Transcript of Question & Answer Session Why Productivity Matters for Central Bankers

Thank you, Sarah. Informative as always. I think this time last year we were talking about inflation, which obviously has been topical over the last 12 months, referenced many times in your presentation. I guess as you reflect back to us sitting on stage this time last year to where we are today, any sort of thoughts you want to share with us? I guess, from my point of view, I heard you say that we’re pretty close or within the band of inflation, but I think households are still experiencing high prices. Inflation tends to look a little over that inflation band that you discussed. So, how are you thinking about that and, when do you think that what can households and us all expect going forward?

Yes, it’s a good question. I think when I was here a year ago, I was talking about inflation expectations and how important it was to us to keep them anchored, and how pleased we were that that seemed to be the case. I’m very pleased to be able to say that still looks like it’s the case. We don’t see any signs that they’re becoming de-anchored. And - in terms of actual inflation - our forecast a year ago, we were expecting and obviously hoping that we’d continue to see underlying inflation - so we measure that through trimmed mean - to continue to moderate through this year and to be approaching - to be inside the target band by this point and certainly approaching the midpoint by the end of the year. It’s very pleasing to see that that has generally played out. We obviously had trimmed mean inflation at 2.7 per cent for the June quarter. But we are monitoring, and we are always looking and seeing how things are developing, and in the latest monthly indicator prints what we have seen is that a couple of the components that we monitor very closely, where you can get a good read from that monthly indicator, have come through a touch stronger. The Governor mentioned housing inflation and market services and so those are particular watch points for us. What we want to try and do from here obviously is keep inflation more or less where it is now and, in the underlying sense, around about that 2.5 per cent, the midpoint of the target band. So that’s why we’re going to be monitoring risks to both sides, but in the latest data in particular we’ve seen it come through a touch higher.

Fantastic. I’m going to ask one more question. I think there’s going to be a tonne of questions out there in the room. I’m particularly interested in just asking a question about AI and productivity. Obviously, we’ve heard here over the last 24 hours a lot of AI discussions in every topic pretty much. You referenced AI - or emerging technologies - in your address. How are you incorporating the impact of these potential emerging technologies into your forecasting? What timeframes do you think we’re likely to see AI or any technological gains coming through?

Yes, excellent questions. It’s very much an emerging nascent question, so I’m not going to pretend I have the answer to these things or that we have a crystal ball back at the office that we can rub and know for sure. We’re thinking about it in terms of looking at historical experience with a generalist technology like this. And a couple of observations from history are one, it does take some time, so it’s unlikely to erupt and emerge in the next three to six months, we think it’ll take a little bit longer than that. And two, the technological gains can potentially come in almost waves. We first use the technology and that gives us some uplift. Then we work out how to use it better and that gives another uplift down the track, so it could well take a number of years for these things to flow through.

My final observation is that if you look through history, each sort of big generalist technology that’s emerged and then been adopted, the speed of adoption has generally increased over time. So, it took a lot less time for the internet to become well established than it did for electrification, for example. I don’t know but it certainly feels like this time is moving pretty quickly again, so perhaps we’re not talking decades and decades, but I think we probably are talking years. And that’s why we’re very conscious that this change to our productivity assumption. We’re very much thinking about it as our view for the next couple of years. We will revisit this because I think even in two years’ time we’ll know a lot more about the impact AI is having here and around the world, and then we’ll have to look to potentially change that assumption in one direction or another. It’s certainly not a set and forget.

Yes. So remaining patient, I guess, and observing what is happening out there and proactive around integrating those into your forecasts.

And actually, rolling out this technology for ourselves within the Bank. So, how do we make use of it and how does it improve what we do, the quality of the analysis, and the breadth of that that we can give to the Board?

As a bank we’re also going through the same process. I might throw to the floor. I’ve got a tonne more questions as I said, but we’ve got the opportunity now to ask your questions of Sarah. So, we’ll start down here.

I was wondering if you could talk about how you think about the risks of a head fake in the inflation data. Clearly the high frequency data suggests that inflation maybe is a bit hotter in Q3. We’ve had that signal before, from the data, and it turned out to not be the case. And then I’m also reminded of the US, which I think has quite high-quality inflation data and has obviously had - that started this year, a bit of a high inflation scare but it’s ended up being quite moderate. How do you think about those signal to noise judgments in the context of what I think to be - I mean, no disrespect to the ABS - but I think that the US data is probably higher quality than the Australian data, at least at this moment in time. And I guess, just to tie up the question, it does seem like an unusual place to end an easing cycle with the unemployment rate trending up and policy a little bit restrictive. So maybe you can fold that into your judgments about the inflation pulse.

Great question. So, could we be head faked by the inflation numbers? Look, that’s always something that’s on our mind. I mean, the fundamental of any type of analysis of any given data point is how much is signal and how much is noise. And so, we dig into all of the series to really try and understand that, and obviously you do have to make some judgments about that too. If you knew precisely, there wouldn’t be this risk, we wouldn’t have to worry about it.

