Should we spend more on infrastructure… or less?

The Auckland Transport Alignment Project (ATAP) report, which was released last week, identified the need to spend more money on transport infrastructure in Auckland. ATAP estimates that we need to spend $24 billion on new transport infrastructure over the next decade, around $4 billion of which would not be funded by current transport budgets.

While $4 billion is a sizeable gap, it’s smaller than previously assumed, due in part to ATAP’s recommendation to defer costly projects like the Additional Waitemata Harbour Crossing. But meeting it will mean raising fuel taxes, fares, rates, general taxes, or implementing congestion pricing to manage demands until funding is available.

ATAP’s obviously identified a need to spend more money on transport infrastructure in the Auckland context. But is spending more money on infrastructure in general a good idea? In other words, should any additional spending in Auckland be balanced by proportionately higher spending everywhere else in the country?

Two prominent American economists, Larry Summers and Ed Glaeser, recently took contrasting views on this question. Summers is most well-known as a policy advisor to Democratic administrations and (in recent years) an advocate of fiscal policy as a cure for long post-GFC recession. Glaeser, on the other hand, is best known for his work on urban economics, including his great book Triumph of the City.

Summers lays out the case for spending more (in the Financial Times). His key argument is that low interest rates signal an underemployed economy where the “opportunity cost” of paying people to build more stuff is relatively low, and that infrastructure spending is a good way to do this as it can enhance a country’s long-run productive potential:

There is a consensus that the US should substantially raise its level of infrastructure investment. Economists and politicians of all persuasions recognise that this can create quality jobs and provide economic stimulus without posing the risks of easy-money policies in the short run. They also see that such investment can expand the economy’s capacity in the medium term and mitigate the huge maintenance burden we would otherwise pass on to the next generation.

The case for infrastructure investment has been strong for a long time, but it gets stronger with each passing year, as government borrowing costs decline and ongoing neglect raises the return on incremental spending increases. As it becomes clearer that growth will not return to pre-financial-crisis levels on its own, the urgency of policy action rises. Just as the infrastructure failure at Chernobyl was a sign of malaise in the Soviet Union’s last years, profound questions about America’s future are raised by collapsing bridges, children losing IQ points because of lead in water and an air traffic control system that does not use GPS technology.

In particular, Summers argues that priority should be placed on funding deferred maintenance, which is a major problem in the US:

What is the highest priority? The fastest, highest and safest returns are likely to be found where maintenance has been deferred. Maintenance outlays do not require extensive planning or regulatory approvals, so they can take place quickly. And they tend naturally to take place in areas where infrastructure is most heavily used.

Glaeser sets out a considerably more skeptical perspective in the City Journal. Contra Summers, he argues that infrastructure spending isn’t a particularly efficient way of getting unemployed people back to work, and that the political incentives facing decision-makers tend to mean that additional funding is misspent in declining areas like Detroit or on projects that don’t do much good:

While infrastructure investment is often needed when cities or regions are already expanding, too often it goes to declining areas that don’t require it and winds up having little long-term economic benefit. As for fighting recessions, which require rapid response, it’s dauntingly hard in today’s regulatory environment to get infrastructure projects under way quickly and wisely. Centralized federal tax funding of these projects makes inefficiencies and waste even likelier, as Washington, driven by political calculations, gives the green light to bridges to nowhere, ill-considered high-speed rail projects, and other boondoggles. America needs an infrastructure renaissance, but we won’t get it by the federal government simply writing big checks. A far better model would be for infrastructure to be managed by independent but focused local public and private entities and funded primarily by user fees, not federal tax dollars.

