Fiscal Policy for the Future seminar series: Strengthening Automatic Stabilisers in New Zealand – Direct Payments to Households

Dominick Stephens (00:00:45):

[speaking in te reo Māori 00:00:33] E ngā iwi, e ngā mana, e ngā hoa mahi. Tēnā koutou katoa. E mihi ana ki a koutou, kua tae mai ki te tautoko tēnei kaupapa. Ngā mihi nui ki a Claudia Sahm, te kaikōrero o te rā. Tēnā koe. Kō Dominick Stephens tōku ingoa. Nau mai, haere mai ki Te Tai Ōhanga.

So, I'm Dominick Stephens, Chief Economic Advisor at the Treasury, and it's my pleasure to welcome you all to the first of our newly themed Treasury Guest Lecture Series, Fiscal Policy for the Future. Very warm welcome to Claudia Sahm, our illustrious and fantastic opening speaker for this theme actually.

(00:01:19):

So just before I kick things off, just a reminder, we've got people online, we've got people in the room as well. For those in the room, quick health and safety briefing. In the event of a fire alarm, please follow Treasury staff to the nearest exit and progress down the stairwell to evacuate the building. Do not use the lifts. Claudia, we had a real-life experience of this recently, so we're well practised, so we know exactly what to do.

(00:01:41):

In the event of an earthquake, please drop, cover and hold, stay in the building until instructed by a warden. And bathrooms can be found just down that way to the left of the kitchenette. Walk straight ahead and then left after the door. So, as I say, this is the first in the new theme for the Treasury Guest Lecture Series. We've been theming these year by year. We had Productivity in a Changing World last year and a wellbeing theme the year before that.

(00:02:10):

So the Fiscal Policy for the Future theme, we're going to invite speakers to talk about how to keep fiscal policy sustainable over the long run, stabilising through the economic cycle, and to make sure that we get value for money over time. Fiscal policy's obviously absolutely core to the Treasury, so this is an apt theme for us, but it's worth noting that it's a third of the economy, so it's really, really important that we get this right for the economy.

(00:02:43):

And fiscal policy is it is special in the sense that it represents the decisions we make democratically as a people. So, the other two-thirds of the economy are decisions made in a market setting, these are decisions made in a democratic setting, which does make it distinct. And, in a sense, makes it something that I think is quite appropriate for us to converse on, in this way, for the coming year.

(00:03:09):

So New Zealand's fiscal policy framework is definitely really highly regarded. I went to a fiscal conference, an IMF conference in Japan, last year and ready to talk about the challenges that we’re faced, and just found that we were being held up as a paragon of fiscal virtue and excellence for the assembled countries there. And the issues that I, or the challenges that I presented to the group were dismissed as minor compared to the challenges that they faced.

(00:03:42):

New Zealand's fiscal policy, it's highly regarded, it is unusual. Public Finance Act really puts the emphasis on transparency as the thing that guides our fiscal policy. So, there are fewer numerical targets in our governance structure for fiscal policy or legal structure for fiscal policy, and a greater emphasis on transparency. And as I say, even though we are highly regarded, that's not at all to say that New Zealand's fiscal policy doesn't face challenges.

(00:04:13):

So, it faces a number of challenges. We have faced a series of one-off inverted comma events, because they just seem to be becoming frequent. The Global Financial Crisis, of course, we had the Christchurch, and then the Kaikōura earthquakes. We had a pandemic and then we had a rain event and cyclone, the North Island weather events, all of which were followed by fiscal responses.

(00:04:45):

Obviously, we all know that net debt is now higher than it once was. The ageing of the population is something that Treasury has done a fantastic job of warning about for years. But we're now, the future is here. The ageing is upon us, and the fiscal challenges associated with ageing are here now. And then the global situation has also become more challenging over time, arguably. We have a changing climate that is expected to continue changing and provokes all sorts of challenges, both in trying to prevent and adapt to climate change. We have a changing geopolitical structure, and a changing economy that works with that.

(00:05:32):

So, as I say, during this theme, we're going to invite speakers to talk on three broad areas, and if I can just go into each of them. Stabilisation first. So, it's important that, or macroeconomic stability is really important for living standards. And monetary policy is usually thought of as the thing that safeguards stability over the cycle. But there are important questions for fiscal policy.

(00:05:56):

I mean, at the very, very least, how do we ensure that fiscal policy is not pro-cyclical. Because there may be a tendency for us to tighten as we learn that things that are getting, that the economy is weakening and to loosen as the economy burns, which could potentially make fiscal policy pro-cyclical. So, avoiding that is just the basics.

(00:06:15):

But monetary policy, we've also learned that it can be constrained. It faces potentially a zero bound on nominal interest rates, which could constrain the effectiveness of fiscal policy and creates a prima facie case for fiscal policy to get involved in stabilisation. Some of the alternative tools that the Central Bank might use to deal with those constraints are, in effect, fiscal. The LSAPs, so bond buying has fiscal elements to the policy.

(00:06:47):

We've also learned, I think, during the pandemic that fiscal policy can be quite good at responding to particular events. It can be really specific. When you got a pandemic and you're asking people to stay home and not turn up at work, well, a wage subsidy is actually a very specific policy that is addressing that specific challenge that the economy faces. We learned it can be rapid and, probably most importantly, there seemed to be increasing expectations for fiscal policy to respond to these types of challenges.

(00:07:18):

So, I think there's a really important question of is there a role for fiscal policy in stabilising and helping out with macroeconomic stabilisation? And if so, what should that be? The challenge, of course, with using fiscal policy to stabilise is the asymmetry. It's really easy to loosen during a downturn and it's hard to tighten during an upturn.

(00:07:38):

And I always say, actually my favourite quote from that conference in Japan was an IMF fiscal guy and he said, "If you want to be a Keynesian, I want to see you being a Keynesian on the upside. I want to see you... I want to see you, thousand-dollar poll taxes as often as you recommend thousand-dollar payouts to people. I want to see symmetry," was his quote.

(00:08:02):

And that asymmetry of, or potential asymmetry of, fiscal policy's role in stabilisation, it's a challenge that can be overcome, but it has an overlap with the next of the three themes, which is sustainability. Because if you're running asymmetric counter-cyclical fiscal policy, you got a ratchet effect.

(00:08:19):

So, sustainability has an overlap with stabilisation. I've already mentioned the ageing population, which Treasury has often talked about, creating a challenge to the sustainability of fiscal policy, which means ensuring that our commitments could be financed over the longer term without significant change. There're political challenges with sustainability, because future generations don't vote yet.

(00:08:46):

And there are uncertainties with sustainability, such as uncertainty about what interest rates are going to be or what the climate's going to be like or the productivity growth and the like. And the last of the three areas that we want to look at is the structure of fiscal policy. So, this is about ensuring that it's giving us what we want.

(00:09:05):

Part of that is ensuring that we get the best value for money from what we spend. It's also ensuring that we collect the revenue in the best possible way, the most efficient way, in the way that's perhaps best for the economy. And that what we spend our money on is not just efficient, but is also effective, producing the outcomes that we want.

(00:09:26):

And, of course, there are overlaps. This good, well-structured fiscal policy that gets value for money, for example, will make it easier to make fiscal policy sustainable. There's also an overlap between structure and the stability angle. Because one of the ways that fiscal policy can be stabilising is through automatic stabilisers. And if the parts of your fiscal policy that are, that fulfil... A larger unemployment benefit, for example, will create larger automatic stabilisers and could help with stabilisation.

(00:10:01):

So, there are overlaps between all of these. They're going to be really interesting. I'm really looking forward to them. As I say, this is core to the Treasury. So next year, we're going to be releasing three stewardship documents, all of which relate to the fiscal theme, the Long-Term Fiscal Statement, which will be essentially measuring that sustainability challenge. The Long-Term Insights Briefing.

