Sunday, February 24, 2019

Deconstructing the Brisbane CBD


What’s it going to take to reignite the demand for Brisbane CBD offices? From an employment point of view, the answer could be surprisingly little. But first, it helps to ask what types of jobs make the CBD jobs market tick. Only around 60% of the foot traffic we see in the CBD is making its way to or from a workplace in an office building, so understanding how this 60% is different from the rest is a good starting point to understanding drivers of future demand. 

In the case of Brisbane, the most accurate record of the number of workers in the CBD is the Census. For 2016, it showed 122,486 people called the CBD their place of work. Of these, 92,118 were full time – the rest part time or otherwise classified. 

But what do they all do? The presumption is that CBD workers are in office jobs. But only some are. Of the 122,486 going to work in the CBD, there were 2,015 cleaners, 819 in food preparation, 7,019  sales workers, 3,462 working in hospitality, 3,795 managers of hospitality and retail services, 2,454 protective service workers, 1,212 health professionals, 2,103 educators, and a range of other occupations - including 16 who called themselves farmers and farm managers, 43 gardeners and 101 "skilled animal and horticultural workers." 

These types of occupations are also more likely to have high proportions of part time or casual workers. These also aren’t the types of occupations likely to leasing space in office towers. For that, we need to look to occupations like the 1,726 CEOs and GMs, the 12,286 specialist managers, the 19,472 business, HR and marketing professionals, the 6,490 design, engineering, science and transport professionals, the 7,243 information and communication professionals, the 6,782 legal professionals, the 2,778 engineering, ICT and science technicians, and the 30,441 clerical and admin workers. These are also more likely to be full time in nature. That collection of occupations gets us to around 90,000 – which is what I’d consider roughly the number of workers who occupy the various office buildings in the Brisbane CBD. 

A quick reality check is found in the Property Council’s office market report which for mid 2016 (when the Census was conducted) showed 1,880,451m2 of occupied space (that is total stock less total vacancies). Divide the occupied space by the 90,000 and you get 21 square metres per person. Allowing for empty desks in leased offices and various blocks of surplus but unlettable space, that sounds about right. (Quoted workspace ratios of below 15m2 per person are typically in fully occupied demonstration projects only and not typical of the entire market). 

The Census also tells us that 10,062 CBD workers (full and part time) were employed by the Federal Government, some 25,902 by the State and 2,681 by Local Government. State Government numbers since then have escalated significantly, helping drive demand for more office space. 

In addition to these workforce numbers, there were around 9,640 people who called the CBD ‘home’ (it’s where they live) and a further 4,556 visitors (most of who would have been in hotels or serviced apartments – not including overseas visitors who aren’t counted in the Census), plus there’s around 4,000 students. Workers, residents, visitors and students all blend together and become part of the CBB mass of people. Put the workforce, residents, overnight visitors, students all together and you get a total of 140,000 people. Add in an allowance for day trippers and international visitors (neither of which are counted in the Census) and you get to a theoretical 150,000 or 160,000 people a day. 

But if you’re in the business of filling office towers then residents, visitors and students aren’t of much interest. You need lawyers, accountants, engineers, marketing and HR professionals, designers, IT and communication professionals and of course the admin workers that support them. By how much do these occupations need to grow to support renewed demand for more office towers? 

Let’s go back to our guesstimate of 90,000 office workers occupying roughly 20m2 of office space each. If there was 10% growth in these occupations, that would mean another 9,000 jobs or potentially 180,000m2 of new demand needing to be satisfied. You could adopt a more conservative 15m2 per person (which you might find in a new tower as opposed to across all grades of space across the entire market) and that 10% growth still gives you 135,000m2 of new demand. 

This starts to sound promising. Of course, there’s still some 350,000 square metres of space in existing buildings available for lease, which will take up a proportion of any new demand from employment growth. But it does start to take the edge off the dimensions of market growth in office employment needed to see new towers rise out of the ground.

It’s been this type of CBD employment growth that explains why Melbourne has close to half a million square metres of new office space underway and why Sydney CBD has 200,000m2 underway. These markets are generating the types of jobs in the types of locations that will drive CBD office demand. So what’s needed to drive that sort of job growth in Brisbane? 

