Transcript of Question & Answer Session Understanding Supply Shocks and Their Implications for Monetary Policy

Along the lines of Isaac Gross’s comments, the research I’ve seen from the US shows that 50% of inflation due to recent supply shocks has gone straight to the bottom line TO Profits, 42% has gone to non-wage factor costs and only 8% has gone to wages. even prior to the recent shocks, about 50% of inflation is a result of profit growth.

I think there is some similar research in Australia, but I haven’t cited it. Professor Warwick Smith’s talk tomorrow touches on exactly this matter. Now, you’ve spoken to structural long-term factors and the difficulty of measuring them. But one obvious measure of this dynamic is excess returns to equity or rate. So, my question is, aren’t interest rates a fairly blunt instrument to address inflation caused by profit growth, especially if it is a result of industry concentration and in fact exacerbates the problem because it raises the cost of capital.

And why do we never hear about profit price spirals and only about wage price spirals?

Thank you for the question and thank you for the comments. I’m really pleased actually that you picked up on some these that’s coming because I wanted to perhaps just flag a few things and I think I’m going to say this a lot in this Q&A. I was deliberately with the speech taking a very sort of ‘40,000, feet view’ of the world and was simplifying a lot of that messiness that I mentioned just to sound how we’re thinking about the current shock, I guess a couple of comments.

One, I’d point you in the direction of our May SMP, which is our latest SMP, it’s got a whole chapter where we unpack those transmission channels. Just to be clear, our assessment on the data is that Australia is very much an oil importer, a natural gas exporter for sure in a very substantial way, and we agree that the terms of trade are definitely moving natural gas prices very closely linked to oil prices so, there is that, that channel. But, then you have to sort of think about how does that income actually flow back through the economy and how does it show up. But I think we, you know, generally would agree with the premise that it does look different in commodity countries and you do have to think quite carefully about distribution. So, it’s cold comfort to households that are paying more to fill up their car. They’re definitely having to do that. They may well, in the coming months, have to pay more for the food that they need to eat.

There is this other channel on the natural gas side that could well be an offset, but there’s a redistribution going on. So that was some of the messiness that I didn’t talk to, but we are aware and I’d definitely point you at our SNP chapter that goes into a lot of detail. To the question on profits and interest rates and how we think about those and, you know, profit price spiral versus wage price spiral, I’d say wage price spiral is not language that we’ve used in recent years. We don’t think that that’s what we’re seeing in the labour market right now. There’s no sort of evidence to support that - so that’s not a view that we take at the moment, that that’s happening. In terms of how we think more generally around inflation dynamics and some of those metrics you mentioned around profits, our research shows that at a very aggregate level - so, I’m trying to pull up again from individual firms – individual firms and their pricing decisions - their profit margins will change over time. If you’re a stock picker or a tracker of individual companies, you can see that in their reporting that they put out twice a year at least.

But pulling sort of right up from that, looking across the whole economy, which is what we do, generally speaking, what we’ve seen over recent years is, we haven’t found much evidence for substantial moves in the profit metrics that we track. We actually put out a Bulletin article in April that dug into this - so, I’d definitely recommend you have a look at that for those who are interested. We present a number of those different metrics. One of the tricky things is if you’re trying to look at that aggregate level, which is where we focus, it is quite hard to get clean data that’s a measure that you can grab hold of that’s meaningful that you can then use to do an assessment. It’s much easier to look at profit on an individual company basis than it is at an aggregate whole economy level.

But our assessment that you find in that Bulletin article is that you know profit margins and mark-ups can move for a number of different reasons, but we haven’t found evidence of a substantive cycle there and so, we’re concerned with that short run cycle when we set interest rates.

And then finally, just to your comments on what’s happening structurally to profits over time, I mean it will come through the numbers and it’s something that we can see and observe, but because it doesn’t move inflation through the cycle, or if it’s not moving inflation through the cycle, I should say, then monetary policy is generally not going to respond to it.