(00:10:23):

We're considering, we need to consult on the topic, of course, but we're considering a topic that will be related to stabilisation role for fiscal policy, or sustainable fiscal policy over the cycle. And the Investment Statement, which, of course, is about effectiveness and sustainability on the capital side of the government's fiscal policy. Okay, so that's how we're thinking about this theme.

(00:10:49):

Without further ado, I mentioned the automatic stabiliser role, or the stabilisation role, of fiscal policy and that's what Claudia is here to talk to us about today. So, I'm really looking forward to it. This is actually work that was commissioned by the Treasury, and Claudia's helped us out with, that considers whether there's scope to extend New Zealand's automatic stabilisers by pre-positioning interventions that would otherwise only be made on a discretionary basis during a recession, allowing a quick response and more time for policymakers to focus on unique aspects of the crisis.

(00:11:23):

So, as I say, directly related to what we'd like to talk about through this theme, and I'm really confident that this is going to produce fantastic conversation and insights, both Claudia's presentation and hopefully, we'll get a bit of conversation and questions at the end. So let me introduce Claudia who are thrilled to have with us today.

(00:11:47):

Claudia Sahm is a Washington-based expert on monetary and fiscal policy and forecasting. She has several years of experience advising decision-makers at the Federal Reserve, the White House, and Congress. Claudia created a widely used and highly accurate recession indicator, the Sahm Rule. She is the Chief Economist at New Century Advisors and founder of Sahm Consulting. Previously, she was a Section Chief at the Federal Reserve and a Senior Economist at the Council of Economic Advisors.

(00:12:16):

So, Claudia, thanks very much for joining us today. Just before get Claudia to the stage, just on the etiquette. We'll keep questions till the end, and if you can just put them into the Q&A function for those online, we'll do it in person and online, but put the questions in the Q&A function rather than in the chat, and we'll have them at the end. Thank you very much. Claudia, welcome and thanks for joining us. Cheers.

Claudia Sahm (00:12:48):

Okay, well first of all, I want to say thank you. I feel so honoured to get to kick off this lecture series, which is an extremely important one, not just in New Zealand, but as usual, you all are leading on the forefront of trying to do good policy, so I appreciate that. And also, I have gotten so much support and help from your colleagues at Treasury.

(00:13:10):

I have learned more about New Zealand since January than I thought I ever would, and I would continue to learn this week. So, I want very much for you to see today's talk, the ideas that I put out here as we're going to do with the Q&A conversation, but I'm not going anywhere. So, I want to keep learning and have this be something that can spark a conversation, thinking about automatic stabilisers here. Yes, so let's get into it.

(00:13:39):

Oh yeah, so as I said, I'm going to talk about strengthening automatic stabilisers. So, a particular, I'm going to give a concrete proposal just to help us form our thinking. And yet, it's important to start at the beginning. I think Dominick laid this out as well. What are we trying to do with fiscal policy, particularly when it's in times of a recession? Steady, sustainable growth, it's good for people, it's good for businesses in terms of planning.

(00:14:06):

We also know that recessions that are very severe and prolonged can be scarring to workers, not just at the time it's happening, but for years to come. So, there's definitely a opportunity and a responsibility, in terms of fiscal policy, to think of economic stabilisation. While I worked at the Federal Reserve for over a decade, I am not in the camp of let the Fed take care of everything or let the New Zealand Reserve Bank take care of everything. So, I think fiscal policy absolutely has a role and it can be a very powerful one.

(00:14:41):

Okay, but again, we got all kinds of tools. I'm going to talk to you about automatic stabilisers, going to make a case that there's room to use more of them. But there's a mix. You've got monetary policy, you've got fiscal policy, you have discretionary, which is going to be ad hoc, in the moment. And then you have automatic stabilisers that will be pre-set. So, before you need to use them, you're going to make a commitment as to what they look like.

(00:15:06):

There are many reason... I'll go more into why the automatic stabilisers have good virtues to them. You shouldn't necessarily think about, I'm trying to convince you to do more. Often, this would be a case of doing better. Because there are some complications of discretionary policy. In the crisis, everyone's a Keynesian in a foxhole. Not everyone is the Keynesian on the good day. So, I hear you on that. And that can lead to decisions that either aren't well-thought out, that are just driven by a political impetus and sometimes aren't able to be administered well. So, you can set all those conversations ahead of time and that can be a very powerful way to do the fiscal policy in a potentially more effective way.

(00:15:53):

And again, the key premise of this, and I'm going to give a very concrete proposal of this, is that these policies are driven by economic conditions. So, what's happening in the economy, signs of going into recession, turn it on. You can have set up so that they then phase out. And even, I'm not going to give a concrete example of this, use economic conditions to trigger on when it's time to pay back. Because these often, again, it's all trying to reduce some of the politics and improve the design and make these pre-commitments.

(00:16:30):

Okay, so just briefly, this is such a moment to have this conversation, particularly when we think about the stabilisation aspects of fiscal policy. COVID was a moment of massively active fiscal policy. I was actually quite surprised and maybe not so surprised. I mean, New Zealand, the discretionary policy that was done in New Zealand was second only to that which was done in the United States.

(00:16:59):

I helped advise Congress through everything from the Cares Act through the Rescue plan. And I knew we were going big. You all did too. And I think the crisis was large in nature. I don't want, I think it's good to focus the mind. We did a lot of fiscal policy. It was very powerful. It could be very targeted to the different parts of the crisis that unfolded. This is not to say that everything that we did in COVID is going to apply to every other kind of recession.

(00:17:31):

And in fact, the automatic stabilisers... I'm going to talk to, we want to look at what generally happens in recessions. Because these are not going to be the pinpointed targeted. We can take on some of the discretionary policy, but it's always going to be with us. But, in general, the last four years have absolutely elevated we need to think more about the successes, the failures, and especially understanding, as we came to the other side, you have deficits, government debt to GDP ratios that are far higher than we went into it. The United States would still envy New Zealand for their levels, but it is something just underscores the benefits and the drawbacks of activist fiscal policy.

(00:18:14):

Okay, so the roadmap for today, so I'm going to go a little more into detail about why automatic stabilisers, more automatic stabilisers would be beneficial. Some basic design principles that one should keep in mind. I'm going to give a case study, looking at automatic direct payments to individuals. This is an example, it's a case study. It's also one that I think has merits on evidence of it, but I want to make sure that we can see how principles laid out can then be applied in a very concrete setting. And then, of course, there are challenges. So, I'll lay those out, too, because, yeah.

(00:18:55):

All right, so the first selling point for automatic stabilisers is the one that New Zealand have already are very effective. They are more passive programmes. Like progressive income taxes, unemployment jobless benefits, these are ones, that as a recession comes in and people's income goes down, the job losses have those programmes ramp up. So, these are ones that, but they're always running in the background. You always have progressive income taxes. They have other benefits, like meeting distribution.

(00:19:33):

And again, coming out of the democratic process what you think distribution would be. They also do have stabilisation properties. And as Andrew Binning, who's here and has so much excellent research on the automatic stabilisers using models, I mean these are big. If you would take all of the automatic stabilisers away from New Zealand, this would be a substantial increase in how volatile the output of the country is. And that, again, is a stress on people. Ben also has research just showing income. So, if you put it in a household setting, these are important programmes. Now, just and... Okay, I will to the next thing.

(00:20:13):

Now, all that said, as a small open economy, New Zealand is subject to shocks that come from the outside world. Yesterday, I joked that I've basically learned that the United States causes all your problems. And so, in terms of recessions coming in from outside, and just a small country. So, you can see even though there's a very successful and really in line with OECD, maybe a little bit more effective automatic stabilisers than the OECD average, you still have a really volatile output.

(00:20:50):

So, this is just showing, and the United States is somewhat below the average for the OECD in this one measure of how responsive government spending is to output changes. And yet, the volatility is half. These are totally not comparable economic setups, but it does show that even though you have effective automatic stabilisers, there still is a lot of volatility. Now, that doesn't mean that you just dial up all of the programmes that you have in place. What I'm going to argue for is a much more targeted, in moments of particularly severe shortfalls that you have these additional stabilisers.