Last time Brisbane’s CBD jobs market grew quickly, it was off the back of the resources boom (generating engineering, legal, and a variety of technical jobs). When that boom ended, the wind was sucked from the Brisbane CBD’s sails. This time around, a recovering resources sector isn’t seeing as much new investment (ask Adani about how hard that is) but rather existing mines are gearing up production – which doesn’t generate as much new job growth but does generate export earnings. The tourism sector is looking good for the State, but part-time jobs in hospitality aren’t going to drive demand for CBD office space. Building new resorts and hotels will (think designers and engineers for example), but in anything other than established urban locations, new resorts and tourism infrastructure seems to meet with a lot of hostility these days (witness the Mt Coot-tha Zip Line). Banking, finance and large accounting firms are increasingly reliant on offshoring and centralization of functions in Sydney or Melbourne – so there’s not a great deal of comfort on that front. Government jobs growth has been a saviour for the CBD in recent years but after an extended period of strong employment growth, there are questions around how sustainable this will be in the long term given Queensland’s finances. Expecting a lot more growth from Government jobs in the CBD might be expecting too much.

Health and education are two of the fastest growing industries for the foreseeable future and the investigation of and delivery of new schools, hospitals, tertiary facilities and so on will no doubt generate jobs in design, construction, legal and project management (yay). The question around many of the larger health and education projects is when the start button gets pressed. Then there’s infrastructure as potentially the next driver of demand for occupations that will fuel the need for more CBD office space. But on the infrastructure front, we are challenged by a state budget which is short of money. There has been talk of tens of billions in new infrastructure investment associated with SEQ ‘city deals’ and other initiatives but much of this remains speculative until someone finds the money. 

Economist Gene Tunny from Adept Economics produced this graph recently which shows the height of the infrastructure boom (driven by resource exploration and new mines) which Queensland has come from, and where we are now relative to NSW and Victoria. It’s a sobering challenge. 



So on the plus side, once this investment kicks off, it’s not hard to see 10% growth in city employment generating enough demand to kick off some new office towers. On the flip side, the big question is when this will happen. 

Sunday, January 6, 2019

Travel by car is getting safer


The Christmas New Year period is for some a time of intense grief, with loved ones lost to pointless motor vehicle fatalities. We are constantly reminded of the importance of not speeding or toying with mobile phones while driving but year after year, the road toll often seems to remain stubbornly high.

It is no consolation to someone who loses a loved one or family member, but the statistics overall show that the road toll is actually falling significantly – so hopefully we are on the right track.

Total road fatalities across Australia peaked in the 1970s and have trended down sharply since then. This is despite a very significant increase in population from the 1970s onwards. Fewer drink drivers and safer vehicles are among the explanations for this.


On a per capita basis, the falling road toll is just as dramatic, falling from a peak of 30.4 deaths per 100,000 people in 1970 to 4.65 per 100,000 people in 2018. This is the lowest ever.


However, the per capita statistics don’t take account of the fact that there are more cars per person now than in the 1970s or almost any time in our history. Car ownership has become available to almost all members of society. People are more inclined to travel by car and fewer travel by public transport. So the rate of road fatalities per 100,000 vehicles is possibly a better indication – and this shows a long term decline to record lows. The early arrival of the first motor vehicles was clearly a dangerous time: poor skills, poor technology and poor roads proving a lethal combination. But road safety campaigns, improved roads (and slower roads thanks to congestion?) and improved vehicle design have all contributed to an impressive fall in the rate of road fatalities per 100,000 vehicles.


The arrival of further advances in motor vehicle technology – including the advent of self-driving cars – could lead to even further reductions to the road toll in the future.

As a means of readily accessible and on-demand transport, there are good reasons why the private motor vehicle has proved so popular. It can also be dangerous but we can find some comfort that the dangers have been falling steadily.  This statistical reality is at odds with some inaccurate suggestions that the road toll is rising and that further curbs to driver discretion and individual liberty are required to bring it under control.

Wednesday, November 7, 2018

Cars and urban mobility


Schlomo ‘Solly’ Angel is a world renowned urbanist and author of countless books including “Atlas of Urban Expansion”, “Planet of Cities” and “Tale of Scale.” He is adjunct professor at New York University (NYU) and senior research scholar at the NYU Stern Urbanization Project, where he leads the Urban Expansion initiative. He has advised the United Nations, the World Bank, and the Inter-American Development Bank (IDB). 

So when Solly Angel observes that across the United States that “the great majority of workplaces (are) now dispersed outside CBDs, employment sub-centers or live-work communities, and (are) beyond walking or biking distance” he is saying so as a well-informed global expert. 

He goes on to conclude that as a result “increasing the productivity of American cities requires a sustained focus on meeting the travel demands of the great majority of commuters rather than on improving mobility at large or on transportation strategies focused on CBDs, employment sub-centers, or live-work communities.” 

