Search

We develop a methodology to estimate robust city-level vehicular speed indices, exactly decomposable into uncongested speed and congestion. We apply it to 180 Indian cities using 57 million simulated trips measured by a web mapping service. We verify the reliability of our simulated trips using a number of alternative data sources, including data on actual trips. We find wide variation in speed across cities that is driven more by differences in uncongested speed than congestion. Denser and more populated cities are slower, only in part because of congestion. Urban economic development is correlated with faster speed despite worse congestion. (JEL O15, O18, R23, R41)

Transport investment has played an important role in the economic development of many countries. Starting from a low base, African countries have recently initiated several massive transportation i...

Quality-adjusted population density (QAPD) is population divided by land area that has been adjusted for geographic characteristics. We derive weights on these geographic characteristics from a global regression of population density at the quarter-degree level with country fixed effects. We show, first, that while income per capita is uncorrelated with conventionally measured population density across countries, there is a strong negative correlation between income per capita and QAPD; second, that the magnitude of this relationship exceeds the plausible structural effect of density on income, suggesting a negative correlation between QAPD and productivity or factor accumulation; and third, that higher QAPD in poor countries is primarily due to population growth since 1820. We argue that these facts are best understood as results of the differential timings of economic takeoff and demographic transition across countries, and particularly the rapid transfer of health technologies from early to late developers.

Previous work on transportation investments has focused on average impacts in high- and middle-income countries. We estimate average and heterogeneous effects in a poor continent, Africa, using roads and cities data spanning 50 years in 39 countries. Using changes in market access due to distant road construction as a source of exogenous variation, we estimate an 30-year elasticity of city population with respect to market access of 0.05-0.20. Our results suggest that this elasticity is stronger for small and remote cities, and weaker in politically favored and agriculturally suitable areas. Access to foreign cities matters little.

Previous work on transportation investments has focused on average impacts in high- and middle-income countries. We estimate average and heterogeneous effects in a poor continent, Africa, using roads and cities data spanning 50 years in 39 countries. Using changes in market access due to distant road construction as a source of exogenous variation, we estimate a 30-year elasticity of city population with respect to market access of about 0.08–0.13. Our results suggest that this elasticity is stronger for small and remote cities, and weaker in politically favored and agriculturally suitable areas. Access to foreign cities besides international ports matters little. Additional evidence points suggestively to rural-to-urban migration as the primary source of this population increase, though we cannot fully rule out natural increase or reallocation across cities.

We explore the role of natural characteristics in determining the worldwide spatial distribution of economic activity, as proxied by lights at night, observed across 240,000 grid cells. A parsimonious set of 24 physical geography attributes explains 47% of worldwide variation and 35% of within-country variation in lights. We divide geographic characteristics into two groups, those primarily important for agriculture and those primarily important for trade, and confront a puzzle. In examining within-country variation in lights, among countries that developed early, agricultural variables incrementally explain over 6 times as much variation in lights as do trade variables, while among late developing countries the ratio is only about 1.5, even though the latter group is far more dependent on agriculture. Correspondingly, the marginal effects of agricultural variables as a group on lights are larger in absolute value, and those for trade smaller, for early developers than for late developers. We show that this apparent puzzle is explained by persistence and the differential timing of technological shocks in the two sets of countries. For early developers, structural transformation due to rising agricultural productivity began when transport costs were still high, so cities were localized in agricultural regions. When transport costs fell, these agglomerations persisted. In late-developing countries, transport costs fell before structural transformation. To exploit urban scale economies, manufacturing agglomerated in relatively few, often coastal, locations. Consistent with this explanation, countries that developed earlier are more spatially equal in their distribution of education and economic activity than late developers.

We explore the role of natural characteristics in determining the worldwide spatial distribution of economic activity, as proxied by lights at night, observed across 240,000 grid cells. A parsimonious set of 24 physical geography attributes explains 47% of worldwide variation and 35% of within-country variation in lights. We divide geographic characteristics into two groups, those primarily important for agriculture and those primarily important for trade, and confront a puzzle. In examining within-country variation in lights, among countries that developed early, agricultural variables incrementally explain over 6 times as much variation in lights as do trade variables, while among late developing countries the ratio is only about 1.5, even though the latter group is far more dependent on agriculture. Correspondingly, the marginal effects of agricultural variables as a group on lights are larger in absolute value, and those for trade smaller, for early developers than for late developers. We show that this apparent puzzle is explained by persistence and the differential timing of technological shocks in the two sets of countries. For early developers, structural transformation due to rising agricultural productivity began when transport costs were still high, so cities were localized in agricultural regions. When transport costs fell, these agglomerations persisted. In late-developing countries, transport costs fell before structural transformation. To exploit urban scale economies, manufacturing agglomerated in relatively few, often coastal, locations. Consistent with this explanation, countries that developed earlier are more spatially equal in their distribution of education and economic activity than late developers.