(00:21:32):

Yeah, so the stabilisers that I'm going to talk about, I'll refer to them as automatic stabilisers through this. They're also referred to as quasi or semi-automatic stabilisers. I think the distinction is they don't always run in the background. So, you're going to have a trigger that under certain economic conditions, a programme starts up and then shuts off. Unlike you've always got the progressive taxes.

(00:21:53):

But just for simplicity, I'm going to just say automatic stabiliser, because they do have the principle of when that support happens or doesn't, depends on economic conditions. It does not depend on political decisions in the moment. And that's really the key distinction is it doesn't have the discretion. Now, what I said is I'm not just saying more but better. And yet, automatic stabilisers cannot replace the discretion.

(00:22:21):

Just like any central banker would tell you, no, I do not want to follow a monetary policy rule, because there are situations that need discretion. That will absolutely be the case for any kind of fiscal policy. So, I think there's a good case, and I'll make, that you could have automatic stabilisers that would be appropriate in the great recession and appropriate during COVID, but in no way, shape, or form could you develop something automatically that would deal with the very different contingencies of those two recessions. But you can get a start with it.

(00:22:57):

And while I'm not going to get into it, doing pre-commitments and setting the policy ahead of time can get you, it can also, like I said before, serve the fiscal prudence. You can, again, tie it to conditions, and this is where I know Treasury has done work and will continue to do work on ways to have the fiscal sustainability. And those are ones within these kind of programmes, specifically ones that are, they turn on in a, say, recession and turn off. These would be ones that also in the beginning to have that, make those commitments. Again, this is just a piece of the fiscal policy, but it is one where you could pre-commit, which is hard to post-commit. Her benefits.

(00:23:43):

Okay, so I'm going to... Yeah, so the first thing, whenever you, I'm working towards a proposal, the first thing you do is define the problem. Where is it that we want to step in and have extra support to the economy? Again, I'm not talking about just increasing all of the progressivity of the taxes. I'm very pinpointed, when would you have another intervention automatically? And so, here even the New Zealand economy is being volatile, trying to figure out what are the kinds of contractions that are broad-based and the kind of contractions policymakers would want to step in to address.

(00:24:23):

So basically, I'm using an algorithm from Hall and McDermott that separates out what would be considered recessions in a sense of a broad-based contraction. So, a real classical business cycle and which ones are technical. So, the grey ones are the ones that I'm going to focus policy on. These are ones you'd want to support, try to address the shortfall in demand. And the blue ones are more of these technical recessions. So, you have your two-quarter decline. Otherwise... But that's what we're aiming at. But again, you got to start with, okay, what would you have wanted to support the economy? And so that's my starting point for this.

(00:25:07):

Here's the thing about the automatic stabiliser, with the principle. And I'm going to get to an example. And I very much in kinds of conversation I've had already, you all tell me there's something better than what I picked. Because what you want to do is find a stabiliser that has been done on a discretionary basis, regularly, or maybe considered regularly, and then take a discretionary policy that you have some experience with, because you're going to put this on autopilot, and then move it into an automatic stabiliser capacity.

(00:25:39):

And so, then it does need to broadly apply. And the broadly apply is you have a broad-based contraction; you have a notable shortfall in aggregate demand. So, you're thinking of what are the programmes that would address that? And just knowing this is not a one-size-fits-all. It also, these programmes, the idea is to get them going as quickly as possible. So potentially, they can move faster than other discretionary fiscal policy.

(00:26:05):

They certainly, in many cases, depending on how quickly the contraction is assessed, they can move faster than monetary policy, which acts with some lag. So, you can have these as a frontline to get things going. But it does need to have a basic, it needs to be more plain vanilla in its effects than something very targeted.

(00:26:31):

And it's beneficial, something that's been used before, something that has research, something that has, we understand the political economy of it. So, all of these details can be hashed out ahead of time. And you do have to be careful with, and I mean this is one of the challenges that I'll bring up at the end.

(00:26:48):

There are many different tools of automatic stabilisation and one, really thinking hard about how fiscal and monetary policy are interacting. You do not want fiscal policy, particularly on autopilot, to get in the way of monetary policy stabilisation. And yet, you wanted these to be done independently. So, they're a delicate balancing act, but important.

(00:27:12):

Okay, so the trigger, and I'm going to go through the trigger that I developed for New Zealand's economy. These are principles that are really important... because the trigger, what turns it on, what economic conditions, this is at the core of the stabilise. If you can't come up with a good trigger, then this is not a good path to go down, so it needs to be accurate.

(00:27:34):

As much as I'm going to talk about direct payments, as much as individuals with hundreds of millions, billions of dollars coming out, at any point in time, the government will want that to only be happening in a recession. You don't want to have those false positives. Whereas, if you, if it doesn't turn on and it is a contraction, or it develops into one, discretionary policy can step in, but it's very hard to claw things back. So, accuracy is important.

(00:28:06):

Then it should be timely. We talked about before, these automatic stabilisers, they can be very good as a frontline. They can potentially be front loaded in their effects, so that then there's space, and then discretionary policy that once you have a better sense of the recession can step in. Okay, so this next piece about they need to be simple and easy to understand.

(00:28:32):

So, I worked on a proposal in Congress, and also when I developed one for, just a general one in research, getting feedback from people that work in the policymaking. You do have to be able to write this down and you have to be able to sell it. So, recession indicator that I have for the United States looks at changes in the unemployment rate. The unemployment rate is, people know what it is. It's a widely followed statistic. It's not hard to explain what it is.

(00:29:07):

It also, at its core, is why we fight recessions. That is a sign of the hardship that recessions cause. That doesn't mean, and I've never, despite it being used widely to think about recessions right now, the Sahm Rule is not, I didn't optimise it to be the best forecast over recession or the best we're in a recession.

(00:29:31):

There are other, and in New Zealand, there's more and more of the other statistics, like factor models, that bring together all kinds of information or diffusion indices which are simpler, but still are abstract. Those are great if you're doing monetary policy, or is it informing? You wouldn't want to rule based on it. But they're not transparent, they're hard to understand.

(00:29:52):

So, you do have, throughout all of this, there's a balance to get things into the real world. There's a balance between what would be most elegant, theoretically, have the economic theory behind it, and then what would actually work. Because the goal is to think about a concrete.

(00:30:08):

The data should be high quality, because the trigger's all about the data. You're looking at patterns, but you have to put good data in or you're not going to get good indicator out. And I'm very much in the camp of, it should be official statistics, something that has government oversight, something that has people, that's what they focus on.

(00:30:30):

It's not the private sector. Which does then relate to the last piece. No one should be able to influence the trigger. So, any kind of, if you feed something into the trigger that people, someone has very clear control over, then you no longer have an automatic stabilise. You have someone who's made a decision.

(00:30:51):

There's been a discussion, I think the research and logic backs us up, that when the Central Bank is constrained, so their policy interest rate is at zero, this would be a time where any fiscal action, whether it's automatic or discretionary, would have a bigger bang for the buck and we would maybe even need it. Okay, but there are a group of individuals who decide what the official cash rate is. So, you can't, to directly tie it onto that would not be consonant with the idea of the automatic stabiliser.

(00:31:23):

Now, remember this, automatic stabilisers can't do it all, right? So, to have a discretionary policy that reacts to that fact, which it would, like what the environment is, is perfectly fine. But you have to be very careful about what you base these triggers on, because it can't be influenced. So really, like I said, the core is getting a trigger.

(00:31:47):

I already talked about this before. So, I won't spend too much time. The time you spend ahead of time getting a good design that people agree on, let some economists in there, we won't agree on anything. And then let the people in that have common sense and we'll come up with something. But there, you can have that discussion ahead of time, because as said before, fiscal policy, it comes out of the political and democratic process.

