Covid 19 - Unpacking the City

 The compact city – a crumbling consensus

The policy consensus promoting compact, centralised cities cannot be sustained in a world ravaged by Covid-19. The current pandemic is accelerating the move to dispersed urbanism. Beyond the direct impact of disrupted trade, travel, and consumption on the economic foundations of cities lies the cascading impact of changing work behaviours. This post considers what remote working might mean for urban development.

Calculating the number of jobs that can be decanted

My previous post summarised the McKinsey Global Institute (MGI) analysis of sector-specific potential for remote working, applying the resulting metrics to New Zealand.  Here, in line with other developed nations, around 30% of current jobs offer remote working potential.  

This post considers the possible impacts on urban function and form by applying the MGI approach to New Zealand’s cities and districts.

Urbanisation and Remote Working

The share of tasks that can be undertaken remotely has been calculated across 19 sectors for 67 territorial local authorities (TLA).  The results have been aggregated into: (1) the three largest cities (Auckland, Christchurch, Wellington); (2) ten provincial cities with populations between 50,000 and 175,000; (3) partly urbanised districts characterised by smaller towns and townships; and (4) mainly rural districts encompassing rural areas and small settlements (Figure 1). 

Figure 1: Urban Dimensions of Remote Working Potential, New Zealand

 

The potential for remote working appears to be a matter of scale: the large urban areas offer the greatest opportunities. 36% of Wellington’s jobs could be done remotely, 32% in Auckland and 29% in Christchurch. The potential is lower among small provincial cities (28%), partly-urbanised districts (25%), and rural areas (22%).

It appears that more urbanised areas have greater potential to substitute remote work for fixed-workplace employment.  This is confirmed when we plot potential for remote working against urbanisation across all TLAs (R²=0.47, Figure 2). 

Figure 2: Urbanisation and Remote Working Potential, New Zealand Council Areas

.. 

Examination of the variation around this relationship between urbanisation and remote working potential indicates the role of differences in local employment structure.  More production-based jobs offer lower potential for remote working, while more business and service jobs lift the potential.

Where within the city?

This section considers variation in the potential for remote working within Auckland (New Zealand’s dominant city with 1.7+m residents). Figures for 20 Local Board areas based on the MGI sector coefficients have been aggregated and organised in Figure 3 from north to south (left to right on the axis).  Reflecting the city’s linear geography, areas at each end are most distant from the CBD.  The CBD and its fringe comprise the “Central” area, sitting within the Isthmus, which contains the city’s older, inner suburbs.

Based on this example, remote working potential varies more within the city than among cities, ranging between 41% in the centre (47% in the CBD fringe) to 10% in the upper north and west and 2% in the rural south. The former reflects the high value, administrative, business, and professional jobs in the inner city and suburbs, and the latter the greater share of manufacturing and personal service (face-to-face) jobs in the suburbs and primary production on the fringe.

Figure 3 The Prospects for Remote Working within Auckland

 

Push and Pull Drivers

Much of this theoretical capacity is likely to be taken up.  On the push side, the risk of exposure to infectious diseases is reduced by limiting exposure to places where people congregate for work, education, and entertainment. One benefit is limiting the spread of other infectious illnesses, a personal and productivity bonus. Remote meetings offer another productivity gain, lowering travel costs and focusing information exchange, supervision, and negotiation.

On the pull side, remote working has been positive for many people and businesses, with reports that the practice is being adopted on an ongoing basis in New Zealand despite limited community transmission of Covid and internationally.

What about the downsides? 

A suitable work (or school) space within a dwelling is needed to maintain both productivity and satisfaction from remote working at the individual level. With housing affordability constraints impacting on younger people and families, in particular, their capacity to work effectively from home will be constrained.

Lack of face-to-face contact with colleagues limits the benefits of work based social interaction. In a Covid-free environment, however, one option is to mix remote working with workplace attendance one, two, or three days a week.

Impact at the centre

However it evolves, the urban impacts of remote working will be far-reaching. As employment in the central city stutters, policy makers and investors will have to rethink the principles of workplace location, investment, and development.

