Construction Workers

Construction Industry Economics and Policy


Innovation and Industry Evolution

The stages in the life cycle of an industry typically start with first applications of a new invention by technology leaders, followed by development and refinement of products and services, before becoming a mature industry with well-understood products and practices. Mature industries are past the early growth phase, their culture of technology has stabilised and the shape of industrial structure and processes has emerged. In many cases these industries are oligopolistic, with a few specialised firms dominating market niches in the supply chain. Consolidation leads to concentration.

The new technology that starts a cycle of industry development can be a general purpose technology (GPT) that becomes the basis of a new system of industrial production. The key feature of a GPT is ‘pervasiveness’, how it is used by other sectors in the economy and leads to ‘complementary investments and technical change in the user sections’ (Helpman and Trajtenberg 1998: 86). The examples originally used by David (1990), and broadly followed since, were steam, electricity and information technology. Lipsey, Carlaw and Bekar (2005) include two organizational GPTs in their list of two dozen since 9000BCE: mass production and the factory system; and lean production and the Toyota system. It is widely believed AI is a new GPT.

Thinking about the construction industry and the production of the built environment as an evolving ‘system of production’ provides a new perspective on the context and direction of innovation and its evolution since the first industrial revolution. Hughes’ (1987) life cycle model had seven phases: invention, development, innovation, transfer, growth, competition, and consolidation. Within those seven phases of the life-cycle are two interior cycles that divide an industry’s evolution into two stages: Cycle 1 is invention, development, innovation, and transfer, Cycle 2 is growth, competition, and consolidation.

Cycle 2 focuses on innovation in production and organization, when mature technological systems emerge and construction materials like cement, concrete and glass, and components like building management systems, interior walls, plumbing fixtures, lifts and elevators have become oligopolistic industries in a mature supply chain. A mature industry produces a specific culture of technology, embodied in the firms and social institutions of the system of production, and creates the tendency for an industry to develop along defined technological trajectories unless or until deflected or disrupted by a powerful external force.

A diverse cluster of industries with deep layers of specialised firms in a dense network of producers, suppliers and materials is a ‘technological system’ (Hughes 1987: 47). Electricity grids and railways have networks, telecommunications and air traffic use interconnected nodes, postal systems use existing networks, some are geographically large, some are local, some are narrow, some broad.

Construction innovation has been narrowly focused because construction is a mature technological system, but this is changing. With a technological trajectory based on AI and associated emerging production technologies, the commercial contracting part of the industry will adopt these technologies as they become viable. The organization and structure of the industry will then change in response to changes in relative costs as the economies of scale of digitized production technologies are realized.

AI as a new GPT may be the start of a new life cycle in building and construction technology, and may be as disruptive as steam power was in the nineteenth century to the master builders and craftsmen of the day. The organization of construction is currently centred on project managers and incremental innovation, but a transformed industry would be focused on integrators who combine site preparation with production and assembly of digitally designed and fabricated components and modules.

David, P. A. 1990, The Dynamo and the Computer: An Historical Perspective on the Modern Productivity Paradox‟, American Economic Review, Vol. 80, pp. 355 - 361.
Helpman. E. and Trajtenberg, J. 1998. Diffusion of General Purpose Technologies, in Helpman, E. (ed.), General Purpose Technologies and Economic Growth, Cambridge: MIT Press. 85-119.
Hughes, T. P. 1987. The evolution of large technological systems, in The Social Construction of Technological Systems: New Directions in the Sociology and History of Technology, W. E. Bijker, T. P. Hughes, and T. J. Pinch (eds.), Cambridge, Mass.: MIT Press.
Hughes, T. P. 1989. American Genesis: A Century of Invention and Technological Enthusiasm 1870-1970, Chicago: University of Chicago Press. 
Lipsey, R. G., Carlaw, K. I. and Bekar, C. T. 2005. Economic Transformations: General Purpose Technologies and Long-term Economic Growth, Oxford: Oxford University Press.
10 views0 comments

A Significant Contribution to an Important Issue

In Democratizing Our Data, Julia Lane argues that good data are essential for democracy. She believes that public policy choices can only be made intelligently when the people making the decisions have accurate and objective statistical information to inform them of the choices they face and the results of choices they make.

“We must rethink ways to democratize data. There are successful models to follow and new legislation that can help effect change. The private sector's Data Revolution—where new types of data are collected and new measurements created by the private sector to build machine learning and artificial intelligence algorithms—can be mirrored by a public sector Data Revolution, one that is characterized by attention to counting all who should be counted, measuring what should be measured, and protecting privacy and confidentiality. Just as US private sector companies—Google, Amazon, Microsoft, Apple, and Facebook—have led the world in the use of data for profit, the US can show the world how to produce data for the public good.”

