CAN CHINA MAKE 12 million vehicles a year by 2020 - up from about four million now? Chinese automotive officials predicted that this week. The sceptic splutters. Fair enough, vehicle output and sales have risen far faster than almost anyone expected this year, on the back of some nifty moves in car financing. But, with no supporting evidence, projecting where an industry in China will be in 17 years is little more than a wild guess based on extrapolation. Or is it? While I suspect this claim is little more than economic boosting of the kind that has characterised China's socialist and proto-capitalist eras, there are also changes taking place which make life harder for the sceptic, certainly compared with even just two years ago, let alone five. The one that strikes me most forcefully is a change in the way western companies are looking at the country. In the 1990s, and still to a large extent today, the mantra which drove foreign business into China was the potential market of its 1.3 billion people. While even at the time it was largely accepted that this market was mythical, it was still the lure which pulled companies in. This is giving way, certainly among manufacturing companies, to a view of China as being part of a global economy in which it is a market, but also - and more importantly - in which it is a key part of the global supply chain. This was true from the start for Hong Kong and Taiwan firms, and for many of the world's biggest multinationals. But they were followed by Japanese and South Korean firms, and now are being joined by ever greater numbers of medium-sized United States and European companies. A lot of these companies are suppliers to other companies, and so have to follow their customers. When one big producer moves to China, a lot of others follow. Others have moved out of fear; competition at home has meant moving production to the world's most efficient manufacturing centre. This movement of manufacturing to China, apart from pushing up its export earnings, is also raising productivity - and increasing productivity, as any economist will tell you, is what lies at the heart of economic growth. There is little doubt that this increased productivity in China is strongest in the foreign-invested sector. But it is quickly being adopted by domestic firms - all of China's huge consumer electronics makers, to take just one sector, have simply adopted highly productive overseas manufacturing processes and put them to work producing their own goods. These companies, such as Haier, Changhong and TCL, have done so with wonderful efficiency, making life tough for foreign manufacturers of those products. They are also now starting to take on foreign companies overseas. From China's point of view, this is all a virtuous circle. It even has the benefit of sidestepping one of its own nastiest problems - its precarious banking system which, aside from its problem of non-performing loans, has yet to learn how to price risk and allocate capital efficiently. Instead, this task is largely handled by the allocators of foreign capital. But this points to the real challenge for Chinese banks and companies: how to start getting the allocation of capital right in China? This means looking for companies which will produce a decent return on capital invested. Most domestic firms - however successful at home - are not judged by this criterion. Instead, they have to produce a lot of goods to produce cash flow. When they are judged by their return on capital - and I would imagine we are looking at such a prospect in about 10 years - then these companies will be forced to look at other ways of selling their goods than by simply selling the cheapest. This will call for research, which in turn will call for money. The only way this will become possible is through the development of China's banking system to a stage where it can allocate capital efficiently. It is here that foreign companies will maintain their edge. Maybe in the short term they will lose out, but companies with access to funding and an ability to produce new goods will hold them in good stead when the financial screws start to tighten on Chinese firms. Does this add up to a vehicle industry trebling output in 17 years? Who knows - but the question has less to do with the volume than the timescale, and who exactly will be doing the producing. Asking instead whether China will be part of a global automotive supply chain, manufacturing components and vehicles for sale in China and overseas, then the answer is undoubtedly yes. Will the companies running this supply chain be domestic or foreign? The answer to that question depends on whether you believe China can develop a global carmaker over the same period, bearing in mind that cars are a lot more complex than fridges, not just to make, but to sell, not just in China, but abroad.