Categories
Podcast episode

How the Australian Bureau of Statistics prepares the National Accounts w/ Robert Ewing – EP162

The National Accounts is the comprehensive data set on a country’s economic performance. It gives us GDP growth estimates and a whole bunch of other important indicators. Australian Bureau of Statistics Principal Advisor Robert Ewing takes us behind the scenes at the ABS and provides some great info and insights into how the GDP figures are prepared. Learn about the huge range of economic data from households, businesses, and governments that go into the National Accounts, the roles played by algorithms and judgment, and how the numbers are crunched using the time series database FAME, short for Forecasting Analysis and Modeling Environment. 

Please get in touch with any questions, comments and suggestions by emailing us at contact@economicsexplored.com or sending a voice message via https://www.speakpipe.com/economicsexplored

You can listen to the episode via the embedded player below or via podcasting apps including Google PodcastsApple PodcastsSpotify, and Stitcher.

Links relevant to the conversation

Robert Ewing’s LinkedIn profile:

Economics Explored EP153 which also considered the National Accounts

Transcript: How the Australian Bureau of Statistics prepares the National Accounts w/ Robert Ewing – EP162

N.B. This is a lightly edited version of a transcript originally created using the AI application otter.ai. It may not be 100 percent accurate, but should be pretty close. If you’d like to quote from it, please check the quoted segment in the recording.

Gene Tunny  00:01

Coming up on Economics Explored.

Robert Ewing  00:04

It’s the work of hundreds of people across the ABS once you count the people in the survey divisions, the data acquisition divisions, all the other publications such as the balance of payments, the capital expenditure, the business indicators publication, which all feed into the national accounts.

Gene Tunny  00:24

Welcome to the Economics Explored podcast, a frank and fearless exploration of important economic issues. I’m your host Gene Tunny. I’m a professional economist based in Brisbane, Australia, and I’m a former Australian Treasury official. This is episode 162 on how the national accounts are put together, my guest is Robert Yuen from the Australian Bureau of Statistics, the ABS. Rob is the principal adviser to the ABS statistical services group. In this episode, Rob takes us behind the scenes of the ABS and provide some great info and insights into how the GDP figures are prepared. While Rob and I chat about the Australian National Accounts, I hope this conversation is useful for you, wherever you’re living. statistical agencies around the world will be using similar data and procedures to the ABS as they prepare their own GDP figures. Please check out the show notes relevant links Information and for details of how you can get in touch. Please let me know what you think about either Rob or I have to say in this episode, I’d love to hear from you. Right now for my conversation with Rob Ewing on the national accounts. Thanks to my audio engineer, Josh Crotts for his assistance in producing this episode. I hope you enjoy it. Robert Ewing from the Australian Bureau of Statistics, good to have you on the programme.

Robert Ewing  01:42

Very happy to be here Gene, always keen to spread the gospel of the great work that the ABS is doing statistics and trying to tell people the story of what’s going on in the economy.

Gene Tunny  01:53

Excellent. Well, I’ve been really impressed, Rob, that you’ve been communicating on LinkedIn, you’ve been talking about the data and giving us some insights into how the ABS compile the data. And I mean, one of the sets of data I’m really interested in is the national accounts and GDP, which is the statistic that measures the amount of economic activity in a certain period of time. And I’d be keen to chat with you about GDP given that you’re at the ABS and you’re involved in, in collecting the data and, and crunching the numbers to create the GDP figures. I chatted previously with a colleague of mine, Brendan Marquis Taylor, we chatted about the three different measures of GDP, would you be able to kick off by telling us what those measures are, please Rob, why do we need three different measures?.

