RDP 2022-08: The Consequences of Low Interest Rates for the Australian Banking Sector 4. The Pass-through of Monetary Policy at Low Interest Rates

From the discussion in Section 3, it is clear that the Australian banking system operates differently to the banking systems typically discussed in the international literature. Unlike systems in which lending spreads are expected to expand as policy rates fall and then compress as these rates settle their new low-rate steady states, lending spreads in Australia remained broadly stable as the cash rate fell. As a result, there are three main channels through which low interest rates affect the Australian banking sector: deposit ELB, losses, and NIMs.

Brassil et al (2022) incorporate all three of these channels into BA-MARTIN, and are therefore able to capture the effect each of these channels has on the pass-through of monetary policy to lending rates when the level of the policy rate is low.

During large downturns, the losses channel makes expansionary monetary policy more effective than usual (Figure 8). This occurs because the expansionary policy has its usual effect via lowering banks' funding costs and an additional effect of increasing credit supply. To explain, expansionary policy lowers losses by ensuring unemployment and interest rates are lower than otherwise, and asset prices are higher than otherwise. By lowering losses, the cash rate reduction reduces banks' capital deteriorations, thereby mitigating the extent to which these banks reduce credit supply in response. A smaller reduction in credit supply means a smaller increase in lending spreads; hence, the pass-through of policy is greater than 100 per cent.

But losses do not remain higher forever. Once banks have provisioned for the extra losses they expect from the downturn, monetary policy effectiveness is no longer amplified. Instead, monetary policy becomes less effective than normal. When losses are sufficiently high to reduce banks' capital, the deleterious effect policy rate reductions have on banks' NIMs and credit growth will amplify any increase in lending spreads – a fall in NIMs reduces banks' abilities to raise capital via retained earnings, while the increase in credit growth inflates the denominator of banks' capital adequacy ratios (see Brassil et al (2022) for a detailed explanation) – thereby partially offsetting the reduction in funding costs. Of course, if banks have also replenished their capital levels by the time losses normalise, policy pass-through need not fall below 100 per cent (as shown in the Figure 8 example).

The losses, NIM and credit growth channels are not specific to low interest rate environments; they are active at any level of interest rates as long as the downturn is sufficiently large. The deposit ELB, however, is only operational at low levels of the cash rate. Figure 9 shows how the deposit ELB affects the pass-through of policy as the level of the cash rate falls. This figure shows the peak pass-through (when the losses channel is at its peak) and the lowest level of pass-through (once the losses channel subsides); these curves converge due to the nonlinear effect of interest rate changes on loan losses.

In short, the losses channel is an especially powerful, albeit short-lived, amplifier of monetary policy during severe downturns. But at low levels of interest rates, the losses channel is unlikely to be sufficiently powerful to fully offset the deposit ELB. Therefore, at low interest rates, we should expect less than full pass-through of monetary policy.

That said, even during the large hypothetical downturns explored by Brassil et al (2022), policy pass-through remained above 80 per cent for positive and small-negative levels of the cash rate. So cash rate reductions remain an effective policy tool during large downturns and at low interest rates. The question then is: are there scenarios that weren't explored by Brassil et al (2022) in which cash rate reductions are no longer an effective tool of monetary policy? This question is explored in the next section.