There is a great deal of ruin in a city, to paraphrase Adam Smith’s calm reply to those heralding the doom of Britain after the battle of Saratoga. In other words, it takes a lot to really wreck either a city or a nation. An earthquake alone cannot do it, or at least not one the size of ours. But, an earthquake, combined with paralysing land-use regulatory structures, a confused intersection of building codes and insurance rules, and an army of bureaucrats each equipped with an all-powerful ‘No’, can come much closer than we’d really like.
After February’s earthquake, we had some hope that the city’s planners were starting to recognise the importance of the bottom-up, decentralised response that lets each of us see how best to adapt to the new realities. Coffee shops that were relocating to garages and sheds in Lyttelton were being allowed to operate regardless of zoning regulations – eminently sensible. People were finding ways to help themselves, and each other.
The first sign that this light-handed approach was not to last was the bureaucracy’s attempt to block the University of Canterbury’s student volunteer army. Sam Johnson’s team of volunteers, coordinated by Facebook and responding to a grassroots-level need for help, were hindered by bureaucrats who wanted to micromanage. This, writ large, explains rather well why, in June 2014, so much of downtown remains a shambles. Christchurch’s glacial post-quake recovery really can be traced to these three main factors.
The first, highly restrictive land-use regulations, existed well prior to the either the February 2011 or the September 2010 earthquakes, and are mirrored by similarly stultifying strictures in every major city in New Zealand. This part of the post-quake disaster is a chronicle of a failure long foretold, and easily avoided.
The paralysis generated by the sequence of earthquakes and New Zealand’s particular public/private insurance market, and the problems caused when CCC changed the building code subsequent to the insured event, were only partially foretold. The Earthquake Commission (EQC) warned the incoming National Party coalition in November 2008, that EQC had no capacity to carry out home inspections after a disaster, that the cap on EQC’s coverage per home was too low, and that EQC needed better capitalisation. Governments have a difficult time dealing with things that are important but not urgent; setting EQC on sounder footing never seems important until it’s really very important and too late. But, many of the larger problems were not well anticipated and constitute hard-taught lessons; we would do well to learn from these and fix regulation accordingly before any substantial Wellington event.
Finally, the confusopoly of Christchurch City Council, the Earthquake Commission, the Canterbury Earthquake Recovery Authority (CERA) and the Christchurch Central Development Unit (CCDU) was entirely unanticipated, or at least I had never expected that this could happen in New Zealand.
I had always held New Zealand governance in the highest of esteem, reckoning it to be the world’s best. And perhaps New Zealand’s overall governance really is as good as it gets and other countries are only able to deal with these kinds of events by virtue of greater size and wealth. But no level, branch or agency of government covered itself in glory in the months and years following the initial disaster phase. If this be the best of all possible governments, as international surveys of such things often tell us, what hells be all the others?
While CERA and the CCDU were surely established with the best of intentions, the result was the creation of far too many people who could veto new developments or changes in land use. Economists are well familiar with the Tragedy of the Commons that results when nobody can exercise veto rights over the use of property: the Commons are then over-grazed. But fewer are familiar with the Tragedy of the AntiCommons that results when too many people are allowed to say no. Traditionally applied in analysis of intellectual property, the Tragedy of the AntiCommons applied with vehemence in post-quake Christchurch. When any one of many bureaucrats can delay or ban you from rebuilding, either by explicitly saying ‘no’ or by failing to get around to providing a needed ‘yes’, it’s hard to get anything done. Christchurch has too many veto players.
While the failures are separate, they share a common root: an utter failure of the government, both CCC and central government, simply to allow property owners to get on with the job at hand and to make those changes necessary to allow them to do it.
My colleague, economist Seamus Hogan, reminds us of the analogy in M*A*S*H that, sometimes, meatball surgery is needed. In the Korean War’s mobile army surgical hospitals near the front lines, an extra hour spent providing perfect treatment to one patient doomed others consequently left untreated. In Christchurch, the three-year-long quest for the perfect central city plan stopped anyone downtown from proceeding with any work at all for far too long, bleeding downtown’s prospective recovery as businesses fled for the suburbs, or left town entirely.
In my view, too many city planners spent far too much of their youths playing SimCity, a city planning videogame popular in the mid-1990s, in which you can always press the pause button while you think about your next move and where nobody much objects if you bulldoze their houses or businesses to put up a new park or stadium. Real cities are not like that.
