4.2 Avoid the emergence of over-crowded mega-cities

4.14 As we have indicated, national income would increase and rural-urban inequality decrease if barriers to migration are further relaxed, encouraging rural-urban migration and urban agglomeration. However, that still leaves the problem of fiscal and capital-market favoritism which could then induce excessive in-migration to provincial-level cities. In China in this decade two cities have been heavily favored: Beijing and Chongching. From the last column of Table 5, the annual population growth rates of cities and towns in these two metropolitan areas are still 4-5% a year, despite their size. However the sources of growth are very different. For Chongching it is internal migration as local rural residents move to cities and towns; for Beijing it is migrants from outside the province. In fact even Shanghai still faces very high in-migration rates from people outside the province. As Table 5 also indicates arriving at good growth numbers is hard. The whole province can be losing population while urban areas grow at high rates; and official numbers from the UN (based on information provided by China) mask the reality. Because of favouritism, a few urban areas in China now may face the prospect of over-population.

4.15 As noted earlier, one solution which encourages migration and urbanization overall, but limits cross-province migration, is to permit unfettered migration within but not across provinces. But in the end, more general market reforms would help avoid the emergence of over-populated mega-cities. In particular would be the opening of the entire capital market to free competition for financing across all firms, cities, and the rural sector, so those further down the hierarchy will be better able to attract investment and migrants. Such a reform also promotes sustained economic growth by ensuring capital goes to the highest return ventures, not just politically favoured ones. Reform of capital markets to create an integrated national market complements labor market reform and is an integral part of integration of the urban and rural sectors.

4.16 A related issue concerns policies that affect the size distribution of cities. For example, the McKinsey Global Institute in 2008 looking to China over the next decades advocates allocating much of the urban population to super-size “cities” (well over 15-20 million).⁴ The evidence discussed earlier suggests that there is a very limited role for super-size cities such as Tokyo or New York in a country. And the difficulties in managing super-size cities without strong institutions and well trained urban managers are profound. It seems economic growth in China would be better served by focusing on development of cities in, say the 1 million to 5 million size range (recognizing that many smaller cities in China may be under-sized).

4.3 Promote efficient use of natural resources

4.17 There are many implications of urbanization for natural resource use and the environment. The focus in this report is on food security and land use, but other issues are important. For example, there are the energy needs of industrialization and the energy demands that come with rising urban incomes (for automobiles, appliances, air conditioning, and heating in the winter) that can result in environmental degradation, absent appropriate policies. Reportedly, there are water shortages in the north of China. However, urbanization and the concomitant advances in agricultural productivity can conserve water. Modernized agriculture with better irrigation, conservation, and mechanization can reduce water usage in agriculture. Consider the high agricultural productivity of Israel despite its poor water resources. And we know from Singapore and other international experience that urbanization and high population density make water treatment and recycling cheaper, and make universal metering feasible, encouraging individual conservation.

Achieve food security

4.18 Usually, urbanization is associated with improvements in agricultural productivity and food security. The most profound change is in the use of agricultural land. Rural-urban migration transforms agriculture from peasant farming on small plots to modern farming, where farm businesses assemble large tracts of land and invest in land improvements, mechanization, and human capital. Better-educated farmers are better equipped to adopt new technologies, make better crop and input choices, and understand market conditions. As an example of what can be achieved concerns Korea as noted earlier. The Korean achievement stemmed from technological advance, education of farmers, consolidation of land holdings and mechanization. China could replicate this success with huge reductions in the farm population and even some loss of agricultural land.

4.19 To improve food production and security, two key reforms were noted earlier: improve rural education and reform the financial sector to allow more investment in agriculture, as would be consistent with the 2008 general policy announcement of the CCP Central Committee and earlier directives issued by the State Council. Here we focus on reforms, some underway or on the table and others not, governing the use of agricultural land itself, in particular strengthening property rights in the rural sector. These three reforms together would raise farm incomes to help close the urban–rural income gap and also encourage some talented rural youth to choose farming as an occupation, rather than migrate. In the next section, we turn to reforms that would discourage any excessive urban expansion into agricultural areas.

4.20 Strengthen rural property rights Lengthening agricultural leaseholds to 70 years or more (from the current 30) would provide stronger incentives to invest in land improvements and machinery. Leaseholds need to be fully transferable and saleable, so as to encourage investment in agriculture and assemblage of land into larger holdings, raising the scale of farm businesses. A current problem is that, while transfer is possible if approved at the local level and experiments exist to trade leaseholds of farm and rural residential land for urban hukou, in many locations migrants are still strongly discouraged from disposing of their leaseholds when they move to cities. When such leaseholds are, instead, rented out, those renting have less financial incentive to invest in the land. The retention of rural leaseholds by migrants to cities has been viewed as some form of economic security for them, as if migrants could return to farming if cities faced some kind of crisis. This is an illusion as is increasingly recognized: sustained urbanization of the type China has experienced over the last 25 years will not be reversed. A migrant who left a village as a teenager and returned 15 years later would not know how to farm. In fact evidence from the recent economic downturn indicates migrants who “went home” to villages typically quickly returned to the city (Cai and Chan, 2009). There is no personal security from the forced retention of these leaseholds – and no food security for China. And, as noted earlier, portability of rural wealth helps the integration into urban society of those who migrate.

