Transcript of Question & Answer Session Structural Change in Australian Industry: The Role of Business Services

So if you would like to ask Alex a question as she’s kindly said that she would take some. Please if you could stand and say your name so that people and Alex knows where you’re from, and ask your question clearly so that the mic can pick it up. That would be great.

And they don’t have to be constrained to the presentation.

No. You can ask all sorts of questions. You probably can’t ask her what the exchange rate is going to be or the interest rate will be at the next monetary policy meeting but, other than that I think she’s pretty happy to answer your questions.

Absolutely.

On your measures of productivity, this is probably a left of field type of question, but … normally outsourcing in some service sectors is really involving outsourcing to the non-enterprise households. So for those of us who spend hours waiting to get through a telephone queue or doing online stuff, that’s not measured. And is there any data on that, of any time-use surveys? That I would have said would become more important over time. Do you have any data on that?

The short answer is we don’t have good data on that. And I think, you raised that the outsourcing of business services to the household sector because you have to do a lot more work when you’re on the phone to get to the right place rather than having a person on the other end of the line. Another area where this question about measurement and household services and business services comes into play are questions around how should we measure the service being provided by somebody renting out their house using Airbnb, because that’s a margin where at some deep level it’s always been thought of as a household, but now some part of what was a household service is being delivered as a business service in some cases.

So the short answer is we don’t have good measurement. But I would point out that another very interesting trend is that a whole lot of production that used to happen in the household sector has actually been formalised in the form of the household services. So if you look at some of the sub-categories that have grown particularly strongly, it’s things like aged care workers, child care workers. So we do actually have good measurement of what things are moving out of, and were not being measured in the informal sector but are moving into the formal sector, but we don’t have a very good way of measuring how things are working in reverse. And I think you’re right. I think it would be detailed time-use surveys that we would need to be able to do that and as far I’m aware we haven’t got a recent time-use survey that would allow us to do that for Australia.

I was kind of wondering when you worked as a [inaudible], there was a lot of talk about [inaudible] especially at the [inaudible] where the exchange rate is considered, well about [inaudible] industries [inaudible]. It’s all about the conditions [inaudible]. So whether it would be [inaudible] originality [inaudible] So [inaudible] all. So whether it is technological factors is actually more [inaudible] sector [inaudible] before it was manufacturing or mining or [inaudible] resources sector. So there’s no sort of change [inaudible]

I think there’s a couple of aspects to that. So in terms of the measurement, you’re right. Things like architecture would fall into business services but actually the Arts, so very literally taking things like going to the opera and art galleries, would actually be classified as a household service. And I think one of the interesting things is that when I set out to do this speech I had actually thought I would be talking about how the manufacturing sector has changed, and what’s really interesting there is, in Australia there’s quite a sizable what you would describe as a niche manufacturing sector. So there’s a lot of creative input, a lot of intellectual capital, in the design side of things that happens in Australia. But then the production element of it tends to get outsourced overseas.

So what a manufacturing industry looks like in Australia is much less likely to be a factory now than a much more creative space where people are coming up with the original ideas and so, in composition terms, I think it’s true that, because of outsourcing and because of dividing up the supply chain across a whole range of different countries as well means that what we have in the goods production sector is probably a lot more creative than what we’re thinking about when we have a picture of the manufacturing sector. So it is another force that’s operating.

In terms of your graph showing the change in the sectors in a big way, what do you think, you know that the, underlying issues that [inaudible]?

Oh, GDP growth or productivity growth?

Productivity growth … What do you think are the policy levers that are, or what might be driving that change that you’re seeing there and then how might that relate to the policy levers that even the Reserve Bank or sort of fiscal policy, or GDP policy or what do you think might be driving that pretty massive structural change there in that relationship?

The first point I’d just reemphasise is it’s a correlation … So the way I set the graph up was, wouldn’t it be amazing if we got a graph that showed that there was a correlation between these two things and we drew the graph and as any of you who have been applied economists know that you don’t often get something looking better than what you expect.

