Working Papers 2018
2018-06: Negotiating Cooperation Under Uncertainty: Communication in Noisy, Indefinitely Repeated Interactions
Fabian Dvorak and Sebastian Fehrler (University of Konstanz)
Case studies of cartels and recent theory suggest that repeated communication is key for stable cooperation in environments where signals about others' actions are noisy. However, empirically the exact role of communication is not well understood. We study cooperation under different monitoring and communication structures in the laboratory. Under all monitoring structures - perfect, imperfect public, and imperfect private - communication boosts efficiency. However, under imperfect monitoring, where actions can only be observed with noise, cooperation is stable only when subjects can communicate before every round of the game. Beyond improving coordination, communication increases efficiency by making subjects' play more lenient and forgiving. We further find clear evidence for the exchange of private information - the central role ascribed to communication in recent theoretical contributions.
Keywords: innitely repeated games, monitoring, communication, cooperation, strategic uncertainty, prisoner's dilemma
JEL Classication: C72, C73, C92, D83
2018-05: Sparse Approximate Factor Estimation for High-Dimensional Covariance Matrices
Maurizio Daniele, Winfried Pohlmeier and Aygul Zagidullina (University of Konstanz)
We propose a novel estimation approach for the covariance matrix based on the l1-regularized approximate factor model. Our sparse approximate factor (SAF) covariance estimator allows for the existence of weak factors and hence relaxes the pervasiveness assumption generally adopted for the standard approximate factor model. We prove consistency of the covariance matrix estimator under the Frobenius norm as well as the consistency of the factor loadings and the factors.
Our Monte Carlo simulations reveal that the SAF covariance estimator has superior properties in finite samples for low and high dimensions and different designs of the covariance matrix. Moreover, in an out-of-sample portfolio forecasting application the estimator uniformly outperforms alternative portfolio strategies based on alternative covariance estimation approaches and modeling strategies including the 1/N-strategy.
Keywords: Approximate Factor model, weak factors, l1-regularization, high dimensional covariance matrix, portfolio allocation
JEL classification: C38, C55, G11, G17
2018-04: How Informative is High-Frequency Data for Tail Risk Estimation and Forecasting? An Intrinsic Time Perspectice
Timo Dimitriadis and Roxana Halbleib (University of Konstanz)
This paper proposes a novel and simple approach to compute daily Value at Risk (VaR) and Expected Shortfall (ES) directly from high-frequency data. It assumes that financial logarithm prices are subordinated unifractal processes in the intrinsic time, which stochastically transforms the clock time in accordance with the markets activity. This is a very general assumption that allows for a simple computation of daily VaR and ES by scaling up their intraday counterparts computed from data sampled in intrinsic time. In the empirical exercise, we discuss the statistical and dynamic properties of the resulting daily VaR and ES estimates and show that our method outperforms standard ones in accurately estimating and forecasting VaR and ES.
Keywords: Value at Risk, Expected Shortfall, Intrinsic Time, Subordinated Process, High-Frequency Data, Scaling Law
2018-03: Job Displacement and Training Activities: Human Capital Accumulation during the East German Transition
Franziska K. Deutschmann (University of Konstanz)
With the East German transition from a command to a market economy in the early 1990s, many East Germans lost their jobs because of firm closures while their human capital was not yet aligned to a modern market economy. I study how displaced East German workers reacted to that double effect of human capital depreciation. Using a difference-in-difference-in-difference set-up, I compare the effect of displacement on adult education between East and West German workers. In line with human capital theory, I find that East Germans increase their time investment in adult training signicantly when they were displaced one or two years earlier. This increase in adult training is lower for older workers and not observable for West Germans. Moreover, the results suggest that when East Germans lose their job due to a plant closure, they prepone training activities to the time when they are not employed. In that sense, firm closures may have useful side-effects in a transition period as they allow people to work earlier against the human capital depreciation associated with the transition.
2018-02: The Micro-Foundations of the Resource Curse: Oil Ownership and Local Economic Well-Being in Sub-Saharan Africa
Tim Wegenast, Arpita Khanna and Gerald Schneider (University of Konstanz)
Empirical tests of the “resource curse” thesis have provided inconclusive evidence for the core claim that natural resource abundance lowers economic growth. Some macro-level studies argue that the key expectation holds if one controls for the level of democracy. However, even if we account for the quality of institutions, some democratic resource-rich countries experience low growth rates. We argue in line with a new contest model and recent findings that these anomalies are a consequence of different control rights regimes and that state-controlled resource production stimulates local income more than privately-controlled extraction. Our micro-level arguments are tested using information on local economic activity and a new data set that establishes the control rights over hydrocarbons at the individual extraction site of the resource. Relying on this novel data, we perform district and grid-level analyses of sub-Saharan Africa covering the period from 1997 to 2014. Our multi-level and two-way fixed effects linear models show that the presence of domestic national oil companies is associated with increased local growth, while international oil companies show no effect on economic development. We also find that state-controlled oil production particularly furthers local economic well-being under democratic institutions, good governance and low levels of corruption.
