Big city life? Challenges and trade-offs for Auckland
In a recent working paper, NZIER have analysed how economic welfare responds to changes in housing and transport costs in Auckland. Three policy scenarios were tested, specifically:
- An increase in land supply outside of the MUL by 20%;
- An increase in productivity in the housing construction sector; and
- A reduction in transport costs of 2%.
In this post I want to focus on their findings with respect to the metropolitan urban limit (MUL), where they find (emphasis added):
Moving the MUL out improves land supply and decreases housing costs. Families across the city benefit from reduced housing costs – even though commuting costs increase for some. The net impact on all families from expanding the MUL is always positive (pg iii).
This statement – if true – would have major implications for land use policy in Auckland. Put simply, it suggests the economic benefits of expanding the MUL (lower housing costs) are greater than the economic costs of doing so (higher transport costs).
This may be true. However, in this post I will argue the working paper (in its current form) does not present a very comprehensive or useful assessment of the MUL as a policy. In my mind there are three key questions to consider when evaluating the economic effects of policy interventions, specifically:
- What are the objectives of the policy intervention? Specifically, what does the policy set out to achieve?
- Is there a prima facie case for market intervention? Specifically, do we have evidence of imperfectly functioning markets?
- What are the benefits and costs of the policy intervention? And what is the sensitivity of these results to changes in our assumptions?
The first question is fairly straightforward: it considers what the MUL hopes to achieve. Advocates of the MUL typically point to a range of benefits in terms of coordinated infrastructure planning (transport, wastewater, communications, electricity, as well as social infrastructure, e.g. schools) as well as environmental benefits from directing developing to areas where it’s external effects are mitigated or at least able to be managed. Hence, any economic evaluation of the MUL needs to at least acknowledge, and preferably quantify, these kinds of benefits. I understand that development at Hobsonville, for example, is to be accompanied by investment in ferry terminals, state highway upgrades, and investment in schools (photo source).
The second question relates to whether there is evidence of imperfectly functioning markets, where individual firms and households (whom I collectively refer to as “market participants”) are unlikely to make wise decisions. Imperfect market functioning typically arises where market participants are influenced by:
- The unintended consequences of other policy interventions, e.g. minimum parking requirements;
- Inaccurate (i.e. non-marginal) price signals, such as road excise duties;
- Externalities, such as road congestion and agglomeration economies;
- Imperfect information, such as future road congestion levels;
- Search/transaction costs, such as moving house and changing jobs; and
- Path dependencies and irreversibilities, such as development of agricultural land.
Well-designed policy interventions should seek to understand the reasons why market participants are not making wise decisions, and influence behaviour accordingly. The externalities associated with developing greenfields land is particularly relevant, I think. After all, while new residents “win” from moving out the MUL, existing residents (including those just on the fringe) may “lose”. This was, I believe, one of the main objections to an expansion in the MUL at Long Bay, illustrated below (photo source).
The third and final question considers the benefits and costs of a specific policy intervention. Costs are unavoidable and arise in the form of 1) welfare losses to market participants, who – as a result of the policy intervention – are now unable to act exactly as they please; 2) compliance costs, such monitoring/enforcement; and 3) unintended consequences from regulatory intervention. The supposed economic benefits of a policy intervention need to be compared to these costs.
The third question is where the NZIER working paper has focused most of its attention, in terms of quantifying the effect of the MUL on housing and transport costs. But because the NZIER paper does not grapple with the first two questions, its conclusions (“expanding the MUL is always positive”) comes across as premature at best.
My scepticism in this particular instance stems from two key issues with the working paper as it currently stands.
The first issue is that proponents of the MUL (and I am not one) point to a range of areas in which managed urban development can help to realise efficiencies in infrastructure investment and environmental management. Such benefits need to be included in the benefit cost equation before any firm conclusions can be drawn.
