Get Complete Project Material File(s) Now! »

**Empirical Analysis**

Using the World Bank’s World development indicators data from 1970 to 2009 for all 47 SSA countries, statistical analysis of GDP per capita and urbanization is conducted in this section. Three regressions, using fixed effects panel LS is conducted for GDP per capita with the following main explanatory variables: urban population as a percen-tage of total population, prime city size as a percentage of total population and prime city size as a percentage of urban population.

**Data collection**

The data presented in this paper has been collected from various sources, namely: The World Bank’s development indicators data, Penn World Tables and various segments of the UN. In particular UN Habitat, UNDP and the UN World urbanization prospects (2009 revised version).

It is important at this point to note the problems with various definitions of Urbaniza-tion. The concept of what is urban and the rural-urban divide has become more of a grey area as it changes over time (Frey and Zimmer 2001). Moreover each country has their own definition of what they constitute as urban. For example in Benin any area with more than 10,000 inhabitants is considered to be urban, in Angola this figure is 2,000 and in Botswana an area with more than 5000 inhabitants 75% of with must be engaged in non-agricultural labor would be considered urban (Cohen, 2003). This will be taken into consideration in the analysis.

The collected data on urbanization, secondary school enrollment, foreign direct invest-ment, trade as well as Gross capital formation, to be used in the regression are from WB (for definitions and measurement information see Appendix A notes 1). The data spans 40 years from 1970 to 2009. All variables where relevant are in the form of as a percen-tage of GDP. Urbanization is measured by population living in urban areas as a percen-tage of total population, using national definitions of what constitutes as urban.^{4}

**Descriptive statistics: Urbanization and GDP per capita**

It is fitting to begin with the trends of what will be considered the dependent variable in following calculations, GDP per capita. By world standards GDP per capita in SSA has been and remains extremely low. Regardless of this fact, GDP per capita has been in-creasing over the years as the figure 3-1 below shows. The range of GDP per capita within the SSA region is large with countries such as South Africa and Botswana show-ing approximately 7500 and 8800 (constant USD prices) compared to the Democratic Republic of Congo at 231 for the year 2009.

As the Two figures show, both variables GDP per capita and urban population as a per-centage of total population have the same upward trend. Figure 3-2 shows that urban population grew from less that 15% in 1950 to approximately 40% in 2010. More de-veloped countries have seen the same 30% increase (approx.) during these same 50, but have had better suited economies to handle the increase.

It is clear from the figure that urban population has been growing and is expected to continue 20 years from today. Furthermore, grouped countries by regions in the SSA, eastern, western and central SSA shows the same upward trend. The southern part of SSA has seen less of an increase in urban population as compared to the others (see ap-pendix A. figure A1-1). The southern region of SSA also contains the countries with the highest GDP per capita.

It is interesting to make a comparison between the trends in SSA and the rest of the world. As noted earlier this trend of urbanization is increasingly being seen in less de-veloped countries. This would coincide with the theories that explain the urbanization process in stages, where at the beginning of economic development there should be slow urbanization, as countries move from low to medium levels of economic develop-ment, urbanization should rapidly increase. Finally at high levels of economic develop-ment one should see a tapering off of urbanization. In Figure 3-3 one sees that for more developed regions the urbanization process has slowed and the tapering off is expected by 2020.For all other regions there is an upward trend and interestingly even for the *least *developed region, urbanization has been rapidly increasing since 1990 (*least *de-veloped as define by the UN GA 2001 comprises of 49 countries 34 of which are Afri-can).

Debatably, one may disagree with describing a process of urbanization based on actual population figures (as in figure 3-3) since population its self has an upward trend. How-ever, even when taking the growth rates of urban population, the fact still remains that urbanization is in fact taking place and at high rates. Seen in figure 3-4 is the growth rate compared to more developed regions (Europe, US, Australia/New-Zealand and Ja-pan). Furthermore, the data shows the rate of change for urban growth to be positive and averaging 1.76% (this should be considered high since the total population growth rate has been fairly constant around 2%). Comparatively, the rate of change for rural growth has been consistently negative from 1950 to 2010 (ranging from -0.37 to -0.87).

The trend path suggested by Williamson (1965) for urbanization and economic growth may or may not be confirmed by the situation in SSA, depending on how it is inter-preted. The figure below shows both urban and GDP per capita growth. One can note that when GDP per capita growth was relatively high in 1970-75 so was urban growth, since then they have declined simultaneously, though at much slower rates for urban growth. It is important to recognize that from 1990-2010 where GDP per capita growth has been increasing, urban growth has been steadily on the decline.

**Descriptive Statistics: Prime city**

In SSA countries, urbanization (urban population as a percentage of total population) and Prime city size (population of prime city as a percentage of total population) fall in the same category. Countries that have high (relatively) urbanization also have high prime city size (relatively). In 1970 the top 10 in terms of urbanization are also in the top 15 in terms of prime city size (with one outlier: Nigeria). The bottom 10 countries in terms of urbanization are also the bottom 15 in terms of prime city size. The same pat-tern holds for 2009, where the bottom 10 in terms of urbanization range between 13% and 30% also fall in the bottom 12 in terms of prime city size (ranging from 3.3% to 8.7%). This is excluding Burkina Faso which showed prime city size increase by ap-proximately 520% (but only 350% increase in urban population). For the top 10 urba-nized countries in 2009, all except Nigeria and South Africa fall in the top 15 prime city size (note that these two countries are the wealthiest in the region). Some of the num-bers are staggeringly high, with countries such as Angola, Republic of Congo and Libe-ria showing 35%, 32% and 24% of the population as living in the prime city. Caution on the interpretation of this should be taken as these countries have civil unrest in common (For example in Angola, rebel activity was mainly in rural areas, therefore this could explain the urban population influx). Figure 3-6 below shows the relationship between urbanization and prime city size (both measures by their population with respect to total population).

**1 Introduction**

1.1 Purpose

1.2 Outline

**2 Background**

2.1 Theoretical Background

2.2 The Phenomenon of Over-urbanization

**3 Empirical Analysis **

3.1 Data collection

3.2 Descriptive statistics: Urbanization and GDP per capita

3.3 Methodology

3.4 Results

**4 Analysis**

4.1 Discussion of Data

4.2 Discussion with reference to theory

**5 Conclusion**

5.1 Suggested further study

List of references

GET THE COMPLETE PROJECT

The Impact of Urbanization on GDP per capita