2018-01: Identification of Structural Vector Autoregressions by Stochastic Volatility
Dominik Bertsche and Robin Braun (University of Konstanz)
We propose to exploit stochastic volatility for statistical identification of Structural Vector Autoregressive models (SV-SVAR). We discuss full and partial identification of the model and develop efficient EM algorithms for Maximum Likelihood inference. Simulation evidence suggests that the SV-SVAR works well in identifying structural parameters also under misspecification of the variance process, particularly if compared to alternative heteroskedastic SVARs. We apply the model to study the interdependence between monetary policy and stock markets. Since shocks identified by heteroskedasticity may not be economically meaningful, we exploit the framework to test conventional exclusion restrictions as well as Proxy SVAR restrictions which are overidentifying in the heteroskedastic model.
Keywords: Structural Vector Autoregression (SVAR), Identication via heteroskedasticity, Stochastic Volatility, Proxy SVAR
JEL classication: C32
Working Papers 2017
2017-18: Low Homeownership in Germany - A Quantitative Exploration
Leo Kaas, University of Konstanz, Georgi Kocharkov, University of Konstanz, Edgar Preugschat, Technical University of Dortmund and Nawid Siassi, University of Konstanz
The homeownership rate in Germany is one of the lowest among advanced economies. To better understand this fact, we analyze the role of three specic policies which discourage homeownership in Germany: an extensive social housing sector with broad eligibility criteria, high transfer taxes when buying real estate, and no tax deductions for mortgage interest payments by owner-occupiers. We build a lifecycle model with uninsurable income risk and endogenous homeownership in order to quantify the policy effects on homeownership and welfare. We find that all three policies have sizable effects on the homeownership rate. At the same time, household welfare would be reduced by moving to a policy regime with low transfer taxes and mortgage interest tax deductions, but it would improve in the absence of social housing, in particular when coupled with housing subsidies for low-income households.
JEL classication: D15; E21; R21; R38
Keywords: Homeownership, Housing markets
2017-17: Testing Out-of-Sample Portfolio Performance
Ekaterina Kazak, University of Konstanz and Winfried Pohlmeier, University of Konstanz
This paper studies the quality of portfolio performance tests based on out-of-samplereturns. By disentangling the components of out-of-sample performance we show thatobserved dierences are driven to a large extent by the dierences in estimation risk. OurMonte Carlo study reveals that the puzzling empirical ndings of inferior performance oftheoretically superior strategies mainly result from the low power of these tests. Thus ourresults provide an explanation why the null hypothesis that the simple equally weightedportfolio cannot be outperformed by theoretically superior portfolio strategies can not berejected in many out-of-sample horse races regardless of the underlying testing strategy.For the applied researcher we provide some guidance to cope with the problem of lowpower. In particular, we show by the means of a novel pretest-based portfolio strategy, howthe information of performance tests can be used optimally.
Keywords: portfolio tests, estimation risk, power, pretest estimation
JEL classication: G11, C58, C12, C51
2017-16: Identification of SVAR Models by Combining Sign Restrictions With External Instruments
Robin Braun, University of Konstanz and Ralf Brüggemann, University of Konstanz
We identify structural vector autoregressive (SVAR) models by combining signrestrictions with information in external instruments and proxy variables. We incorporatethe proxy variables by augmenting the SVAR with equations that relate them tothe structural shocks. Our modeling framework allows to simultaneously identify differentshocks using either sign restrictions or an external instrument approach, alwaysensuring that all shocks are orthogonal. The combination of restrictions can also beused to identify a single shock. This entails discarding models that imply structuralshocks that have no close relation to the external proxy time series, which narrowsdown the set of admissible models. Our approach nests the pure sign restriction caseand the pure external instrument variable case. We discuss full Bayesian inference,which accounts for both, model and estimation uncertainty. We illustrate the usefulnessof our method in SVARs analyzing oil market and monetary policy shocks. Ourresults suggest that combining sign restrictions with proxy variable information is apromising way to sharpen results from SVAR models.
