It’s time for a quick round of everyone’s favourite game, Ask An Economist. Today’s question is: What happens when the government decides to spend up large in a growing economy?

If you guessed that the answer is that it will drive up inflation and crowd out private sector spending, congratulations! You win a hug from the invisible hand. You’ve obviously either paid attention to the lessons of history or the words of latter-day popularisers of economic theory such as Bill English, who recently said that:

It was also important for the Government to run a counter-cyclical fiscal policy which, right now, meant running surpluses, paying down debt, and limiting future initiatives in spending and tax cuts to what would not push interest rates higher than they need be.

However, the government is acting as though the normal rules of prudent fiscal management don’t apply to the road budget. Instead of taking a conservative approach to transport spending, and focusing on the projects that offer the best long-term value at a relatively low cost, they’re pushing ahead with plans to build a number of expensive road projects. For example, the 2014 Budget announced another $800m in motorway projects in Auckland, paid for in part by borrowing, along with $212m in regional road projects paid for out asset sale proceeds.

As an economist, I’m nervous that the motorway spend-up will have perverse economic effects, driving up prices and crowding out other activity. When I went to look at the data, I found reasons to worry.

First, I looked at the data on New Zealand’s spending on roads (from OECD.Stat). The following graph shows investment in new or improved roads as a share of GDP. Essentially, New Zealand spent a fairly consistent amount on roads – and much less on public transport! – in the 1990s. In 2004, road spending started to increase rapidly, rising from about 0.3% of GDP to 0.7%. Unfortunately, the OECD’s data doesn’t cover the last few years, but NZTA’s data suggest that road infrastructure spending has risen further.

NZ road spending, 1995-2011

So the government has boosted spending on roads over the last decade. Has this had any impacts on inflation?

I used Statistics NZ data on inflation to examine the effects. The following graph compares the Capital Goods Price Index for civil construction, a measure of construction cost inflation, with the Consumer Price Index, which measures inflation in the general economy. (NZTA uses a slightly different composite measure of road construction prices, but I’ve chosen to look at civil construction prices as they provide a better indication of potential crowding-out effects on private construction.)

As you can see, the CGPI for civil construction tracked closely with CPI from 1995 to 2003, when road spending made up a relatively constant share of GDP. Between 2004, when the road spend-up started, and the Global Financial Crisis in 2008, civil construction prices rose much more rapidly than the CPI. Construction price inflation briefly cooled off in the aftermath of the GFC, due to falling private-sector demand for construction. But over the past two years it has again started rising faster than CPI, as a result of the economic recovery and spending on major road projects such as Waterview.

NZ civil construction inflation, 1995-2014

In short, spending an increasing amount of money building roads has coincided with a big increase in the cost to build roads. This is exactly what most economists, along with the current Finance Minister, would have predicted to happen.

Moreover, this is likely to have a significant negative effect on private construction. If you want to build an apartment building or a warehouse, you’ll have to compete with NZTA for bulldozers, cranes, and construction labour. Expect to pay higher prices as a result. If civil construction prices had continued to rise in line with CPI over the last decade, rather than being bid up by road spending, big construction projects would be almost 20% cheaper than they are now.

To be fair, other factors may have also played a role, such as the run-up in house prices in the 2000s and increased oil prices. But as Bill English says, government spending shouldn’t exacerbate the inflationary pressures that exist in a growing economy.

This is a tricky dilemma for any government. On the one hand, we do need to invest in a transport network that can accommodate future growth in Auckland. On the other hand, it would be better if the government’s spending didn’t create perverse outcomes for the private sector. If it wants to steer clear of the Scylla of underinvestment and the Charybdis of inflation, it would make sense to look at cheaper alternatives – e.g. the Congestion-Free Network.


Share this


  1. Pork barrel politics. It’s nice to think that decisions are the result of rational analysis but the politicians know that pandering to uninformed opinion, ignorance and prejudice really works. A bit disappointing, really.

  2. What happens if spending up at large in a growing economy? The answer depends on whether the economy is at full employment (in which case there should be an inflationary effect), or if the economy is not at full employment (in which case inflation is not an issue, demand deficiency is).

    Are you seriously saying that today’s economy is a full employment economy?

