Transcript of Question & Answer Session Additional Monetary Policy Tools: Reflections and a New Framework

So let me actually thank and welcome Michelle Lewis up to the stage. Michelle’s going to help moderate the session and was a key architect and author of much of the work that sits in behind the framework. And also, thanks for contributing to my speech. All errors were mine though. So, Michelle, I’m in your hands.

Thank you very much.

And please, looking forward to questions, including quite a few people online, I believe.

Yes. So we’ll – for those in the room, we’ll have two options. So one will be a microphone, which will be passed around. The other option is you can use Slido, and then those questions will be fed in to the rest of the audience who is online, who can also ask questions using Slido. But first of all, let’s start with questions in the room. Maybe I will start off while you gather your questions. So, the first question would be: this is quite a fulsome framework from both a process and a governance perspective. Are you concerned that it could slow the Monetary Policy Board down?

No, indeed, I think, I think if anything the opposite. The whole idea of the framework is to draw in a timely way all of the lessons which we’d already tried to draw on each of the tools that we had used during the pandemic, but do that more fulsomely and look at everyone’s experience and bring in this external, these external perspectives which were extremely helpful. And then get the Monetary Policy Board and the Governance Board familiar with the framework, the principles, the process that we’d go through, and start to even think a bit about some of the tools and their pluses and minuses. So I think, if anything, having this framework doesn’t slow the process. I think it will speed – sort of allow us to make timely decisions. And sometimes those decisions have to be made under considerable time pressure depending on the nature of whatever crisis we might face in the future.

Yes, there’s a microphone coming.

Hi, thanks both for your explanations there. Just a quick one on the framework here. Was there ever any discussion of entirely ruling out any of these options, or were they always all going to be on the table for future episodes? Particularly referring to the yield curve target, I suppose, and discussions there.

Yeah, look, I think starting back with the – so after the various tools that we used during the pandemic, after they’d sort of rolled off and we had a bit of time, but before too much time had passed, we did a review of each and every one of them. And then this framework sort of brings all of that together again and has a fresh look at it. And I think even at the time with that individual review, just of the yield target by itself, we gave a sense that it was problematic for the reasons that are outlined in the framework and that I outlined in my speech. But I just think you never want to entirely rule things out.

Too much forward guidance?

Well, it’s a bit – it’s more that you don’t need to tie your hands. And look, I think the yield target was an important and valuable part of the package, and we’ve got to be careful because hindsight is always perfect, but there were aspects where maybe the yield target could have been configured and used and been more effective than it had been at the time and less problematic in particular. It may have been that we didn’t have to – if you’d have your time again and you still had a yield target – you might not have rolled it forward as readily as we did, for example. So no, we don’t – we really don’t want to rule things out. And that’s why even I wanted to mention those other tools. You just wouldn’t want to rule them out.

You mentioned the reviewers earlier. Can you talk us through what their key contributions were?

Yeah, so look, they were really helpful. We had some sort of draft work and thinking together, and we provided it to them, and they came into the Bank, came to Australia indeed, and came down to Sydney and met with staff. They met with all the senior executives, including the governors for some time in February. And what they did was they tested the key issues that we put before them, all of our assumptions and our design choices.

I think the main thing they brought out was that fourth principle that I’ve sort of discussed, to think about broader public policies and how our use of additional tools might have an effect on other agencies and and how we would want to best take account of our effect on them and what they’re doing as well. I mean, those were always sort of there at the back of our minds, but I think they made us crystallise that a lot more clearly. So we added that consideration of policy settings across the public sector agencies. They also just wanted us to drill down and be a bit clearer about the role of uncertainty, the distinction between using tools for market dysfunction versus using tools over a longer period to really drive financial conditions into an easier place. I wanted us to talk about and think about scenarios, communication, and even the fire drills.

Great, I think we had a question.

Thank you so much for your remarks this morning. I wanted to ask you, in the section where you talk about the Term Funding facility, the fixed-term funding facility introduced in COVID, you’ve said that there is a variable rate facility that would involve less financial risk, that wasn’t available at the time, but is now. What do you mean? Just a bit more colour on what you mean by variable rate facility. Presume it’s fixed for some term, perhaps not 3 years, but just more colour.

So, you could do it a little bit like our OMO rates now, just over a longer period of time. So, when you bid at OMO – at Open Market Operations – on a Wednesday, you don’t know what rate you’ll be paying, you just know it’ll be paid out according to how markets behave over a period of time for which you have that. So you could have done that sort of thing in theory. In practice, we weren’t really ready for that operationally, nor were the markets. But at the same time, I think even if we had, I think the thinking was we’d never faced circumstances like this globally in modern times. We were facing a financial crisis of real substance.

