China Daily ran an article on the continuing urbanization of Beijing. In Build upward or outward: City’s growth dilemma, Daniel Garst notes that Beijing is not as centralized as other urban areas, with its multiple business districts and comparatively low density in its inner areas. He indicates a preference for the urbanization of Shanghai, with its stronger center (both Pudong and Puxi), but suggests that it would be a mistake to replace the historic low density development with the high rises that would be necessary to change Beijing's urban form.
Actually, Beijing's form is not that unusual for Asian urban areas. Tokyo has multiple office centers rather than a single dominant center and has comparatively low residential densities, even within the Yamanote Loop. Bangkok, Manila and Jakarta are similarly multi-centric. Chinese urban areas like Shenyang, Xi'an, Wuhan, Suzhou and Changsha are closer (but smaller) replicas of Beijing than Shanghai. Garst also misunderstands the dynamics of traffic congestion in his belief that roads and metros (subways) would be less congested with a more centralized form. In fact, higher densities routinely produce more intense congestion, not only on the roads but also on the rails and buses, a point recently made by Michael Matusik on this site.
However, Garst may be onto something with respect to a suggestion that Beijing's growth should be directed to new satellite towns, in which residents work rather than commuting to Beijing. This is good theory, but there is an important caveat, which we outlined in a comment at China Daily on the article.
Satellite cities are not a reasonable answer unless they are so far from the Beijing urban area that commuting to Beijing is not possible. The idea of self-contained satellite cities, where people live and work in them has not worked anywhere. There are good examples of failure in London, Cairo, Stockholm, etc. So long as the large urban area can be reached, people will commute there.
Cairo provides a useful example. Egyptian planners have long decried the continuing commute pattern into the urban area from the new towns of 6th of October and 10th of Ramadan, which are within commuting distance. On the other hand, the new town of Anwar Sadat, more remote from the urban area, has been more successful in keeping its residents in its labor market.
Locating new satellite towns far enough to make commuting infeasible will be a real problem for Beijing. There just is not enough territory in the provincial level municipality. That means the new towns would have to be in the province Hebei, which along with the province level municipality of Tianjin surrounds Beijing.
Short of remote new towns and forcing population and economic growth away from Beijing, the key to minimizing traffic congestion will be to minimize work trip distances by achieving a dispersion of comparatively lower density employment to match the lower density suburban dispersion. Economists Peter Gordon and Harry W. Richardson have found that "suburbanization has been the dominant and successful mechanism for reducing congestion." in the United States. This applies no less to Beijing.
Photograph: Forbidden City, Beijing (by author)