Publications

Research Spotlight

Aerial view of Tokyo at nightExplaining the prevalence, scaling, and variance of urban phenomena


As we experience rapid urbanization around the globe – and especially in the developing world – it becomes more critical to understand the dynamics of cities and their dual capacity to both spur creativity and wealth, and to amplify crime and disease. With this motivation in mind, a research collaboration between the Growth Lab at Harvard’s Center for International Development and the T.H. Chan School of Public Health has resulted in the formation of a unified model to analyze how social phenomena occur in urban areas. This exciting new research - published in the journal Nature Human Behavior - could influence the ways in which urban policies are developed, by helping local governments anticipate when these phenomena would occur and take precautionary measures to address them.

Recent Publications

Schetter, U., 2019. A Structural Ranking of Economic Complexity.Abstract
We propose a structural alternative to the Economic Complexity Index (ECI, Hidalgo and Hausmann 2009; Hausmann et al. 2011) that ranks countries by their complexity. This ranking is tied to comparative advantages. Hence, it reveals information different from GDP per capita on the deep underlying economic capabilities of countries. Our analysis proceeds in three main steps: (i) We first consider a simplified trade model that is centered on the assumption that countries’ global exports are log-supermodular (Costinot, 2009a), and show that a variant of the ECI correctly ranks countries (and products) by their complexity. This model provides a general theoretical framework for ranking nodes of a weighted (bipartite) graph according to some under- lying unobservable characteristic. (ii) We then embed a structure of log-supermodular productivities into a multi-product Eaton and Kortum (2002)-model, and show how our main insights from the simplified trade model apply to this richer set-up. (iii) We finally implement our structural ranking of economic complexity. The derived ranking is robust and remarkably similar to the one based on the original ECI.
Popularity Spikes Hurt Future Chances For Viral Propagation of Protomemes
Coscia, M., 2018. Popularity Spikes Hurt Future Chances For Viral Propagation of Protomemes. Communications of the ACM , 61 (1) , pp. 70-77. Publisher's VersionAbstract
A meme is a concept introduced by Dawkins12 as an equivalent in cultural studies of a gene in biology. A meme is a cultural unit, perhaps a joke, musical tune, or behavior, that can replicate in people's minds, spreading from person to person. During the replication process, memes can mutate and compete with each other for attention, because people's consciousness has finite capacity. Meme viral spreading causes behavioral change, for the better, as when, say, the "ALS Bucket Challenge" meme caused a cascade of humanitarian donations,a and for the worse, as when researchers proved obesity7 and smoking8 are socially transmittable diseases. A better theory of meme spreading could help prevent an outbreak of bad behaviors and favor positive ones.
Coscia, M., 2018. Using arborescences to estimate hierarchicalness in directed complex networks. PLoS ONE , 13 (1). Publisher's VersionAbstract
Complex networks are a useful tool for the understanding of complex systems. One of the emerging properties of such systems is their tendency to form hierarchies: networks can be organized in levels, with nodes in each level exerting control on the ones beneath them. In this paper, we focus on the problem of estimating how hierarchical a directed network is. We propose a structural argument: a network has a strong top-down organization if we need to delete only few edges to reduce it to a perfect hierarchy—an arborescence. In an arborescence, all edges point away from the root and there are no horizontal connections, both characteristics we desire in our idealization of what a perfect hierarchy requires. We test our arborescence score in synthetic and real-world directed networks against the current state of the art in hierarchy detection: agony, flow hierarchy and global reaching centrality. These tests highlight that our arborescence score is intuitive and we can visualize it; it is able to better distinguish between networks with and without a hierarchical structure; it agrees the most with the literature about the hierarchy of well-studied complex systems; and it is not just a score, but it provides an overall scheme of the underlying hierarchy of any directed complex network.
