Comparisons of Construction to Manufacturing Use Flawed Data

 

Construction productivity has been negatively compared with manufacturing (e.g. McKinsey), and the comparisons are typically between all of construction and all of manufacturing. The problem is that both are averages of extremely varied economic activities of firms, based on data collected by the standard industrial classification (SIC) system. This makes useful comparisons between the two difficult, as this post using UK data argues. The post first breaks down industry statistics on UK construction and manufacturing to show the structural differences, and then compares construction to the car industry, showing a comparison between the two requires including repair and maintenance with vehicle manufacture. Lessons from other industries and their production methods and processes can be useful and informative, however, comparing performance between industries is very difficult without adjustments to make the subjects comparable.

 

The production of building elements and components somewhere other than the construction site has been variously called prefabrication, pre-cast and pre-assembly construction, and offsite manufacturing (OSM). The degree of OSM and preassembly varies from basic sub-assemblies to entire modules, and the use of OSM varies greatly from country to country. Types of offsite construction are panelised systems, volumetric systems with partial assembly of rooms, units or pods offsite, and factory built modular components or homes. Offsite manufacture is used to describe factory production and preassembly of components, elements or modules. Prefabrication is used to describe offsite production of components that are installed onsite. The idea that OSM and prefabrication are the solution to problems of poor quality and low productivity in construction became central to the movement to ‘reform’ construction by making it more like manufacturing.

 

Advocates of industrialized building argued for construction to adopt similar production practices to manufacturing, particularly car manufacturing. However, while there are some factory made buildings, the number and type of standardized buildings is limited, whereas opportunities for producers of standardized construction products are widespread. Onsite production is organized around those standard parts and materials but manufacturing, in contrast, is organized around standardised products and continuous production runs. 

 

In UK construction the largest grouping by number of enterprises and employment is specialised construction, typically single trade contractors (there are 17 individual industries or trades under SIC 43). The largest group by turnover is building contractors, including residential and non-residential building with only two SIC sub-categories. Civil engineering contractors have the smallest number of enterprises and employment but the highest average number of employees and highest average turnover per enterprise. Civil engineering work is typically of larger scale compared to building work.

 

Table 1. UK Construction turnover 2019

Source:  Meikle, J. and de Valence, G. 2022. Construction products and producers: One industry or three, in Best, R. and Meikle, J. (eds.) Describing Construction: Industries, projects and firms, London: Taylor and Francis. Data from ONS Annual Business Survey 2018.

 

Data on construction turnover by size of firm includes the value of subcontracting and construction work by non-contractors. The distribution of construction turnover by number and size of firm and average turnover per firm is: 99% of construction firms have less than 50 employees and are responsible for just over 50% of turnover; and 94% of firms have less than 10 employees and are responsible for around 35% of turnover. At the other end of the size scale, less than 1% of firms, those with 50 or more employees, are responsible for the other 50% of turnover. Around 0.1%, a few hundred, are responsible for around 30% of turnover and each of these has an annual turnover averaging around £275 million. The structure of the construction typically takes this form.   

 

Table 2. Construction firms by employment 2019

Source:  Meikle, J. and de Valence, G. 2022. Construction products and producers: One industry or three, in Best, R. and Meikle, J. (eds.) Describing Construction: Industries, projects and firms, London: Taylor and Francis. Data from ONS Annual Business Survey 2018.

 

Although the SIC groups all construction firms into a single category, that is for statistical convenience based on conventions developed originally for classifying manufacturing. The exclusion of design from construction output while included in manufacturing and the inclusion of R&M in construction but not in manufacturing is one result.[i] Another is the view of construction as a single industry, producing and maintaining buildings and structures, despite their many different types and the differences in the producers and processes used in their delivery. 

 

Manufacturing in the UK comprises 24 two-digit industrial groups (SIC 10 to SIC 33), for example, food products (SIC 10), manufacture of paper and paper products (SIC 17) and manufacture of motor vehicles, trailers and semi-trailers (SIC 29); and 325 individual industries.  Manufacturing of fabricated metal products except machinery and equipment (SIC 25) is the largest two-digit group with 22 individual industries, 26,301 total group enterprises and total group turnover of  £23.6 billion; the smallest is the single industry group of manufacture of tobacco products (SIC 12) with nine enterprises and a turnover of £12 million.  Manufacturing is not only relatively large but extremely diverse and industry policies have reflected that by targeting specific industries such as IT and automobiles for example.