So in terms of policy levers, we’re really dealing with long term structural trends so that takes it out of the space where the Reserve Bank would really have relevant policy levers because ultimately, things like monetary policy are business-cycle focused.

So the message to take away from it is, when you’re getting structural change and it is leading to reorganisation of industries and where things are allocated, it does seem to have a positive productivity effect. So I guess in the current context the question is: ‘Well how is technology transforming things?’ and there is a lot of debate about where the margins are between, you know, Uber and Airbnb are kind of right on the edge of that for example. Is that something we should let happen or is that something that we should feel quite nervous about. And I think the answer is in a long-term sense, you have to take advantage of technological change to get the productivity growth, but you also have to be conscious that it can be a disruptive process. So I think one of the lessons from the labour market information that I touched on is that technological change really does create new jobs but there are other jobs that become obsolete. So there’s a real margin question about how do you make sure that the people who lose their jobs in that process are in a position to retrain and get the new jobs. So I think it’s hard to come up with a specific policy recommendation, but I think it would be, try to understand what sorts of changes are happening and try not to get in the way unnecessarily, but to be aware that there are consequences of that that may need to be managed. And maybe it’s my personal bias and interest in labour markets, but I would say that’s certainly one of the areas that people in state governments and the federal government should be paying a lot of attention to.

Abuse my privilege as chair and say do you think these changes you’ve mentioned, especially [inaudible] struck by the 1960s to now, have made the economy more robust or less robust? Should we more worried in the macro or less worried in the macro? If I take some of the most recent literature that kind of implies that the tail wags the dog. You know, it’s firm level shocks that transmit up and cause volatility at macroeconomic level. Should I be worried about longer supply chains if I’m worried about stability or should I be thinking that this is actually making everything that’s small a lot bigger.

I think there’s probably offsetting forces at play. My natural reference point, and I’m suspecting yours also is the financial stability literature. So a more distributed system tends to be more robust to distributed shocks. Because if one piece of it is affected, the rest of it can work around that, but they tend not to be as robust to correlated shocks.

So I think this is one of the things that you would want to understand at that firm level much more effectively, because I think the graphs are intriguing, but they don’t really explain what’s happening and why it’s happening at the firm level and ultimately that’s what your question requires. What exactly are these interactions? Is it that some of those business services firms are just providing services to one employer or one goods production sector firm, for example, or is it like the IT provider example where I’m producing a solution that I can sell to a very large number of different firms in which case I’m going to be reasonably insulated from a specific shock. I think the answer to the question really requires more information on what’s happening at the firm level. But it …

More work for us, that’s always good.

Well yes, more work but, you know, I think some of this is about what are the right questions that we should be asking. When it comes to productivity growth, it’s a fundamentally important thing. That’s where the welfare of the Australian people comes from. That’s one of the core pieces of the RBA’s mandate. So we do need to understand it better and we have the Productivity Commission, we’ve got the recent information from the Treasury. So this is a really important question and I guess what I am flagging here is that maybe this is an area that we haven’t paid a lot of attention to that deserves more research. And who knows, we may be able to create network graphs and based on firm level data and we can actually be more specific about this question about resilience.

It is one of these questions. If you were to take [inaudible] maybe the upstream [inaudible] is knowing that [inaudible] get some short term gains out of it rather than the long term [inaudible] we don’t know yet wouldn’t you say? [inaudible]

But I think the interesting thing about productivity is, when you go into the literature where people are trying to understand what drives productivity, one of the things that works quite differently is that often the question is, well we’ve seen computers become integrated in everyday office activities so, why is it that we can’t see a boost to productivity growth? So it tends to be that, the dividends are much longer term and it takes a really long time to see the benefits rather than the other way around seeing them very quickly and perhaps not finding it the longer run.

Any more questions?

Now we talk about [inaudible] cross-matching the data [inaudible]… How would you explain that? [inaudible] cross-matching and [inaudible]

So there are data available. So there’s international input-output tables. I’ve really just restricted myself to looking at the Australian case. But you can actually ask questions about well if the production of my sector becomes an input to a sector in another country, which then becomes an input to a sector in a third country, how does that all show up?