Hospitality, personal services, and discretionary retailing in the city centre will suffer from reduced commuter and work-related spending. They will also suffer from any changes in the attraction of the city centre for housing. Modest apartments in multi-storied buildings marketed for a city lifestyle will lose appeal, becoming a welfare, last resort, or first housing step rather than lifestyle choice.

Add the deskilling of high order services, displacement of predictable or repetitive transactional tasks by AI, and the prospect that the international travel industry shifts away from mass tourism, and the outlook for city centres as we know them dims.

Against this, the resurgence of suburban centres, provincial cities, satellite towns, and country life may build on the intrinsic appeal of living locally as more people decamp from intensively urbanised areas, provided, perhaps, that the policy-makers do not seek to impose the densities of inner cities on under-resourced suburbs and cling on to the notion of commuting-based city centres.

And Infrastructure?

As it stands, the compact city comes at considerable cost. The demands on ageing infrastructure from intensification were never anticipated by the city builders.  Maintaining or rebuilding energy and water supplies, the capacity and reliability of wastewater systems, boosting transport networks and retrofitting ageing transit systems all demand substantial expenditures, mortgaged in large part against expectations of growth that must now be in doubt. 

If nothing else, the impetus Covid has given to dispersal must shift attention to infrastructure challenges in existing suburbs.  While highlighting the hard questions in a policy environment beset by an unnatural aversion to greenfield development (in which the potential for sustainable settlement has suddenly become compelling); it also raises challenges for small, erstwhile sleepy settlements unprepared for the demands of a growing flow of ex-urbanites. 

Restricting retailing to save the CBD

Resisting decentralisation
In my last posting I argued that planning for a hierarchy of retail centres simply protects the interests of a few landlords, investors, and operators.  Regulating to retain the status quo of centres stifles retail investment and productivity without necessarily improving the city and town centres it aims to support.

In this posting I look at what’s happened to Auckland’s retailing over the past decade.  Here the argument for preserving an established retail hierarchy has prevailed in plans that rely on directing growth to existing centres as one way of promoting a more compact city.

Despite this, it seems the forces of decentralisation have prevailed. 

The data
I have looked at 15 retail and leisure sectors using Statistics New Zealand employment figures for 2000 and 2010.  I have grouped Census Area Units into eleven categories to explore changes in location.  This involves some imprecision in area definitions but the figures still tell a pretty compelling story. 

The big picture

First, look at changes in retail and leisure employment in different parts of the region (Figure 1).

It was up 11% in the CBD; but in “out of zone” industrial areas it was up 46%.  Rural areas were up 59%, although we are dealing with much smaller numbers here. 

The biggest absolute gains were in suburban centres (up 39%, insofar as as they could be isolated through Census Area Units).  The middle and outer suburbs fared pretty well, too (up 22% and 26% respectively).  Major centres lagged (4%).

Figure 1: Sub-regional Employment in Retail and Leisure Activity, 2000 and 2010

The significant changes taking place in the distribution of retail and leisure activity raise some interesting questions.  

How far can retailing – in which employment actually fell by 7% in the three years to 2010 – be expected to prop up the CBD?   Does planning actually undermine retailing by constraining investment in favour of the CBD and centres?  Or has it simply been ineffective, with retailing and leisure expanding where they can achieve superior performance regardless of councils’ objectives, favouring more dispersed sites?

Changes in retail mix
When we look at the changing retail mix we begin to see why the CBD and major centres fared poorly over the decade.  There have been big differences in growth among subsectors (Figure 2), suggesting that the shops we plan for tomorrow will not be the same sorts of shops (or leisure activities) planned for today.

Cafes, restaurants, and take-aways have maintained their dominance.  They were the biggest sector in 2000 (28% of total employment in retail and leisure) and grew 32% over the decade.  Hence, eating out accounted for nearly 40% of all growth.  Did our centres-based planning anticipate that?

And look at clothing and footwear shops.  With just 7% of leisure/retail employment in 2000 they generated 17% of all growth.  Electrical and electronic retailing also stood out, along with pharmaceuticals and personal goods.  Presumably the success of these groups influenced the decline in department stores, a category associated with major centres.

The creative and performing arts also declined, raising even more questions if we believe that these industries, usually associated with the CBD, are essential to innovation and growth.

