A Change to the Cash Rate Assumption Method for the Forecasts

Our forecasts are conditioned on an assumption for the cash rate path.

RBA staff provide forecasts of economic activity and inflation to the Reserve Bank Board as an input to their monetary policy decision-making. These forecasts are built with a suite of models and assessments, which are underpinned by assumptions that describe the current and expected future state of financial conditions, including paths for the exchange rate, commodity prices and the cash rate. These assumptions are not the staff’s or the Board’s views about the future path of these variables but are drawn from financial markets.

From 2015 to 2022, the staff’s forecasts were conditioned on the cash rate implied by market pricing for overnight index swaps (OIS), hereafter referred to as the ‘market path’. This methodology allowed staff to incorporate an independent and readily available cash rate path into the forecasting process.

In February 2022, the assumption was changed to incorporate the average of market economists’ expectations of the cash rate target, as surveyed by Bloomberg. At that time, the gap between market economists’ views and the market path was especially large and the market path was unusually volatile. Combining information from market economists and the market path dampened large revisions from rapid changes in the market path.

In May 2024, we reverted to setting the cash rate assumption using the market path only. Nevertheless, market economists’ cash rate expectations, and forecasts of the economy more broadly, remain an input to our assessment of the outlook.

The market path provides a robust estimate and is internally consistent with other assumptions that underpin the forecasts.

There are a number of ways that a cash rate assumption can be generated. For central banks that condition their forecasts upon an external assumption about the policy rate, most use the market path (e.g. the Bank of England and the European Central Bank). For other central banks, their forecasts of the policy rate are consistent with their forecasts of other key variables, including inflation and unemployment (e.g. Sverige Riksbank, the Reserve Bank of New Zealand and Norges Bank).

We recently reviewed three options for the cash rate assumption: (1) the market path; (2) expectations of market economists; and (3) a simple average of the market path and market economist expectations. The market path has several benefits. It is:

• simple, replicable and transparent, which means that it is easy to interpret, implement and communicate
• robust – OIS markets are deep and liquid (and therefore unlikely to be influenced by changes in who is participating in the market) and rates from these markets are readily available
• consistent – this method maintains internal consistency with the exchange rate and other financial market variables at the time of the forecasts
• extendable – a longer forecast horizon via the market path is available relative to the survey of market economists, facilitating longer term forecast scenarios and other extensions
• accurate – a more accurate assumption will underpin more accurate forecasts, although this effect is marginal compared with the overall uncertainty in forecasting.

To assess the forecast accuracy, we compared the three options to the actual cash rate prevailing over the forecast period, from 2007 to 2023.^[1] We found that the market path has typically been a slightly better predictor of the actual cash rate from one quarter to a 2½ year horizon (the horizon for the RBA’s forecasts) (Graph 1). While differences in performance are fairly small out to a horizon of 1½ years, they are statistically significant.^[2] It is perhaps not surprising that the market path is a slightly better forecast of the actual path of interest rates at the points in time that we have assessed. The market path is available in real time and responds quickly to new information. The surveys of economists, on the other hand, are taken at specific points in time that do not align exactly with the RBA’s forecasting schedule. They are therefore unlikely to incorporate all of the most up-to-date information. There are also a number of technical factors in market economists’ responses that could mean the path implied by market economists is a less accurate predictor of the cash rate.^[3]

We will continue to monitor and assess the cash rate assumption.

If volatility in the market path was to become elevated in the future (like in February 2022), we may consider other ways to mitigate the effects of this volatility on the forecasting process. One option is to use a smoothed market path – such as a rolling average of previous days of market paths – which is used by some other central banks. The benefits from this smoothing must be traded off against the reduction in the cash rate path’s expected accuracy and a reduction in consistency with other technical assumptions. We will continue to monitor and evaluate the cash rate assumption.