【阅读】08/30起粤笃寂静整理（09/09更新，38篇原始，36篇考古）

gege1111

billyisfragile!

https://diegopuga.org/papers/divvsspz.pdf
Policy implications and conclusions

Two main issues arise time and again in the empirical and theoretical work on the structure of systems of cities reviewed in this paper. The first major issue is the efficiency of the urban system as a whole. Existing theories are divided on this point. Models in the spirit of Henderson (1974) tend to suggest that urban systems without central regulation will be efficient in terms of size and numbers of cities of each type. This is true, provided that some mechanism enabling the creation and development of new cities is present. The two usual mechanisms are autonomous local governments or the market, with land developers. Of course, there are likely to be distortions such as imperfect competition between developers or political-economy issues, making the interests of local communities not perfectly aligned with those of the ruling politicians. Nevertheless, these two mechanisms should bring urban systems reasonably close to efficiency. In contrast, theories where distance matters tend to have a more reserved judgement about the overall efficiency of the urban system. Unfortunately, detailed welfare analysis has not yet been carried out in this type of framework. Despite this division, a few policy prescriptions are clear. For instance, anything that hinders the creation and development of new cities, such as poorly-defined property rights, is likely to result in inefficient urban systems. The desirability of other policy actions is less clear. For instance, the need for (and type of) intervention needed in the presence of localisation and urbanisation economies depends delicately on the source of these economies, something that is far from well understood.

P2 否定了P1的内容，提出local government可以出台某种policy来调控specialized city。最有一句提到了labor mobility（举了欧洲的例子）这块有考点，考场上想了好久最后随便选了一个。（网上搜了一些论文，但相关度不大，只能回忆出这么多了。）

p1 讲城市有两种:一种 specialized， 一种 large 和 diverse 的(呃大概是跟神马经济环境相关的)，两种城市都需要存在。后面好像提到 specialized 的什么问题。。 p2:主要讲specialized的城市需要central government怎么怎么样，但是否定了;然后alternatively, local government 可以如何如何，好像也是否定了== 后面失忆 (就两段 一瓶左右)

后来第二段就提出了 government 要 limiting ，毕竟硅谷是少数嘛

第二篇是第一段 speicalized city 和 large diverse city 比较，然后指出 speicalized city 有很大 risk， 第二段写可能减小 risk 的办法:一是中央政府的保险，但是成本大;二是给予地方政府足够的控制权，但是怕地方势力 too ambitious，毕竟硅谷这样的少见;三是加强劳动力流动性，但是会使各个城市变得一样。总之就是没有提出一个可行的办法

The second major issue is about the composition of economic activity in individual cities. From our review it seems that specialisation has both advantages and disadvantages. The advantages are less urban crowding and stronger localisation economies arising from the proximity of closely related producers. The disadvantages are less innovation and more exposure to risk as the fortunes of specific sectors and technologies rise or fall. But overall, there appears to be a need for both large and diversified cities and smaller and more specialised cities. The main problem with this dichotomy of diversified and specialised cities is that the latter are exposed to a greater risk with the rise and fall of specific sectors and technologies. To limit the risks of urban specialisation, policy may want to curb this. But limiting specialisation may forego important present benefits for the sake of avoiding a possible future downturn. Insurance by the central government may be a good alternative. However, this is costly, and complete insurance may lead to complacency. Another alternative is to give local governments the power to restructure. The disadvantage here is the involvement of local government in excessively ambitious economic-development policies. In particular, not every community can create the next Silicon-Valley in its backyard. Moreover, the link between innovation and diversity seems fairly robust, so that highly innovative clusters cannot be bred in previously highly specialised environments. An alternative solution is to encourage labour-force mobility. But in Europe this awakens fears about lost regional identities.

Lubi

Ding