There were a number of odd things in the report released several weeks ago by the New Zealand Council for Infrastructure Development (NZCID), a lobby group. Matt has already reviewed the report in detail. Perhaps the oddest part of it was this sentence:
Motorway capacity is essential because motorways generate economic activity.
NZCID presents this as a factual statement – or perhaps an article of faith? – but does not attempt to justify it or offer much supporting evidence.
From an economic perspective, this is an odd statement because transport infrastructure does not and can not generate economic activity. Roads are a means to an end, rather than an end in themselves. They can enable some economic activity, by allowing people to make journeys that otherwise wouldn’t have been possible, but they can’t actually generate it themselves. (Unless you think that the roads physically lift themselves up off the ground and start moving around and working in factories and stuff, in which case I recommend a psychiatric evaluation.)
Consequently, we must ask: Is there evidence that past motorway investments have raised productivity elsewhere in the economy?
Although the NZCID hasn’t cited it, there is relevant empirical research that addresses this question, including in New Zealand.
Before I get on to that, here’s some macroeconomic data. The top graph, sourced from OECD data, shows New Zealand’s investments in roads in dollar terms. Observe how it started to rise sharply after 2003 – that’s approximately when we started building more motorways.
The bottom graph shows Statistics NZ’s labour productivity index for the measured sector – a measure of changes in GDP produced per worker. Observe how there has been absolutely no change in the productivity growth trend, in spite of a threefold increase in the amount of money being spent on roads.
Correlation is not causation, but an absence of correlation is often evidence for a lack of causation.
This graph makes me doubt NZCID’s assertions about motorways and economic activity. For one thing, if building motorways truly was an economic panacea, shouldn’t tripling roads spending since 2003 be observable in the data by this point?
Fortunately, we don’t have to guess at the effects of motorway spending on economic output. Three OECD researchers, Balázs Égert, Tomasz Koźluk, and Douglas Sutherland, have taken a look at the issue. In a 2009 paper entitled “Infrastructure and growth: empirical evidence“, they examined the impact of infrastructure investment on economic growth using data for 24 OECD countries from 1960 to 2005. They looked at how investment (or disinvestment) in roads, motorways, rail, electricity generation, and telephone networks flowed through into subsequent economic growth.
Importantly, Égert et al found that the effects of infrastructure investment varied between countries – investments that had a positive impact on growth in one country can have a negative effect on growth in another. This could reflect differences in, for example, economic structure or quality of investment decisions.
Their key findings for New Zealand (from Table 1) were that:
- Road investment had a positive impact on economic growth throughout the period
- So did rail investment, although the effect was not quite as strong
- However, motorway investment had a negative impact on economic growth.
This is, again, the exact opposite of what NZCID have asserted. Transport investment in general appears to have had a positive impact on economic growth, but motorway investment in particular was a drag on growth.
Moreover, the authors considered the possibility that the returns from further investment changed over the course of the period. This is a reasonable hypothesis – after all, in 1960 many OECD countries were undergoing rapid economic change, and trying to build new infrastructure networks to keep up with it. Today, they are largely investing in incremental improvements to existing road and rail networks.
When Égert et al modelled the effects of infrastructure investment over the last decade or so of the period – around the time New Zealand was thinking about ramping up road spending – they found that:
“…in a number of countries the effect became stronger, suggesting for example that further increases in electricity generation capacity can be related to a decrease in output in Australia and Austria, similarly to motorways in Austria, New Zealand and Switzerland and rail tracks in Ireland and the Netherlands, whereas increases in road capacity may be associated with an increase in output in Greece, Ireland and the United Kingdom and additional electricity generation capacity in Portugal may support growth”
Again, not great news for NZCID’s argument that motorways generate economic activity. If the OECD researchers had simply found that past motorway spending in New Zealand had an ambiguous or negligible effect on growth, I’d be willing to accept the possibility that we could achieve more positive outcomes from further spending. But their finding that past motorway spending has been a drag on growth makes me worried about NZCID’s policy prescriptions.
There is, in short, a risk that NZCID is confidently recommending the wrong strategy for New Zealand. A strategy that has little robust empirical evidence to back it up, and which could easily backfire and reduce our growth prospects.
What could a responsible lobby group do differently?
First, rather than arguing for an increase in the quantity of investment, it could argue for an increase in the quality of investment. We know that this is a challenge for current transport spending. For example, a Ministry of Transport review that I covered last year (parts 1, 2, 3, 4) found that benefit-cost ratios for new and improved state highway have fallen significantly over the last decade:
Second, it could consider the role of transport investment in improving the choices available to people. As I’ve argued in the past, cities are diverse places, and the people living within them don’t all want the same thing. Some people love the big car and the big house – which is great, as long as they pay for the carbon pollution and don’t run anyone over. Others would be happier living in an urban neighbourhood and getting around on foot, bicycle, or public transport – and that’s also great.
Having more choices raises individual and social wellbeing. Unfortunately, transport policy has historically been “one size fits all” rather than “made to measure”. As there’s no real evidence that motorway spending has a positive effect on economic growth in New Zealand, wouldn’t it make more sense to invest in improving transport choices instead?
Motorways and economic growth: What do you think?