 

The table below shows that total UK Construction turnover is less than 50% of Manufacturing turnover, although it is much larger than any individual manufacturing industry.  Manufacturing has 21% of firms that are small and medium size, construction has 6%, and manufacturing turnover is more concentrated in the larger firms.

 

 

Table 3. UK construction and manufacturing compared by size of firm

Source: Meikle, J. and de Valence, G. 2022. Construction products and producers: One industry or three, in Best, R. and Meikle, J. (eds.) Describing Construction: Industries, projects and firms, London: Taylor and Francis. Data from ONS Annual Business Survey 2018.

 

The largest UK manufacturing industry in 2018 was motor vehicles, with 22% of construction turnover and 5.5% of construction employment, and it is manufacture of motor vehicles that is often compared with construction and used as the example to be followed in OSM. Based on turnover per employee (an imperfect but indicative measure of productivity), vehicle manufacturing (&07,965) is over three times as productive as construction (197,902). This might be the case, or it may be a statistical illusion, created by the framework of the SIC.

 

 

Table 4. Comparing UK construction and vehicle manufacture 2018

Source: Meikle, J. and de Valence, G. 2022. Construction products and producers: One industry or three, in Best, R. and Meikle, J. (eds.) Describing Construction: Industries, projects and firms, London: Taylor and Francis. Data from ONS Annual Business Survey 2018.

 

 

Table 5 breaks down construction to its main components and adjusts manufacturing by including both the manufacture and repair and the maintenance of motor vehicles.  All construction includes both new construction and the repair and maintenance of existing buildings and works. The manufacture of motor vehicles does not. In order to adjust for this, maintenance of vehicles (SIC 45.2) should be added to manufacture of vehicles (SIC 29.1) to make the groups more comparable. When vehicle maintenance is added to manufacture, turnover increases by 58% but employment increases by almost 180%.  

 

This allows a more realistic comparison and reveals that motor vehicles and their maintenance (SIC 29.1 plus SIC  45.2) has almost the same turnover per worker  (1.6mn) as building construction (1.4mn), twice that of specialist construction (0.9mn) but less than engineering construction (2.1mn).  Turnover per worker is a metric of productivity and, on this basis, all construction is less productive than all manufacturing and much less productive than motor vehicle production. However, when repair is added to manufacture, the car industry is on a par with building, the largest part of construction. 

 

 

Table 5. Turnover and employment by SIC division 2018

Source:  Meikle, J. and de Valence, G. 2022. Construction products and producers: One industry or three, in Best, R. and Meikle, J. (eds.) Describing Construction: Industries, projects and firms, London: Taylor and Francis. Data from ONS Annual Business Survey 2018.

 

With the differences in these industries in terms of firm size, turnover and employment, it is difficult to draw clear conclusions from a comparison of their structure, economic performance or productivity. Vehicle manufacture and, to a lesser extent vehicle repair and maintenance, are capital intensive businesses. Construction, generally, is not, although a few activities like tunnelling and prefabricated housing are. Comparisons between manufacturing and construction based on the figures from the SIC are not helpful or accurate without adjustment.   

 

Nevertheless, on the basis of these comparisons, for the last three decades advocates for applying production methods from car manufacturing to offsite manufacturing in construction have argued this is necessary to improve construction productivity and products. Despite the distinctly different characteristics of manufacturing and construction there have been and are many attempts to industrialize construction. However, after decades of efforts to promote OSM, the market share of OSM remains small, estimates are low single digits of total construction work in the UK, US and Australia. Success elsewhere is restricted to specific markets such as fast food outlets and hotels, or house manufacturers like the Japanese and Scandinavian firms Sekisui and Ikea. 

 

The US and UK governments have both supported OSM, with the UK government funding research, publishing case studies and promoting OSM in construction for decades.[i] In the US a Technology Roadmap for Advanced panelised construction was produced in 2003 for the Department of Housing and Urban Development as a Partnership for Advanced Technology in Housing (PATH[ii]). Despite these efforts, offsite production is not industry practice in either country. Although pre-cast concrete and panelised construction are widely used, OSM has not led to significant advances in mechanization or required a thorough reorganization of project management methods.  