Keywords: Structural vector autoregressive model, sign restrictions, external instruments
JEL classication: C32, C11, E32, E52
2017-15: Directed Graphs and Variable Selection in Large Vector Autoregressive Models
Ralf Brüggemann, University of Konstanz and Christian Kascha
We represent the dynamic relation among variables in vector autoregressive (VAR) models as directed graphs. Based on these graphs, we identify so-called strongly connected components (SCCs). Using this graphical representation, we consider the problem of variable selection. We use the relations among the strongly connected components to select variables that need to be included in a VAR if interest is in forecasting or impulse response analysis of a given set of variables. We show that the set of selected variables from the graphical method coincides with the set of variables that is multi-step causal for the variables of interest by relating the paths in the graph to the coecients of the 'direct' VAR representation. Empirical applications illustrate the usefulness of the suggested approach: Including the selected variables into a small US monetary VAR is useful for impulse response analysis as it avoids the well-known 'price-puzzle'. We also find that including the selected variables into VARs typically improves forecasting accuracy at short horizons.
Keywords: Vector autoregression, Variable selection, Directed graphs, Multi-stepcausality, Forecasting, Impulse response analysis
JEL classication: C32, C51, C55, E52
2017-14: A Latent Factor Model for Forecasting Realized Volatilities
Giorgio Calzolari, Università degli Studi di Firenze, Roxana Halbleib, University of Konstanz, Aygul Zagidullina, University of Konstanz
This paper proposes a new method of forecasting realized volatilities by exploiting their common dynamics within a latent factor model. The main idea is to use an additive component structure to describe the long-persistence in their autocorrelation function, where the components, extracted from high-dimensional vectors of realized volatilities, follow stationary autoregressive processes of order 1. The model we propose allows also for autoregressive structures in the idiosyncratic noises and conditional hetersokedasticity. Differently from HAR and ARFIMA, our factor model profits from the high-dimensionality of the system that provides more information of the commonality of their dynamics with direct efficiency gains in the estimates and forecasts. For estimation purposes, we use the indirect inference method that is easy to implement and provides accurate estimates. We apply the new models to vectors of up to 30 daily realized volatility series of stocks composing the Dow Jones Industrial Average index and show that they outperform standard long-memory models both in-sample and out-of-sample.
Keywords: Long Memory, Component Model, Dynamic Factor Model, Factor-GARCH Model, Indirect Inference
2017-13: Public opinion and social investment: How political-institutional context shapes support and opposition towards expanding childcare
Erik Neimanns, University of Konstanz
Social investment policies are generally considered to be widely popular among the public and policymakers alike as they are expected to generate social and economic benefits, and to create political payoffs for governments implementing such policies. However, empirically, we observe strong and persistent variation across countries in their design of social investment policies. This variation presents an important empirical puzzle, given the postulated positive returns associated with such policies. Focusing on early childhood education and care as a central element of social investment, I argue in this theoretical contribution that once we take into account the country-specific political and institutional context, the popularity of social investment should not be taken for granted. Contingent on who benefits from expansive childcare policies there can be a substantial potential of conflict in public attitudes between different societal groups. Building on theories of policy feedback, I elaborate the concept of perceived relative policy payoffs. How individuals perceive the costs and benefits associated with childcare policies can attenuate or amplify the potential of conflict in public attitudes towards expanding childcare. If an expansion of childcare comes at the cost of some groups in society and if reforms benefit only more narrowly defined groups, preferences are likely to be more conflictive and the political viability of expansive reforms appears more uncertain. The framework developed in this paper has implications for theories of policy feedback and the politics of social investment. Furthermore, it suggests that instead of a convergence towards a fully fletched paradigmatic social investment welfare state, we are likely to observe persistent variation across countries in their design of social investment policies.
2017-12: Measuring willingness to pay for childcare: Findings from a vignette study in the city of Konstanz
Marius R. Busemeyer, University of Konstanz, Achim Goerres, University of Duisburg-Essen
This paper is about the perceived level of fair fees for childcare usage among residents of a German city. The analysis is based on original data from a vignette study of Konstanz residents, administered in the fall of 2014. The paper studies, which level of fees is perceived as fair and which factors influence these fairness perceptions – either directly or moderated by respondent characteristics. We find that residents in Konstanz on average consider a fee of 191 Euro to be fair, which is slightly below the average fee level that citizens had to pay at the time of the survey. The most important factors influencing the level of (perceived) fair fees are related to resourcefulness: High-income citizens are expected to pay more. Single parents and those without support from grandparents are allowed to contribute less. Religious background, employment status as well as having local roots do not affect the levels of perceived fair fees. The acceptance of the model of income-dependent fees indicates strong feedback effects from policies and institutions at the national level as the local fee level deviates from this most common model in Germany by raising fees independent of income.
