• Increasing the contribution developers make for infrastructure to reflect the true cost of providing that infrastructure, rather than continued subsidies by existing general ratepayers, is unlikely to increase the cost of housing.
• Developers in Auckland, particularly in greenfield areas, pay only a fraction of the cost of infrastructure to serve the new development through development contributions (DCs).
• Setting the price paid by developers at a full cost-recovery level could unshackle infrastructure development from the limits of ratepayer funding, and reduce planning incentives to regulate land use.
• In this paper, we examine the international evidence and the economic theory that suggests more accurate DCs are unlikely to result in higher house prices, but will be absorbed by land owners who currently receive windfall gains from general ratepayer-funded infrastructure.
At their current levels, DCs do not fully recover the cost of building growth infrastructure, and the funding gap is made up by existing general ratepayers. Auckland Council’s Future Urban Land Supply Strategy (FULSS) suggests the subsidy by ratepayers of new greenfield development in future urban zones is likely to be about $50,000 per dwelling over and above the current level of DCs. This $50,000 subsidy per greenfield dwelling from general rates could instead be used for critical maintenance and renewals of our existing assets.
But increasing DCs to a level that recovers costs fully may have consequences for home buyers. In fact there will be impacts on at least one of the following three players: undeveloped land owners, developers, or house buyers.
The added DC costs must either be passed on to home buyers or land owners – or absorbed in developers’ profit margins. But who might eventually pay the increase in DCs – undeveloped land owners, retail purchasers or developers? And how do increased DCs affect the quantity of housing development?
Nice in theory…
In a previous Insights paper, we said that if DCs accurately reflect the full costs of new infrastructure , they help:
• incentivise development where it is “least costly”
• fund growth infrastructure more fairly from those who benefit most from it
• faster sequencing and delivery of infrastructure, which would expand development opportunities faster than if funding was dependent on general rates revenue.
Economic theory says that if retail property prices are already near the limit of what the market can bear or if buyers have other housing options to choose from, the developers’ ability to increase prices to pass on DC increases is restricted.
Are property prices already at levels the market “can bear”?
Very likely so. If prices were not already the maximum the market could bear at this particular point in time, this would mean developers are acting irrationally and “leaving money on the table”. Also, the Auckland housing market is likely at price levels that are at or close to the ability to pay limits of home buyers, as evidenced by the flat lining of prices over the last year or so.
Further, potential buyers’ willingness to pay does not increase if the developer pays a higher portion of the infrastructure costs rather than ratepayers – the house purchaser is still just buying the same product - a home with taps that run and toilets that flush, regardless of who funds the infrastructure.
Do buyers have other choices?
Yes. New homes made up about a third of all the houses sold in Auckland over the last three years. Developers of new homes must compete with the stock of existing homes on price. This means their ability to increase prices is limited beyond what the new build can command over an older home – and that too only up to the level the market can afford.
…and supported by real-world evidence
So theory says the developer will not be able to pass an increase in DCs on to house prices. But does real-world evidence suggest they absorb the cost themselves, pass it upstream to raw land owners or downstream to property purchasers?
Land owners benefit from infrastructure…
Evidence shows that rural land in Auckland that gets re-zoned as future urban receives a significant uplift in value. The announcement of impending infrastructure development and then building that infrastructure multiplies land values by a factor of 40 or more, from around $50,000 per hectare to between $2.25 and $3.25 million a hectare.
At the assumed FULSS density of about 10 dwellings per hectare, an increase of $50,000 per dwelling in DCs to cover the cost of FULSS infrastructure would thus be around $500,000 per hectare. Even if developers lower their bids for land by the full $500,000 per hectare, land owners would still be to sell at prices well above farm land values, generating a healthy profit.
…but when DCs go up, the cost passes upstream…
Empirical evidence on what happens when DCs change in New Zealand is limited. A report produced for Waitakere City Council found that a 25% increase in DCs had little or no impact on house prices (or rents) and a small negative impact on building consent growth, after controlling for other variables that might affect these indicators.