(00:32:11):

There are a lot of pieces of this. Those conversations are impossible to have in a thoughtful way, in the middle of the recession, in the middle of a crisis. And another piece to it that I think is really important is the preparation, getting these enacted ahead of time. Once you've made a commitment we're going to do these policies, well, if you have made a commitment as a government to do these policies, then you need to make sure you have the systems in place to actually do them.

(00:32:39):

And certainly, I've met some of the heroes today of administering programmes during COVID and I know many in the United States. This is, when you put programmes together on the fly, it is very stressful on the administrative systems. You all, in many cases, have very good ones. But it still is, you can be more thoughtful and be prepared if you're set up ahead of time.

(00:33:05):

And even the best administrative systems will have failures. I mean, in looking at the direct payments and reading up some on the cost-of-living payments, there was some bad press on the administration of those programmes, in terms of not getting exactly to who it should have or getting payments to those who should not have.

(00:33:25):

And these are things where you can clean up, if you're prepared ahead of time, you can clean up matching of border records with income records. You can do that clean up so you'll never have no errors, but you can reduce them. But again, it's asking a lot to try and figure that out in the moment.

(00:33:48):

Okay. So now we'll get to a concrete proposal. I lined out the principles. As I said before, I want to be clear, this is both a case study. I want to not just give you the high level, "Here's what you should think about," I want to actually walk through what would be an example. I think there's a lot of merits to direct payments, in terms of them being used in recessions. And yet, there are other options that you can make a very good case, and this is actually what I worked on, with Senator Bennett's office, of automatically increasing the amount of the jobless benefits for a period of time.

(00:34:30):

And then when you come up, turning them back off. So not just always having more generous unemployment benefits but just periodically. So again, and those are just two that have been used several times in the United States. So again, I'm looking at this principle of things we know well, and we've used, direct payments are not, I mean there are plenty of other countries, and the cost of living payment.

(00:34:54):

New Zealand, Australia has done in recessions, direct payments and other countries. But there's nothing, I don't want to be the ugly American and pretend that what our policies are, what your policies should be. So it could be that there's a very clear alternate approach to this.

(00:35:18):

Okay, so I will walk through, so we can do, be concrete. Okay, so the policy is to do a one-time, lump-sum payment to the eligible individuals. And you do this as soon as the trigger happens, you send it out. So, this is an early in-recession, early as possible, in-recession payment. And these are big. So, in this type of a proposal, what I am, so 1% of GDP, dividing it across the eligible individuals.

(00:35:47):

And the benefit of, say, a direct payment versus topping off jobless benefits is you front-load that money. So, you could take the same amount of money, the 1% of GDP, and spread it out, say, over two years of extra jobless benefits. But if you put it all at once, you're front-loading it and you're getting in ahead of what would be discretionary policy.

(00:36:14):

Now, this is where it becomes very important, that it has to be calibrated in size, and I'll get to why choosing 1% in a minute, and it needs to be general enough that early in a recession, you would want this kind of push, regardless of where... I mean, you want the criteria that you're going into a severe contraction or a notable contraction, but you're not going to have all the details of exactly how it looks.

(00:36:40):

Okay, so the trigger is... Actually, I left, that's funny. I never defined what the eligible individuals are. I'll come back to this. But the proposal was to send it to individuals who are in the bottom 80% by income. So, I use a rather wide, broad targeting. And so, this is certainly, and I'll, actually I'm going to get to some of the research and debate on this, but you can think about what is, is this 60%, 80%?

(00:37:11):

And again, this trade-off between the breadth and the size versus the targeting is often when you think about what's the value of stimulus versus relief. Anytime you give money out in a recession, you're doing both things. But again, that's a value judgement as to what. And targeted relief can miss people, because we're not super, we're not that precise at targeting.

(00:37:36):

Okay, so the trigger, and going back to this, the identified the problem, the recession. So, the trigger to send the payment is you see a notable, so a half a percent decline, in real consumption, in a quarter. And that is paired with an increase in the unemployment rate, relative to its low, over the prior four quarters, which is like a quarterly modified Sahm Rule.

(00:38:01):

But again, because you have to have, I needed to have a multi-pronged trigger, and I tried lots of different data for this. Because the economy is very volatile. It is harder in the New Zealand economy, and also the institutions are different in the labour market to identify what's really these severe contractions that you want to see, that you'll act on.

(00:38:26):

Now, and this is something, when I've thought about these, before COVID, had not thought much about is inflation. So, there is a test in here that if I use core inflation, and debate which CPI measure you'd want to, if it's outside of the band for the Reserve Bank, then that would stop the payments, even when the trigger goes on. And so, this does recognise the fact that you have to, these type of automatic stabilisers, anything that would be general like this, we're going to be addressing a shortfall in demand.

(00:39:04):

And there are times, like in COVID, where we saw at least part of the contraction came from a shortfall in supply. So, you have to be careful in... This is not a... We can't measure potential output in the economy precisely, but at least knowing that you don't want fiscal policy to be just blatantly at odds with monetary policy that's trying to rein inflation in.

(00:39:29):

Okay, so I'm going to go through justifications. As I said before, these kind of tools, particularly the direct payment, this is about supporting aggregate demand. And we're looking across these lines and, sorry, these are going to be kind of small. These are the five different recessions. The ones, I mean, can guess the purple is covid and the, let's see, the orange is the '90s recession, that was the next, and the blue is the 2008.

(00:40:02):

The one thing that this shows you is, so I'm going to get to in a minute. The trigger shows will come on early in these recessions. You'll notice that all of them have a, wait a minute, it's hard to see here, but the cumulative decrease in, to aggregate demand from the peak to the trough is notable across all these, except the Asian financial crisis, which was a very shallow, at least in terms of output and spending.

(00:40:32):

But as you notice, as time goes on, they look very different. And this is just in the aggregate statistics. This isn't knowing about what the different causes are. So, this, again, is the automatic stabiliser can be very beneficial, both early on as trying to get in, soften the blow, some hopefully work towards a less, a shorter recession, but they're not going to be able to anticipate all of these different patterns, like the severity. Yeah, and... there's different causes under the hood. So, they can't do it all, but they can address a very common component across the recession, and then let policymakers focus on what is novel.

(00:41:11):

These all with, you see the COVID is very unique. And you also see the unemployment rate is rising. So, these are the two pieces I used were the declines in consumption and the increase in the unemployment rate. Again, these are also, it's not just statistical data mining, these are also measures that are very connected to why stabilisation policy is so important. These go very back to the effects, the wellbeing on households, wellbeing on workers.

(00:41:49):

Okay. See we're doing on time. In terms of the direct payments, and again this is to give a concrete example, I don't want to push this too hard, because I definitely have gotten lots of feedback on other options. It might be a more appropriate for New Zealand. The first thing is we have considerable experience, that means there's research. I will attest much of the research that has come out of this, that I'm familiar with, comes out of the United States.

(00:42:23):

We've used these policies multiple times and have some very good ways of trying to estimate their effects, just because of some administrative quirks in how the payments went out. But they have been used in a wide range of countries. They're also really simple, right? This is, you're just, you're giving a lump sum to people. It gives transparency, both to policymakers, whether it's doing fiscal or monetary policy. You have this set up, you know where it's coming, it's fairly easy to understand and to potentially model.

(00:43:02):

And it's also, I mean, people understand. There is a backstop or there is a first line of defence that's going to help people potentially. And we haven't had these kind of programmes. I don't want to push out over my skis, but an important part early in a recession is the increase in the uncertainty. I might not have lost my job yet, but I'm worried I might lose my job and you pull back.

(00:43:27):

So, knowing that the government is here to help in a real way, it can give some, both transparency and potentially we know that that first line of defence is there. And again, I talked about their other alternatives I think are also... I mean, increasing unemployment benefits is pretty straightforward. It's one that the political economy of it is harder in the United States, and maybe it is in New Zealand just because of stigma about welfare state.