Glaeser takes more specific aim at the notion that infrastructure investment inevitably generates broader economic returns:

Infrastructure spending is a form of investment: just as building a new factory can boost productivity, laying down a new highway or opening a new airport runway can, at least in principle, generate future economic returns. But the relevant question is: How do those future returns compare with the costs? Just because infrastructure is a form of capital doesn’t mean that spending a lot on it is always smart. When a firm estimates the rate of return for a new factory, it can calculate the expected net profits and compare those with the expense. The analog for, say, new or improved roads is to estimate the benefits to users from reduced travel times, add the likely modest spillover benefits to nonusers, and then subtract the spending needed to construct and maintain the infrastructure. The results can differ significantly across projects. A well-known 1988 Congressional Budget Office survey found that spending to maintain current highways in good shape produces returns of 30 percent to 40 percent—but that new highway construction in rural areas showed a much lower return. A clever study that used firm inventories estimated that the rate of return to new highways was sizable during the 1970s but sank below 5 percent during the 1980s and 1990s.


The existence of plausible transportation alternatives and the law of diminishing returns have also tended to reduce the benefits of infrastructure investment over the past two centuries. The opening of the Erie Canal in 1821 brought enormous value because the inland transportation options at the time were dismal. In the early nineteenth century, it cost as much to ship goods 30 miles over land as to send them across the entire Atlantic Ocean. Yet the very existence of canals, as much of a breakthrough as they represented, reduced the benefits of the later rail system, as Nobel economist Robert Fogel has shown. The returns for new transportation infrastructure in places with terrible roads, such as much of Africa and India, will be much higher than in the United States, which already enjoys an impressive, if under-maintained, array of mobility options.

While Summers and Glaeser take different views on the value of spending more money on infrastructure, there are some important points of agreement, such as:

  • Prioritising maintenance spending to replace or upgrade run-down infrastructure
  • Better cost benefit analysis to ensure that money is being spent in more beneficial ways
  • Where appropriate, funding new infrastructure more from user charges and fees, rather than general taxes.

Lastly, it’s worth asking whether these issues look different in New Zealand than in the United States. I don’t have a complete answer to this, but in previous posts I have looked at the issue of infrastructure spending from a variety of perspectives. For instance, I’ve asked:

On balance, I’d say that those posts present a moderately skeptical view of the case for significantly ramping up transport investment – ie more in line with Glaeser’s view than Summers’. That’s not to say that we shouldn’t spend a bit more, but any additional spending should be backed by robust analysis.

What do you think about infrastructure spending? More or less?

30 comments to Should we spend more on infrastructure… or less?

  • Well it seems Summers and Glaeser are more in agreement than not; on maintenance and renewals. Which is surely the first order of business.

    Even if we ‘are all Keynesians now’, still does matter enormously what the value of what we build is. And in this the other clear point above is the low value of duplication. And this where the RoNS look especially suspect; they are increasingly duplicate highways, beside existing ones that could in many cases be upgraded at much less cost, and without then adding to the ongoing maintenance burden. For example the existing SH1 north will become the ratepayers of Auckland’s burden when the flash Holiday Highway opens. This is an unsafe and tricky road to be responsible for.

  • Stuart Donovan

    in New Zealand’s case, I’d support more infrastructure spending. We’re a relatively young and rapidly growing country that could benefit from a larger population.

    The more interesting question, I think, is how revenue is raised. I wouldn’t support, for example, diverting general funds to invest in infrastructure. Instead, I’d rather see more accurate pricing mechanisms being used to raise revenue for new infrastructure while also managing demand for what we have.

  • dr

    Speaking of spending less, Did ATAP include a ‘do nothing + road pricing’ option?

  • Case of not spending more or less but rather better spending of the existing pool already.
    Just think what $11b worth of RoNS could have gotten Auckland and the rest of the nation instead of over engineered highways.

  • Be interesting to see a fair comparison of investing in transport vs digital connectivity. Reliable, affordable hi-res video can be a gamechanger.

