The 'two wolves' inside monetary policy are demand shocks and supply shocks. It is important to know which is which.
There is a story in Native American folklore that inside each of us are two wolves, one evil, one good. The one that wins the one you become goes the saying, is the one you feed.
So too for monetary policy. In that case, though, the two wolves are demand shocks and supply shocks. The demand shocks are well known and well understood. Policymakers know how to respond to these. When demand is strong, tighten policy to keep inflation in check. When it is weak, ease policy to support demand.
In contrast, supply shocks create a trade-off between keeping inflation at target and stabilizing output or employment. It is an easy trade-off to stomach when the supply shocks are benign and inflation is showing a tendency to undershoot your target. Adverse supply shocks are less comfortable, with higher prices and lower output.
We can see this conflict of narratives in the Reserve Bank's own analysis of the economy, as well as in some of the discourse from other quarters. Some observers frame the situation as demand-driven. They focus on fiscal and monetary support during the pandemic and argue that this has resulted in an economy where demand is simply too strong for the economy's supply capacity to meet it.
No wonder we see repeated stories worrying about every extra dollar going into households' pockets, afraid that this will boost inflation. Perhaps this is right, but there are a lot of assumptions going into that argument.
A demand-led narrative is also more comfortable for the economics profession. Most of the models we learned at university primarily capture demand-side relationships. Supply capacity is usually treated as given or driven by unpredictable (and transitory) 'productivity shocks'.
Don't assume, test
We should grant that advanced economy governments and central banks provided massive stimulus during the peak of the pandemic. In Australia and a handful of other economies, fiscal support managed to overfill the income hole created by lockdowns an understandable outcome given the many uncertainties at the time.
We should recognize the possibility that some of the resulting strength in demand might not have unwound fully. But we should test this proposition, not assume it. Recognizing and understanding the supply shocks much like making the virtuous choices involved in feeding the good wolf can be hard.
Hitting the strong demand 'nail' with the policy hammer is so much more straightforward. Even when supply shocks are recognized, it is sometimes assumed that the response still needs to be to hammer down demand to fit supply.
We see some flavor of this view in the June 2024 RBA Minutes, which said, 'The case to raise the cash rate could be further strengthened if members judged that aggregate supply was likely to be more constrained than had been assumed.'
Again, there are some unstated assumptions here: that the supply constraint is persistent, or fading so slowly that demand-based policy still needs to respond to keep inflation in check. It would be a mistake to assume that a particular supply constraint lasts forever; better to understand what is driving it before reaching that conclusion.
Curves, shifts and other fruit
We must bear a few things in mind when assessing whether something is a supply constraint and how it might play out.
Firstly, and as we have said before, simultaneous increases in prices and volumes for some goods and services are not necessarily evidence of strong overall demand. We could be dealing with an 'other fruit' problem, where supply shocks affecting some goods and services displace demand elsewhere.
Another thing to be aware of is that some economic relationships are nonlinear and others can shift. The RBA knows this: it has long recognized that the Phillips Curve is indeed a curve, not a straight line. It has also, commendably, recognized that the unemployment rate consistent with the full employment concept in its mandate can shift.
There are other labor market phenomena that we should recognize are nonlinear, or shifting, or some combination of the two. As discussed in a recent note, Fed Governor Waller and others at the Federal Reserve have argued that theory would suggest that when job vacancies are high enough, they can fall back to more usual levels without unemployment rising much. The relationship between the two variables is nonlinear.
Subsequent events, including in other countries, have supported Waller's thesis. For some countries, like the Netherlands, the high vacancy rate was in line with the normal nonlinear relationship, and it could reverse without unemployment rising by much.
For others, like the United States, United Kingdom and possibly Australia there was a bit more going on. But that 'bit more' was a pandemic-related shock to labor supply. As the shock faded and participation recovered, the historical relationship re-asserted itself.
There is a deeper question about why labor supply took a long time to come back in those countries, but not others, even after social distancing restrictions were removed. Indeed, in the United Kingdom it is going in the wrong direction again. In Australia, at least, the constraint was more about the borders than about participation. This means that we can reasonably conclude that this supply shock, to the extent it was relevant, has been fading of its own accord.
Of course, there might still be other supply constraints at play in Australia. Some supply shocks might persist, while others unwind without help from monetary policy. However, policymakers might not recognize that the shock will not persist. And even if they do, policymakers might still conclude rightly or wrongly that demand needs to be reduced to meet it.
It would be good to know what the RBA is assuming about supply constraints, their causes, and their likely persistence. Newish Deputy Governor Hauser mentioned the need to do more work on supply issues during the Q&A to his recent speech. Hopefully the post-Review changes will direct more resources into such activity than was previously available.