This paper documents strong but differentiated links between climate and urbanization in large panels of districts and cities in Sub-Saharan Africa, which has dried substantially in the past fifty years. The key dimension of heterogeneity is whether cities are likely to have manufacturing for export outside their regions, as opposed to being exclusively market towns providing local services to agricultural hinterlands. In regions where cities are likely to be manufacturing centers (25% of our sample), drier conditions increase urbanization and total urban incomes. There, urban migration provides an “escape” from negative agricultural moisture shocks. However, in the remaining market towns (75% of our sample), cities just service agriculture. Reduced farm incomes from negative shocks reduce demand for urban services and derived demand for urban labor. There, drying has little impact on urbanization or total urban incomes. Lack of structural transformation in Africa inhibits a better response to climate change.

The past decade or so has seen a dramatic change in the way that economists can learn by watching our planet from above. A revolution has taken place in remote sensing and allied fields such as computer science, engineering, and geography. Petabytes of satellite imagery have become publicly accessible at increasing resolution, many algorithms for extracting meaningful social science information from these images are now routine, and modern cloud-based processing power allows these algorithms to be run at global scale. This paper seeks to introduce economists to the science of remotely sensed data, and to give a flavor of how this new source of data has been used by economists so far and what might be done in the future.

Most readers of this article are already familiar with images of the “Earth at Night,” which are really montages showing satellite views of visible lights.1 Looking at such images, it is easy to see that the brightness of visible lights is strongly related to both population density and income per capita. The effect of population density shows up in the visibility of urban agglomerations, coastlines, river valleys, and even the higher density of population along the paths of highways and railroads. Similarly, the Northeastern United States is far brighter than the mountain West, even though they have similarly levels of income. The effect of income is similarly easy to see. Sharp economic boundaries, such as the inter-Korean border show up plainly. India, with a population density slightly higher than Japan, is significantly dimmer, The observation that income per capita is one of the determinants of visible light suggests that one might potentially use visible light to help measure income. In this paper, we explore the usefulness of these night lights data to economists studying issues related to economic growth and the national or local level. In particular, we ask the extent to which lights data can augment or entirely replace data on Gross Domestic Product.2 There are a number of obstacles to using lights visible from space to measure income, broadly falling into two categories. First, the relationship between economic activity and the true amount of light emanating from the Earth’s surface is not constant across time and space. Light visible from space is a byproduct of consumption and production activities, but light is hardly produced in a fixed ratio to output. Some productive activities such as steelmaking produce a lot of light; others such as software design produce very little. Countries or regions may differ in the fraction of their economic activity that takes place after dark (for example, Las Vegas vs. Salt Lake City). Patterns of settlement (for example, whether people live in multi-storey buildings), the availability of hydropower, etc. may all affect how much light is visible from space for a given level of income and population. Second, true light is imperfectly measured by the satellites. Humidity, reflectivity, and time periods excluded from coverage because of sunlight, moonlight, and cloud cover contamination all differ across the globe. Sensor scale for recording lights in practice varies across satellites and as satellites age over time. If one is measuring economic growth, some of the location-specific factors are differenced out. The rest we treat as measurement error in the relationship between GDP growth and light growth. We note that while we focus on measuring total income growth of regions, one might be also interested in growth in income per capita. We don’t pursue that here although we have explored the issue by using population growth numbers to supplement lights growth. Population is both conceptually and practically easier to measure than is income, so it is also feasible to alternatively focus on income growth per capita. Despite problems, the night lights data have several advantages. The first has to do with the nature of measurement error in lights data. A number of studies have recently reminded economists that much of the data on GDP growth in developing countries is plagued by serious measurement error. For example, Johnson et al. (2009) study revisions in the Penn World Tables, a standard measure of GDP. They find that comparing version 6.1 of that data, released in 2002, with version 6.2 released in 2006, the standard deviation of the change in countries’ average growth over the period 1970–1999 was 1.1% per year (which is very large relative to the average growth rate of 1.56% per year). This indicates that at least one of the measures, and presumably both, contained a great deal of error. We do not claim that the measurement error in the night lights data is smaller than that in conventional data, but we do think that the two forms of measurement error are poorly or not correlated. It is well know that combing two problematic measures can produce a composite with smaller measurement error than either of them. A second advantage of the night lights data is that they are available for regions for which standard GDP measures are not. These include sub-national units such as cities and regions, as well as entities that cross national borders, like biomes. Economic analysis of growth and of the impacts of policies and events on cities and regions of many countries is hindered by a complete absence of any regular measure of local economic activity. Much of the interesting variation in economic growth takes place within rather than between countries. However, for the vast majority of economics research, “empirical analysis of growth” has become synonymous with use of national accounts data. The night lights data offer a tool that allows the researcher to set aside this limitation. The rest of this paper is organized as follows. In Section I, we briefly discuss the night lights data. Section II outlines our statistical approach to combining data from lights with conventional measures of GDP, and presents results of an exercise applying our technique to a set of low and middle income countries. Section III discusses an application beyond the measurement of national income. We refer the reader to our companion paper, Henderson, Storeygard, and Weil (2009), for more detailed discussion of many of the technical and statistical issues raised here.