And so actually the Reserve Bank work that I referred to is a Research Discussion Paper that looks at that question and, one of the conclusions was that in an international sense, Australia has actually become more upstream. And that at some fundamental level has a lot to do with the fact that a lot of our exports are mining exports that become inputs into other people’s production sectors.

And from memory, these international input-output tables are actually not available in the post mining boom period … but, I mean all of things that I’ve talked about here fundamentally can be done in an international context. Another conclusion of that paper that I found quite interesting, and it was sort of obvious when you saw it, but somehow wasn’t until you see it, is that when we look at trade, we’re not netting out, it’s not a value-added concept. If we sort of add the trade that I do with China and then the trade that China does with the US, we’re not actually measuring the value-added along that process, we’re just adding it up as a gross concept. So anyway that can deliver some interesting insights as well.

I was wondering in terms of the changes [inaudible] did that translate [inaudible]… going gangbusters [inaudible]

I’ve actually done some other work looking at this. The connection with wages is also related to supply. At the same time as this process of increasing demand for say non-routine cognitive jobs is occurring, we also have had a significant expansion in tertiary education so, the supply of people with the sorts of qualifications that match this demand has also been increasing.

So basically you would expect wage growth to be stronger in situations where there was more demand than supply. And so, there actually are some business services where, it’s been much more equal than others. So at this particular instance in time for example, we have a liaison programme where we talk to businesses to try and understand what’s happening on the ground as it were. And when it comes to wage growth, one of the big macro questions that we have is, why has wage growth declined the way it has. When we understand that, maybe we’ve got a better way of forecasting how we should expect wage growth to evolve going forward. And one of the things we find is that there are pockets of quite rapid wage growth, but it tends to be in things like IT security and cloud computing services where there is a huge amount of demand, but the supply of graduates and the supply of experience hasn’t kept up. But it’s not so true for, I hesitate to sort of pick out a very specific example, but like legal services for example. That’s actually an area where there’s been a huge increase in supply, so that hasn’t had as much in the way of wage consequences. And I think, well this is a bit of an aside, but on the non-routine manual side for the household services, that’s an interesting one because growth in wages there has not been very strong. And one of the explanations for that is that the supply of people who can do those jobs has also been quite significant.

Any more questions?

[inaudible]

We haven’t done that. So basically what we can do is, as long as you can identify the public sector as an industry … there is a public administration industry, you can look at the linkages and interrelationships between that industry and other industries. You can cut it in slightly different ways, this is a reasonably aggregated view. So the answer is we could look at whether that’s different, but we haven’t looked at whether that’s different. So I can’t answer the question. But I will go back and ask some people to have a look at it.

Have you looked at [inaudible] into the changes or the skew, [inaudible]. Clearly is there a difference between male dominance, male [inaudible].

So I have looked at it. Just sort of going back to this point about tertiary education. So one of the things that has been really important for women’s participation in the workforce is access to tertiary education and participation in tertiary education. So, for both males and females, participation has increased, but for females the increase has been greater.

So it is quite interesting if you compare things to the 1980s. There are three occupations that have become distinctly more even, they were quite male dominated and have become relatively equal. The three that stood out to me, from memory, were law, medicine and accounting. So these are three areas where females have been able to take advantage of their tertiary education and gain employment. But interestingly there’re some sectors that have become more male dominated over time and some that have become more female dominated. So for example, vet apparently has become more female dominated, whereas IT remarkably has become more male dominated over time. So the question about how all this plays out is, I think you have to kind of almost go a little bit lower into the, you know deeper into the question and ask well, so what’s motivating women to go into accounting, law and medicine and vet for example, but not to go into IT, and as we’ve discussed about economics, certainly at the high school level, there’s been an attrition of females doing economics has been much more significant than the attrition for males. So I think there’s a real question about how these choices have been made and what’s influencing them because it’s working quite differently in different sectors of the economy.