The city would have fared far better had neither CCC nor central government embarked on adventures in planning and instead concentrated on core city functions: ensuring that land-use planning did not stand in the way of expanding the supply of housing after the earthquakes, focusing on core infrastructure rather than masterplanned precincts and buildings, avoiding regulatory and planning uncertainty so that property owners knew where they stood, and quickly sorting out the legal and regulatory morasses that were too late to avoid entirely after the earthquakes.
The failure long foretold: Rigid land-use planning before and after the Christchurch earthquakes
Restrictive zoning practices are hardly unique to Christchurch, or to New Zealand. The effects everywhere are similar: housing has become extraordinarily expensive. While more sensible parts of the Right and Left recognise that we simply have not been building enough new dwellings for some time, they disagree on solutions: the Right lauds new subdivisions while blocking others’ plans to increase density; the Left cheers intensification while blocking others’ moves to increase land supply on the city’s fringes. Meanwhile, councils recognise that home-owning voters reward them for the ever-increasing housing prices that come from restricted supply. Consequently, as Matthew Yglesias put it in his recent Kindle Single, ‘The Rent Is Too Damned High’.
Christchurch was certainly no worse than many places in New Zealand prior to the earthquakes, and was certainly in better shape than Auckland. But the regulatory constraints that served only to mildly increase the cost of housing in Christchurch before the earthquakes became very important after the earthquakes.
Prior to the earthquakes, expansion outwards from Christchurch was fairly constrained, as was any substantial densification outside of specific zoned areas. Solutions found in cities like Vancouver, in which many homeowners built self-contained flats within their existing homes, were prohibited in Christchurch by rules preventing a permitted dwelling from having more than one kitchen.
Surprisingly, neither the September 2010 nor the February 2011 earthquakes resulted in any expedient substantive relaxation of either zoned density regulations or the number of permissible sections on Christchurch’s fringes. Over 12,000 homes were destroyed in the earthquakes, about 7 per cent of the housing stock, with 8000 of those homes located within the residential red zone where rebuilding was prohibited. New subdivisions, increased density, or both were necessary to accommodate the homeless, the families whose homes were being repaired and the influx of construction workers.
CCC did little to ease the resulting pressure on housing. In the short term, new construction would have been very difficult even without council impediment as the ongoing aftershocks made insurance on any new building project unobtainable. But even on those margins where CCC could have helped, they seemed instead more concerned with upholding the existing pre-earthquake rules and protecting amenities for homeowners in the wealthier suburbs.
The worst exemplar here was Council’s steadfast refusal to allow homeowners to build secondary rental flats within their existing homes or on their properties. While insurance on a new building was unobtainable, owners of existing homes with ongoing insurance policies could maintain their existing insurance while undertaking either earthquake repairs or home renovations. This would have provided an opportunity for owners to build self-contained flats. While it is unlikely that thousands of such units would have been built, even a few hundred would have been helpful where people otherwise lived in uninsulated garages, sheds, caravans and broken homes over a couple of winters. These aren’t just apocryphal or inventions of Campbell Live: people were living in our neighbour’s garden shed for rather some time after the earthquakes.
The usual objections to densification manifestly failed to apply in the case of secondary suites. Unlike an apartment block, they provide no concentrated burden that must be accommodated by trunk infrastructure like water, sewerage or roading. They impose no shading or other real, substantive, demonstrable impediment on neighbours. But, the only cases in which they were allowed were for family units where the owner could guarantee that a family member would live in the flat. Under later CERA regulations, secondary flats were permitted where the owner could guarantee, prior to construction, that the dwelling would be used by someone displaced by the earthquakes rather than by an incoming construction worker; the flat also needed to be removed by no later than 2016.
The only sense I can make of Council’s intransigence on secondary flats is a combination of bureaucratic inertia and fear that homeowners near the University of Canterbury would object if their neighbours used the regulatory provision to accommodate students. It seems a pretty thin basis on which to block what could have been a helpful and expeditious way of getting new affordable accommodation onto the market after the earthquakes.
The regulations that were, pre-quake, relatively innocuous, became highly constraining after the earthquakes. These were rules that had nothing to do with building safety or standards but rather aesthetic considerations around minimum lot sizes, mandatory parking minimums, maximum allowed density, and the pace at which new greenfield properties could be developed. While these may have arguable benefits in normal times, surely after a destructive earthquake the balance should have been tipped in favour of increasing housing supply. Council utterly failed to relax the pre-quake zoning rules or to quickly release land to enable new construction.
And this was a chronicle of a housing failure long foretold. Shortly after the September earthquakes, developer Hugh Pavletich argued for the release of more land on the more stable western city fringes. Nothing happened. And, amazingly, neither did anything happen after the February earthquakes.
