Costs of growth: greenfields

Much has been made of the need to open up more land on the fringes of Auckland to help drive down residential land prices and free up housing supply. Setting aside the issues of transport accessibility of fringe areas and the daily costs of getting to work or education, usually the first question asked is who is going to pay for all the infrastructure. Auckland Council is hobbled by its debt limits and nobody wants to put up rates to pay for the necessary bulk infrastructure.

Given the funding constraints present, Auckland Council’s Chief Economist, David Norman, has called for "more innovative solutions to deliver the infrastructure Auckland desperately needs as traditional funding options are impracticable" (See note 1). I think that means: "get someone else to pay".

It has also long been claimed that urban expansion is subsidised to an extent, with new households and businesses in greenfields areas not facing the full costs of the necessary infrastructure. This is in terms of capital costs of infrastructure, but also operating costs. Environmental costs are better addressed these days than they were 20 or 30 years ago when urban expansion meant streams were filled in, bush cleared and sediments discharged into coastal areas. But the issue of infrastructure funding remains. In the past we tended to average out infrastructure costs across the city, with core bulk infrastructure seen as a basic public service. While this may have helped expansion to get along, it possibly lead to too much growth in some areas. If costs are artificially low, then it is easy to over consume a good.

One of the potential benefits of the current focus on a more responsive planning system is for the 'subsidies' for urban expansion to be identified and better addressed (i.e. reduced or eliminated as economic theory suggests should happen). The issue also helps to highlight the point that tools like the old Metropolitan Urban Limit Line were partly designed to counter act these incorrect price signals. The new RUB has kind of missed this point (or rather relies upon better prices signals to be put in place to work properly).

So how to address the funding issue and the price signal issue? Easy, say many people, just make the people who are going to benefit from the infrastructure pay for the infrastructure. Set up some sort of special agency. Get them to build or fund the required infrastructure. The costs of infrastructure could be recouped through either the developed land sale price, or through targeted rates on that developed land.

There are a few questions:

1. Will this cover bulk and local infrastructure costs? How much if both?
2. Does the agency get access to lower rates of finance that the council enjoys?
3. If private capital is involved, will there be a return to these 'investors' ?
4. If costs are recouped through rates, over what pay back period - 10 years, 30 years, 100 years?
5. Who carries the can if costs blow out and / or growth is much less than anticipated?
6. And what happens if the infrastructure built by an 'special agency' turns out to be a bit shoddy?

Lets not get side tracked in the details. The first question is how much infrastructure spending for a new dwelling in a greenfields area are we talking about?  One estimate is as follows:

1. Local (civil) infrastructure costs: $80,000 per dwelling
2. Bulk infrastructure costs : $105,000 per dwelling

The local (civil) costs cover the costs of pipes, cables and roads in a subdivision. This estimate comes from MBIE’s handy ‘development feasibility calculator’ (note 2).

The bulk infrastructure cost is calculated by looking at the Auckland Council's latest (2017) land supply strategy (note 3). This identified $19 billion dollars needed for motorways, bus and rail, arterial roads, the three waters and social and community facilities to provide for all the housing in the future urban areas identified by the Auckland Unitary Plan. There is space for 120,000 dwellings. There will also be employment areas and new town centres and the like. A simple calculation suggests that about 2/3rds of the demand will come from the dwellings. 2/3rds of $19 billion is $12.5 billion, divide that by 120,000 and you have $105,000 per lot.  

As an aside this cost estimate is roughly double what a Council study suggested a few years back. Council’s 2015 Cost of Growth study said that greenfields growth generated infrastructure costs of up to $50,000 per lot (note 4). I think these were bulk infrastructure costs, but what they counted as bulk infrastructure may have been different from the Council’s latest land supply strategy.  The 2016 land supply strategy identified $17billion in costs.

The local civil works already get built into the price of the section. At the moment, it is a bit unclear who pays for the bulk infrastructure. Council’s development contributions may be about $20,000 to $30,000 per lot. These cover some of the bulk infrastructure costs. Potentially the rest is something we all pay for through tax on petrol, income tax, user charges and general rates.

Rather than be subsidised by us all, the theory goes that the user of requred infrastructure extensions should pay. I know that a motorway extension that is built to serve, say, Kumeu will end up being used by lots of different people, not just residents who live in Kumeu, but let’s not quibble. Without the planned growth, the motorway extension may not happen. So if the user pays, then this is an extra $70,000 to $80,000 per lot to cover the bulk infrastructure costs.

Total local and network infrastructure costs are close enough to $200,000 per section. Then we need to add in finance costs, as someone has to borrow the money to build the infrastructure (or pay for it to be built by the likes of Watercare). This bit is a bit tough for me to work out, as it all depends upon the pay-back period and if commercial interest rates apply. If the extra costs are capitalised into the sale price of the section, then the developer may only need to borrow the money for 5 to 10 years to cover the development and sell down period (and at that point the new landowner effectively takes over their part of the  debt in their mortgage). If it is paid  back is through rates, then maybe the loan term is 30 years. If the extra costs are paid back through rates, then in theory the price of the section should drop back a bit to reflect this financial commitment. If the cost of the infrastructure is reflected in the section prices, then presumably the developer needs to pay less for the raw land as they may not be able to put up the price of the section by too much. Otherwise buyers will go elsewhere.