We describe the compilation of a spatially explicit dataset detailing infant mortality rates in over 10,000 national and subnational units worldwide, benchmarked to the year 2000. Although their resolution is highly variable, subnational data are available for countries representing over 90% of non-OECD population. Concentration of global infant deaths is higher than implied by national data alone. Assigning both national and subnational data to map grid cells so that they may be easily integrated with other geographic data, we generate infant mortality rates for environmental regions, including biomes and coastal zones, by continent. Rates for these regions also show striking refinements from the use of the higher resolution data. Possibilities and limitations for related work are discussed.

Using two complementary methods in a framework that allows incorporating both environmental and household-level factors, we explore the correlates of underweight status among children. We use individual children as the units of analysis in 19 African countries, and subnational survey strata in 43 African, Asian and Latin American countries. We consider the relationship between household-level demographic and health survey data, environmental factors from external geospatial data sets and two indicators of malnutrition among children aged 1–3, deviations from the international standards of weight-for-age and height-for-age. We discuss methods for data integration. In general, household determinants explain more variation than environmental factors, perhaps partly due to more error-prone measurement at the community level. Among individual children, some measures of agricultural capacity are related to lower incidence of child hunger, while among regions, measures relating to urbanness and population density show a stronger relationship. We give recommendations for further study, data collection and policy making.

It is suggested that environmental or geographical factors (e.g. population density, climate and disease environment) play an important role in determining infant and child survival above and beyond that played by individual and household level factors. Only recently, however, have relevant spatial data become available and have demographic survey data systematically recorded geographical location of surveyed households, or has the technology to integrate these data become accessible. This study estimates the risk of infant and child death in ten West African countries attributable to individual, household and spatially explicit geographical factors; 120,000 births occurring in the 10 years prior to the 1997–2001 Demographic and Health Survey dates are evaluated. Results from a generalised linear model show that spatial variables explain away a good deal of the country-specific variation in mortality, and that they are associated with (and may be mediated by) the household characteristics. Implications for research and policy are addressed. Copyright © 2004 John Wiley & Sons, Ltd.

We examine the effect of rainfall shocks on dowry deaths using data from 583 Indian districts for 2002–2007. We find that a one standard deviation decline in annual rainfall from the local mean increases reported dowry deaths by 7.8%. Wet shocks have no apparent effect. We examine patterns of other crimes to investigate whether an increase in general unrest during economic downturns explains the results but do not find supportive evidence. Women's political representation in the national parliament has no apparent mitigating effect on dowry deaths.

This paper documents a significant impact of climate variation on urbanization in Sub-Saharan Africa, primarily in more arid countries. By lowering farm incomes, reduced moisture availability encourages migration to nearby cities, while wetter conditions slow migration. The paper also provides evidence for rural-urban income links. In countries with a larger industrial base, reduced moisture shrinks the agricultural sector and raises total incomes in nearby cities. However, if local cities are entirely dependent on servicing agriculture so their fortunes move with those of agriculture, reduced moisture tends to reduce local urban incomes. Finally, the paper shows that climate induces employment changes within the rural sector itself. Drier conditions induce a shift out of farm activities, especially for women, into non-farm activities, and especially out of the workforce. Overall, these findings imply a strong link between climate and urbanization in Africa.