    1. Seemed pretty clear that civil construction inflation is being discussed. Try recruiting civil engineers in NZ at the moment. It’s a challenge.

      1. Civil construction is greater than just roads so what impact from other civil construction during the 2000s? What is contribution of civil construction inflation to overall inflation?

        Still comes down to if not full employment economy inflation is NOT a concern.

  3. What, in practical terms, does this constraint on supply mean? I would expect that it is harder to build large apartment blocks than it otherwise would be. But how specific are engineering skills? Does the large increase since 2005 also mean that there is a corresponding increase in supply of engineers and other skilled workers in this industry, who will have capacity if the RONS and other pork-barrel projects are scaled back or cancelled?

    1. Constraint on supply leads to having to pay more, recruiting from overseas or possibly delaying projects with budgetary constraints. It’s not all bad though…if you are a civil engineer you can command a higher price.

  4. I’ve had conversations with large construction companies (who are building lots of roading projects) who aren’t overly happy about the amount of projects on as it’s increasing costs for them due to having to hire staff from overseas at great cost just to even try and tender for the jobs. This is because their usual workforce is tied up in other projects. They would prefer a constant amount of work to be coming in so they can use their resources more efficiently and have staff move from one project to the next.

    1. “I’ve had conversations with large construction companies (who are building lots of roading projects) who aren’t overly happy about the amount of projects”

      Flat battery on the BS detector, Matt? Maybe the salaried aparatchiks are being squeezed (more hours for the same pay) but the senior stakeholders in those companies love it. They can cherry pick which projects to tender for (we recently had a major construction company just decline to even put in a bid).

      1. Yes but these players are very concerned about their rivals cornering work and relationships, and, in particular, foreign contractors gaining a foothold. Maybe we shouldn’t care about that? Perhaps that would help with the inflation issue? But it does seem crazy to rush to build these roads all at once, like paying more for them is great idea? 20% more!

        1. I’ve got a substantial construction project on my plate at the moment (industrial) and I was just thinking last night that it’s time I upped my rate. 20% sounds like a nice round figure! A big increase in productivity for my business and I don’t have to get myself agglomerated! A win-win.

      2. It’s not that they don’t want to build stuff it’s that they don’t want it all being built at once. They want a smooth and steady stream of project. They are concerned that after the current mad rush is finished that they’ll have to lay off staff unless we move into a permanent cycle of this level of construction (which isn’t likely)

  5. We also know that construction inflation is seriously affecting the housing sector too. I know of a project stuck in limbo despite all the units being sold, but at prices the developer can no longer make money on because of the sudden ramp up in the costs of everything. Not just people, but cranes, concrete, and other inputs.

  6. ramping-up roads expenditure at such a rapid rate is particularly unwise given the resources required for the Christchurch re-build. I’d suggest the inflation stimulated by public expenditure on roads is not just squeezing out private investment, but actually other forms of public investment as well. The only slightly contrary point I’d make (which is especially relevant to people of a Green/CFN bent) is that ramping-up investment in public transport will tend to have similar inflationary effects, albeit in slightly different sectors of the economy, e.g. rail and bus operations.

    So we just need to be careful we’re not the pot calling the kettle black. The need for and benefits of prudent fiscal management (especially smoothing changes in public expenditure) cuts both ways. I think?

  7. Paying more and more, for less and less. There’s a good analysis at ((Table 1) showing that most transport schemes used to produce returns of over 4 times their cost, but most now produce only low returns at about half that. So we’re getting less and less value and at the same time pushing up housing and other costs. One of the excuses my local council uses for much higher than inflation rate increases is the rising cost of construction.

    1. It’s not an excuse, it’s a fact. Council purchases a different basket of goods to the consumer (you and I). They buy a lot of construction material (all those parks with concrete paths, wooden bollards, etc). If these go up in price due to scarcity, then either rates goes up, or CAPEX declines. If CAPEX declines, constituents complain. AKCouncil has done admirable job sweating the rates income, but at some point it can’t escape inflationary impact of construction costs.

  8. The increase in road spending from 2004 coincided with the increase in rail spending from 2004 (when Ontrack was formed). This was largely because successive governments had underspent on transport infrastructure over the previous couple of decades, and it was decided a “catch up” was necessary – for both road and rail. Labour started this process, and National have continued it.