 

OSM markets exist mainly in housing and institutional building, wherever it is the most effective or efficient piece of technology available and there is a lot of repetition from project to project. This manufacturing-centric view of progress in construction, endorsed by numerous government and industry reports, was the end point of the development trajectory from the first to the third industrial revolutions. Despite all efforts this has not become the primary system of construction of the built environment because OSM does not deliver a decisive advantage over onsite production for the great majority of projects. Instead, construction has a deep, diverse and specialised value chain that resists integration because it is flexible and adapted to economic variability.

---

[i] Farmer, M. 2016. Modernise or die, London: Construction Leadership Council.

[ii] PATH, 2004. Technology Roadmap: Advanced panelised construction, 2003 Progress Report. Partnership for Advanced Technology in Housing (PATH), Department of Housing and Urban Development, Office of Policy Development and Research, Washington, D.C.

[i] Despite the importance of repair and maintenance, only Canada has an annual business capital and repair expenditures survey. Between 2006 and 2016 construction R&M by firms averaged nine percent of their total capital expenditure, or around 1.2 percent of GDP, ranging between one percent of GDP in 2006 and 1.3 percent in 2012. Statistics Canada. Table: 34-10-0035-01 Capital and repair expenditures, non-residential tangible assets

Construction Industry Policy and Industry Culture

 

In a time of rapid urbanisation and great social and environmental challenges, the built environment and associated housing, infrastructure and urban policies havebecome central issues in public policy. The quality of the built environment is a major determinant of the quality of life. Further, cities are at the centre of themodern economy and, in a fundamental sense, how well cities function depends on how well the many and diverse industries, firms and organizations across thebuilt environment sector can design, deliver and operate the projects required. The resilience of cities to climate change is being tested as temperatures increaseand fires and floods become more intense. However, because of the range and complexity of these issues it is difficult for governments to develop and implementcoordinated built environment industry policies that address these issues satisfactorily.

Industry policy was out of favour for a couple of decades before the financial crisis in 2007-08 in the US, UK and Australia, although the European Union (EU)and many Asian countries followed well developed national strategic plans. This was partly ideological, a view that policy is another government economicintervention that requires picking winners, and partly because some issues traditionally addressed by industry policy like tariffs and market access moved intonegotiations around trade policy, at both the global level and in the increasing number of regional and bilateral trade agreements.

Following the financial crisis governments looking for sources of economic growth and employment creation began focusing on specific sectors in manufacturingand services where they saw opportunity in global value chains. Environmental standards and policies supporting renewable energy were developed. Industrieslike pharmaceuticals and biotechnology, semiconductors, aerospace, IT, AI, cars and steel have featured in industry policies in many countries. Any policyintervention intended to strengthen the economy is an industry policy, and governments regularly establish priorities and target industries. Countries protect orfavour industries with legislation for many reasons but some of them are strategic and long term, like innovation programs with their associated challenges,roadmaps and milestones, and many of these programs currently involve digitization and automation in some form.

These is little practical difference between a country’s industry policy and national industrial strategy. They are both typically framed around competitiveness andproductivity, focus on innovation and R&D, and follow pathways and roadmaps through scenarios and scoping studies. Some industries like agriculture, steel andautomobiles are regarded as strategic and have always been surrounded by rules and regulations and subject to government intervention. Governments’ havescience and technology policies that influence industrial structure and macroeconomic policies that affect economic development. For many countries theemphasis in industry policy has shifted to industry 4.0 technologies and AI, as governments and industry respond to these technologies.

Government policies like these that target supply side issues are not as high profile as others, they don’t get regular updates like monthly unemployment orquarterly GDP statistics and capture attention like announcements of interest rate changes. Because productivity has become the measure used for industryperformance, despite the statistical questions that raises, it has often been the target for government policy. However, many of these policy measures will onlyaffect productivity in the long run, examples are education, training, infrastructure, innovation, R&D, capital expenditure subsidies, and pilot or demonstrationprojects. Therefore, results take time and thus take longer than the electoral cycle to develop, so there is often little benefit to the government of the day even if apolicy is working well.

