Using a large individual-level survey spanning several years and more than 150 countries, we examine the importance of social networks in influencing individuals’ intention to migrate internationally and locally. We distinguish close social networks (composed of friends and family) abroad and at the current location, and broad social networks (composed of same-country residents with intention to migrate, either internationally or locally). We find that social networks abroad are the most important driving forces of international migration intentions, with close and broad networks jointly explaining about 37% of variation in the probability intentions. Social networks are found to be more important factors driving migration intentions than work-related aspects or wealth (wealth accounts for less than 3% of the variation). In addition, we find that having stronger close social networks at home has the opposite effect by reducing the likelihood of migration intentions, both internationally and locally.
In August 2016, the Government of Sri Lanka (GoSL) and the Building State Capability program of CID convened five teams of civil servants, tasking them with solving issues related to investment and export promotion. One of these teams, the “Targeting Team,” took on the task of formulating and executing a plan to identify promising new economic activities for investment and export promotion in Sri Lanka. With the assistance of CID’s Growth Lab, the Targeting Team assembled and analyzed over 100 variables from 22 datasets, studying all tradable activities and 29 representative subsectors. Their analysis highlighted the potential of investment related to electronics, electrical equipment and machinery (including automotive products), as well as tourism. Ultimately, the team’s recommendations were incorporated in GoSL strategies for investment promotion, export development, and economic diplomacy; extensions of the research were also used to help plan new export processing zones and target potential anchor investors.
This report summarizes the methodology and findings of the Targeting Team, including scorecards for each of the sectors studied.
Economies grow by adding new products and services to their production portfolio, not by producing more of the same kinds of products. The key to such diversification is access to know-how, but know-how often has to come from abroad. This is because it is often easier to move brains to new countries than to move new know-how into brains. In the experience of Singapore, India, Vietnam and most other dynamic economies, three channels of know-how transfer stand out: FDI, immigration and diaspora networks.
In this lecture, Professor Hausmann explores the relationship between economic development and the accumulation of know-how. In particular, he discusses how to tackle Sri Lanka’s limited export diversification.
Video - Acessing Know-how for Growth in Sri Lanka
Video - Full Q&A on Sri Lanka's export diversification
This note collects evidence related to possible constraints to economic growth, and their relation with GoSL’s industrial zone development agenda. We find that new zones are especially well-suited to help address Sri Lanka’s lack of industrial land and high policy uncertainty, both of which may be holding back growth. Less clear, however, are zones’ impact on Sri Lanka’s limited transport links beyond the Western Province. Finally, partnering with well-connected zone management companies may also help create opportunities to connect with firms in new, non-traditional sectors.
International aid is a complex system: it involves different issues, countries, and donors. In this paper, we use web crawling to collect information about the activities of international aid organizations on different health-related topics and network analysis to depict this complex system of relationships among organizations. By systematically collecting co-occurrences of issues, countries, and organization names from more than a hundred websites, we are able to construct multilayer networks describing, for instance, which issues are related to each other according to which organizations. Our results show that there is a surprising amount of homophily among organizations: organizations of the same type (multilateral, bilateral, private donors, etc.) tend to be co-cited in groups. We also create a taxonomy of issues that are generally mentioned together. Finally, we perform simulations, showing that messages originating from different organizations in the international aid community can have a different reach.
Setting a country’s structural growth rate on a higher path, i.e. sparking and sustaining a growth acceleration can have quantitatively huge implications for national income and, more broadly, for people’s wellbeing. We develop a novel statistical framework to identify systematically the set of binding constraints that were unlocked before the 135 growth acceleration episodes that took place between 1962 and 2002 worldwide. We employ this information to characterise the acceleration process, which tends to be preceded by a deep recession and major economic policy changes. Once we combined this information with a set of counterfactual analyses, we find however that successful acceleration strategies should not contain off-the-shelf approaches or necessarily all-encompassing “shock therapy” solutions. On the other hand, they call for a careful tailoring to local conditions. Richer countries tend to experience fewer accelerations, but once these have been ignited, they are better positioned to make the most out of them. Despite standard growth determinants doing a fairly good job at characterising successful accelerations, we note how take-offs remain extremely hard to engineer with a high degree of certainty.