2017-11: Combining Value-at-Risk Forecasts Using Penalized Quantile Regression
Sebastian Bayer, University of Konstanz
Elastic net penalized quantile regression is proposed for the combination of Value-at-Risk forecasts. The main reason for this regularization is the multicollinearity among the standalone forecasts, which leads to a poor forecast performance of the unpenalized quantile regression estimator due to unstable combination weights. The new approach is applied to combining the Value-at-Risk forecasts of 17 frequently used standalone models for 30 stocks comprising the Dow Jones Industrial Average Index. Within a thorough comparison analysis, the elastic net quantile regression performs better in terms of backtesting than the standalone models and several competing combination approaches. This is particularly the case during the global financial crisis from 2007 / 2008.
Keywords: Value-at-Risk, Forecast Combination, Quantile Regression, Elastic Net
JEL: C51, C52, C53, G32
2017-10: Productivity, Taxation and Evasion: An Analysis of the Determinants of the Informal Economy
Alessandro Di Nola, University of Konstanz, Georgi Kocharkov, University of Konstanz and Aleksandar Vasilev, American University in Bulgaria
We evaluate the relative importance of labor productivity versus income taxes and social contributions for tax compliance in an economy with a large degree of informality. Empirical evidence points out that tax evasion in Europe happens through partially concealing wages and profits in formally registered enterprises. To this end, we build a model in which employer-employee pairs of heterogeneous productive capacities make joint decisions on the degree of tax evasion. The quantitative model takes as inputs the income tax structure and social contributions. The model is used to analyze the case of Bulgaria which has the largest informal economy in Europe. The estimation strategy relies on matching the empirical series for the size of the informal economy and other aggregate outcomes for 2000-2014. Our counterfactual experiments show that the most important factor for the changing size of the informal economy is labor productivity, which accounts for more than 75% of the change. The variation in corporate income tax accounts for the rest. Changes in personal income tax levels and progressivity are found not to be quantitatively relevant for tax evasion. We also characterize optimal taxation in 2014 with respect to minimizing tax evasion. The productive gains of imposing optimal taxes are small.
JEL Classifications: H24, H25, H26, C63, E62, E65.
Keywords: Informal economy, progressive taxation, tax evasion, flat tax reform.
2017-09: The CAPM with Measurement Error: There's life in the old dog yet!
Anastasia Morozova and Winfried Pohlmeier, University of Konstanz
This paper takes a closer look on the consequences of using a market index as a proxy for the latent market return in the capital asset pricing model. In particular, the consequences of two major sources of misspecification are analyzed: (i) the use of inaccurate weights and (ii) the use of only a subset of the asset universe to construct the index. The consequences resulting from the use of a badly chosen market proxy reach from inconsistent parameter estimates to misinterpretation of tests indicating the existence of abnormal returns. A new minimum distance approach of estimating the CAPM under measurement error is presented, which identifies the CAPM parameters by exploiting the cross-equation cross-sectional restrictions resulting from a common measurement error. The new approach allows for quantifying the impact of measurement error and for testing the presence of spurious abnormal returns. Practical guidelines are presented to mitigate potential biases in the estimated CAPM parameters.
Keywords: CAPM, Measurement Error, Roll's critique, Identification, Minimum Distance Estimation
2017-08: A Joint Quantile and Expected Shortfall Regression Framework
Timo Dimitriadis and Sebastian Bayer (University of Konstanz)
We introduce a novel regression framework which simultaneously models the quantile and the Expected Shortfall of a response variable given a set of covariates. The foundation for this joint regression is a recent result by Fissler and Ziegel (2016), who show that the quantile and the ES are jointly elicitable. This joint elicitability allows for M- and Z-estimation of the joint regression parameters. Such a parameter estimation is not possible for an Expected Shortfall regression alone as Expected Shortfall is not elicitable. We show consistency and asymptotic normality for the M- and Z-estimator under standard regularity conditions. The loss function used for the M-estimation depends on two specification functions, whose choices affect the properties of the resulting estimators. In an extensive simulation study, we verify the asymptotic properties and analyze the small sample behavior of the M-estimator under different choices for the specification functions. This joint regression framework allows for various applications including estimating, forecasting and backtesting Expected Shortfall, which is particularly relevant in light of the upcoming introduction of Expected Shortfall in the Basel Accords.
