Inflation indicators for fun and profit: what to make of the new ABS monthly CPI indicator

Plus, the RBA’s review of forward guidance

Aug 18, 2022

The ABS has released details of its new monthly CPI indicator. ‘Indicator’ is used by ABS to distinguish this partial series from a fully-fledged monthly CPI. The indicator covers between 62% and 73% of the CPI basket, depending on the reference month. The first formal release will take place alongside the Q3 CPI in October and then at the end of each month after that. For now, we have the monthly percent change from December 2018 to June 2022 to work with.

What can we say about the new indicator? Firstly, it’s volatile. The ABS itself recommends referencing the annual or quarterly rather than the monthly percent change. It will also be subject to revision. An obvious benchmark for the ABS indicator is the Melbourne Institute’s monthly inflation gauge, for which we have a 20-year history. As the following chart shows, the ABS series is more volatile on a month-to-month basis:

We can also use the Melbourne Institute inflation gauge to benchmark the ABS indicator in terms of forecast performance. A simple model that explains the quarterly CPI in terms of contemporaneous averages of either indicator suggests the ABS series has the smaller mean forecast error, even though the Melbourne Institute series has somewhat better fit for the common sample available for both. A model of the quarterly CPI that includes both indicators performs better still and the two indicators have independent explanatory power for the quarterly CPI, suggesting they are picking out somewhat distinct components of the inflation process. So we should be able to improve on near-term forecasts for the headline CPI based on the new monthly indicator.

The ABS has yet to produce a trimmed version of its monthly indicator, although is looking into it, so the Melbourne Institute’s trimmed measure will still be useful for forecasting the ABS quarterly trimmed mean inflation rate. The RBA tends to focus on the trimmed mean as a measure of the persistent component of inflation that best forecasts future inflation.

One big advantage of the ABS over the Melbourne Institute indicator is that it is not behind the Melbourne Institute’s paywall. Back when the series was sponsored by TD Securities, I would get it for free thanks to the generosity of their economists, but for those who don’t have access to a data vendor or are not able to stump-up for a subscription, the ABS series will obviously appeal.

TD Securities and the Melbourne Institute performed a valuable service in developing their indicator back in the mid-2000s. They showed what could be done with a limited budget, in contrast to the (from memory) $15 million price tag the ABS put on a fully-fledged monthly CPI. But even granting the ABS their cost estimate, a monthly CPI easily passes any plausible cost-benefit test. Hopefully, there will continue to be non-government innovation in this space, possibly including development of a real-time inflation measure.

Another implication of a monthly CPI is that will increase the salience of inflation data relative to other indicators. Having a monthly labour force and a quarterly CPI gives the labour market greater prominence for financial markets and arguably skews the market’s focus. This reinforces the RBA’s own tendency to focus too much on the labour market in conditioning the outlook for inflation.

Peak inflation?

The flat outcome for the US July CPI holds out the promise of a peak in the US inflation process. The Cleveland Fed’s trimmed mean moderated from 0.8% in June to 0.4% in July. In Australia, the Melbourne Institute inflation gauge trimmed mean recorded a 1.8% m/m and 4.5% y/y rise in July. If the July reads for the US and local trimmed mean are indicative of the rest of Q3, then we can look forward to a 1.4% q/q rise in Australia’s trimmed mean inflation rate, down from 1.5% in the previous quarter, taking the annual rate to 5.4% from 4.9%. We will update this expectation based on the monthly data for August and September. The RBA still expects a headline inflation rate of 7.8% by the end of the year and the Melbourne Institute’s diffusion index of price changes showed a further broadening in inflation pressures in July:

Wages growth in line with RBA forecast

The Q2 wage price index printed in line with the RBA’s August Statement on Monetary Policy forecast at 0.7% q/q and 2.6% y/y, although disappointed the median market expectation of 0.8% q/q and my own of 0.9% q/q. That still left private sector wages growth running at its fastest annual pace since 2013. While there is a version of my model that yields 0.7% q/q, this requires incorporating a very slow long-run adjustment process that risks missing a short-term acceleration in wages growth. My sense is that the rapid acceleration in inflation has caught the wage setting process somewhat flat-footed, but looking under the hood, that adjustment is coming through for those wages that were actually adjusted over the quarter. It is also very evident in business surveys like NAB’s shown below:

The unemployment rate fell to a new 48-year low at 3.4% in July such that the number of job vacancies in the most recent reference quarter exceeds the number of unemployed people. Home Affairs is said to be sitting on nearly one million non-humanitarian visa applications. The labour market is still tightening.

Whereas Governor Lowe was once a cheerleader for +3% wages growth in the mistaken belief it led the inflation process, he may soon be in the position of cautioning against, you guessed it, +3% wages growth, on the basis that above-target inflation is being embedded in the wage setting process. At least the data is far less equivocal on the subject of the dynamic relationship between inflation and wages.

RBA economists’ panels and review of forward guidance

The RBA has moved to formalise its existing informal private sector economists’ and academic panel. I attended the first of the academic panel meetings in November 2019. This was a good initiative from the RBA and was explicitly designed to mitigate the dangers of group think within the Bank, although was somewhat de-railed by the pandemic.

The first meeting of the panel of private sector economists is planned for 6 October and the meeting of the academic panel will be held on 27 September. The agenda will include the role of forward guidance ahead of the Board’s review of forward guidance at its November meeting. Zac Gross suggests some answers to the RBA’s consultation questions.

ICYMI