In terms of the latest data and our read of it and our sense of it, so we - particularly with the monthly CPI indicator - really don’t look at the aggregate indices that are published, so that’s the headline series and also the monthly trimmed mean. That’s because the coverage is incomplete and various other challenges there. We really do focus on the components where they give a genuine read. So, that’s why month one you get less information because it’s less complete. Month two is a good month because you get market services and that’s for the quarter, not just for that month. And so, I suppose our read of the data is looking at those components where we have, at least more confidence than in others, that the two months together are giving us some signal. But could we see a bit of a drop back down in the December quarter for instance, that there’s some one-off temporary factors in there? Absolutely. We’re not discounting that fact and obviously we’re working through that at the moment in the context of our November forecast. So, you’ll see where we land up in all of that in a few weeks’ time.

In terms of thinking about that and in the context of restrictiveness of policy and those sorts of questions, I think the key question for us really, in terms of policy setting, is what are the outcomes that we think we’re going to achieve over the one to two-year horizon. So, you’ve heard this from many central bankers I’m sure, monetary policy acts with a lag. So, we’re not influencing the data that we get today. Well, clearly not because that was from backward-looking anyway. But beyond that we’re not really having much of an impact on the economy in the very near term, the three to six months. It’s only in the six, 12 to 18-month horizon that whatever decision is made today has an impact. So, we’re always forward looking and our assessment and our advice to the Board, in terms of the policy settings, is looking ahead to that point in time. So, whether we think we have to stay a little bit restrictive or change that is really a function of what we think the economy is going to look like in 12 months. And that’s really the key judgment for us in terms of that policy advice.

My final comment on that is that I often get asked about neutral, what’s the neutral rate for the cash rate, and how quickly you’re going to get there, those sorts of things. Any estimate for neutral, number one, has incredibly wide error bounds. It’s an incredibly hard thing to estimate. And number two, it’s a long run concept that takes - that looks through, if you like, all the shocks and shifts and everything else that might be playing out at any given point in time. Of course, at any given point in time, there are always shocks and shifts running through the economy and today is no different. So neutral is only ever a guide. It’s, sort of, up there at the top. It is not the anchor point for actual policy setting.

Do we have another question from the floor? Microphone down the front. Thank you.

Arguably interest rates before COVID were far too low and, of course, they got to very low levels in Australia and elsewhere. What do you think central banks, including the Reserve Bank, learned from that period? And what do you think is - I mean, we used to read the textbooks that you had a zero bound on interest rates, but you know that was proven not to be true. But what do you think is a - sort of, a lower bound on interest rates in the future?

It’s a good question. It comes a little bit back to the question around neutral and what is neutral. But I think more generally, yes, the rates were low and then went to the zero lower bound during COVID, that was clearly an exceptional time where all arms of policy were providing exceptional support to the economy. I would never discount the risk that we end up with another such event that requires that type of support. So, I would never say that we won’t go back to the zero lower bound. I think you always have to prepare for those very, very bad outcomes and they can materialise in the future. And colleagues in the financial markets group at the Bank are always looking at what monetary policy implementation looks like if we did have that very bad outcome. So, I guess that’s one part to your question.

In terms of where interest rates are now and what is neutral now compared to pre-COVID, that is very, very hard to know. Different models will give you very, very different answers to that question. That was my point around the error bands around these models are really wide. So, I don’t have a definitive answer for you. The best answer that we will ever get is what we observe happening in reality to interest rates, but then also to economic outcomes. That’s ultimately going to tell us where neutral is. But where it is today precisely, I don’t know. I can’t tell you that and most central bankers will say the same. We don’t know. If it was that easy, the job would be a lot easier. Unfortunately, it’s not.

Thank you. Bringing it back to productivity, you mentioned that we’ve seen productivity growth slowing since the mid 2000’s. There are a number of factors, I guess, that you called out there - technology lag here versus in the rest of the world, competition decline etc. What can Australian businesses do, beyond what government can do, to improve productivity in the private sector?

Oh, what a question. If there was an easy answer to that question, we’d all be doing it already and we wouldn’t be talking about it and trying to work it out today. I think what’s interesting for an economy like Australia, and I mentioned some of the research that my colleagues at the Bank have done on this, is that we are a small open economy, relatively small in the grand scheme of things, and so diffusion of new technologies that may be coming in from overseas is particularly important. It’s very hard for us to break new ground right across all sectors of the economy, because of our size and scale, we’re just not that big. So, diffusion is really important and so thinking about how we bring in and propagate through the economy new technologies I think is going to be pretty critical. But beyond that and to the specifics for specific businesses, I think that’s pretty tricky to completely work through. And to be honest with you, it’s not an area that we at the Bank focus on. It’s not our core mandate. The Productivity Commission and others have got lots and lots to say on that so I would always send people in their direction.

Thank you. Another question over here, in the middle.

Thank you for your time. I had a question around Q2 consumption data was obviously quite strong, but people were calling out a lot of one-off factors within that data. Are you able to talk about how you’ve seen Q3 consumption data tracking so far and whether that’s as you expect - and how you’re thinking about that going forward?

