model for funding roads this has a great deal to commend it – it means that user charges work in much the same way as direct infrastructure charges elsewhere, if with a lot of averaging (a point I will come back to). Essentially, the model for funding was to use road user funds to secure benefits for road users, rather than the “community”, however defined, as a whole. It should be noted that the Government probably took more out of the sector financially than it put in; this was a bugbear for the AA for many years. When the funding of the New Zealand network is portrayed in these terms, it is quite clear that car and heavy vehicle traffic has not been given the network for ‘free’ or even ‘subsidised’, as many in the rail community seem to think.
However, this model had a series of unintended consequences for rail. As Treasury became more interested after 1966 in the financial outlook for rail (p188), they had the choice, as they saw it, between a roads network which was effectively self-funding )with other modes -shipping, aviation- seen as mostly paying their way as well), and a railway system they no doubt came to regard as a money pit – and there was absolutely no scope, then or later, to use road user funds to support the railway network. This was especially the case once it became clear, as it was by the 1990s, that big public transport projects were not going to yield that much in the way of benefits to road users. That they saw rail as a money pit also explains a lot of why Tranz Rail was privatised in 1993; the system had needed two huge refinancings in the previous years (1981 and 1992, more of which below), and they very much wanted to wash their hands of it – to the point that when rail was again facing another financial black hole, as it did by the end of the 2000s, they would have quite happily abandoned everything, out with the Auckland-Hamilton-Tauranga triangle – including the NIMT down to Wellington, and nearly everything in the South Island [i]. And although it was not involved to any degree in the major decisions, the Ministry of Transport was not exactly sympathetic to rail either.
The lack of enthusiasm for funding passenger services was not limited to rail. I was once told that during the 1980s, the Treasury representative on the old Urban Transport Council (responsible for funding passenger transport operations, later wrapped up into Transit New Zealand), would happily have cut support for operations completely. A further attempt was made during the early 1990s, when Ruth Richardson was Minister of Finance, to completely remove any central Government support for public transport operations. Given the mindset of most of New Zealand’s local government, this step would have doomed them instantly.
Second, many of the decisions which affected the passenger operation were as a result of changes in the wider freight market
Historically, railways in New Zealand were seen as a freight operation with a passenger operation attached; the opposite of the situation in Great Britain, it is worth noting. The other issue is that most of the New Zealand rail network is single-track, which limited, significantly, the scope for expanding services; and the maps do not show that in most cases, the inter-regional service level was rarely more than one or two trains per day. André does not, in my judgement, have enough background in railway operations to appreciate the difference which this makes in practice.
Apart from the day-to-day operational issues, the key issue for passenger rail was – again – money. And here, a quick guide to railway economics is in order. When the cost base of the average railway is examined, what is striking – especially compared with other transport modes – how much of the total cost is tied up in the infrastructure. Based on the British evidence, and New Zealand is unlikely to differ in this, the infrastructure accounts for about half the total cost base. For rail’s trucking competition, the share of the cost tied up in the equivalent infrastructure charge (that is, road user charges) is no more, I understand, than fifteen percent. Whether the asset is owned, as it is for rail, or rented, as it effectively is for truck services, is not the issue – the asset still has to be paid for. This means that when the railway’s management admit that they can cover their above-rail costs, but not the below-ones, this proportion explains why this is such a problem. In practical terms, this represents a huge fixed cost, and the only way to save money in this instance, is simply not to operate over the infrastructure … at all. Costs in a railway are not at all scaleable, as they are for truck or bus services, or even for aviation. Other modes cover the financial costs of their infra-structure, most of the time. Rail, very obviously, does not.
The point is that as time went on and the net losses within the freight business did not go away – some operational costs could be met, but the debt load was growing over time – one of the consequences was that there was, increasingly, little or no public money for passenger rail. The connection may not have been specifically drawn, but it would be surprising if it had not been a factor. Specifically: in 1981, nearly $NZ1bn in debt was written off as the old Department was corporatized; perhaps equivalent to $3bn in today’s terms. In 1992, a further $1.2bn was written off and $NZ360m in cash was invested into the business in the runup to privatisation – perhaps equivalent in total in today’s terms to a further $3bn. The situation post-1981 reflected a lot of investment which had been undertaken on in the clear expectation that the debt which had been taken on to pay for it, would be covered. Not for the first time, and certainly not the last, it wasn’t [ii].
The basic problem for railway financing is that what people are prepared to pay for rail services bears no resemblance to what it costs to provide those services. Corporatisation in 1981, privatisation in 1993, and “taking back the track” in 2006 were all predicated on the notion that a freight railway could pay its way. We know better now. Dave Heatley’s analysis from 2009 [iii] argues that the ‘break point’ in the finances of the freight railway, and the start of their long-term decline, can be dated to 1920; that is, the point at which trucks started to feature in the domestic freight task.
