Working Papers:
Censored Fractional Response Model: Estimating Heterogeneous Risk Aversion of European Households
(IWH Discussion Papers , No. 11, 2015, and was presented at Econometric Society European Winter Meeting 2014 in Madrid)
This paper estimates relative risk aversion using the observed share of risky assets and characteristics of households. Given that the risky share is a fractional response variable belonging to unity interval, this paper proposes a censored fractional response estimation method using extremal quantiles to approximate the censoring thresholds. Considering that participation in risky asset markets is costly, I estimate both the heterogeneous relative risk aversion and participation cost using a working sample that includes both risky asset holders and non risky asset holders by treating the zero risky share as the result of heterogeneous self censoring. Estimation results show lower participation costs and higher relative risk aversion than what was previously estimated. The estimated median relative risk aversions of eight European countries range from 4.6 to 13.6.
(IWH Discussion Papers , No. 11, 2015, and was presented at Econometric Society European Winter Meeting 2014 in Madrid)
This paper estimates relative risk aversion using the observed share of risky assets and characteristics of households. Given that the risky share is a fractional response variable belonging to unity interval, this paper proposes a censored fractional response estimation method using extremal quantiles to approximate the censoring thresholds. Considering that participation in risky asset markets is costly, I estimate both the heterogeneous relative risk aversion and participation cost using a working sample that includes both risky asset holders and non risky asset holders by treating the zero risky share as the result of heterogeneous self censoring. Estimation results show lower participation costs and higher relative risk aversion than what was previously estimated. The estimated median relative risk aversions of eight European countries range from 4.6 to 13.6.
College Choice Allocation Mechanisms:Structural Estimates and Counterfactuals
(with Thierry Magnac and Jose-Raimundo Carvalho, R&R at Quantitative Economics; available as IZA Discussion Paper No. 8550)
We evaluate a simple allocation mechanism of students to majors at college entry that was commonly used in universities in Brazil in the 1990s and 2000s. Students first chose a single major and then took exams that select them in or out of the chosen major. The literature analyzing student placement, points out that this decentralized mechanism is not stable and is not strategy-proof. This means that some pairs of major & students can be made better off and that students tend to disguise their preferences using such a mechanism. We build up a model of performance and school choices in which expectations are carefully specified and we estimate it using cross-section data reporting choices between two medical schools and grade performances at the entry exams. Given those estimates, we evaluate changes in selection and students' expected utilities when other mechanisms are implemented. Results highlight the importance of strategic motives and re-distributive effects of changes of the allocation mechanisms.
(with Thierry Magnac and Jose-Raimundo Carvalho, R&R at Quantitative Economics; available as IZA Discussion Paper No. 8550)
We evaluate a simple allocation mechanism of students to majors at college entry that was commonly used in universities in Brazil in the 1990s and 2000s. Students first chose a single major and then took exams that select them in or out of the chosen major. The literature analyzing student placement, points out that this decentralized mechanism is not stable and is not strategy-proof. This means that some pairs of major & students can be made better off and that students tend to disguise their preferences using such a mechanism. We build up a model of performance and school choices in which expectations are carefully specified and we estimate it using cross-section data reporting choices between two medical schools and grade performances at the entry exams. Given those estimates, we evaluate changes in selection and students' expected utilities when other mechanisms are implemented. Results highlight the importance of strategic motives and re-distributive effects of changes of the allocation mechanisms.
Housing Consumption and Housing Market Regulations in the Post-Crisis Europe
With Antonios Mavropolous, preliminary working paper
In this paper, we use the panel of the first two waves of the Household Finance and Consumption Survey by the European Central Bank to study housing demand of European households and evaluate potential housing market regulations in the post-crisis era. We find that around 60 percent of both renters and owners adjust their housing consumption at the intensive margin between 2010 and 2014, and the proportions of households trading-up and trading-down are almost the same. At the extensive margin, we find that housing preference shocks dictate the tenure transition; while the background risks do not have a significant impact. We structurally estimate a life-cycle model using the simulated method of moments and then evaluate the effect of LTV/LTI regulations via counterfactual simulations. We find that those regulations are unlikely to be effective in Europe.
