To innovate in fashion, don’t reshore – omnishore!

After a long and strong wave of offshoring – manufacturers moving their factories abroad, often to low-cost countries such as China – limits to that model have emerged, especially in the fashion industry. These include hidden costs in logistics, quality and in coordination, as well as counterfeiting, negative impact on the environment and poor working conditions. Trends are also changing more rapidly, making long supply chains problematic.

The fashion industry is a particularly interesting case as it’s labour-intensive – particularly the stitching, which has not yet been automated – and has offshored substantially to lower costs. Yet it’s also an industry in which clusters exist and flourish – Prato in Italy, or the garment centres in London, New York and Paris.

A study from Boston Consulting Group shows a reduction in the gap between labour costs in China and in the United States because of an increase in Chinese labour costs and rising productivity in the United States. It therefore suggests in numerous industries, there may be a wave of reshoring – bringing manufacturing back after it was offshored. Reshoring is also becoming a buzzword in Europe, and the benefits of manufacturing for European countries are clear.

While offshoring can cut some costs, it has been shown to reduce firms’ capacity for innovation. Researchers from MIT (Pisano and Shih, 2012) propose a framework that allows industries to be classified based on their innovation type – pure product innovation, pure process innovation, process-embedded innovation, and process-driven innovation. Fashion is considered a process-embedded innovation industry, which means that design and manufacturing have to be co-located for firms to sustain their capacity to innovate. As a consequence, the recommendation from Pisano and Shih is to co-locate the two activities in the country of origin. Innovation is presented as an additional argument in favour of reshoring.

Digitization, automation and the industry 4.0 technologies are also presented as changing the manufacturing landscape in fashion, encouraging closer manufacturing locations.

A more complex picture: omnishoring

An in-depth analysis of 20 European fashion companies (realized by Celine Abecassis-Moedas and Valerie Moatti within the ESCP Europe Lectra “Fashion and Technology” Chair) shows a more complex picture.

First, European firms perceive sourcing countries differently depending on whether they’re distant (in Asia) or close (in Europe, Turkey or North Africa), rather than the usual dichotomy of local versus distant sourcing. Manufacturing in the home country is virtually non-existent. None of the companies examined reshored from far-sourcing countries to the home country, but quite a few shifted to close-sourcing places – nearshoring. The majority of firms are manufacturing more than half of their volumes in close-sourcing destinations.

All firms examined manage a complex portfolio of sourcing locations, either allocating certain products to particular locations (jeans in Turkey for instance) or allocating same products to different sourcing locations depending on the product life cycle – close in smaller series at the beginning of the season, then distant in large quantities, and again close for replenishment at the end of the season. This is what we call “omnishoring”, i.e., a complex portfolio of multiple sourcing (close and far) according to products and their degree of innovation and complementing each other.

Strategies to coordinate design and manufacturing and manage distance

On the other hand, the need to coordinate design and manufacturing when not co-located essential, but firms use a variety of strategies beyond co-location in the home country. Some have developed a reverse co-location, in which firms move design close to manufacturing (rather than the other way around). For example, some large players in the European fashion industry have moved their design department (part or all) to Hong Kong or China. Another coordination mechanism is to locate prototyping – the early step of manufacturing, considered critical for the innovation process – close to design, and the rest of the manufacturing phases can be distant.

Some firms manufacture in distant countries that have some cultural proximity – for example, British firms can choose Thailand because of a greater level of English fluency. Others manufacture in distant locations in their own facilities or through strong partnerships, which allows for greater control and a stronger coordination. Finally, the use of manufacturing and communication technologies (PLM, 3D design…), intermediaries, or regular and systematic trips of designers to manufacturing plants is also a strong coordination mechanism.

In conclusion, our analysis shows a more subtle approach of what distance is, beyond its purely geographical dimension, and therefore of how it is managed. Four strategies to manage distance between design and manufacturing emerged:

  • Avoiding distance by co-location or reverse co-location.

  • Segmenting geographical distance, by segmenting manufacturing between prototyping and product development, and bulk manufacturing.

  • Hedging against geographical distance through other forms of proximity, such as cultural or institutional proximity (manufacturing in distant locations in a firm’s own plants).

  • Facilitating geographical distance, through the use of technologies.

The term omnishoring reveals a rich portfolio of sourcing destinations (close and far) and of institutional models (a firm’s own facilities and subcontracted ones) that is definitely complex and allows firms to manage different products, seasons, series and more.

This study of the fashion industry shows that the need for co-location between design and manufacturing can be managed in different ways. These strategies are also food for thought for other industries that want to manufacture outside of their home country without sacrificing innovation.

Leave a Reply

Your email address will not be published. Required fields are marked *