Glaeser and Xiong, members of Harvard’s Economics Department, in March 2017, released a working paper entitled “Urban Productivity in the Developing World”. Quoting the introduction, the paper seeks to see if “Cities help turn poor countries into rich countries”. The paper looks at several factors to determine this: productivity, density, human capital, and local entrepreneurship.
First the paper looks at the disparity in production between cities. All countries have a production disparity between urban and rural areas. More people are more densely located in cities. Industry and technological improvement is more likely to be in cities. However, the gap between urban and rural areas is more pronounced in developing countries. By understanding why parts of poor countries have become richer, we can improve the whole country. The part examines the correlation between urban density, population relative to area, and productivity in Brazil, China, and India. They use earnings, which in a classical model is equal to the value of a laborer’s production and total firm productivity. The urban-rural wage gap is established in earlier papers (urban areas earning 45%, 122%, and 176% more than rural areas in China, India, and Brazil, respectively). Data from China shows that earnings dispersion is matched with labor productivity dispersion closely (Figure 1). They show differences in labor productivity (by industry) between prefectures (small provincial areas) in China, showing that the productivity gap between the 1st and 2nd most productive is large (60%), and then lessens.
Then they look at the differences across industries in agglomeration (the extent firms locate near firms in the same industry (Silicon Valley, Wall street) and what extent firms locate in densely populated prefectures). 2000-2007 sees a rise in agglomeration in Chinese prefectures. Some industries agglomerate more than others (artificial fibers, electronic music equipment), and more traditional industries (silk-dyeing) have negative agglomeration. Agglomeration is attributed to the ability to share ideas and inputs and access a larger labor and customer pool. Looking at the industry-prefecture’s average labor productivity and the prefecture’s share of total employment in that industry, they conclude that the relationship isn’t linear. Any area with more than 2% of an industry’s employment has good productivity. The relationship between industry employment share and population density of the prefecture correlates positively, though not as much, since manufacturing isn’t urban in most cases (Figure 3).
Lastly, the correlation between population density and labor productivity is positive, being even slightly higher than that of the U.S. While this could be a misleading effect, most likely these agglomeration economies are legitimate because large cities tend to be more connected to the rest of the world, allowing technology to enter and dramatically boost productivity. Glaeser and Xiang run several correlations and regression between several variables already discussed as well as other variables that would alter them like college degrees and export percentage. The main takeaways from these charts are that industries and firms with higher percentages of employees with college degrees tend to be more agglomerated Information sharing is valuable with knowledge-based fields, so they urbanize to facilitate it. It also provides easier access to the rest of the world. Industries based on exporting tend to be urbanized as they need access to a larger customer base and to special economic zones like ports. While these are the main interpretations, the data shows that the tendency to and returns to urbanization and agglomeration are very industry based.
Next human capital (knowledge and skills) externalities (the benefits of being around smart and skilled people) is discussed. Areas are defined by skill level, so human capital increases will increase earnings. In India, Brazil, and China, these effects are more pronounced than in the U.S. Human capital is important in developing countries because high human capital enables spread of knowledge and skills, where there is a large gap in developing countries. The more urbanized developing countries become, the more pronounced these effects become. Denser populations are in closer proximity to people with higher human capital. Glaeser and Xiang say that developing countries shouldn’t use policy to impose artificial barriers on growth, such as housing limits, to maximize this phenomenon.
Next, they address entrepreneurship’s effects on development. Previous papers looking at entrepreneurship’s affects shows that it makes urban areas resilient to declines and increases employment and establishment size. It doesn’t increase income growth, potentially from the elastic labor supply or the ability of entrepreneurs suppress keep labor costs. Modern ideas conclude that entrepreneurship is a type of human capital, so it behaves as such.
Specifically, in developing countries in Africa, entrepreneurship is low because human capital is low. Economists originally thought that foreign direct investment (FDI) would increase local entrepreneurship potential and allow exporting businesses to flourish. It appears that neither FDI or local entrepreneurship have any effect unless a certain threshold of human capital, not present in many African countries, exists. They lack the knowledge and ability to produce for global markets. What about immigrant entrepreneurs? Several immigrant entrepreneurs are noted: Sergey Brin (Google) and Fernando Duarte (Nando’s), and others from India, Europe, and the Middle East. However, Africa has difficulties attracting entrepreneurs due to its unattractive locales and pushback from local politics and regulations. Making Africa more attractive and easier to access is crucial in policy decisions. Other policy considerations include investing in education to improve human capital and make areas more attractive to entrepreneurs who can better utilize skilled and knowledgeable people. This however doesn’t encourage native entrepreneurship, so potential strategies (little evidence of their effects exists) are entrepreneurial training, providing spaces for clustering entrepreneurs to facilitate learning from each other, and deregulating and removing restrictions that deter entrepreneurs. A combination of these strategies can increase local entrepreneurship and lift businesses to global markets.
Should policy encourage the increase of city size? Benefits and costs exist. Spatially biased policies are dangerous as they favor more politically powerful regions and subsidies may be misplaced. Barriers to urban growth need to be reduced through urban life quality improvement. Cities appear to be beneficial to the economy, so their growth shouldn’t be hindered. Downsides of density in urban areas are disease, congestion, crime, etc. Decreasing these makes cities more attractive to immigrants, as well as encourages productivity. Points to improve urban areas include infrastructure, which can reduce water-borne illnesses, traffic congestion, etc. Infrastructure is expensive and in several instances costs outweigh benefits. Different types of infrastructure providers are discussed, each having strengths and weaknesses. The For-Profit Independent and the Public Integrated are determined to have the most potential, given the presence of strong independent leaders or a non-corrupt government, respectively. Property rights are less clear and aren’t well protected in the developing world. This reduces incentives to invest in property and its improvement. Labor supply decreases because people spend more time protecting property. Property rights make transactions easier when properties are defined and documented, as well as reducing crime.
In summary, this paper promotes the benefits of urbanization of developing countries. Urbanized areas are more productive and attract more educated people and local and immigrant entrepreneurs. Investing in these people and human capital overall is crucial to development. Therefore policy-makers must enact policies that decrease the negative effects of urbanization and not restrict city growth to promote these skilled people.
Original Paper: http://www.nber.org/papers/w23279