(00:44:02):

And then, as I said before, I am in favour of, and the paper, when it is finished, will have the literature and in a more formal sense, but the... Okay, so in a perfect world, if I knew exactly who needed the money, who was in a place that something in the recession that contraction had caused them to not be able to spend what they had before. I mean, that on its own would just give you the stimulus.

(00:44:31):

We want to get people back on track or it puts them in dire straits, right? So, it's two levels of targeting. Targeting to maximise stimulus and targeting to maximise a relief. Okay, if we could do those things, that would make sense and that would be highly effective. Now we do not have on either dimension everything we would need to do that, particularly on the stimulus.

(00:44:54):

Now there is a long-standing view that lower-income individuals are more likely to spend. You got multiple rounds of effects on stimulus, you have to get people to spend, and then you have the multiplier, in the sense that you don't crowd out other spending or it goes overseas and imports. So, there's the idea, you give it to lower-income people, they're more likely to spend.

(00:45:14):

In fact, the research, particularly in the last 15 years, has been able to do this in a more, looking at cause-and-effect way says income is really unstable as a predictor. That what you want to know is liquidity, the cash on hand. And those people can be high-income. I mean, in the United States, there are, this category is notable, and I think New Zealand would have, I don't have the data on it, but would have the background for it.

(00:45:40):

So, you can think of people who are house rich and cash poor, all of their assets, they're kind of living paycheck to paycheck, even though they have a big asset that they just can't tap in a recession. So those people, you may not think of them as needing relief, but if you gave them more money, it would have a kick in terms of spending. But New Zealand, and frankly I don't know of governments that do, do not have records of the liquid assets, the easily accessible assets of different households. Wealth data just isn't collected in that sense.

(00:46:15):

So, you're left with income and if you wanted to push on a, we want these payments based on need, then I think income, completely intuitive, and yet the more narrow you target it, the less you're going to miss out on some of the stimulus effects. And you're almost certainly going to miss out on people who really do need that money. Because recessions are a time that's very disruptive in people's income, not just job loss, but just circumstances will change in a way that's very hard to track that and is a totally subjective decision on what need is.

(00:46:51):

But again, you can have all those ahead of time. But I, whoops, my read of the research and what I think are priorities in the fiscal policy have always leaned towards looser targets or looser targeting. But that's... I think with all this, you can see the translation from the principles into a concrete plan. There's research and evidence that goes with it and there's also a lot of subjective judgement. So, I'm just trying to flesh out the process.

(00:47:21):

Okay. I already said that who needs it? And there is... While all of this comes with the level of subjectivity, the idea, not just research can't answer everything, because we just don't have everything we need to know. But you do want to do some quantitative calibration exercises to the point you can. One percent of GDP in payments, that's big, although you did 22% of discretionary payments. So that's a large number. But if you look just in the period from the peak to the trough, the average cumulative decline in GDP was 5%.

(00:48:07):

So, if you're thinking about the whole, and that doesn't even count in the recovery... Often in a period, you've got output running below the pre-recession peak. And, of course, that's completely setting aside whatever growth path we should've been on without the recession. So, recessions are, they're big, they're bad. So, it was trying to think... But here, it's more of a how much I think do you want to swap out of discretionary policy into an automatic programme like this? That would, you know. But I don't think it's too large relative to the problem. It's just 1% of GDP is good bit of money.

(00:48:49):

Okay, so this is just showing, I don't know if this is justification, this is just showing you the two pieces of the trigger. I think the left piece, which is the quarterly percent change in real consumption, this kind of illustrates what I was up against here, in terms of coming up with a trigger that actually only took hold in these times that we'd identified as broad-based contractions.

(00:49:13):

Because there's a lot of quarter-to-quarter volatility in spending. And yet, this idea of using a half a percent quarterly decline, substantial decline, that was able to get into a space where, oh, this isn't just noise bouncing back and forth, this is something that really signifies the shortfall in demand.

(00:49:36):

And then pairing that with what's, is the unemployment rate increasing? Those two pieces together, because I'm trying to identify where is there a notable shortfall in aggregate demand? Because that's where these payments would be, they're fit to the purpose of solving that problem, that piece. So, I've got both of these components and I'm going to show you, put them together and they're pretty good.

(00:50:04):

Okay, you should see the spreadsheet that I have worked through all of this on. Because, again, it's not, well, yeah, not going to take you through that. But I do, where we ended with this, and this was also, like I said, I think it's important to me that when you look at this trigger, it does reflect the cost of wellbeing that a recession causes. So, it has, it tells a story, too, it's not just some random data series that correlates well.

(00:50:36):

And what you can see here, so the top side of the table, so these are the five recessions that we identified, and the two columns are showing you the decline in consumption and then the increase in the unemployment rate that went along with the period in which the trigger went into effect. And so, that's the third column of numbers. And the one that I've highlighted in blue, the Asian financial crisis of those five is the one where it did not trigger.

(00:51:10):

Now this isn't, it was just, it was very shallow in terms of output, in terms of consumption. So, it's not to say that this wasn't a crisis or could be deservant of discretionary policy, it's just meeting that metric of it looks like we have this aggregate shortfall in demand that we want to step in automatically. It just didn't make the cut. Oh, the bottom set of panels, these are four of the technical recessions. So, two quarter declines that we've seen. Oh, this starts in 1978.

(00:51:48):

And there are two times in there that, oh, I should back up. Okay, staying in the top panel. If you look, the next, after the column that has the blue, the next two columns are the total CPI quarterly change and the core CPI quarterly change. And that's our, you have the trigger down the centre, and we know when it turns on, because it's met the consumption and the unemployment criteria.

(00:52:16):

And then even if the trigger has gone on, if they don't, if inflation is above the target, the payments don't go out. So, this went into effect in the late '80s, one that was the case. And so that's the one time where the inflation and the five contractions would've shut it off.

(00:52:40):

And then you had the Asian crisis, it wouldn't have gone out, the payments wouldn’t have gone out, because it wasn't, didn't meet the demand criteria. And then if you look down, and these are the false, whoops, these are the false positives that I'm trying to avoid, and there are two times where the trigger turns on in these technical recessions. Those are also cases where the inflation is high enough that it would stop the payments.

(00:53:09):

Now, these are... All these exercises are expos, going back and looking at data and trying to see what would work. These are empirical patterns. And yet, there are, for instance, now it probably should not be surprising that as inflation is high and the Reserve Bank has increased interest rates, and you see demand coming down.

(00:53:30):

But this is, again, where it look, that inflation test is supposed to help us differentiate between, either there's a supply constraint or there's a different policy goal that's being put in place. You don't want the fiscal policy to go at odd. So that one is probably not that surprising that it triggers. And yet, the demand, the consumption's gone down and employment is up. That's kind of unfortunately by design of monetary policy to get inflation down.

(00:53:58):

Okay, so there's some open questions here. Oh, can see these clearly. So, I said before, I had a little bit of bait and switch on this, that a feature of the automatic stabiliser you could define ahead of time, pre-commit to the offsetting revenues. So, this is beyond the scope of the simple example, and as was mentioned before with Dominick, Treasury is doing more work on fiscal sustainability. Those are a place to then take principles and apply specifics. Not up to speed enough on the New Zealand tax system to do this, but it absolutely can fit in this schema.

(00:54:38):

And then there's questions about exactly what's the right CPI. This is why we're having discussions ahead of time, it's not like you're having coordination across fiscal monetary policy, but you can have a conversation about, are we creating... that are working at cross purposes? Right? Personally, I think taking out food and energy or just looking at domestic inflation would make a lot of sense, as opposed to all these external shocks. But it wouldn't have made much difference in terms of when these turned on.

