Publications - Working Papers IER SAS

We study the effect of consumer sentiment on bank lending in a panel of eleven European economies that have experienced significant credit expansion in recent decades. Using standardized surveybased consumer data, we distinguish between the sentiment about i) current and ii) future (expected) economic conditions. Nonlinear machine-learning techniques allow us to improve the identification of exogenous sentiment shocks by using a large set of macro-financial variables. The component of sentiment that is unexplained by economic conditions reflects exogenous (or irrational) sentiment shocks. Using a local projections approach, we show that positive sentiment shocks lead to both short- and long-term increases in housing loans, with little effect on consumer loans. We also find that shocks to sentiment related to future economic conditions have a stronger effect on bank lending than shocks to present sentiment do. Shocks to sentiment explain up to 10% of housing loan growth over the next two years following the shock and are most pronounced under persistently loose monetary policy. Finally, we also show that sentiment has implications for financial stability, as positive sentiment shocks increase the vulnerability of the banking system.