June 8, 2014 Leave a comment
(Or “how to buy a Smarter City that won’t go bump in the night”)
There are good reasons why the current condition and future outlook of the world’s cities have been the subject of great debate in recent years. Their population will double from 3 billion to 6 billion by 2050; and while those in the developing world are growing at such a rate that they are challenging our ability to construct resilient, efficient infrastructure, those in developed countries often have significant levels of inequality and areas of persistent poverty and social immobility.
Many people involved in the debate are convinced that new approaches are needed to transport, food supply, economic development, water and energy management, social and healthcare, public safety and all of the other services and infrastructures that support cities.
As a consequence, analysts such as Frost & Sullivan have estimated that the market for “Smart City” solutions that exploit technology to address these issues will be $1.5trillion by 2020.
But anyone who has tried to secure investment in an initiative to apply “smart” technology in a city knows that it is not always easy to turn that theoretical market value into actual investment in projects, technology, infrastructure and expertise.
It’s not difficult to see why this is the case. Most investments are made in order to generate a financial return, but profit is not the objective of “Smart Cities” initiatives: they are intended to create economic, environmental or social outcomes. So some mechanism – an investment vehicle, a government regulation or a business model – is needed to create an incentive to invest in achieving those outcomes.
Institutions, Business, Infrastructure and Investment
Citizens expect national and local governments to use their tax revenues to deliver these objectives, of course. But they are also very concerned that the taxes they pay are spent wisely on programmes with transparent, predictable, deliverable outcomes, as the current controversy over the UK’s proposed “HS2” high speed train network and previous controversies over the effectiveness of public sector IT programmes show.
Nevertheless, the past year has seen a growing trend for cities in Europe and North America to invest in Smart Cities technologies from their own operational budgets, on the basis of their ability to deliver cost savings or improvements in outcomes.
For example, some cities are replacing traditional parking management and enforcement services with “smart parking” schemes that are reducing congestion and pollution whilst paying for themselves through increased enforcement revenues. Others are investing their allocation of central government infrastructure funds in Smart solutions – such as Cambridge, Ontario’s use of the Canadian government’s Gas Tax Fund to invest in a sensor network and analytics infrastructure to manage the city’s physical assets intelligently.
The providers of Smart Cities solutions are investing too, by implementing their services on Cloud computing platforms so that cities can pay incrementally for their use of them, rather than investing up-front in their deployment. Minneapolis, Minnesota and Montpelier, France, recently announced that they are using IBM’s Cloud-based solutions for smarter water, transport and emergency management in this way. And entrepreneurial businesses, backed by Venture Capital investment, are also investing in the development of new solutions.
However, we have not yet tapped the largest potential investment streams: property and large-scale infrastructure. The British Property Federation, for example, estimates that £14 billion is invested in the development of new property in the UK each year. For the main part, these investment streams are not currently investing in “Smart City” solutions.
To understand why that is the case – and how we might change it – we need to understand the difference in three types of risk involved in investing in smart infrastructures compared with traditional infrastructures: construction risk; the impact of operational failures; and confidence in outcomes.
At a discussion in March of the financing of future city initiatives held within the Lord Mayor of the City of London’s “Tommorrow’s Cities” programme, Daniel Wong, Head of Infrastructure and Real Estate for Macquarie Capital Europe, said that only a “tiny fraction” – a few percent – of the investable resources of the pension and sovereign wealth funds often referred to as the “wall of money” seeking profitable long-term investment opportunities in infrastructure were available to invest in infrastructure projects that carry “construction risk” – the risk of financial loss or cost overruns during construction.
For conventional infrastructure, construction risk is relatively well understood. At the Tomorrow’s Cities event, Jason Robinson, Bechtel’s General Manager for Urban Development, said that the construction sector was well able to manage that risk on behalf of investors. There are exceptions – such as the delays, cost increases and reduction in scale of Edinburgh’s new tram system – but they are rare.
So are we similarly well placed to manage the additional “construction risk” created when we add new technology to infrastructure projects?
Unfortunately, research carried out in 2013 by the Standish Group on behalf of Computerworld suggests not. Standish Group used data describing 3,555 IT projects between 2003 and 2012 that had labour costs of at least $10 million, and found that only 6.4% were wholly successful. 52% were delivered, but cost more than expected, took longer than expected, or failed to deliver everything that was expected of them. The rest – 41.4% – either failed completely or had to be stopped and re-started from scratch. Anecdotally, we are familiar with the press coverage of high profile examples of IT projects that do not succeed.
We should not be surprised that it is so challenging to deliver IT projects. They are almost always driven by requirements that represent an aspiration to change the way that an organisation or system works: such requirements are inevitably uncertain and often change as projects proceed. In today’s interconnected world, many IT projects involve the integration of several existing IT systems operated by different organisations: most of those systems will not have been designed to support integration. And because technology changes so quickly, many projects use technologies that are new to the teams delivering them. All of these things will usually be true for the technology solutions required for Smart City projects.
