The Reserve Bank has both a date and data dilemma: Editor’s note
Musty data means the Reserve Bank is fighting inflation with one eye closed.
Good morning. My new digs in Bondi is up three flights of stairs and I spent most of the long weekend out of breath hauling boxes. There hasn’t been a chance to catch it back this week: rate hikes in the US, UK and Switzerland, all of 2021’s gains wiped off the S&P 500 and the S&P/ASX 200 getting uncomfortable close to bear market territory. Panic is acceptable. Panic selling is a mistake. Take a deep breath and check out our survival guide.
Today we’re talking statistics. Bear with me, it matters more than you think.
We should be bagging lagging data
The Reserve Bank is steering an overheating economy and its most important crystal ball is past the use-by date.
It’s bad enough that Governor Philip Lowe and the Board have a date problem. The 2021 prediction of no cash rate increases “until 2024 at the earliest” will haunt him for years. What receives less focus is he also has a data problem to make his job even harder.
In a rapidly-changing market, it is unacceptable that Australia’s best measure of inflation is months out of date by the time it reaches the cash rate decision-makers at 50 Martin Place. A mild inconvenience normally, stale data risks policy mistakes when inflation is advancing at the fastest pace since Howard.
Inflation watchers rely on the Consumer Price Index (CPI). It tracks price changes for almost 900,000 goods and services over each quarter—say January to March. The data is then published publicly one month later. In other words, if something changes in February, policymakers find out in late April.
Monthly data is the norm globally. Australia is the sole G20 member—a group that includes Indonesia, Mexico and South Africa—to publish inflation quarterly.
Old-world Europe leads the pack. The European Central Bank sees how prices changed in May, on the last day of May. This, in a union of 19 countries and 750 million people.
Fresher data would make it easier to spot trends and turning points. It could help soften the nasty surprises the Reserve Bank is making a habit of delivering. April’s CPI belatedly spurred the Reserve Bank into action in May. Could higher-frequency data have caused a change of heart earlier? Stephen Miller, an investment strategist at GSFM Funds management thinks so.
“Monthly CPI data would have alerted the RBA much earlier its transitory narrative regarding inflation was not accurate,” he says. “It could have responded earlier and rectified the course it was on. I have no doubt whatsoever.”
None of this is particularly controversial. The Reserve Bank has been asking for monthly inflation data since at least 2010. In a 2017 speech, then Deputy Governor Guy Debelle said it would help “identify changes in the trend in inflation sooner”. The Reserve Bank declined to comment on this story.
The potential move to monthly reporting has been mooted for over a decade. The Australian Bureau of Statistics (ABS) first rejected the idea back in 2010 due to cost. By mid-2018 it had changed its mind. New technology had slashed data collection costs and the agency “committed to development work”. It expected to take one year.
Fast forward almost four years to March 2022 and the ABS announced it was “examining the feasibility of a monthly CPI”. An information paper outlining work-to-date and methodology is due in August 2022. Given a year of test data is required to verify and baseline the new series, we could be years away from policymakers getting access. Asked by Morningstar this week about the three-year delay, the ABS blamed the pandemic.
But given the ABS’ own schedule had development finished by mid-2019, months before the pandemic, one can’t help but wonder if the work simply fell off the agenda.
Budget cuts may be partly to blame. Outgoing ABS chief David Kalisch warned in 2019 that a 30% fall in operational funding over the previous decade had put essential statistics at risk. The agency noted in 2018 that more funding would be needed to cover the ongoing costs of maintaining a monthly CPI.
And there are those who argue the ABS should prioritise other issues. Chief economist at AMP Shane Oliver would like to see the ABS narrow the hefty lags between when data is collected and reported: it takes a month for the CPI, six weeks for wages and two months for Gross Domestic Product. Other developed countries manage it in half the time or less. More funding for the ABS sounds like a good place to start.
Monthly inflation data won’t make Philip Lowe omniscient, nor will it eliminate policy mistakes. Rapid-fire data is noisy and prone to misleading fluctuations. Helpful but no “panacea” says David Plank, head of Australian economics at ANZ.
But when every change in the cash rate influences billions of dollars of investments and thousands of jobs, we should equip our policy makers with the best tools available. The ones they’ve been asking for since 2010.
The ABS estimated then the shift would cost $15 million annually. Web scraping and card scanner data mean collection costs are likely a fraction of that today. A few million dollars is a pittance to pay for the chance to improve monetary policy decision-making, even if only a little.
More from Morningstar this week
When will the bear market end? With Bubbleville rapidly deflating, Mark Lamonica looks to history to gauge how far markets remain above normal and finds there is still a way to go. Retail investors throwing in the towel is a common sign the bear market is bottoming. Local investors are no longer buying the dip but they’re stopping short of mass capitulation.
For the brave of heart, Anton Tagliaferro and Peter Warnes separately share their thoughts on where defensive investors could think about putting money to work. Also, bonds are back!