Essentially, the point he is making is that few cities observe the deterministic overlays imposed by urban planners for the sake of convenient analysis. The monocentric model (a high density CBD to which a majority of workers commute from outlying dormitory areas) he claims almost never exists in reality (although oddly, this model describes widely held prejudices about our own urban form). 

The poly-centric model (with multiple centres) is he says equally more often described in theory than in practice. There are various other urban models, including what he calls “the maximum disorder model” (which I love the sound of) but the one that he argues best describes the majority of US cities is the “constrained dispersal model” where “the great majority of jobs are dispersed throughout the metropolitan area and where workers and workplaces in a metropolitan-wide labor market adjust their locations to be within an tolerable commute range of each other.” Makes sense, hey?

In a nutshell, because we live and work in largely randomized locations across cities, a focus on urban productivity needs to acknowledge this reality and try to create transport systems that cater for the majority of commuters, not just a proportion who can be serviced by public transport in high density cores. In his words:

“While we do not have to accept this state of affairs as “the best of all possible worlds”, as Voltaire’s  Candide would have it, we do have to acknowledge it and to understand that the future of our cities is path dependent: the cities of the future will be variations on the cities of today and, barring catastrophes and calamities of one kind or another, any changes in  their spatial structure and their built form are likely to be gradual and marginal, building upon their existing spatial structure. The same observation also applies to commuting patterns: most commuting patterns are quite likely to be between dispersed residences and dispersed workplaces for a long time to come.”

In Australia, much like the USA, CBDs do not dominate as metro wide employment hubs (as they might in a hypothetical mono-centric model). They are significant but not dominant. In some cases, their significance is eroding – not because they are shrinking, but because the suburban economy is growing faster. In Sydney and Melbourne for example, inner city jobs (being the CBD plus surrounds, described as the SA3 by the ABS) represent 22% and 21% of metro wide jobs respectively. In Brisbane the figure is lower – with the inner city representing 14% of regional jobs. Interestingly, while Brisbane’s inner city added 12,802 jobs (full and part time and casual) in the 2011-2016 period (growing by 7%) the Greater Brisbane region added nine times that number at 112,517 jobs in the same period, growing faster at 12%. 

Much like the USA, our major cities mostly fit Solly’s “constrained dispersal” model. Plus, with the two fastest growing future industries being health and education, that pattern of dispersal is likely to increase over time. So to ensure our cities remain economically productive, urban transport policy should ideally support the efficient movement of the greater majority of people from home to work and during work. Cycling, walking, and traditional modes of public transport are suitable for some of the working population, but nowhere near a majority. For the majority of workers, the rational transit solution is the car.

Which makes the relentless public policy and media assault on the private car a strange thing. For the majority of workers in dispersed urban locations, it offers door to door convenience, it is on demand (ready when they are), it is comfortable (and often air conditioned), and generally quite affordable. As a transport choice, it also generates substantial government revenues via taxes (fuel excise, registration fees, etc) compared with public transport which consumes a great deal more in subsidies than it generates via the fare box. 

We have fallen into a habit of blaming congestion on the car but we also need to accept that more people (a bigger population) trying to get around on the same road space is - mathematically and inevitably - going to mean more congestion. Public transport has an important role to play but its ability to “solve” congestion has been oversold. Unless worldwide patterns of employment dispersal are suddenly and radically reversed and the monocentric urban model materializes overnight, PT will forever be limited by its convenience for the minority of urban workers with jobs in urban cores or high density suburban centres. Outside this, the 80% of the remaining urban workforce will continue to use cars as a rational and affordable choice. (Additionally, intra urban freight will continue to be totally reliant on private delivery vehicles using the road space). 

The horizon for the private car is also not bleak, as some might suggest. Advances in driverless technology, car sharing and other innovations in urban mobility that revolve around better ways to make use of private vehicles (which for many sit idle 20 out of 24 hours) are worthy of exploration, but receive little public policy attention (or investment). So far anyway.

Addressing urban congestion therefore should require a balanced policy which accepts the critical role played by private transport and the road network, along with the critical role played by public transport. Demonising the car, or suggesting that it can be largely replaced by walking or cycling or PT for a majority of the population, is delusional given the spatial reality of our urban economies. If this thinking finds its way into public policy practice, it goes from delusional to dangerous. 

