Abstract

There are two milestones which make spatial dimension of economics much more influential in recent years. One is that Paul Krugman won Nobel Prize in economics in 2008; another is the publication of the World Bank's report, “World Development Report 2009:Reshaping Economic Geography”, which stresses the effects of density, distance and division（3Ds）. Unfortunately, while population density represents an important socio-economic parameter, its role is rarely studied in the economic literature（contrary to natural sciences）（Yuri A. Yegorov, 2009）. Especially when we focus on the research of population spatial distribution and its relationship with economic growth, few studies considered simultaneously spatial effect and spatial analysis methodology. This is because there are two difficulties at least:One is that it is hard to include the spatial factor in Neoclassic Economics, so while New Economic Geography made the breakthrough, economists are so excited. Furthermore, general Econometrics with lack of spatial perspective cannot overcome the problem like spatial-dependence and spatial-heterogeneity.

However, difficulty always accompanies with new idea. This current paper tries to do some kind of those researches. On one hand, I studied on the（Chinese）population distribution with the importance of population spatial attribution. On the other hand, I focused on the relationship between population density and economic growth, and tried to add a theoretic factor into economic growth research. What's more, spatial analysis methodology, such as Spatial Econometrics, which considers spatial-dependence and spatial-heterogeneity, could better explain the spatial dimension of demography than traditional method ology, is applied on almost all my empirical researches in this paper. Meanwhile, these studies are based on some realistic arguments:（1）To some countries like China, is it too much population or too less space?（2）Which promotes economic growth, population quantity itself or population agglomeration?（3）Which is better, optimal population quantity or optimal population density（distribution）? So this current paper maybe provides a research support for Chinese population distribution policy. Specifically, Chinese population distribution is considerably uneven, and this trend will continue. However, the control polices of population migration into mega-cities like Beijing and Shanghai make little success. So what policies shall we take when we face the coexisting of agglomeration and congestion effects of population density? Here this book will give an answer.

Specifically, here are the abstracts of some core chapters.

Chapter 2. Literature review on population spatial distribution and its relationship with economic growth. I give three points:first, the main context and importance of research of population spatial distribution; second, the importance of spatial effect and spatial analysis methodology in population spatial distribution research; third, the effect of population distribution on economic growth. And finally I give my comments, one of which is that few studies considered simultaneously spatial effect and spatial analysis methodology in population distribution research.

Chapter 3. Some new description of Chinese population distribution and its spatial auto-correlation analysis. Based on the data of the fifth and sixth population census（2000,2010）of China, this chapter compares and contrasts the spatial characteristics of Chinese population distribution in three spatial scales:counties（NO.2,844）, cities（NO.337）and provinces（NO.31）. That is due to the Modifiable Areal Unit Problem（MAUP）. Especially the spatial auto-correlation analysis, which includes the spatial weights, is influenced much more by MAUP, such as scale effect and pattern effect, size, boundaries, distance and adjacency to be exact.

Chapter 4. Quantifying the uneven degree of population distribution and its simulation and projection of Probability Density Function. Based on the census data of county scale, this chapter quantifies the uneven degree of population distribution in the whole China, and Eastern, Central, Western and Northeastern China as well as some provinces with Gini-Coefficient. Furthermore, this study does some simulations in Probability Density Function of population density, and finds that Log Normal Distribution fits well. And then, I give the projection of Probability Distribution of population density in 2020,2030,2040,2050, 2075 and 2100 based on Log Normal Distribution. And one of the projection results shows that the uneven degree of population distribution in future will be larger. However, this trend is consistent to the population flowing law. So it is not difficult to understand that the controll policies of population migration to megacities like Beijing and Shanghai make little success.

Chapter 5. The influence mechanism of population density:in micro perspective. This chapter does not pay attention to the macro scale like a country but a micro and special region in Western China where the environment and terrain is so complicated. It is due to the fact that this kind of researches on macro scale is affluent and their results and conclusions are familiar. However, micro scale may give us more complex details, in that perspective, spatial-heterogeneity will show out, even may accompany with some paradoxes we met less before.

