This year we have seen two key planning reforms rolled out, with the new residential zones and VicSmart both reaching the climactic phase of extended gestations. I have previously argued – here and here – that both have been problematic, but those concerns are secondary to my purpose here. What’s more important is that the ideas of code assessment and “three-speed” zones have dominated discussion about planning reform in Victoria for most of the last decade. Across multiple reviews variations on these ideas have bobbed up repeatedly as the solution to our problems.
Now both are done (even if code assessment morphed into something different in VicSmart.) And whatever you think of these reforms, either as conceived or as in fact implemented, we now have an opportunity to outline some new directions.
Some Observations About the Victorian System
So where might we go from here? Well, I would like to start by making a pair of observations about the VPP system.
Firstly, our system is better suited to managing incremental change than wholesale transformation. The VPP system (and its principal residential planning tool, ResCode) is based upon highly discretionary controls and a strong emphasis on neighbourhood character. The correct “answer” for any given site is therefore driven by a process of site analysis and detailed consideration of context. The frustrations this creates in terms of lack of certainty for users of the system have been much discussed. Less commented upon is the performance of such a system over time in areas subject to heavy development pressure.
ResCode’s discretionary and contextual approach favours a “match what’s happened and go a bit further” response to infill development. We can all picture the classic ResCode two-storey pitched-roof unit development in a street full of detached single-storey homes. It’s a safe and inoffensive response, and works well in established suburbs where there are relatively few medium density incursions. For this reason the ResCode model has effectively defused the intense battles about medium density infill in wealthy middle ring suburbs that occurred during the 1990s.
In areas under greater development pressure, however, such a context-driven model has problems. As subsequent waves of development occur, each responds to a different context: the earliest unit proposals respond to a “pristine” single dwelling precinct, while the later ones respond to a streetscape of units. The first infill developments are therefore conservative, but the “match what’s happened and go a bit further” response will gradually lead to an increase in intensity that is often poorly thought out.
When such wholesale change is occurring, ResCode therefore gives inequitable outcomes across different sites, depending on when in the cycle they redevelop. And because the goal-posts shift over time, such an approach struggles to help us outline or achieve a clear vision of the desired “end-state” for a neighbourhood.
The second key issue I wish to highlight is that we lack good tools to speed development up or reward good proposals. Under the VPP system, if we wish to outline a vision of change for an established area that we wish to see increase in density, we have two key options (putting aside the new zones, to which I will return in a moment). We can describe the desired outcome in policy, or we can put in an overlay such as a Design and Development Overlay (DDO) to describe the new vision.
The first problem here is that policy is weak as a facilitative measure. It may increase prospects of council support but its impact on process, or procedural risks such as third party objections or rogue council decisions, is minimal.
Yet tools like DDOs, while stronger than policy, have their own drawbacks. Firstly, the fundamental mechanism of traditional overlays is constraint, rather than facilitation. At a very minimum, traditional overlays must act as a permit trigger if they are to have any effect at all. Most also impose additional development standards. So, for example, a DDO in an area earmarked for growth might apply a height limit that is higher than would apply elsewhere, but it is still applying a limit rather than something genuinely facilitative. Ironically, then, a control such as a DDO designed to facilitate change will actually increase regulatory burden in that area.
Secondly, we struggle to create consistency with DDOs. Mandatory controls always carry risk of unforeseen restriction of reasonable outcomes, and as such have traditionally been frowned upon, especially where growth is envisaged. Yet discretionary controls usually end up as a “discussion-starter” and encourage constant testing of the boundaries. If, as is usually the case, a building needs a permit regardless of whether it is at or above the discretionary height, why wouldn’t a developer try to seek a little extra in their application? This increases system conflict and leads to patchy outcomes. Controls such as height limits, for example, have proven very ineffective at creating built from consistency under the VPPs, because every application of discretion ends up a little bit different.
Our Existing Responses
If we look at the two signature reforms I have mentioned, their impact upon these issues is limited. The new zones do give better options in terms of directing growth, yet even the Residential Growth Zone is a constraining tool: like a growth-directing DDO, its constraints such as height limits kick in at a higher level than other zones, but they are constraints nonetheless.
VicSmart, meanwhile, is limited in its application by its focus on small proposals. And because it ultimately emerged as a conventional merits assessment model, modifying it for larger proposals would lead us back to a similar model to conventional VPP clauses. Neither reform, then, breaks from the basic combination of merits assessment within constraining provisions.