It passed very quickly, but I think it passed because central banks and governments acted quite quickly. And central banks acted very quickly because, you know, we had the most critical market in the world economy – the US Treasury market – just grinding down to almost a halt, and other government bond markets were doing a similar thing. So, I think at the time we thought, ‘No, we have to go all out here. We have to provide a lot of support. That will come at some cost.’ But a fixed rate with a long fixed term was actually going to provide the sort of support the economy needed, particularly when we’d had years of undershooting the inflation target, I think that came into mind as well. So, if we’d been able to flick a switch and have variable-rate Term Funding Facility, I don’t know that we would have done it even at that time.

Thank you.

Oh, sorry, and then Joe’s got a question. We’ve got a few in the room, so that’s good. Slido might have to work harder, but Michelle, sorry, I should let you direct traffic.

Yeah, thank you, thank you, Mr. Kent. Probably a bit of a derivative of Phil’s question, but there’s a common theme across downsides to a lot of the tools here of the mark-to-market cost or the carry cost or all those sort of things. I mean, you know, in the long run, a central bank can obviously have negative equity and survive quite well, but that does turn up as a negative for all these policies. To what extent is that a limitation in what the RBA may do going forward? Has that concern changed due to that particular issue getting a little bit of an airing in a few sort of media and parliamentary circles over the past couple of years?

I think it’s not so much that. I think it’s more thinking about the decision-making process, the risk management process, and the governance process. So particularly now, it’s – we’ve focused our minds. We have a Governance Board that is responsible for managing the Bank’s risks. Now, the Monetary Policy Board has a different lens and a different need. It needs to focus and do what’s best for the economy, and it will do that. But in formulating those policies, if there are different options that are available to it, the Governance Board would direct it – not direct it, but like it to take the one that provides similar stimulus, for example, but has less cost to the Bank.

Now the challenge here, of course, is you often – say through a bond purchase program – the whole idea here is you’re taking risk off private balance sheets and putting them on the public balance sheet, which can wear those, in order to provide that sort of stimulus. So it necessarily implies the potential to make losses. But what the framework’s doing is just say, yes, but document those carefully, think through scenarios that could be more adverse for the balance sheet, and consider all of those things when drawing conclusions, putting a package together.

So it’s not ruling them out by any means. It’s just saying let’s follow a slightly more thorough and considered approach. And if we do that ahead of time, including in these fire drills, we’ll be better placed to make good decisions. I’m not saying the decisions at the time were the wrong ones; when the Reserve Bank Board made the decision to buy bonds at the time, they were aware of these risks. But I don’t think they considered quite the upside scenarios that you might want to now. Again, hindsight, right, it’s perfect. You now know you need to consider the possibility of a major incursion in Europe that forces oil prices and energy prices up dramatically, for example.

Good morning. I can see in the heat map in the framework for additional monetary policy tools at lower interest rates, so that’s Table 4, you can see that term lending facilities is almost unambiguously better or perhaps suitable relative to the other tools. Is it fair to say that that will sort of be the first choice of tool? And I guess an alternative question as well is, you know, are there circumstances where perhaps some of the other tools might be preferred ahead of term lending facilities?

Well, I think if – so as I tried to emphasise in my speech, the absolute most important tool is the cash rate target. So, that’s first and foremost. And possibly even doing that somewhat preemptively in order to avoid even being stuck at or getting to the effective lower bound and needing to use other tools. I think I wouldn’t be nearly as strong as what you’ve suggested. I think it does depend on the circumstances. I mean, it was a key plank of the initial package.

The initial package didn’t involve bond purchases, partly for the reasons I talked about, because at the time – and we’d done some reviews using some BIS studies which pulled together central bank experience during the GFC – at the time we didn’t think the bond purchase program was going to be as powerful, and it would potentially be more costly because of Australia’s – the nature of Australia’s financial system. Being very bank-based, something like a term lending facility can be very powerful. Because when you’re in a real crisis, you not only want to reduce the cost of funding, you want to make sure there is funding being provided to businesses and households. When you’re very bank-based, something that goes direct to the source like that can be quite effective. But I don’t think I’d be nearly as strong as you’ve suggested.

It’s a really great summary. I noticed though, each of these tools have been assessed individually, and in your speech you’ve talked a lot about the package, the interaction of these, with the cash rate being at the centre of that. Have you done work about which combination of these may be most effective in any given scenario?

I don’t think we have. I’ll just check with Michelle. We haven’t really got to that point.

Michelle Lewis: So, in the latter part of the framework, we do talk about – we need to consider tool interactions, and in the staff assessments, the more detailed assessment that’s been published alongside, we talk about in a bit more detail there. But I think that’s when the fire drills and scenario analysis, that’s when we’re really going to bring that to life.