Figure 2: Employment Changes - Retail and Leisure Sectors in Auckland, 2000-2010

The redistribution of retailing
To examine the relationship between retail mix and growth I divided the city into: the CBD and inner city, the suburbs, all centres, and industrial areas (including the small rural areas for convenience).

The bars in the Figure 3 show the shares of city growth (or decline) in each retail sector by area.  

Shares of supermarket and grocery store employment increased in industrial areas, and fell everywhere else.  This does not mean employment fell elsewhere, but the weight of growth was greatest “out-of-zone”.  Furniture and clothing and footwear outlets also did well in non-retail areas.

Industrial and suburban areas recorded small gains in department store employment, which fell everywhere else.  Industrial areas and centres reported gains in stores selling recreational goods, which also declined elsewhere.

Only specialty food favoured the inner city (and suburbs) although this is not a big sector. 

Figure 3: Changes in Subregional Shares of Employment by Retail Category

2000-2010

And the leisure sectors?
The central city is seen as increasingly important for leisure.  But this is not necessarily supported by the employment evidence.  The big growth sector, cafes, restaurants, and take-aways, performed better elsewhere.  It appears that the latté set has spread well beyond the CBD and that dining out no longer automatically means tripping into inner city restaurants.

Not only that, employment in the creative and performing arts declined in the city centre faster than elsewhere, belying another popular preconception about its role.

There’s more: inner city sport and recreation underperformed compared with the rest of the region despite the boom in gymnasiums.  The same goes for pubs and taverns.  Overall, industrial areas and centres seem to have been the growth leaders in leisure.

Figure 3: Changes in Subregional Shares of Leisure Sector Employment. 2000-2010

Let’s not throttle the patient
We could go into more detail.  But the message is clear: in our planning we need to allow for market realities, and the variety of places investors and consumers favour for retail and leisure.  Planning may be more effective if it responds to community preferences rather than aiming to impose order by arbitrarily excluding some activities from some places to prop up other activities in other places. 

So, what of the CBD?
This evidence suggests that we should abandon simplistic models of hierarchies for planning either for retailing or for urban form.  If planning is reduced simply to resisting change or enforcing hierarchies, we impede productivity based growth and counter the vitality essential to the city’s and the city centre’s progress.  

But there’s no need to panic about the future of the CBD.  We just need to recognise that cementing past functions in place is not the way to revitalise it (or our town centres).  We may also have to acknowledge that the CBD’s lagging performance in some areas may reflect success in others, in this case through displacement of commercial activity or limitation of new investment in retail and leisure by the expansion of inner city housing. 

More generally, the view of what drives a successful city centre has to be far less mechanistic than appears to have been the case in planning till now.  The city centre is a living, evolving place.  Ten years ago financial services were the key to its growth.  Before that retailing played an important role.  Today education and tourism have centre stage.  Tomorrow it may be communications and technology.  Whatever the future holds, it is out of a culture of change and diversity that the CBD will grow, not as a result of imposing outmoded views of what should go where.

Auckland – not the only town in the game

Do people prefer small cities?

Leading US urbanist, Witold Ribczynski, in his recent book Makeshift Metropolis (2010), places the past growth of major metropolitan areas in context:

“the majority of us still appear to prefer dispersed, small cities, a significant minority want to live in concentrated big cities, and a tiny fraction is prepared to pay the price of living in the very centre of things” (page 179).

Despite signs of dispersal, he still sees increasing urban densities as the way forward for America “while taking into account people’s desire for dispersed living in smaller cities”. 

One implication is that advanced urbanisation does not depend on centralisation or the ever-increasing expansion (upward or outward) of the largest urban areas.  Equally, decentralisation does not necessarily mean sprawl.  It may even be that the more compact settlement planners have been pushing over the past two decades is best achieved through dispersal – decentralised intensification – because that lines up with what more people want.

And, if more people prefer small cities, we need to be asking questions about what that means for public spending and economic growth, as well as for the quality of life.

Do small cities feature in New Zealand’s growth?

I thought it would be interesting to see if New Zealand’s changing settlement pattern reflected a similar preference for smaller cities. 

