Research

Publications 

Abstract: We introduce a new real-time method to measure business opening and closure rates by relying on Google Places, the data behind the Google Maps platform. We collect data on establishments of customer-facing industries (food, retail, accommodation) and provide evidence that the opening and closure rates reflect well the temporary closures and reopening during the pandemic. We find that the operational or closed status of establishments is correlated with business reviews: fewer reviews are associated with impending business exit, and more reviews are associated with expanding businesses posting new job vacancies.

Abstract: We present a new estimation of business opening and closure rates using data from Google Places—the data set behind the Google Maps service. Our algorithm, through a bisection routine, counts the appearance and disappearance of “pins” that represent unique businesses. As a proof of concept, we compute business opening and closure rates for the city of Ottawa during the reopening phase of the COVID-19 pandemic in mid-2021. The lifting of restrictions coincides with a wave of re-entry of temporarily closed businesses, suggesting that government support may have facilitated the survival of hibernating businesses. Our entry estimates are validated by a survey of new businesses. This methodology allows policymakers to monitor business dynamics in quasi-real-time during rapidly unfolding crises.

Media coverage: [VOX EU column] [Bank Underground]

Abstract: We develop a daily composite index of financial stress for the United Kingdom over 50 years, the UKFSI. The index includes market stress indicators based on their incremental information to capture financial crises. During the COVID-19 crisis, financial stress peaks but remains less severe than the Global Financial Crisis. The UKFSI is used in a threshold vector autoregression to differentiate the economic dynamics between tranquil and stressful periods. We highlight the importance of nonlinearities that amplify shocks. But we find no evidence of financial shocks contributing to the COVID-19 crisis, possibly reflecting effective policy interventions.

[Link to United Kingdom Financial Stress Index (UKFSI) data]

Abstract: We use logit and Markov switching models to assess, in (pseudo-)real-time, the ability of 27 indicators to predict systemic financial crises in the European Union. Before the global financial crisis (GFC), some models provided early warning signals, but it is unclear whether a specific model would have been favored over other candidate models providing contradictory evidence. Only after the GFC do debt service ratios, credit-to-GDP gaps as well as house price-to-income and house price-to-rent ratios appear as robust early warning indicators. Our results highlight that the predictive ability of indicators may change due to new risk factors or policy actions. 

Abstract: I construct a new composite measure of systemic financial market stress for Canada. Compared with existing measures, it better captures the 1990 housing market correction and more accurately reflects the absence of diversification opportunities during systemic events. The index can be used for monitoring. For instance, it reached a peak during the COVID-19 pandemic second only to the 2008 global financial crisis. The index can also be used to introduce non-linear macrofinancial dynamics in empirical macroeconomic models of the Canadian economy. Macroeconomic conditions are shown to deteriorate significantly when the Canadian financial stress index is above its 90th percentile.

[Link to Canadian Financial Stress Index (CFSI) data]

Abstract : We integrate two different stochastic models to analyse the joint dynamics of the financial and real cycles, captured by the financial stress index (FSI) and the industrial production index (IPI). The joint dynamics of the FSI and IPI exhibit stochasticity, mean reversion, seasonality, and occasional jumps. All parameters are modulated by a discrete-time hidden Markov chain that switches between economic regimes. Through change of reference probability technique, adaptive multivariate filters are derived which in turn provides online optimal parameter estimates. We apply our new framework to Canadian data and extract real time early-warning signals to detect financial crises. 

Abstract : This paper introduces a new methodology to date systemic financial stress events in a transparent, objective and reproducible way. The financial cycle is captured by a monthly Country-Level Index of Financial Stress (CLIFS). Based on two Markov-switching and one threshold vector autoregressive model, information from the CLIFS and industrial production are combined to identify those episodes of financial market stress that are associated with a substantial negative impact on the real economy. By applying this framework to 27 European Union countries, the paper is a first attempt to provide a chronology of systemic financial stress episodes as a complement to the expert-detected events that are currently available.