Add to this is the cost of the raw land, design fees, consent fees, contingencies and of course a profit allowance for the developer, and GST.  Nothing is cheap. I'm not too sure how much is set aside for earthworks in the above MBIE estimate. Seems to me in Auckland’s hilly terrain an awful lot of time and effort goes into flattening things out.  

I guess the idea is that if enough land is freed up by getting the funding of bulk infrastructure away from the financially strained council, then raw land prices will fall. But would they fall by enough to off-set the costs of the bulk infrastructure that are now to be directly paid for by the user?  Too much land freed up means possibly slower sales of sections and houses for each individual developer, making the the costs of paying for all that infrastructure up front much more risky and hence expensive.

Of course there is also the problem that extending trunk infrastructure is not a straightforward exercise, especially if it needs to pass through or over multiple properties to get to where the demand is. No private agency can achieve that. What about quality, any worries there?

So are we back to the public using its powers to borrow more cheaply for the long term, to carry more risk and to use its planning and designation powers to build infrastructure being not a bad way to help development along?

And what about the infrastructure costs of urban infill and redevelopment - are these costs higher or lower than urban expansion? The 2015 Costs of Growth suggested that they were lower, but a lot depends upon the state of the infrastructure present in the areas subject to redevelopment: is there spare capacity that can be used at limited cost; or does the infrastructure need (expensive) extra capacity to cope; or is it old and needs replacing anyway, so more people and businesses help spread the costs? One simple way to look at the issue would be to say that in comparison to greenfields, infrastructure costs of up to $40 billion for infill and redevelopment would create an even playing field with greenfields, given that around double the growth is expected in the existing urban areas over the next 30 years compared to urban expansion (around 230,000 houses compared to 120,000). That is quite a bit of infrastructure that could be upgraded.

(1)

http://www.aucklandcouncil.govt.nz/EN/AboutCouncil/businessandeconomy/Documents/akleconomicquarterlymay2017.pdf

(2) http://www.mbie.govt.nz/info-services/housing-property/national-policy-statement-urban-development

(3) http://shapeauckland.co.nz/media/1782/future-urban-areas-anticipated-dwelling-employment-capacities-cost-table.pdf

(4) F I N A L R E P O R T, Cost of Residential Servicing,Prepared for Auckland Council, January 2015. THE CENTRE FOR INTERNATIONAL ECONOMICS

Auckland and cycles of urban change

In considering Auckland and how to manage its growth and development, I wonder if we should focus more on the bigger cycles of change that flow through the city and the economy. For example, post war saw the expansion of the suburbs and the decline of the inner core of the city - for a while some of the cheapest real estate in the city was in Ponsonby and Parnell. In the 1980s, prices in the inner city started to rise.  The process of decline opens up the opportunity for renewal. It is this linked process of decline then rejuvenation that really propels the type of urban redevelopment that we now seek for urban Auckland. The mix of relatively cheap but increasing land values, older building stock and growing demand create the perfect environment for urban redevelopment. But in a way we missed the boat with the inner Isthmus area, or maybe more correctly changed boats; we gentrified the inner suburbs rather than redeveloped them.

A side question: is gentrification a market-based mechanism? Is it wrong for a bunch of residents to effectively form a cartel to maintain their current environment? Something to come back to.

What I want to look at in this blog is whether there is another period of decline then growth around the corner for us to plan for and exploit?

First lets look at whether urban super cycles exist. If we look at Auckland’s population growth since 1900, then we get the above graph.

It looks like a pattern of consistent growth, with a few soft patches. If we break the figures down into population change by decade, then the following is the outcome. Some cycles of faster, then slower growth are a bit more apparent.

There are three periods of growth:

·        Up to the mid 1920s

·        From mid 1940s to mid 70s

·        From mid 80s to today.

And two down turns in between.

One theory places the length of these super cycles at about 50 years, from trough to trough.

Of course within the super cycle there are smaller periods of ups and downs, as the following graph shows for the period 1996 to 2016: Two ups and two downs.

The super cycles of growth then down turns can be related to a number of factors.

1. External shocks like the depression and WWII in the 1930s and 1940s, and then the period of stagnation in the mid 70s the followed the oil shocks and, I think, the entry of the UK into the EEC.

2. The super cycles can also be correlated with infrastructure and technology effectively leading the growth of the city:

a.      Up to the late 1920s, the city grew due to sewage and water supply being securely provided, as well as the tram network expanding.

b.      The mid 1950s to mid 70s are related to the initial development of the  motorway network which radically changed accessibility profiles.

The other interesting relationship is with  "kondratieff waves". This is the concept of long run economic cycles driven by technological change. Typically these long run waves are organised something like this:

1. (1600–1780) The wave of the Financial-agricultural revolution

2. (1780–1880) The wave of the Industrial revolution

3. (1880–1940) The wave of the Technical revolution

4. (1940–1985) The wave of the Scientific-technical revolution

5. (1985–2015) The wave of the Information and telecommunications revolution.