  • Bex

    A recent paper on China’s infrastructure spending was fairly critical of the actual long term value generated, it points out that massive investment into immobile investments (i.e. large structures that you can’t easily remove) present a long-term risk if the quantity, or nature, of demand changes. A more useful approach to infrastructure spending might start by asking whether the problem can be solved using options that avoid new construction – demand management, reprioritising road space, new technologies – instead of defaulting to the idea that the solution has to involve building something new

  • Angus Robertson

    Whilst construction rates continue advancing quicker in other parts of the country, Waikato/BoP – rather than Auckland, infrastructure investment should go there.

  • Proven, experts such as Paul Krugman make a compelling case that the the world in general, and the USA in particular, are caught in a classic lack of demand trap. In that setting, ultralow interest rates are not inflationary, and additional government spending is vital to make up for the lack of private spending. For the US, Summers is completely correct. Keynes suggested burying money and giving out contracts to dig it up again. He just meant that even something so worthless was better than nothing, 8-).

    Instead they should be upgrading their trains, subways and electrical infrastructure, and bringing it at least to the end of the 20th century, if the 21st would be out of reach!

    NZ hasn’t been as badly hit, but except for housing there are lots of things suggesting that we are way under demand, starting with inflation well under 2%, sigh. All the same, we should not waste a penny more on roads for cars. Instead Lets spend _more_ on trains, installing modern cycle paths, and generally upgrading our built environment. Until inflation is clearly over 5%, we do not need to worry that we are stoking. At the moment we are drowning it for no reason.

    Housing is a totally different. Yes, house prices are out of control but this is because of growing population and bad restrictions, as highighted many times by our hosts, not general inflation. THAB for auckland!

  • Brendon Harre

    Peter you seem to be contradicting yourself regarding this statement. “Is spending spread around the country proportionately? (Yes, with exceptions)”

    Not matching the following statement you made in your “MoT’s review of capital spending on roads, part 4” article.

    “…..there are some big disparities between revenue and expenditure in some regions. In particular: Auckland and Wellington are getting about 20-25% more in NZTA expenditure than they pay in revenue, while Canterbury is getting only slightly more than half as much expenditure as it pays in revenue (note you even put the first sentence in bold).”

    Evidence over a ten year period (probably longer if further evidence was examined) showing a massive regional disparity in revenue and expenditure is surely more significant than being brushed off as an exception.

    It is evidence that NZ’s centralised transport funding fails to allocated transport expenditure fairly.

  • Brendon Harre

    When I returned from Europe in 2012 one of my first impressions was that NZ has very poor quality and quantity transport systems for our cities.

    I did some investigation and came to the conclusion that after a binge of rural road construction last century we became frugal on the second and third generation transportation systems like motorways (outside of Auckland), bus lanes, cycle lanes and passenger rail needed to keep modern cities running efficiently. I wrote an article about it

    If NZ invested in the bus lanes, cycle lanes and rapid transit systems to make our cities more efficient. I think it would lead to multiple benefits. It would help provide the infrastructure needed for new affordable housing (up or out), it would help the economy transition from primary produce dependence to a more diversified value-added economy and it would boost demand in this period of insufficient global demand (deflating prices, historically low interest rates….).

  • Brendon Harre

    My preference for funding new infrastructure. Is firstly variable price congestion charging for private vehicles. This should both reduce the demand for moar roads and identify to the various private and public players where new transport system and services are genuinely needed.

    If congestion charging cannot be implemented quickly I would allow a regional fuel tax with a planned transition over to congestion charging.

    I would create regional infrastructure bodies with appointees from both local and central government -that would receive the above funding plus I would allow these bodies to receive the regionally generated construction GST receipts. The infrastructure bodies should have a Making Room philosophy wrt urban development - but that concept should be interpreted as both up and out -so areas which are intensifying should also get upgraded transport systems.

    • buttwizard69420

      Fuel taxes are a sledgehammer solution – they punish travel no matter what time you take it, or what direction you happen to be heading in, be it in or out of the district. You’re going to burn an enormous amount of goodwill with the public if it is isn’t immediately linked to widespread transport improvements (which it probably won’t be, but that’s the reality of glacial infrastructure development) so you might as well do it right first time.

