Is a rail link between North & South Islands really "essential"?🍋
And what game has KiwiRail been playing with government finances?
It’s rare that NZ Herald columnists Simon Wilson and Richard Prebble agree on anything. In my experience, Wilson’s opinion columns are a reliable barometer of fashionable opinion on the left; whereas those of Prebble, a former Labour minister and ACT Party leader, reflect the polar opposite.
But the recent columns from Wilson and Prebble could almost have been written by the same person. Both came out with strong support for KiwiRail’s iReX project — the replacement of its Interislander ferries and associated port facilities — following the project’s abandonment by government due to cost estimates blowing out to over $3bn.1
In the pastiche of Wilson and Prebble’s columns that follows, which commentator said what? I can’t tell.
KiwiRail wants to build major new port infrastructure in Wellington and Picton that will future proof the service for the next 100 years.
The new diesel-electric hybrid ferries reduce emissions by around 40 per cent and carry an increase in rail freight. Today the three ageing, unreliable ferries cannot carry all the freight that shippers want to send by rail.
The ferries themselves look like the result of exceptionally good strategic planning. They would have been hybrid electric, thus quieter and calmer in the harbours, with much lower emissions. They would have doubled the capacity for passengers, tripled it for rail wagons and almost doubled it for cars and trucks.
… building new seawalls and raising the docks by a metre will help cope with the changing climate. These things will all become essential. Shifting more freight to rail is a foundation for any serious plan to reduce transport emissions. It’s also critical to easing road congestion and making the roads safer. And the need for resilience in the face of rising sea levels and the storm surges of wilder weather is self evident.
The economic, environmental, and social impact on the South Island, on exporters, importers, industry, tourism, indeed the whole country, of not having an inter-island rail service will be profound.
It’s been nearly 15 years since I researched and published an economic history of New Zealand rail. I remain bemused at what the topic of rail can do to otherwise sensible people. It seems that many people have strong preferences in favour of rail, and they lie right across the political spectrum. But society cannot accommodate every strong preference, especially preferences that can seemingly only be satisfied with billions of dollars of public funds.
KiwiRail did not get the cost right. Why trust it to get the benefits right?
Infrastructure projects are prone to cost blowouts, especially publicly funded ones. One reason is that political incentives encourage proponents to undercook their cost estimates:
Proponents know that early realism may mean that the project never gets off the ground. Far better to start with a low-ball estimate and get further buy-in, then let ever-more realistic estimates dribble out over time. It is difficult for decision-makers to reverse their support, once given.2
KiwiRail clearly low-balled the costs of its Interislander ferry upgrade project. Proponents also have incentives to over-estimate benefits.3 Given KiwiRail’s performance on costs, it would naïve to uncritically accept the benefits they claim, nor those Wilson and Prebble claim on their behalf.
Let’s take a look at some of those claims.
“Future proof the service for the next 100 years”
Anyone who thinks they can reliably forecast the next 100 years is either brave or foolish. To illustrate, let’s roll the clock back 100 years. If you had looked at the following data back in 1923, you could easily have forecast a rosy commercial future for NZ rail.
But your 1923 self would have been totally wrong. The profitability of rail collapsed, under competition from motor vehicles. The NZ Government responded in 1936 by restricting the transport of goods by road. Those restrictions slowed, at best, rail’s decline in financial performance. Line closures and technological improvements (e.g. replacing steam with diesel locomotives) arrested the decline during the 1950s and 1960s, after which it plummeted yet further.
Despite being a preeminent technology in the 19th Century, rail had a rocky time in the 20th. In many parts of the world, including NZ, rail’s survival into the 21st Century became increasingly dependent on government subsidies. Will governments be as generous into the 22nd Century?
Few people in 1923 would have anticipated that air travel would all-but-wipe-out long-distance passenger rail. And nobody was concerned about climate change. It is no less fraught today to predict the world of 2123. Will, for example, rocketships provide a one-hour point-to-point service anywhere on the globe? I don’t know. But I’m wary of anyone who confidently predicts 100 years hence, and uses those predictions to justify immediate action.
“Shifting more freight to rail is a foundation for any serious plan to reduce transport emissions”
Shifting more freight to rail has been the foundation of many plans to reduce transport emissions. To take one example, The New Zealand Transport Strategy 2008 adopted the target: “to increase rail’s share of [domestic] freight to 25 percent of tonne kilometres by 2040.”4 How has that gone? Using the latest data5, I graphed a comparison of actual vs. target performance for rail’s proportion of the road plus rail freight task. That looks like a comprehensive failure.
Despite a strategy that led to governments pouring billions of dollars into KiwiRail, just 7 commodities made up 90% of rail tonnage in 2019.6 There seems little hope that KiwiRail will ever capture significant market share beyond its core task of point-to-point transport of bulk commodities.
Any serious plan to reduce transport emissions would learn from the failure of past plans. NZ just cannot afford to back any and every half-baked idea that sounds like it might reduce greenhouse emissions. Those who are serious know that relentless pursuit of the cheapest options for greenhouse emissions abatement, alongside rejection of expensive ones, offers the best outcome for NZ, and the planet.
