The new era of financial regulation is marked by the importance of macroprudential policy. Many central banks and supervisory authorities have acquired additional responsibility for the stability of the financial sector as a whole and have been equipped with new instruments to maintain stability. The instruments at their disposal include inter alia the new set of Basel III capital buffers and often a range of borrower-based and liquidity instruments.

Against this backdrop, there is an increasing need to evaluate the effectiveness of various instruments in countering systemic risks. The literature on this topic has developed quickly and evolved along several paths. Structural models, including dynamic stochastic general equilibrium (DSGE) models, have been employed to investigate the transmission channels of new instruments. Time-series models or studies based on microdata and cross-sectional variation have supplied empirical assessments of the effects of higher bank capitalisation on liquidity.

Another important strand in the literature on the effectiveness of macroprudential policies is related to a seminal paper by Romer and Romer (2010). This and similar studies use narrative information on the exact timing and type of distinct policy events. Their advantage is that they pin down the effects of policies directly, rather than the impact of changes in the intermediary targets of these policies such as bank capital. Furthermore, they constitute a convenient platform to compare the effectiveness of different types of policies and the interactions between them.

The key challenge for narrative-based studies is the availability of data on relevant policy events. Constructing a dataset commonly involves substantial effort, starting with proposing a working definition of a macroprudential measure and looking closely at the regulatory history of the country in question, and, for cross-country studies, ensuring comparability of information. For this reason, a large share of narrative studies on macroprudential policies work with very few databases, with prominent examples being the databases developed by the International Monetary Fund (Lim et al., 2011) and the Bank for International Settlements (Kuttner and Shim, 2013).

This conference will bring together policymakers and researchers and invite them to discuss the possibilities and challenges of narrative-based studies on the effectiveness of macroprudential policies. It will take stock of this strand of the literature, evaluating to what degree it might inform ongoing policy discussions. The conference will also provide an opportunity to present for the first time the comprehensive Macroprudential Policies Evaluation Database (MaPPED), including information on macroprudential policy actions in all EU Member States from 1995 to 2015. This database has been developed in close cooperation with central banks and supervisory authorities in the EU over 2016 and 2017 in order to facilitate narrative-based studies on macroprudential policies in the region.