One other specific example is timely, for which I must admit to a personal involvement. During the period around 2000, when several long-distance passenger services were abandoned, it was not because (as André implies), that the Passenger Group had run out of ideas and enthusiasm, and weren’t interested in marketing the services (p251). We had plenty of ideas and enthusiasm, if we had been able to get the investment capital to do so (as we were also getting to the point that significant investment in the passenger rolling stock would be required). However, at the time Tranz Rail was essentially going bust, not that anyone could dare to admit to this, and as a result there wasn’t any money; for anything. Central and local government money wasn’t forthcoming either, but that is a separate issue.
Third, there is not enough attention paid to the way in which motorisation in New Zealand flowed through into reduced passenger demand
The growth in car ownership and use in New Zealand after 1920 was inexorable. The AA observed many years ago that by 1939, New Zealand was one of the most motorised countries in the world. It is hardly surprising that this happened, and that passenger demand was affected as a result – cars are simply far too convenient, for most journeys at most times. It is also forgotten than in the post-war period, it was actually quite expensive to own and run a car – Australians used to joke that New Zealand “was the land where Morris Minors went to die”. But this didn’t stop the growth in car numbers, and no amount of investment in passenger rail would have prevented this happening.
For example, when Stan Goosman convinced his Cabinet colleagues in 1954 (p154ff) to go for a motorway-based solution for Auckland, he was also picking up on the public mood at the time for more roads (e.g. the opening of the Auckland Harbour Bridge, p189; and the opening of the Lyttelton Road Tunnel in 1964 had been similarly welcomed [iv]). Had it been put to a public vote – “you can have a motorway system or rail-based public transport network, but not both” – my judgement is that the roads option would have won, hands down. (At this point, there is also a need here for the book to mention the equivalent report for Wellington a decade later (de Leuw Cather), which also recommended extending the rail system southwards of the railway station, and in preference to motorway investment. Again, this was not proceeded with, and Wellington’s rail system has suffered ever since). Even during the discussions in the early 1970s which looked at Robbie’s Rapid Rail, there was no way that the Government of the time was prepared to take money from the motorway construction programme, then in full swing, to pay for it [v].
Historically, it was quite expensive to buy and then run a car. The costs of car use only fell after 1988, when a change in industry rules allowed the import of second-hand Japanese vehicles. This lowered the real cost of motoring by about a quarter, if I recall correctly, and it is no surprise that car ownership rates picked up significantly in the next few years (with the use of motorcycles then declining as well, with very positive consequences it must be said for our road safety statistics). At this point, something as simple as a table, showing the numbers of cars and all vehicles, and as a rate against the total population, would have helped both illustrate and explain a great deal; and it is a major weakness of the history that these time-series are not provided. The reality is that people like their cars and they like to use them. It was not the case, at all, that declining provision of railway services led to more car ownership (p202); far more the case that increasing car ownership affected demand for rail services.
It is also striking that the policy of ‘shift to buses’ was evident as early as the 1920s and certainly by 1939 (pp 16,50). The development of motorised modes created difficulties for freight as well, hence the development of a protection regime in 1931 – which lasted in one form or another until the 1980s – and it is certainly clear how the development of the motor bus affected urban tram services as well. They were simply a lot cheaper.
Fourth, there is not enough attention paid to the way in which the growth of aviation in New Zealand affected railway demand over the longer journey sectors
The key year here is 1968, with the introduction into revenue service of the Boeing 737 on the main-trunk sectors. In 1970, air fares between Wellington and Auckland were still about four times as high as train fares, as André very usefully points out (p202), but they did not stay there for long [vi]. And the effects of the 737’s introduction were not that long in making themselves apparent: by 1974 the Rangatira (the inter-island ferry between Wellington and Lyttelton) was on the verge of cancellation. The writing had been on the wall here for some time here anyway, and it was finally cancelled in 1976. The end of the Silver Star overnight train between Wellington and Auckland occurred in 1979 (p229); it had only come into service in 1971. The subsidy requirements were mounting, and the growth in aviation provision in the previous ten years would have been a contributing factor to its losses.
To illustrate this more anecdotally:
Why go through this? Because it provides the context, which André does not, for decisions such as the cancellation of the Northerner in November 2004 and the shift after 2008 of what was left of the business to a ‘tourist model’. Demand had simply dried up. Even if measured in terms of city-centre-to-city-centre journey times, air was simply far faster, and their use of the discount model also allowed Air New Zealand to go after some intercity bus demand as well. Time is money, then as well as now, and the cost of providing long-distance rail service was now well beyond the market’s willingness to pay for it. Intercity itself has been under pressure for years, from the growth in car ownership, but the pandemic removed its tourism business – quite a large part of its market, about a third, as events proved – and it has not recovered since.