Keywords: Expected Shortfall, Joint Elicitability, Joint Regression, M-estimation
2017-07: Family Planning and Development: Aggregate Effects of Contraceptive Use
Tiago Cavalcanti, University of Cambridge and EESP/FGV-SP, Georgi Kocharkov, University of Konstanz and Cezar Santos, EPGE/FGV-Rio
What is the role of family planning interventions on fertility, savings, human capital investment, and development? To examine this, endogenous unwanted fertility is embedded in an otherwise standard quantity-quality overlapping generations model of fertility and growth. The model features costly fertility control and families can (partially) insure against a fertility risk by using costly modern contraceptives. In the event of unexpected pregnancies, households can also opt to abort some pregnancies, at a cost. Given the number of children born, parents decide how much education to provide and how much to save out of their income. We fit the model to Kenyan data, implement several family planning policies and decompose their aggregate effects. Our results suggest that given a small budget (up to 0.5 percent of GDP), legalizing and subsidizing the price of abortion is a more cost-effective policy for improving long-run living standards and reducing inequality than policies that either subsidize the price of modern contraceptives or subsidize basic education.
Keywords: education, income per capita, contraception, abortion
JEL Classification: E24, I15, J13, O11.
2017-06: Testing learning mechanisms of rule-based judgment
Janina Hoffmann, University of Konstanz, Bettina von Helversen, University of Zürich and Jörg Rieskamp, University of Basel
Weighing the importance of different pieces of information is a key determinant of making
accurate judgments. In social judgment theory, these weighting processes have been
successfully modeled with linear models. How people learn to make judgments has received
less attention. Although the hitherto proposed least mean squares or delta learning rule
can perfectly learn to solve linear problems, we found in a first study that it does not
adequately describe human learning. To provide a more accurate description of learning
processes we amended the delta learning rule with three learning mechanisms —a decay, an
attentional learning mechanism, and a capacity limitation —and tested in a further study
how well those learning mechanisms can describe and predict learning in linear judgment
In the study, participants first learned to predict a continuous criterion based on four
cues. To test the three learning mechanisms rigorously against each other, we changed the
importance of the cues after 200 trials so that the mechanisms make different predictions
with regard to how fast people adapt to the new environment. On average, judgment
accuracy improved from trial 1 to 200, dropped when the task structure changed, but
improved again until the end of the task. The capacity-restricted learning model best
described and predicted the learning curve of the majority of participants. Taken together,
these results suggest that human learning when making inferences is governed by cognitive
Keywords: Multiple-cue Judgment; Rule-based Processes; Learning
2017-05: Tracing the path of forgetting in rule abstraction and exemplar retrieval
Janina Hoffmann, University of Konstanz, Bettina von Helversen, University of Zürich, Regina A. Weilbächer, University of Basel and Jörg Rieskamp, University of Basel
People often forget acquired knowledge over time such as names of former classmates. Which knowledge people can access, however, may modify the judgment process and affect judgment accuracy. Specifically, we hypothesized that judgments based on retrieving past exemplars from long-term memory may be more vulnerable to forgetting than remembering rules that relate the cues to the criterion. Experiment 1 tracked the individual course of forgetting in a judgment task facilitating rule-based or exemplar-based strategies by systematically prolonging the retention interval between a training in which participants learned to make judgments and later tests (immediate, one day, and one week). Practicing the acquired judgment strategy in repeated tests helped participants to consistently apply the learnt judgment strategy and retain a high judgment accuracy even after a week. Yet, whereas a long retention interval did not affect judgments in the rule-based task, a long retention interval impaired judgments in the exemplar-based task. If practice was restricted as in Experiment 2, judgment accuracy suffered in both tasks. In addition, after a week without practice participants tried to reconstruct their judgments by applying rules in the exemplar-based task. These results emphasize that the extent to which decision makers can still retrieve previously learned knowledge limits their ability to make accurate judgments and that the preferred strategies change over time if the opportunity for practice is limited.