However the debate about code assessment gives a reminder of a better approach. Before code assessment became the reform of choice, 2003’s Better Decisions Faster review had instead focused on “deemed to comply” provisions. Ultimately that approach was not taken up, with the 2006 Cutting Red Tape in Planning Report instead diverting attention to a pairing of code assessment and fast-track processing that dominated the reform agenda for much of the next decade.
I would argue, however, that the “deemed to comply” approach was always more promising, and would have avoided the conceptual muddle that ultimately ended up leading to the codeless code assessment of VicSmart.
With VicSmart, the new zones, and other reforms we have been overwhelmingly focussed on a particular set of tools – permit processes, prohibitions, and assessment standards – and have allowed for greatly increased customisation of these for local circumstances. Yet we have neglected the potential of changes to permit triggers, particularly at a local and highly customised level. Our work on permit triggers, for example, has largely been limited to tweaks to standard exemptions under clause 62. Yet statewide permit exemptions have to be done carefully, so such changes have generally been tentative and therefore of limited benefit. The basic triggers for planning permission for housing (constructing more than one dwelling on a block, and construction on a lot of below 300 or 500 square metres) have remained essentially untouched.
This obsession with standards and prohibitions, but unwillingness to play with permit triggers, means we have operated with one hand tied behind our back. Permit triggers are actually the most powerful facilitative measure in our statutory toolkit, and it is time that we started using them better. What we need is a model that allows for “deemed to comply controls” that have the same potential for localisation we have built into development standards.
“Preferred Development Overlays:” A Better Way?
What form might a revived version of a “deemed to comply” model take, and how might it address the observations I have made about the planning system? My suggestion is what I have called “Preferred Development Overlays.” (“Development Facilitation Overlay” perhaps makes more sense, but I think “development facilitation” has developed some uncomfortable connotations over recent years).
I have drafted a rough version of what this might look like: the changes to the General Residential Zone are here and the draft overlay itself is here. However the core idea is simple enough. Firstly, an exemption would be added at the planning permit triggers for medium density and small lot housing (in the General Residential Zone this would be clauses 32.08-3 and 32.08-4) stating that these triggers do not apply if a development meets criteria outlined in a Preferred Development Overlay (PDO). The PDO itself would then set up criteria for exempt development. Development that met the criteria would then be as-of-right in the absence of any other permit trigger. ResCode building requirements would still apply as a baseline standard.
No change would occur to assessment of development that did not meet the PDO requirements. They would simply have the usual permit requirements and be assessed under clause 55 or 54 as appropriate. The PDO would thus be genuinely facilitative: it wouldn’t add any extra burden on development, or constrain any existing rights. Instead it would allow councils a pro-active way to identify, describe and encourage development patterns with which they are particularly comfortable.
The scope for criteria would be open ended, although with some road-testing it should be possible for the Department to establish some suggested templates for Councils to adapt. However I would envisage a key application might be making low intensity medium density housing as-of-right subject to some key requirements about site coverage, lot size, and height. The point here would be to try to establish a desirable and sustainable pattern for individual streets, and then incentivise construction that matched that form.
The overlay approach would mean these could be either applied widely, or tinkered with on a street-to-street basis if a finer-grain customisation was desired. Once some basic models had been tested, it would be envisaged that the strategic work to tailor them to particular streets would be relatively limited. When some confidence in a particular set of criteria was gained, desktop assessments from aerial photography could quickly identify the basic extent of application. If it was anticipated that the criteria may lead to issues on individual sites, they could simply be left out of the overlay. For example, if a portion of a street abutted a series of adjoining properties with very small back yards, and it was envisaged that those sites might therefore be particularly sensitive, the extent of the overlay could be easily customised to avoid those properties. With traditional ResCode application routes always available as a fall-back, Councils could afford to be careful in their use of the tool, since nobody would be losing any existing opportunities. This can be contrasted with tools such as the Neighbourhood Residential Zone, where the risk of unforeseen inequities through unforeseen prohibition of reasonable development outcomes is high.