Great question. The consumption data at the moment definitely looks like it’s being impacted by seasonal shifts. So, we’re getting these very strong December quarters that we think are related to Black Friday, and then you get a softer March quarter and then there’s potentially a less severe, if you like, version of that for the June quarter within the end of financial year sales and that sort of thing. Timing of school holidays can have an impact too. So, we are aware of that and also, of course, running through the consumer spending numbers at the moment and the timing of the end of electricity rebates and, sorry, this is a technical point, but when that happens in a national accounts measurement perspective, the payment for the consumption of electricity transfers back from government to households so that’s another additional challenge to work through at the moment.

In terms of the latest data for the September quarter, with all of that, we were actually expecting maybe the momentum might soften a bit, just from some of these technical adjustments really. So, not in a fundamental sense, but in an adjustment sense. The data hasn’t been too surprising for us. We’ve certainly not been surprised on the downside with it. It’s a bit choppy, as it always is. And that monthly spending indicator that we get from the ABS is still new and so we’re still learning about it. I think everybody is. But we certainly haven’t been this surprised on the downside by what we’ve seen, but obviously we’ll find out in a couple of months’ time how it plays out.

Paul.

Thanks, Sarah, for your comments. I just wanted to explore a little bit further the point about lags in monetary policy. One thing that we know already is that house prices have responded quite promptly to the easing so far and that presumably means that, sort of, down the track, and to this earlier - to the previous question, consumer spending is going to pick up so - through several channels. So just interested in your thoughts around how you’re thinking about this, sort of, transmission mechanism, particularly through the housing market.

Yes. So, the housing market, the established markets for house prices effectively, it’s the first and one of the most responsive parts of the domestic economy in terms of monetary policy transmission. We’re obviously tracking what’s happening in that market with house prices. It’s sort of in line with what we’ve seen in previous easing cycles – so we’re not surprised there’s been a pickup. We’ve seen that before and we understand why from a cyclical sense because lower interest rates increase people’s borrowing capacity and then that naturally flows through. So, we’re not surprised that that’s happening. We’re certainly tracking it. It’s not out of line with what we’ve seen in previous cycles but obviously there’s always a lot of other things running through that market at any given point in time and that’s certainly something we’re going to be focused on moving forward.

I think where you do get the sort of lag starting to materialise, in terms of economic activity and actual inflation outcomes, because obviously house prices are not in the inflation basket is in via, as you mentioned, consumption but also dwelling construction and that does take a bit of time, you know, if you’re going to build your own home. Even a detached home you’ve got to get the approvals; you’ve got to get the work underway and that takes a number of months. If we’re talking about high density housing construction, that can take years to, sort of, get it off the ground and fully make it through to completion. So that’s the kind of lag that I’m talking about, when I’m talking on a 12, 18, even 24-month horizon. But, no, we’re certainly monitoring it, and we’re not surprised that we’ve seen a pickup, and it’s certainly something that we’ll be paying attention to - how much of a pickup and how much continues from here, in terms of the outlook.

Thanks, Paul. We’ve probably got time for one more question. Over here.

Thanks, Dr. Hunter, for your speech. Stephanie from the ABC. Sorry to bring up the topic, but you mentioned uncertainty and obviously a major source of uncertainty is the US trade war and we’ve seen some reignition of tensions between China and the US in the recent week. How much is that factoring into the difficulty around forecasting? You know, you said your August forecasts were pretty good, but given this is now dragging on for a significant amount of time what challenges is that creating, and I guess what have you seen so far, in terms of the balance of risks around inflation and growth risks?

Yes, it’s certainly an unpredictable situation to be dealing with. We’re still in a phase seemingly where we wake up at the start of the day and there’s another fairly major announcement come through. So, my China team and my international team more broadly are getting a real workout at the moment. They’re working very hard on this, and it is pretty challenging. And the latest escalation is obviously still pretty new as well. So we are, as everyone is, still working it through. So, I don’t have definitive answers for you. We’ll obviously take a view when we come to November in our next forecast.

But generally speaking, what we’ve seen historically time and again is that uncertainty does drag on the outlook. It makes it very hard for businesses to make decisions, in particular around investment. It can be challenging for consumers as well, if you’re concerned that you might be at risk of losing your job or what have you then that can dampen your spending too. What we’ve seen in the data through to now, I think, is that there’s some of that may be coming through in the US, around their labour market data in particular. We’re seeing less evidence of that here in Australia. And, you know, we can ask in our liaison program, for instance, what decisions might be delayed because of uncertainty. We’re not really hearing too much of that at the moment. We’re not seeing too much of an impact in the business surveys, for instance. So - but we are very watchful of it and mindful of it. We’ve been calling it out in the last two SMP sets of forecasts, and we’ll certainly have more to say in November. Yes, it’s a very unpredictable situation and that does make it challenging.

Thank you, Sarah. We’re out of time unfortunately, but she spared the question I was going to ask on CBDC and stablecoins, so we’ll save that for another day. But please join me in thanking Sarah for her time here this morning.