The failure partially foretold: The consequences of planning rigidities, new building codes and insurance
Council failed to move with sufficient alacrity to allow new housing to come to market. Sorting out the regulatory mess blocking the construction of new dwellings was not going to be easy. Council’s consenting office was overwhelmed, though they could potentially have coped under a streamlined and simplified set of rules. Heritage preservation regulations worked at cross-purposes to earthquake-strengthening regulations after the September 2010 earthquakes. Density regulations and restrictions on ‘granny flats’ prevented densification on brownfield sites where the land was sound on the west side of town. Council changed the building regulations after the February insured event, but before insurance-funded rebuilding, thus guaranteeing legal uncertainty as to the extent of insurers’ liability where meeting the new code would constitute betterment but where building to less than the new standard was forbidden. Finally, the slow trickle of land released for new development had important and easily foreseeable implications, though ones that apparently were not foreseen by the planners.
It is worth briefly walking through how the slow release of land at the edges of town can have pervasive negative effects regardless of whether the released land would be sufficient for some number of years’ supply. When only a small amount of land can be released for development every year on a known path, it is relatively easy for developers to buy up the newly zoned land and to release it even more slowly, keeping land prices very high. When land prices are very high, and where those prices are high due to zoning rather than due to the inherent nature of the land, developers face particular incentives to provide larger and more expensive homes on zoned sections: why put a $100,000 house on a $300,000 section? The top-end of the market, with higher margins for developers, gets served first. That part of the market puts fairly high value on restrictive building covenants preventing their neighbours from putting up houses that might reduce their own property values.
Consequently, much of the new development on the edges of Christchurch provided higher priced homes bound up by covenants. Homeowners in the residential red zone, then, were largely precluded from moving their homes onto new land in new developments. While it’s easy to blame developers for those covenants’ restrictions, they’re fundamentally a consequence of a severely limited supply of zoned land. In the absence of those restrictions, a farmer on one of the thousands of hectares adjoining Christchurch could have turned a paddock into a subdivision for houses from the red zone. Instead, Christchurch was exporting red zone houses in 2012 to places as far away as Gore.[i] In the midst of a terrible housing shortage, we were sending houses away because our regulations made it too hard to let people live in them.
It consequently has been extraordinarily frustrating to hear the Christchurch rebuild described as exemplar of the failures of leaving things to the market. Really, we have been in the worst of all possible worlds here. A well-run government rebuild would have been better than what we’ve had. An unhampered market approach led by developers and property owners would have been far better than what we’ve had. Instead, we’ve had developers and property owners trying to provide new housing as and where they could under the somewhat important government constraint that building wasn’t really allowed.
Blame cannot lie solely with CCC though. Shortly after the September 2010 earthquakes, the national government, in an all-party consensus, passed the Canterbury Earthquake Response and Recovery Bill enabling central government to do, effectively, anything it wanted in Christchurch. CERA was established a month after the February earthquakes to coordinate the recovery. Such an agency could have been helpful in finding the problems in regulation, or in insurance markets, that were stymying the rebuild. It manifestly failed to do so.
When CERA was established, I was hopeful. In June 2011, CERA CEO Roger Sutton said, ‘I have quite extraordinary powers to actually bypass these planning laws, but my preference would be for the normal legal process to work.’[ii] He noted the lack of lower priced sections on Christchurch’s fringes and the potential for cutting the costs involved with planning and resource consents.
But nothing happened. Where CERA was supposed to cut through the regulatory morass so that the rebuild could happen, they seemed oblivious to the severity of the regulatory problem. Roger Sutton’s appearance on TVNZ’s CloseUp in May 2012 provided ample demonstration of CERA’s problems. The Christchurch Press, and the #eqnz Twitter hashtag, had been filled with stories of the problems involved in trying to get red zone houses relocated to new developments. The Haywoods in particular were very active in documenting and publicising through social media the exact regulatory difficulties they had been encountering in trying to move their house from the residential red zone. But when Sutton appeared on CloseUp and was presented with the story, he said ‘The first I heard of this and the difficulties was today’. The agency that was supposed to have sorted a way through the regulations to allow the rebuild simply didn’t know what was going on. How could we hope that they might fix the regulatory mess if they were seemingly clueless as to its effects?
I attended one meeting of CERA’s Canterbury Economic Indicators External Review Panel in 2013 at which many of us asked very pointed questions about what was being planned to allow for the accommodation of an expected inflow of 15,000 construction workers in 2014, with nobody in CERA, and none of the representatives of the other agencies, knowing where they might possibly live given the constraints against housing construction. I was left with the impression of an agency that wanted to do good but that really didn’t know what it could do for fear of judicial review, despite its broad enabling legislation. What a waste.