Birds of a feather scam together: Trustworthiness homophily in a business network
Barone, M. & Coscia, M., 2018. Birds of a feather scam together: Trustworthiness homophily in a business network. Social Networks , 54 (July 2018) , pp. 228-237. Publisher's VersionAbstract
Estimating the trustworthiness of a set of actors when all the available information is provided by the actors themselves is a hard problem. When two actors have conflicting reports about each other, how do we establish which of the two (if any) deserves our trust? In this paper, we model this scenario as a network problem: actors are nodes in a network and their reports about each other are the edges of the network. To estimate their trustworthiness levels, we develop an iterative framework which looks at all the reports about each connected actor pair to define its trustworthiness balance. We apply this framework to a customer/supplier business network. We show that our trustworthiness score is a significant predictor of the likelihood a business will pay a fine if audited. We show that the market network is characterized by homophily: businesses tend to connect to partners with similar trustworthiness degrees. This suggests that the topology of the network influences the behavior of the actors composing it, indicating that market regulatory efforts should take into account network theory to prevent further degeneration and failures.
Coscia, M. & Neffke, F., 2017. Network Backboning with Noisy Data. 2017 IEEE 33rd International Conference on Data Engineering (ICDE) , (May) , pp. 425-436. Publisher's VersionAbstract
Networks are powerful instruments to study complex phenomena, but they become hard to analyze in data that contain noise. Network backbones provide a tool to extract the latent structure from noisy networks by pruning non-salient edges. We describe a new approach to extract such backbones. We assume that edge weights are drawn from a binomial distribution, and estimate the error-variance in edge weights using a Bayesian framework. Our approach uses a more realistic null model for the edge weight creation process than prior work. In particular, it simultaneously considers the propensity of nodes to send and receive connections, whereas previous approaches only considered nodes as emitters of edges. We test our model with real world networks of different types (flows, stocks, cooccurrences, directed, undirected) and show that our Noise-Corrected approach returns backbones that outperform other approaches on a number of criteria. Our approach is scalable, able to deal with networks with millions of edges.
The Mobility of Displaced Workers: How the Local Industry Mix Affects Job Search
Neffke, F., Otto, A. & Hidalgo, C., 2018. The Mobility of Displaced Workers: How the Local Industry Mix Affects Job Search. Journal of Urban Economics , 108 (November 2018) , pp. 124-140. Publisher's VersionAbstract
Are there Marshallian externalities in job search? We study how workers who lose their jobs in establishment closures in Germany cope with their loss of employment. About a fifth of these displaced workers do not return to social-security covered employment within the next three years. Among those who do get re-employed, about two-thirds leave their old industry and one-third move out of their region. However, which of these two types of mobility responses workers will choose depends on the local industry mix in ways that are suggestive of Marshallian benefits to job search. In particular, large concentrations of one’s old industry makes it easier to find new jobs: in regions where the pre-displacement industry is large, displaced workers suffer relatively small earnings losses and find new work faster. In contrast, large local industries skill-related to the pre-displacement industry increase earnings losses but also protect against long-term unemployment. Analyzed through the lens of a job-search model, the exact spatial and industrial job-switching patterns reveal that workers take these Marshallian externalities into account when deciding how to allocate search efforts among industries.
The workforce of pioneer plants: The role of worker mobility in the diffusion of industries
Hausmann, R. & Neffke, F., 2018. The workforce of pioneer plants: The role of worker mobility in the diffusion of industries. Research Policy. Publisher's VersionAbstract