Chapter 6. The theoretical mechanism and models about population density and economic growth. It cannot be denied that population density could influence economic growth, but how, and what is the mechanism and how to prove? This chapter gives the theoretical framework to these questions in two aspects:agglomeration effects and congestion effects of population density. Agglomeration effect means population density promotes economic growth, and congestion effect means when the population density gets a certain level it will obstruct the economic growth. There are two theoretical models for both aspects. The first model for agglomeration mechanism is Neoclassic Growth Model in which the population density was introduced directly. The second one comes from Ciccone（1996, 2002）which pays attention to the output produced on the land（space）. And the first model for congestion mechanism is the expansion of the first model considering the dynamic spatial externality. From this model an inverted U-shaped relation between population density and economic growth is got. And the last one is derived from the Local Spillover Model（LSM）of New Economic Geography. Then the paper introduces the population density into LSM, and gets an inverted U-shaped curve with quadratic function which could prove exactly the relationship between population density and economic growth. All the models are the theoretic foundation of the empirical study later.

Chapter 7. Empirical test of population density and economic growth based on the theoretical model result:inverted U-shaped curve with quadratic function. This result provides a framework for empirical basic model specification. Then empirical analysis is based on the Spatial Panel Data Model with the data of 126 countries and regions from 1992 to 2012 and 256 Chinese cities from 2001 to 2012. The key test for my theoretic framework stands on the hypothesis that the coefficient of population density（proxy for agglomeration effects）is significantly positive while the coefficient of population density square（proxy for congestion effects）is significantly negative. And my test results support the theoretic framework. So inverted U-shaped relation is confirmed right. Meanwhile Spatial Econometrics which considers spatial-dependence and spatial-heterogeneity could be better to explain the spatial dimension of demography than traditional methodology and empirical data and model. From Chapters 6 &7, we can see that theoretical model and empirical test support each other.

Chapter 8. Conclusion and next steps. This chapter makes a conclusion for all sub-topics of this thesis, including the research logic and main results. Meanwhile, it puts out the foothold of Chinese population distribution polices and the key aim of spatial equilibrium of population. At last it points out some shortcomings of this study and some research promotion in the future. For instance, more attention should be paid to the indirect effects of population density, such as the influence of population density on technology progress or environmental congestion which then influences the economic growth.

There are some innovations in this thesis. First, new research perspective of economic growth. There are few researches on relationship between population distribution and economic growth, though population, population structure, and population quality（human capital）appear frequently. Second, this paper pays much attention to the core attribution of population distribution:spatial dimension of demography. On one hand, I study on the population distribution with the importance of population spatial attribution. On the other hand, I try my best to apply spatial methodologies, especially the Spatial Econometrics. Third, the paper explains the relationship between population density and economic growth. According to the theoretical framework and empirical test, there is an inverted Ushaped curve with quadratic function between them. Fourth, the paper provides a research support for Chinese population distribution policy. Specifically, Chinese population distribution is considerably uneven, and this trend will continue. However, the controll policies of population migration to mega-cities like Beijing and Shanghai make little success. And agglomeration effects and congestion effects of population density co-exist. So the more effective policies will be redistribution of resource and governmental support which will guide the population migration and redistribution naturally and rationally. And then both mega-cities and emerging cities, or different scale cities, would keep growing in a relative high level rate.

However, there are also some shortcomings in this paper due to the boarding perspective of population distribution and its relationship with economic growth. So some new fields and questions should be explored in future. For example, population density can proxy for population distribution, but population distribution does not mean population density only. That means we should choose some more variables of population distribution to prove the outcome in this paper. What's more, population density may not influence economic growth directly. It may do throughout some indirect factors, such as technology progress or environmental congestion which then influences the economic growth. So we should continue to expand the growth model with some indirect variables like technology and environment.