Transcript of Question & Answer Session Where is the Growth Going to Come From?


Thanks very much, Luci, for a fantastic speech. I guess a huge number of really important, interesting insights about Australia's growth experience and very adept at weaving in Kelly's story as well. So, thank you for that. Luci has generously agreed to take questions and she's also very skilfully stuck to the time limit. Not something lecturers always do in my experience. So she's left 15 minutes for questions so, lots of time for questions. Maybe when you ask your question if you could just give your name and organisation if you're from one. So who'd like to ask a question? Yep, in the middle.


Thank you, Luci. You talked about productivity, I guess my question is around how do we measure economic growth, GDP, whether that is a good measure of economic wellbeing and how happy people are? We talked about productivity and how to measure and increasingly the growth comes from the services sector and productivity is very hard to measure, there's no goods produced. So I'm thinking we are enjoying more consumption, more services, faster computers, faster phones, and so, which is all, maybe all growth, but very hard to be reflected in the original statistics. Do you think maybe working more and consuming more doesn't necessarily make people better off? So bigger TVs, people getting stuck in traffic, and faster smartphones, but not necessarily generating output, maybe are turning to measure of economic growth or output rather than GDP, should take me to a passive environment or social equality or …


So I think that could be the subject of another talk …

Luci Ellis

Yes that could be, you're right. Quite right. So for the recording, the question was about the fact that it's hard to measure productivity in GDP, particularly when services are becoming more important and the question of whether we're really getting at welfare using GDP as a measure when there are sort of more goods and services, and whether more work and more consumption is necessarily welfare improving.

So there's a couple things going on there. So the first point is, I think it is harder to measure services productivity, but the Bureau and other statistical agencies around the world certainly do make state-of-the-art adjustments to try and get at that. So firstly, you mentioned faster computers, faster phones. They do actually make adjustments for that, so if a company brings out a phone at the same price, but it's better, the Bureau does actually adjust for that and so it's recorded as if it's a fall in prices. That already does happen. It's not easy, but statistical agencies around the world already do that.

The question of the greater variety of goods and services available, that's exactly the point I made. Overseas travel and Netflix were not part of the CPI basket in the 1950s. How do you adjust for that? Well, I think in some ways it is the revealed preference of what people are consuming and that's sort of also the answer to the third part of your question about, well wouldn't it be better if we all did something other than buy more stuff? The interesting thing is actually retail trade is not growing that quickly. And we're consuming more services, so we are voting with our wallets that we would much rather sit in a café and hang out with our friends or look at our phone while at a café, rather than go shopping. And so I think that it's evidence that it's welfare improving.

But I agree: there are things about health and education and other services that are hard to get at. My favourite example in the health space, you've got keyhole surgery, and you've got moonboots. You know, those boots that people wear now instead of having a plaster cast and crutches. Both of those are clear improvements in terms of health outcomes and how quickly you heal. And you can measure that. So the Department of Health certainly has a huge number of metrics about death rates from surgeries, cancer survival rates, and all of those complication rates.

These are all things you can measure. They're not easy to get into GDP, but at some level you can get some of that in there and look, this is an ongoing question in the field of official statistics and the Bureau does a lot of work on all of that. So I think revealed preference says we are choosing that and it's not clear to me that we're actually choosing. I mean, average hours worked by full-time workers is actually coming down. The share of part-time work is going up. It's unevenly distributed, I can assure you of that. But it's not clear to me that we are actually choosing a configuration of working lots more hours and just spending it on things we don't really enjoy.

There was a question down here, somewhere.


Thank you for that Luci. One thing that worries me … The most worrying thing I think was this final graph you've got on gross R&D expenditure. Do you have some feel for the sectoral breakdown of that?

Luci Ellis

Not off the top of my head, but I think it's one of those things that's worth looking into. Again, it's a productivity thing. You can measure gross R&D spend, but the productivity of that in terms of what it then means for future outcomes, future inventions, future productivity may not be one for one. And of course, you don't have to do all your own R&D, you can adopt other people's work. And that's the whole thing: Australia, going back to that turn of the century period when foreign investors were telling me that we were an old economy. Australia wasn't inventing all technologies but we were adopting them really fast and that turns out to be almost as good a strategy, if not a better strategy. Being a fast follower is actually not a bad idea. You've got to be fast follower, not a slow follower.

Okay. Any other questions?


This issue of growth, makes me think back to the Club of Rome and the fact that resources were either running out or getting very much harder to get down to. And also the issue of climate change and how that's going to affect our future long term growth prospects.

Luci Ellis

Both very important questions. I think at least from the Australian perspective, we've still got a heap more coal and iron ore. There's still plenty in the ground that isn't being developed at the moment and in the Statement on Monetary Policy we had a box on the Chinese steel market. We think China has actually reached peak steel in the sense that its population isn't growing much anymore and it's got to the peak steel intensity that most other countries, that are not major exporters of steel have gotten to. And China is just not likely to have exports as being a big fraction of its steel output because its domestic demand for steel is so enormous, there just isn't actually enough world for China to export to in the same percentages that say, Korea would, or Japan, which are both the two main steel exporters, relative to their domestic production.