Encourage efficient land use and reduce urban sprawl

4.21 There are two aspects to urban land-use policy: (a) policies allocating land between the urban and rural sector, so as to use land efficiently, thus providing the correct incentives to conserve agricultural land and green space; and (b) policies governing the land internal to the city, so as to avoid disordered use and unhealthy, poorly serviced neighborhoods. Reforms involve both strengthening property rights of individual holders and better regulation of land usage patterns.

4.22 There is a popular worldwide concern about ‘urban sprawl,’ the ‘over dependence’ on the automobile, and the removal of farm land as cities expand. Yet the incentives faced by city mayors in China encourage the taking of farm land for urban use and the dependence on automobiles. The chief restraints on this behaviour are directives from above, for example, ‘prohibiting’ agricultural land conversion, or requiring that cities meet some arbitrary minimum population density. These restraints well illustrate the difficulty of the command-and-control approach to urban management. These edicts generally do not sufficiently recognize that expansion into farm areas will For differ across cities (depending on geography and population and per capita income growth), or that higher average densities are appropriate for larger cities with certain histories and geographies, but not for many other cities.

4.23 Incentives under current policies promote the inefficient use of land (Ding and Song (2005). First, many cities have fiscal incentives to focus on ‘green-field’ development rather than ‘brown-field.’ Under green-field development, cities seize agricultural land at the fringe and sell leaseholds to urban developers. The alternative way to acquire land for new uses is brown-field development, the redevelopment of land within the city. In some cities, negotiating the acquisition of urban land held by SOEs or by housing authorities is much more costly and time consuming than the profitable taking of agricultural land, which often may be acquired at below opportunity cost. Second, people are encouraged to travel by automobile, through the significant under-pricing of motor fuel (at about half the European price). This under-pricing makes it individually cost effective for people to commute long distances from low-density suburban areas. Key reforms to improve the efficiency of land use are as follows.

4.24 Ensure that cities pay the true cost of taking agricultural land If cities faced the opportunity cost of taking agricultural land, less might be taken and cities would be denser. First, the range of compensation for farmers for land taken -- set by national law based on income generated in agriculture – may be insufficient (Ding and Song, 2005). Defining appropriate compensation is tricky since part of the market value of agricultural land on the urban fringe reflects nearby urban infrastructure improvements (such as highway construction) for which farmers may not be paying their share of costs. Yet the value of agricultural production also appreciates with urban growth and enhanced demand for local agricultural products. Second, actual compensation is negotiated with the village head. The head may not adequately represent the village interests (as opposed to his personal interests), and may be induced to sell at a low price. Following one policy experiment under way, if all rural village residents were recognized as shareholders in the village land and if these shareholders had to approve land sales, this transparent decision process would provide the village head with a stronger incentive to negotiate in good faith. Third, as part of acquiring rural land, if villagers are offered urban hukou that helps ensure that cities city face the full opportunity cost of the land reallocation. If such rural villages are absorbed, in some places they will contain significant “illegal” housing” built to service migrants to the associated urban area. Such housing should be legalized and absorbed into the regular housing stock, just as we will advocate below for housing in “urban villages”.

4.25 Encourage brown-field development In the 1990s, brown-field development was encouraged in some cities (e.g., in Guangdong and Shanghai provinces), by allowing state enterprises to redevelop their land (“land development rights”) — providing a fiscal incentive to pursue land redevelopment more quickly. (Zhu, 2004, 2005). Reinstatement of greater property rights for historical (pre-1988) users of urban land would facilitate redevelopment, especially in cities that have redeveloped slowly. However reinstatement might conflict with the current policy of all urban land being developed through the auction of leaseholds and strengthening of property rights of those users.

4.26 Raise automobile costs Urban residents pay less than the social costs of operating vehicles – including both their contribution to congestion and their environmental costs. While the gasoline tax has been increased, an even greater increase towards Europena levels would encourage use of other transport modes, conversion of fuels, and air-quality improvements. In the largest cities, other incentive schemes would also reduce congestion and auto use: higher fees for parking, dedicated bus lanes to encourage bus travel, and development of ‘congestion-free zones,’ as in London, and congestion tolls, as in Singapore. Encouraging the maintenance of high population densities in the largest cities in turn helps make public transport viable, even underground systems in some cases, where cost-effective provision requires high usage over short distances.

