A Munnel Retrospective

Will we have buyers remorse?

The Melbourne Metro Tunnel opened this week to widespread acclaim, everyone seems to be having fun which makes it an ideal moment to ruin the party byt revisiting the decades old original business case to check how the numbers panned out!

In 2016, the economic case for the Melbourne Metro Tunnel appeared exceptionally strong. Even under conservative assumptions, the benefits were expected to exceed the costs with a CBR of 1.1. Using more generous and, arguably, much more realistic inputs (around discount rates and wider economic benefits) the project’s case looked even more compelling, with a Benefit-Cost Ratio (BCR) of 3.3 at a the more realistic 4% discount rate and including wider economic benefits. A slam dunk of idea!

Evaluting the Crystal Ball

Many aspects of the business case turned out to be surprisingly accurate. Population forecasts which largely correct. Despite the disruption of the pandemic and the temporary loss of international students and workers, Melbourne’s long-term growth trajectory remains broadly consistent with the projections made in the original business case, which forecast a population of 6 million by 2031 and almost 8 million by 2051. In fact, more recent demographic modelling suggests the city may even exceed those earlier expectations. Though I suppose a cynic might argue that the unforecastable pandemic provided a even larger miss on the population numbers.

Construction timelines were also largely met. The opening aligns with the completion windows outlined in the early planning documents, an achievement that is unusual for a megaproject of this scale. Costs did rise with the final overrun being around average by global standards and lower than the blowouts seen in many comparable projects. The project’s cost increased from an original P90 nominal estimate of $10.9 billion to approximately $15.5 billion—a rise of around 42%. Again not bad considering the pandemic and inflation that occurred during the period.

Something must be done

However, the projections for train patronage tell a very different story. The business case assumed that metropolitan rail usage would grow steadily over time, forecasting a doubling of weekday boardings from 750,000 in 2011 to 1.5 million in 2031.

Instead, current patronage remains well below pre-pandemic levels. More striking still, usage appears to have plateaued or even slipped slightly despite strong population growth and a substantial return to office work in many sectors. This stagnation is sharply at odds with the sustained growth path underlying the original forecasts.

​​

It’s not just working from home, patronage remains weak even on weekends

The implications for the Metro Tunnel’s benefits are significant. Much of the project’s value came from expanding capacity to move large numbers of commuters into and out of the CBD during peak times. With demand for those trips materially lower than expected—and with hybrid work now a persistent feature of the labour market—the realised benefits may fall well short of what was assumed. And because patronage would normally have grown over the past several years, the true shortfall relative to the forecasts embedded in the business case is considerably larger than the simple gap between today’s usage and pre-pandemic norms. We calculate this shortfall to be approximately 50% when comparing 2024 actuals (plus a generous 15% growth buffer) against the original 2031 forecasts.

Take for example Flinders St station. That had just over 50,000 entries each work day in 2011, forecast to rise to over 150,000 in 2031 without the Melbourne Metro. In 2024 however it averaged just 61,000 entries. Even allowing for 15% growth in the next 6 years that implies a shortfall in the forecast entries of 54.1%!!!

The only station to exceed its forecast is Caulfield - which presumably due to the unstoppable success of the Monash Business School ;)

This shortfall in patronage strikes also reduces the project’s other economic justification: agglomeration. A significant portion of the Metro Tunnel’s projected value was derived from ‘Wider Economic Benefits’ (WEBs)—essentially the productivity bonus triggered by densifying the CBD’s labor market. The theory was that by transporting more skilled workers into the Hoddle Grid, the state would unlock a ‘network effect’ of innovation and efficiency. However, with peak-hour commuting down and hybrid work dispersing the workforce, the ‘effective density’ of the CBD has not hit the levels anticipated in 2016. Nor indeed is the train network a barrier to it increasing. If the physical clustering of workers is significantly lower than forecast, those agglomeration benefits may not just shrink linearly; they risk evaporating entirely, leaving a hole in the project’s benefit ledger.

Even under the most forgiving economic settings—applying a 4% discount rate to capture intergenerational equity and including the full suite of Wider Economic Benefits (WEBs)—the Melbourne Metro Tunnel now hangs by a thread of economic viability. When we adjust the original business case to reflect the increase in nominal costs from $10.9 billion to $15.5 billion (a 42% blowout) and a 50% structural reduction in benefits due to the post-pandemic patronage cliff, the project’s Net Present Value collapses from a massive $18 billion surplus to a razor-thin $0.72 billion. This results in a Benefit-Cost Ratio of just 1.06, meaning that for every dollar invested, the state is now retrieving only six cents in surplus value.

While an NPV greater than zero technically classifies the project as “viable,” this is obviously just a back of the envelope calculation. If train patronage remains low and it turns out we don’t hit capacity constraints on the old network for another 15 or 20 years then the benefits could easily fall by more than half. Without the 4% discount rate—if assessed under the standard 7% Treasury test—the project would be deeply unviable, destroying billions in value. It is undeniable that the unforeseen pandemic took a hatchet to what was otherwise a very socially profitable piece of infrastructure.

In 2016, this project was a “slam dunk” with a BCR of 3.3 (at 4% discount), promising to fundamentally reshape the city’s productivity. Today, it stands as a break-even asset, saved only by the long-term horizon of its utility.

Doing nothing is an option

The saga of the Melbourne Metro Tunnel serves as a stark warning for any heavy rail project dreamt up in a pre-pandemic world. For a long time, there was hope that the pandemic was a blip—a temporary interruption to the inevitable growth of mass transit.

But here we sit, five years after it began, and we are nowhere near our pre-pandemic peak. There is no sign of a rapid recovery; if anything, 2025 is shaping up to be a slightly weaker year for public transport than 2024. The structural shift is undeniable.

This reality exposes a fundamental contradiction in modern government policy. If governments are going to commit to making “Working From Home” a permanent and prominent feature of the workforce, they must also have the courage to pare back their infrastructure ambitions.

We must recognize that Work From Home is, in itself, a massive infrastructure benefit. By keeping workers at home two or three days a week, we naturally decapitate the peak-hour congestion that these projects are built to solve. If we lock in hybrid work, we simply need fewer roads, fewer trains, and fewer expensive tunnels.