  • Matthew W

    Tyler Cowen recently fairly eloquently picked apart the case for fiscal authorities using lower than private sector hurdle / discount rates. Basically unless there are genuinely idle resources (not really the case in the US and NZ, maybe elsewhere), the opportunity cost of public sector investment is private sector investment. So unless the return is as high or higher than private sector investment at the margin, it is a bad idea.

    Similarly Scott Sumner makes the point that that for fiscal stimulus to work, the central bank has to not be doing its job properly. If a central bank is doing its job it will offset any tendency to higher demand from fiscal stimulus with monetary policy. So, lets just get central banks to do their jobs properly and reserve infrastructure investment to that which passes a cost benefit analysis using private sector discount rates.

    • Peter Nunns

      Central banks face a zero lower bound on the OCR – once they’ve hit that point conventional monetary policy loses its effectiveness. Unconventional monetary policy (helicopter money!) can in principle fill that gap, but in practice it doesn’t seem to have been fully effectual. (Due in part to political constraints.) Hence the case for fiscal policy. Krugman covers this quite well:

      On discount rates: Cowen presents one view on the issue; it’s not universally agreed. Some substantive critiques:
      1. A social rate of time preference approach may be more appropriate than a social opportunity cost approach. NZIER addressed this in a paper a few years back that transportblog covered.
      2. Even if we do adopt a social opportunity cost approach, we have to acknowledge that most government funds come from taxes on household income/consumption, not taxes on businesses. Households tend to value the future more highly than businesses, with real discount rates more in the range of 2.5% than 10%. If government spending substitutes for household consumption/investment, rather than business investment, then a lower discount rate may be appropriate.

      • Matthew W

        Political constraints have hindered the application of unconventional monetary policy, they havent affected its effectiveness.

        The US has demonstrated that once they sorted out what they were doing unconventional monetary policy is effective. In particular the “fiscal cliff” turned out OK in early 2013. Krugman himself pitched this as a test of market monetarism – and it passed.

        In NZ of course we have not been at the ZLB so havent had to worry about this.

        The fact that households are taxed rather than businesses does not imply that consumption is being offset rather than investment. Obviously households save and ultimately invest in and own those businesses. The fact that consumption is occuring at all means it is valued more highly than investment – indeed consumption vs investment really is about the now vs the future! So not sure I follow you there.

      • mfwic

        In addition to point 2 Peter business borrowing typically comes at a higher risk premium than either household borrowing or Government borrowing. So it is not simply about comparing returns but also about the potential downside. A lot of household debt is secured against property. Government borrowing can fail to produce a return to society but the asset is still there. In contrast business debt has a high hurdle because they can not only fail to get a profit but can lose the capital or debt.

        • Matthew W

          I dont know that distinction is that real mfwic. If business invests in a building which doesnt get the same level of return, the building doesnt disappear. If government produces an asset that doesnt provide the level of benefits its the same thing.

          • Peter Nunns

            It’s not as silly a distinction as you suggest. Different types of businesses face different costs of capital.

            For instance, a business whose debts were secured against intellectual property, reputation, or expectations of future returns would expect to face high borrowing costs – 20% or more – whereas a business with debts secured against real estate would face much lower borrowing costs – probably around 5%. The government is more similar to the latter type of business, although the individual investments that it makes may be different.

      • Matthew W

        Interesting regarding the low discount rates for freehold vs leasehold properties though.

        • Peter Nunns

          Yes, that was a really interesting study. One interpretation of it (which I tend to highlight) is that the opportunity cost of government spending may be lower than implied by businesses’ cost of capital. Or, in even simpler terms, we should address climate change because there’s good evidence that humans care about more than the next quarterly report.

          I don’t think that using a lower discount rate for infrastructure projects is an argument for spending *more*. If we use a lower discount rate, it will affect our choice of projects – favouring capex-heavy solutions over opex-heavy ones, or projects with longer delays between implementation and ramp-up in use. But in the absence of a policy decisions to increase the overall funding envelope, it won’t directly affect total infrastructure spending.

