Developments during the financial crisis of 2008 and 2009 have, at least temporarily, changed the way monetary policy is implemented in many countries. Before the crisis, most countries implemented policy through some form of a corridor system. The key feature of a corridor system is that it is costly for banks to use the central bank's standing facilities for deposits and lending. Provided that this cost is sufficiently large, the system will encourage interbank trading. However, for given values of the central bank's rate, the system also implies a tight link between the quantity of central bank reserves and the overnight interest rate: The central bank cannot control the two independently. In contrast, a floor system has almost the opposite features: there are limited incentives for interbank trading, but the central bank can control the overnight rate and the level of excess reserve balances independently. During the crisis, the need to supply more reserve balances meant that many central banks found it necessary to move towards a floor system, in order to break the link between reserves and the overnight interest rate.