Again, a table, showing the volume of air traffic over time, would have helped both illustrate and explain a great deal. For example: between 2006 and 2018, total volumes through New Zealand airports grew 57 percent; a growth rate of 3.8 percent per year, cumulative. Over fifty years, that rate of growth would be equivalent to the total market growing by a factor of 6.5; or a factor of 4, allowing for growth in population over this time (from 3.2m in 1970 to 5.1m in 2023). With higher incomes and lower fares, it is hardly a surprise that airline traffic grew at such a rate, or that it impinged so much on intercity rail. In terms of how much domestic air fares have fallen, the indicative advice is: by 70 percent in real terms since 1970 and by 60 percent since 1980. International air fares have fallen by even more.
Fifth, not enough attention is paid to the institutional environment for urban transport, especially in Auckland
We all know that people prefer trains to buses, and by a considerable margin. We also know that people will drive rather than take a bus when rail is not available. But suppose a way could be found around people’s apparent preferences?
In 2008 the North Shore Busway in Auckland came into service, more or less following the route suggested for the Rapid Rail link nearly forty years previously. The busway concept was simple enough; a 7-km two-lane road running parallel to the main motorway with proper stations every kilometre or so. Although described in some quarters as a “Clayton’s railway” it has proved to be extremely effective, and has worked nearly as well as a railway investment would have – but, at $300m, at about a tenth of the cost of a railway link, remembering that a link will involve tunnelling under the harbour. The busway has since been extended up to Albany.
Elsewhere in Auckland, another busway is being developed from the Panmure station out to Pakuranga (now in place), the Botany shopping centre and in time down to the Manukau shopping centre. What is of interest here is that this route was also proposed for rail as part of the Rapid Rail proposals. However, with segregated busways generally around a third to a half of the cost of comparable light rail schemes, and light rail often not carrying many more in the way of passengers, it is clear that busways are here to stay [vii].
The importance of this for André’s work, is that it shows that the argument over urban rail options has now become much more nuanced. Auckland has specifically opted for a “mode-neutral” approach in its urban transport investment, and while this would not normally be covered in any detail in a history of passenger rail, it does need to be noted.
I have to confess that I found this chapter of the book frustrating. The wish list for rail investment is impressive – although it doesn’t include a re-instatement of passenger services along the Stratford-Okahukura Line, p286 – but the presentation here lacks a lot:
There is no appreciation of the wider environment of public finance
At one point, Lord Rutherford is supposed to have told his colleagues: “Gentlemen, we haven’t got any money, so we’re going to have to think”. Treasuries struggle to find the money needed for all the demands which other parts of the government put on them, so it is not a surprise that they tend to have a guarded view about spending money. A separate history of Treasury’s attitudes over time to the railways system would answer a number of questions raised by André’s history.
A case in point, from my own recollection. In 1976, when the idea of replacing the railcars was last looked at, the choice was between spending $20m ($200m in today’s money) on the railcars, or spending $3m ($30m in today’s money) on coaches, which would at least have been able to pay for themselves. It should not be a surprise that Muldoon’s Government opted for the latter. That year also saw the demise of the Rapid Rail proposals: years ago, I heard an Auckland Regional Council leader called Jo Brosnahan tell how in 1976, as a newly-minted planner, she was part of one meeting looking at Rapid Rail, “… where the costs of the scheme doubled in a morning, and it was doomed after that”.
There is no assessment of the cost of this investment, either totally or over time, and there is no assessment of how many passengers it would actually carry
Nor is there any assessment of the extra passenger traffic that this would generate, and where that traffic would come from. This is an issue; as other commentators have remarked what sort of money would be needed, or perhaps what the expenditure of a billion dollars per year over thirty years could achieve. In particular, there is no assessment of how this would interact with current intercity coach services, because it is very likely that a lot of the traffic for an expanded intercity rail system would be abstracted from that market. Finally, while rail investment would generate additional traffic, that is beyond question, it is not as yet clear how many people would switch over from driving, or indeed from flying. This needs to be known in order to make the case for the investment. This might not be André’s forte, but in the current climate his proposals won’t get anywhere if this question can’t be answered.
The comparison of road and rail funding includes a significant ‘category error’.