Keywords: Judgment, forgetting, rule-based and exemplar-based processes
2017-04: Justifying the judgment process affects confidence but neither accuracy, nor strategy use
Janina Hoffmann, University of Konstanz, Wolfgang Gaissmaier, University of Konstanz and Bettina von Helversen, University of Zürich
The quality of a decision in daily life is often evaluated based on whether one can adequately explain how one came to this decision. Holding decision makers accountable for the decision process often improves judgment quality because decision makers weigh and integrate information more thoroughly. Our research aimed to identify the conditions under which process accountability may lead to favorable or disadvantageous judgments. Specifically, we hypothesized that process accountability may hinder accurate judgments in tasks that require remembering previously encountered cases (Experiment 1) whereas it may improve judgment accuracy in tasks that require weighing and integrating information (Experiment 2). To test these hypothesis, participants learned to solve either a multiplicative (Experiment 1) or a linear multiple-cue judgment task (Experiment 2) with or without accountability instructions. We manipulated process accountability by randomly selecting trials in which participants had to justify their judgment. In both experiments, process accountability neither changed how accurately people made a judgment, nor the cognitive processes underlying their judgments. In trials asking participants to justify their judgment process participants were less confident about their judgments. Overall, these results imply that process accountability may impact judgment quality only to a small. We discuss the limits of our results with respect to the eectiveness of the manipulation.
Keywords: Judgment; Accountability; Cognitive processes
2017-03: Foundation Owned Firms - A Detailed Decomposition of Differences in Return Distributions
Matthias Draheim, University of Konstanz and Phillip Heiler, University of Konstanz
We study if and how differences in firm policies have an impact on differences in return on assets of foundation owned firms (FoFs) and other firms (Nons) in Germany. We perform the analysis for return differences at the mean and at several quantiles. We document that for high-performing FoFs and Nons the return on asset difference is more pronounced in favor of Nons, for low-performing firms the difference vanishes. Performance differences are substantially driven by differences in firm policies. We find that (1) lower risk in FoFs increases FoF underperformance at high quantiles, but offsets it at low quantiles, (2) a lower leverage of FoFs offsets FoF underperformance at low quantiles and the median, (3) higher labor intensity in FoFs offsets their underperformance at low quantiles, (4) the larger size of FoFs increases FoF underperformance for the mean and all quantiles, (5) lower growth rates of FoFs increase FoF underperformance at high quantiles, and (6) residual differences
beyond firm policies tend to be insignificant.
2017-02: We Belong Together - A Cross-Smoothing Approach for Non-overlapping Group Mean Estimates
Phillip Heiler, University of Konstanz and Jana Mareckova, University of Konstanz
We propose a general framework for estimating means of orthogonal groups stemming from e.g. categorical regressors or (quasi-)experimental data. It penalizes the loss function by adding squared L2-norm differences between group location parameters and a first stage estimate for potentially all other groups. Under quadratic loss, the penalized estimation problem has a simple interpretable closed form solution that is related to methods established in the literature on discretized support smoothing kernels and model averaging methods. We provide optimal smoothing parameters that serves as a benchmark method, propose a plug-in approach and study their comparative statics. The behavior of both methods is analyzed in an asymptotic local to zero framework that allows for the
presence of moderate test statistics for arbitrary sample sizes. We introduce a class of sequences for close and distant systems that is sufficient for describing a large range of data generating processes. We show consistency, derive the asymptotic distribution under fixed, theoretically optimal and estimated smoothing parameters and provide upper limits for the statistical complexity of the estimators. The method is applied to the estimation of time trends in a short panel based on the Haifa field experiment by Gneezy and Rustichini (2000a) and to the difference-in-differences minimum wage study by Card and Krueger (1994).
Keywords: Categorical data analysis, shrinkage, smoothing
2017-01: Teenage Childbearing and the Welfare State
Andra Filote, Georgi Kocharkov and Jan Mellert (University of Konstanz)
Teenage childbearing is a common incident in developed countries. However, the occurrence of teenage births is much more likely in the United States than in any other industrialized country. The majority of these births are delivered by female teenagers coming from low-income families. The hypothesis put forward here is that the welfare state (a set of redistributive institutions) plays a significant role for teenage childbearing behavior. We develop an economic theory of parental investments and risky sexual behavior of teenagers. The model is estimated to fit stylized facts about income inequality, intergenerational mobility and sexual behavior of teenagers in the United States. The welfare state institutions are introduced via tax and public education expenditure functions derived from U.S. data. In a quantitative experiment, we impose Norwegian taxes and/or education spending in the economic environment. The Norwegian welfare state institutions go a long way in explaining the differences in teenage birth rates between the United States and Norway.
Keywords: Teenage risky sexual behavior, teenage birth rates, progressive taxation, education, redistribution.