Of course, New Zealand is not North America.  It is distinctively dependent primary production, something which has traditionally supported a wide level of dispersal of a small total population.  At the same time Auckland in the northern North Island  has evolved into a strong primate city, something I have touched on elsewhere.  A productive hinterland, nodal advantages, import substitution, the cumulative concentration of services and consumption, and immigration have all contributed to this.  Today, Auckland accounts for a third of the nation’s population, something that guarantees that it will continue to dominate growth.  

I have looked at recent population change to see if there is any stirring among New Zealand's smaller cities, though. 

The data I used

I have used Statistics New Zealand’s population estimates for territorial local authority areas (TLAs) from 1996 to 2010 for this.  I divided Auckland into the inner city area, the balance of the built up metropolitan area, and two predominantly rural areas or ex-urban areas.  These are based on TLAs in existence (Rodney and Franklin) before the 2010 reforms rolled them up into the city's jurisdiction. 

I then grouped into several categories: the second tier cities (Wellington and Christchurch); ex-urban areas (TLAs immediately adjacent to metropolitan Auckland, Wellington or Christchurch); those TLAs dominated by provincial cities, ranging from Marlborough/Blenheim (TLA population 39,000, urban population 30,000) through to Hamilton City (143,000 people); those TLAs with mixed rural and urban (town) populations; and the predominantly rural TLAs. The inner city population figures for Wellington and Christchurch were added to those of Auckland (and deducted from the balance of the city population in each case).

Auckland remains dominant

In 1996 metropolitan Auckland accounted for 27% of the population, but over the next 15 years it accounted for 52% of the country’s population growth.  There’s not a lot of obvious support for the small city hypothesis in those numbers. 

Of course, Auckland at around 1.4m (inner, metropolitan, and ex-urban areas rolled together) is not an especially big city by North American standards.

But something is happening.  This shows if we compare the first five years of our data with the last four years.

Between the 1996 and 2001 censuses the three inner cities, with less than 1% of the population jointly, exploded (at least in relative terms) to account for 7% of growth recorded between 1996 and 2001 (see chart, below). 

And, Auckland’s metropolitan area, with just 27% of the population, recorded almost two thirds of the nation’s growth.  Exurban areas also grew their share of population (with 20% of the growth).  The second tier cities (once we take out inner city growth) and rural areas declined.

Shares of Population (1996) and Population Growth (1996-2001)

But centralisation is slowing down

Between 2006 and 2010, however, this pattern moderated.

Metropolitan Auckland still dominated but with 29% of the national population in 2006, its share of growth slipped back to 47%.  Inner city areas dropped their share of growth, too, back to 4%.  Exurban areas dropped back to 13% (chart, below).  

So where did the balance go?  Well, between 2006 and 2010 provincial cities accounted for 22% of growth compared with 15% between 1996 and 2001.  Mixed rural/township areas were also up.  They accounted for 12% of national growth, up from just 3.5%.  Rural areas were back in the black. 

Shares of Population (2006) and Population Growth (2006-2010)

This is by no means an end to Auckland’s dominance.  But centralisation is no longer a one way street.  New Zealand, like the United States, is experiencing an increasingly complex pattern of population growth reflecting, presumably, greater variety in residential choices.

A time for rethinking?

We’ve raised this earlier, but its time to take a more balanced view of New Zealand’s growth, one that is less Auckland-centric.  Sure, with around a third of the population and employment, it is important that Auckland operates efficiently, that we invest wisely in the city’s infrastructure, and that it provides a great lifestyle to its citizens and continues to attract (and hold) skilled migrants. 

We can also acknowledge, like Rybczynski, the desirability of inner city living for some citizens.  This is reflected in the growth rates of inner Auckland and Wellington in particular.  But, like him (page 178) we also have to acknowledge what a small minority that is.  Using our generous definition of the inner city,[1] less than 60,000 people lived collectively in inner Auckland, Wellington, and Christchurch in 2010, or 2.6% of their joint population. 

Let's not stop thinking about Auckland and its needs, but let’s also invest some thinking and planning in how to realise the opportunities that arise out of more dispersed settlement. 

