Preprint: André O. Laplume, Bent Petersen, Joshua M. Pearce, Global value chains from a 3D printing perspective, Journal of
International Business Studies 47(5), 595–609 (2016). doi:10.1057/jibs.2015.47
Global Value Chains from a 3D Printing Perspective
For the JIBS Special Issue on Internationalization in the Information Age
André O. Laplume
School of Business and Economics,
Michigan Technological University
1400 Townsend Drive.
Houghton, Michigan, 49931
906-487-3267
Contact author: aolaplum@mtu.edu
Bent Petersen
Department of Strategic Management and Globalization
Copenhagen Business School
Solbjerg Plads 3, 2000 Frederiksberg, Denmark
Room: KIL/14A-2.59
+45 38152510
bp.smg@cbs.dk
Joshua M. Pearce
Department of Materials Science & Engineering
and Department of Electrical & Computer Engineering,
Michigan Technological University
1400 Townsend Drive.
Houghton, Michigan, 49931
906-487-1466
pearce@mtu.edu
Abstract: This paper outlines the evolution of additive manufacturing technology, culminating in 3D
printing, and presents a vision of how this evolution is affecting existing global value chains in
production. In particular, we bring up questions about how this new technology can affect the geographic
span and density of global value chains. Potentially, wider adoption of this technology has the potential to
partially reverse the trend towards global specialization of production systems into elements that may be
geographically dispersed and closer to the end-users (localization). This leaves the question of whether in
some industries diffusion of 3D printing technologies may change the role of multinational enterprises as
coordinators of global value chains by inducing the engagement of a wider variety of firms, even
households.
Keywords: additive manufacturing; 3D printing; global value chains; geographic span; geographic
density
1
Preprint: André O. Laplume, Bent Petersen, Joshua M. Pearce, Global value chains from a 3D printing perspective, Journal of
International Business Studies 47(5), 595–609 (2016). doi:10.1057/jibs.2015.47
INTRODUCTION
A key capability of multinational enterprises (MNEs) is the organization and effective coordination of
global value chains (GVCs) (Oviatt & McDougall, 1994). However, this view is bumping up against the
information age (Globerman, Roehl, & Standifird, 2001), which may have dramatic consequences for the
configuration of GVCs in terms of their geographic span and density. According to its advocates, 3D
printing now threatens to upend retailers, distributers (middle-men), and manufacturers of tangible goods,
by introducing a new paradigm of industrial production via the layer-by-layer additive construction of 3D
objects from digital designs (Lipson & Kurman, 2013).
3D printing technology has moved beyond its early success as an acceleration of innovation
cycles via rapid prototyping and is now being applied in the manufacture of a wide array of products. The
Economist (2012), and others, suggest that the technology is spawning the next industrial revolution
(Hopkinson, Hague, & Dickens, 2006). Developers of the technology argue that 3D printing provides a
path to sustainable development for low-income countries, and many authors argue that the technology
will lead to a world in which personal fabrication and peer production will replace most industrial
processes (Gershenfeld, 2008; Moilanen & Vadén, 2013; Lipson & Kurman, 2013). Despite its current
diffusion into marginal markets, the potential effects of 3D printing technology can be viewed in various
lights. Proprietary 3D printing and open-source 3D printing are being used in some high-end
manufacturing (e.g., for aircraft-engine components and automobile production), but desktop 3D printing
(e.g., households making household items) also holds promise (Economist, 2012). A few years ago,
desktop 3D printers remained a pursuit of hobbyists, innovators, and early adopters, and had yet to cross
into mainstream applications, possibly owing to the technology’s limitations (e.g., size limits, resolution,
ease of use, speed, and complexity of materials). However, the technology now appears more economical
than more labor-intensive “cut-and-mold” manufacturing techniques (Berman, 2012; Nyman & Sarlin,
2013).
Recent work shows that 3D printing is economically viable for US households (Wittbrodt et al.,
2013). As they adopt the technology, households and 3D print shops may gain a bigger share of potential
industry earnings (Dedrick, Kraemer & Linden, 2010) because 3D printing puts the means of production
back in their hands and undermines some of the complementary asset advantages of MNEs (Dunning,
2001; Ghoshal & Bartlett, 1998). However, it does not do so equally in all industries, rather, as we will
show, the technology’s rise has been steeper in some industries than in others. We theorize that several
dimensions, including the type of materials, the need for customization, and for speedy delivery, and low
cost (for printing complex objects), may be the disruptive drivers of this technology.
