How the increasing trend to rent factories is challenging economic measurement

By David Nguyen

Blog post for paper “No Plant No Problem: Factoryless Manufacturing and Economic Measurement” By Diane Coyle and David Nguyen.

Did you know that Mercedes does not actually produce its heavy-duty G-Class? To be fair, it does keep design, development and marketing of the SUV in-house, but the vehicle is entirely built in the factory of Magna Steyr, a contract manufacturer based in Graz, Austria. In the same plant one will also find entire production lines for the Jaguar I-Pace and E-Pace, as well as BMW’s Series 5. They are all produced as part of contract manufacturing arrangements and subsequently shipped to their final owners, which could be anywhere around the globe.

The most famous example of this type of ‘manufacturer’ is of course Apple. It famously engraves the iPhone as being “designed in California”, while the actual production of parts and assembly of the final product is outsourced along a complex global supply chain. In the UK, Dyson follows a similar model as it does not actually produce any of its high-level household electronics and appliances.

Companies can have different strategic reasons for using contract manufacturers to produce their final products. Some of the key considerations are saving on costs for large upfront capital investments in machinery and equipment and the ability to produce close to important markets. However, there are also risks arising from being dependent on the contract manufacturer to deliver products on time and with consistent quality.

In a recently released paper we take a closer look at this type of manufacturing model, as they present potentially severe challenges to existing economic statistics. We review existing definitions and show conceptually that factoryless manufacturing (i.e. the use of a plant owned by someone else against a fee) can lead to significant measurement issues when allocating economic activity across firms, sectors, and also countries.

In the UK around 6.7% of total revenues in the service sector come from production activities (ONS, 2018). While it is not unusual for some service companies to have some production activities as well, it could be that some of these are in fact factoryless manufacturers. We review the UK’s statistics on trade in services (incl. merchanting, manufacturing services, and R&D services) and discuss whether they are likely to capture some aspects of factoryless manufacturing. We conclude that while some of it might be recorded it is not clear where, and there is most likely some degree of underreporting.

The allocation of economic output between manufacturing and services sectors should not affect total output (though there could in principle be reallocation across economies). It is an important issue, however, as the ‘decline of manufacturing’ across industrialised economies has received considerable attention from politicians, media and the general public. Studies for the UK show that the total number of manufacturing jobs has declined from 7.9 million in 1971 to around 2.7 million in 2016. Studies for the US economy have shown that reclassifying factoryless manufacturers from services to manufacturing can lead to a significant increase in manufacturing output and employment (Bernard & Fort, 2015).

In our paper we also provide case in-depth studies on the wide-spread use of contract manufacturing the automotive and pharmaceutical sectors, which appear to be intensive users of contract manufacturing services. We also construct novel measures of contract manufacturing for the UK and US using information scraped from the universe of company websites in both economies. An intelligent text processing software allows us to look for topics related to contract manufacturing if businesses discuss, e.g. in the ‘About’ section. We find that in the UK the use of contract manufacturing appears to be mainly used by companies in sectors such as chemicals, pharmaceuticals, biotechnology, and electronics. In the US it appears to be more concentrated in electronics, pharmaceuticals, plastics and medical devices.

Assembly line production at a car manufacturing plant

To illustrate this in more detail let us return to the example of the premium vehicles produced in Austria by the Canadian-owned Magna Steyr. Technically, the contract manufacturer is providing a ‘manufacturing service’ for Germany’s Mercedes-Benz or UK-based Jaguar Land Rover (owned by Tata), who in turn provide it with the necessary design, development and other intellectual property to produce the final vehicle. It is easy to see the formidable challenge this poses to working out the international transfers of inputs of goods and services, intellectual property, ownership rights and final products. At this point it is not entirely clear what type of economic activity is recorded in which country and whether we are dealing with a trade in goods or services.

We recommend that:

  • National Statistical Offices need to start collecting data on the use of contract manufacturing services across sectors, including if they are traded;
  • There is a need to agree on a consistent and unambiguous definition and terminology to describe factoryless manufacturing;
  • As this is a fundamentally global phenomenon, there is a clear need to coordinate at the international level economic measurement at the global level (i.e. involving OECD, WTO, EC, UNECE).

At this point we do not know how many UK businesses are classified within services despite the possibility that they could be reclassified into manufacturing. Our study shows that more research is needed in this area as the issue is highly relevant in the context of a declining manufacturing sector which we can observe in most advanced economies.

David Nguyen (@d_vidnguyen) is a Research Associate at ESCoE and a Senior Economist in the Trade, Investment & Productivity directorate at NIESR.

A full version of this paper is available here.

 


ESCoE blogs are published to further debate.  Any views expressed are solely those of the author(s) and so cannot be taken to represent those of the ESCoE, its partner institutions or the Office for National Statistics.

Friday, September 13, 2019

TwitterFacebookLinkedInEmailPrint