With Antonios Mavropolous, preliminary working paper
In this paper, we use the panel of the first two waves of the Household Finance and Consumption Survey by the European Central Bank to study housing demand of European households and evaluate potential housing market regulations in the post-crisis era. We find that around 60 percent of both renters and owners adjust their housing consumption at the intensive margin between 2010 and 2014, and the proportions of households trading-up and trading-down are almost the same. At the extensive margin, we find that housing preference shocks dictate the tenure transition; while the background risks do not have a significant impact. We structurally estimate a life-cycle model using the simulated method of moments and then evaluate the effect of LTV/LTI regulations via counterfactual simulations. We find that those regulations are unlikely to be effective in Europe.
The Liquidity Premium of Safe Assets: The Role of Government Debt Supply
(Under review at Journal of International Money and Finance; available as IWH Discussion Papers, NO. 11, 2017)
This paper studies the impact of government debt supply on the liquidity premium, as measured by the yield spread between public and private safe assets. I test, at a quarterly frequency, how the liquidity premium of Treasury bills against i) Aaa-rated corporate bonds and ii) commercial paper responds to government debt supply changes. The response is significant in each case – even after controlling for the opportunity cost of money – but heterogeneous: negative for Aaa-rated corporate bonds and positive for commercial paper. This points to different degrees of substitutability with government debt across apparently similar private safe assets.
(Under review at Journal of International Money and Finance; available as IWH Discussion Papers, NO. 11, 2017)
This paper studies the impact of government debt supply on the liquidity premium, as measured by the yield spread between public and private safe assets. I test, at a quarterly frequency, how the liquidity premium of Treasury bills against i) Aaa-rated corporate bonds and ii) commercial paper responds to government debt supply changes. The response is significant in each case – even after controlling for the opportunity cost of money – but heterogeneous: negative for Aaa-rated corporate bonds and positive for commercial paper. This points to different degrees of substitutability with government debt across apparently similar private safe assets.
The Effect of Wealth Fluctuation on Asset Holding: Micro-Evidence from Dutch Household Survey
(Under major revision, possibly a new title)
This paper uses the DNB Household Survey from Netherlands to investigate household portfolio adjustment in response to financial wealth fluctuation. I study both the intensive and extensive margin of the portfolio adjustment. I use the habit formation model to test whether household risky share adjusts when liquid financial wealth changes. The results show that changes in household risky asset holding has negative correlation with the changes in financial wealth, which rejects the decreasing relative risk aversion prediction of habit formation model. Meanwhile increasing relative risk aversion is also rejected when the sampling periods extend to 20 years. Constant relative risk aversion seems to best describe the risk preference of Dutch households.
(Under major revision, possibly a new title)
This paper uses the DNB Household Survey from Netherlands to investigate household portfolio adjustment in response to financial wealth fluctuation. I study both the intensive and extensive margin of the portfolio adjustment. I use the habit formation model to test whether household risky share adjusts when liquid financial wealth changes. The results show that changes in household risky asset holding has negative correlation with the changes in financial wealth, which rejects the decreasing relative risk aversion prediction of habit formation model. Meanwhile increasing relative risk aversion is also rejected when the sampling periods extend to 20 years. Constant relative risk aversion seems to best describe the risk preference of Dutch households.
Research in Progress:
The Value of Public Transport in House Pricing: Evidence from Property Listing Websites
with Stefano Colonnello
This paper builds on the housing price data we collect under the IWH Housing Index Initiative and studies the role of access to public transport in housing pricing. House prices vary for different reasons, most of which can be accounted for by its physical features such as the size of the house, the number of bedrooms and the location of the house. However, there is also many intangible aspects of the housing that are critical to the valuation of housing, such as safety of the neighborhood, availability of high-quality schools and the air and noise pollution levels. In this paper, we look at one specific feature of the intangibles --- access to public transport. We investigate whether such intangible element is priced at the market or it remains as positive externalities. We first propose a theoretical framework based on standard Campbell-Shiller asset pricing decomposition to explain how the public transport access enters the house pricing. Empirically, we study the effect of in the proximity of public transportation on the housing listing prices (both rental and sales) in Metropolitan areas in Germany. We use the voluntary description of the transportation condition provided by the listing agencies or owners as the measure of proximity to public transport. In addition to qualitative analysis, we take advantage of the large sample size and adopt the econometric matching method to estimate the price difference attributed to public transport access.