And I think it can be quite hard because it’s quite hard to incorporate these in your standard macroeconomic models, for example, and I think it really will depend on the circumstances at the time. The one thing actually, and just going back to the previous one, but the two questions blend together quite well, the one thing I sort of forgot: now, this isn’t so much about a tool that you use when interest rates are very low, but the one thing we did very, very quickly is provide a lot of liquidity. We immediately altered our operations at the time, which were quite limited. They’re not like they are today. We essentially turned them to be very close to what we have today, almost like full allotment. And we did that quite live.

I remember it during one of the sessions where we decided we were going to up the amount we were offering banks on that morning dealing from numbers like $2 billion to $10 billion. And then we just saw the demand and we just live during the auction decided, right, we’re going to just provide everybody what they asked for at the price they asked for it. So it turned it into a full allotment auction. So, lots of liquidity came in, that’s a really important thing to marry with reducing, in a crisis, reducing the policy rate. I think the other thing we did is we did bond purchases. Now, some of them came through the yield target, but we did bond purchases – we started that process very early on to try and alleviate the stresses running through all of the markets that I talked about earlier when I was discussing with Phil. So, you know, I think it’s hard though to sort of know exactly what circumstances you’ll face and run those.

The challenge will be to modellers now and into the future to see if they can do that. But even then, I suspect that you’re going to sort of have to assess the circumstances at the time and think about the different sort of market conditions you’re facing. And in a bank-based system, if the banks are under stress and facing the prospect of not being able to raise funds internationally or even domestically in bond markets, then that’s why a term funding facility might be particularly useful, but it might be very different circumstances from that.

Okay, we’ll let some people online jump in. Chris, how often do you expect that these additional tools might be used if the neutral cash rate estimate is, say, around 3.6% and developed economy central banks are historically cut by around 400 basis points in a recession, there seems to be a fair chance that they will be needed.

Yeah, so I think it’s a good question, uh, not surprising really. Um, I mean, this has been an issue for some time. Indeed, it was an issue out of the Global Financial Crisis. So the issue is, if neutral rates are coming down, sort of this what rate would sort of sustain an economy at the inflation target at full employment is a lower rate over time, you’re that much closer to hitting something like an effective lower bound if you think negative rates aren’t particularly useful in your economy. So if that’s the case, I guess there’s a few things you could do.

One I suggested in the speech is you might be cognisant of that and be that and be very careful when there are disinflationary pressures and rates are already low, to be maybe a bit more reactive than you might otherwise be. Yes, you might feel the need to bring in some additional support from these other tools. It may just be the reality that we face though, if that’s the case. Two points. One, I think, is the wrong one. At the time, a former professor of mine, Stan Fischer, and a colleague of mine that I went to school with and was at the IMF were wondering, should you raise the inflation target? Because that raises your nominal neutral rate and takes you further away from the effective lower bound. I think that’s quite dangerous because if people think you can move your inflation target around, you’ll move it as it suits.

It does argue slightly in the favour of having 2.5 per cent versus 2 per cent. I think it’s a very strong argument not to change the target to 2 per cent in Australia. The other perspective is maybe there are other things that other branches of the public sector can do to try and boost productivity, and anything we do to boost productivity might help with the neutral rate; push that a bit higher. But that’s not something central banks can control.

Hi, Chris. So buying bonds basically left you with large paper losses and made you technically bankrupt. Have you considered ways you could buy bonds without such risks, maybe getting the Government to cover any losses?

I wouldn’t – I don’t think a central bank can become bankrupt, so I just challenge that premise so people don’t think I agree with you on that. So, I think what you’re really talking about is indemnities, and I think it’s a good question. I think it would be something that you would imagine if the Monetary Policy Board was deciding to pursue a significant bond purchase program, for example. Or something else that put the balance sheet at considerable risk.

On the one hand, they’re thinking, well, we do that because our assessment of the benefit for the macroeconomy far outweighs what risk there might be to the bank’s balance sheet. But it should still be something that they take into account explicitly, and I would imagine there’d be conversations with the government on the day. These things were considered during the pandemic. Would we go and ask for something like an indemnity at that time? And the decision was no, partly because we thought not asking strengthened our hand in terms of our longer-term central bank independence. It strengthened our hand in trying to say to the market and to everybody else, ‘We’re buying in the secondary market. It’s at arm’s length from the Government.’ And then the other experience is it’s varied with those who did have indemnities, the perceived value.