When the intention of such policies is to influence a country’s economic structure and industry development they can be described as industrial strategy or industrypolicy. What history generally does show is that it is hard to get an industry strategy right and implementation is difficult. Traditionally manufacturing was thefocus for industry policy, but after 2007 the approach became more about coordinating a wide range of policies to achieve both economic and social objectives.[i]Climate change and environmental issues have become a focus for a range of industry policies aimed at reducing emissions.[ii] The rollout of protectiveequipment and vaccines during the Covid pandemic in 2020-21 both tested and accelerated this new approach.

Construction Industry Policy

As well as common industry policies targeting innovation, training or business investment, construction of the built environment is also subject to many othergovernment regulations, legislation and policies. On the demand side interest rates, taxes, public infrastructure spending, urban development and housing policiesare all important, but are also external to the built environment sector itself and are determined by a wide range of factors beyond the sector. Then there are theeffects of planning and environmental regulations and restrictions limiting the supply of new housing or infrastructure, an issue that has featured in recent debatesin many countries and spills over into other issues around the affordability of housing and the location and cost of major projects. The number of differentgovernment departments and agencies involved in regulating the built environment is often a major barrier to innovation because coordination is difficult and thereare many opportunities for incumbents to delay or derail progress when reforms are proposed.

The public sector in many countries is collectively the largest client for construction, but the expenditure is spread over departments like health, education,transport and defence, and there is unrelenting pressure from public sector clients for the lowest possible cost of work. In practice, there are significant institutionalconstraints on government buying power. Although reports in many countries have recommended leveraging public procurement of buildings and structures topush industry reform this is not widely used, despite being common practice in Asian countries like Singapore and Japan.

While it is a fact that governments can have major impacts through regulation, tax, training,  innovation and R&D policies, their effect is uneven and can be hardto discern. For example, large firms in capital intensive industries like cement respond to industry policies differently to large contractors, as do professionalservice SMEs compared to construction trade SMEs. Two areas where governments have had some success in promoting industry development are discussed inprevious posts: BIM mandates and building standards and codes.

Industry Policy and Industry Culture

Contractual relationships were the focus of much of the reform agenda of the 1990s and 2000s. In the UK the Simon Committee report in 1944 on buildingcontracts called for cultural change, as did the Latham Report 50 years later. Egan in 1998 introduced benchmarking against best practice to improve productivityand Constructing Excellence documented demonstration projects. In their book on UK Construction Reports Murray and Langford thought the ‘demands on theindustry cannot be met and so lead to an industry that cannot attract staff to deliver buildings on time, with increased costs and questionable quality.’[iii] Othercritics attacked the reform movement for its technocratic and managerial approach[iv]and the language used. By 2011, when the new UK industry strategy waslaunched, there had been little change in the industry, clients awarded projects to the lowest bidder while contractors offloaded risks and maximised profits.

That a series of UK reports were required, averaging over two a decade for 50 years (many others were not included in Murray and Langford), shows howineffective they were in developing policies to address the issues raised. The explanation for this policy ineffectiveness offered by Latham and Egan in theirreports was industry culture, broadly seen as the custom and practices underlying the business model in UK construction. Latham focused on procurement andcontractual relations with recommendations to change an adversarial culture, calling for more collaboration between clients, contractors, subcontractors andconsultants, and more cooperative practices. He recommended ‘Partnering’ between clients and contractors to realise this.

Culture is clearly important, but it is also clear that culture is not malleable and does not change easily or quickly. A better explanation for the lack of impact ofthese reports and their recommendations, and the ineffectiveness of public policy in reforming construction is required. Is the problem the policy making process,resistant to evidence and subject to ministerial whims and churn, with issues becoming politicised once they enter public debate? In a technocratic system ofproduction like construction regulatory proposals often lack a convincing evidence base, and can be poorly integrated with impact assessment and policydevelopment processes. The generic ‘problem-inspired’ industrial strategies developed by central policymakers also have to be interpreted by the ‘problem-solving’ implementers responding to nuances of local context and capability.