Using a large individual-level survey spanning several years and more than 150 countries, we examine the importance of social networks in influencing individuals' intention to migrate internationally and locally. We distinguish close social networks (composed of friends and family) abroad and at the current location, and broad social networks (composed of same-country residents with intention to migrate, either internationally or locally). We find that social networks abroad are the most important driving forces of international migration intentions, with close and broad networks jointly explaining about 37% of variation in the probability intentions. Social networks are found to be more important factors driving migration intentions than work-related aspects or wealth (wealth accounts for less than 3% of the variation). In addition, we nd that having stronger close social networks at home has the opposite effect by reducing the likelihood of migration intentions, both internationally and locally.
Throughout 2016, CID conducted a growth diagnostic analysis for Sri Lanka in collaboration with the Government of Sri Lanka, led by the Prime Minister’s Policy Development Office (PDO), and the Millennium Challenge Corporation (MCC). This presentation report aggregates collaborative quantitative and qualitative analysis undertaken by the research team. This analysis was originally provided to the Government of Sri Lanka in April 2017 in order to make available a record of the detailed technical work and CID’s interpretations of the evidence. A written executive summary is provided here as a complement to the detailed presentation report. Both the report and the executive summary are structured as follows. First, the analysis identifies Sri Lanka’s growth problem. It then presents evidence from diagnostic tests to identify what constraints are most responsible for this problem. Finally, it provides a summary of what constraints CID interprets as most binding and suggests a “growth syndrome” that underlies the set of binding constraints.
In brief, this growth diagnostic analysis shows that economic growth in Sri Lanka is constrained by the weak growth of exports, particularly from new sectors. Compared to other countries in the region, Sri Lanka has seen virtually no diversification of exports over the last 25 years, especially in manufactured goods linked through FDI-driven, global value chains. We found several key causes behind this lack of diversified exports and FDI: Sri Lanka’s ineffective land-use governance, underdeveloped industrial and transportation infrastructure, and a very high level of policy uncertainty, particularly in tax and trade policy. We believe that these issues trace back to an underlying problem of severe fragmentation in governance, with a critical lack of coordination between ministries and agencies with overlapping responsibilities and decision-making authority.
The fact that firms benefit from close proximity to other firms with which they can exchange inputs, skilled labor or know-how helps explain why many industrial clusters are so successful. Studying the evolution of coagglomeration patterns, we show that which type of agglomeration benefits firms has drastically changed over the course of a century and differs markedly across industries. Whereas, at the beginning of the twentieth century, industries tended to colocate with their value chain partners, in more recent decades the importance of this channels has declined and colocation seems to be driven more by similarities industries' skill requirements. By calculating industry-specific Marshallian agglomeration forces, we are able to show that, nowadays, skill-sharing is the most salient motive in location choices of services, whereas value chain linkages still explain much of the colocation patterns in manufacturing. Moreover, the estimated degrees to which labor and input-output linkages are reflected in an industry's coagglomeration patterns help improve predictions of city-industry employment growth.
Macroeconomic adjustment in the euro area periphery was more recessionary than pre-crisis imbalances would have warranted. To make this claim, this paper uses a Propensity Score Matching Model to produce counterfactuals for the Eurozone crisis countries (Greece, Portugal, Ireland, Cyprus, Spain) based on over 200 past macroeconomic adjustment episodes between 1960-2010 worldwide. At its trough, between 2010 and 2015 per capita GDP had contracted on average 11 percentage points more in the Eurozone periphery than in the standard counterfactual scenario. These results are not dictated by any specific country experience, are robust to a battery of alternative counterfactual definitions, and stand confirmed when using a parametric dynamic panel regression model to account more thoroughly for the business cycle. Zooming in on the potential causes, the lack of an independent monetary policy, while having contributed to a deeper recession, does not fully explain the Eurozone’s specificity, which is instead to be identified in a sharper-than-expected contraction in investment and fiscal austerity due to high funding costs. Reading through the overall findings, there are reasons to believe that an incomplete Eurozone institutional setup contributed to aggravate the crisis through higher uncertainty.
The recent economic depression in Greece hit the population of Albanian migrants in Greece particularly hard, spurring a wave of return migration that increased the Albanian labor force by 5% in less than four years, between 2011 and 2014. We study how this return migration affected the employment chances and earnings of Albanians who never migrated. We find positive effects on the wages of low-skilled non-migrants and overall positive effects on employment. The gains partially offset the sharp drop in remittances in the observed period. An important part of the employment gains are concentrated in the agricultural sector, where most return migrants engage in self-employment and entrepreneurship. Businesses run by return migrants seem to pull Albanians from non-participation, unemployment and subsistence agriculture into commercial agriculture.
