Estimating the Financial Cycle Under Limited Data Availability: Alternative Methods

Introduction:The Great Financial Crisis of 2008 highlighted the importance of financial cycle fluctuations. While the regulatory response was to mandate higher bank capital requirements during the financial cycle upswing, academic research focussed on identifying the best performing early warning indicators to forecast financial cycle fluctuations that have proven to be often unrelated to business cycle changes. To safeguard the global financial system against the financial cycle fluctuations, Basel Committee of Banking Supervisors, based on first strands of empirical evidence, proposed the credit-to-GDP gap as the headline indicator tied to the countercyclical capital buffer. However, later research on this indicator identified certain concerns, among them subpar performance for economies with short available data series. Aim of the Study:To this end this study aims to analyse various financial cycle indicators from a unique perspective of their potential viability under limited historical data availability. Methods:For this purpose, a meta-study of existing research is carried out as well as an empirical study to compare performance of certain indicators for the sample of six countries in the Central, Eastern and South-Eastern European region, where long data series are not available. Main Findings:It was found that certain approaches, among them calculation of raw credit growth rate and application of Hamilton filter, can supplement or possibly even outperform the Basel credit-to-GDP gap indicator under limited data availability. Conclusion:Author concludes that for limited time series Basel credit-to-GDP gap can be potentially outperformed by other indicators and further research in this currently under-studied field is warranted. Originality of the Paper:By using various financial cycle indicators that already proven their early warning prediction powers from previous research, this study focusses on their potential viability under limited historical data availability. Respective findings might be appreciated for supplementing policy-makers’ toolkits as complementary indicators in cases where there is no available long time series for financial cycle estimation, for example, such as countries that entered market economies relatively late.