And there are other sort of arms of policy that that would be responsible for looking at that, but that’s not in our purview so we wouldn’t be really digging in on that. But look all of these things are really important to think about and they’re certainly we’re not blind to them, to what we see happening, but we really do focus on the things that matter for inflation, particularly through those shorter run cycles where monetary policy is having an impact on the economy and the longer-term structural stuff is the stuff we’re tracking so that we can get a clear sense of where the economy is today. That’s how we have to judge it by what’s happening to those underlying trends.

Has the effect of rising prices affected production, changed the effect of short term and long-term shocks on inflation, and does this threaten the role of reducing unemployment as an answer to inflation?

I can certainly talk to that. And maybe my fellow panellists can provide some comments too. So, I think that what you’re getting at is structural change in the economy and we know that that’s been happening, we can see that quite clearly over time. If you roll the clock back far enough, the manufacturing sector was much larger, the services sector was much smaller, the composition within those broad aggregates also changes over time. And because those sectors make use of different factors of production then in a relative sense, so the relative cost of different factors of production, they will move and change over time. That’s a natural, sort of development pathway for any economy and we are not unique here in Australia to experiencing that. In terms of how we factor that into our models, well we’re seeing it in the data, and the models that we build are therefore innately capturing those transitions and those shifts. There’s a lot that goes into that model development process. Sometimes we have to take a view that we think that an event has distorted what the data might be telling us. Covid is a really great example. I’m sorry, I’m talking to a bunch of academics here, so I’m sure you all know, you could just estimate your models through that and do nothing and see what you get. That’s one, option. But I do think for that kind of shock, and we certainly do this internally, that we take a very close look at the data through that period and really think, well is this a fair is this actually a reflection of what’s happening fundamentally, or is something being distorted by the pandemic? And to state the obvious, lockdowns are not a normal sort of way for the country to function. So, we wouldn’t want to take some inference from lockdown data and apply that to today if it’s really just reflecting the impact of lockdowns.

I had a question about the new approach to trying to adapt your approach to monetary policy in a new, what you and your colleague were referring to, as a new era. UBS put out a note recently talking about an unprecedented policy miss - inflation stayed above target for quite a long time now. I mean, isn’t that evidence that at the very least you’re not managing to get this balance right at the moment, trying to navigate this new era. And it just looks like at the moment you’re just applying a weaker form of monetary policy, just kind of toning back your response to high inflation and trying to wait to see what happens.

Is that being unfair or do you think that essentially the way that the monetary policy was set up, you know, (panellist) Cherelle was talking about when she was that the RBA that you weren’t conceiving of these kind of things, or maybe you were, but it’s a demand management tool basically in a world of supply shocks. Is it fit for purpose?

It’s a great question. I mean, I’ll maybe just sort of, I guess, talk through what we’ve seen recently because I think you’re talking about the lift in inflation through the back half of last year and then obviously, we’ve had a a further shock this year coming from the conflict in the Middle East. I mean, I think, when I look back at the last 12, 15 months, there were a number of factors that played out differently to how we were expecting and I think, differently to how other people were expecting as well.

For example, the trade and tariff challenge that we faced into in May last year, we had Liberation Day at the start of April, and then it’s sort of, backwards and forwards. It felt like through that period you woke up overnight and you’ve got a new set of tariff numbers. I think a lot of people were expecting that to have quite a substantial negative impact on the global economy.

We, in our May S&P last year, we explored that in a scenario. It certainly was a factor in the Monetary Policy Board’s mind if you look at the statements of the minutes from that period. I think what played out there was that actually the world economy, the trade system, was much more resilient. The actual tariffs ended up being a lot lower as well than what was initially announced.

So, you know, we’re doing this in real time. We had the information we had to hand but it played out perhaps a bit differently to what was expected. There’s a number of other factors, too. I think the AI boom has been, in terms of its investment boom into data centres, has been spectacular and it’s quite hard when something is very new like that it is quite hard for the statistical agencies, never mind anyone else, to actually track it. And so that’s been something that’s come through that we weren’t anticipating 12, 15 months ago but we’re now seeing not only overseas playing through the Asian supply chains, but here in Australia as well.