Interislander has a commercial competitor: Bluebridge
Both Wilson and Prebble write as if Interislander is a monopoly, and one only suited to government ownership. Neither mentions Interislander’s privately owned competitor Bluebridge. A working paper recently published by Chris Stone included 2018/19 financial accounts for Interislander and Bluebridge.
Rail freight contributed just $34m to Interislander’s revenues, albeit with only one rail-capable ferry. Rail revenue would rise with two shiny new rail-capable ferries, but only at the expense of truck revenue — leaving Interislander no better off.
To increase total revenue, Interislander needs to capture market share from Bluebridge. To do that, it would have to drop prices, reducing revenue, at least in the short run. But even if Interislander could use predatory pricing to drive Bluebridge out of business (and the Commerce Commission would likely have something to say about that), the total Cook Strait transport market is not huge. It offers just $273m of annual revenues and $82m of EBITDA (i.e. earnings before interest, taxes, depreciation and amortization).
Bluebridge has been capturing market share from Interislander over the past two decades. It is able to fund replacement ferries out of earnings, whereas Interislander has been letting their ferries age in place. These facts, alongside Chris Stone’s analysis, suggests Bluebridge’s operating costs are below Interislander’s.
A $3bn “investment” in new ferries would add at least $100m of annual depreciation plus $150m of interest.7 Covering those costs from revenue looks impossible.
Is a rail link between the North & South Islands really "essential"?
It’s hard to argue that rail across Cook Strait is essential in any sense. There was no inter-island rail link until 1962. This was well after NZ’s “golden age of rail”, generally accepted to be the first half of the 20th Century.
Similarly, in the year following the 2016 Kaikoura earthquake, which severed the Picton to Christchurch rail link, it was trucks that kept freight moving — making the Interislander rail capability irrelevant.
It’s also hard to argue that Cook Strait ferries need to be publicly owned. Interislander itself was privately owned between 1993 and 2008. And Bluebridge has operated in competition with Interislander since 2002.
Those who are serious about climate change know that relentless pursuit of the cheapest options for greenhouse emissions abatement, alongside rejection of expensive ones, offers the best outcome for NZ, and the planet.
Rail-ferry dreaming
My theory is those running KiwiRail, for whatever reason, saw the ferry upgrade primarily as a rail upgrade. This was their chance to get two rail-capable ferries.
Rail-capability is a rare requirement for ferries worldwide. This requirement meant KiwiRail needed custom-built ferries, at around $250m each. Bluebridge, by contrast, can purchase perfectly good second-hand ferries for fraction of that price. (For example, they paid just $35m for the Strait Feronia.)
There was no way KiwiRail could fund this project from revenue. But they had a very good chance of persuading a supportive government to fund it. Custom-building the ferries means they could have chosen ferries compatible with existing port infrastructure. Port upgrades could have followed later.
KiwiRail, however, specified their dream ferries. These designs were incompatible with existing infrastructure, forcing the Government into a complete rebuild of port infrastructure to make use of the ferries they had agreed to purchase. And the Government did agree to an additional $1bn, but balked as the total project price blew out to $3bn. (And, given that design work was incomplete, the price would likely go even higher.)
Accountability for this shipwreck
Georgina Campbell writes:
It is tempting to call for heads to roll after a project of this magnitude has experienced such spectacular failure.
However, the governance and leadership at KiwiRail have changed several times since the business case for the two new ferries was first submitted in November 2018. Greg Miller, for example, served as both the chief executive and the chairman of the board during that time.
It is therefore not so clear cut as to how accountability for this shipwreck should be served.
Governance and leadership positions at KiwiRail are well-paid. In 2022/23, KiwiRail’s CEO received $1.37m; its directors received between $42k and $89k.8
It seems clear that KiwiRail played a game of drip-feeding low-balled estimates to government, gaining permission to continue, upping the estimates, and then daring government to back out. Executive staff and directors knew what was going on, which makes them culpable. Or they didn’t, which reflects badly on their competence. Either way, they have demonstrated their unsuitability for public governance roles. Accountability is meaningless without consequences.
By
Funding for the final round of cost blowouts was rejected by the Labour Government in September 2023, and again by the National/ACT/NZ First Government in December 2023.
Flyvbjerg, B., Garbuio, M., & Lovallo, D. (2009). Delusion and Deception in Large Infrastructure Projects: Two Models for Explaining and Preventing Executive Disaster. California Management Review, 51 2, pp. 170-193.
Ministry of Transport. (2008). The New Zealand Transport Strategy 2008. Available at https://www.mcguinnessinstitute.org/wp-content/uploads/2021/04/267.-The-New-Zealand-Transport-Strategy-2008.pdf
Data on rail and road is from the Ministry of Transport’s Transport Indicators webpage. I could not find recent data on tonnes-km of freight moved by coastal shipping, so I converted the target into a percentage of the rail plus road freight task. I used linear interpolation for the target, and to fill gaps in the data.
Ministry of Transport (Sep 2023). FIGS: Rail — Commodity movements.
I assumed straight line depreciation over 30 years (20 years for ships, presumably longer for port facilities), and 5% interest rates. Both assumptions are generous.