2017 Bogota Workshop: Abstracts

DECEMBER 10-16, 2017


Clicking a link will scroll the page to the relevant section:
 | Accad  | Batheja  | Belmonte  | Espinoza  | Franco  | Ge  | Huaroto  | Huo  | Leung  | Majid  | Markgraf  | Martins  | Molina  | Oh  | Ostler  | Romero Haaker  | Rozencwaig-Feldman  | Su  | Szmurlo  | van der Merwe  | Wang  | Wiebe  | Xiang  | Yu  | Zhang | 

Can We Discipline Politicians with Competition?
Institutional Conditions Affecting Electoral Accountability
Mel Lorenzo M. ACCAD
University of Hawaii

Can election competition (1) be a credible threat on incumbent politicians to lessen fiscal wastage, (2) or encourage more fiscal wastage due to less incentive of staying in power? Fiscal wastage is defined and measured by the amount of money involved in unresolved audit findings. This is the first study to use the 2006-2015 panel data of audit reports from the Philippines.

We hypothesize that the impact of election competition highly depends on quality of the institutions, more specifically (1) the barriers to having a viable competitor, (2) preference of voters to the level of public goods vs. private goods (one form is vote buying), and (3) the monopoly of violence by the incumbent politician/political clan. All of these hypothesized factors may be revealed in the election results.

Because election competition cannot be randomly assigned in the actual electoral process, this study plans to implement a Randomized Encouragement Design (RED) in 2019 Philippine local elections. We will randomize a household-level intervention of information dissemination about the incumbents’ previous unresolved audit findings on two sets of municipalities. The first set is composed of municipalities ruled by politicians who previously won by very small vote margin, while the second set by very large vote margin. We will also measure the qualities of institutions (both informal and formal) by asking the households about their perceptions and experiences on public goods provision, vote buying, violence, trust, approval of the incumbent, and many more.

Preliminary (not RED) results from a panel data of 78 provinces from 2006 to 2015 show that election competition does not have any impact on the fiscal wastage after controlling for politician, province, and year fixed effects.

Can Familiarity Improve Gender Attitudes?
Gender Peer Effects in the Labor Market in India
Deepshikha BATHEJA
University of California - Riverside

Gender segregation is practiced in many traditional and patriarchal societies like India. In such a scenario, for many boys and girls, the very first prolonged interaction (as equals) with the opposite gender outside of family members, happens in a workplace. The gendered division of the society can increase the cost of communication among employees in a gender diverse work place and might have a negative effect on productivity, especially in team performances due to co­ ordination and free riding problems. The objective of the paper is to ascertain if there are any economic costs (in terms of productivity losses in the labor market), associated with the gender segregation policies practiced in many educational and social institutions.

Despite most industries and occupations being male dominated in India, Business Process Outsourcing (BPO) sector employs large number of female employees due to their comparative advantage in interpersonal skills and most employees are hired straight after high school. This paper will use this setting and through a quasi­experimental design to study how economic outcomes of labor productivity differ in the face of gender diversity at workplace. The random seating assignment rule will be used as the identifying variation for exposure to opposite gender. It is expected that in a gender segregated society like India, the initial exposure to the opposite gender will cause decreases in work productivity in the short run but favorable changes in gender attitudes in long-term. To understand the underlying implications of gender segregation on gender attitudes, the analysis will be supplemented by a survey involving implicit association tests (IATs) to assess if there is a change/improvement in gender attitudes if seated with employees of the opposite gender.

Recalling Negative Collective Memories:
Evidence from Russian Politics
Alessandro BELMONTE and Michael Rochlitz

For over a century, historians, sociologists and psychologists have been studying the ways in which societies remember, represent and interpret the past. Commemoration ceremonies and the building of monuments, tradition and myths have been set up to transmit, preserve and alter the cultural and national identity of societies, shaping communities of memory. Scholarly debate has been dominated, in particular, by the relation between history and memory, and on the extent to which modern societies use history at the service of memory. Hundreds of events make headlines each year, but only very few among them contribute to create the collective memory of a society. What factors determine the selection of such events?

In this paper we develop a political economy model where the autocrat strategically and selectively recalls negative collective memories to increase his political support. Citizens decide whether to support the autocrat, or to initiate a revolution that will transit the political regime to a democracy, their political reference point. We find that autocrats prevent the revolution by reactivating negative collective memories that convince the citizenry that the democratic political alternative, their political reference point, is economically inferior to the autocratic status quo.

We test this prediction using data from Russian politics. After the end of the Soviet Union, Russia experienced a period of economic and political transition that was characterised by serious economic difficulties, political chaos and social disruption. We exploit regional variation in the magnitude of changing in the cost of living following the collapse of the Soviet Union and in the recollection of this negative collective memory disseminated through local newspapers. We find that citizens protest to a lesser extent while voting more for the government party in regions in which past recollection of negative collective memories from the 1990s is more intense, even controlling for the extent of past economic and social trauma.