Instead of moving nimbly to shed the most restrictive regulations and consequently to allow rapid densification on good land and expansion out into the suburbs, CCC’s planners instead hunkered down and stuck with what they knew best: an overly zealous approach to regulatory compliance, an obsession with masterplanning that makes the best the enemy of the good, and a refusal to consider that maybe, just maybe, getting people out of garages and sheds in the east might be worth regulatory changes that might upset people in Gerry Brownlee’s constituency. The fix in housing would have been relatively simple. In any future earthquake event, we should have a regulatory switch that simply flips automatically enacting the following:
- A four-year window in which all density restrictions are removed. So long as a building meets building code, it can go up. We do not need extensive planning and handwringing over the essential characteristics of particular neighbourhoods and whether they’re consistent with intensification: people stuck living in uninsulated garages count for more than that. Let developers and insurers decide which bits of land can stand taller buildings and let them go up. If Council moves to return to the ex ante land use restrictions after that window closes, any buildings already consented during that window are grandfathered in.
- A similar window in which all green belt or Metropolitan Urban Limit restrictions are removed, barring those that exist to avoid substantial and demonstrable environmental harm. The window here can be shorter because greenfield development is faster than brownfield.
- All restrictions against building secondary units within a dwelling or on-site at existing properties are removed.
Had Christchurch taken this approach, a lot of houses would quickly have gone into construction to the south-west of Christchurch while other subdivisions would have opened up where red zoned houses could have been placed. During the early period, homeowners would have added flats within existing homes or granny flats on existing properties to let out to accommodate the spike in demand caused by the combination of incoming construction workers, displaced families, and ongoing student accommodation demand. We would not have had families living in uninsulated sheds for two or three winters. All government needed to do was to get out of the way.
Downtown confusopoly: The unexpected failure
Commercial redevelopment in Christchurch has been no less shambolic. In the immediate post-quake period, business owners, even those with their own privately hired search and rescue technicians, were not allowed to access their companies’ files and records within the downtown cordon. But if you were a young bride needing to get her wedding gown out of a cordoned dressmaker’s shop, you could get through. While this was only one example, it demonstrates the arbitrariness of the cordon’s restrictions.
After the initial crisis phase, we found the commercial confusopoly. Because CCC changed the building code after the insured event, restoring a building to ‘as new’ status (the terms of at least some insurance contracts) was insufficient; upgrading it to the new code would constitute a betterment. The government should have sought a declaratory judgement over a few standard insurance contracts to resolve uncertainty and allow construction to proceed, and similar declaratory judgments over insurance cases where the insurer wished to rebuild on-site but where the government deemed the land unfit; uncertainty instead prevailed for years.
The bigger problem, though, was the regime uncertainty brought about by the government’s refusal to commit to a central city plan. Economists use the term ‘regime uncertainty’ to describe a state of affairs in which nobody really knows what the rules are or what they will be over the next few years. In the first six months after the February earthquake, downtown property owners really could not do much while they waited for Council to decide on its central city plan. By April 2012, the Government had thrown out the proposed CCC plan and established the Christchurch Central Development Unit to come up with a new city plan. The eventual plan that was released was long on visions of precincts, but a bit short on respect for the property rights of existing owners.
Outside of the central city and away from Brownlee’s thumb, business owners were simply getting on with things. Cassels and Sons opened a new brewpub in Woolston only one week after the July 2011 earthquake, then expanded to a full new retail development. But, downtown, nobody could tell you whether your proposed development was consistent with the grand plan. You’d have to wait to find out. Would there be a new convention centre? If there would be, and you’d owned a hotel, you’d want to rebuild your hotel near it. If there weren’t, then you needed to make other plans.
And, if your hotel happened to be in the newly designated Performing Arts Precinct, whether you’d be allowed to rebuild on your current site would depend on some yet-to-be-made decision as to whether hotels were consistent land use within an arts precinct.[iii] At Day 757 after the February earthquakes, the owners of the Copthorne Hotel simply did not know whether they were allowed to rebuild, despite an urgent shortage of hotel spaces in the city. Their insurer had settled and the hotel was keen to rebuild. But they risked expropriation if they did, because nobody yet knew whether it would be decided that hotels weren’t meant for arts precincts. As CCDU official Greg Wilson said in the Press, ‘The test is whether the proposed use would prevent or hinder the public work – in this case the development of the performing arts precinct’.[iv] The Copthorne couldn’t do anything until Earthquake Recovery Minister Gerry Brownlee provided consent, and his office was not known for expeditious decisions about anything. In SimCity, you can pause while you figure out precincts. Christchurch’s pause button was rather more costly.