(00:55:07):

Okay, so just briefly, because I want to make sure we have time for questions. I think this came up a few time, you could hear this coming up time and again, tailoring to the specifics of the crisis is hard. Yes, it would be, these kind of programmes are most effective when we're at the zero lower bound. Yes, these are most effective when there is a substantial shortfall in demand, relative to supply, because you don't have supply constraints.

(00:55:32):

It is very hard to either meet the, we want an independent trigger criteria or we have to actually be able to measure this in the real world. So, it's trying to figure out what you can do to get as close to that. That's what I try to do with the inflation trigger. Yeah, so that, and... Oh, and... the trigger, and I've had to remind people, if this were to get popular, the New Zealand, do not call it the Sahm Rule. I hate the Sahm Rule. It's been really... These are empirical regularities.

(00:56:10):

I mean, that is one of the risks with tying policy to them is, because things can change in the dynamics of the economy, the institutions, and then they may not be as reliable as they were before. Now I think there are, you look for something that looks highly reliable, it makes sense with the concept of what the recession is, and you can go forward with it, but these aren't laws of nature. So, it's just being careful about how, and like, again, you would never want to tie all of your policy to this, to any kind of automatic programme.

(00:56:48):

Yeah, so we talked about those. All right, so I suspect I will not have many people push back on this. The better the data are for the trigger, the better the trigger is. And the data constraints in New Zealand are not trivial. I mean, in terms of the breadth of data, I think is official statistics... I don't want to push back on that, it's just the timeliness is pretty lacking or lagging.

(00:57:17):

So given the trigger I used and all this quarterly data, and with needing the official statistics, so the consumption data, if a recession started in January, it would be six months later before you'd even get a chance of seeing it in the data. That's a long time. But even the trigger they use in the United States, it's typically maybe three months into a recession. So, nobody's going to know unless you have extremely high frequency that we're there.

(00:57:46):

Amazon chief economist told me they could see the Great Recession in real time. I was like, you could've picked up the phone. I was like, "We would've..." Anyway, but the things not just with the timing, the reliability in terms of data revisions, if you're going to do this really carefully, you need to calibrate triggers off of real-time data. And I get very nervous about national accounts data, in general, in real time. It's very hard to make this incomplete data.

(00:58:14):

New, more timely data, I think you all would like that for lots of reasons. These are an alternative, but it gets very hard then to test the reliability looking back over time. Okay, I'll skip by this. The last thing, and I won't go through all the points, we want as much as possible these, the specifics, the size, the targeting, the stabilisation effects. You'd really want those to be done in a rigorous model bait.

(00:58:46):

And having worked, Andrew has been a saint for trying to run these programmes through, it's really difficult. These are not, for just a lot of different reasons, these added-on programmes, the benefits and the macro model sense are pretty modest. And yet, it's very hard to make those compare. How do you judge what it actually is adding? Okay, thank you.

Dominick Stephens (00:59:20):

Thank you very much, Claudia. That was absolutely fabulous. We've got time now for some Q&A. I'll just start by saying there were three things I loved about that presentation. First of all, you, the mix of theory and implementation is something that we don't see that often. Just the implementation practicalities that you've covered, along with the theory side, was great. I loved the specificity of your proposal, and I'd like to acknowledge the bravery there of doing that actually, because it's much easier-

Claudia Sahm (00:59:56):

[inaudible 00:59:57]-

Dominick Stephens (00:59:59):

Well, yeah, it's really good, because it gives us something to sink our teeth into and go, "Okay, yes." Yeah, it's much easier to build a barn than it is to either imagine one or design it or to burn one down. So, building an actual specific proposal is a wonderful thing to do. And then, thirdly, well actually our last theme was productivity.

(01:00:24):

And one of the things we talked about was openness to international perspectives or innovation. And so, actually getting somebody from overseas who says, "Oh, why don't you consider doing this? This is how we do it in our country," is really very much in line with something that, as a small, isolated country, we really need to make sure that we're doing. So, thank you very much. I really, really enjoyed the presentation.

(01:00:49):

The specificity, I can see the Q&A is, there's plenty in there, and I'm sure there are plenty of questions in the room. So, what we'll do, we'll take, start with one in the room and I might see some online as well, but I'll take the chair's prerogative and ask a warm-up question to start with, because I've got so many questions, most of which you answered. I can see that you've thought through all of the issues. And you have thought through the issue I'm thinking of, and that's the inflation one.

(01:01:18):

Because the thing is, sometimes the recession is the solution, not the problem. If you're redlining your car, it's a really good idea to slow down. So, if the economy's overheating, you need to slow it down. And so, what I thought in there was why 3%? Why not two? Because if the Reserve Bank's doing a really good job, we could see things heat up. And they're slowing things down and we're at 2.8, 2.9, and anything above two, they should be slowing things down. So why three rather than two?

(01:01:54):

And then secondly, on the symmetry, I found myself thinking, you did cover how do we raise the revenue to pay for this? Every once in a while, we have to shut 1% of GDP. But I was thinking more about symmetry, that preventing the redlining in the first place is often the best way to prevent or could be, in my mind, the best way to prevent the recession. So is there a, the flip side of the symmetry we could do on that side?

(01:02:31):

Oh, redlining, operating your car too fast so that if you... Another analogy you might, you can't sprint a whole marathon. So, if you find yourself running above the pace that you're able to sustain, it's a good idea to slow down. It's not necessarily a problem that you've slowed down, it's actually the solution, because it'll get you to the finish line in better shape.

(01:02:53):

I mean, recessions that are actually the result of a previously overheated economy and are required to bring inflation back down, the question is why 3% rather than two is the target? And secondly, is there more of a role we can do just to prevent that overheating in the first place?

Claudia Sahm (01:03:14):

Right. Well, so as I hinted with the what's the right inflation measure, so it's also what's the right target range. So as much as I have explained to people how the 2% target came from New Zealand and all, this has been a thing recently. I don't, in terms of what's the right numeric target for the inflation, that does feel like something that's thinking harder about the coordination with monetary policy, which is not what I've done.

(01:03:47):

Really, the fiscal stabilisers come from the space of when we have a shortfall in aggregate demand, it is detrimental to the wellbeing of households. That's what this is organised at. The inflation check is to make sure that you're not adding fuel to the fire. Like you said, there are times where this is part of the solution, or this is just part of this inflationary process, that often shows up in aggregate demand. So, I do think that that check at the end is worthwhile.

(01:04:21):

So exactly how strict you want to be. I think to me the more interesting question, or another interesting question was, which type of inflation? Because you don't want, what you're trying to avoid is giving money to people, stimulating the economy in a way that it creates more inflation. Well, if your inflation problem, and this certainly shows up in the 2008 recession site. Because early on, there was a global commodity.

(01:04:43):

So, if it's coming from outside of the country, well, giving the people of New Zealand more money is not going to affect global oil prices. So, it's trying to figure out where is the, what's the flashing red signal if not to do it. So, I had thought more about that as opposed to if you get down to 2% and you bounce... I hadn't thought of that complexity, but that's a clearly... But the stronger, the more narrow you make those criteria, the less likely you are to get money out and support the blunting recession. There was your second part? Oh, oh, about slowing things down. Okay, sorry.

Dominick Stephens (01:05:27):

Yeah, that's fine. And actually, one of the questions from online is very much related. So, the slowing things down, how do we pay for it and how do we prevent overheat, use this to prevent overheating before the recession. And actually, another perspective that's come from online is, would you do something like add 10% to your net GST payments for the next six months or something like that? So, a practical suggestion of how do we pay for it and how do we use this to prevent upturns?

Claudia Sahm (01:06:00):

Yeah. So, in terms of, again, I'm not going to get into what specific way the tax would be formatted. One thing that... because I did take a stab at this in an earlier draft. One thing that is important is you do want to have, I mean, fiscal prudent is part of your mandate, what you need to be following. It is important not to start bringing those tax revenues back in too early. Because you don't want to undo the good of the fiscal stabilisation by then going into austerity earlier than you should have.