Who introduces structural change in regional economies: Entrepreneurs or existing firms? And do local or nonlocal establishment founders create most novelty in a region? We develop a theoretical framework that focuses on the roles different agents play in regional transformation. We then apply this framework, using Swedish matched employer–employee data, to determine how novel the activities of new establishments are to a region. Incumbents mainly reinforce a region’s current specialization: incumbent’s growth, decline, and industry switching further align them with the rest of the local economy. The unrelated diversification required for structural change mostly originates via new establishments, especially via those with nonlocal roots. Interestingly, although entrepreneurs often introduce novel activities to a local economy, when they do so, their ventures have higher failure rates compared to new subsidiaries of existing firms. Consequently, new subsidiaries manage to create longer-lasting change in regions.
Although Venezuela’s experience since the 1980s would seem to make it a classic example of the resource curse, that perspective fails to explain the country’s impressive economic, social and institutional performance—including healthy public debate—during the first five decades after oil was first produced on a large scale. This chapter takes a long view of the Venezuelan experience and argues that this initial performance was lengthy and positive but fragile, given the incapacity of the country’s institutions to adapt to the different environment that developed afterwards, characterized by high oil price volatility and significant and sudden declines in oil revenues. The prolonged initial period of equilibrium became a curse of sorts. Lacking adaptive efficiency, political institutions were forced to rely increasingly on maintaining an illusion of harmony, and as faltering performance became evident, Venezuelans began questioning the model and its hegemonic arrangements. The scope and magnitude of the economic, social and institutional devastation that followed were such that public debate became one the first casualties. One of the main problems in contemporary Venezuela is the polarization of politics. This makes it difficult for the country’s population to arrive at reasonable solutions through public discussion.
Many countries, like Sri Lanka, are trying to diversify their economies but often lack the capabilities to lead diversification programs. One of these capabilities relates to targeting new sectors to promote and pursue through a diversification policy: countries know they are ‘doomed to choose’ sectors to target,1 but lack effective capabilities to do the targeting. This paper narrates a recent (and ongoing) initiative to establish this kind of capability in Sri Lanka. The initiative adopted a Problem Driven Iterative Adaptation (PDIA) process, where a team of Sri Lankan officials worked with Harvard Center for International Development (CID) facilitators to build capabilities. The paper tells the story of this process, providing documented evidence of the progress over time and describing the thinking behind the PDIA process. It shows how a reliable targeting mechanism can emerge in a reasonably limited period, when a committed team of public officials are effectively authorized and engaged. The paper will be of particular interest to those thinking about targeting for diversification and to those interested in processes (like PDIA) which are focused on building state capability and fostering policy implementation in public contexts.
1 The term here comes from Hausmann, R. and Rodrik, D. 2006. Doomed to Choose: Industrial Policy as Predicament. Draft.
Many countries, like Sri Lanka, are trying to diversify their economies but often lack thecapabilities to lead diversification programs. One of these capabilities relates to engaging new investors—in new sectors—to bring their FDI and know-how to a new country and kick-start new sources of activity. This paper narrates a recent (and ongoing) initiative to establish this kind of capability in Sri Lanka. The initiative adopted a Problem Driven Iterative Adaptation (PDIA) process, where a team of Sri Lankan officials worked with Harvard Center for International Development (CID) facilitators to build capabilities over a six-month period. The paper tells the story of this process, providing documented evidence of the progress over time (and describing thinking behind the PDIA process as well). It shows how an investment engagement approach can emerge in a reasonably limited period, when a committed team of public officials are effectively authorized and engaged. The paper will be of particular interest to those thinking about investor engagement challenges and to those interested in processes (like PDIA) focused on building state capability and fostering policy implementation in public contexts.
Many countries, like Sri Lanka, are trying to diversify their economies but often lack thecapabilities to lead diversification programs. One of these capabilities relates to preparing the investment climate in the country. Many governments tackle this issue by trying to improve their scores on ‘Doing Business Indicators’ which measure performance on general factors affecting business globally (like how long it takes to open a business or pay taxes). Beyond these common indicators, however, investors face context specific challenges when working in countries like Sri Lanka that are not addressed in global indicators. Governments often lack the capabilities to identify and resolve such issues. This paper narrates a recent initiative to establish these capabilities in Sri Lanka. The initiative adopted a Problem Driven Iterative Adaptation (PDIA) process, where a team of Sri Lankan officials worked with Harvard Center for International Development (CID) facilitators to build capabilities over a six-month period. The paper tells the story of this process, providing documented evidence of the progress over time (and describing thinking behind the PDIA process as well). The paper will be of interest to those thinking about the challenges associated with creating a climate that is investor or business friendly and to those interested in processes (like PDIA) focused on building state capability and fostering policy implementation.