Informal Institutions and Collective Action in the Traditional
Food Retail Sector in Lima: How Is Internal Governance Shaped?

This article seeks to conceptually and empirically analyze the institutional elements involved in the informal governance and collective action of traditional retail food markets in Lima, Peru. By analyzing the relationships between historic background, social capital, trust, and enforcement practices, this paper seeks to clarify the institutional factors and mechanisms that shape the collective-action capacity of these food supply centers. In a city with more than 1,200 traditional markets (accounting for around 110,000 traders) operating in contexts of high informality, the ability of markets – as a collective - to implement joint actions or investments (common goods) to attract customers could affect the competitiveness and profitability of the whole sector.

This study builds up on an ongoing research project conducted by GRADE and the IADB, whose goal is to analyze the relationship between institutions and productivity in an informal setting. So far, the project has established some preliminary causality relationships between historic background and collective action, and between collective action and profitability. Nevertheless, I have noticed that it should be possible to explore the mechanisms through which these historical factors affect the collective-action capacity of these traditional markets. This understanding may help to connect the theoretical elements identified in the institutional literature with the empirical evidence we have on the performance of the traditional markets.

Analyzing such complex relationships requires abundant information. Early this year, the study has collected detailed information from 95 traditional markets, more than 1,000 shopkeepers, and about 1,400 customers. The information was gathered through the application of five instruments, including surveys (on shopkeepers, customers, and leaders), observation guides (on the common goods and services of the establishment), and economic games sessions with shopkeepers (trust game and public goods game). These instruments allowed us to obtain plenty information on the historic background of the markets (provenance of the merchants, form of initial organization, among others), the social capital (social events) and trust (trust game and questions) of the merchants, the enforcement practices within the market, and evidence of collective action (investment in common goods and services). I have processed all this information and it is ready to be analyzed.

Long-Term Effects of the Inca Roads on Peru’s Current Development
Ana FRANCO, Pablo Lavado, Ximena Medina

What are the long-term effects of the Inca roads on Peru’s current development? The Incas were the last and largest native pre-colonial culture of South America (15th century). Current territory of Argentina, Bolivia, Chile, Colombia, Ecuador and the whole Peru were part of it. Their transportation system was called the Inca Road. It was about 30 thousand kilometres long. In times of the Inca Empire, the road served mainly for religious and military purposes (expansion). It was also a sign of the omnipresence of the empire. Regardless its importance for the Inca Empire, its impact on current Peru has not been explored in the economic history.

Our identification strategy is a multidimensional regression discontinuity in latitude and longitude as in Dell (2010). Our treatment group are the districts crossed by the road and the control group are districts located just beside the treated and whose capital is within a range of 15 km or 20km. A concern with an RD is whether other characteristics beside treatment vary smoothly at the discontinuity. We do not have data on development for the 15th century and this is a major concern. But we have data on altitude and slope that are the main determinants of climate and, therefore, agriculture (the main economic activity in pre-colonial Peru). We do not find statistically differences between the treatment and control groups on these variables.

To estimate the effect, our treatment is at the district level and our dependent variables of interest are consumption and educational attainment. These come from the National Household Survey of 2007. Our results show that the long-term effect of the Inca road increases consumption by 30% and increases educational attainment by 1.2 years (the mean is 6.9 years). Now, why? What is the mechanism?

Our hypothesis is that tambos are the mechanism. During the Inca Empire, tambos were small warehouses built and managed by the Incas to support the chasquis (their messengers) and their troops with food and water. These were across the road every 20 km and were not intended for commercial purposes. However, during the Spanish conquest, this changed. They established a trade regime, and the road was used for this purpose. In this line, the tambos became private points of sale and districts benefit from trade around it. Furthermore, according to new Ordinance of Tambos, their owners had to benefit the people from their district with the profits obtained through public goods. This did not happen in districts not crossed by the road. Therefore, a virtuous circle of economic development based on trade and provision of public goods was established in the road districts which led us to expect different paths of development

Does Media Reporting on Corruption Have a Destabilizing Effect?
Xincheng GE
Hong Kong University of Science and Technology

How do people interpret corruption information? This study uses corruption news released by the anti-corruption campaign in China that began at the end of2012 to causally examine its impact on people's self-reported perceptions of social mobility and fairness. Using the generalized difference-in-differences approach, this study finds that corruption news did not affect views on social mobility, but deteriorated views on fairness. The results are robust after controlling for regional cultural differences and local corruption severity. The possibility of reversal causality is excluded by controlling for prior corruption awareness and by testing for the pre-trends. To examine the channel of knowledge of corruption, the Baidu Search Index with fanfu (anti-corruption) is used as a proxy for corruption awareness. As a result, this study finds that corruption exposure increased corruption awareness, and that more increase in awareness is correlated with more negative attitude change. This suggests that the anti-corruption campaign updated people's knowledge of corruption, and that the campaign might have had an unintended destabilizing effect. The study uses President Xi Jinping's news to refute an alternative explanation for the effect of other information. The placebo test on people's attitudes toward sex further shows the robustness of the results.

