.
Comparative advantage
.
The boomerang effect
.
As Chinese wages rise, some production is moving back to the rich world
.
Apr 21st 2012
THIRTY YEARS AGO Shenzhen was little more than a village, abutting the border of Hong Kong’s New Territories. When China’s first Special Economic Zone was established in the early 1980s, workshops started to grow and glistening skyscrapers began to rise up. Its population is now around 12m, including perhaps 6m migrant workers. They often live in dormitories close to the factories that have helped make this city one of the richest in China.
.
Countries that make things more cheaply than others are often accused of running sweatshops, and labour in China was undoubtedly cheap: that was why Hong Kong’s clothing and toy factories moved to the mainland. But with increasing prosperity Chinese workers want more pay, shorter hours and more benefits, just as Taiwanese, Japanese and South Korean workers did before them. Labour costs in China have recently been growing by around 20% a year.
.
Some labour-intensive businesses are now moving from the coastal regions to inland China, where costs are lower, though the infrastructure may not be up to the mark. A number of firms, especially those making clothes and shoes, have upped sticks and moved to Bangladesh, Cambodia, Indonesia and Vietnam. Nike, for instance, used to make most of its trainers in China, but many of its big suppliers have moved elsewhere, and in 2010 Vietnam became the company’s biggest production base worldwide. Unless some way of making shoes and clothing without manual labour emerges (which, as this report will suggest later, is entirely possible), these businesses will move again in the future; Myanmar looks tempting, provided that reforms there continue.
.
Yet for some manufacturers low wage costs are becoming less important because labour represents only a small part of the overall cost of making and selling their products. Researchers for the Personal Computing Industry Centre at the University of California, Irvine, took apart an iPad and worked out where all the various bits inside came from and what it had cost to make and assemble them (see chart 3). They found that a 16-gigabyte 2010 iPad priced at $499 contained $154-worth of materials and parts from American, Japanese, South Korean and European suppliers (Apple has more than 150 suppliers in all, many of which also make or finish their parts in China). The researchers estimated the total worldwide labour costs for the iPad at $33, of which China’s share was just $8. Apple is constantly tweaking its products so the figures shift all the time, but not by much.
.
If China accounts for such a small share of the overall labour costs, surely Apple could afford to make iPads in America? It turns out that low wages are not the only attraction. What Shenzhen has to offer on top is 30 years’ experience of producing electronics. It has a network of firms with sophisticated supply chains, multiple design and engineering skills, intimate knowledge of their production processes and the willingness to leap into action if asked to scale up production.
What Shenzhen provides, in other words, is a successful industrial cluster. It works for Apple because many of the electronic parts it uses are commodities. The real innovation lies in designing the product and creating smart software, which is the speciality of another successful cluster, in Silicon Valley, where Apple is based.
.Where China scores
.
Li & Fung, a Hong Kong firm that helps companies find suppliers in Asia, says in a recent research report that clusters like Shenzhen are “an integral part of China’s international competence in manufacturing”. It counts more than 100 industrial clusters in China—including one, in Zhuji in Zhejiang province, that just makes socks. It consists of more than 3,000 small and medium-sized enterprises in the production chain for socks. As long as China’s clusters maintain their edge, these jobs, whether producing iPads or socks, will not go back to America or Europe.
Yet some jobs are returning to developed countries. With Chinese wage costs rising, America’s productivity improvements can help tip the balance, especially when American firms invest in more automation. Yet robots can be used anywhere to reduce labour costs. For example, Terry Gou, Hon Hai’s boss, says he is planning to use more robots for assembly work in China. He is also setting up factories in some of the inland provinces.
..
Again, wage costs are not the only consideration in transferring production from China back to America. Chesapeake Bay Candle used to ship its scented candles for the American market from China, and then from Vietnam when America raised import tariffs on Chinese-made candles. In June 2011 the company opened a highly automated factory near its base in Maryland, partly because of rising labour costs in Asia and increased shipping charges, but also because having a research and development facility in the American factory allows the company to respond to new trends much faster.
.
.
The candle-maker is keeping its factory in China to serve the vast domestic market there. Many firms are adopting this “China plus one” strategy, usually putting an additional production base in a lower-cost country in Asia. The idea is now being extended to repatriating manufacturing facilities to rich countries. This also saves companies from having all their eggs in one basket. A string of natural disasters in recent years has shown that lean supply chains can snap all too easily.
For Peerless AV, a company based in Aurora, Illinois, moving production back from China began with worries about protecting its intellectual property. Peerless makes metal brackets and stands for all sorts of televisions, ranging from screens hung in offices to information displays at railway stations and the giant “video walls” used at music and sporting events. To make lighter, better-looking supports for the thinner screens it saw coming, the company decided in 2002 to produce a range made from aluminium instead of steel. Unable to find an American firm to supply suitable extrusions and castings at the right price, it turned to China. As the flat-screen boom took hold, sales soared—but then the company began to find copies of its products turning up all over the world.
..