                      [Link to the monthly updates of the CLIFS dataset]

                      [Link to other related dataset and supplementary material]

                      Media coverage: [VOX EU column] [BSI column (French)]

                      ECB Financial Stability Review: [November 2015] [May 2016]

Abstract : This paper investigates the lending pattern of state-owned banks over the business cycle. I take the endogeneity of public banking into account by including records on both privatizations and nationalizations during banking crises. I find that public bank lending is (i) significantly less cyclical except for low-income countries, (ii) asymmetric along the business cycle, (iii) heterogeneous across stages of economic development, and (iv) related to banks’ vulnerability on their funding side. Public banks reduce their lending less during economic downturns, but their ability to absorb negative shocks marginally decreases as the size of the shock increases.

Working Papers

Abstract: We quantify the distributional impacts of climate change physical risks on employment, using shocks derived from narrative evidences of monthly natural disasters collected across 10 Canadian provinces over 40 years. We use a novel panel quantile local projection approach, which takes into account that natural disasters are tail events, localized in space and time, and hard to predicted in a given month-province. We find that the odds of a high provincial employment rate decreases on impact, but the effect tends to become positive as the economy recovers from a disaster. We also find substantial heterogeneity across different sectors, employment status and disaster types. 

Abstract: We incorporate quantile regressions into a structural vector autoregression model to empirically assess how monetary and fiscal policy influence risks around future GDP growth. Using a panel of six developed countries, we find that both policy instruments affect the location of the distribution of future GDP growth, whereas fiscal shocks also impact the shape of the distribution. Fiscal stimulus generates upside risk, paving the path to a faster recovery, especially when the policy rate is constrained by the zero lower bound (ZLB). Unconventional monetary policy during ZLB episodes has a comparable effect on future GDP growth as conventional monetary policy.

Abstract: We propose a novel framework to analyze how policy-makers can manage risks to the median projection and risks specific to the tail of gross domestic product (GDP) growth. By combining a quantile regression of GDP growth with a vector autoregression, we show that monetary and macroprudential policy shocks can reduce credit growth and thus GDP tail risk. So policymakers concerned about GDP tail risk would choose a tighter policy stance at the expense of macroeconomic stability. Using Canadian data, we show how our framework can add tail event information to projection models that ignore them and give policy-makers a tool to communicate the trade-offs they face. 



Policy articles and non-peer reviewed

Abstract : This paper presents a composite indicator of Canadian financial system vulnerabilities—the Vulnerabilities Barometer. It aims to complement the Bank of Canada’s vulnerabilities assessment by adding a quantitative and synthesized perspective to the more granular (distributional) analysis presented in the Financial System Review. The Vulnerabilities Barometer for Canada is above the level reached in 2007. The current state is driven by housing market vulnerabilities and elevated household indebtedness. The oil price shock contributed to the recent increase in vulnerabilities, though this risk factor has eased since the end of 2016. When assessed across countries, the Vulnerabilities Barometer sends earlier and better signals of future stress episodes than its components taken individually, or than the credit-to-GDP gap. It is also consistent with the narrative of stressful episodes for peer countries.

Abstract: Natural disasters occur more often than before, potentially exposing households to financial distress. We study the intersection between household financial vulnerabilities and severe weather events.

Abstract: Monetary policy decisions need to consider all potential outcomes, not just the most likely path for the economy. This is especially true in the presence of elevated financial system vulnerabilities, which lead to increased downside risks for future growth. In a novel risk-management framework, we decompose the outlook for the distribution of future gross domestic product (GDP) growth into macroeconomic and financial stability risks. When analyzing the efficacy of policy tools, we find that macroprudential tightening is substantially more effective than monetary policy at reducing downside risks to future GDP growth.

Bank of Canada Monetary Policy Report: [October 2018, Box 5, pp 26]

Bank of Canada Financial System Review: [June 2018, Box 2, pp 19-20]

Cited in a speech by Governor Poloz: [May 2018]

Abstract: We use a suite of risk-assessment models to examine the possible impact of a hypothetical house price correction, centred in the Toronto and Vancouver areas. We also assume financial stress significantly amplifies the macroeconomic impact of the house price decline. The rates of arrears rise for households and businesses, which puts some pressure on banks. But the large banks remain resilient through the risk scenario, supported by their international diversification and their ability to replenish capital with retained earnings. As with any simulation exercise, the results are subject to significant uncertainty and depend on the specifics of the scenario being considered.