The key implication of this technology for international business research is that it has the
potential to reshape GVCs by altering their geographic span and density. On this background the paper
proceeds as follows: First we provide a brief introduction to, and status of, the new manufacturing
technology: its properties and applications. In the next section we discuss the scope of the additive
manufacturing: which industries are fully ‘exposed’ to 3D printing, which only moderately, and which
industries seem immune to the technology. Next follows an account for the GVC phenomenon and its
underlying drivers and impediments. In the next three sections the GVC determinants—factor cost
differentials, scale economies, and factors impeding global specialization—are analyzed from an additive
manufacturing perspective. We juxtapose the various GVC determinants and additive manufacturing in
order to clarify potential implications of the new technology to configuration of production systems,
raising questions for future research. A final section concludes and point out limitations of the study.
BACKGROUND ON ADDITIVE MANUFACTURING
A 3D printer is a device that is able to construct a three-dimensional solid object of any shape from a
digital design. Historically, 3D printing has been referred to as “additive manufacturing” because it uses
an “additive” process in which layers of material in different 2D shapes are successively added. These 2D
2
Preprint: André O. Laplume, Bent Petersen, Joshua M. Pearce, Global value chains from a 3D printing perspective, Journal of
International Business Studies 47(5), 595–609 (2016). doi:10.1057/jibs.2015.47
shapes build upon each other into 3D objects. This distinguishes 3D printing from subtractive methods of
manufacturing in which one starts with a block of material and mills away unnecessary material until the
final shape is obtained.
Some versions of 3D printers have been around since the 1970s, but they were not
commercialized or widely diffused. Improvements in technology (e.g., Hull, 1986) led to the development
of 3D printers that were largely used for rapid prototyping or secondary manufacturing techniques (e.g.,
forming tools for traditional manufacturing techniques, such as injection molding). In the 1980s, 1990s,
and early 2000s, 3D printing evolved within the confines of the R&D departments of a small oligopoly of
firms (e.g., 3D Systems, zCorp, Stratasys, and Objet Geometries), leading to some variations in terms of
resolution, color availability, and time required for printing. These printers cost between USD 20,000 and
USD 300,000 (Bradshaw, Bowyer & Haufe, 2010). Those machines that were used for production in
metal often cost more than USD 500,000. Despite their high costs, these machines were widely adopted,
primarily by firms seeking rapid prototyping capabilities. Prospects for much wider adoption in terms of
manufacturing at the household or local shop level were limited by the high price tag.
1
However, in 2004,
a professor in the UK launched an open-source 3D printer project called the RepRap (self-replicating
rapid prototyper) (Bowyer, 2014; Jones et al., 2011; Sells, Bailard, Smith, Bowyer, 2010). The process
used in RepRap 3D printers is called fused-filament fabrication (FFF). The name refers to essentially the
same process as fused-deposition modeling (FDM), but it is used to avoid trademark infringement. The
goal of the RepRap project is to make a 3D printer that is capable of not only printing various products
but also replicating itself. Recent versions of the RepRap can print approximately 50% of their own parts,
dramatically reducing costs (Pearce, 2015).
Open-source innovation includes more participants than proprietary or closed-source innovation
within firms, and it is less encumbered by intellectual property issues (Chesbrough, 2003; Huizingh,
2011; Yu & Hang, 2011). Thus, the trajectories of improvement are steeper (Foss and Pedersen, 2004)
than they are for traditional manufacturing technologies. Improvements are essentially continuous, as new
designs are published almost daily. While the number of improvements is hard to quantify, Reprap.org
(2012) shows that unique versions of the Darwin were introduced 20 times between 2006 and 2009, 41
times in 2010, and 99 times in 2011. In 2012, 43 unique versions were introduced in Q1 alone. Similarly,
repositories such as Thingiverse, which houses nearly one million free digital designs, are continually
posting new designs for objects that can be printed (Wittbrodt et al., 2013). Rough sales figures suggest
that only around 70,000 low-cost 3D printers were sold prior to 2013, whereas 2013 was on track for
sales of about 145,000 units. This represents a doubling of the total amount in just one year.