Does technology require labour mobility to diffuse? To explore this, we use German social-security data and ask how plants that pioneer an industry in a location – and for which the local labour market offers no experienced workers – assemble their workforces. These pioneers use different recruiting strategies than plants elsewhere: they hire more workers from outside their industry and from outside their region, especially when workers come from closely related industries or are highly skilled. The importance of access to experienced workers is highlighted in the diffusion of industries from western Germany to the post-reunification economy of eastern German. While manufacturing employment declined in most advanced economies, eastern German regions managed to reindustrialise. The pioneers involved in this process relied heavily on expertise from western Germany: while establishing new manufacturing industries in the East, they sourced half of their experienced workers from the West.

Originally published as CID Working Paper 310
Kosack, S., et al., 2018. Functional structures of US state governments. Proceedings of the National Academy of Sciences of the United States of America. Publisher's VersionAbstract
Governments in modern societies undertake an array of complex functions that shape politics and economics, individual and group behavior, and the natural, social, and built environment. How are governments structured to execute these diverse responsibilities? How do those structures vary, and what explains the differences? To examine these longstanding questions, we develop a technique for mapping Internet “footprint” of government with network science methods. We use this approach to describe and analyze the diversity in functional scale and structure among the 50 US state governments reflected in the webpages and links they have created online: 32.5 million webpages and 110 million hyperlinks among 47,631 agencies. We first verify that this extensive online footprint systematically reflects known characteristics: 50 hierarchically organized networks of state agencies that scale with population and are specialized around easily identifiable functions in accordance with legal mandates. We also find that the footprint reflects extensive diversity among these state functional hierarchies. We hypothesize that this variation should reflect, among other factors, state income, economic structure, ideology, and location. We find that government structures are most strongly associated with state economic structures, with location and income playing more limited roles. Voters’ recent ideological preferences about the proper roles and extent of government are not significantly associated with the scale and structure of their state governments as reflected online. We conclude that the online footprint of governments offers a broad and comprehensive window on how they are structured that can help deepen understanding of those structures.
Patterson-Lomba, O. & Gomez-Lievano, A., 2018. On the Scaling Patterns of Infectious Disease Incidence in Cities.Abstract
Urban areas with larger and more connected populations offer an auspicious environment for contagion processes such as the spread of pathogens. Empirical evidence reveals a systematic increase in the rates of certain sexually transmitted diseases (STDs) with larger urban population size. However, the main drivers of these systemic infection patterns are still not well understood, and rampant urbanization rates worldwide makes it critical to advance our understanding on this front. Using confirmed-cases data for three STDs in US metropolitan areas, we investigate the scaling patterns of infectious disease incidence in urban areas. The most salient features of these patterns are that, on average, the incidence of infectious diseases that transmit with less ease– either because of a lower inherent transmissibility or due to a less suitable environment for transmission– scale more steeply with population size, are less predictable across time and more variable across cities of similar size. These features are explained, first, using a simple mathematical model of contagion, and then through the lens of a new theory of urban scaling. These frameworks help us reveal the links between the factors that determine the transmissibility of infectious diseases and the properties of their scaling patterns across cities.
Nedelkoska, L. & Khaw, N., 2015. The Albanian Community in the United States: Statistical Profiling of the Albanian-Americans.Abstract

When the Albanian Communist regime fell in 1991-92, many Albanians saw their future outside the borders of Albania. At that time in history, no one anticipated the scale of migration that would take place in the subsequent two decades. Today, one third of Albania’s 1991 population lives abroad. Most of these migrants live and work in neighboring Greece and Italy. The third most popular destination is however the United States. Besides this new wave of migrants, the US has an old Albanian diaspora–the offspring of migrants who came to the US between the First and the Second World War. This is what mainly gives rise to the second generation Albanian-Americans.

To the best of our knowledge, there is currently no systematic documentation of the socio-demographic and economic characteristics of the Albanian community in the US. To bridge this gap, we use data from the American Community Survey 2012 and analyze these characteristics. The profiling could be of interest for anyone who focuses on the Albanians abroad – the Government’s Programs dealing with diaspora and migration issues, researchers interested in migration questions, the Albanian Community Organizations in the US or the diaspora members themselves.

We find that the first and the second generation Albanian-Americans have distinctive features. The first generation (those who arrived after the fall of Communism) is more educated than the non-Albanian Americans with comparable demographics. This is particularly true of Albanian women. The education of the second generation resembles more closely the US population with comparable demographic characteristics.

Despite the qualification advantage, first generation Albanian-Americans earn much less than non-Albanian Americans with comparable socio-demographic characteristics. We find that this is not associated with being Albanian per se but with being an immigrant more generally. The migrant-native gap narrows down with time spent in the US.

An important channel through which the current gap is maintained is qualification mismatch. We observe that first generation Albanian-Americans are over-represented in occupations requiring little skills and under-represented in occupations requiring medium and high skills, in direct contrast to them being more educated than non-Albanians.

When it comes to the earnings of second generation Albanian-Americans, the situation is more nuanced. The low skilled Albanian-Americans earn significantly more, and the highly skilled Albanian-Americans earn significantly less than the non-Albanian Americans with comparable socio-demographic characteristics. We currently do not have a straightforward explanation for this pattern.

The Albanian population in the US is highly concentrated in a few states: New York, Michigan and Massachusetts account for almost 60% of all Albanian Americans. The community in Massachusetts is the best educated; best employed and has the highest earnings among the three, but is also the oldest one in terms of demographics.

However, due to its sheer size (over 60,000 Albanian-Americans), New York is the host of most Albanians with BA degree (about 10,000). New York also hosts the largest number of high earning Albanians (about 1,800 earn at least $100,000 a year).

More