As economies become more service-oriented, they become less commodity intensive and so, we have gone through this period where some very large economies, China and India, are going through very steel-intensive phases of development and then it peaks. In some sense that's a bit of a one-off because while there's still a way for India to go, there isn't anyone as big as China or India coming up behind. So in that respect the demand for commodities is … I don't think the world will run out of iron ore. Remembering also that the vast majority of steel nowadays in industrialised countries is made out of scrap. So you don't actually need more iron ore to make more steel, you just scrap what you've already got.

I think one of the interesting things is China has a relatively low percentage of steel made from scrap and they want to increase that. So again, that's likely to: The demand for commodities is going to moderate before we run out of them. Now, you then mentioned climate change and that is a nice segue given that coal is one of the important things that we export. Obviously we do need to think about what that might mean. You think about the Paris Agreement. The baseline, versus some scenarios of what might happen to consumption of carbon releasing fuels like coal versus switching to renewables, or becoming more efficient in our energy use.

In terms of the raw ‘what would this mean if everybody just stopped importing our coal?’: If it happened slowly it would actually not amount to much. In terms of the coal exports are sort of, not even a percent of GDP in levels, and so if that dissipated … If you think of BP they have some scenarios where their baseline looks a lot like our current forecast for coal exports, so it was their scenarios of stronger reductions in use of coal around the world. Yeah, it would halve your coal exports perhaps, or cut them by a third, but if the scenarios they have, that happens over several decades, you would not notice that in GDP. You would notice coal exports are lower, but that wouldn't materially weigh on our growth going forward.

And again, I think one of the things that becomes very relevant as economies are trying to reduce their carbon usage is innovation. I think it's quite relevant just how much innovation is happening in the energy use space both in terms of energy efficiency – I hope you're all using LED light bulbs now – but also just in terms of the efficiency of energy production. Moving to electric cars, moving to renewables as a source of electricity. That's all happening and I think it would not surprise me, particularly depending on where China goes with this, after the Congress. They've made it quite clear that they're concerned about the environmental degradation that their industrialisation has imposed upon their economy and on their environment. And so, it wouldn't surprise me if the authorities there do quite a bit to switch their energy mix, and that makes a big difference to the amount of carbon going into the atmosphere.

So there was one other … It was over here.


My question was really about the public infrastructure and productivity. And it seemed to me we have enormous opportunities if we switched from politicians picking something that was in an electorate and be a ribbon-cutter to what's a sensible benefit–cost assessment. And I would have thought the Kellys would jump up and down about the nonsense that goes on about it.

Luci Ellis

Look, as a macro-economist I'm not going to comment on the micro-economic cost benefit analysis and how you best pick infrastructure projects. That's certainly a view that a lot of people have put and there are mechanisms, you know, Infrastructure Australia, and there are mechanisms for working out what are the good things.

I think one of the hard things is if you take cost benefit analysis really seriously, it's very hard to model how business opportunities change. It's not just, ‘Oh, well, you know. You put an airport here and people don't need to drive to the airport over there.' It's like, well, if you put an airport over here suddenly a whole new industry of exporters starts air freighting food to China. That's a real example. So you have a whole new industry and it's very hard to get at that. You can't really predict how businesses are going to take opportunities to create whole new business opportunities as a consequence of that infrastructure. So it's a hard problem because of that unknowable.

Businesses take decisions under uncertainty all the time, but if you get too mechanical about it, you may miss things that actually turn out to be really good, but are just the benefits are very diffuse, or the benefits are reaped by whole new industries that aren't there yet. So I think that's important. But I agree that it is important to think carefully about what infrastructure you're putting in and what the cost benefit analysis of that is. And that also means being open eyed about where the productivity benefits will come from and recognising that we are in many respects an urban economy, and thinking about that.


I think we had, I know one … Time for one last question. One at the back.


You mentioned earlier about increased dependency of part-time workers. The IMF made a similar observation, they said that was contributing to depressed inflation rates through this whole labour share problem as well[unclear 00:14:09]. Does the Reserve Bank have a position on how that can be addressed if that's true it directly affects the consumption, if consumption's about 60% of the economy?

Luci Ellis

Okay, so the question in case it didn't come out on the recording, was whether the increasing share of part-time work and employment is holding down wage growth and thus consumer price inflation. I don't think that we think that the increasing share of part-time work per se is weighing on wage growth directly. That's been a long-term trend over many decades and additionally there's just a normal pattern when the labour market is slow, full-time employment falls, part-time employment holds up. And it's not because people are being sacked, it's because just the flow from people who currently have a part-time job getting a full-time job slows down at the same time as people continue to retire, and so there seems to be a normal mechanism there.

And that's what we saw last year. And then, as the employment growth picked up and this year, almost all of the growth in employment has been full-time and it's been extremely strong. So in a cyclical sense, over this year that probably isn't the case. But you're absolutely right. The wage price index came out today and it was still pretty low. Wage growth has been low for a while and I think as we said in the Statement on Monetary Policy, it's one of the central questions about when is inflation going to pick up? Is when is wages growth going to pick up and how much that's going to flow through?

I think it's not so much the part-time growth itself, but it's more that as labour demand picks up there are plenty of people who are willing to work more hours and so the hours margin is where part of that extra labour demand is met rather than through employing more people, and that seems to have a different dynamic in terms of wage growth. So it's not necessarily holding down wage growth as such, but it just means that as labour demand picks up, the effect on wages is less. That's our current thinking on that.