Ruben Hipp

Non-parametric structural identification via heteroskedasticity of unknown form

the latest version can be downloaded here,  A SWP version of this paper circulated under the name "On Causal Networks of Financial Firms". 

Abstract

We introduce a novel non-parametric approach to identify structural models via heteroskedasticity. The approach avoids parametric assumptions about the form of heteroskedasticity. Instead, by allowing all structural matrices to vary smoothly over time, we can identify parameters locally and globally by imposing different alteration rates of the time variation. In this regard, the key identification assumption is that the alteration of the structural matrix that describes connectedness is slower than the alteration of the matrix that describes the volatilities of structural shocks. Identification is achieved by a local approximation of the derivatives of the structural volatility matrix leading to the additionally required parameter restrictions. Based on a rescaling approach, we propose local-linear kernel estimators for the structural parameters. In a Monte Carlo study, we illustrate the identification approach by showing the estimation performance for the structural parameters in comparison to other natural competitors.

From Stress to Strategy: How Banks Balance the Scales

joint with Javier Ojea-Ferreiro, draft available soon

Abstract

This paper introduces a partial equilibrium stress-testing model that captures the strategic portfolio responses of heterogeneous banks. Banks in this model maximize utility functions that balance expected returns against the proximity to regulatory constraints, with endogenous price effects arising from their collective actions. A Nash equilibrium ensures internally consistent strategies across institutions. Applied to Canada’s Big Six banks, we showcase three usages of the model: scenario-based policy analysis, counterfactual analysis and reverse stress testing. In the latter, we identify multiple narratives which can cause systemic stress, detailing the vulnerability of the banking sector towards different horizons and economic conditions. In doing so, we provide new insights into which macro economic narratives are most likely to cause stress in the banking sector. Interestingly, we find that a strong appreciation of the Canadian dollar with or due to Oil price increases, may be enough to add stress to the financial system in Canada.

Estimating Large-Dimensional Connectedness Tables: The Great Moderation through the Lens of Sectoral Spillovers

joint with Felix Brunner, 

published QE (2023), previously appeared as SWP

Abstract

We estimate sectoral spillovers around the Great Moderation with the help of forecast error variance decomposition tables. Obtaining such tables in high dimensions is challenging because they are functions of the estimated vector autoregressive coefficients and the residual covariance matrix. In a simulation study, we compare various regularization methods on both and conduct a comprehensive analysis of their performance. We show that standard estimators of large connectedness tables lead to biased results and high estimation uncertainty, both of which are mitigated by regularization. To explore possible causes for the Great Moderation, we apply a cross-validated estimator on sectoral spillovers of industrial production in the US from 1972 to 2019. We find that the spillover network has considerably weakened, which hints at structural change, for example, through improved inventory management, as a critical explanation for the Great Moderation.