Construction is better viewed as three industries when the differences between residential building, non-residential building and engineering construction are takeninto account. Within the broad culture of construction they have their own permeable but distinct subcultures, based on differences in processes, products andmarkets. If the culture in each of the three industries is different, recommendations and policy directed at construction as a single industry are unlikely to berelevant across them and will thus be disregarded by many firms and clients. Clients are also different and can be generalised as households, businesses and thepublic sector, and their relationships with contractors varies accordingly. Another example is design, where house builders have pattern books, commercialbuilding uses architects, and infrastructure is designed by engineers.

These structural differences between the three industries affects the way clients, contractors, designers and suppliers will interact, thus each industry has developedindividual characteristics over time within the broader culture of construction that become that particular industry’s subculture. The specific nature of theseindustry subcultures often makes recommendations and policy directed at construction as a single industry ineffective. With separate industries and separatesubcultures, separate policies are required. A broad industry policy of the sort that targets construction as a single industry will be challenged by three deeplyentrenched subcultures with limited, though important, similarities. Research and reports that treat construction as a single industry share this problem.

The UK government mandate on use of BIM on public projects has been much more effective in the last 10 years than the previous six decades of exhortations andrecommendations to change industry culture. Recognising this, the provision of clauses covering contentious issues in construction contracts (such as intellectualproperty and data ownership) worked with rather than against industry practice and culture. The BIM Framework provided a roadmap for the firms and clients andthe development of standards provided a toolkit. Also, local governments, universities, regulators and industry bodies were all given significant but looselyspecified roles in these policies to support industry engagement.

The UK construction strategy applied to all firms involved in public projects, and thus included designers, consultants and suppliers as well as contractors andsubcontractors. The strategy targeted technology adoption not the ‘construction industry’, which is really three separate industries of residential building, non-residential building and engineering construction each with distinctive characteristics.[v] The differences in the subcultures of these separate industries accountsfor the differing rates of uptake of BIM found across firms in the UK since the launch of the strategy.

Industry culture is a complex outcome of social, institutional and economic factors. Because of the range and dynamic interplay of those factors it is not anappropriate target for industry policy, as the history of construction reform efforts that argued cultural change was necessary for industry improvement in the UK,documented over decades in a series of reports, clearly shows. When a new construction strategy was launched in 2011 the focus shifted from using publicprocurement to foster cultural change to requiring BIM on public projects, and over the next decade succeeded in increasing the use of BIM to around half of firmsand the majority of public projects. Despite all the claims made for BIM changing industry culture and increasing collaboration, if it were to come about it wouldbe as a consequence not a cause of industry improvement from the new construction strategy. Recognising this, the provision of clauses covering contentiousissues in construction contracts (such as intellectual property and data ownership) worked with rather than against industry practice and culture.

Another aspect of construction industry culture is that the nature of the work attracts many people with technical skills who use ‘technological thinking’ to findsolutions to the various problems a project will encounter between inception and delivery. Technological thinking is essentially problem-solving through trial anderror. Regardless of which part of construction they work in, for the vast majority of these people there is a great deal of satisfaction in doing this work well,following relevant codes of practice and meeting the required standards. Basing policies to improve industry performance and the quality of buildings ontechnocratic measures like ISO accreditation and BIM use levels works with industry culture.

References

[i] Chang, H-J. and Andreoni, A. 2020. Industrial Policy in the 21st Century, Development and Change, 51(2): 324–351. [ii] Acemoglu, D., P. Aghion, L. Bursztyn, and D. Hemous. 2012. The Environment and Directed Technical Change. American Economic Review. 102(1): 131-166. [iii] Murray, M. and Langford, D. 2003: 7. Construction Reports 1944-98, Oxford: Wiley-Blackwell. [iv] Green, S.D. 2011. Making Sense of Construction Improvement, Oxford: Wiley-Blackwell. [v] Although there is an economic activity called construction in the SIC the characteristics of the three divisions makes them different industries. The manufacturing SIC includes glass, wood products, steel, plastics and concrete, but they are regarded as separate industries and are not grouped together under a construction products SIC. An industry policy for the steel industry is not thought to apply to plastics or concrete because it is not relevant to those industries. The same applies to the differences between residential building, non-residential building and engineering construction.