And finally, I think the local economy was more robust than we thought it was going to be. Through the back half of last year in particular, we did see a pretty marked pickup in household spending, I’ve mentioned data centres and that came through in business investments, and housing market too. And so, all of those things sort of came together and we had, you know, stronger demand ran up into a supply constraint - I’ve showed my charts earlier on - and that then flowed through and lifted underlying inflation. You know, perhaps you can say we should have spotted some of those things, you know, I hope at least we were clear all the way through on the SMPs where we saw the risks. In fact, in August last year, we highlighted the risk of inflationary pressures emerging - and that is what happened. I also think some of the more recent events, too, these are very tricky things to forecast. I’m not sure anybody had their money on a war in the Middle East and that’s obviously had implications which are further compounded where we were starting from. So, when you’ve got this sort of volatility and uncertainty, I think it’s tricky for any framework to navigate through that because innately things are moving that you can’t predict, certainly can’t predict the timing, very hard to predict even the size even if you do know the timing.

In terms of the framework more broadly, is it fit for purpose?

I think that it is. I think that we’ve shown, and our historical experience of the 70s and the 80s shows, how difficult it is for economies to function when inflation is high and volatile. It just makes decisions really hard, from really big decisions like, you know, big businesses deciding whether or not to invest and build a new factory or invest in a new commercial property or whatever it might be, if you don’t know what that’s going to cost you, really tricky to then work out if it’s worth doing or not, all the way down to much smaller decisions. And we hear this all the time from people that we go and talk to in the not-for-profit sector, in community organisations, it’s really tough for households that are at the lowest income levels, that are in a vulnerable position when they go to the shops and they don’t know if they can afford the food they need, and they have to make some really hard choices sometimes. That’s because inflation is higher and those prices are moving more quickly; it’s harder for them to budget and plan. And that’s not by fault, but that’s the reality of a higher-inflation environment.

So, I do think having a central bank – an institution that’s tasked with keeping inflation low and stable – understanding what’s happening and responding as it needs to with its policy instrument is really, really critical and vital. And that’s the job we’re here to do. And I think, you know, we are doing it. There was a comment earlier on about ‘interest rates are a blunt tool.’ They are a blunt tool, but they work because they go everywhere in the economy. Every single part of the economy will be impacted by whatever the interest rate settings, that’s people with a mortgage to people with savings, to what happens with the exchange rate that we were discussing earlier on and how that flows through the economy, they get everywhere. And that’s why it’s a tool that does work ultimately. But it is blunt and it is hard. And as I’ve been talking about, we are putting a lot of time and effort into understanding this new environment because we know we have to because we have a really important job to do.

I’ve got a question for Sarah. I was very interested in your comment that an increase in interest rates might lead to an increase in labour supply, and that would increase the capacity of the economy to respond. It also might increase unemployment, because if you have people coming in looking for jobs, you increase unemployment. So, my question is how nuanced is the RBA models to be able to understand where that fits on the pendulum, which markets, which occupations, which industries – so basically, I’m saying: how many occupations and industries do you have in your modelling? Thank you.

No, great question, thank you. I will say just upfront for those that are interested in the details of that result, it is published on our website now. And it was by a colleague, Jon Hambur, Jon Hambur led that work. So please go there. And so I’m also going to put the health warning on: if I say something that contradicts that paper, that paper is the source of truth.

So that paper was making use of the ABS’ PLIDA data set; that’s why the disclaimer for PLIDA is actually at the front of my speech online. That’s the data set that looks at individuals – they’re anonymised, don’t worry, I don’t know who you are – but it makes use of various administrative data sets. It allows us to effectively track anonymised people through time. And what we were able to do there was match up people’s employment using their tax record against whether or not they had a mortgage, and then we could look at how that changed, effectively. And so we could unpack whether or not you had a mortgage, if it was high, and then we were obviously using interest rate changes as our event, if you like, to then track what happened to your labour supply.

So that particular result – we were not able to dig into anything about somebody’s job, we didn’t have that information, you don’t tell the ATO what you do for a living, you just tell them how much you earn, so we weren’t able to dig in on that particular question, but you’re right. It is actually an area that we are very much looking at and again, making use of this micro data that’s now available.