  9. It is something like the Civil Construction cost price index that drives council rates, not consumer CPI. I don’t know whether it is actually the Civil Construction PI, or something that StatsNZ creates specifically for the local government sector.

  10. Perhaps the answer is in the second graph you posted. The construction cost index doesn’t appear to be driving the CPI at all. The slope of the red line ebbs and flows and doesn’t lead the CPI line so perhaps it has no impact or very little impact on inflation.

    1. It’s my understanding that the two indexes are *not* linked. The CPI is the *Consumer* Price Index. Changes in this basket of goods and services drives *consumer* inflation.

      Council buys a substantially different basket of goods and services: bitumen, steel, wood, consultants, concrete etc. I think the name of *this* basket is Civil Construction Index, but I’m not sure. Councils have to pay for *this* basket, *not* the CPI basket. So changes in the Civil Construction basket drives the amount Council is required to pay, and consequently rates.

  11. So the government doesn’t really care about housing affordability, otherwise it would be rescheduling some of these projects. Housing affordability is only a mantra to drive its own policies. Is this a correct reading of the situation ?

  12. We need to be very clear, inflation is a rise in the general price level not a rise in one price or even a rise in prices in one sector. That is supply and demand in action. In the long run inflation is always a monetary problem, to much money for the amount of goods or services and the money is worth less so prices for goods and services are higher. There is argument about how a short run demand can increase prices which spillover into a long run price increase but even if you accept that then you need full employment for it to be a problem (not just of civil engineers but for everyone). Yes government spending can crowd out private spending but it is a big leap from that statement to arguing government spending causes inflation. The argument we should be having is would we be better off spending the money allocated to roads with a low B/C’s to the health sector or funding education. These are fiscal choices that we should hold the Government to account over. Inflation is well under control due to monetary policy regardless of government spending.

    1. Yes, agree. The good thing about RONS is spending in a demand deficient economy, a classic Keynesian answer.

      The bad thing about RONS is that it is not the best value for the money, compared with lots of alternatives.

    2. I’ve been pondering your comment for the last 24 hours and must say that I disagree, mainly because I think you’ve painted an overly disconnected view of the relationships between government transport spending, direct/wider inflation, and implications for monetary policy.

      First point: I don’t think we have a demand-deficient economy in the construction sector. In this context, large lumpy demands – such as those caused by ramping up the RoNs – can cause material increases in prices, some of which can be expected to flow through into higher general prices. I appreciate there will be a lag and some muting of effects due to substitution and/or demand destruction. But given Government transport spending is not particularly price sensitive (especially once consent has been awarded), then I expect the substitution/demand destruction will tend to occur in the private sector.

      Second point: The price effects of large, lumpy transport investments are not necessarily limited to construction sector. This is because they are basically being funded by higher RUC and fuel excise taxes, the price effects of which will be relatively widely dispersed throughout the economy. So government transport spending not only causes a direct and material rise in prices in the construction sector, but also an indirect rise in prices in the wider transport sector. The latter is of course often an input into prices in other sectors. So we have a direct price-effect on construction costs, a linked price-effect in transport costs, and an indirect (but not necessarily small) indirect general price effect.

      Third point: To say inflation is purely a problem for monetary policy seems to me to be passing the buck big-time. What this effectively seem to imply is that large and lumpy increases in Government spending in growing and constrained parts of the economy are perfectly fine, we just need to accept that 1) construction price inflation and 2) interest rates will need to be higher than they would be otherwise. Yes they will, and yes that will reduce inflation, but is that a good outcome? Not by my reckoning. And if the RoNs themselves are having price-effects, then as a minimum this effect needs to be built directly into the business cases for all transport projects.

      How say you mfwic?

      P.s. Notwithstanding the fact we disagree on some things, I do appreciate your contributions to the blog threads; I’ve found your comments to be insightful, considered, and ultimately very useful. Thanks!