2
At the
moment, most designs (for printed objects) are hard to copyright and copyright laws can be bypassed
through the introduction of small changes in the overall design (Bradshaw et al., 2010). Therefore,
Thingiverse, YouMagine, Stanford 3D Scanning Repository, Github, Repables, Pirate Bay Physibles, Fab
Fabbers, Cubehero, Bld3r, Libre 3D, and other repositories of public-domain designs may blossom.
3
Interestingly, one of the goals of the core 3D printing research community (e.g., RepRap) is to
make 3D printers printable and to take control of the machines themselves out of the hands of incumbents
(i.e., by building self-replicating printers). Although this goal has yet to be fully achieved, the technical
potential is developing rapidly. However, some of the technologies needed to build better but still
inexpensive 3D printers are currently being held up by several patents. For instance, laser patents owned
by incumbents make it difficult for researchers to experiment with alternatives to fused filament additive
manufacturing (e.g., laser sintering).
4
When the core FDM patent expired in 2009, the RepRap was quickly followed by hundreds of
derivative innovations created by individuals and companies all over the world. Some of these remained
open-source (like Ultimaker) or at least “accessible source”, such as Type A Machines. Others went
“closed source”, such as the MakerBot. Many of these 3D printing companies that chose to remain open-
source in order to leverage the rapid innovation cycles in the RepRap community have continued the
tradition of designing printable parts. For example, the Lulzbot, a commercialized 3D printer
3
Preprint: André O. Laplume, Bent Petersen, Joshua M. Pearce, Global value chains from a 3D printing perspective, Journal of
International Business Studies 47(5), 595–609 (2016). doi:10.1057/jibs.2015.47
manufactured by Aleph Objects, is made in a factory where hundreds of Lulzbot printers print parts for
future printers. These open-source 3D printers not only provide the mechanical designs for the maker
community to improve upon, but they are also made with open-source electronics (e.g., Arduino
microcontrollers). Similar to open-source software, those who use open-source hardware are expected to
provide the community with information on any improvements based on the open designs. This leads to
rapid innovation and improved machines.
Between 2009 and 2013 the standard RepRap cut the cost of 3D printers to less than USD 1,000,
which was less than one tenth of the cost of the 3D printers provided by the commercial oligarchy at the
time. As they were free to innovate without negotiating licenses or paying royalties, open-source
supporters of low-cost 3D printers exploded onto the scene during these years. Dozens of companies,
offering different versions of open-source printers, appeared. Many of them received funding from
crowdsourcing websites, such as Kickstarter and Indiegogo. This intensive fermentation period allowed
the technology to improve rapidly from year to year, and the cost of a basic machine fell to just a few
hundred dollars, although assembly is sometimes required. In general, commercial proprietary 3D printers
are limited in terms of applicability in order to maintain quality and reliability, while the RepRap and
other open-source 3D printers are more flexible, as shown in Table 1.
-------------------------------------------
Insert Table 1 about here
-------------------------------------------
Notably, although both conventional and RepRap 3D printers print in a variety of materials,
including ceramics and metal, the vast majority are limited to printing in plastic. With RepRap driving
down the cost of 3D printing, the major firms now also offer lower-cost (i.e., less than USD 2,000) 3D
printers capable of printing in plastic in a limited fashion (e.g., only in one plastic, which the company
makes available in a cartridge). Examples of recent improvements in open-sourced, low-cost 3D printers
include: printing using a greater variety of materials, such as metals (Anzalone, Zhang, Wijnen, Sanders,
& Pearce, 2013) and conductive materials; simultaneous printing with more than one type of material;
printing with multiple printer heads (Anzalone, Wijnen & Pearce, 2015), which allows for higher
volumes; improved resolution; greater area or volume; and ease-of-use improvements aimed at making
assembly, maintenance, and use accessible to consumers.