Working Papers 2016
2016-05: Volatility forecasting using global stochastic financial trends extracted from non-synchronous data
Lyudmila Grigoryeva, University of Konstanz, Juan-Pablo Ortega, University of St. Gallen and Anatoly Peresetsky, International Laboratory of Quantitative Finance, Higher School of Economics, Moscow
This paper introduces a method based on various linear and nonlinear state space models that are used to extract global stochastic financial trends (GST) out of non-synchronous financial data. More specically, these models are constructed to take advantage of the intraday arrival of closing information coming from different international markets to improve volatility description and forecasting. A set of three major asynchronous international stock market indices is used in order to empirically show that this forecasting scheme is capable of signicant performance gains when compared to standard models like the dynamic conditional correlation (DCC) family.
Keywords: multivariate volatility modeling and forecasting, global stochastic trend, extended Kalman filter, dynamic conditional correlations (DCC), non-synchronous data.
2016-04: Singular ridge regression with homoscedastic residuals: generalization error with estimated parameters
Lyudmila Grigoryeva, University of Konstanz and Juan-Pablo Ortega, University of St. Gallen
This paper characterizes the conditional distribution properties of the nite sample ridge regression estimator and uses that result to evaluate total regression and generalization errors that incorporate the inaccuracies committed at the time of parameter estimation. The paper providesexplicit formulas for those errors. Unlike other classical references in this setup, our results take place in a fully singular setup that does not assume the existence of a solution for the non-regularized regression problem. In exchange, we invoke a conditional homoscedasticity hypothesis on the regularized regression residuals that is crucial in our developments.
Key Words: ridge regression, singular regression, training error, testing error, generalization error, regularization methods, high-dimensional regression.
2016-03: Sources of German Income Inequality across Time and Space
Franziska K. Deutschmann, University of Konstanz
Income inequality rose in Germany since the 1970s. To quantify the impact of socio-economic trends on inequality, I construct counterfactual distributions of net household income with rich German data. The procedure controls for marital sorting in education and allows for indirect influences, such as the influence of education on employment. I find that the prevalence of singlehood can account for the observed increase in inequality since the 1970s to a vast extent. The inequality increase is also associated with a change of employment among males and single females. Compared to West Germany, in 2011, the stronger labor attachment of East German married females combined with the high East German unemployment produce more income inequality. East-West differences in children per household also boost East German inequality. East German education works against it. I find no evidence that positive assortative mating in education or the aging society augment income inequality significantly.
JEL classication: D31, E25, I24, J11, J12, J21.
Keywords: inequality, demography, household structure, assortative mating,
education, employment, Germany
2016-02: Anti-poverty Income Transfers in the U.S.: A Framework for the Evaluation of Policy Reforms
Salvador Ortigueira, University of Miami and Nawid Siassi, University of Konstanz
We develop a dynamic model of labor supply, consumption, savings and marriage decisions to study the behavioral responses of low-income workers to anti-poverty income transfers in the U.S. The model is calibrated to match moments from a sample of non-college-educated workers with children drawn from the 2014 Annual Social and Economic Supplement. The categorical, asset and income eligibility criteria of the transfer programs, along with the income and payroll taxes, yield complex budget constraints and introduce a web of interactions whose eects we identify and measure. We examine the workers' behavioral responses across the model's equilibrium distribution over living arrangements, labor productivities, wealth and number of children. Then we use the model to assess the eects of three recent proposals to reform the U.S. tax-transfer system, including the "21st Century Worker Tax Cut Act" and the "Tax Reform Act of 2014". A core objective of these proposals is the mitigation of the disincentives introduced by the Earned Income Tax Credit to married mothers' labor market participation.
JEL-Codes: E21, H24, H31, J12
Keywords: Anti-poverty income transfers; household decisions; cohabitation and marriage.
2016-01: Revisiting the Oil Curse: Does Ownership Matter?
Arpita Khanna, University of Konstanz
A large body of scholarship finds a negative relationship between oil abundance and economic growth. The existing empirical evidence on the oil curse, however, does not account for variations in the ownership of oil. This article investigates whether the effect of oil abundance on growth varies with ownership structures. It also investigates whether institutional quality influences the effect of different ownership structures. Using a novel database on ownership structures and employing a panel fixed-effects estimation method, it analyzes a sample of 20 oil-exporting developing countries during the period 1984-2005. The results show that different ownership structures have differential effects on growth, and that their effect is influenced by the quality of institutions. State ownership and control reduces growth when the institutional quality is poor, but increases growth when the institutional quality is good. Private ownership, on the other hand, increases growth when the institutional quality is poor, but reduces growth when the institutional quality is good. The results contrast the existing knowledge that institutional quality alone is decisive for the resource curse; the results show that the choice of ownership structure in the oil sector plays an important role in determining whether oil-exporting countries benefit from their oil wealth or suffer from the curse. The policy advice in this article is: retain state ownership and control if the institutions are strong, if the institutions are weak, transfer ownership to foreign oil companies.
