The limits of monetary policy .etfraTtat Eit1.7n I ?Ave, "N.) However, things get somewhat mossy when you think about the practicalities. So far, negative interest rates are only charged on balances of commercial banks with the central banks. Commercial banks have been reluctant to pass this cost on to their clients, so most of the private sector, including practically all individual savings accounts, is not charged. This, in turn, means two things. First, households are shielded, so lower interest rates will not have an impact on household behavior (encouraging current consumption, say, by discouraging saving, or prompting households to purchase riskier assets). Second, bank profitability will suffer. 2. What's the evidence so far on the impact on banks? In Sweden, Denmark and Switzerland banks have coped reasonably well with negative interest rates, in part because their domestic banking markets are quite concentrated." This allowed the top two or three key players to make up for the shortfall by pushing up profits on other products. For example, Swedish banks have been able to protect their net interest margin by increasing mortgage loan rates to offset charges on deposit. The problem is that, first, this means that the NIRP results in tighter, rather than looser financial conditions. Second, it would not work in other, less concentrated markets. And third, and perhaps most troubling for the ECB, it means NIRP will have a differential impact in different Eurozone countries, depending on the degree of concentration in the local banking market 3. How low can central banks go? Therein lies another problem. After all, there was a reason why most economists were doubtful of attempts to push interest rates below zero. Reduce the interest rates too much, and the private sector might simply withdraw their bank deposits and hold the money in cash. Of course, there are some costs to storing cash, with some estimates at 20 basis points (bps), and