          Falling government bond rates tend to be a signal that the economy is experiencing a shortfall in aggregate demand (to use the Keynesian parlance) and hence that additional government spending may be economically beneficial. But, as Glaeser says, that fact doesn’t necessarily imply that *infrastructure* is the best thing to spend money on. So I tend to see this debate as distinct from discussions about the appropriate level of discount rate. They’re often confused, because they’re fairly similar, though.

  • mfwic

    The standard view seems to be that infrastructure spending doesn’t create economic growth but it can remove a bottleneck to growth if one exists. It is an important distinction. If you accept that growth is due mainly to capital being invested in productive ways then spending on infrastructure can crowd that out as it did with Japan’s highways and bridge programme. But if you cant get the pig to market due to a missing road or rail line (Adam Smith argued canals!) then public spending has a place.
    I am more with Larry Summers here, not in thinking all public spending is good, but if there are bottlenecks that we need to fix then now is a really good time to do them. CRL should have started as soon as interest rates dropped below 5%. Same for the east west link and rail to the airport. If these things are going to happen then do them while finance is cheap and while there is capacity.

    • Yes exactly as i agued here about Northland economy and Pu-Worth:

      Hon GERRY BROWNLEE: ‘It may surprise the member to note that people cannot get their goods out of Northland unless there are roads outside of Northland. It sort of makes sense that if you want to get from Northland down to some other part of New Zealand, you need a road to get there.’

      So true, imagine if there were no roads, or rail line, or shipping routes, or flights to and from Northland, or at least if these were so suboptimal as to prevent any goods getting to market, milk spoiling, logs rotting, tourists unable to get to the Bay of Islands, then yes, making that connection for the first time would indeed be a breakthrough, and have a truly transformational outcome. Is this an accurate picture of the situation? Well there is one day of the year when traffic does get stuck heading north out of Auckland, December 27th, but even that dissipates well before it reaches Northland and only holds a cargo of impatient holiday makers.

    • Matthew W

      Here is Tyler Cowen explaining why government interest rates are not a good reason for upping borrowing. Its all about opportunity cost:

      • mfwic

        I don’t think that applies if the work is required in any case. The choice is borrowing while demand for finance is low or waiting and borrowing when demand for funds is high. In both cases there is an opportunity cost but it is higher when Government borrowing crowds out other borrowing at a higher rate.

        • Matthew W

          If the project is a goer, its a goer, so interest rates shouldnt come into it in that situation either. The goverment should borrow to invest in positive net benefit projects, as long as it is applying an appropriate discount rate.

          • mfwic

            The point is that the appropriate discount rate should reflect the Government’s cost of capital. In the past in NZ the two haven’t moved together. But they should. When the cost of capital is low the discount rate should also be low. It is a time to invest in the longer term projects like CRL or whatever. When the cost of capital is high then they should be looking at projects that have high benefits in the early years- these tend to be mostly lower cost fixes rather than the big projects.

          • Matthew W

            The point is the government is an equity investor. Its cost of capital is not what it costs to borrow but the opportunity cost of taxing taxpayers.

            See here for an explanation:


          • mfwic

            Well that’s a few minutes of my life I wont be getting back. The lesson I got from macro is that fiscal policy isnt an economic issue it is a political issue. Economists are the same as everyone else. Some believe in small government and their economics are focused on that, a few think the government is the solution to all ills. There are more economists in the former group due to adverse selection- those who are interested in money (their own included) study economics. He has totally missed the point that the government sometimes has projects that can significantly increase welfare with low risk. And to get back to your point the opportunity cost of taxing is less when the interest rate is lower. QED.

          • Peter Nunns

            My approach in this post has been to teach the controversy.

          • Brendon Harre

            Well done Peter : )

Leave a Reply