I note that NZIER’s analysis is limited to households, and households may not be directly responsible for some of the relevant costs – e.g. local and central government may shoulder the costs of providing infrastructure. However, households will ultimately (and indirectly) pay for these costs through their rates, and this does not seem to be taken into account. Evidence from Australian studies such as the Perth cost of growth study (Trubka, R, Newman, P and Bilsborough, D. 2008. “Assessing the Costs of Alternative Development Paths in Australian Cities.” Curtin University Sustainable Policy Institute.) and a similar study conducted in Sydney (CIE and Arup (2012) “Costs and benefits of alternative growth scenarios for Sydney – existing urban areas.” Prepared for NSW Department of Planning and Infrastructure.) suggests that development in greenfield areas imposes significant additional infrastructure costs. A similar study is forthcoming from Auckland Council.
The second issue with the NZIER working paper is more subtle but possibly more significant. It relates to the justification for policy intervention. Specifically, the pervasive and long-standing presence of inaccurate transport/land use price signals that have stimulated urban expansions rather than urban intensification. This includes:
- Regulatory barriers to urban intensification, such as building height limits, minimum parking requirements, floor area ratios, minimum apartment sizes, and building set-backs. In a recent talk at the University of Auckland, for example, the urban economist Ed Glaeser indicated that – in his opinion – regulatory barriers to urban intensification were more binding than regulatory barriers to urban expansion, and more critical for Auckland’s situation. Indeed, barriers to intensification may one reason why house prices in Auckland’s central suburbs have increased so much more than peripheral areas, as noted in the working paper. The NZIER working paper mentions such barriers only in passing, as “an area for further research”, in spite of the fact that Glaeser and Gyourko (2002) [link: http://www.nber.org/papers/w8835] developed a methodology for quantifying their impact over a decade ago. I think that is an understatement: economists have ignored regulatory barriers to intensification for too long, preferring instead to focus on the easier to observe, but less important, barriers to urban expansion.
- Inaccurate transport price signals, causedby:
- Vast over-supply of under-priced parking caused by the application of minimum parking requirements; and
- Absence of time-of use-road pricing, which tends to benefit long distance commuting by people with a low value of time.
- Inaccurate land use price signals. Ideally Councils would charge the marginal cost of developing in different parts of the city. In practise, however, it is incredibly difficult to develop robust estimates of how the marginal costs of development varies across the city. Hence Councils tend to set development contributions in a rough and approximate way and instead rely on regulatory tools to stage/manage development in those areas where existing infrastructure is better able to accommodate growth.
While I support and advocate for addressing the above issues, I am under no illusion this will occur anytime soon. For this reason it is important that criticisms of the MUL (or its more recent incarnation the “RUB”) consider that it may have a role as a second-best (interim) policy solution. I say *may* because I’m wary of advocating for a nightmarish “babushka policy doll” scenarios, whereby “second-best” policy interventions, such as the MUL/RUB, are justified primarily on the grounds of other flawed policy interventions, such as building height limits, minimum parking requirements, floor area ratios, minimum apartment sizes, and building set-backs. Naturally, it would be more efficient to address the underlying causes of unfettered urban expansion.
So where does this leave us?
Well, on one hand the working paper makes a useful contribution to the policy debate by quantifying some of the benefits from urban expansion, specifically lower housing costs. On the other hand, it does not discuss – let alone quantify – many of the supposed “benefits” of the MUL. For this reason I believe its conclusions with respect to the effects of expanding the MUL are, at best, premature and therefore unlikely to convince those people who support the MUL.
More generally, however, I believe the working paper is – like much of the economic literature on this topic – rather superficial. It does not consider regulatory barriers to urban intensification, for example, when evidence and professional intuition suggests these barriers are more binding than the MUL. At a recent talk at the University of Auckland, the urban economist Ed Glaeser suggested that regulatory barriers to urban intensification were likely to be more of a binding constraint on urban development than regulatory barriers to urban expansion. This recent study into the effects of removing minimum parking requirements in London, for example, found that developments provided almost 50% less parking after the regulation was removed.
It’d be great if all the good people at the NZIER, the Reserve Bank, and Treasury etc invested a little more effort into “binning the mins”, rather than the (apparently) myopic focus that seems to be placed in fighting the MUL.
And when I say “regulatory barriers to urban intensification”, I am actually referring to the combined effects of building height limits, floor-area ratios, minimum parking requirements, building set-backs, and minimum apartment requirements. In this context, the debate over whether to expand/remove the MUL seems *relatively* unimportant.
Thankfully it’s a working paper, so I’m already looking forward to the next version.