      1. Gosh two complements in a week. Goosoid questioned how an “otherwise intelligent..” yes a backhanded complement but I take what I can get! I haven’t had a chance to really think through what you have said yet but my first thoughts are you might be mixing up inflation and growth or lack of growth. Government spending will crowd out private spending- it always does whether it is on a sorely needed hospital or a road no one is going to use. We all accept government spending for some things as they have a benefit that exceeds the cost. The issue with dud investment is it probably reduces national earnings. That is it has a real impact not just a monetary impact. Japan is the case study here. They built a whole bunch of unnecessary roads and bridges hoping to stimulate growth through spending but it failed. People saw the stupidity of it and saved more money as they knew they would have to pay for it all. So growth languished.

        The question you raised is can lumpy investment lead to inflation. I dont know is my short answer. I do know it leads to higher costs for the projects, higher salaries/wages for those in the industry, but as it is money not spent elsewhere then probably not inflation- some other industry suffers. Again we are back to its negative impact on growth. I completed an undergrad degree in economics in 2008 and then did post grad and we were taught that the monetary side of macro was now settled, right wingers say loose money gives a short term growth boost but long term only inflation based on the money equation and IS/LM model. Left wingers got to the exact same answer based on aggregate demand models. The part of macro that is still hotly debated is fiscal spending and what you see as right or wrong depends on your political views as much as any model.
        So the likely answer if we are classical in our approach is that road spending affects the real economy not the monetary and if we are new Keynesian then it depends how you pay for it. My view is that poor investments by the private sector has a simple Schumpterian result of creative destruction. It is inefficient but the economy recovers. But poor investment by governments just impoverishes us all as our money cant be spent on other things. In the case of Rons I would rather we gave free education to tertiary students or we could feed hungry kids or even just give me back my share as a tax cut- almost anything is better.

        1. I have thought more about your questions and offer the following thoughts. At one level anything you spend money on is inflationary as aside from any models the Reserve bank sees any growth of more than 3% as likely to be unsustainable and so they look to raise interest rates. Either they are right and prevent inflation getting out of hand or alternatively they are wrong and happily kill off growth, I dont think anyone checks. If we build a huge works programme when we have full employment then the Keynesian view is the price level rises, but you don’t meet many Keynesians anymore. Friedman claimed inflation was always and everywhere a money issue, too much of it basically. The more modern view is there are linkages and that variations in the real economy cause the minor fluctuations we see. But I can’t see how full employment in the construction sector can have a huge effect on the basket of goods measured for the CPI particularly if the money being spent is money not spent on other things. The real issue is its effect on growth and the opportunity cost of building Rons when the money is needed for other things that would actually improve wellbeing.

  13. Malcolm, I suspect you are right that the government cares little about housing affordability, but they have to be seen to care about it because a sizeable number of voters do. As a consequence Nick Smith, in my view a very able Minister, was appointed to sort it out (read, “we are doing something”). Then the government does something by opening huge green sites anywhere, it does not matter where (but the message is, “we are doing something, we are doing it fast and it will fix the problem.”) Far less important is whether it will fix the problem. When the analysis is done this election will be long over. Of course most sensible people know that building countless homes near Silverdale will make absolutely no difference to spiralling prices in Ponsonby, but again that’s inconsequential, because the last thing that National voters who reside in Ponsonby want is for house prices to stagnate, or for the gay parade to be cancelled, but that’s another matter. And those who wanted to buy in Ponsonby but couldn’t afford it probably live in a less affluent Labour electorate so they don’t matter anyway..

    The bit of the puzzle that I admit I don’t get is that very quietly Nick Smith is allowing significant sites within inner city suburbs to be developed. I suspect that he sees that such development makes sense.

  14. Is comparing the prices of capital goods and consumer goods is necessarily very informative? In general, capital goods can increase productive capacity, and this can put downward pressure on general inflation just as much, if not more, than a rise in the prices of these capital goods as a (relatively tiny) component input cost of downstream consumer goods.

    The real question is: Do these infrastructure investments enable the economy to run more efficiently or not, i.e., do more roads solve congestion or simply attract more congestion?

    Especially when taking “externalities” like increased fuel imports, vehicle emissions and the adverse effects of these on health, community severance and mobility discouragement, it would seem obvious that investment in the Congestion Free Network would be a far better choice – even if it had a similar effect on the price of civil construction capital goods.

Leave a Reply

Your email address will not be published. Required fields are marked *