We examine the effect of rainfall shocks on dowry deaths using data from 583 Indian districts for 2002-2007. We find that a one standard deviation decline in annual rainfall from the local mean increases reported dowry deaths by 7.8 percent. Wet shocks have no apparent effect. We examine patterns of other crimes to investigate whether an increase in general unrest during economic downturns explains the results but do not find supportive evidence. Women's political representation in the national parliament has no apparent mitigating effect on dowry deaths.

In the past dozen years, a literature has developed arguing that urbanization has unfolded differently in post-independence Sub-Saharan Africa than in the rest of the developing world, with implications for African economic growth overall. While African countries are more urbanized than other countries at comparable levels of income, it is well-recognized that total and sector gross domestic product data are of very low quality, especially in Africa. When instead viewed from the perspective of effective technology, as suggested in endogenous growth frameworks (and as proxied by educational attainment), the African urbanization experience overall matches global patterns. There are differences, however, at the sector level. Agricultural trade effects that improve farm prices deter African urbanization, while they promote urbanization elsewhere. Potential reasons include differences in land ownership institutions and the likelihood of agricultural surpluses being invested in urban production. Positive shocks to modern manufacturing spur urbanization in the rest of the developing world, but effects are dependent on the level of development. Thus many countries in Africa, with their lower level of development, do not respond to these shocks. Finally, historical indicators of the potential for good institutions promote urbanization both inside and outside Africa.

How does isolation affect the economic activity of cities? Transport costs are widely considered an important barrier to local economic activity but their impact in developing countries is not well-studied. This paper investigates the role of inter-city transport costs in determining the income of sub-Saharan African cities. In particular, focusing on fifteen countries whose largest city is a port, I ask how important access to that city is for the income of hinterland cities. The lack of panel data on both local economic activity and transport costs has prevented rigorous empirical investigation of this question. I fill this gap with two new datasets. Satellite data on lights at nigh proxy for city economic activity, and new road network data allow me to calculate the shortest route between cities. Cost per unit distance is identified b plausibly exogenous world oil prices. The results show that an oil price increase of the magnitude experienced between 2002 and 2008 induces the income of cities near a major port to increase by 6.6 percent relative to otherwise identical cities one standard deviation farther away. Combined with external estimates, this implies an elasticity of city economic activity with respect to transport costs of -0.25 at that distance. Moreover, the effect differs by the surface of roads between cities. Cities connected to the port by paved roads are chiefly affected by transport costs to the port, while cities connected to the port by unpaved roads are more affected by connections to secondary centers.

We examine the e ect of rainfall shocks on crimes against women using data from 583 Indian districts for 2002-2007. We nd that a one standard deviation decline in annual rainfall from the local mean increases reported domestic violence by 4.4 percent and dowry deaths by 7.8 percent. Wet shocks have no apparent e ect. These patterns are consistent with consumption smoothing by households relying on agriculture, but inconsistent with hypotheses emphasizing general unrest. Women's political representation in the national parliament has no apparent mitigating e ect.

The author created a sculpture using a rigorous mathematical procedure informed by aesthetic decisions. The form is based on the relationships found in the golden triangle and its spiral folded into three dimensions. In creating the form, the author discovered a three-dimensional motif (the “golden gnomonoid”) with interesting mathematical properties and sculptural ramifications.

GDP growth is often measured poorly for countries and rarely measured at all for cities or subnational regions. We propose a readily available proxy: satellite data on lights at night. We develop a statistical framework that uses lights growth to augment existing income growth measures, under the assumption that measurement error in using observed light as an indicator of income is uncorrelated with measurement error in national income accounts. For countries with good national income accounts data, information on growth of lights is of marginal value in estimating the true growth rate of income, while for countries with the worst national income accounts, the optimal estimate of true income growth is a composite with roughly equal weights. Among poor-data countries, our new estimate of average annual growth differs by as much as 3 percentage points from official data. Lights data also allow for measurement of income growth in sub- and supranational regions. As an application, we examine growth in Sub Saharan African regions over the last 17 years. We find that real incomes in non-coastal areas have grown faster by 1/3 of an annual percentage point than coastal areas; non-malarial areas have grown faster than malarial ones by 1/3 to 2/3 annual percent points; and primate city regions have grown no faster than hinterland areas. Such applications point toward a research program in which "empirical growth" need no longer be synonymous with "national income accounts."