The preview of the housing list prices in Germany
with Stefano Colonnello
This paper builds on the housing price data we collect under the IWH Housing Index Initiative and studies the role of access to public transport in housing pricing. House prices vary for different reasons, most of which can be accounted for by its physical features such as the size of the house, the number of bedrooms and the location of the house. However, there is also many intangible aspects of the housing that are critical to the valuation of housing, such as safety of the neighborhood, availability of high-quality schools and the air and noise pollution levels. In this paper, we look at one specific feature of the intangibles --- access to public transport. We investigate whether such intangible element is priced at the market or it remains as positive externalities. We first propose a theoretical framework based on standard Campbell-Shiller asset pricing decomposition to explain how the public transport access enters the house pricing. Empirically, we study the effect of in the proximity of public transportation on the housing listing prices (both rental and sales) in Metropolitan areas in Germany. We use the voluntary description of the transportation condition provided by the listing agencies or owners as the measure of proximity to public transport. In addition to qualitative analysis, we take advantage of the large sample size and adopt the econometric matching method to estimate the price difference attributed to public transport access.
The preview of the housing list prices in Germany
The delivery risk in public procurement of long-term projects
with Sinem Hidir
We study competitive public procurement of long-term projects, whose delivery is not certain either in completion time nor quality. We show that if public procurement ignores the delivery risk of long term projects in auctions, they create incentive for private bidders to bid lower than their optimal risk-cost trade off to win the project and then shirk in the effort of delivering the projects. The costly foreclosure of public-private partnership is likely to force the public authority into bailing out the shirking and inefficient private partner, which leads to big loss of public fund. We propose that delaying pay out for the projects is efficiency improving and prevents partly the adverse selection and mitigates the moral hazard problem. We plan to test our theory using the Italian public procurement record data and UK national archive data on public procurement.
with Sinem Hidir
We study competitive public procurement of long-term projects, whose delivery is not certain either in completion time nor quality. We show that if public procurement ignores the delivery risk of long term projects in auctions, they create incentive for private bidders to bid lower than their optimal risk-cost trade off to win the project and then shirk in the effort of delivering the projects. The costly foreclosure of public-private partnership is likely to force the public authority into bailing out the shirking and inefficient private partner, which leads to big loss of public fund. We propose that delaying pay out for the projects is efficiency improving and prevents partly the adverse selection and mitigates the moral hazard problem. We plan to test our theory using the Italian public procurement record data and UK national archive data on public procurement.
Exogenous macroeconomic and macro-prudential policy shocks on regional housing market
with Michael Koetter and Antonios Mavropoulos
This project encompasses a few the papers that investigate policy impacts on regional housing markets to better understand housing market demand, especially on the demand elasticity and speculative housing demand. The first paper of this project is to investigate the housing market response to Brexit in financial centers like Frankfurt and Munich. Our goal in this paper is to exploit Brexit as an exogenous demand housing shock for Frankfurt and investigate its impact on house prices in its metropolitan area. Brexit would induce a positive shock in Frankfurt housing market due to the reallocation of labor capital from London to Frankfurt. We would like to find out whether such demand shock has a significant impact on the metropolitan area of Frankfurt where the financial center is located. Moreover, since the financial workers often have a high income, we want to see whether there is substantial re-distributive effect in the broader areas of Frankfurt. For instance, whether the top earners drive out lower-income workers to the outskirts of Frankfurt or nearby cities. The project is currently collecting data via two channels: 1) we conduct web scrapping to collect data through major online housing listing websites; 2) we also try to contact the online housing listing platforms to seek access of historical listing data.