By analogy, then, an IT project often feels like an exercise in building an ambitiously new style of building, using new materials whose weight, strength and stiffness isn’t wholly certain, and standing on a mixture of sand, gravel and wetland. It is not surprising that only 6.4% deliver everything they intend to, on time and on budget – though it is also disappointing that as many as 41.4% fail so completely.
However, the real insight is that the characteristics of uncertainty, risk, timescales and governance for IT projects are very different from construction and infrastructure projects. All of these issues can be managed; but they are managed in very different ways. Consequently, it will take time and experience for the cultures of IT and construction to reconcile their approaches to risk and project management, and consequently to present a confident joint approach to investors.
The implementation of Smart Cities IT solutions on Cloud Computing platforms by their providers mitigates this risk to an extent by “pre-fabricating” these components of smart infrastructure. But there is still risk associated with the integration of these solutions with physical infrastructure and engineering systems. As we gain further experience of carrying out that integration, IT vendors, investors, construction companies and their customers will collectively increase their confidence in managing this risk, unlocking investment at greater scale.
We are all familiar with IT systems failing.
Our laptops, notebooks and tablets crash, and we lose work as a consequence. Our television set-top boxes reboot themselves midway through recording programmes. Websites become unresponsive or lose data from our shopping carts.
But when failures occur in IT systems that monitor and control physical systems such as cars, trains and traffic lights, the consequences could be severe: damage to property, injury; and death. Organisations that invest in and operate infrastructure are conscious of these risks, and balance them against the potential benefits of new technologies when deciding whether to use them.
The real-world risks of technology failure are already becoming more severe as all of us adopt consumer technologies such as smartphones and social media into every aspect of our lives (as the driver who followed his satellite navigation system off the roads of Paris onto the pavement, and then all the way down the steps into the Paris Metro, discovered).
The noted urbanist Jane Jacobs defined cities by their ability to provide privacy and safety amongst citizens who are usually strangers to each other; and her thinking is still regarded today by many urbanists as the basis of our understanding of cities. As digital technology becomes more pervasive in city systems, it is vital that we evolve the policies that govern digital privacy to ensure that those systems continue to support our lives, communities and businesses successfully.
Google’s careful exploration of self-driving cars in partnership with driver licensing organisations is an example of that process working well; the discovery of a suspected 3D-printing gun factory in Manchester last year is an example of it working poorly.
These issues are already affecting the technologies involved in Smart Cities solutions. An Argentinian researcher recently demonstrated that traffic sensors used around the world could be hacked into and caused to create misleading information. At the time of installation it was assumed that there would never be a motivation to hack into them and so they were configured with insufficient security. We will have to ensure that future deployments are much more secure.
Conversely, we routinely trust automated technology in many aspects of our lives – the automatic pilots that land the planes we fly in, and the anit-lock braking systems that slow and stop our cars far more effectively than we are able to ourselves.
If we are to build the same level of trust and confidence in Smart City solutions, we need to be open and honest about their risks as well as their benefits; and clear how we are addressing them.
Smart infrastructures such as Stockholm’s road-use charging scheme and London’s congestion charge were constructed in the knowledge that they would be financially sustainable, and with the belief that they would create economic and environmental benefits. Subsequent studies have shown that they did achieve those benefits, but data to predict them confidently in advance did not exist because they were amongst the first of their kind in the world.
The benefits of “Smart” schemes such as road-use charging and smart metering cannot be calculated deterministically in advance because they depend on citizens changing their behaviour – deciding to ride a bus rather than to drive a car; or deciding to use dishwashers and washing machines overnight rather than during the day.
There are many examples of Smart Cities projects that have successfully used technology to encourage behaviour change. In a smart water meter project in Dubuque, for example, households were given information that told them whether their domestic appliances were being used efficiently, and alerted to any leaks in their supply of water. To a certain extent, households acted on this information to improve the efficiency of their water usage. But a control group who were also given a “green points” score telling them how their water conservation compared to that of their near neighbours were found to be twice as likely to take action to improve their efficiency.
However, these techniques are notoriously difficult to apply successfully. A recycling scheme that adopted a similar approach found instead that it lowered recycling rates across the community: households who learned that they were putting more effort into recycling than their neighbours asked themselves “if my neighbours aren’t contributing to this initiative, then why should I?”
The financial vehicles that enable investment in infrastructure and property are either government-backed instruments that reward economic and social outcomes such as reductions in carbon footprint or the creation of jobs ; or market-based instruments based on the creation of direct financial returns.
So are we able to predict those outcomes confidently enough to enable investment in Smart Cities solutions?
I put that question to the debating panel at the Tomorrow’s Cities meeting. In particular, I asked whether investors would be willing to purchase bonds in smart metering infrastructures with a rate of return dependent on the success of those infrastructures in encouraging consumers to reduce their use of water and energy.
The response was a clear “no”. The application of those technologies and their effectiveness in reducing the use of water and electricity by families and businesses is too uncertain for such investment vehicles to be used.
Smart Cities solutions are not straightforward engineering solutions such as electric vehicles whose cost, efficiency and environmental impacts can be calculated in a deterministic way. They are complex socio-technical systems whose outcomes are emergent and uncertain.