Objective(s): The Millennium Development Goals (MDGs) have put maternal health in the mainstream, but there is a need to go beyond the MDGs to address equity within countries. We argue that MDG focus on maternal health is necessary but not sufficient. This paper uses Demographic and Health Survey (DHS) data from Kenya, Ethiopia and Ghana to examine a set of maternal health indicators stratified along five different dimensions. The study highlights the interactive and multiple forms of disadvantage and demonstrates that equity monitoring for the MDGs is possible, even given current data limitations. Methods: We analyse DHS data from Ghana, Kenya and Ethiopia on four indicators: skilled birth attendant, contraceptive prevalence rate, AIDS knowledge and access to a health facility. We define six social strata along five different dimensions: poverty status, education, region, ethnicity and the more traditional wealth quintile. Data are stratified singly (e.g. by region) and then stratified simultaneously (e.g. by region and by education) in order to examine the compounded efect of dual forms of vulnerability. Results: Almost all disparities were found to be significant, although the stratifier with the strongest effect on health outcomes varied by indicator and by country. In some cases, urban-dwelling is a more significant advantage than wealth and in others, educational status trumps poverty status. The nuances of this analysis are important for policymaking processes aimed at reaching the MDGs and incorporating maternal health in national development plans. Conclusion(s): The article highlights the following key points about inequities and maternal health: 1) measuring and monitoring inequity in access to maternal health is possible even in low resource settings—using current data 2) statistically significant health gaps exist not just between rich and poor, but across other population groups as well, and multiple forms of disadvantage confer greater risk and 3) policies must be aligned with reducing health gaps in access to key maternal health services. Conflict of Interest: The authors declare that there are no conflicts of interest. The lead author had full access to the data used in this paper and the final responsibility for the decision to submit for publication. East African Journal of Public Health Vol. 5 (3) 2008: pp. 133-141

Emerging infectious diseases (EIDs) are a significant burden on global economies and public health1–3. Their emergence is thought to be driven largely by socio-economic, environmental and ecological factors1–9, but no comparative study has explicitly analysed these linkages to understand global temporal and spatial patterns of EIDs. Here we analyse a database of 335 EID ‘events’ (origins of EIDs) between 1940 and 2004, and demonstrate non-random global patterns. EID events have risen significantly over time after controlling for reporting bias, with their peak incidence (in the 1980s) concomitant with the HIV pandemic. EID events are dominated by zoonoses (60.3% of EIDs): the majority of these (71.8%) originate in wildlife (for example, severe acute respiratory virus, Ebola virus), and are increasing significantly over time. We find that 54.3% of EID events are caused by bacteria or rickettsia, reflecting a large number of drug-resistant microbes in our database. Our results confirm that EID origins are significantly correlated with socio-economic, environmental and ecological factors, and provide a basis for identifying regions where new EIDs are most likely to originate (emerging disease ‘hotspots’). They also reveal a substantial risk of wildlife zoonotic and vector-borne EIDs originating at lower latitudes where reporting effort is low. We conclude that global resources to counter disease emergence are poorly allocated, with the majority of the scientific and surveillance effort focused on countries from where the next important EID is least likely to originate.

OBJECTIVE: This analysis seeks to set the stage for equity-sensitive monitoring of the health-related Millennium Development Goals (MDGs). METHODS: We use data from international household-level surveys (Demographic and Health Surveys (DHS) and Multiple Indicator Cluster Surveys (MICS)) to demonstrate that establishing an equity baseline is necessary and feasible, even in low-income and data-poor countries. We assess data from six countries using 11 health indicators and six social stratifiers. Simple bivariate stratification is complemented by simultaneous stratification to expose the compound effect of multiple forms of vulnerability. FINDINGS: The data reveal that inequities are complex and interactive: inferences cannot be drawn about the nature or extent of inequities in health outcomes from a single stratifier or indicator. CONCLUSION: The MDGs and other development initiatives must become more comprehensive and explicit in their analysis and tracking of inequities. The design of policies to narrow health gaps must take into account country-specific inequities.