(01:06:37):

But you can, again, use economic condition. This may be too simplistic for the way you bring the taxes back in, but just something like, well, when consumption is 10% above its pre-trigger value. And in looking across the New Zealand episodes, that was about four years in. So maybe that's too long or too short, but again, you can use, the same principle applies if you want to use economic conditions to bring it back.

(01:07:03):

It's just with these automatic stabilisers... these type, you have to specify it. Because unlike, say, progressive taxes that just have this natural, in the recession, taxes go down and when the expansion happens, taxes go up. You don't have that built in, but it's the same principle applies.

(01:07:22):

Oh yeah, and you can, if you do that ahead of time, you're making a commitment and weighing out how much of this do you need to offset? Was this an expectation? Does this make sense? As opposed to, and there was a massive amount of discretionary policy that was done the world over in COVID, and it was not done with explicit commitments like a plan to do the pay-fors.

Dominick Stephens (01:07:50):

Okay. We've got number of questions from around the floor. We'll start up the back over here.

Speaker 4 (01:07:58):

Thanks for a very interesting talk, Claudia. Just a question about distributional impacts, which are something I'm always concerned about. There's been some debate about New Zealand's approach, relying heavily on the wage subsidy and did that result partly in quite a lot of that support going to perhaps higher income earners or to business owners who are at the upper end of the spectrum.

(01:08:23):

Conversely, I suppose you've got the Australian approach of doubling their job seeker payment, which would obviously be very pro-poor weighted towards the other end of the spectrum. I get the sense of your proposal for a lump sum going to 80% of households would sit somewhere in the middle. But have you seen distributional comparisons of those three different approaches across different countries? Do you have a view about how the alternatives stack up on a distributional basis?

Claudia Sahm (01:08:55):

Yeah. So, I think your rank ordering is probably right in terms of who directly benefits the most from the particular programmes. One research, and this is more in these distributional macro models that are pretty abstract but can be useful. And I think, one intuition I've taken from some of this work makes sense is these programmes, these are stabilisation programmes, at least what I'm talking about with the direct payments. If you blunt the recession, you get the recovery going faster.

(01:09:37):

It is typically, at least in the United States, it will be the workers at the bottom who will benefit, because they're the ones that are most likely to have lost their jobs, lost their income. And so, in some ways, you want to give them the money directly, but as long as you're giving it to people that spend, and as I said, you can have that up.

(01:09:53):

It's not just the low income that spend, the indirect benefits of getting the recovery going, they have distributional differences too, especially in the labour market. Asset prices, that's a whole other thing. But I think that's one where you want to think about, both the direct distributional effects and then the feedback distributional effects.

(01:10:18):

But again, as I said, this goes, in a recession, in a crisis, when the government is putting resources to support, in the recession, do you want to focus... How much of the focus do you want on stimulating the economy, getting the recovery going, and that the economy's this big thing, or do you want to focus on getting relief to people in need? Which sits much more, that's a direct distributional. And so that's just a value judgement.

(01:10:48):

I definitely sit and you can hear I'm sympathetic to the research that says your indirect distributional effects, a good labour market is really good for people at the bottom. I take that very seriously and I also put a substantial amount of weight on us not being able to identify who needs. I just think we're very bad at targeting.

(01:11:06):

And I'd rather, I think it makes in a wellbeing sense, it's better to get money to people who maybe don't really need it, as opposed to missing people who do. But again, that's a value statement. But I do think the distributional effects, both direct and indirect, are important, and there is more research on those. I haven't seen any with the wage subsidies, which, and I know Germany has had a similar programme of keeping workers attached to employees, employers.

Speaker 5 (01:11:41):

Hi. Thanks for your presentation. Just two very brief questions. In regards to setting up policy, the Reserve Bank mandate was changed very easily earlier this year with the new government. And I'm just wondering whether you think automatic stabilisers might be subject to the same political pressure over time? And then also, have you considered regional targeting with a system like this, because of earthquakes, natural disasters, Auckland being in lockdown when the rest of the country wasn't? Things like that? Thank you.

Claudia Sahm (01:12:10):

No. Fiscal policy is a creature of politics. There's nothing, the idea of this is not to tie... Okay, you either have politics... When you're doing fiscal policy, you have politics during the recession or you have politics before the recession, right? The idea is just to find a way to make it more effective, more thoughtful. Because you only have so much bandwidth during the...

(01:12:35):

And yes, I saw, sadly, the dual mandate went away, that with any of these, policymakers can create a policy and they can take the policy away, and they should be able to. That's just, I mean, that's the way fiscal policy should be set up. So yes, absolutely, there is a risk of that. And there could be good reasons that you turn it off.

(01:12:54):

I think, one suggestion I've gotten when I was working on these proposals in the US is, well, policymakers are never going to want to hand over the keys in a recession, because, if nothing else, they're going to want to take the vote and be like, "Oh look, we voted to send you all money."

(01:13:08):

So, this balance of would policymakers actually be willing to do this kind of policy? And that is a much more open question. I make the arguments on the merit of it, but some hybrid, which I was disabused of here, but, I mean, some hybrid of it's all ready to go and they have a vote, "Let's go." You might not be able to go full on, but maybe, but that would be the way to go. The second you asked about the...

Speaker 5 (01:13:40):

The regional.

Claudia Sahm (01:13:41):

Oh, the regional. Okay, so I think given the exposure to an understanding now of New Zealand's economic data, so having regional triggers would not make sense in the US. There're definitely proposals to do that with jobless benefits, state-level unemployment, because we have very, the recoveries look different. Economics, that doesn't make sense. You can see, unfortunately, if the disasters that you all have dealt with became of sufficient frequency, you could see an automatic...

(01:14:10):

But really, the only thing I could say is, if you think those payments or any other system you'd use from a recession of automatic stabiliser makes sense, then they're ready to go on a discretionary basis. Of course, that's also a risk for them. They're ready to go. But I think it'd be hard to, because again, you need an event that's general enough that you could, with confidence, put them on autopilot and I don't, and hopefully that will not be the case here with the natural disasters.

Jeff (01:14:49):

Thanks. I just wanted to offer two sceptical comments and see how you react to them.

Claudia Sahm (01:14:54):

Okay.

Jeff (01:14:58):

The first is that while I agree with all of the things you say about designing helicopter money drops of various sorts and how important it is that they're funded rather than unfunded, in the sense that everybody knows that in the future, there will be a revenue increase. I think your attempt to make them automatic and to find some trigger that works falls, because of the need for a policymaker to recognise that the trigger has arrived.

(01:15:27):

And because you've got yes or no, depending on inflation rate for example. At some point, there's inescapable discretionary element in the proposal you're making. And I think much better to think about if there's a discretionary policy properly designed with a windfall profit tax or something to claw it back than to think some way you can make this particular type of instrument automatic. That's the first of my two points.

(01:15:52):

The second one is that in an economy like New Zealand, with higher, as you pointed, volatility than in bigger economies, the importance of automatic stabilisers becomes more important rather than less.

Claudia Sahm (01:16:07):

Right.

Jeff (01:16:07):

But you want stabilisers that really do move ahead of the information coming into government. And there are a couple that do that. They're in the benefit system, and there's an argument that if you want automatic stabilisers, rather than discretionary ones in New Zealand, what you want is a really generous benefit system, a really generous unemployment benefit. So as the numbers on unemployment increase before your numbers even come in through Stats, they're down there collecting from WINZ.

(01:16:43):

And as more and more people are forced into other parts of the benefit system, the quantitative number of increasing beneficiaries sucks the money from government into circulation, without any need for decision-makers to ask, "What are the numbers for the current quarter?"

(01:17:01):

So, I mean, the quest for automaticity seems to me to be inherently a quest for generous benefits and progressive taxation. The quest for making helicopter money work is a quest for government credibility in promising windfall profits tax to fund the helicopter money.