So in terms of the cost to the bank’s reputation and to the balance sheet, that was always revealed fairly quickly in our case because of the way we mark to market. And then we’ve moved on from that, and people have seen, well, the central bank’s been able to do everything it needs to do, even with negative equity. And of course there is a plan to rebuild that, and that’s in – that’s coming. But the one thing I would sort of note, if you have to go into the public domain and to your fiscal authorities every month and say, this is the bill for the month, then that’s not so great either. The most important thing, I think, is for everyone to recognise when you use these tools, they’re not without risk. And the risk is not only to the central bank balance sheet, but as I said multiple times in the speech, to the consolidated public sector balance sheet. So, I think as long as you’re very clear about that, and note and say, ‘Well, if we’re doing it, we still think it’s for the economy’s benefit,’ well, so be it. There’s a question over here unless you want to go to the other one.

Thank you, Chris, for the speech. I’m just curious, you talk about the risks to the consolidated balance sheet and you’ve talked a bit about the exit of some of these additional tools. Does the framework consider once the objectives have been met, any more active reduction in the RBA’s balance sheet, or is hiking the cash rate always going to be the preferred option?

It’s a good question. I think when we’ve looked at this in the past, and Michelle can correct me, I think the evidence is that the effect you have on yields and on market functioning, the transmission of monetary policy, by pursuing something like bond purchases, for example, tend to be strongest when the economy is already quite weak and struggling and needs that support. But then if you needed to really tighten policy, just starting and doing it through, say, quantitative active sales of bonds, an active QT program, I think that’s possibly going to be less effective in raising rates to the extent you need to. And that’s certainly the path we followed this time around, and I don’t And I think that’s probably worked about as well as we could have hoped.

And we’ve looked at that thoroughly over a period of time. Would we actively sell bonds? And we decided no, in our circumstances, that we didn’t think that that was necessary. It might be quite different circumstances if you were purchasing bonds not so much for financial conditions to make them easier under a bond purchase program, but to address market dysfunction. The general consensus is you say, ‘We’re going to come in and buy while the market needs support.’ But depending on the nature of that circumstance, if it’s not a broad economic crisis that’s going to last for a long time, the intention, I think, would be – and the Bank of England did this reasonably well, I think – to sell the bond shortly thereafter, which is not unlike a foreign exchange intervention, which we’ve done in the distant past.

In the speech you note that the Monetary Policy Board may act earlier and more decisively to deal with a disinflationary situation. Going back to the 2019–2020 period when the bank moved the cash rate down from an already low 1.5 per cent, How much lower could the bank move, and would it mean larger cuts?

I think I suggested in the speech, our analysis suggests while you cannot rule out negative rates in Australia, they’re probably not one of the first tools or even the tools you’d regularly consider. So going back to these circumstances, I think a bit the review of the RBA and looking back over a period even before the 2019-20 period, the question was you’d been under target for a while. Maybe the suggestion is, and again hindsight’s always perfect, you’ve been a bit more aggressive even leading up to the pandemic to raise inflation, and that might have been a better course. Now again, it’s easier to say that with hindsight, but that’s, that’s I think one of the lessons learned. Be careful getting too close to zero if you think it’s because of, you know, if inflation’s low and you can provide some more stimulus. But you have to consider the circumstances at the time, so I think it’s hard to say.

This framework is about monetary policy at low interest rates and the question is more about wondering if we’d also need additional tools at high nominal interest rates because inflation is elevated, and this is because the cash rate solely may not be effective. Has any thought been given to this?

I’m not sure if this is quite what they meaning, but I think there’s a general sense in ‘Can’t we do something else?’ Because the cash rate, when it’s high, hits really hard those who have a mortgage. And I think that’s natural in an economy like Australia where we have variable rate mortgages, and so the effect is quite instantaneous and the effect is quite obvious. What we’ve said in the past, though, is that that probably overstates the role of the cash flow channel, that is, you know, people who have a mortgage. Their cash flows are impacted, they have to pull in their spending.

It seems there are other channels that are equally, if not much more important, in Australia than the cash flow. The effect that interest rates have on dwelling investment, tangentially through business investment, through the Aussie dollar, and then the so-called intertemporal effect, where everybody has an incentive to consume less and save more when interest rates are high. So, I think my answer is, look, the cash rate’s a pretty effective tool when you’re trying to address high inflation, and monetary policy is the right approach. So, no, I wouldn’t be using other tools.

Does the RBA ever intend to do a sale of its AGS holdings?

Ah, well, they’re coming down reasonably quickly and aren’t particularly substantial now, particularly when you think about the sort of residual maturity of these bonds. Look, we looked at it long and hard, we considered it over a number of different board meetings, and we’ve come out with the decision earlier that passive QT, just letting them roll off, is sufficient for our needs. So, no obvious burning platform, I think, to sell bonds.

Okay, that brings the Q&A part to a close. Thank you everyone for joining us here today, and there’ll be a full transcript online later this afternoon.

Thanks for all your questions, thanks for coming along, and thanks for all those online.