Some other research that we published – this is a while ago, might have been February last year – where we were looking at transitions through the labour market from sector to sector and how that was playing out. And so that particular bit of research, for example, found that there were some people moving from some of the market services sectors, hospitality and retail, into what was at the time the fast-growing non-market services sector, healthcare in particular. So, we do dig in and look at those sectoral compositions when they’re particularly useful for us for understanding dynamics. And that one was a very useful dynamic, because we did have this fast-growing period of non-market sector employment. We wanted to understand what was underpinning that.

But the data, and the models, we are always constrained by what’s available. And that PLIDA data set, for those that don’t know, is very, very new. It’s only been available just in the last couple of years. It’s getting bigger and more comprehensive over time. As it does that, we’ll have access to more and we’ll be able to play with it more, frankly, and learn more from it. But our more traditional, longer standing, workhorse, general equilibrium macro models, they don’t go into the kind of detail that you’re talking about because they’re built on data sets that don’t have it. And, necessarily, we have to make some decisions and calls about the level of detail and granularity, versus the tractability of the model and its usability. So, it’s horses for courses. The good news is we’re getting access to more and more horses, which means we can run on more and more courses, which for me is great, we’re just learning more all the time.

Sarah, you spoke about the importance of credibility in achieving the RBA’s dual mandate, and I think that applies to institutions more generally. So, I pose this question to all of you. I feel like we’re here today because we care about the wellbeing of people, and I feel like people are exposed to the most noise that they’ve ever been exposed to at the moment. And I wonder whether this has posed any noticeable challenges in your work, and further, whether doing a good job is enough or whether you know more positive action is required.

Yeah, great question. Just a couple of comments and then I’ll hand to my fellow panellists. I don’t disagree, I mean, I said it: this is definitely a time, it feels, where we’re sort of seeing one shock after another. It’s more uncertain than it has been certainly in the recent past, and that is challenging. I think of myself as an individual: five years ago, I didn’t wake up and rush to check the news. I didn’t- well, maybe I did five years ago, sorry. Maybe we go back pre-COVID; I’m thinking five years ago was actually during the pandemic. So, in 2018–19, I wasn’t rushing to my phone to check what was going on because the world wasn’t moving that fast, and I didn’t feel like I needed to be that immediate when I woke up for the day. Now, I don’t remember the last time I didn’t check my phone when I first woke up to see if there’s anything I need to be aware of. That’s really challenging to live with, right? To not have that certainty, to have things change so frequently. These are difficult to manage.

Certainly, where we think we might be seeing some of that – and we’re thinking very hard about what it means, what it means in terms of how it translates into decisions and then ultimately what that means for policy – is on some of the consumer and the sentiment series that we can track. So, for instance, consumer sentiment did hit – in one of the series – a record low at the outbreak of the conflict in Iran. It’s recovered a bit since then, but it’s still very, very low by historical standards. How do we understand that? What’s that telling us about how – obviously consumers are not feeling good about things – but how is that translating into the decisions they’re making? That then comes back to the policy decision we have to make. That’s just one very active area for us. Uncertainty more broadly is a is a general topic that we pick up and focus on a lot. But I think you’re not wrong. And I think it really does, at least for me, emphasise how important it is that we keep working to get it right, to do our jobs. But I might hand over to others now.

Dr Hunter, in the latest RBA Minutes I believe they said they talked about a potential material weakening in the housing market, including whether this could inhibit growth in consumption. I was wondering if you could talk through how you’re reading the tax changes, the market response to them, and this particularly in relation to the shocks that you talked about today?

Yeah. So, the housing market, I think, has experienced a few things in quick succession, some of which we know empirically, from past experience, will slow down momentum, and others, perhaps, are a bit newer. To start with interest rate rises, we’ve seen it historically, so we weren’t surprised to see them have an impact on the housing market – this is the established housing market, to be clear – it’s fairly normal for that to happen. The housing market actually, in terms of domestic monetary policy transmission, is the most powerful transmission channel that we have. So that response was not unexpected. We also have, as we were just talking about, the uncertainty of the conflict in the Middle East, and we generally, think that that type of uncertainty will again dampen down activity in a market like the housing market, and so that then flows through into prices.