Understanding How Bad Institutions Affect Today's Development:
The Case of the Mining Mita
Universidad Nacional de la Plata

History matters for a better understanding of present development. But how do they still matter now? A growing and solid body of literature in the long-run effects of historical events or institutions has expanded on recent years, but, there is fewer evidence on what are the mechanisms of them on the present and how do they interact with positive and negative economic shocks.

I study the case of the Mining Mita in Peru, a colonial institution created on the Viceroyalty of Peru in 1573 to force local indigenous to work on the Potosi and Huancavelica mines. The formal institution was eliminated in 1812 but its consequences can be traced down to the present.

I use Geographic Regression Discontinuity as identification strategy and data of Household Surveys and Night Luminosity from Satellite as local development indicators to estimate the impact of the Mita for the last 25 years and find that the point estimate is not static, but instead it seems correlated with economic cycles.

Additionally, I present tentative evidence on the mechanisms behind this dynamic, and they suggest that the reasons behind this are related with present day political instability and corruption.

Why Do Sellers Choose Buyers on Airbnb?
Derek Da HUO
The University of Hong Kong

A rather unique feature of Airbnb is that the hosts (sellers) choose the guests (buyers). Although in many markets it is very common that buyers choose sellers for pursuing a better deal, yet sellers usually have few incentives to reject buyers. Even in the hotel industry whose business model is similar to Airbnb, the hotel managers seldom screen out guests. The question arises why there exists such a choice mechanism on Airbnb that the hosts (sellers) choose the guests (buyers). This research will answer it from the perspective of incomplete contracts.

Recent technological progress in information technology has tremendously reduced transaction costs, particularly between traders who are geographically far apart. One institution that emerges as a result is Airbnb, where customers rent accommodation spaces and associated services from property owners over the Internet. However, unlike the transactions in a mature market such as the hotel industry, the contracts between hosts and guests on Airbnb are relatively incomplete. The reasons are (1) Airbnb is faced with legal issues as it violates the regulations on short-term leasing in many cities, and (2) the damage to housing is hard to measure because of the housing heterogeneity. Thus, it is costly to settle the disputes on Airbnb. This research argues that it is because the contracts are so incomplete on Airbnb that the hosts voluntarily screen out guests in order to avoid disputes.

The two arrangements between which hosts can freely switch provide the possibility to empirically test my argument. One arrangement is “instant booking" where bookings are instantly accepted without any filtering of customers. The other is “filter booking” where the host’s permission is needed for a booking. This research will test whether hosts tend to adopt “filter booking” when the contracts are more incomplete.

Self-Enforcing Mechanism of Informal Housing in Hong Kong
Ka-man LEUNG
The University of Hong Kong

Different from subleasing and co-housing, sub-divided unit (SDU) is a form of informal housing in Hong Kong which supplies low-cost rental units by subdivision of individual quarters into two or more flats illegally. Despite tenure insecurity and substandard housing quality, SDU is highly demanded by the urban poor because of the affordable housing shortage in the formal sector. As at 2015, there are 199,900 people of this affluent city living in SDUs. Although informal settlements have been drawing scholarly attention, limited studies focus on the choice of contract of the informal market. Most of the SDU tenancy agreements are unregistered and arranged verbally. Both landlords and tenants face difficulties in enforcing the tenancy agreements as the Court may be incapable of handling disputes. In other words, SDU tenancy arrangements are self-enforcing, which exist when it pays the parties to live up to them. Tenants obtain low-cost shelters while the landlords obtain excess return by the illegality of the building works and foregoing the socially acceptable minimum living standard.

The excess return to the landlords in the nature of quasi-rent is temporary. It exists when there is shortage of affordable units in the formal sector. It is a function of the rent gap, enforcement of law and self-enforcing mechanism, which arises because of the high information and transaction costs in the informal market. Self-enforcement mechanisms, devised by the landlords such as special agency arrangement and collection of key money by overcharging the utilities etc., emerge to safeguard the quasi-rent. The choice of mechanism by the landlords is contingent on the demand elasticities of the tenants. This study aims to examine the factors affecting the contractual arrangement and self-enforcing mechanism of SDU, which in turn affect the size of quasi-rent hence the supply of SDU. The significance of this study gives fuller account of the choice of contract in the informal housing market and provides thorough analysis on the informal housing supply.