It was these knock-offs that led to a decision to bring production back to America, says Mike Campagna, the firm’s president. Other benefits were to follow. By chance the car industry had gone into a slump and the company was able to pick up the manufacturing equipment it needed at low cost. It also managed to track down people with production experience. For the first time since its launch in 1941, the firm took on debt: $20m-worth to build and equip a new factory, which opened in 2010 to house all its operations under one roof.
“The total cost of manufacturing in China is not as cheap as it might appear to be,” says Mr Campagna. Shipping costs have been rising, containers are expensive and staff have to be maintained in both countries to manage the operation. It is also difficult to react quickly if the market changes.
.
Typically there would be 30 days or so of inventory at each stage of the supply chain: the stock held by the suppliers to the Chinese factory, that factory’s inventory, the content of a shipping container on its way to America, and so on. A design change could take at least six months to implement. Now the company can get a prototype to a customer in a couple of weeks.
Mr Campagna would be happier if the economy were brighter, but says that making 95% of its products in America instead of 65% has transformed the firm’s business. The company used to have 250 workers in America and 400 in China; now it has 350 in America and robots doing hot and dirty jobs, like pouring molten aluminium and laser-cutting steel. The new arrangement, Mr Campagna reckons, “makes us very nimble”.
.
That not only speeds up the production of customised brackets, it also helps with the standard stuff. The company’s standard products used to have a ten-year life cycle, but with new televisions appearing at an ever faster rate its stands and brackets now need replacing every 18 months or so.
..
Sunshine and silicon
.
.
Can repatriation work for commoditised goods too? Until a decade or so ago most of the world’s solar panels were made by American, European and Japanese firms. Then Chinese manufacturers piled into the business, helped by various government-backed incentives. China has now captured more than half the world market for the most widely used solar panels, which rely on photovoltaic cells made from crystalline silicon. But that could change again.
.
Partly because of China’s onslaught, the bottom dropped out of the market: the price of silicon-based solar panels fell from $1.80 per watt at the start of 2011 to 90 cents by the end of the year, according to GTM Research, a market-research firm. This clobbered some firms that used different solar technologies. One of those casualties was Solyndra, a Californian firm, which manufactured photovoltaic panels in the form of thin-film coatings inside arrays of transparent tubes. Although more expensive than the silicon-based panels, the tubes were able to capture sunlight more effectively at different angles throughout the day. But Solyndra could not compete against the glut of Chinese panels. It filed for Chapter 11 bankruptcy last year, despite having (controversially) received $535m in federal loan guarantees.
.
The solar-panel producers are slogging it out, often losing money, in anticipation of a huge market to come when solar panels reach “grid parity”—that is, the ability to match fossil fuels in supplying power to national grids without subsidy. Zhengrong Shi, the boss of China’s Suntech Power, which has become the world’s biggest producer of solar panels, thinks that the market is now showing signs of picking up and that China could attain grid parity within three or four years.
.What chance, then, for solar-panel producers in Europe and America? For a start, it is not an all-or-nothing choice. To make a solar panel, the silicon is cut into wafers onto which photovoltaic cells are fabricated. The cells are then wired up, encased in frames and covered with glass.
.
Turning the cells into panels might be done more economically in the country where they will be used to save on shipping costs. And fitting the panels to buildings, which accounts for most of the cost of putting in solar power, is always going to be a local business. The installation price in America is currently around $6.50 per watt for a house.
.
So Western firms could import solar cells from China and make a good living installing them. But there are manufacturing advances in the pipeline that might level the cost of producing silicon-based cells in America and China, says Tonio Buonassisi, head of the Photovoltaic Research Laboratory at MIT.
.
.
.It is possible to work out from publicly available data that the cost of making a complete solar panel in America is around 25% higher than making it in China and shipping it to the west coast of America. Much of China’s cost advantage is thought to come from cheaper raw materials, lower wages and the lower cost of capital. Doug Powell, a researcher at the Photovoltaic Research Laboratory, is undertaking a detailed analysis of production costs in both countries. After factoring in the manufacturing advances already in the pipeline, the cost of an American-made solar panel will fall by more than half to around 50 cents per watt within a decade (see chart 4). Solar panels that can be made for 40-75 cents per watt are expected to provide grid parity in America. The variation reflects regional differences in the amount of sunshine and the price of electricity.
.
There is nothing to stop China from adopting the same manufacturing breakthroughs, and Mr Powell is investigating the effects of that too. But it is already clear that many of the production innovations now under way would chip away at China’s advantages. For instance, new production methods involve thinner wafers, reducing the amount of silicon required. Cells will become more efficient, simplified production will reduce capital costs and more automation will cut labour costs. “You only really need one breakthrough in each area of innovation to work and we are back in business,” says Mr Buonassisi.
.
Although Solyndra and others have stumbled, the thin-film technology they used remains attractive. GE, for one, is betting on it. As part of a $600m investment in solar businesses it is completing America’s biggest solar-panel factory near Denver, Colorado. It will use thin-film technology to make larger and lighter panels which it reckons will cut installation costs by about half. Employing just 350 people, the GE facility will be capable of producing enough panels every year to power around 80,000 homes.