You can read Solly Angel’s work on Commuting and the Spatial Structure of American Cities, via the link below. 
http://marroninstitute.nyu.edu/uploads/content/Commuting_and_the_Spatial_Structure_of_American_Cities,_20_December_2014_Version2.pdf

Thursday, October 25, 2018

Property taxes are about to fall! (and why this is bad news)


Auction clearance rates are falling, sales volumes slowing, and prices are coming off the boil in the formerly red hot markets of Sydney and Melbourne. This is bad news for State Governments who have increasingly relied on property taxes to fund a growing list of infrastructure promises from transport to schools to hospitals. It could also be bad news for the Australian economy because one of two things could happen: falling forward property revenues will see infrastructure spending curtailed (even though we continue to have an infrastructure deficit) or we will go into more debt to keep the tap on even when the tank’s running dry. Neither is a great outlook and let’s hope it doesn’t happen.

The remarkable thing is just how dependent State Governments have become on property taxes – chiefly stamp duty on transfers. (Remember, this was the tax they promised to abolish in exchange for GST revenues. Paul Keating once said: “Never stand between a State Treasurer and a bucket of money.” He was so right).

Nationally, total State Government taxes on property (according to the ABS data) have soared from around $30 billion per annum in 2009 to over $50 billion in 2016-17. That figure will probably climb for 2017-18 and could even approach the $60 billion mark, given how white hot Sydney and Melbourne were in 2017 and into the first quarter of 2018. Imagine if a slowdown in markets (mostly a fall in volumes) brings that revenue back to even 2013 levels… that would be a $20 billion per annum hit to state budgets. Ouch.


NSW is going to feel the pain. Its last State Budget announced a raft of infrastructure projects which had the country weeping in envy. Their stamp duty revenues alone grew from $4billion per annum in 2011-12 to $9billion in 2016-17. Property tax dependency also grew – from around a third to nearly 45% in the same period. A sustained fall in sales volumes combined with a fall in prices could see $3billion per annum wiped from future state revenues.



Victoria doesn’t fare much better. With both stamp duties and land taxes steadily climbing, their annual revenues have swelled by over $3billion per annum. Their property tax dependency has grown from around a third to over 45% of all income. Nearly one in two state tax dollars comes from property. That’s going to take some delicate fiscal balancing if revenues fall on the back of a slowing property market.



Queensland has less to lose. Its market (mostly the Brisbane region) never reached the fever pitch of Sydney or Melbourne. Stamp duties rose by just over $1billion per annum (but I bet State Treasurers had hoped for more). Property tax dependency in Queensland actually fell as other income sources (payroll tax, gaming taxes and motor vehicle taxes) rose faster. Plus, having never overheated, the prospects of a substantial cooling are less, meaning the likelihood of falling forward state revenues tied to a slumping property market are less. So in this case, it’s perversely not a bad thing if property taxes don’t fall in the years ahead?



Tuesday, October 9, 2018

What would low population growth mean for Australia?



Australia relies mostly on net overseas migration to sustain its rate of population growth. Our population grew by 1.6% in the year to March 2018, or by around 380,000 people. Of this, natural population growth contributed over a third (144,000) while overseas migration contributed the rest (237,000 people).

So what would happen if growing community opposition to population growth – and in particular high rates of overseas migration – meant a slowdown in our population growth rate? The pressure is growing for just such a change in policy, and many in industry fear the worst for Australia – especially property – should our growth rates slow.

To try understand what this might mean, I’ve taken a look at some low population growth countries and also at high growth countries. The data sources aren’t consistent from one measure to the next nor are the time periods the same for each country due to differences in reporting times (and data availability) but the results are pretty confronting nonetheless.

First, some countries with low rates or negative rates of population growth are listed below. Australia is there for comparison purposes only – our rate of population growth easily eclipses this group.

Is there an obvious correlation between population growth and the economy and housing? The answer it seems is no. Japan’s population is shrinking, but its GDP per capita is healthy and on a par with the UK and NZ and certainly among the wealthier of nations by this measure. Its housing, relative to incomes, is on average cheaper than the Australian average (for all major markets… Sydney and Melbourne are of course in a league of their own) and Japan is experiencing housing price growth of around 1.46% - even with a shrinking population.

Denmark also has a very low rate of population growth but a high GDP per capita and modest housing price growth. The Asian tiger economies like Taiwan and Hong Kong have negligible population growth rates but high GDP per capita and in HK’s case, outrageously expensive housing which has been rapidly rising in price.

Canada and New Zealand have rates of population growth that are roughly half ours while their GDP per capita is high by global standards. They also have expensive housing markets and high rates of price growth. (Australia’s high GDP per capital reflects our high value resources exports with a relatively small population, leading many to argue that a growing population only dilutes our prosperity given resource sector exports are not related to population).