Keywords: choice of contract, self-enforcing mechanism, informal housing, sub­divided units

The Long-Term Effect of Early Life Exposure to Cow Slaughter Bans
Muhammad Farhan MAJID and Wafa Hakim Orman
Rice University

We exploit an ongoing natural experiment from the rollout of legislation banning cow slaughter across states in India to study the long-term effect of disruptions in beef intake during pregnancy on the next generation. Reduced consumption of iron rich animal proteins is likely to be particularly harmful for pregnant women, who have a significantly greater need for iron. Indian women in particular may be most vulnerable as over 50% of Indian women suffer from at least mild to moderate anemia. Using a triple difference in difference strategy which exploits variation across religious and caste groups which do and do not consume beef, time (cohort) variation in implementation of laws over more than 40 years from time of 1950s to 1990s, in different states of India, we find that overall, women exposed to cow slaughter bans in their year of birth have lower levels of hemoglobin (Hb) and are more likely to be anemic in their prime reproductive ages between 15 and 35, particularly those who have not completed primary schooling or who come from poorer families.

We use data from a nationally representative survey - India's DHS 2005-2006 - for this purpose. These impacts are evident in our triple difference-in­difference models and are robust to the inclusion of linear state specific time trends and an array of SES variables including own and partner's education and a wealth index. We don't find similar effects for men. This is consistent with gender differential in parental investment response to early life insults.

For robustness we explore a range of similar policies likely to restrict red meat consumption - beef sale bans, beef possession bans, export of beef bans and buffalo/bull slaughter bans, which may restrict beef consumption among the beef consuming groups and find evidence in support of our findings. We conclude that cow slaughter bans can have long-term harmful effects, particularly, for women of reproductive age among minorities, which historically consume beef (scheduled caste Hindus, Christian and Muslims in particular). To the best of our knowledge, this is the first paper to document the long-term harmful health effects of laws restricting beef consumption in India.

Borrowing Welfare: Effects of Credit Access on Redistributive Preferences
Jonas MARKGRAF, Guillermo Rosas, and Sebastian Lavezzolo
Hertie School of Governance

In many advanced democracies, support for redistribution is decreasing although income inequality is growing. The empirical observation stands in direct contrast to the Meltzer-Richard model that predicts that the median voter will demand more redistribution if income inequality is increasing because she stands to gain more from redistribution when income is more unequally distributed; the model, therefore, predicts a self-correcting mechanism in democracies that balances growing income inequality through fiscal redistribution. The puzzling finding of decreasing demands for redistribution in the light of increasing income inequality has, therefore, inspired many scholars.

Recently, role of credit to the poor arouse academic interest, in part because of the devastating consequences of the US real estate boom that was fueled by mortgage credit to subprime borrowers. Scholars across academic disciplines argued that the credit boom in the US was politically motivated and supported by loose financial regulation. The expansive provision of credit in the US was arguably a politically convenient instrument to replace costly fiscal redistribution: in that way, more consumer credit -- not a more generous welfare system-- was the response to growing income inequality.

Thus, scholars have offered theoretically and empirically convincing supply-side evidence for the political drivers of credit growth. An analysis of the demand-side effects of credit access and credit growth is, however, missing and it remains unclear if voters accept credit as a substitute for redistributive policies.

This project, therefore, studies the relationship between access to consumer credit and preferences over redistribution in Europe; we shed light on the demand-side effects of credit by quantitatively analyzing the link between credit availability and redistributive preferences using European cross-country survey data over a span of eight years. We estimate a hierarchical model and find evidence for the claim that better access to credit is associated with lower preferences for fiscal redistribution; the relationship between credit access and attitudes toward redistribution, however, exhibits considerable variation over time and across countries. Currently, we are working on a theoretical model that accounts for this variation.

Buying Cheaply and Not Paying Dearly?
E-Procurement, Cost-Efficiency, and Contract Execution in Public Sector
Darcio Genicolo MARTINS

Do the adoption and use of e-procurement platforms improve or not the efficiency of public procurement process? The expected benefits derived from public e-procurement platforms (PeP) include a wider choice of suppliers, better quality, improved delivery, and reduced cost. PeP enhances efficiency through two main avenues: (a) transaction cost savings and (b) reduced direct procurement costs, including prices. Since the procurement process becomes more transparent and faster, this makes less likely collusive behavior and corruption. On the other hand, contract incompleteness is very present in public procurement, mainly in contracts on services. Depending on the degree of contract incompleteness lower prices could induce poorer quality execution of contracts (renegotiation, more contractual additives, delays, sanctions, judicial disputes etc.); which can generate relevant costs to the public. Therefore, there is a trade-off between auction prices and quality in the execution of public procurement contracts. Very low Prices in the auctions are not necessarily good for the social welfare.

Using a unique database from Secretary of Finance (Sao Paulo state), this paper aims to estimate the direct effects of the e-procurement adoption and its usage in Sao Paulo state municipalities: (i) prices (cost-efficiency measure) and, (ii) delivery time (quality of contract execution measure). To estimate prices and delivery time differentials (before and after joining PeP), we use propensity score matching (comparing similar municipalities) and Differences-in-Differences methods. The main contribution of this paper is to estimate the trade-off between cost-efficiency and quality of contract execution in different types of public contracts.