Lane’s book really only covers the US. It is very focused on the institutional problems there in chapter 3, and features a couple of good case studies on developing useful data sets from disparate sources in chapter 4. While the problems collecting and managing data for national statistics in the US is unique, broader issues around extent and quality are not. Chapter 2 addresses those issues, and looks at why measurement is difficult, and why it is hard for agencies to innovate (no incentive) and develop (no funding) new measures. Its very much an insiders account. I thought it a big improvement on recent books on GDP etc that tend to highlight the problems, it’s a good read (and quick, at 120 pages).

There is discussion on new data sources, and how the private sector finds ways to use it. However, because public data requires confidentiality agencies need new tools and skills to be able to use it. That is chapter 5, and chapter 6 proposes a new organizational model. Lane makes a compelling argument for building a new public data system in order to safeguard privacy and improve the US government's ability to implement policy initiatives.

I suspect National Statistical Agencies everywhere are under pressure.

Lane emphasises the increasing costs and diminishing returns for surveys, the traditional source of data. However, bureaucratic inertia and vested interests, lack of funding for pilot projects, and privacy and confidentiality issues combine to make developing new sources and products difficult. How difficult in different countries I don’t know. I’d like to think most would have something based on administrative data by now, but am probably being way too optimistic.

As the cover says “A Manifesto”. For those who care about data and the statistics used for policy decisions on the economy, health, education, transport, community and social assistance and so on, this book is a must read.

2 views0 comments

Updated: Feb 8

Three industry scenarios

McKinsey’s Artificial Intelligence: Construction Technology’s Next Frontier (Agarwal et al 2018) is one of a series of recent papers from the management consultants on AI, automation and infrastructure. They identify five AI-powered applications, and use cases that have already arrived in other industries, that can be applied to construction.

This is a practical approach that seems to target major contractors, and is a different approach to previous reports that could have been primarily intended for public sector clients. McKinsey has been seriously developing their infrastructure practice for some years now, positioning themselves for the global infrastructure boom they forecast over the next few decades. The five industry applications are:

  • Transportation route optimization algorithms for project planning optimization;

  • Pharmaceutical outcomes prediction for constructability issues

  • Retail supply chain optimization for materials and inventory management

  • Robotics for modular or prefabrication construction and 3-D printing;

  • Healthcare image recognition for risk and safety management.

Each of these has a short discussion with some examples of crossover potential. They are all plausible extensions of current technology, and in robotics, 3-D printing and drones, leading construction firms are already well advanced. Using AI for optimization is obvious (Gans 2018), and is addressed below (see figure 2), but construction firms typically contract out specialized tasks such as design and logistics, rather than invest in the hardware and software development needed (Manly and XXX). Its questionable whether McKinsey makes a convincing case for using AI in construction. Are these are the pathways into construction for AI, or the only ones?

McKinsey also looks at some machine learning algorithms that are relevant to contractors, and briefly assesses their potential engineering and construction applications. Despite their extensive reporting on BIM elsewhere there is no discussion of the potential use of AI in design and engineering, or in restructuring processes. They do have a generic framework for types of machine learning, and they suggest algorithms will be useful for: refining quality control and claims management; increasing talent retention and development; boosting project monitoring and risk management; and constant design optimization.

If McKinsey has a more nuanced story to tell on pathways for AI into construction it might look something like the scenarios depicted by the World Economic Forum and the Boston Consulting Group in their Future Scenarios and Implications for the Construction Industry (WEF/BCG 2018). This scenario analysis is the second, final step in their Future of Construction project, which has involved people from industry and researchers from a wide range of organizations, after the Shaping the Future of Construction report (WEF/BCG 2016). They use infrastructure and urban development Industry (IU) to describe what has elsewhere been called the built environment sector.

The three future scenarios the WEF describe make technological context central to the future form of the industry. The scenarios depict three extreme yet plausible versions of the future. Each scenario is used to extrapolate implications for the industry, identifying potential winners from technological transformation, and the range of examples and ideas shows the value of such a widespread collaboration between industry, government and academia. The WEF does not say how far into the future they are looking, although it seem to be a lot further than McKinsey:

  1. In Building in a virtual world, virtual reality touches all aspects of life, and intelligent systems and robots run the construction industry. Interconnected intelligent systems and robots run the IU, software players will gain power, and new businesses will emerge around data and services.