The above account for additive manufacturing does not establish whether or not this technology
should be considered “disruptive” or a “general purpose technology” understood as a new method of
production with a protracted aggregate impact (Jovanovic and Rousseau, 2005). Examples of such
technologies include electricity and information technology. The degree of “disruption” depends not least
on the extent to which the new technology changes the affordability of the products produced (and
thereby the consumption pattern) but also on the extent to which the new technology entails a change in
the composition of factor inputs (Danneels, 2004; Tushman & Anderson, 1986). Hence, in the cases
where 3D printing substitutes for labor-intensive manufacturing processes this certainly implies a
disruption in the industry in question. However, in cases where the substituted manufacturing processes
are already highly automated the factor inputs (with low labor-capital ratios) basically remain the same.
So, when we juxtapose 3D printing and “traditional” or “conventional” manufacturing technology this is,
in practice, an overgeneralization. In a way, we may consider 3D printing as a variety of automated
manufacturing and, as such, we implicitly assume “conventional” and “traditional” manufacturing
technology to be less automated and more labor-intensive. The other issue, whether additive
manufacturing is an emerging “general purpose technology” depends foremost on its expected diffusion
in the various manufacturing industries. We look at this in the next section.
DIFFUSION OF 3D PRINTING TECHNOLOGY IN DIFFERENT INDUSTRIES
4
Preprint: André O. Laplume, Bent Petersen, Joshua M. Pearce, Global value chains from a 3D printing perspective, Journal of
International Business Studies 47(5), 595–609 (2016). doi:10.1057/jibs.2015.47
This section is intended to give a sense of the diffusion of the 3D printing technology in different
manufacturing industries and is thus necessary coarse-grained. Although the construction industry may
also be affected by this technology, with several innovators printing houses and buildings with concrete,
the manufacturing industry is more relevant for global value chains and hence is the focus here.
Based on their independent assessments of the current and future diffusion of 3D printing
technology the authors coded each of the industries listed in the manufacturing sector according to the
International Standard Industrial Classifications (ISIC). We then compared response explanations and
recoded based on discussions. The resultant coding was used to categorize the industries into the four
quadrants of Table 2.
-------------------------------------------
Insert Table 2 about here
-------------------------------------------
The following subsections identify and explain (1) industries with no, or low, current or future adoption
of the 3D printing technology, (2) industries currently being affected by the technology, and (3) industries
into which the new technology is likely to diffuse in the future.
Unaffected Industries
Some industries are unlikely to be affected by the 3D printing technology—neither today, nor in the
future. Products that are made of natural materials (e.g., solid wood, cork, leather, natural textiles, paper,
and tobacco products) are largely unsuitable as filament for 3D printing and therefore unlikely to be
affected (although there are already several types of wood powder based filaments). These products tend
to be desired in part due to their natural properties such as tensile strength, grain, or texture and, therefore
are less likely to be affected.
Another significant area that is unlikely to be directly affected by the technology is the production
of most industrial raw materials (e.g., petroleum products, and basic metals). The production of the tools
to harvest these materials may be 3-D printed and these materials can become the filament for 3-D
printing, but cannot be directly printed. Similarly, industries that break down or fragment materials are
unlikely to be affected by the technology.
Industries Currently Being Affected
Simple products are currently most affected by 3D printing technology. These products tend to be small in
size, made of just one material, and do not have many interacting parts, making them ideal candidates for
low-cost 3D printing. Hence, the manufacture of jewelry, musical instruments, sports goods and toys, and
medical instruments, has to a large extent already adopted the 3D printing technology. For instance, toys
and games are so popular that they warrant their own section in the storefront of the largest repository
(Thingiverse). Most of the open-source printers currently on the market print with plastic and the vast
majority of the designs currently available for download on repositories (e.g., chess pieces, construction
toys, dice, games, mechanical toys, playsets and puzzles) are intended to be printed in plastic. In general,
those made of a single raw material such as plastic, ceramics, or metals, are already feasible. Both the
manufacture and repair of machinery and equipment is also currently being affected by 3D printing
technology. 3D printing replacement parts for machines is becoming commonplace. However, the best
example of this trend is the advancement of the printing of 3D printers themselves. As earlier mentioned,
many of these machines are designed so that their parts can be printed.
Industries to be Affected in the Future
Some industries are currently minimally affected, but are likely to adopt the 3D printing technology in the
future. Due to space limitations, the focus here will be on four exemplary industries (foodstuff, wearing
apparel, automobiles, and medicine), which show the breadth of application of the technology.
5