A national view
When we look at the annual compound growth rates of the top 15 growth areas in New Zealand since 2006 the movements that stand out for me are: the revival of rural and small town South Island; the increasing pervasiveness of growth across the wider northern North Island; and the significance of ex-urban growth around our major cities. 

Fastest Growth Areas in New Zealand, 2006-2010

Metropolitan and inner city growth are up there, but Auckland is no longer the only game in town.

In any case, a preoccupation with fixing Auckland must now make space for the imperative of recovering and rebuilding Christchurch.  The numbers used in this blog predate the September 2010 and February 2011 earthquakes there, and right now we do not know what long-term effects they will have on housing preferences and settlement patterns.  But the analysis here suggests that decentralisation to smaller, nearby centres may be an attractive option to a significant share of Christchurch citizens, and perhaps a reasonably affordable one from a national point of view.

In a mature urban society – and especially one facing an uncertain economic outlook – we may need to think about a future in which we don’t bank too much on the prospects and opportunities associated with large-scale urbanisation.  That, in the New Zealand case, would be too much like putting all our eggs into one basket.  The time may be upon us when we have to seriously consider  the role of smaller cities and towns play in our economic growth, if only to support the fact that more people are choosing to live in them.

The prospects for New Zealand and the prosperity of New Zealanders demand that we think long and hard about how we are going to exploit the advantages and attractiveness of our smaller cities.

[1]               For those interested, the following Census Area units: Auckland – Harbourside, Central West, Central east, Newton, Freemans Bay, Grafton East and West; Wellington – Thorndon-Tinakori, Lambton, Willis-Cambridge Tce, Aro-Nairn Streets; Christchurch – Cathedral Square, Hagley Park, Avon Loop

---

In what sense a region? Drawing a line around economic development

Unleashing the spatial plan

Auckland’s Spatial Plan is intended to help shape the city/region’s future development.  The council has just-released a Discussion Document to engage the community in its preparation. 

It’s an ambitious document that goes well beyond discussing just land use and infrastructure.  The Auckland Plan is proposed as a blueprint to transform the region into “the world’s most liveable city” and “the economic powerhouse of the nation”.

This focus raises questions of just where the city’s economic borders begin and end.  In what sense is Auckland City with its sprawling borders a discrete economic unit?

Inclusiveness – where do the neighbours fit in?

Among many other things, the discussion document acknowledges that “Auckland is also part of an emerging northern North Island urban and economic system – a cluster of cities and towns north of Taupo (Hamilton, Tauranga and Whangarei)”. 

This, it says, calls for “a strong planning and policy relationship” with adjoining regions and ”an inter-regional agreement for the upper North Island ... to cover land use, planning, infrastructure and a range of other issues”(p21) and that it is “essential to develop inter-regional transport links to support the burgeoning Northland, Auckland, Waikato and Bay of Plenty urban and economic systems” (p48),.  It also notes Auckland’s dependence for some services and resources on the Waikato region.

But that’s about all it says about these key relationships in its 220 pages.  

At a seminar of the Institute of Public Administration this week, Hamilton Mayor Julie Hardaker said that this was not enough.  She pointed out the economic significance of her region, the Waikato, as the heart of New Zealand’s dominant dairy export sector for example.

Mayor Hardaker also reminded us that what happens to Auckland’s land use influences what happens in other cities in the Northern North Island. 

Hamilton, for example, has had rapid investment in light industry, warehousing, and distribution around its northern gateway over the past 10 to 15 years.  This is a cost effective location for accessing a wide market, including Auckland.  The resulting investment in Hamilton has been boosted by a shortage of reasonably priced industrial land in Auckland. 

In a similar vein, Tauranga and Whangarei both receive significant population contributions by way of migration from Auckland (as detailed in Margins Matter, p18).

Auckland as a regional centre

The notion of inter-regional integration is not new.  Auckland’s original growth was driven by the output of its resource-rich hinterland.  It also came to enjoy prosperity as an import substitution centre through much of the last century. That role has almost disappeared today. 

More recently Auckland has prospered as a centre of consumption. Nevertheless, the city continues to depend for much of its income on that hinterland, and on providing services to trade and business in the northern North Island.