  2. In Factories run the world, a corporate-dominated society uses prefabrication and modularization to create cost-efficient structures. The entire IU value chain adopts prefabrication, lean processes and mass customization, with suppliers benefiting the most from the transition and take advantage of new business opportunities through integrated system offerings and logistics requirements.

  3. In A green reboot, a world addressing scarce natural resources and climate change rebuilds using eco-friendly construction methods and sustainable materials. Innovative technologies, new materials and sensor-based surveillance ensure low environmental impacts, so players with deep knowledge of materials and local brownfield portfolios thrive on the new business opportunities around environmental-focused services and material recycling.

It is important to keep in mind that scenarios are not predictions of the future. Rather, they outline a broad spectrum of possible futures. In the real future, the construction industry will most probably include elements of all three, as the supply side of changes in demand for different types of building.

One issue is where the industry is at in regard to technology take-up, now that there is widespread recognition of the reality of a digital future. Will construction industry development over the next decades absorb the impacts of new technology and be gradual, changing industry practice over time without significantly affecting industry structure or dynamics? Given the entanglement of economic, social, political, and legal factors in the construction technological system this might be the case, however there are good reasons to think this may be wrong. Machine learning, AI, automation and robotics are an interconnected set of technologies that are evolving quickly, enabled by expanding connectivity and the massively scaleable hardware available today.

In 2016 a scenario analysis called Farsight for Construction, looking at the future of the building and construction industry in Queensland, Australia, was released (Quezada et al, 2016). The scenarios describe “four plausible futures for Queensland’s construction industry over the coming two decades, with a focus on impacts for jobs and skills. Each scenario consists of a description of Queensland’s construction industry in the year 2036, a narrative of how the scenario came about, and a commentary on plausibility.” In the figure below Australia is substituted for Queensland.

If we think of the structure of the construction industry as a pyramid of different sized firms, there is a broad base of tradesmen and small firms at the bottom, followed by a deep layer of medium sized firms, and a small top triangle with a few large firms. Some of those large firms, and some of their major clients, are clearly on the technological frontier, and their investment in capability and capacity should deliver significant increases in efficiency and productivity, and probably scale.

Some medium-size firms are also making these investments, and also have access to technologies like algorithmic optimisation, platform-based project management, robotic, VR and AR applications and so on. The WEF Shaping the Future of Construction report (WEF/BCG 2016) included snapshots of what a range of firms at the frontier were doing. These examples reflect the diversity of the industry, and were missing from McKinsey’s high level analysis.

A period of technology-driven restructuring of the building and construction industry may be about to start, similar to the second half of the 1800s when the new materials of glass, steel and reinforced concrete arrived, which led to new methods of production, organisation and management. There are many implications of such a restructuring. Some firms are rethinking their processes in response to developments in AI, robotics and automation as capabilities improve quickly and the range of new products using these technologies expands. Many firms, however, are not. Meanwhile, firms at the frontier are exploring new technology and pushing the boundaries of what is possible, and are inventing new processes.


Agarwal, R., Chandrasekaran, S. and Sridhar, S. 2016. Artificial Intelligence: Construction Technology’s Next Frontier, McKinsey & Co.
Quezada, G., Bratanova, A., Boughen, N. and Hajowicz, S. 2016. Farsight for Construction: Exploratory scenarios for Queensland’s construction industry to 2036, CSIRO, Australia.
WEF/BCG, 2016. Shaping the Future of Construction: A Breakthrough in Mindset and Technology, World Economic Forum and the Boston Consulting Group, Geneva.
WEF/BCG, 2017. Future Scenarios and Implications for the Construction Industry, World Economic Forum and the Boston Consulting Group, Geneva.

3 views0 comments

About the Blog

This blog is interested in the organisation of the building and construction industry and its role in the creation and maintenance of the built environment.

Like other industries, it is being reshaped by rapid and widespread advances in materials, technology and capability. How those advances might affect an industry changes slowly over time is, I think, an interesting question.The blog collects data and discusses these trends and their effect on industry structure and performance.


The economic perspective focuses on firms and industries rather than individual projects.

Previous Posts

Gerard 2.jpg

Gerard de Valence

I studied politics, philosophy and economics at Sydney University and worked in the private sector for a decade before becoming an academic in the School of the Built Environment at the University of Technology Sydney and UCL’s Bartlett School of Construction and Project Management in London. The ABOUT page has my bio.