Murray (2016) looked at the impacts of a policy change by the Queensland government in 2011 that introduced caps for DCs as a response to the property development lobby that had argued that high DCs led to high prices. This meant that some local councils were forced to lower their DCs for some dwelling types, but it also allowed them to increase their DCs for other dwelling types.
The four year transition period involved multiple upwards and downwards revisions to DCs for different dwelling types. This provided the natural variation to isolate the effect of DCs on prices and new dwelling supply. Murray found no impacts from changes in DCs on prices and positive impacts on quantity of new dwellings consented in Brisbane and on the Gold Coast. Murray also critiqued other empirical work on this issue for not controlling for the fact that the DC structure there meant that larger dwellings pay higher fees.
Burge (2014) looked at the relationship between specific impact fees (similar to DCs) and the values of undeveloped land parcels using more than 1.5 million residential parcel sales and over 130,000 commercial parcel sales between 1994 and 2009. He found that school impact fees (which were levied only on residential construction and not on commercial development) lowered the price of residentially zoned undeveloped parcels but increased the value of commercial parcels. In other words, developers passed the cost increases up to land owners, but could charge more for commercial properties because of announcing the benefit of a new school, which would generate more passing traffic for businesses.
Nelson (1994) documents a practice by developers in the Atlanta metropolitan area who routinely inserted a paragraph in their land purchase option agreements requiring land owners (from whom they purchased) to reduce their sales prices by any impact fees (similar to DCs) charged. Nelson also documents responses from developers who said that after a transition period, impact fees would be mostly paid by the sellers of land.
Not every study found that all costs passed upstream to land owners.
Delanay and Smith (1989) used two samples in Florida to test whether impact fees are associated with higher prices for both new and existing homes. They found that developers passed on the impact fees to buyers in the short run but over a significant period of time, increased new home prices were observed in one of the four cities that imposed impact fees.
Evans-Cowley et al. (2005) examined the relationship between impact fees and land values in forty-three Texas cities. They found that developers absorbed around 60% of the increases themselves, passed on around 31% of the increased cost to retail purchasers, and around 11% up to land owners on the average lot.
Other common objections
Won’t more accurate DCs discourage development?
If developers are able to pass on the higher DCs to land owners, there should be no significant impact on the quantity of development. If developers have to accept a lower margin, this may reduce the amount of development.
But Nelson argues that funding new infrastructure via more accurate impact fees may in fact speed up the supply of buildable land and help make the housing market more competitive. Further, if growth self-funds new infrastructure, the political economy of planning decisions has less of an incentive to restrict land-use through planning controls and can encourage the adoption of progrowth policies.
Aren’t developers price takers?
Some developers argue that they do not have many options of developable land on the market that they can buy at any given time. They might argue that higher DCs will make land owners less likely to sell (at a lower price). If this is indeed the case in Auckland, then more accurate DCs could be followed by moving towards a targeted rates regime that recovers the equivalent amount of money over the duration of the targeted rate.
This could incentivise land owners to bring land to market faster, among other benefits we have previously discussed. Moving to targeted rates would also add to the revenue base that Council can borrow against (Council is unable to borrow against DC revenue), helping speed up infrastructure provision on greenfield land, making land markets more competitive, faster.
Charging accurately for infrastructure such that those who benefit from it pay for it, is fair. When we undercharge, land owners receive a windfall gain they don’t pay for. Economic theory and majority of the evidence shows that the market price for housing is unlikely to rise by charging more accurately for infrastructure, particularly in the longer term.
Some developers who have already purchased undeveloped land at prices that reflect the infrastructure subsidy from general ratepayers will be impacted in the short term by a shift to more accurate DC pricing. But an announcement that signals a move to cost recovery will allow more “correct” pricing of land in future transactions, and would stop exisiting ratepayers from providing a windfall gain to raw land owners.
Over a period of time, unshackling infrastructure development from the limits of ratepayer funding can help speed up the rate at which land is serviced with infrastructure, providing developers more options and reducing the ability of land bankers to hold development back.
Harshal Chitale is a senior economist in the Auckland Council's Chief Economist Unit. This is a repost of one published an their Insights series, here. It is reposted with permission.