Skip to Content

Is it all falling apart for central banks?

Years of rapid financialisation has led to colossal debt growth with low interest rates and quantitative easing (QE) helping to push asset prices higher. The tide is rapidly reversing, at least for now. Central banks are unable to ignore inflation, but both inflation and employment are lagging indicators.

We are now working through the consequences of the excessive fiscal and monetary stimulus delivered 18 months ago in the pandemic. A good example of the scale of that excess is real disposable income rising more than 15% in the US over the 2020/2021 period despite the economy being in recession as a deluge of fiscal transfer payments overwhelmed the actual loss of salaries and wages.

As well-intentioned as fiscal policy was, with the benefit of hindsight, it clearly exceeded requirements across the world. Global Debt/GDP is at record levels and markets have unquestionably been fuelled by an extended era of ultra-easy money. Unfortunately, the probability of this ending well is extremely low as the monetary action being taken right now will manifest itself in significantly weaker growth over the coming twelve months.

The chart below shows that across 125 central banks, we have seen a monetary tightening campaign to take conditions into the most restrictive since the GFC (blue line inverted). This is before moves this week, particularly from the Fed which likely moves by 50bps.

Advancing this index by 11 months against the US manufacturing ISM PMI shows how this story will end. Recession and early signs are already confirming that downturn. Just as inflation is a lagging indicator so changes in monetary policy work with lags which will show up over coming months and quarters.


The difficulty is central banks must respond to the here and now even as lags prevail. Last week’s first quarter Australian CPI cannot be dismissed easily. 5.1% annual headline CPI is the fastest in more than 20 years and the measure showed some ~88 items in the ABS’ basket recorded price gains. It’s not the end of the world and its certainly not the 1970s. In many respects, everything that could have conspired to push prices higher in Q1 did. But there are some slivers of hope to consider:


1. Housing

Housing has been one of the strongest sources of inflation pressure in the last 12 months. We are now experiencing the consequence of over stimulating housing at a time when supply of construction materials and labour was materially constrained. Construction costs for building a new home was one of the larger components that exceeded forecasts contributing 0.62% of the 2.1% quarterly number alone.

Source: Australia Bureau of Statistics


There are many reasons to suggest that we are moving past peak housing mania but it will take time to show up. Higher costs, the winding down of homebuilders, higher interest rates (which have already been moving as banks raise rates on fixed rate products in particular) and just the natural normalisation of housing demand will all cool construction.

The chart below highlights the relationship between residential building approvals which exploded in 2020/21 (but is now in the rear-view mirror) and construction costs. In short, the worst might be past. We pay particular attention to the housing market because, as per the saying, as goes the housing market so goes the economy. Even the hint of higher interest rates is clearly starting to add to the list of headwinds for the sector with auction clearance rates already at multi year lows and Sydney house prices now down for two consecutive months (Corelogic data).

2. Non-discretionary prices

The RBA, along with other central banks, will be mindful of the moves in areas like food and energy and key non-discretionary prices. Increases in these areas act as de facto tightening as households can’t choose to not eat or fill up the car. That is why core inflation ex these items is the focus for them. Core inflation of 3.7% is a concern but it’s not seismically above the 3.25% in their 2022 forecasts. The ABS points out that non-discretionary CPI grew at more than twice the pace of discretionary.

In some ways, in Q1 everything that could go wrong, did. From the Ukraine conflict impacting energy prices to floods in NSW and QLD adding to food inflation. Which doesn’t mean the RBA won’t respond but it highlights that prices of non-discretionary items are already doing some of the work for them in curbing demand. The risk when price pressures spread is a more broad-based cycle leading into a wage-price spiral. So doing nothing is not an option.

Source: Australia Bureau of Statistics


Within this category, energy prices faced a perfect storm in Q1. Prices may stay elevated for some time but a 30% plus move from here is clearly much less likely.