Subnational REDO+ Programs, Institutions and Incentives:
The Case of Brazil
Adriana MOLINA
University of Colorado - Boulder

The forests coverage of Latin America has shrunk by roughly 10% between 1990 and 2015, making the region's deforestation rate one of the highest in the world (World Bank Indicators). Different initiatives have been implemented to preserve existing forests and reduce continued deforestation. One such program is Reducing Emissions from Deforestation, forest Degradation and enhancement of forest carbon stocks in developing countries (REDO+) from the United Nations Framework Convention on Climate Change (UNFCCC). Despite its popularity, questions about the sustainability of the program remain, especially considering the slow transition towards carbon markets by countries in Latin America implementing REDO+. One of the main challenges is the multilevel governance dynamic created by REDO+, in which both national and subnational institutions and incentives should match in order to provide an adequate framework for the reduction of deforestation.

Brazil's case is relevant to this analysis since it provides nearly two-thirds of the total Amazon area (CIFOR, 2016) and its decentralization process has led to the implementation of several subnational REDO+ programs, along with an increasing number of local responsibilities including land regularization actions. This research project focuses on evaluating the effect of subnational institutions, such as land tenure security over the household's likelihood and degree of forest exploitation, using CIFOR'S household data in municipalities where REDO+ is implemented. I evaluate whether the positive effect of REDO+ programs at the subnational level is conditional on the existing types of land rights at the local level, and the degree of perceived security over those titles for the program's beneficiaries. In addition, I assess the alignment of the incentives provided by the national and subnational institutions using the Institutional Analysis and Development (lAD) framework (Ostrom, 2005), including the role of the REDO+ subnational programs in Brazil. A better understanding of how REDO+ and its specific instruments influence the action arena could provide answers with regard to the governance dynamic and the consequences over the individual's behavior with respect to the forests and their incentives to comply with the rules of use and management.

Radio “Fences” and Inventor Attention to Property Rights:
Evidence from Wireless Patents
Sarah OH
Technology Policy Institute

Do inventors of radio “fences” care about property rights? I hypothesize that inventors do pay attention to policy regimes such as licensed and unlicensed spectrum when building technologies to handle interference. This inquiry follows from Ronald Coase’s claim that property rights align incentives for efficient spectrum use. To answer this question, I conduct textual analysis on 500,000 wireless patents to estimate inventor attention to property rights. Indeed, since the Radio Act of 1927, inventors have filed hundreds of thousands of patents related to radio technology.  I find that inventors are more likely to discuss interference and sensor technology if they discuss property rights, than if they do not discuss property rights. Inventors are less likely to discuss filter technology if they discuss property rights, than if they do not discuss property rights. While further research is needed to determine causality, textual analysis provides empirical evidence of inventor attention to property rights and policy regimes.

JEL: K11, K23, L51, L96, O31, P14, P48. Keywords: property rights, spectrum policy.

How Do Land Property Rights Influence Alternative Development Efforts
in Illegal Crop Farming Communities in Colombia?
King’s College London

Effective land property rights bring benefits that result in increased economic productivity. But to what extent is the legal economic productivity generated by property rights sufficient to displace the illegal use of land? This study examines how the nature and impact of different systems of property rights influence the incentives for illegal crop substitution in Colombia. Based on observations at the community level and other primary data sources, it will compare the distribution and enforcement mechanisms of informal, transitional i.e. recently formalized, privately owned and collectively owned land. The data will illustrate how each system of property rights, shapes a community's access to credit, efficient allocation of resources and levels of external investment. Of interest is the role that informal institutions play in the effectiveness of formal property rights, and the interaction between the perception of fairness in distribution and large-scale capital interests in the legal use of land. 

On the Consequences of Electing Criminal Politicians for Public Office:
Evidence from Peru
Duke University

This paper aims to analyze the consequences of electing criminal politicians for public office as a means of understanding why weak governments persist over time. An important body of evidence shows that the identity and characteristics of political leaders matter for the type and efficacy of implemented policies, while another strand of the literature documents the long-lasting effects of institutions on economic development. This study uses a unique data set with information on the characteristics of every politician running for public office in Peru, including criminal records and political histories, that is matched to detailed project-level government spending records and infrastructure data. Using a regression-discontinuity design to gain causal identification, I document that that criminal politicians expend considerably more public funds than non-criminals; however, the quantity and quality of the supplied public goods remains virtually the same. Anecdotal evidence suggests that an incentive mismatch may be at play: given the lack of local tax collection and pervasive economic informality, the electorate may not be facing the economic distortions associated with electing criminal politicians and thus fails to hold them accountable. Future steps of this project will test for this and competing mechanisms in detail, as well as document effects on well-being measures.

The Author and the Other: Collaboration and Ownership in Copyright Law
Bar Ilan University

To what extent does the joint authorship doctrine fulfill its objectives in today's digital age? Over the last years we have witnessed an increase in the importance and prevalence of the joint authorship doctrine due to the Internet evolution and globalization processes that allow the development of technological platforms as well as quick and simple capabilities to share content and information between various creators from around the world. Today, many types of works such as songs, movies, software, and computer games are created on a regular basis through joint authorship comprising a number of creators working together collaboratively. Notwithstanding, the current copyright law relates to this complex and fascinating matter in a limited and partial manner, leading to an inconsistent interpretation of the doctrinaire tests by the court.

The joint authorship doctrine relies on only one collaborative model (One size fits all) according to which only similar contributions to a work grant recognition of joint authorship. This ignores and misses various other types of contributions, e.g. the contribution of ideas or the participation in mass collaborative models such as 'Wikipedia'. Despite the extensive legal literature and the frequent and thorough discussions on comprehensively modifying the joint authorship doctrine, a wide consensus regarding the appropriate solution has yet to be achieved.

In this research, I wish to combine both a theoretical as well as a quantitative and qualitative empirical study in an attempt to reveal the creator’s expectations of the collaborative process. The theoretical discussion will examine how different fields of law (property law, patent law, criminal law and tort law) cope with collaborative activities. The aim of examining the various aspects of the different laws is to provide a new perspective to collaborative activities and to enrich the legal conversation by refining insights which can be incorporated into copyright law.

In summary, my work aspires to propose a new innovative model in order to promote a distinctive and feasible way to allocate joint authorship rights, namely preserving the incentive of joint authors to create collaborative work and at the same time better reward authors who create in common.

How Does Co-Financing Affect the Performance of Projects
Funded by the World Bank?
Yue SU
Party School of the Central Committee of CPC

Multilateral development banks have been very active in funding large-scale investments in many countries in recent years. Co-financers of such projects often include local governments, commercial banks, and export credit and international agencies. There has been little investigation about whether co-financing affects the performance of these projects. Co-financing brings in supplemental knowledge and other resources, but this could also increase transaction costs and conflicts of interest due to increasing numbers of stakeholder. This research examines how co-financing affects such projects’ returns.

This study focuses on the World Bank’s projects, utilizing its comprehensive project database. I classify the projects that employ further funding in addition to the World Bank loans as co-financed projects. As a proxy for project performance, I use the Economic Rate of Return (ERR). I first estimate the relationship between co-financing and the uncertainty of projects’ performance measured as the difference between the actual ERRc (estimated after project completion) and the ERRa (estimated at the project approval stage). I further estimate the relationship between the degree of the World Bank’s involvement and the return on projects. As instrumental variables I use World Bank presidents and the world distance to the UN’s Millennium Development Goals.  The results indicate that the co-financed projects tend to have better project performance than those financed by the World Bank alone. Further, among all co-financed projects, the proportion of funding that comes from the World Bank is positively associated with the projects’ actual return.

I further find that the lower the proportion of funding that comes from the World Bank, the higher is local governments’ gaming of the system. Examples include delays in issuing credit, poorly implemented institutional reforms, and waste of time and financial resources in prolonged renegotiations. Local governments also sometimes claim that co-financing adds fiscal or economic burdens to them.  Thus project performance appears to be better when the World Bank acts as the dominant funder, especially for poorly managed economies.

Subsidized Flood Insurance and Housing Growth:
The Unintended Consequences of State Disaster Management Programs
University of California – Santa Barbara

Does the government provision of subsidized flood insurance ultimately increase the cost of flooding events? Using a dataset of residential building permits for a sample of coastal communities in the Southeast US, I examine the role of the National Flood Insurance Program in encouraging housing and population growth in flood-prone areas.  Existing evidence suggests that NFIP subsidies on average prevent a $20,000 decrease in the value of a flood-prone home and cost taxpayers over $450 million annually. However, the problem of overdevelopment has not yet received a comprehensive empirical treatment. 

Disaster management and relief programs enacted by the state have the potential to alter the relative returns to land by shifting disaster risk from property owners to taxpayers, decreasing the expected cost of land ownership in risk-prone areas. State provision of disaster insurance can result in such distortions if provision is motivated by political inclusiveness and affordability rather than actuarial principles. By charging less than actuarially fair rates, the NFIP offers a subsidy that could induce inefficient levels of development in such areas, increasing the expected social costs of disaster. The unique procedures enacted by the NFIP regarding community acceptance and enrollment into the program allow for the identification of these development effects. When the NFIP was created, the agency needed time to construct a flood map and determine risk zones for each of the 20,000+ communities entering the program. A structure built between the creation of the Program and the date the community’s official flood map was completed qualified indefinitely for an insurance premium priced well below the actuarial rate. A community’s “subsidy phase” lasted anywhere between a few months to over a decade.

Drawing from the NFIP’s community records as well as Census data, my paper’s identification strategy exploits variation in the length of each community’s subsidy phase. Preliminary results indicate that a year’s enrollment in the subsidy phase lead to a .9-1.3 percentage point increase in a community’s housing growth rate. Back of the envelope calculations reveal the legacy of the subsidy phases to be costly- adding on average $2 million of annual flood costs per community.

What Happens to Farm Productivity During Land Reform?
University of Pretoria

The question “What effect does land reform policies have on farmer productivity?” has been debated for many years, and rightly so. The issue of failed land reform programs, especially in struggling economies like those in Africa, is particularly troublesome. Many of these countries are food insecure and failed land reform programs are expected to worsen the situation. Intuitively, when land reform programs fail, overall food production is expected to decrease which drives up prices and ultimately makes food unaffordable. It is therefore of utmost importance to investigate the reasons why these programs fail and to find solutions to rectify these politically misdirected agendas.

The South African land reform program was implemented in 1994 to rectify skewed ownership patterns and to ensure that the beneficiaries become self-sufficient to generate decent livelihoods from their land and reduce poverty. Unfortunately, like in many African countries, South Africa’s land reform program has suffered many failures (between 70% and 90% of land reform programs have failed), and its beneficiaries have seen little or no improvement in their livelihoods.

South Africa’s land reform program is characterised by ‘social tenures’, an informal land tenure system which confers de facto tenure security to large numbers of people. A key objective of a successful land reform program is to secure tenure rights. Rather than focusing on private titling (many of the land reform beneficiaries cannot pay the costs involved), the focus should be on legal recognition of social tenure, and these rights should connect to production, employment and livelihoods in practical ways. At the same time, these rights should also allow the beneficiary to obtain credit to use the funds to invest in the land, to purchase inputs, and to ultimately increase productivity. With social tenure, this is unfortunately not always possible.

The question about the impact of security of land tenure on investments and productivity has been debated for many years, mainly due to the adverse effect that it can have on especially poor communities. Empirical studies that analyse the impact of land reform programs (and its diverse tenure systems) are however scarce and leaves a definite gap for empirical research on this controversial issue.

The study aims to focus on a group of sheep farmers in the Northern Cape region of South Africa, who have been beneficiaries of the land reform program, to assess the impact of land reform programs (under social tenure) on the overall productivity of these farmers. During initial conversations with some of these farmers, it became clear that these farms are not productive, with many of these farms producing zero lambs per year. The results of this study are expected to inform a future larger scale study to determine the effects of the land reform program on South African farm productivity.

Vertical Structure, Firm Innovation in China
Mengqi WANG and Yong Wang
Peking University

This paper documents one important source of institutional cost of firms to innovate.

Speaking of firm innovation in China, many agree that the state-owned enterprises (SOEs) innovate less than the private firms. Different from the traditional horizontal structure, where SOEs and private firms compete in the same industry, this paper proposes a new perspective to understand this innovation pattern, which contains the production linkage between SOEs and Non-SOEs.

Revealed from the firm-level micro data, the upstream industries (electricity, energy, natural resources etc.) are dominated by SOEs, while downstream industries are composed mainly of competitive private firms.

Thanks to their market power, the upstream SOEs raise the input costs of firms in the downstream firms through the input-output linkage by charging higher mark up.

This paper develops an endogenous growth model to explain how the institutional cost caused by the market power of the upstream firms hampers the innovation behavior of the downstream firms; in turn, how the stagnation of the innovation in the downstream affects the performance of the firms in the upstream.

We highlight an exaggeration effect caused by input-output linkage. Taking advantage of the input-output linkage as well as their market power, upstream SOEs can extract rents from the downstream non-SOEs, and simultaneously enjoy the positive externality from downstream technological progress due to demand expansion. At the same time, the downstream firms also gain benefits from upstream innovation as a result of costs reduction. Thus both upstream and downstream industries have the insufficient incentives problem in innovation, lowering the aggregate TFP growth rate.

We use firm-level data to test these propositions derived from the model. This model fits well with the puzzling facts that upstream SOEs in China surpass private firms in profitable performance while lag behind in innovation achievements after 2001.

This paper, by exploiting the interaction of innovation activities between the firms with input-output linkage, can explain the innovation pattern of Chinese firms, as well as aggregate productivity growth under certain institutions.

The Effect of Marijuana Legalization on Mexico’s Drug War
University of British Columbia

Does U.S. marijuana legalization reduce violence in Mexico? Under prohibition, criminals dominate drug markets: without recourse to official legal institutions, violence is needed to enforce contracts. If drugs were legalized and produced on the white market, however, cartels would be outcompeted by legitimate firms, and organized crime would lose all revenue from drug trafficking.

In this paper I use a difference-in-differences strategy to study the effect of U.S. marijuana legalization on marijuana-growing regions in Mexico. If legalization reduces demand for cartel marijuana, and if marijuana is an important revenue source for cartels, then we should observe legalization affecting violent crime and drug trafficking in Mexico. Theoretically, legalization is expected to reduce homicides, by decreasing cartel revenue needed to pay foot soldiers; in addition, with the loss of U.S. marijuana markets, cartels are less willing to fight over trafficking routes. Legalization should decrease marijuana cultivation and trafficking, while having the opposite effect for heroin.

I use data on drug trafficking and cultivation to examine how the 2012 legalizations in Colorado and Washington affected the Drug War in Mexico. I identify marijuana-growing regions using data on historical marijuana eradication by the Mexican military. To measure outcomes in Mexico, I use per capita homicides, crime, and data on drug confiscations and eradications. These variables allow me to proxy for cartel behaviour, as I expect legalization to affect cartel violence and drug trafficking.

From preliminary analysis, I find that legalization is associated with a 10% decrease in homicides. However, my placebo test suggests the common trends assumption does not hold for drug confiscations. I plan to use trafficking routes as an alternative treatment group to address this issue.

Place-based Industrial Policies and Firm Productivity
Peking University

With Chinese firm-level data from 2000 to 2005 and the geographical information of the state-level special economic zones, I find that in general, SEZs do improve firm TFP. The results survive a rich set of econometric specifications. Implementing the placebo test, I demonstrate that the TFP premium of SEZ firms is not due to the "winner-picking" effect. The surge of TFP can be attributed to the better policy environment (lower taxes) in the SEZs. In addition, I find that agglomeration effect on the TFP of SEZ firms is rather marginal qualitatively and quantitatively. Furthermore, I provide some evidence to show that SEZs do generate "beggar-thy-neighbor" effect. To be more specific, SEZs have negative spillover effects on neighboring firms. To address the potential endogeneity problems, I use the minimal distance of the SEZ and the non SEZ to their nearest highway as the instrumental variable for SEZ dummy. The results remain robust.

Increasing Heterogeneity:
The Transformation of Chinese Pastoral Communities Through State Polices
Lu YU and Katharine Farrell
Humboldt University

In recent decades, a host of global and local social-ecological changes have brought about rapid transformations in pastoralist environments and practices across the planet. It creates a social dynamic in which previous traditional practices are clashing, leading to fast transformation in local communities. Approaching the practice of pastoralism as an institutional design strategy for securing livelihood in regions of low soil fertility, this paper explores the institutional consequences of major changes in the socio-ecological context of pastoral communities in the Ningxia Hui Autonomous Region of northern China. Based on institutional analysis of local responses to a combination of land-use policy interventions and demographic changes with an Institutions of Sustainability framework, we identify a number of pastureland-related transactions arising in the studied communities, some of which pre-date the policy interventions and demographic changes, and some of which appear to have emerged in response to them.

Among the observed transactions we distinguish between two types, 'classic' and
'nature-related' transactions, finding that lack of attention to the latter appears to have led to unintended consequences that may run counter to stated government policy objectives. In particular, we find that implementation of a grazing ban has led to decreased sheep holdings, greater heterogeneity of livelihood in the studied communities, as well as abandonment of past cooperation practices and decreasing reference to norms of reciprocity. If these practices continue, we would expect them to become institutionalized innovations that will in turn influence how these communities respond to future policies intended to protect northern China's grasslands. In closing we consider the degree to which these observations might be generalizable to other pastoralist contexts and explore the possible implications of unintended institutional changes in Ningxia pastoralism for the likely effectiveness of current Chinese government policies intended to foster sustainable use of pasturelands both in Ningxia and throughout the country.

Goal Setting in Tournament with Sabotage: Theory and Experimental Evidence
Le (Lyla) ZHANG and Cheng-Tao Tang
Macquarie University

Tournaments are widely implemented in organizations to encourage productive effort; however, they are often plagued by sabotage activities among competitors. Competing employees can achieve a relatively higher ranking and win the tournament by sabotaging their peers. Such sabotage activities are detrimental as they not only reduce the organization's output but also demotivate other employees to exert productive effort. It is therefore imperative to design an institution to encourage employees for more productive effort and fewer sabotage activities.

In this study, we investigate the effects of setting goals in tournament. Our theoretical model predicts that goal setting (within an appropriate range) in tournament can play a positive role. It incentivizes employees to exert more productive effort as well as fewer sabotage activities. Because goals can immune from sabotage activities, it is not worthwhile for employees to waste effort on sabotage activities. Hence it is in the benefit of employees to substitute productive effort for sabotage activities when goals are implemented in the tournament. As a result, the output of the entire organization increases.

We design a controlled laboratory experiment to test our conjectures. The experiment includes 6 treatments which vary in the level of goals and the size of financial bonuses for winning the tournament. Consistent with theory, the experimental results confirm the positive effects of goal setting in tournament. However, contrary to the theory, we find the effect of goal setting diminishes as the financial incentives (i.e., bonuses) increases. Financial incentives become dysfunctional when high goals are implemented. Employees reduce both productive effort and sabotage activities when both high goals and high financial bonuses are implemented. Overall, high goals increase performances more than higher financial incentives. Our study offers experimental evidence indicating that design of performance goals and financial incentives is important in the implementation of tournament reward schemes.