On p138 and p272, André compares the funding of the national road network with the funding of the national railway network, stating that it is unfair that more is expected of rail than is the case for the roading system. The problem with this argument is that it compares the whole of the rail network with the whole of the road network, when rail can really be said only to compete with that part of the road network to which it is parallel – SH1 south of Whangarei through to Invercargill; SHs2, 3, 4; 5 as far as Rotorua, SH16 in Auckland and SH73 from Christchurch across to the West Coast. Now, the road system is like any network business – the trunk supports the branches. It is very likely that this part of the road system collects much more in the way of RUC and petrol tax revenues than is spent on it. If location-specific pricing were in place, and the current averaging removed, trucks (in particular) would pay less to use the core parts of the road system, thus putting even more pressure on the freight railway’s business model, but much more for running on the lighter (and less trafficked) rural road pavements. The current cross-subsidy cannot be seen as being ‘unfair’ to rail – because rail does not run to these locations, and it needs to use trucks to access them. This is the classic ‘last mile’ problem of network economics.
An example of the cross-subsidy in the road sector which I came across many years ago, was from looking at a proposal to build a log transfer station at a railhead. That way, trucks could drive to this station and offload their cargo there, instead of driving all the way through to the port. This option was seen as saving about $200,000 per year (at the time) in avoidable road maintenance costs – but it was going to lose the roading authority some $400,000 annually in revenues. Not surprisingly, the idea didn’t proceed.
There is not nearly enough discussion of the intercity bus system
This is because there are significant journey pairs within New Zealand which do not have any sort of railway link – Auckland-Taupo-Napier, Dunedin-Queenstown, Gisborne-Tauranga, Gisborne-Napier, New Plymouth-Auckland, Blenheim-Nelson-the West Coast, are all cases in point. There is a good case for looking more at the system, especially as investing public money in services now would both improve network frequencies – currently, a key weakness in most areas – and begin to develop a market which in time might be upgraded to rail.
Where to from here?
If I were asked to develop a strategy for intercity passenger rail, I would proceed as follows:
The case for improving urban rail links is now accepted, but there is a way to go to see the same for inter-urban rail.
Ross Clark has worked in his career for the old Ministry of Transport (1986-1988), Transit New Zealand (1989-1996), Tranz Rail as-was (1996-2005),and since then for Transport Scotland. Most of this work has been in analytical roles. That background of two civil services and two railway companies provides the background for these comments. The opinions expressed here are completely his own!
Address for correspondence: firstname.lastname@example.org
[i] My own take on the privatisation of Tranz Rail can be found in this paper:
Clark, Ross (2010), “Full Circle: Rail industry privatisation in New Zealand, and a new theory of its fundamental conceptual weaknesses”, presented at the European Transport Conference, Glasgow. URL:
FULL CIRCLE: RAIL INDUSTRY PRIVATISATION IN NEW ZEALAND, AND A NEW THEORY OF ITS FUNDAMENTAL CONCEPTUAL WEAKNESSES (aetransport.org)
This text was finalised in mid-2010, and the situation has moved on a lot since then.
[ii] Readers are referred to Dave Heatley’s history of rail and its financial structures. Dave Heatley, “The future of rail in New Zealand” (2009):
.https://researcharchive.vuw.ac.nz/xmlui/bitstream/handle/10063/4024/Rail_Seminar_Feb_19_Slides.ppt?sequence=1. Also: The future of rail in New Zealand (core.ac.uk), for the full paper.
[iv] Refer the relevant Wikipedia article: Lyttelton road tunnel - Wikipedia
[v] Refer to this article on the Greater Auckland website for more details:
Sir Dove-Myer Robinson on his Rapid Transit Scheme – Part 3 - Greater Auckland
In his letter to the Auckland Regional Authority in July 1973, the Rt. Hon. Hugh Watt, then Minister of Works and Development, set out details of the Government’s terms for financing the bus/rail plan. He said in part:
“… the Government would make available the capital and meet capital service charges on any rail transit scheme that might ultimately be agreed between the Government and the ARA.
“… operating expenses and maintenance costs would be met by the local authorities through the ARA.
“… any expenditure by the Government on a rail system would not be undertaken at the expense of a curtailment of the motorway plan.” (emphasis added).
[vi] American analysis shows that in their market, airfares fell in real terms by half between 1980 and 2010 and likely 60 percent between 1970 and 2010. Another analysis reports that real airfares fell 70 percent between 1970 and 2020, and sixty percent between 1980 and 2020. International air fares decreased by even more. The New Zealand trends would be very similar.
[vii] And even in cities with well-established rail networks, as this piece from Brisbane illustrates:
What Is the Brisbane Metro? The City's $1.8 Billion Dollar Electric Mega Bus? | Talking Tactics - YouTube