Sri Lanka has an excessively complex tariff structure that distorts the structure of the economy in important ways. It is a priority for the Government of Sri Lanka (GoSL) to rationalize the system in order to facilitate a transition to greater economic diversification, stronger export growth, and the emergence of new, higher paying jobs. Sri Lanka’s New Trade Policy makes this tariff rationalization a priority. It also recognizes that tariff rationalization should go hand in hand with new trade adjustment assistance measures to support the adjustment of firms and of people. The New Trade Policy outlines the basic contours of tariff rationalization and trade adjustment assistance measures but does not provide a detailed roadmap.
This discussion paper was prepared at the invitation of the Ministry of Development Strategies and International Trade (MoDSIT) as part of the Center for International Development’s research project on sustainable and inclusive economic growth in Sri Lanka. The aim of the paper is to study policy tools that the GoSL could use to structure trade adjustment assistance in the context of tariff rationalization. In order to accomplish this aim, we begin by outlining the type of tariff rationalization that needs to take place in order to address key constraints to growth in a way that is sensitive to both government revenue needs and political economy considerations. We stress that tariff rationalization must be approached in a holistic way that treats the various tariffs and para-tariffs as interrelated, rather than an approach that attempts to address one part of the system at a time. A holistic approach would provide many degrees of freedom to solve the underlying problems in the system while increasing revenues and potentially generating strong public support. Critically, a holistic approach would allow for a single tariff rationalization plan to be phased in over a period of years in a predictable way, whereas attempts to rationalize the system one part at a time would lead to extreme uncertainty.
With the principles of smart tariff rationalization in place, we draw upon international lessons and Sri Lanka’s own institutional capabilities to recommend a two-tiered approach to helping industries and workers adjust. In each case, the first tier represents low-cost measures that can begin in the short term to help industries and workers, regardless of whether they will be negatively impacted by tariff rationalization, while the second tier of assistance applies only to trade-affected industries and workers and can be developed in the medium term. For industries, Tier 1 support involves the use of an innovative process of public-private problem solving of industry-specific constraints, and Tier 2 support involves the use of special safeguard measures to provide an objective and transparent process for determining which industries require longer phase out periods for tariff reductions versus the tariff rationalization plan. For workers, Tier 1 support involves improved access labor market information and training opportunities through the development of regional (or local) job centers. Tier 2 support provides government funding for training and job placement services. We conclude that this package of trade adjustment assistance measures could be used to complement a holistic tariff rationalization plan. But we caution that attempts to rush the implementation of these measures without careful design and communication could deeply undermine the potential for the reforms to work in solving underlying economic problems.
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.
How can the productive capabilities of each municipality be unleashed taking into consideration the resources available to them? A first pass at this ambitious question begins by understanding the set of outputs a municipality is capable of producing. We answer this by discovering relationships between agricultural inputs and outputs and ask a relatively simpler question: how similar are agricultural outputs in terms of the inputs they use? Answering this question is made difficult by the fact that most UPAs cultivate just one or two crops. This may be a rational response to economies of scale. Given a plot of land and inputs, it may be easier to cultivate one crop on the entire land than plant a number of them with each requiring a different care regimen3. It may be that the inputs available only allow for a few types of crops.
In this paper, we use the rural census data from Colombia to build an agricultural product space capturing the similarities between outputs. We test the predictive power of the product space and use this to answer the question above. In section 2, we discuss the various sources of data and how they are merged, cleaned, and transformed before processing. In section 3, we look at some high level features of the dataset and how inputs, outputs, and land use are related. In section 4, we explore the mechanics of diversification.
We construct similarity and density matrices and show that they do indeed predict what a municipality produces. Finally, in section 5 we use Machine Learning algorithms and the density matrices to predict municipalities that are best suited to produce a given output. Further, we identify "missing" municipalitiesoutput pairs i.e. municipalities that should be producing a given output at high yield but currently are not. Finally, we summarize our findings and suggest areas for further work.
