Cash Demand and Consumption Response to Unanticipated Monetary Policy Shock: Evidence from Turkey
Posted: 9 Feb 2017 Last revised: 5 Aug 2018
Date Written: June 25, 2017
We study the impact of an unanticipated interest rate policy shock on cash demand and consumption using an administrative panel dataset of financial transactions from Turkey. In response to the unanticipated policy announcement, consumers significantly reduce their monthly ATM withdrawal amount by 92 lira or 5.67 percent on average in the subsequent six months. The difference-in-difference estimate implies an even higher response among treated group relative to the counterfactual and the response is achieved mainly through fewer ATM visits. The differential effect on cash demand persists throughout the post-announcement period and is much stronger among consumers who are less liquidity-constrained and have stable income. Our difference-in-difference analysis also shows a negative consumption response to the unanticipated policy shock that is primarily concentrated in non-durable and discretionary spending. Both the cash demand and consumption response results are consistent with theory. The estimate of the interest rate elasticity, using either the monetary policy shock or classic money demand specification, are comparable with those in the literature. The welfare cost of inflation is moderate and varies across subsamples.
Keywords: Cash Demand, Interest Shock, Liquidity Constraint, Consumption, Big Data, Household Finance
JEL Classification: D12, D14, E43
Suggested Citation: Suggested Citation