Our ability to predict their performance and impact will certainly improve as more are deployed and analysed, and as University researchers, politicians, journalists and the public assess them. As that happens, investors will be more willing to fund them; or, with government support, to create new financial vehicles that reward investment in initiatives that use smart technology to create social, environmental and economic improvements – just as the World Bank’s Green Bonds, launched in 2008, support environmental schemes today.
Evidence and Leadership
The evidence base need to support new investment vehicles is already being created. In Canada, for example, a collaboration between Canadian insurers and cities has developed a set of tools to create a common understanding of the financial risk created by the effects of climate change on the resilience of city infrastructures.
More internationally, the “Little Rock Accord” between the Madrid Club of former national Presidents and Prime Ministers and the P80 group of pension funds agreed to create a task force to increase the degree to which pension and sovereign wealth funds invest in the deployment of technology to address climate change issues, shortages in resources such as energy, water and food, and sustainable, resilient growth. My colleague the economist Mary Keeling has been working for IBM’s Institute for Business Value to more clearly analyse and express the benefits of Smart approaches – in water management and transportation, for example. And Peter Head’s Ecological Sequestration Trust and Robert Bishop’s International Centre for Earth Simulation are both pooling international data and expertise to create models that explore how more sustainable cities and societies might work.
But the Smart City programmes which courageously drive the field forward will not always be those that demand a complete and detailed cost/benefit analysis in advance. Writing in “The Plundered Planet”, the economist Paul Collier asserts that any proposed infrastructure of reasonable novelty and significant scale is effectively so unique – especially when considered in its geographic, political, social and economic context – that an accurate cost/benefit case simply cannot be constructed.
Instead, initiatives such as London’s congestion charge and bicycle hire scheme, Sunderland’s City Cloud and Bogota’s bikeways and parks were created by courageous leaders with a passionate belief that they could make their cities better. As more of those leaders come to trust technology and the people who deliver it, their passion will be another force behind the adoption of technology in city systems and infrastructure.
What’s the risk of not investing in a Smarter City?
For at least the last 50 years, we have been observing that life is speeding up and becoming more complicated. In his 1964 work “Notes on the Synthesis of Form“, the town planner Christopher Alexander wrote:
“At the same time that the problems increase in quantity, complexity and difficulty, they also change faster than ever before. New materials are developed all the time, social patterns alter quickly, the culture itself is changing faster than it has ever changed before … To match the growing complexity of problems, there is a growing body of information and specialist experience … [but] not only is the quantity of information itself beyond the reach of single designers, but the various specialists who retail it are narrow and unfamiliar with the form-makers’ peculiar problems.”
(Alexander’s 1977 work “A Pattern Language: Towns, Buildings, Construction” is one of the most widely read books on urban design; it was also an enormous influence on the development of the computer software industry).
The physicist Geoffrey West has shown that this process is alive and well in cities today. As the world’s cities grow, life in them speeds up, and they create ideas and wealth more rapidly, leading to further growth. West has observed that, in a world with constrained resources, this process will lead to a catastrophic failure when demand for fresh water, food and energy outstrips supply – unless we change that process, and change the way that we consume resources in order to create rewarding lives for ourselves.
There are two sides to that challenge: changing what we value; and changing how we create what we value from the resources around us.
The Transition movement, started by Rob Hopkins in Totnes in 2006, is tackling both parts of that challenge. “Transition Towns” are communities who have decided to act collectively to transition to a way of life which is less resource-intensive, and to value the characteristics of such lifestyles in their own right – where possible trading regionally, recycling and re-using materials and producing and consuming food locally.
The movement does not advocate isolation from the global industrial economy, but it does advocate that local, alternative products and services in some cases can be more sustainable than mass-produced commodities; that the process of producing them can be its own reward; and that acting at community level is for many people the most effective way to contribute to sustainability. From local currencies, to food-trading networks to community energy schemes, many “Smart” initiatives have emerged from the transition movement.
We will need the ideas and philosophy of Transition to create sustainable cities and communities – and without them we will fail. But those ideas alone will not create a sustainable world. With current technologies, for example, one hectare of highly fertile, intensively farmed land can feed 10 people. Birmingham, my home city, has an area of 60,000 hectares of relatively infertile land, most of which is not available for farming at all; and a population of around 1 million. Those numbers don’t add up to food self-sufficiency. And Birmingham is a very low-density city – between one-half and one-tenth as dense as the growing megacities of Asia and South America.
Cities depend on vast infrastructures and supply-chains, and they create complex networks of transactions supported by transportation and communications. Community initiatives will adapt these infrastructures to create local value in more sustainable, resilient ways, and by doing so will reduce demand. But they will not affect the underlying efficiency of the systems themselves. And I do not personally believe that in a world of 7 billion people in which resources and opportunity are distributed extremely unevenly that community initiatives alone will reduce demand significantly enough to achieve sustainability.
We cannot simply scale these systems up as the world’s population grows to 9 billion by 2050, we need to change the way they work. That means changing the technology they use, or changing the way they use technology. We need to make them smarter.