Abstract: When financial system vulnerabilities are elevated, they can give rise to asymmetric risks to the economic outlook. To illustrate this, I consider the economic outlook presented in the Bank of Canada’s October 2017 Monetary Policy Report in the context of two key financial system vulnerabilities: high levels of household indebtedness and housing market imbalances. Uncertainty on the profile of consumption by indebted households—and, therefore, risks to growth in gross domestic product (GDP)—arises from higher interest rates and from recent changes to the Office of the Superintendent of Financial Institutions’ B-20 mortgage underwriting guideline. I use non-linear Bayesian techniques to capture the potential amplification of negative shocks in a vulnerable environment. I find that the materialization of larger-than-expected impacts on consumption from higher interest rates and/or the tighter mortgage qualifying criteria would imply asymmetric risks to GDP growth.

Abstract : Over the past several years, the Bank for International Settlements has noted that Canada’s credit-to-GDP gap has widened and is above thresholds indicating future banking stress. In this note we take a closer look at the subcomponents of credit and find that (i) excluding non-financial government enterprises narrows the credit-to-GDP gap to well below worrisome thresholds and (ii) excluding borrowings between interrelated corporations (i.e., focusing only on borrowing through banks or financial markets) significantly decreases the level of the credit to GDP ratio and modestly widens the gap. We also review the literature and discuss the benefits and limitations of using the credit-to-GDP gap as a measure of vulnerabilities in the Canadian financial system.

Abstract : This paper presents the main features of macroprudential policy with a focus on the French case. We first recall the ultimate objective of this policy, which is to prevent and to mitigate systemic risk, i.e. the risk of “widespread disruptions to the provision of financial services that have serious consequences for the real economy”(CGFS, 2012). We put forward two goals to achieve this ultimate objective, namely (i) increasing the resilience of the financial sector and (ii) leaning against the financial cycle. Then, in the context of the ongoing reflections on the organisation of macroprudential policy at the national and European level, we analyse the macroprudential institutional framework recently adopted in France. We discuss the instruments available for macroprudential authorities in the light of the two main goals of macroprudential policy. Drawing on theoretical considerations and past experience, we favour a macroprudential toolkit broadly consistent with the European CRD IV/CRR package. Finally, we emphasise the need for macroprudential authorities to be able to monitor and detect systemic risk. To this end, several indicators and their reliability are analysed.

Unpublished manuscripts

Abstract : Production efficiency and financial stability do not necessarily go hand in hand. With heterogeneity in banks’ abilities to screen borrowers, the market for loans becomes segmented and a self-competition mechanism arises. When heterogeneity increases, the intensive and extensive margins have opposite effects. Bank informational rents unambiguously decrease welfare and distort effort incentives. But the bank most efficient at screening expands its market share by competing against itself to offer effort-inducing contracts, which decreases the share of non-performing loans. A macroprudential authority acting alone reinforces this tension. Optimality is restored by targeting lending policies toward borrowers with intermediate abilities.

Abstract : This paper proposes a new micro-founded measure to quantify the aggregate capitalisation of banking sectors taking into account both market discipline and regulatory constraints. It allows studying the connection between micro capital shortfalls from an implicit bank specific capital target and macro impacts of capital shortages on aggregate lending. (i) Our quantitative country-wide index of bank capitalisation is consistent with the qualitative reports of the ECB Bank Lending Survey. (ii) This index correlates with future fluctuations in aggregate lending, especially when a banking system is under-capitalised. (iii) The adjustment of capital constrained banks mostly impact loans to domestic non-financial agents. Thus our measure suggests that (a) countercyclical capital requirements may be less effective if market constraints are more important, and (b) slow moving balance sheet variables can help detect vulnerabilities and reversals in the lending cycle.

Tutorial

Abstract : The Bankscope dataset is a popular source of bank balance sheet informations among banking economists, which covers the last 20 years for more than 30000 worldwide banks. This technical paper intends to provide the critical issues one has to keep in mind as well as the basic arrangements to be undertaken if one intends to use this dataset. To that extent, we propose some straightforward ways to deal with data comparability, consolidation, duplication of assets or mergers, and provide Stata codes to deal with it.