Here is the table showing a selection of low population growth countries compared with Australia:

Country
POP GROWTH RATE
GDP PER CAPITA (IMF data)
Housing median multiple (Demographia, major markets 2018)
House price growth
(IMF Global Housing Watch, latest for each country)
Japan
-0.21%
$38,440
4.2
1.46%
Russia
-0.08%
$10,608
na
-5.45%
Taiwan
0.17%
$24,577
na
0.69%
Denmark
0.22%
$56,444
na
3%
Hong Kong
0.32%
$46,109
19.4
11.80%
China
0.41%
$8,643
na
3.18%
United Kingdom
0.52%
$39,735
4.6
1.94%
Switzerland
0.69%
$80,591
na
-0.86%
Canada
0.73%
$45,077
4.3
5.53%
New Zealand
0.79%
$41,593
8.8
4.78%
Australia
1.60%
$55,707
6.6
3.02%

So economic prosperity and housing markets are not, on this sample at least, connected to high rates of population growth. There are nations very similar to ours with much lower rates of population growth but which are equally prosperous and with equally healthy (or too healthy?) housing markets.

What about nations with higher rates of population growth than Australia? Again, the list is not convincing. Without even looking into their housing markets or economic measures, the list of nations with higher rates of population growth than ours gives little comfort to any link between high rates of population growth and economic prosperity.

I’ll leave you to peruse the list below (which lists all countries with population growth above our 1.6% per annum) and choose where you’d rather live. 

In the meantime, where does this leave us? Do we still need “a big Australia?” and is high growth essential to our prosperity? Personally, I have always believed we need a larger population to give us a critical economic mass across a range of measures and for many different reasons. Equally though, it seems odd in a country so large as Australia to concentrate nearly all growth in mostly two cities and ask them to bear the infrastructure burden to cope with that growth while other regions perfectly capable of absorbing growth with less stress are overlooked.

It’s going to be an interesting and ongoing debate and opinions will not be in short supply. Pro-growth and anti-growth proponents will dig their trenches and wage their wars. The evidence (also known as truth) is - as the saying goes - the first casualty in war and this will probably be no different. But hopefully a quick look at the evidence might leave some of us better informed. Balanced arguments, while possibly overlooked in the short term, may just find an eventual foothold in public policy debate about a very important topic.

Now here’s that list of countries growing faster than Australia, in descending order:

1
South Sudan
3.83
2
Angola
3.52
3
Malawi
3.31
4
Burundi
3.25
5
Uganda
3.2
6
Niger
3.19
7
Mali
3.02
8
Burkina Faso
3
9
Zambia
2.93
10
Ethiopia
2.85
11
Tanzania
2.75
12
Benin
2.71
13
Western Sahara
2.7
14
Togo
2.64
15
Guinea
2.61
16
Cameroon
2.56
17
Iraq
2.55
18
Liberia
2.5
19
Madagascar
2.5
20
Mozambique
2.46
21
Rwanda
2.45
22
Egypt
2.45
23
Equatorial Guinea
2.44
24
Nigeria
2.43
25
Senegal
2.39
26
Sierra Leone
2.38
27
United Arab Emirates
2.37
28
Congo, Democratic Republic of the
2.37
29
Afghanistan
2.36
30
East Timor
2.36
31
Gaza Strip
2.33
32
Yemen
2.28
33
Qatar
2.27
34
Bahrain
2.26
35
British Virgin Islands
2.25
36
Mauritania
2.17
37
Ghana
2.17
38
Djibouti
2.16
39
Turks and Caicos Islands
2.16
40
Central African Republic
2.12
41
Congo, Republic of the
2.11
42
Gambia, The
2.05
43
Jordan
2.05
44
Oman
2.03
45
Cayman Islands
2.01
46
Somalia
2
47
Luxembourg
1.98
48
Anguilla
1.97
49
Namibia
1.95
50
Solomon Islands
1.94
51
Gabon
1.92
52
Chad
1.86
53
Guinea-Bissau
1.86
54
Vanuatu
1.85
55
West Bank
1.84
56
Cote d'Ivoire
1.84
57
Singapore
1.82
58
Belize
1.8
59
Guatemala
1.75
60
Sao Tome and Principe
1.72
61
Papua New Guinea
1.71
62
Algeria
1.7
63
Kenya
1.69
64
Comoros
1.64
65
Sudan
1.64
66
Tajikistan
1.62
67
Honduras
1.6