2 September 2016 - DNB workshop on Estimating and Interpreting Financial Cycles

Location: De Nederlandsche Bank

The concept of financial cycle plays an important role in the current debate on how to increase the resilience of the financial system and dampen its procyclicality. The financial cycle captures systematic patterns in the financial system that can have important macroeconomic consequences. In recent years, empirical research has documented time-series patterns in financial variables – mainly credit growth, leverage, and real estate prices – around financial crises. A main conclusion of this research is that on average, financial crises tend to occur close to the peaks of the medium-term cycles in credit, the credit-to-GDP ratio and residential property prices.

Central banks are increasingly turning to estimates of the financial cycle as a broad indicator, in real time, of whether risks to financial stability increase, remain stable, or decrease. This can be especially useful in times of persistently low interest rates, which could lead to the build-up of financial imbalances.

There is however no consensus yet on how to measure the financial cycle or how to interpret different patterns that have been found over time and across countries.

2 September 2016 - DNB workshop on Estimating and Interpreting Financial Cycles

To take stock of these challenges, DNB hosted a workshop on “Estimating and Interpreting Financial Cycles” on 2 September 2016. The workshop brought together leading experts from central banks, international organizations and academia, with a view to discussing insights, challenges and concrete implications for policy. Key speakers were Siem Jan Koopman (Vrije Universiteit Amsterdam), Oscar Jorda (San Francisco Fed) and Stijn Claessens (Federal Reserve Board).

As part of the workshop, participants used a common data set to produce their best estimates of the financial cycle. The results of this exercise helped gain a better understanding of how and why estimates of financial cycles may vary.

Several important conclusions emerged from the workshop. First, estimating financial cycles involves issues comparable to estimating business cycle. However, the task of distinguishing trends and cycles in the financial system is trickier compared to business cycles, given that there is no common definition of the financial cycle and that financial indicators are typically noisier than macroeconomic variables. It is therefore not surprising that estimates of the financial cycle can vary quite markedly across estimation methods. More work is required to understand the benefits of alternative empirical approaches.

This said, these different methods all reveal the existence of cycles in the financial sector that tend to be more ample and more protracted than the business cycle. How to best combine information from different financial variables remains a challenge.

Research presented at the workshop also shows that financial cycles differ quite markedly across countries and over time. This suggests that structural differences in the macroeconomic and financial sphere, as well as in the policy regime, may play a role in shaping the financial cycle. It also underscores the need to build a theoretical underpinning for the concept of the financial cycle.