with Michael Koetter and Antonios Mavropoulos
This project encompasses a few the papers that investigate policy impacts on regional housing markets to better understand housing market demand, especially on the demand elasticity and speculative housing demand. The first paper of this project is to investigate the housing market response to Brexit in financial centers like Frankfurt and Munich. Our goal in this paper is to exploit Brexit as an exogenous demand housing shock for Frankfurt and investigate its impact on house prices in its metropolitan area. Brexit would induce a positive shock in Frankfurt housing market due to the reallocation of labor capital from London to Frankfurt. We would like to find out whether such demand shock has a significant impact on the metropolitan area of Frankfurt where the financial center is located. Moreover, since the financial workers often have a high income, we want to see whether there is substantial re-distributive effect in the broader areas of Frankfurt. For instance, whether the top earners drive out lower-income workers to the outskirts of Frankfurt or nearby cities. The project is currently collecting data via two channels: 1) we conduct web scrapping to collect data through major online housing listing websites; 2) we also try to contact the online housing listing platforms to seek access of historical listing data.
Household Financial portfolio adjustment after financial crisis
with Reint Gropp
Temporary fluctuation in return or experience of the financial crisis should not induce significant household portfolio re-balancing for a rational household with stable risk preference and stock market belief. However, recently literature find that those critical parameters vary over time and react to the macroeconomic environment. We intend to use the existing household finance surveys to investigate whether and how do households adjust their portfolio when low interest rates of safe assets persist. This project is closely related to the empirical exercise conducted in Calvet and Sodini (2009), which investigate the dynamics of individual portfolios in a unique dataset containing the dis-aggregated wealth of all households in Sweden from 1999 to 2002. They study the propensity of re-balancing among the households holding risky assets, the exit and entry to risky financial markets of all the households, and the influence of past stock market experience on portfolio adjustment. We intend to conduct a similar exercise with a novel data and macroeconomic environment to contribute to the literature with new empirical evidence across European countries. Additionally, our project has the potential to provide new insights on the country difference of household portfolio choices. To the best of our knowledge, this project will be the first to look at the timing of portfolio adjustment after the arrival of a macroeconomic shock. In the post-crisis era, the supposedly higher risk macroeconomic environment would push households to reduce the overall risk exposure of their portfolio. However, the low interest rate on safe assets would also push households to shift to high-risk investment if they expect a certain level of return from their liquid asset holding. Similar to the extensive margin, it is not crystal clear as to which effect triumphs the other. We intend to contribute to this discussion with new household survey data and structural estimation of the life-cycle model that dictates the investment decisions.
with Reint Gropp
Temporary fluctuation in return or experience of the financial crisis should not induce significant household portfolio re-balancing for a rational household with stable risk preference and stock market belief. However, recently literature find that those critical parameters vary over time and react to the macroeconomic environment. We intend to use the existing household finance surveys to investigate whether and how do households adjust their portfolio when low interest rates of safe assets persist. This project is closely related to the empirical exercise conducted in Calvet and Sodini (2009), which investigate the dynamics of individual portfolios in a unique dataset containing the dis-aggregated wealth of all households in Sweden from 1999 to 2002. They study the propensity of re-balancing among the households holding risky assets, the exit and entry to risky financial markets of all the households, and the influence of past stock market experience on portfolio adjustment. We intend to conduct a similar exercise with a novel data and macroeconomic environment to contribute to the literature with new empirical evidence across European countries. Additionally, our project has the potential to provide new insights on the country difference of household portfolio choices. To the best of our knowledge, this project will be the first to look at the timing of portfolio adjustment after the arrival of a macroeconomic shock. In the post-crisis era, the supposedly higher risk macroeconomic environment would push households to reduce the overall risk exposure of their portfolio. However, the low interest rate on safe assets would also push households to shift to high-risk investment if they expect a certain level of return from their liquid asset holding. Similar to the extensive margin, it is not crystal clear as to which effect triumphs the other. We intend to contribute to this discussion with new household survey data and structural estimation of the life-cycle model that dictates the investment decisions.