Claudia Sahm (01:17:21):

Yeah. So, one thing I would say this, and it goes all the way back to the beginning of talk, towards the beginning of talk, whereas, let's define the problem? What is it we're trying to solve? So, the New Zealand economy is very volatile. And it's not just recession, it's moving all over the place. And that's even with a substantial automatic stabilisers that are already there, the benefit system, the progressive taxes.

(01:17:45):

Okay, so are we to smooth out that volatility just, in general, make these smaller movements? You'd have to make substantial increases in the size of the automatic stabilisers that are already there. So then, you're talking about a substantial increase in the footprint of government. And a lot of these programmes have other, not stabilisation goals, like their distribution or their... So, it's just, whereas, so that's like you're going after a big problem.

(01:18:16):

Just like, in general, there's a lot of volatility and you're going after with a big solution really ramping up these programmes, which means much bigger government. Okay, so the kind of proposals I've talked about, these, and this is why they come up with the semi-automatic, the quasi-automatic stabiliser, they're going and saying, "Okay, we're not going to worry about all the volatility, we're going to go pick some parts of the volatility that are really bad and we're going to have..." I mean, it's a targeted response, "We're going to go after those and address them."

(01:18:46):

But again, if you define the problem you're trying to solve differently, then I think, you have to be careful of how much scope you have, feasibly improve the automatic stabilisers that are there. But yeah, it's a different approach... This is just, and that's why, I mean, I breeze by it, but the model estimates of how much the proposal I had reduces volatility is really pretty small, because I'm only going after a few pieces of volatility as opposed to all across the spectrum.

Speaker 7 (01:19:30):

Hi. Thank you so much for the presentation. Just a super-quick question. I'm wondering if these one-time, lump-sum liquidity injections would be weighted by national income distribution? So, people who are at the lower end of the spectrum might receive more and people who are at the higher end might receive less. Or, is it just a one-time helicopter money drop on the economy, regardless of where you sit in the existing tax bracket system?

Claudia Sahm (01:19:51):

Yeah. So, what I was proposing was just lump sum. Everybody gets a check for $600 or 900, I forget what my calculation was for the 1% GP. And so, the, it's simple. Again, these are how do you weight different design principles? If you want to go on the direct distributional effects, I mean, frankly, probably the easiest way would just be to target it to 60%, 40% of income, as opposed to having this tiered, how big is your check based on your income? Yeah, you could have it be more targeted.

(01:20:30):

In an example, in the United States with, these checks were used in 2001, 2008, 2020, 2021, and they got progressively bigger. There were times where there were, adults would get a certain amount and children would get another amount, and then by the time we got to the end, it was just everybody got 1,200 bucks. But that's the way you could, that was one that's been used. I mean, you could also do it by income. You just start to create complexity and administrative complexity [inaudible 01:21:00].

Dominick Stephens (01:21:04):

Thank you very much. We'll take one from online, and then we'll take, after that we'll do one more inside the room. So, we had a question about the regional specific. I've got a different question here about industry specific. I know that, actually, you may or may not be aware. During the mid-2000s, we had this enormous boom related to the dairy industry.

(01:21:28):

And, at that time, people were talking about a milk stabilisation fund, similar to Norway's oil stabilisation fund, because of terms of trade, or one of the big things that we talked about. But this question is a different industry and that is counter-cyclical investment.

(01:21:50):

So, the construction sector in New Zealand has large boom-bust cycles which result in diminished construction sector capacity through. And would ramping up or down the level of construction of public housing be an appropriate automatic stabilisation mechanism?

Claudia Sahm (01:22:06):

So, the problem with using infrastructure investments or these, where you're trying to get business investments going, is they tend to have very long lags. And the idea of these automatic stabilisers, like what I talked about, is you want to get it out fast. And it certainly was, and we have some evidence that, I mean, intuitively, this wouldn't work stabilising.

(01:22:30):

But also, this was a key part... When Obama came into office, his plan to address the recession did have these small tax credits. So, chopping up, over time, money going to people, but it did have investments in infrastructure. Now the idea was thinking long-term.

(01:22:48):

We do want investments in infrastructure, but that proved very problematic, because what they needed was a lot more front-loaded stimulus. And again, I don't, these would be, seem very likely candidates for more discretionary policy. It gets really, putting on autopilot who's the winner and loser of a programme, just feels like a road one would not want to go down.

Speaker 8 (01:23:20):

Kia ora Claudia, thanks for putting forward some stimulating ideas. Just to build on Jeff's question, a couple of years ago, we had a proposal floating around about social insurance in New Zealand, and there were various pros and cons talked about at the time. It just strikes me, in the context of this conversation, we didn't talk about it as its potential to act as a stabiliser at a macroeconomic level. In terms of evidence you've seen of social insurance schemes overseas, how much of the volatility... The big advantage of them is that they build in the tax side as well, right? So, any sense on how much of the volatility they could potentially remove?

Claudia Sahm (01:24:11):

So, I think, so when you're looking at the macroeconomic volatility, even the job seekers programme, even though it doesn't have this payback, those programmes ramp up. The total spending on them gets bigger because more people are unemployed. The benefits don't change, but it is, even the job seeker... think that one will consider that a automatic stabiliser. So, I think with the unemployment insurance, you have more of this you pay in, there's work requirements.

(01:24:39):

But it has that same principle of, in a recession more people are on the programme, thus the totals, it is supporting more of aggregate demand and that's how it stabilises. But I don't... Yeah, and then with both the job seekers, it's a lump sum, or in unemployment insurance, you can top off benefits, or you can make them more generous all over time, depending on what your, the stabilisation versus the distributional. And then built-in repayments.

Speaker 9 (01:25:17):

Thanks very much. Very interesting. You studiously avoided, for good reason, the so-called technical recessions there. So, I'd just like to ask you about those, because you were quoted as saying the cost of the Fed going it alone to bring down inflation is not only inefficient in quotes as Elizabeth Blanchard argues, but "pointlessly downright cruel."

(01:25:43):

And now, so clearly, there is a problem there that current methods for bringing down inflation have quite significant paired social effects. And so, big question is whether there are fiscal means to at least help the Reserve Bank and its task of bringing down inflation. So I just wondered if you've got any thoughts on that side of things.

Claudia Sahm (01:26:12):

Great. So again, the proposal I put forward is one that should be multipurpose. It should work across recessions, it should, on its own, not automatically get in the way of monetary policy. But that does not mean that there are, there are ways, on a discretionary basis, to support... I think this is one where you want to emphasise the relief and not the stimulus. Anytime you give support out, you're going to be causing both. But there are ways to very target the money.

(01:26:43):

So, I think the cost-of-living payments, I mean, that was a case, and I would agree with that. If there's inflation, and I think direct payments can still be useful, you target it to much narrower group by income and you spread them out. So, you don't put one big impulse into the system. But putting that on autopilot I think would be a big mistake, because that's just too complicated to hit go on.

Dominick Stephens (01:27:20):

Okay. So, I think we'd better close there. Absolutely fascinating presentation, Claudia. Thank you very much. Yeah, as I say, the specificity of that proposal was fantastic and really something that the audience can obviously wrap their heads around and ask you about. So, thank you very, very much.

(01:27:43):

So, our next seminar in the series is going to take place on the 5th of July. We'll hear from David Vines, Professor of Economics at the University of Oxford. He'll be presenting on changing global economic order and global economic governance. So, it'll be another hybrid event, with David presenting in person. So, visit the Treasury website and register your interest early.

(01:28:07):

Well, I'd like to finish by congratulating Claudia in the usual way. And I'll finish with a brief karakia.

[speaking in te reo Māori 01:28:27] Piki te kaha. Piki te Ora. Piki te Wairua. Hui e, tāiki e!

Thanks for attending, everyone.

[speaking in te reo Māori 01:28:29] Mā te wā.