Measuring financial cycles in a model-based analysis: Empirical evidence for the United States and the euro area

Abstract

Standard-Nutzungsbedingungen:Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden.Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen.Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in AbstractWe adopt an unobserved components time series model to extract financial cycles for the United States and the five largest euro area countries over the period 1970 to 2014. We find that credit, the credit-to-GDP ratio and house prices have medium-term cycles which share a few common statistical properties. We show that financial cycles are longer and more ample than business cycles, and that their length and amplitude vary over time and across countries.

Cyclicality in lending activity of Euro area in pre- and post- 2008 crisis: a local-adaptive-based testing of wavelets

“…Contrary to this group of researchers, another group proved that in a certain small group of countries (Germany, the Czech Republic, Hungary or the Netherlands), there are shorter financial cycles which are close to the business cycle frequencies (Galati et al, 2016;Gonzalez, Lima, & Marinho, 2015;Rünstler & Vlekke, 2018). The third group of researchers make their findings more general (Galati et al, 2016;Hiebert, Klaus, Peltonen, Schüler, & Welz, 2014; as they see the variation of the financial cycles' length at the country level which reflects heterogeneity between countries. Hiebert et al (2014) point out the facts that financial cycles tend to differ from the business cycle counterparts, and that the identified length of financial cycles differs according to the definition of financial cycles which is given by the used methodology.…”

“…The methodological approaches started with (i) time domain analysis of turning points identification (Claessens et al, 2012) and was followed by (ii) detrending via frequency-based filtering (Drehmann et al, 2012;Aikman et al, 2014). Several authors use more sophisticated models such as unobserved component models or structural models (Galati et al, 2016;Gonzalez et al, 2015). The next step was (iii) the application of frequency domain methods which allows the identification of spectral components, i.e.…”

“…This insufficiency was partially solved by applying detrending methods such as frequency-based filters (high-pass or band-pass filters) (Drehmann et al, 2012) which allow the selection of pre-defined frequency range from the data, or multivariate model-based filters (Galati et al, 2016;. Some authors (Drehmann et al, 2012;Galati et al, 2016;Hiebert et al, 2014) combine both filtering and dating approaches to improve the achieved results and bring more robust conclusions. Gonzalez et al (2015) use alternative methods, i.e.…”

“…Both approaches (detrending/ filtering and dating) fight the problems concerning statistical character, e.g. resulting from assumptions for dating methods application (sensitivity for trend extraction), or expected frequency range for the financial cycles filtering (no consensus for frequency range of financial cycles), or application of filtering methods themselves (ideal filter approximation, edge effect problem) (Drehmann, 2012;Galati et al, 2016;Gonzalez et al, 2015;Hiebert et al, 2014;. Even if the frequency techniques highlight the cyclical behaviour of data, they, unfortunately, are not able to describe the temporal character of identified cyclical behaviour.…”

“…we use a quantity-based measure and not a price-based measure of the cycle. As proved by relevant empirical studies, the chosen methodology is applicable to our date range (see Aloui et al, 2016;Altar et al, 2017;Berdiev & Chang, 2015;Fidrmuc et al, 2014;Ftiti et al, 2014;Galati et al, 2016or Kunovac et al, 2018. From the first overview of an input time series in Figure 2a-b, we can see that the time series of Households and Corporates contain a long-term trend which goes through visible expansion and recession phases.…”