I had never expected that a purportedly market-oriented National Party government would preside over a dirigiste take-over of city planning. Rather than forcing CCC to get on with things, they instead put their own planners in place for the downtown, with exactly the same predilection for making the best the enemy of the good-enough. As I write this, in June 2014, we still do not know whether the National Party government will force an expensive stadium on Christchurch, who will run a new convention centre, or what will be happening with rather too many of the Government’s anchor projects. On some of these, simply getting any decision two years ago would have been better than the dithering. Businesses can at least get on with the rebuild within a less-than-ideal plan. It’s harder to do that under continued regime uncertainty.
Some of the goals of the CCDU seemed laudable. They wanted a vibrant, sustainable downtown of more compact form than that which we had prior to the earthquakes. But they made an utter hash of the job. They established a Green Frame within downtown to reduce the area of land potentially available for downtown development, claiming it a virtue that land prices would thereby stay high. But in a functioning market, land prices are imputed from potential rent. Tenants willing to pay prices consistent with those valuations were few. Consequently, many moved quickly to the suburbs. If, instead of pursuing the grand precinct visions, the CCDU had simply let existing property owners make what best use of their land that they could, we would have had less flight to the suburbs.
Again, government would have done better by simply getting out of the way. But there were important and constructive things that the government could have facilitated for downtown redevelopment. Insurance test cases would have been extraordinarily helpful.
Much of downtown was under pretty fragmented ownership in small lots, and it is entirely plausible that redevelopment would have been better pursued with more concentrated ownership. But rather than look either to compulsory purchase or to mandatory joint-ventures in the downtown retail area, they could simply have maintained a database of ready contact details for existing owners and encouraged the use of dominant assurance contracts for land assembly. Or, they could have moved quickly to set up the anchor projects, committed to the locations and to the funding, and simply then let precincts emerge from the distributed decisions of Christchurch’s downtown property owners given certainty around the public projects.
If we learn anything from the intersection of the work of Jane Jacobs and of Ed Glaeser, it’s that cities are organic. The best parts of cities emerge from the distributed decisions of thousands of property owners, building near each other to take advantage of complementarities in location that they could foresee and that the planners couldn’t envision. SimCity takes no account of the wishes and dreams of the Sims. All of the small actions of distributed individuals can add up to something wonderful, if only Council and the bureaucrats would get out of the way and let it happen. Instead, we had the worst of all possible worlds: the insistence that a perfect central plan supercede these decentralised decisions, but absolutely no bureaucratic capacity to set or follow through with a plan.
It has taken me far longer than this book’s editors would have liked to write this chapter. I’m an economist who works best when considering issues dispassionately. I cannot maintain any reasonable mental state when reflecting on what the planners, both from CCC and those imposed on us from elsewhere, have done to Christchurch. I have had to keep looking away from this Dementor’s gaze. But if we keep looking away, worse will happen to Wellington when its earthquake comes. There are substantial regulatory problems that need addressing ahead of any future earthquakes. Let’s not have another tragedy well-foreseen. We should know better by now.
From 2003 until July 2014, Dr Eric Crampton served as Lecturer and Senior Lecturer in Economics at the University of Canterbury, where he lectured on economic policy, including the economics of the city, and the economics of political decision-making. In July 2014, he left the University to serve as Head of Research with the New Zealand Initiative in Wellington. He blogs at OffsettingBehaviour.blogspot.com.
This is an extract from the new book Once in a Lifetime: City-building after Disaster in Christchurch. There are two launch events in Auckland for the book today:
Auckland lunchtime panel discussion
Time: 12pm 17 September
Location: The Exhibition Studio, Level 3 , School of Architecture and Planning, University of Auckland
Media commentator Russell Brown will lead a discussion with the book’s editors Dr. Ryan Reynolds and Barnaby Bennett.
Auckland launch
Time: 6pm 17 September
Location: Q Theatre on Queens Street.
Join NZ Herald Business editor Liam Dann and the various contributors and editors to launch the book over a glass of wine.
Wellington lunchtime panel discussion
Time: 12.30pm 18th September
Location: LT1, School of Architecture and Design, 139 Vivian Street.
Join economist Eric Crampton, writer Giovanni Tiso, architect Chris Moller and public health housing specialist Graciela Rivera-Munoz in discussion with co-editor Barnaby Bennett as they consider the issues and lessons learned from the Christchurch recovery process, and reflect on why these need to be heeded in the case of any Wellington seismic events.