Keywords: oil curse, ownership, economic growth, institutional quality
Working Papers 2015
2015-04: Discounted Stochastic Games with Voluntary Transfers
Susanne Goldlücke, University of Konstanz and Sebastian Kranz, Ulm University
This paper studies discounted stochastic games with perfect or imperfectpublic monitoring and the opportunity to conduct voluntary monetarytransfers. This generalization of repeated games with transfers is ideallysuited to study relational contracting in applications that allow for longterminvestments, and also allows to study collusive industry dynamics. We show that for all discount factors every public perfect equilibrium payoffcan be implemented with a simple class of equilibria that have a stationarystructure on the equilibrium path and optimal penal codes with astick and carrot structure. We develop algorithms that exactly compute orapproximate the set of equilibrium payoffs and find simple equilibria thatimplement these payoffs.
Keywords: dynamic games, relational contracting, monetary transfers, computation,imperfect public monitoring, public perfect equilibria
JEL-Codes: C73, C61, C63
2015-03: One Fits All? Explaining Support for Immigration Control in a Group Comparative Perspective
Claudia Diehl, University of Konstanz, Katrin Auspurg, Goethe University Frankfurt and Thomas Hinz, University of Konstanz
By using data from a factorial survey we analyze the role of economic and cultural threat in explaining support for immigration control in Switzerland. Threat is modeled more directly than in many previous studies: Economic threat is assumed to be high when migrants and natives have similar levels of education, and perceived cultural threat is assumed to be high when nationally pride natives are confronted with migrants unwilling to adapt culturally.
Furthermore, it is analyzed whether threat varies across immigrant groups that differ in size, aggregate skill level and cultural background. Results show that both economic and cultural threat play a role in explaining support for immigration control. In line with previous studies economic threat seems to be an issue for highly-skilled natives when they are confronted with large groups of migrants with similar skill levels to their own. Likewise, nationally pride natives seem threatened not only by culturally distant migrants, but also by large migrant groups who are not willing to adapt culturally while not being dissimilar enough to stay culturally apart. Further studies are necessary to examine whether these patterns generalize to other countries and immigration groups.
Keywords: Immigration, Switzerland, cultural threat, economic threat, migration related attitudes, factorial survey experiment
2015-02: The Changing Psychology of Culture in Germany: A Google Ngram Study
Nadja Younes, University of Konstanz and Ulf-Dietrich Reips, University of Konstanz
The paper provides evidence for the long-term affiliation between ecological and cultural changes in Germany, based on the assumptions derived from social change and human development theory by Greenfield (2009). Greenfield (2009, 2013) assumes that the increase in urbanization is associated with significant cultural changes of psychology. Whereas urbanization is linked to greater individualistic and materialistic awareness, rural environments are strongly associated with allegiance, prevalence of religion, and intense feelings of belonging and benevolence. Due to a similar rate of urbanization in Germany compared to the US, this study aims to replicate the results of Greenfield (2013) for Germany and to improve on the method. Results indicate that prognoses about the implications of an urbanizing population for the psychology of culture hold true. Individualistic values increased, whereas collectivistic values decreased. Further, a predictable reversal for the time during and after World War II is observed, reflecting Nazi propaganda and influence.
Keywords: Google Ngram, social change and human development theory, urbanization, internet science
2015-01: A Quantitative Model of Sovereign Debt, Bailouts and Conditionality
Fabian Fink, University of Konstanz and Almuth Scholl, University of Konstanz
In times of sovereign debt crises, International Financial Institutions provide temporary financial support contingent on the implementation of specific macroeconomic policies. This paper develops a model of sovereign debt and default with endogenous participation rates in bailout programs. Conditionality enters as a constraint on fiscal policy. In the model, the insurance character of bailouts generates incentives for debt accumulation. Quantitative results suggest that bailouts prevent sovereign defaults in the short-run but may come at a cost of a greater default probability in the long-run. Increasing the intensity of conditionality lowers the bailout participation rate and generates a hump-shaped pattern of sovereign default risk.
Keywords: sovereign debt, sovereign default risk, bailouts, conditionality, fiscal policy
JEL-Codes: E44, E62, F34