The “carrot” for councils is that they could achieve more of their desired housing growth with the kinds of developments that their communities are comfortable with. For example, a PDO might be based on a single storey form in the rear half of the property, encouraging more redevelopment to take on that form. This would be a boon for one-off developers, such as homeowners who simply want to better use an under-utilised rear yard. The elimination of the delays and uncertainty of the permit process would mean small developers – those most at threat from the NRZ’s restrictions – would have a quicker and more certain path to an easy return. The message to residents would be that the PDO was improving outcomes by rewarding those developers who did not push the envelope.
The PDO controls would be stackable, so multiple different solutions could be facilitated. This might allow, for example, customised overlays to support certain specific scenarios: for example, one PDO might describe a model for as-of-right “Fonzie garages” or granny flats, while another might outline a model for simple low-impact unit developments. More ambitious infill models would be possible, if Councils felt bold enough. A carefully drafted PDO could be a very powerful supplement to the Residential Growth Zone, for example.
PDOs could stake out out effective as-of-right portions of properties, while leaving other areas as the contested zone. For example, on streets with Victorian terrace housing, the traditional building envelope could be described in the PDO, but then buildings and works intruding into the traditional back garden area would need permission. Such a control would still serve a purpose even when another trigger existed: if a heritage overlay covered the property, that trigger would still apply, but the PDO would still have taken the clause 54 or 55 issues off the table and simplified the assessment.
Over time, it might be expected that a PDO would be better than traditional incremental / contextual controls at directing development to a particular outcome, as there would be encouragement for properties to redevelop to the common PDO-described moderate growth alternative. This has the potential to be more equitable than the current model, which often sees a stark split between traditional single-dwelling sites and infill development sites that have needed to redevelop at a density that covers the speculative and holding costs of the current permit process. The aim with the PDO would be to see more sites redeveloped, but to a lower and more consistent intensity.
Meanwhile, those developers did wish to pursue non-PDO compliant development would still see benefits from the approach, since the reductions in permit loads for councils would allow more resources to be devoted to the medium density proposals that did need permits. This should lead to better outcomes and more timely processing of such applications.
I have described the PDO in residential terms, but there is no reason the same basic mechanism couldn’t be established to other triggers, such as the onerous buildings and works requirements in the commercial and industrial zones. Other refinements could include allowing the same basic mechanism to be included as a zone schedule for PDO-style exemptions that were sought for an entire council area (this might apply, for example, to exemptions such as the aforementioned “Fonzie garage” control). Over time, such council-wide exemptions could prove a fertile ground for councils to test exemptions that the Department could consider for use as state-wide exemptions.
While the application of the PDO could no doubt be imposed from state level if a government felt bullish enough, the real intent here is to give a “bottom-up” tool that gives councils – who currently have little ability to control what needs a permit – more ability to control their own outcomes. The Department could, however, serve a valuable role in facilitating a central “clearing house” of useful PDO controls. As councils developed models that worked well, the Department could draw these together in a central database of PDO templates for other councils to consider in their own areas.
I don’t pretend that there is anything radical or amazing about such an idea. However I do think that we have been overly oriented to a couple of key answers to the planning facilitation, and that our existing solutions have some key gaps. In particular, I think the PDO approach I have described would be more genuinely useful for councils that are currently under enormous pressure. Where VicSmart is the statutory equivalent of someone standing at the officer’s desk shouting “JUST DO IT FASTER,” PDOs would offer the chance for a control that says to the same officer “now, which of these applications don’t you need to see?”
It strikes me as a better way forward, and a constructive tool to further progress residential development reform in the post-new zone era.
Addendum, 17/11/14: One bit of immediate feedback I have received on this is that this tool would make more sense if implemented through a schedule to the zones themselves. Certainly that would be a more classical bit of VPP drafting, as my suggestion creates a cross-control reference which is not normally the done thing.
Technically there’s no reason that it couldn’t be done within the zone schedules. However, any given site can only be in one schedule to a zone, which means any mixing and matching of controls and exemptions needs additional schedules. This could quickly cause the number of schedules to multiply out of control, unless the exemptions applied happened to align with the boundaries of existing schedules. I wouldn’t necessarily expect them to do so, since the schedules are currently applied to quite a different task, and I envisage application of individual PDO exemptions to be highly tailored for individual site circumstances.
If done with the zones plus PDOs, those differing controls can simply be stacked one on top of the other. If done entirely through schedules, every variation of a schedule standard with or without an envisaged exemption needs a different zone schedule drawn up.
Not sure how clear that explanation is – let me know if it still doesn’t make sense.