Tax is not our thing, we don’t do tax policy, and we sort of got the details of the tax policy when it was announced in the Budget. But we can see, and we can obviously read, what commentators and experts that work in the market are telling us, which is that those tax changes have caused some people to stop, and pause and think about the decisions they might make and then that sort of adds to that slowdown in activity and you can see that coming through in terms of prices.

I think what’s really important for us is the aggregate impact of all of those things. First off, on the housing market itself and then how does that flow through the economy? So, one channel, the way it flows through, is through wealth effects, and that then shows up in consumption. And so if you have a fall, or at least slower growth of your wealth than you were expecting, then maybe you might pull back on at least some categories of spending. Although I have to say it is very category-specific. So, we don’t find much of a relationship between housing wealth and spending on food, for example. That’s probably not too surprising. We do find that it has a bit more of a relationship with spending on things like homewares and household goods, and that partly, actually, reflects the fact that if there’s less people moving home, then there’s less need, when you move into your new house, to buy a new sofa or curtains or whatever it is you might need, because you’ve just changed where you live.

So that’s that channel, but it also gets into dwelling investment. And so, if you think about a developer who’s thinking about a new dwelling investment project, if their expected price at the end of it – the value of the project at the end of all of the development activity – is a bit lower, then maybe at the margin, some of those projects don’t go ahead, and so we get less dwelling investment activity. So that’s how we think about the housing market and how it flows through. We’ll obviously keep monitoring it. We’re just going into our next forecast round ahead of the August meeting. So, in just over a month’s time, we’ll have the meeting and then we’ll be publishing our updated forecast, so you’ll see then what the profile looks like. But yeah, no, it’s certainly something that we’re thinking about in that aggregate, broad sense and how it plays through across the whole economy.

You mentioned about inflation pressure from the household side. You mentioned that increase in labour supply, someone asked a question earlier, but that’s a long run adjustment, I thought. But the short run, mostly household savings, used to address the inflation pressure from the household side. So, do you have data for that – that sort of short run adjustments – and has any research been done on that?

Yeah, it’s a good question. So that labour supply response, that empirical finding, you’re right. We ran that that particular analysis over a couple of years. Although, depending on what your definition of long-run is, I would actually say, given that we were seeing some response to labour supply within a one-to-two-year horizon, that’s a horizon that does matter for us, so I would characterise that more as a short-medium term. And that’s why it was so great that we could get that data, because we’ve never been able to really explore that type of response before, because we just didn’t have the granularity we needed to see if that was happening, and if so, how large.

To the other part of your question, how do households generally manage through these types of shocks and how do they respond to monetary policy more generally? You’re absolutely right. In terms of monetary policy specifically, we do see evidence and we can, get a handle on what this room would call the intertemporal substitution effect. For, for the non-economists, that’s the decision on how much you save versus spend – when interest rates are increased, we’d expect to see more saving and less spending, the return to saving is higher. We can see evidence of that. One data point that we look at, for example, is for those people with a mortgage offset or a mortgage redraw account, do people choose to put more in, over and above their minimum repayments when interest rates are higher, less when they’re lower? We can see that pattern through the data. So, we have seen, over the last few months that payments that are above and beyond the minimum have lifted a bit. So that’s one evidence point for intertemporal substitution.

We can also look more broadly at what happens to savings rates across the whole economy because that particular mortgage redraw offset, that’s only around 30% of households, so it’s everybody else we need to consider as well, and we can find some evidence of it too. It’s a bit harder in an aggregate sense, we don’t have as good, clean data, which is why that offset and redraw data set’s particularly valuable to us.

So, I think the short answer your question is yes, and we are looking at it empirically. But it was interesting to see that that labour supply response was what I would call cyclical, as well as there being, obviously, structural trends in the labour market, which we’re also very much paying attention to. We want to understand structurally where participation rates are going, where population is going – which is where Yiyong and his team come in – and other dynamics through the labour market.