Much of the value added in the processing of primary output from this “region of regions”, especially food products, happens in Auckland.  The city is also the hub of many financial, business, and technical activities that service the resource-based industries of the northern North Island. 

Even the much vaunted “marine industries” value added cluster associated with Auckland reaches deep into the neighbouring regions with their skills and marine infrastructure. 

Four into one might just go

Equally important is the cumulative economic capacity, infrastructure, and resource base of this extended region, a region which we have argued for some time may be the best level at which to plan for regional economic development in a globalising world (see, for example, our report to the Metropolitan Auckland Project in 2006). 

Taken as a whole, the northern North Island has substantial fertile grasslands, horticulture and arable land, forestry, and associated processing industries.  It has a wide range of mineral deposits.  It has good energy resources with high levels of current and potential renewable energy.  Its marine resources are extensive.  It has specialist concentrations in research, scientific expertise, and education that when considered jointly are of international substance and offer the potential to substantially lift the value of New Zealand’s exports and expertise.

The fortunes of the ports of the northern North Island -- Auckland, Tauranga, and Whangarei -- are connected by ownership links and by a degree of complementarity in their respective trades.  As internal transport costs increase and as shipping companies rationalise international movements the regions’ (and country’s) interests might be best served by more collaboration among them.

And this might be the catch cry for relationships among the councils of the region: more collaboration.  Because if we really believe the rhetoric about globalisation and world cities, New Zealand will only succeed if it digs deep into what are ultimately finite physical, human, and capital resources.  Administrative boundaries should not define or constrain the definition of a region within which we identify how public policy and investment might best support economic development. 

Decentralisation is happening

Sure, Auckland remains dominant in the area, as it is in New Zealand, and it will remain so.  It accounted for around two thirds of its jobs in 2010, and 60% of job growth over the past decade. 

However, the 6% difference between these two numbers hints at a degree of decentralisation.  This is confirmed in relative gains (or losses) in employment between 2000 and 2010. The three other regions had better growth rates than Auckland.  The best performing urban areas within the northern North Island were Tauranga, Hamilton, and the urbanised area of South Auckland (Manukau and Papakura).

	All Employment		Manufacturing
	Number	%	Number	%
Auckland Region	92,590	18%	-10,500	-13%
Auckland City	30,350	11%	-8,160	-24%
Manukau City & Papakura District	31,380	29%	180	1%
Waitakere & North Shore Cities	23,330	22%	-2,230	-13%
Rest of Auckland	7530	25%	-290	-6%
Northland	10,170	24%	170	3%
Whangarei	5,910	25%	210	0
Rest of Northland	4,260	22%	-40	0
Waikato	29,130	22%	100	1%
Hamilton	16,560	29%	-100	-1%
Rest of Waikato	12,570	17%	200	2%
Bay of Plenty	5,480	22%	-800	-6%
Tauranga	12,970	37%	140	3%
Rotorua	2,110	8%	-420	-12%
Rest of Bay of Plenty	5,480	22%	-520	-12%
Northern North Island	152,450	20%	-11,030	-9%
Northern North Island Excluding Auckland	59,860	23%	-530	-1%

Source: Business Statistics, Statistics New Zealand

And as manufacturing slipped into the negative under the pressure of cheap imports and then global recession, Auckland led the way down.  The figures in the table include the negative impact of the post 2007 recession: until 2007 manufacturing was growing everywhere but Auckland City, led by the Waikato with 16% growth as Auckland Region stopped growing.

Working together

If it wants to tackle economic matters, Auckland can ill-afford an inward focus.  If it wants to micro-manage land use, it may also pay to look hard at the opportunities that the wider region offers.  And if liveablity is how we define Auckland’s place in the world, it cannot afford to exclude from the portfolio the qualities of the smaller centres, their rural settings, mountains, and coastal hinterlands.

By avoiding being locked into an administrative construct with only limited relationship with what drives Auckland’s growth the Mayor can play a leadership role in economic development that can make a difference.  But it’s time to lead the city beyond “better because we are bigger” and reflect on what it really means to be connected in today’s world, starting at home. 

*************************************

Dushko's banana - here is the image rferred to in the comments below: