Past PhD theses

This thesis is composed of three essays, one in Choice Theory (Chapter 1) and two in Social Choice Theory (Chapters 2 and 3). All chapters consider situations where an issue has arisen with a number of alternative ways to deal with it, and some individual (henceforth, some choice correspondence) needs to choose one, or more, of these alternatives in order to deal with this situation.

In Chapter 1, we are interested in choice correspondences that satisfy some of the following properties: two versions of the independence of irrelevant alternatives (IIA and WIIA), as well as the weak axiom of revealed preference (WARP). Loosely speaking, we show that combinations of these properties can partially or completely determine the choice correspondences satisfying said properties.

In Chapters 2 and 3 choice made by the choice correspondence must be according to the preferences of a group of agents. Specifically, we are interested in choice correspondences satisfying some of the following properties.

• In Chapter 2: Pareto efficiency, population-monotonicity, and replacement domination.
• In Chapter 3: Pareto efficiency, strategy-proofness, and anonymity.

Loosely speaking, we show that combinations of these properties characterize target set correspondences (Chapter 2) and generalized median correspondences (Chapter 3).

This thesis consists of three research papers, with household finance as common topic. The first paper studies a lifetime combination of financial, work and health-related decisions made by households. The results, obtained from a dynamic life cycle model of allocations and welfare, show that households' lifetime decisions are challenging to match, even after accounting for the complex interactions between them. Whereas financial savings and pension claims are both well matched, the model overestimates health levels, and consequently life expectancy. Moreover, health is maintained through more spending, and less leisure than currently observed. As a consequence, observed post-retirement income is higher than expected, and explains the divergence in consumption after 65. The second part analyzes the homeowners' life cycle responses to financial, health and real estate shocks. After successfully matching the US households life cycle data using a unique structural life cycle model, the results show that asset rebalancing and time allocation decisions are shock-dependent. Unsurprisingly, shocks decrease the net worth and postpone the full retirement age, but in different magnitudes. Notice that when the shock is a real estate market collapse, the life cycle impact is bigger. Finally, the third research paper examines the impacts of conventional (i.e. man-made law) and Islamic (i.e. Sharia rules) financial literacies on the Pakistani households participation in the formal and informal financial sectors. Based on a statistical model, the results highlight that the conventional financial literacy positively and strongly affects the use of formal financial services, while the Islamic financial literacy has a weaker positive impact. Therefore, financial literacy programs could be a way to reduce financial exclusion.

Along the life-cycle, individuals go through a series of important events and turning-points that determine their health characteristics and trajectories, and shape their economic and social well-being. This thesis is about applying economic analysis and tools to understand how some of these turning-points affect one’s health and health behaviors. The three important events that I consider in my thesis are birth, schooling and retirement.

I show in the first chapter of my thesis that children born in Malawi in a year in which their mother experienced a negative economic shock had poorer subjective and objective health outcomes later in life. The second chapter looks at the effects of education on technology use and adoption among older adults in Europe.  I show that persons who acquired more education due to a compulsory schooling reform are more likely to use the Internet and to have better computer skills later in life. The third paper of my thesis uses social security eligibility ages as instruments for retirement in econometric models for physical activity and longitudinal data from the USA. I find that retirement causes older persons to engage in more physical activity, especially among highly educated and wealthy individuals. Finally, the last chapter of my thesis highlights potential reporting heterogeneity in subjective survey measures based on Visual Analogue Scale (VAS) and cautions regarding the use of subjective measures in empirical studies for drawing conclusions on how health, health behaviors and well-being change over the life-cycle.

Along the life-cycle, individuals experience a series of major events that shape their health and ability to engage in economic and social activities. Understanding the effects of these events on individual’s life requires objective measures that are comparable across individuals. My thesis shows that health and health behaviors can be influenced by non-health-related causes occurring throughout life, such as economic shocks at birth, education and retirement. My thesis therefore identifies some of the events that causally determine health and contributes to the current understanding on how health evolves along the life-cycle.

Understanding the drivers of cross-country capital movements is crucial from a policy perspective. Sharp fluctuations of capital may raise concerns on the country’s financial stability. My thesis’ first two chapters study the factors potentially driving capital flows. In particular, the first chapter focuses on the consumer confidence index. This index includes a component of households’ expectations about future economic prospects. A priori, these expectations may reflect true information about the country’s future productivity. They may also represent surges of optimism or pessimism. My findings for 33 countries show that these expectations are significant drivers of capital flows. The aim of the second chapter is to formally determine whether expectations reflecting true information about future productivity, labelled “news”, or expectations mirroring “sentiments” are more important in driving capital flows. The use of a more structured model allows to disentangle these two effects. The findings show that news and to an ever greater extent, sentiment, are able to explain a significant portion of capital flows. Moreover, sentiments are not mere reflections of worldwide factors, but rather, are capturing some country-specific information. The last chapter looks at the consequences of large capital inflows, focusing on emerging economies. The analysis shows that, because of the low interest rates resulting from the U.S. monetary policy, investors have started to invest in these countries, seeking for higher yields. Local companies have thus started to issue more bonds in foreign currency, raising their exposure to exchange rate variations. This risk appears to not be particularly concentrated among certain types of companies and can be limited using capital controls measures.

The first chapter studies how lobbying can influence government decisions on regulatory issues. Firms may find themselves opposed to organized consumer groups in battles for influence over the design of new policies regulating their industry. Understanding the incentives of the different actors involved and their influence over the policy outcomes is therefore important. We show that competition in the production of information enables the government to extract an informational rent from the business lobby. Our results suggest that lobbying battles that look very unequal at first sight may not always yield outcomes in favor of bigger player.

The second chapter evaluates the effects of an intensive job search assistance pilot program offered to long-term job seekers in Geneva, Switzerland. We find that the program improved job re-entry in the short-run, but its participants reached less stable positions, which significantly increased the probability of job loss after two years and canceled out the initial positive effects. These findings highlight the need to consider long-term horizons when evaluating such policies, as some mechanisms may take more time to materialize. Failure to do so may result in overly optimistic results.

The third chapter studies how culture affects male labor supply near retirement before and after reforms that introduced early claiming age of pension benefits. These policy changes significantly reduced the incentives to work for individuals approaching the full retirement age. I find that cultural preferences affect LFP but not when the incentives to work are strong. Results suggest that culture and economic incentives do not simply add up. Incentives dominate, while culture creates heterogeneity in reaction to policies as soon as individuals are given room to adjust their behavior.

The first part of this thesis uses laboratory experiments to investigate how beliefs about skill, risk attitude, and the perception of probabilities affect decisions under risk and uncertainty. This is important for many real-life decisions such as choosing a career path where the available options have rewards that are more or less risky and dependent on skill (e.g., working as a lawyer versus working in the public sector). Further, the thesis investigates in more detail how individuals choose between entering a market where rewards are uncertain and a certain outside option. The findings can be summarized as follows: First, low (high) skill individuals are overconfident (underconfident) and therefore overestimate (underestimate) the value of gambles where probabilities depend on skill. Second, overestimation of small success probabilities leads to excess entry into winner-take-all markets, i.e., into markets where rewards are concentrated on very few competitors.

The second part of the thesis investigates the tax sensitivity of firms using Swiss data. The research question is motivated by a general downward trend in corporate tax rates and increasing provision of privileged tax treatments for specific activities in developed countries. Action plans of the EU and the OECD to combat “harmful” tax competition are now forcing countries and jurisdictions to adjust their corporate tax laws. The following empirical findings of this thesis offer guidance for sound corporate tax policy making. First, corporate income reacts significantly to changes in corporate tax rates. However, the reaction is not so strong that a decrease in corporate tax rates would lead to an increase in corporate tax revenue. Second, a significant part of the tax sensitivity is due to mobility of corporate income across jurisdictions. The role of mobility is important, for instance, to gauge the extent to which tax competition is desirable.

The first two chapters of this thesis are related to the debt crisis that hit the Euro area in 2009-2012.
The first chapter examines the effects on financial stability of programs such as the European Financial Stability Facility (EFSF) and the European Stability Mechanism (ESM). I show that such programs generate financial contagion in sovereign debt markets and that, by inducing countries to borrow more, they can increase the frequency of debt crises and defaults.

In the second chapter we ask whether a country similar to one in the Euro periphery could have escaped the debt crisis if it was able to conduct its own monetary policy. We find that an independent monetary policy could in theory avoid the crisis, but doing so involves in general many years of sustained inflation, the cost of which might exceed the cost of a default.

The third and fourth chapters are related to reforms of the banking sector that were proposed after the financial crisis of 2007-2010.
The third chapter focuses on “sovereign money reforms”, and in particular on the Swiss “Vollgeld initiative”, that advocate that the function of issuing and managing deposits be given entirely to the central bank. Contrary to the claims of the proponents of the reform, we find that the Vollgeld reform would have negative welfare effects.

The fourth chapter finds a novel rationale for the coexistence of deposits and lending in current banks, that speaks against narrow banking proposals. Thanks to the money properties
of deposits, interest on deposits is sticky. Banks performing maturity transformation are less exposed to the interest rate risk stemming from monetary policy, and cut lending less after an interest rate increase, than they would if they did not issue deposits.

The thesis deals with different questions about where exactly economic activity happens in geographical space. The main part of the thesis consists of two chapters that analyse which geographic regions in a country benefit most when a country trades more with its neighbor countries. In particular, the chapters investigate whether trade leads to previously poorer regions in remote areas catching up in development with rich urban centers or whether countries become more unequal following an increase in trade. While economic theory previously was not able to answer this question, the thesis shows that trade helps to poor regions to develop if they are located closely to international borders. This provides a strong argument for more international trade in the political discourse about the winners and losers of globalisation.

In a third chapter, I analyse how people bid for otherwise identical apartments located on different floors within a building. I find that the vertical location of an apartment plays a crucial role in explaining its price as people value both the easy access to the road on lower floors as well as the amenities coming with higher floors. Furthermore, we provide evidence that different social classes and income groups exhibit different preferences for vertical locations, which could in the future help to alleviate problems of gentrification in cities.

Households consume housing, food and clothing. Housing consumption is the result of a decision making process that is influenced by the surrounding economic environment. According to the Swiss Household Panel, the average share total annual household net income devoted to housing in 2000 (2014) was approximately 28% (31%). For households in the first decile of the income distribution, the average share was roughly 41% (49%), compared to 19% (20%) for households in the tenth decile. Similarly, in the USA, housing expenditure shares in 2014 amounted 41% and 30% for low-income and high-income households, respectively. In terms of financing, household mortgages represent a significant share of total Swiss domestic loans, amounting to around 85% as of 2018Q2. Using various econometric techniques and theoretical modelling, this thesis studies: (i) how local policy affects house prices through its effect on household location and tenure choices, and (ii) how changes in interest rates affect loan supply, given that most owner-occupiers finance their housing consumption via mortgage debt.

This thesis employs a number of statistical tools and economic theory to evaluate the causes and consequences of different types of vulnerability to economic shocks and circumstances. Different people may be more or less likely to be economically vulnerable for different reasons. Each of the three chapters of this thesis aims to examine a distinct source of economic vulnerability that reflects notable concerns to policy-makers in the twenty-first century.

Economic vulnerability may describe a demographic group which is systematically associated with economic disadvantage. Hence, the first chapter focusses on the vulnerability of a particular demographic group : immigrants. I examine how firms contribute to the systematic empirical observation that immigrants are paid lower wages than natives : and how firms respond after a large immigration shock to the host country. This chapter sheds light on how understanding the reasons behind the immigrant-native wage gap can help identifying the appropriate policy to tackle it.

An unfortunate stroke of nature, such as income loss, is an alternative source of economic vulnerability. The second chapter focusses on how the sudden loss of income affects the individual members of a community in contributing effort towards appropriating a resource shared by many groups. Understanding how income shocks affects group appropriation of the common resource sheds light on behaviour of communities and how their impoverishment can affect increasingly scarce resources such as water, oil or fisheries.

Finally, an individual may be vulnerable due to a personality trait or genetic feature which is expressed by behaviour leading to poor economic outcomes. The third chapter studies how anxiety impacts productivity and risk attitudes in a randomised control trial. Since anxiety is such a prevalent emotion in many aspects of daily life, such as the workplace, understanding what its economic costs are could prove useful in dealing with them.

The voting decision has an inherent ideological component. When choosing to cast their ballot for one candidate instead of another, individuals take into account both the economic policy platform and ideology. The choice on the ideological dimension is determined by the voter's values and beliefs. In recent years, a strand of literature has investigated the impact of violence on beliefs and political attitudes in conflict-torn countries. My thesis contributes to this literature by evaluating the effect of violence on in-group identification and political trust in developed countries, thus making a link between violence, political attitudes, and voting outcomes.

In "The legacy of war exposure on political radicalization" I examine the impact of war participation on identification with the polity and voting behaviour. I answer this question using a particular historical event: the forced conscription of men from the French eastern borderlands to the German military forces during World War II. In places where more men were conscripted, radical right-wing candidates still receive larger electoral shares today.

In "Seeds of Populism: Crime News Exposure and Anti-Immigration Politics", we study how news coverage of immigrants' criminality impacted voting patterns in the 2009 referendum on "Minaret Ban" in Switzerland. Our results imply that, should newspapers report foreigners' crime propensity at its true value, this would have decreased the "yes" vote by 4 percentage points.

In "Technological innovations in electoral campaigns: Direct canvassing and partisan mobilization" we argue that the main reason direct canvassing persists as a campaign strategy is because it is a very efficient mean to mobilize voters that are already supportive of a candidate's programme. We exploit a unique historical context: the first direct canvassing campaign by a presidential candidate in the 1896 election. Our findings imply that the electoral gain of the Democratic candidate can be attributed as 65% coming from increased mobilization, and 35% coming from persuasion.

What drives human behavior? This is the central question in behavioral economics and the answer is - naturally - context dependent. In this thesis I present four different quasi-experimental studies which each provide an insight into individual motivation in a specific environment. Quasi-experiments generally provide results that are easier to generalize than lab experiments but have inferior internal validity, due to a lack of explicit random assignment. Instead, they focus on situations where nature or institutions intervene and create exogenous variations that affect individuals differently. Quasi-experiments allow us to study a wide range of research topics. However, careful designs are necessary to cleanly identify effects despite the lack of experimental control. This often requires large data sets and strong interventions.

The four chapters exploit different quasi-experimental setups. Chapters one and two study spillovers of prosocial motivation among blood donors. Documenting and quantifying such spillover effects is highly policy relevant. Policy makers generally have a limited capacity for interventions. Thus, attributing these resources in the most effective way is crucial. Motivational spillovers can enhance the provision of public goods and affect the cost-benefit analysis of policy interventions. Chapter three looks at a potential factor that might harm pro-social motivation: A substantial fraction of blood donation attempts are rejected because they do not meet required donation criteria. Donors may perceive such rejections as social exclusion what could decrease their future motivation to donate. The last chapter of the thesis looks at a repeated competition setting. Repeatedly participating in competitions is essential for many life goals. Losing in such competitions may discourage individuals and make them postpone further competitions, thereby harming their future prospects.

While the four chapters study four quite different research questions on individual motivation, they all share similar underlying methods that allow to cleanly identify causal effects in natural settings.

This thesis studies the role of social interactions on agent motivation. The first two chapters focus on prosocial motivation of voluntary blood donors who do not receive pecuniary incentives but social incentives from their peer groups. In Chapter 1, the peer groups are blood donors living in the same residence building. Chapter 2 extends the peer groups to donors in larger social networks. Chapter 3 uses a tournament setting where agents are not only motivated by high-stakes pecuniary incentives but also social incentives from their teammate. Chapter 4 investigates how social exclusions affect prosocial motivation of agents in a similar setting to in the first two chapters. All chapters use observational data and rely on the methods of instrumental variables to identify causal effects.

1. This chapter investigates the effect of asylum seekers hosting centers on nearby housing prices. We gather precise information on collective hosting centers, which we relate to a large dataset of housing transactions occurring in Switzerland over 2004-2014. We find a precisely estimated negative effect of 4.9% on the price for housing units located within 400 meters of an active center. Housing prices react less strongly in the proximity of police stations, which we interpret as evidence for perceived threats of increased crime as a relevant determinant for the observed housing price responses to hosting centers. However, we find no significant effect of the opening of centers on local crime rates. We also do not find evidence of price responses being driven by anti-immigrant political sentiment as expressed in a variety of referenda.

2. This chapter provides an account of the world's evolving economic geography by documenting employment growth patterns across 18,978 regions. These regions cover the main economies that together account for three quarters of global GDP, and we observe the period 1990-2010. We find that employment growth has been fastest in the world's densest regions. This flight to density has been accelerating over time and is driven by the growing importance of the non-agricultural sector in emerging economies. There are important differences between mature and emerging economies: whereas growth is fastest in the highest-density regions in emerging economies, it peaks for high medium density regions in mature economies.

3. This chapter studies the extent of spatial market integration among Brazilian municipalities. Relying on individual-level data from two rounds of the census, 2000 and 2010, I structurally estimate a set of parameters that capture the extent of regional market interactions and thereby provide evidence of the spatial frictions occurring between Brazilian municipalities.

This thesis consists of three papers on whether to use and how to use self-reported measures (for example, self-reported health) in social surveys.

Because it is cheap and easy to collect, the self-reported memory is routinely included in surveys and also widely used by general practitioners to evaluate the cognitive function of people. The first paper of the thesis examines whether the self-reported memory reflects the underlying cognitive function. Using data of middle-aged and older adults in China, we have found the self-reported memory is a relatively poor measure of cognitive function. Also, we have shown that individuals from different SES groups behave differently when reporting their memory condition.

The second paper proposes a new econometric model which makes self-reported health comparable between individuals. The new model is based on the premise of an influential theory of psychology and can be easily tested. I have compared this new model with the existing model through both simulations and analyses of real data. I have found the new model performs better than the existing model in general.

The third paper discusses how to use anchoring vignettes (a tool to align self-reported measures) in different ways. Firstly, we have studied the consequences when the econometric model used is not correct and proposed a new model which is more likely to be immune from the wrong model setup. We have done a large number of simulations which mimic different model setups.

Matching models describe several real-world allocation problems that are not appropriately described by the classical competitive market model. Examples are the allocation of workers to firms, of students to colleges, or the matching of kidney-donor-pairs for life kidney donation. A central feature of these markets is that prices are not or not exclusively responsible to equate demand and supply. Pricing mechanisms cannot be used, since either monetary transfers are restricted or forbidden, since market participants are not profit maximizing, or because price taking behaviour is not plausible if market participants are very heterogenous.

In this thesis, we study, in three independent chapters, different theoretical aspects of matching markets and matching mechanisms. The first two chapters are motivated by theoretical questions in the design of clearing houses for entry-level labour markets. For these markets, we study allocation mechanisms with endogenous contracting. An allocation mechanism does not only determine which workers are matched to which firms, but also the contractual details of the labour relation. The third chapter studies the design of fair randomized admission procedures for the assignment of students to public schools.

In recent years, many real-world allocation problems have been studied within the paradigm of matching theory. Policy interventions in real world matching markets have been successful whenever they were informed by theoretical results. We are hopeful that the results in this thesis also find future applications in applied market design.

In the first chapter, I estimate the Swiss franc safety premium, defined as the compensation that investors require when they sell the presumably safe Swiss franc and in turn invest in a basket of foreign currencies. The three-step instrumental variable approach suggested by Maggiori (2013) yields unrealistic results and therefore does not work for Swiss franc. After some slight modification in the methodology, however, the results become more realistic and closer to those obtained when applying a GARCH approach. I find that the CHF safety premium is around 10% during crisis episodes.

In the second chapter¹, we develop a simple two-country model with frictions in the banking sector. We identify three external sources of appreciation pressure : Financial frictions in the foreign investment market, financial frictions in the international credit market as well as capital inflows. In the two latter cases, foreign exchange interventions by the central bank can reverse the exchange rate movements. Under certain conditions, central bank purchases of bonds issued by firms (« credit easing ») can achieve this goal as well. The two policies, however, are likely to differ in their costs.

In the third chapter², we analyse whether the SNB's policy measures to counter the appreciation of the Swiss franc before the introduction of the EUR/CHF lower bound were effective. We apply the so-called synthetic control approach which consists of building an artificial EUR/CHF exchange rate out of other exchange rates and safe assets that were not affected by the SNB's policy interventions. By comparing the actual EUR/CHF exchange rate after the intervention to its synthetic counterpart, we find that the March 2009 and August 2011 interventions were effective, while the spring 2010 interventions were not.

¹Joint with Nicole Aregger

²Joint with Nicole Aregger

The three chapters of this thesis use the European soccer market as a laboratory to shed light on the implications transfer fees (i.e. compensations paid to an incumbent firm) can have in a more standard labor market.

This first chapter investigates the impact of transfer fees on wage determination, by constructing a model with two periods that reproduces salient features of the soccer market. The model predicts that, in a market where transfer fees apply, a worker gets a smaller wage and fewer opportunities to increase it. Moreover, imposing a regulation decreasing wages can increase transfer payments. Using a rich and unique database on the Italian first division, a structural estimation confirms the validity of the model.

The second chapter extends the model of Chapter 1 by adding the different labor life stages. While workers’ wages tend to increase with age and experience, transfer fees rather sharply decrease at the end of a worker’s career. These different profiles can be explained by experience effects, transfer opportunities and horizon effects. In addition, when a transfer fee must be paid, the worker’s wage is lower; however, due to the possibility of a future transfer, this negative effect is tempered in the middle of his career. Finally, being transferred as much as possible during his career maximizes the worker’s total life salary. If only one transfer occurs, an early transfer is more beneficial than a late one.

Finally, the third chapter explores, both theoretically and empirically, whether the worker-incumbent firm matching quality affects the strategic use of contracts as rent-seeking device (i.e. a device aiming at increasing the one’s share of profit without creating new wealth). Rent-seeking incentives are in fact reinforced in case of a mismatch. These behaviors may create inefficiencies, such as deterring future market entry or lowering investment incentives.

These results are not restricted to the world of European soccer. Contract clauses that prohibit a worker from working for a company’s rivals often include breach penalties which are very similar to the transfer fees observed in European soccer. 

This thesis contains three separate empirical papers. The first paper studies how local income tax rates, income tax bases and housing prices react to each other, by fitting a VAR model to data for 2,517 Swiss municipalities. We find a long-run elasticity of taxable income with respect to the local income tax rate of -0.08. The corresponding estimate based on households in the top-1% income category is -0.55. The reverse elasticity of tax rates with respect to exogenous changes in the tax base, is estimated at -0.07.

The second paper investigates how co-ethnic networks influence the Tiebout sorting behaviour of immigrant households that settle in Swiss municipalities. A new estimation approach is proposed to disentangle co-ethnic network effects from alternative location choice factors for each income category and family type. The results indicate that the importance of co-ethnic networks with respect to households' incomes follows an inverted U-shape. The lowest- income households seem to be predominantly attracted by criteria common to all immigrants. Thus, adverse selection originating from co-ethnic network effects may be less pronounced than what earlier findings suggest.

The third paper measures how greenhouse gas (GHG) policy stringency affects anthropogenic CO2 emissions, using a new GHG policy stringency indicator and a spatial VAR approach. Results indicate that a country with no GHG regulations can achieve an average reduction of 15% of its CO2 emissions by adopting the stringency level of the most regulated country. In addition, increasing GHG policy stringency improves sectoral CO2 efficiency, and decreases production in CO2 intensive sectors thereby altering the sectoral composition. Policy induced CO2 reduction costs in terms of GDP are relatively large compared to ex-ante cost reduction studies, but four times lower for developing compared to developed countries.

Civil conflicts are dramatic events pushing individuals out of their home countries. But what if emigration itself worked as a pacifying force? In the first two chapters, I am interested into how emigration affects civil conflict incidence in the countries of origin. First, I construct a general equilibrium theoretical model of social conflict with emigration, where conflict results at equilibrium as the share of fighters in the economy. Emigration discourages fighting by reducing the prize to fight for, while increasing the opportunity cost of fighting. The main prediction of the model is that conflict decreases with the foreign wage net of migration costs. Second, I empirically test this prediction. By constructing a theory-driven instrumental variable, I show that emigration to developed countries decreases civil conflict incidence in origin countries. These findings point that, by opening their borders, host countries could contribute to saving the lives of the migrants as well as of those left home.

Moreover, civil conflicts destroy societies and traumatize the involved individuals, which increases the risk of a future conflict. In order to understand this vicious circle and find solutions for tackling it, the third chapter looks into whether the individuals exposed to conflict during their childhood are more likely to become violent later in life. The analysis focuses on asylum seekers in Switzerland since this group of immigrants are randomly allocated across cantons so they cannot self-select in a crime-promoting environment. The results show that indeed individuals exposed to civil conflicts or mass killings are more prone to develop an unlawful behavior later in life in comparison to their co-nationals born after conflict. Nevertheless, by analyzing the impact of public policies, we find that the implementation of friendly policies can break the vicious circle of violence persistence. In particular, offering labor market access and long-term perspectives to asylum seekers eliminates the risk of increased criminality.

Since the 19th century, the use of mathematics to formally model hypotheses about behavior became an essential element in the economics research toolkit. Ever since, and for the sake of tractability, economists have simplified their models into sets of equations. The representative rational agent is, perhaps, the greatest simplification of the profession. 
During the first half of the 20th century, the study of games became popular among economists, who started moving away from the representative agent by acknowledging that people do not make decisions in isolation, but they take into consideration the decisions of others, whose objectives and beliefs may differ. However, rationality was still maintained in most economic models of behavior. It was until recent years that insights from psychology were introduced into economics to question the rationality assumption, giving raise to what we know as ``behavioral economics''. 
This thesis gravitates around the two departures from traditional economics mentioned above. On the one hand, people's behavior and beliefs are usually not determined by rational reasoning but their drivers need to be observed and understood. On the other hand, individuals do not make decisions in isolation but they consider the decisions of others, with whom they interact in potentially complicated ways. 
The different chapters show that, considering these two aspects of human interactions, leads to different outcomes than those predicted by traditional economic theory. In particular, that; i) coordination between people cannot be explained by rational responses to others’ actions, ii) cooperation is reduced if people cannot easily learn whether other people cooperate, iii) people can be altruist towards members of their groups, while being needlessly aggressive towards those outside their groups, and iv) people can be overconfident about the qualities of members of their group. Lastly, the thesis presents a statistical test for inference from network and experimental data.

The thesis is made of three independent chapters interested in the impact of globalization on workers in industrialized countries. The dissertation is especially focused on identifying the causal impact of international trade on workers’ mobility, wages, and employment with both a short- and medium-term perspective. The first paper explores the relation between intra-industry trade (IIT) expansion and associated worker flows, taking the latter as an indicator of labor-market adjustment costs. Being the first study to combine theoretical simulations and a novel identification strategy, we find that both theoretical and empirical analyses are consistent with the “smooth adjustment hypothesis", according to which IIT expansion is less disruptive than inter-industry trade expansion. The study therefore lends support to the use of IIT indices as first-pass proxies for the adjustment effects of trade expansion. The second chapter contrasts the impact of increased import competition coming from China and the European Union (EU) on workers in the United Kingdom over a 15-year period. The most salient findings show that increased imports from China had significantly negative effects on workers’ earnings, wages and employment. In contrast, larger imports from the EU are associated with positive worker-level outcomes, which is largely explained by the fact that increased imports from the EU were mostly offset by increased same-industry exports to the EU. Besides, we find that increased imports from China exert additional pressure on workers through spillovers to employment and wages in downstream industries. Finally, the last chapter is focused on the impact of exposure to trade and real exchange rate shocks on wages for Swiss manufacturing workers. A particular attention is made to consistently estimate the causal effect in using a two-step gravity-type identification strategy. The study shows that the impact of trade and exchange rate movements is concentrated among high-skilled workers almost exclusively.

The present PhD dissertation consists of three papers, organized in chapters, in the field of behavioral economics. This discipline studies economic behavior of individuals subject to limitations, such as bounded self-interest and bounded willpower. The behavior studied in the present thesis ranges from the complex decision to register as an organ donor, decision-making in the presence of uncertainty and the decision to give money to a charitable organization.

The first chapter aims at testing the effectiveness of an active-decision (AD) mechanism on the decision to become an organ donor in Switzerland, using field experiments. We found that stimulating participants’ reflection on the topic of organ donation had a negative effect on the decision to become an organ donor. Moreover, a non-binding commitment nudge reduces putting off the decision, but does not lead to donation rates higher than in the control group. The results suggest that AD may be far more limited than previously thought and raise doubts about the efficacy of engaging potential donors to reflect on the topic of organ donation.

Beyond carrying for others, behavioral economics also recognizes that individuals do not evaluate outcomes in absolute terms but rather by comparing them to some reference levels, called reference points. Above the reference points, economic outcomes are perceived as gains, while below these levels the same outcomes are felt as losses. The last two chapters analyze the importance of reference points in the evaluation of economic outcomes.  Using a laboratory experiment where subjects played two consecutive lotteries, Chapter 2 studies the speed of adjustment of the reference point. We find that varying the probability of winning the first lottery has no effect on subjects’ risk behavior regarding the second lottery. This result indicates a very fast adjustment of the reference point to the latest information. Chapter 3 investigates whether reference points are relevant for charitable preferences. Using actual donation decisions of participants in a laboratory experiment, the results suggest that reference points are not crucial for shaping charitable giving.

The beauty of the economic science lies in the interplay between theoretical models and empirical evidence. Economists have pushed the social science towards formalizing ideas to allow the derivation of behavioral predictions and hypothesis testing. New theories develop from simple intuitions, or evolve as extensions to existing models to explain empirical facts. The field of behavioural economics excels in challenging existing models through randomized controlled trials. Economic experiments simulate decision environments to understand the underlying psychological mechanisms, which drive behaviour. On the contrary, the field of labour economics uses clean data sources that allow analyses, as if situations were taking place in controlled environments.

The unifying theme of this thesis is to show how empirical results can challenge theoretical predictions and common sense intuitions.

The first two chapters analyse the role of expectations in reference point formation. People do not necessarily evaluate situations in absolute terms, but rather compare outcomes to certain anchors they can relate to, i.e. reference points. Kőszegi and Rabin (2006) put forward the idea that the reference point is formed through expectations. We analyse their implications theoretically, test it experimentally and show that the model fails in its key mechanisms. Expectations do not seem to be the key driver in reference-point formation as proposed by the KR-model.

Chapter 3 analyzes the popular phenomenon of the emerging “internship generation” using panel data. Ever more university graduates conduct internships in order to enter the regular labour market but their usefulness remains pending. Indeed, the results suggest that internships have detrimental effects on several dimensions. Graduates with an internship experience are less likely to be employed within one year of graduation, and if employed, they earn significantly less than their non-intern peers. Luckily, the negative effects are short-lived and vanish within five years.

Behavioral Economics is concerned with increasing the psychological realism of mainstream Economics without impairing its formalism. It considers departures from the central assumptions characterizing the homo economicus, loosely termed rationality. This increase in realism often has the power to rationalize observational data considered as puzzling by the standard model. Experimental methods offer tools to the researcher to test such theories. In this thesis, behavioral models are formalized and tested using laboratory experiments.
In the first chapter, a model in which workers can exert productive effort and poaching effort (e.g. stealing a client to a colleague) is developed. Using an experiment, it is shown that individual-based incentives can trigger high levels of poaching which are detrimental to performance. Team-based incentives however do not have this property.
The second and the third chapters are centered on model of expectation-based reference-dependent preferences developed by K˝oszegi and Rabins (2006). They postulate that people evaluate outcomes with respect to a reference point : their expectations. In the second chapter, the question of the speed of adjustment of the reference point is tackled. The results suggest that the speed of adjustment of the reference point might depend on the size of the stakes at play. These results help understanding the conditions under which past beliefs and emotions should be expected to linger and to affect subsequent behavior.
The third chapter demonstrates that the K˝oszegi and Rabin model closely links myopic loss aversion to the endowment effect for risk. While myopic loss aversion can be replicated in a laboratory experiment, the endowment effect for risk is clearly rejected. These findings cast doubt on the ability of the KR model to provide a unifying theory of behavior under risk.

Abstract

This thesis studies the effects of health on financial decisions and on the short-run fluctuations of economic activity. The first paper analyses the effects of health status on household's portfolio choices; their decision to insure; their health and housing expenditures. Using data from the Panel Study of Income dynamics (PSID), the estimates highlights three results: 1) a significant effect of health on household's decisions, 2) a simultaneity in these decisions and 3) an evidence of health effects through the risk channel. Given these results, the second article explores the joint effects of health status and health-related risks on financial and health-related decisions. In so doing, this paper estimates the theoretical model of Hugonnier et al. (2013) on PSID data. The results show an effect of health through the income and risk channels. Mortality and morbidity risks impact the consumption, the investments in stocks and health insurance, and health expenditures decisions. Based on these empirical findings, the third paper explores the aggregate implications of health capital. This research develops a business cycle model in which agents are endowed with a stock of health capital and exposed to mortality and morbidity risks. Simulations reveal that this model fits well the business cycle stylized facts, as well as the cyclical patterns of health. In summary, these three studies suggest an active role for public health and insurance policies in the improvement of welfare. These policies may improve households' consumption and participation in the stock market. Further, they may contribute to smoothing economic fluctuations.

This thesis consists of three chapters.
The rst chapter analyzes the implications of entrepreneurial optimism for equity market signaling. Project quality is signaled either by retaining shares (Leland and Pyle, 1977) or by retaining shares and underpricing (Grinblatt and Hwang, 1989). On the one hand the existence of optimistic entrepreneurs allows realistic entrepreneurs to insure idiosyncratic risk more, but on the other hand makes it less pro table to sell equity because the presence of optimists reduces stock prices. Entrepreneurial optimism makes entrepreneurs' worse o because it raises the cost of signaling. In contrast, it can make outside investors better o since it increases the volume of stocks that are underpriced.
The second chapter models the impact of optimism on occupational choice. In-dividuals choose to become entrepreneurs or employees. Competition in the en-trepreneurship and employment sectors determines an entrepreneur's probability of success and the wage of employees. Realistic individuals know the probability of succeeding as entrepreneurs, while optimists overestimate it. We nd that when there is a moderate or a high number of optimists in the economy the number of entrepreneurs (number of projects undertaken) raises, the number of employees low-ers, the probability of success of projects lowers, the interest rate on entrepreneur's borrowings raises, entrepreneurs borrow less and undertake projects of smaller size. We show that in this case optimism lowers the individual expected payo of entre-preneurs and raises the individual payo of employees.
The third chapter extends lobbying theory by assuming that the policymaker has reference dependent preferences: he derives utility not only from social welfare and contributions given by the lobby but also from deviations in the lobby's contributions from a reference level. We show that if the reference contribution is small and the policymaker is very responsive to contributions that are higher than the reference one, then the lobby obtains a more favorable policy o ering lower contributions. This result provides a new explanation for Tullock's paradox: the empirical fact that lobbying contributions are small relative to the value of the policies at stake.

Evidence-based policy is understood as an approach that helps people make well informed decisions about policies and programs using systematic empirical evidence. The implementation of successful economic policy however requires a solid understanding of the causal effects of a policy intervention. Improvements in data quality, development of more robust estimation methods and better research designs have helped to develop a large toolkit of new instruments to identify causal effects of policies.

This thesis consists of three self-contained chapters in applied micro-econometrics. The first two chapters discuss related topics in labor economics. Both chapters rely on large administrative databases and use quasi-natural experiments to identify the causal effects of policy interventions using two distinct methodologies. While the first chapter focuses on a discussion of the empirical findings of reducing potential unemployment benefit duration on post-unemployment outcomes, the second chapter develops – inspired by job search theory – a new approach to learn about the relative importance of reservation wages for non-employment duration and survival probabilities. The third chapter is in the field of behavioral environmental economics and discusses the role of information for electricity consumption. This chapter is based on a randomized controlled field experiment and comes next to a purely randomized experiment situation. While the third chapter is not thematically linked to the first two chapters, all three chapters answer policy-relevant questions and contribute to a better understanding of the causal effects of economic policy interventions.

The first chapter, TAF Effect on Liquidity Risk Exposure, measures how much the Federal Reserve's Term Auction Facility helped to decrease liquidity risk in banks. The paper stresses the importance of the liability term structure and provides empirical support to those arguments in favour of the introduction of liquidity risk measures in international financial regulations, as is now being done with the Basel III accords.

In the second chapter, TARP Effect on Bank Lending Behaviour: Evidence from the last Financial Crisis, we assess the question whether the government is capable to increase lending of the banking sector to the real sector. The main result suggest that banks that benefitted from the government sponsored Troubled Asset Relief Program (TARP) provide on average 19% higher loan originations to small businesses compared to other banks. This result is of particular interest to policy makers in the Euro Area, where lending to the real sector is still contained.

The third chapter is a theoretical contribution in the field of macroeconomics. In the paper Liquid Assets in a Cash-in-Advance Model, I analyse an extension to the cash-in-advance (CIA) model. While the CIA framework usually imposes that only money can be used for purchases, I relax that strong assumption and allow that also a real asset provides some degree of liquidity. The main finding is that if the asset is also liquid, then the asset is valued more than what its real return would suggest. The difference between the effective value in the model and the fundamental value can be interpreted as the liquidity premium of the asset. I also calculate under which circumstances the liquidity premium is positive.

We analyse the strategic behaviours of agents in a market through the appropriate¬ness of their skills to the market. If agents' skills are well adapted to market and they can reach their target, they will not need to adopt strategic behaviours. The agents will behave as selfish individuals. However, if their skills are not well adapted and they cannot attain their target alone, they will adopt strategic behaviours to reach their objectives. These behaviours will have a different impact on the utilities of other agents, depending on the skills and the objectives of the agent.

If these agents need other agents to reach their objectives, they will behave as altruistic individuals who internalise the utilities of other agents in reaching their objectives and will adopt cooperative behaviours. However, if these agents fear that other agents could prevent them from reaching their target because they can foresee that the skills of other agents are better adapted than their own skills, the agents will then behave as predator individuals and will adopt destructive behaviours to attain their objective. It is in the interests of these agents to manipulate information to increase disorder and dissimulate their lack of skills. They will reproduce the strategies of animals that modify their appearance to escape predators or simulate being bait to attract their prey. These agents will seek to induce chaos into the behaviours of other agents to amplify the impact of their strategies.

The appropriateness of skills to the market allows an understanding of the emer-gence of networks and associated strategies. The members of a networks are inputs

who are excluded when their costs are higher than their benefits. A network simul-taneously allows cooperation and selfish, predatory behaviours among its members. A network may adopt informational strategies when seeking to become the leader in a market or when it cannot survive. The creation of networks and the manipulation of information are two overlapping evolutionary strategies, with the first strategy favouring the second.

In our model, an agent does not behave like a firm that aims only to maximise the profits of the firm but rather as a member of a network who adopts strategic behaviours as a function of the interests of this network. If his skills are well adapted to the market and he can innovate, he will not invest in erroneous input; in contrast, if his skills are not adapted, the agent will invest in the erroneous input of information into the market in order to survive.

Therefore, when any informational asymmetries between the agents and their principals characterise the market, the price cannot be the main element that allows equilibrium to be reached in the market; instead, the appropriateness of skills to the market enables equilibrium. We will now apply these hypotheses to explain the strategic behaviours of physicians and pharmaceutical companies.

Zero correlation between measurement error and model error has been assumed in existing panel data models dealing specifically with measurement error. We extend this literature and propose a simple model where one regressor is mismeasured, allowing the measurement error to correlate with model error. Zero correlation between measurement error and model error is a special case in our model where correlated measurement error equals zero. We ask two research questions. First, we wonder if the correlated measurement error can be identified in the context of panel data. Second, we wonder if classical instrumental variables in panel data need to be adjusted when correlation between measurement error and model error cannot be ignored. Under some regularity conditions the answer is yes to both questions. We then propose a two-step estimation corresponding to the two questions. The first step estimates correlated measurement error from a reverse regression; and the second step estimates usual coefficients of interest using adjusted instruments.

A significant share of deliveries are performed by Cesarian section (C-section) in Europe and in many developed and developing countries. The aims of this thesis are to highlight the non medical, especially economic and financial, incentives that explain the use of C-section, as well as the medical consequences of C-section on women's health, in regard with other factors of ob¬stetrical care quality such as hospital concentration. Those diagnoses enable us to exhibit ways of improvement of obstetrical care quality in France. Our analysis focus on two countries, France and Switzerland. In the first part of the thesis, we show the influence of two non medical factors on the C-section use, namely the hospital payment system on the one hand and the obstetricians behaviour, especially their demand for leisure, on the other hand. With French data on the year 2003, we show firstly that the fee-for-service payment system of private for profit hospitals induces a higher probability of using C-section. Obstetricians play also a preeminent role in the decision to use a C-section, as the probability of a C-section rises with the number of obstetricians. We then focus on a French reform introduced in 2004, to investigate the impact of Prospective Payment System on obstetric practise. We show that the rise of C-section rate between 2003 and 2006 is mainly caused by changes in hospitals and patients features. Obstetricians practises do not vary a lot for patients with the same risk code. In the mean time however, the number of women coded with a high risk rises. This can be caused by improvements in the quality of coding, obstetricians chosing codes that match better the real health state of their patients. Yet, it can also show that obstetricians change their coding practises to justify the use of certain practises, such as C-section, with no regard to the health state of patients. Financial factors are not the only non medical fac¬tors that can influence the resort to C-section. Using Shelton Brown ΠΙ identification strategy, we focus on the potential impact of obstetricians leisure preference on the use of C-section. We use the distributions of days and hours of delivering and the types of C-section - planned or emergency C-sections - to show that the obstetricians demand for leisure has a significant impact on the resort to C-section, but only in emergency situations. The second part of the thesis deals with some ways to improve obstetric care quality. We use on the one hand swiss and french data to study the impact of C-section on the patients' probability of having an obstetric complication and on the other hand the influence of hospital concentration on the quality of obstetric care. We find the same results as former medical studies about the risks entailed by C-section on obstetric complications.These results prove women ought to be better informed of the medical consequences of C-section and that the slowing of C-section use should be a priority of public health policy. We finally focus on another way to improve obstetric care quality, that is hospital lmarket concentration. We investigate the impact of hospital concentration by integrating the Herfindahl-Hirschman index in our model, on health care quality, measured by the HCUP indicator. We find that hospital concentration has a negative impact on obstetric care quality, which undermines today's policy of hospital closings in France.

Keywords: Hospital; C-section; Payment System; Counterfactual Estimation; Quality of Care.

Thesis co-supervised with the Ecole des Hautes Etudes en Sciences Sociales, Paris

Output, inflation and interest rates are key macroeconomic variables, in particular for monetary policy. In modern macroeconomic models they are driven by random shocks which feed through the economy in various ways. Models differ in the nature of shocks and their transmission mechanisms. This is the common theme underlying the three essays of this thesis. Each essay takes a different perspective on the subject: First, the thesis shows empirically how different shocks lead to different behavior of interest rates over the business cycle. For commonly analyzed shocks (technology and monetary policy errors), the patterns square with standard models. The big unknown are sources of inflation persistence. Then the thesis presents a theory of monetary policy, when the central bank can better observe structural shocks than the public. The public will then seek to infer the bank's extra knowledge from its policy actions and expectation management becomes a key factor of optimal policy. In a simple New Keynesian model, monetary policy becomes more concerned with inflation persistence than otherwise. Finally, the thesis points to the huge uncertainties involved in estimating the responses to structural shocks with permanent effects.

These three chapters, while fairly independent from each other, study economic situations in incomplete contract settings. They are the product of both the academic freedom my advisors granted me, and in this sense reflect my personal interests, and of their interested feedback. The content of each chapter can be summarized as follows:

Chapter 1: Inefficient durable-goods monopolies
In this chapter we study the efficiency of an infinite-horizon durable-goods monopoly model with a fmite number of buyers. We find that, while all pure-strategy Markov Perfect Equilibria (MPE) are efficient, there also exist previously unstudied inefficient MPE where high valuation buyers randomize their purchase decision while trying to benefit from low prices which are offered once a critical mass has purchased. Real time delay, an unusual monopoly distortion, is the result of this attrition behavior. We conclude that neither technological constraints nor concern for reputation are necessary to explain inefficiency in monopolized durable-goods markets.

Chapter 2: Downstream mergers and producer's capacity choice: why bake a larger pie when getting a smaller slice?
In this chapter we study the effect of downstream horizontal mergers on the upstream producer's capacity choice. Contrary to conventional wisdom, we find anon-monotonic relationship: horizontal mergers induce a higher upstream capacity if the cost of capacity is low, and a lower upstream capacity if this cost is high. We explain this result by decomposing the total effect into two competing effects: a change in hold-up and a change in bargaining erosion.

Chapter 3: Contract bargaining with multiple agents
In this chapter we study a bargaining game between a principal and N agents when the utility of each agent depends on all agents' trades with the principal. We show, using the Potential, that equilibria payoffs coincide with the Shapley value of the underlying coalitional game with an appropriately defined characteristic function, which under common assumptions coincides with the principal's equilibrium profit in the offer game. Since the problem accounts for differences in information and agents' conjectures, the outcome can be either efficient (e.g. public contracting) or inefficient (e.g. passive beliefs).

Although the chapters of this thesis address a variety of issues, the principal aim is common: test economic ideas in an international economic context. The intention has been to supply empirical findings using the largest suitable data sets and making use of the most appropriate empirical techniques.
This thesis can roughly be divided into two parts: the first one, corresponding to the first two chapters, investigates the link between trade and the environment, the second one, the last three chapters, is related to economic geography issues. Environmental problems are omnipresent in the daily press nowadays and one of the arguments put forward is that globalisation causes severe environmental problems through the reallocation of investments and production to countries with less stringent environmental regulations. A measure of the amplitude of this undesirable effect is provided in the first part. The third and the fourth chapters explore the productivity effects of agglomeration. The computed spillover effects between different sectors indicate how cluster-formation might be productivity enhancing. The last chapter is not about how to better understand the world but how to measure it and it was just a great pleasure to work on it. "The Economist" writes every week about the impressive population and economic growth observed in China and India, and everybody agrees that the world's center of gravity has shifted. But by how much and how fast did it shift? An answer is given in the last part, which proposes a global measure for the location of world production and allows to visualize our results in Google Earth.

At least since the Great Depression, explaining why there are business fluctuations has been one of the biggest challenges that the science of economics has had to face. The hope is that if we could better understand recessions, then we could also be more successful in overcoming them. This dissertation consists of three papers that are part of the general endeavor of economists to understand these fluctuations. The first paper discusses, for a particular model, whether a result related to fluctuations would still hold if time were modeled as continuous rather than discrete. The two other papers focus on price stickiness. The second paper discusses why, after a large devaluation, prices of non-tradables may change by only a small amount in comparison to the magnitude of the devaluation. The third paper examines price adjustment in a model in which information is imperfect and it is costly to change prices.

This paper analyses the outcomes of the EEA and bilateral agreements vote at the level of the 3025 communities of the Swiss Confederation by simultaneously modelling the vote and the participation decisions. Regressions include economic and political factors. The economic variables are the aggregated shares of people employed in the losing, Winning and neutral sectors, according to BRUNETTI, JAGGI and WEDER (1998) classification, Which follows a Ricardo-Viner logic, and the average education levels, which follows a Heckscher-Ohlin approach. The political factors are those used in the recent literature. The results are extremely precise and consistent. Most of the variables have the predicted sign and are significant at the l % level. More than 80 % of the communities' vote variance is explained by the model, substantially reducing the residuals when compared to former studies. The political variables do have the expected signs and are significant as Well. Our results underline the importance of the interaction between electoral choice and participation decisions as well as the importance of simultaneously dealing with those issues. Eventually they reveal the electorate's high level of information and rationality.

The unifying theme of this thesis is the pursuit of a satisfactory ways to quantify the riskureward trade-off in financial economics. First in the context of a general asset pricing model, then across models and finally across country borders. The guiding principle in that pursuit was to seek innovative solutions by combining ideas from different fields in economics and broad scientific research. For example, in the first part of this thesis we sought a fruitful application of strong existence results in utility theory to topics in asset pricing. In the second part we implement an idea from the field of fuzzy set theory to the optimal portfolio selection problem, while the third part of this thesis is to the best of our knowledge, the first empirical application of some general results in asset pricing in incomplete markets to the important topic of measurement of financial integration. While the first two parts of this thesis effectively combine well-known ways to quantify the risk-reward trade-offs the third one can be viewed as an empirical verification of the usefulness of the so-called "good deal bounds" theory in designing risk-sensitive pricing bounds.

This thesis presents three empirical studies in the field of health insurance in Switzerland. First we investigate the link between health insurance coverage and health care expenditures. We use claims data for over 60 000 adult individuals covered by a major Swiss Health Insurance Fund, followed for four years; the data show a strong positive correlation between coverage and expenditures. Two methods are developed and estimated in order to separate selection effects (due to individual choice of coverage) and incentive effects ("ex post moral hazard"). The first method uses the comparison between inpatient and outpatient expenditures to identify both effects and we conclude that both selection and incentive effects are significantly present in our data. The second method is based on a structural model of joint demand of health care and health insurance and makes the most of the change in the marginal cost of health care to identify selection and incentive effects. We conclude that the correlation between insurance coverage and health care expenditures may be decomposed into the two effects: 75% may be attributed to selection, and 25 % to incentive effects. Moreover, we estimate that a decrease in the coinsurance rate from 100% to 10% increases the marginal demand for health care by about 90% and from 100% to 0% by about 150%.
Secondly, having shown that selection and incentive effects exist in the Swiss health insurance market, we present the consequence of this result in the context of risk adjustment. We show that if individuals choose their insurance coverage in function of their health status (selection effect), the optimal compensations should be function of the se- lection and incentive effects. Therefore, a risk adjustment mechanism which ignores these effects, as it is the case presently in Switzerland, will miss his main goal to eliminate incentives for sickness funds to select risks. Using a simplified model, we show that the optimal compensations have to take into account the distribution of risks through the insurance plans in case of self-selection in order to avoid incentives to select risks.Then, we apply our propositions to Swiss data and propose a simple econometric procedure to control for self-selection in the estimation of the risk adjustment formula in order to compute the optimal compensations.

The field of public finance focuses on the spending and taxing activities of governments and their influence on the allocation of resources and distribution of income. This work covers in three parts different topics related to public finance which are currently widely discussed in media and politics. The first two parts deal with issues on social security, which is in general one of the biggest spending shares of governments. The third part looks at the main income source of governments by analyzing the perceived value of tax competition.