Toutes les versions de cet article : English , français
Contact : email@example.com
Current position :
Since Sept. 2016 : Teaching and Research Fellow (ATER), Université Paris 1 Panthéon-Sorbonne, affiliated with GATE, UMR 5824.
Research interests : International Macroeconomics and Finance, Asset Price Volatility, Financial Crises, Expectations formation, Macroeconomic Modeling and Applied Time-Series Econometrics.
Research Papers :
1. Learning, house prices and macro-financial linkages (advanced version available)
To account for the linkages between the credit market, the housing market and the real sector, which were strikingly evidenced by the subprime financial crisis, we develop a small-scale DSGE model in which agents form misspecified beliefs about future house prices but otherwise behave rationally given those beliefs. To explain the intertwined dynamics observed during the Great Moderation and the subsequent financial crisis, which are unlikely to result only from productivity shocks, we introduce loan-to-value ratio shocks and lenders’ preference shocks. Both macroeconomic and financial shocks prove able to generate endogenous booms and busts in house prices under learning. Because real-estate assets serve as collateral for household and entrepreneurial debt in the context of financial frictions, long-lasting excess volatility in house prices generated by the learning mechanism in response to shocks in turn affects the financial sector and thus propagates to the real sector. This powerful propagation mechanism significantly improves the ability of the model to simultaneously replicate the volatility in house prices and in business cycle variables in response to small shocks beginning in the mid 1980s in the US, and can help explain some features of the Great Recession.
2. Asset prices under recency-biased learning (Best Paper Award at the XVth RIEF Meetings, advanced version available)
To explain the boom period in the US stock market in the early 2000s and the subsequent bust beginning in 2008, this paper presents a consumption-based asset pricing model in which fluctuations in asset prices are persistently driven by time-varying expectations because of learning on the fundamental process. Agents estimate the actual parameter of the fundamental process as Bayesian learners, except that they weight the precision of recent observations more heavily relative to earlier ones (the so-called recency bias), due to cognitive limitations when processing sequential information over time. The stylized model proves able to replicate the boom-and-bust episode on the US S&P 500 stock market in the run-up to the recent global financial crisis and enables an assessment of the specific role of the recency bias for explaining the stock market excess volatility. Those results suggest that the impact of public information disclosure aimed at warning market participants against asset prices misalignments - which has become a new financial risk management tool for mitigating expectations-driven fluctuations - strongly depends on the ability of the regulator to pay full attention to past financial events.
3. Increase in home bias in the Eurozone debt crisis : the role of domestic shocks, with C. Cornand and C. Gimet, published in Economic Modelling, 2016, Vol.53, pp. 445-469.
One of the most striking consequences of the recent episode of sovereign debt market stress in the Eurozone has been the increase in the share of public debt held by the domestic sector in fragile economies. However, the causes and potential consequences of this increase were only given scarce attention in the literature on the Euro area sovereign debt crisis. In order to fill this gap, we first determine the shocks that impact the variation in the share of sovereign debt held at home in an SVAR model on a sample of Eurozone countries between 2002 and 2014, distinguishing between external and domestic shocks. Thanks to several alternative tests, we show that home bias in sovereign debt responds positively to country-specific fundamentals and expectation shocks but we find no evidence that the increase in home bias is destabilizing per se in the short-run. Second, a stylized theoretical model backed by the empirical results predicts that the consequences for sovereign debt crisis depend on the relative impact of domestic initial destabilizing shocks and increased home bias. The analysis suggests that an increase in home bias in times of sovereign debt stress, despite reflecting deteriorating fiscal conditions, may make default less likely.
4. Domestic creditors as last lenders in debt crises : a simple model with multiple equilibria, published in Economics Bulletin, 2015, Vol. 35 No. 4, pp. 2915-2928.
It is widely acknowledged that the ratios of public debt over GDP reached historically high levels in the Euro area during the recent sovereign debt crisis. More unnoticed however is the simultaneous increase in the share of government debt held by residents that has started in late 2008 in most fragile economies of the area. This paper develops a simple theoretical framework, in which multiple equilibria arise, to explain why exogenous increases in the debt level may cause this share to increase, due to distinct expected returns on domestic sovereign debt for domestic and foreign creditors in times of turmoil.