Land-use planning and market operation to achieve livable cities

4.27 There is a tension in all urban land markets between allowing a free-wheeling market on the one hand, so land is sold to the highest and best use, and imposing land-use planning to regulate the market on the other hand. Even in the most market-driven countries, land markets are regulated. This regulation achieves two purposes. The first is to ensure that land allocated among different use types is separated, so ‘obnoxious’ users (e.g. industrial, polluting users) are separated from other users (e.g. residential users). This assures residents that a polluting factory cannot move nearby, imposing hazards and pollution and lowering property values. The second is to integrate private land uses with public ones, such as for transport infrastructure, so that industry is served by highways and other transport nodes, and so that residential users are near rail and bus facilities for commuting and shopping.

4.28 There are two issues for China. First, although the land market is certainly free-wheeling, it is not fully competitive and developers may not face the true costs of land acquisition (Cai et al, 2009). Second, in terms of regulation, as in more developed countries, cities in China have master plans for land use. While, in China, these master plans are referred to as ‘zoning plans’, they do not have the force of law that zoning plans have in developed countries which bind land allocations, based on master plans.

4.29 Ensure land goes to the highest and best use Urban land held by local land bureaus in China is allocated to developers through leasehold sales.. One issue has concerned the method of sale, given a history of negotiated sales with favored developers and corruptible auctions (Cai et al., 2009). All leaseholds would best and be sold at English auction (e.g., “pai mai”), with open entry (subject to meeting past contract requirements). Developers should face the true costs of their projects: besides paying market prices for land, they should pay for the necessary infrastructure required by their development (secondary roads, sewerage systems and roads, etc.). Alternatively, the cost of required infrastructure can be recovered through impact fees on the development set by the city.

4.30 Strengthen property rights in urban villages and integrate into city administration What should be done about urban villages located within cities? While perhaps cities should have purchased the land in such villages long ago, given these lands have been developed for urban purposes, property rights of land users should be respected. In particular, to reduce current uncertainties in local land markets, it should be made clear that villager leasehold sales are legitimate. However, urban villages within the city and on its fringe should be incorporated into the city administration. They should receive full city services, and be subject to all municipal taxes. Redevelopment projects of the village should be subject to the master and zoning plans of the city (discussed next).

4.31 Institute zoning plans Within the context of a competitive market for land, overall spatial patterns in the city should be governed by zoning laws. Zoning ensures that land-use decisions are not completely controlled by individual developers, who can ignore the objectives of master plans, resulting in disorderly land use. In developed countries, zoning plans restrict property rights, by limiting the legal right to build in a location to particular types of use and intensities. Plans are amended periodically as long-term demand conditions change. Plans are subject to the granting of variances – allowing a non-conforming use at a particular location. To ensure that the intent of the plan is met, the granting of variances occurs through a transparent public process. Those adversely affected are represented and generally must consent to the variance (perhaps with inducements provided by the applicant).

4.32 Zoning should be based master plans for city land development and should govern all future land redevelopment (including national-owned, military, and village-owned land) within the jurisdiction. Approval of master plans by the local People’s Congress would be a first step in giving legal bite to these plans. Cities will need lead time to develop up-dated plans and appropriate institutions, with investment in training professionals in city and transport planning and real-estate demand forecasting.

4.33 These recommendations are based on the presumed constraint, that granting “effective” private ownership” of urban lands is a policy reform that is not on the table. A reform that would increase competition and reduce bureaucrat interference would extend leaseholds indefinitely, make them fully transferable, and make redevelopment uses subject to city land use plans and zoning. Property taxes would replace leasehold sales revenue. The land market in Chin is viewed as corrupt (see Cai et al, 2009) and rather disorderly (Bertaud, 2009). Imposing market discipline coupled with zoning restraints would help these problems.

Low income housing

4.34 Across countries, a policy issue concerns housing for low-income people. The housing which the lowest income families in many countries can afford may be of such low quality in such crowded neighbourhoods, as to offend public sensibilities, or else to pose physical and mental health risks. One policy adopted historically in other countries has been public housing for the poor, which has a long, extensive record of policy failure, often eventually offering conditions as unhealthy, neglected and crowded as the ‘slums’ they replaced. Private housing markets work well; but bureaucratically managed ones, with few exceptions, have a record of failure. The increasingly favoured solution is to dismantle public housing systems for the poor, leaving housing provision to the market. In order for low-income people to afford better housing, one policy is to offer housing vouchers or other housing subsidies.

4.35 There are two other aspects related to housing subsidy programs. First as part of trying to upgrade the housing, the government should not impose demolition of existing low-income housing in cities, particularly in more historical neighbourhoods, even if it is clearly of a lower standard than new construction. The quality of viable low-income housing will be upgraded by market forces if the effective demand of low-income workers is increased by housing subsidies. Second, we already noted the problem of subsidizing housing for migrants, which may induce excessive in-migration). It may be no accident that the most successful public housing and subsidized housing programs are in Hong Kong and Singapore, two cities not subject to the potential of free in-migration.