Below is one attempt to map these cycles with the American share market.

The concept of kondratieff waves is not widely accepted, but it is an interesting idea.

Every wave of innovations lasts approximately until the profits from the new innovation or sector that propels the wave fall to the level of other, older, more traditional sectors. The new technology reaches its limits and it is not possible to overcome this limit without an application of another new technology. At the end of an innovation phase of any wave there is typically an economic crisis and stagnation, often associated with excess capital built up over the period of expansion. Land can be the home of some of that capital and real estate booms then busts are a common way to 'correct' the excess accumulation of capital.

Part of the concept is that the waves of innovation and crises are speeding up. Whether the current growth period that started in the mid 1980s extends more than three decades is debatable.

One interesting thought is where does urban planning fit into these patterns and what are the implications of every shorter periods of innovation and stagnation?

Part of me thinks that planning often takes a bit of time to react to a period of innovation - the consequent change to urban patterns from that innovation and developing the necessary management tools. For example, it took a while for public services like clean water and waste water systems to be installed in response to the urbanisation of the late Victorian period, while transport systems like the tram were invented to cope with density, but also to allow the city to spread out a bit and breath. Zoning was invented as a way to manage incompatible activities.

Come the industrial city of the post WW2 period, it is only really in the 1970s that serious effort starts to be put into structure planning to shape the greenfields growth enabled by the car and the motorway. Up to that point, the natural environment paid a heavy price. Since the mid 1980s and the most recent phase with its emphasis on intensification, it has taken a while to sort out how to better manage urban infill (such as the rise of urban design). Even then you could say that the response has been patchy.  Have we really got to grips yet with the post industrial city of mixed uses and mixed densities?

In this sense I do agree with the likes of the Productivity Commission that planning can be a bit slow to respond. But rather than speed planning up, should we shift when we do planning to give us more time? One thought would be that we should use the down periods to better plan for the next phase of change.

Part of the problem is that planning and downturns do not really mix. For example, Auckland kind of went to sleep over the late 70s and early 80s as growth slowed down - no need to brush up plans and invest in new infrastructure like better rail and bus systems, ready for the next phase, if growth is slow - may as well relax instead. Growth spurs the need for planning, with plans taking some time to develop once growth pressures come on stream and so plans are often ready only in the last phases in the innovation cycle.

The other interesting issue is whether planning makes the most of the opportunities thrown up by these cycles. Again this is kind of counter intuitive. The post war expansion left much of the inner core of the city as an area of decline as the city spread out. It is easy to get focused on managing the expansion, not the area in decline. But that decline makes fertile ground for redevelopment  once conditions change. However rather than redevelop (apart from a few pockets of ‘slum clearance’) the emphasis went on historic preservation as part of gentrification of the inner suburbs. We can’t undo that gentrification (and neither should we). But what would Auckland be like if plans for partial or selected redevelopment of the core had progressed more strongly during the period of suburbanisation? Perhaps we would have had much better thought out approaches to apartments in the CBD and in the inner area.

The opportunities thrown up by the present 5th phase are harder to pick. This phase has been marked by a return to the ‘core’ of the urban area as transport accessibility at the edges starts to deteriorate, and technology has reinforced the benefits of proximity (rather than dispersal) for some urban activities. One area where I think we have not done enough is to think through the implications of the rising land values in inner and middle ring suburbs for industrial and commercial activities. Many industrial and distribution activities located in the Isthmus would, I think, logically seek out cheaper, more peripheral sites if land was available for them to do so, particularly to the south. Such a move out would release large areas of brownfields land for mixed use type development. The northern shores of the Manukau harbour are surely an example of this, along with coastal peninsulas like Rosebank Road.  Instead we seem to want to hang onto the industry in these areas and build more motorways to serve these areas, but which cut these areas off from the coast and must stymie potential redevelopment.

These points could be taken as reasons for less planning and more room for market forces. If planning had not supported gentrification through heritage zonings, would the city’s inner suburbs now be able to be redeveloped into mid rise apartments in response to the fifth wave? If the urban boundary was not in place, would much industry already have shifted out of the city into greenfields, opening up even more land for redevelopment?  

But each of the longer wave phases has also been marked by countervailing public measures that have helped to spread the benefits of the wave and created a better society - basic zoning and public health measures in the third wave, the welfare state, the start of environmental management and public housing in the fourth (post war) wave and the rise of the importance of education and training in the fifth wave, along with a demand for urban quality. So in thinking about the role of urban planning, while planning may have slowed some desirable change, it has also supported and enabled the long run economic cycles through tools like zoning, environmental management and urban design.

What is around the corner, in the next (sixth) phase after an inevitable crises? Can we guess what is around the corner? The concept of super cycles doesn't tell you what will be the strength of the 'up cycle', for example, nor necessarily its spatial pattern. It is easy to get things wrong - many people in the mid 1980s predicted that the fifth wave would see the further dispersal of the city as new technology allowed for remote living and tele-commuting. Instead the city bunched up some. For one take  on what the next cycle may mean for cities, see: Creative Destruction, Long Waves and the Age of  the  Smart City, by Michael Batty.