Robert Ewing  02:48

Sure, so I should just say at the outset that, you know, today, I’m going to skip over an awful lot of the technical detail. And so I’m sure that the more statistical methodology knowledgeable are going to be screaming occasionally. But it is very complicated. The manual for the national accounts is several 100 pages long, and you know, can take a lot to get into. I think to start with, I tend to think of the GDP as it’s one concept that you can measure three ways. So GDP, as you say, is the measure of all economic activity. But we have to get a bit more specific than that. One of the things I’ve been talking about on LinkedIn is this idea of the production boundary, we have to put boundaries around GDP, because really, anything you could think of could be economic activity, we have to narrow that down a bit. And so we focus on market activity. And we focus within market activity on production. So GDP is a measure of the total value of production for the market sector in an economy over a period of time. It turns out that the way that the national accounts frameworks are set up, and it very clever frameworks developed over you know, more than 50 years by international groups. You could also measure this two other ways. So the first way is to measure how much income everybody in the economy has gotten from these productive activities. So if you measure the total amount of money that goes to labour, what’s called compensation of employees in the national accounts, and if you measure broadly, the national accounts concept of profit, what’s called gross operating surplus. And so if you add those two together across the entire economy, then you also get a measure of GDP. And that’s because if you think about production, well the measure of production is how much what you produce is worth minus how much you had to buy, in order to produce it, or the two things that are left over after that are income and profit. So it’s quite easy to see how those two match up. The third way, which is the expenditure method of GDP, is to think about it from the other side and think that will, we know, in economies and in markets, that there’s both a supply the production, but also a demand the consumption of the goods. And so the expenditure approach looks at where all this production of goods and services in the economy gets used. And so it adds together, the total amount of money spent by households, the total money spent by governments, some of the things that we produce gets sent overseas. And so we count the exports. Some things go into a warehouse, even though be produced, so we count inventories. And then we didn’t produce all the inputs that we use. So we subtract imports. And so that also gives you that same measure of how much was produced in the economy. And so those are the three measures of GDP, which are normally called the E measure, the I measure and the P measure in Australia. In theory, if you have measured everything absolutely perfectly, they will be equal.

Gene Tunny  05:58

Yeah, conceptually, from the way that they’ve that just because of the theory, and there’s a national accounting framework, isn’t there. So you were talking about this has been developed over 50 years? I mean, that’s since the you’re talking about since the UN codified it, I think, did they try and codify it in the late 60s? And then there’ll be various iterations of that approach? And we follow that there’s an international methodology.

Robert Ewing  06:27

Yeah, that’s right. So there’s the system of national accounts. So I think the first one was 1968. And that was building off kind of a lot of international cooperation. But I believe 1968 was the first time they really sit down like a consistent set of international rules. There was an update to it in 1993, and update in 2008. And if you do the math between those, you won’t be surprised to know that there’s an update currently underway thinking about further changes to the system of national accounts. And that will be due out for hopefully for endorsement by the UN statistical commission in 2025.

Gene Tunny  07:04

Yeah, yeah. And I remember from my time being on ABS reference committees for different things, or the technical reference groups, just all of the the issues you have to think about and R&D is a tricky one. I remember that, but we don’t need to go into into that today. I’m keen to understand what sort of data go into this Rob. I mean, you mentioned there’s production, there’s, there’s income, there’s expenditure. So is the ABS collecting data on all of these different transactions? You’re collecting data from businesses, you’re collecting it from households? What sort of data go into the mix, Rob for GDP?

Robert Ewing  07:46

Well, I think the short answer is anything that you can get your hands on. So a good national accountant is a bit of a data scavenger, because we’re only ever looking through kind of little windows into the economy. Kind of weird, we only have these narrow snapshots. So probably the two absolute most core measures of that production side, firstly, the annual economic activity survey. And so that’s a quite a large survey that we do once a year. And we go out to businesses, and we asked them quite a lot of questions about what they produce, how much money they made, you know, kind of how their money was spent, and so forth. And that allows us, in particular, to do what is quite a complicated little fiddly job, which is to try and convert things from the world of business accountancy, into national accountancy, because there are some things where there are some very, very important differences between the way that GDP is measured. And the way that business accounts are put together. The two most important are probably the concept of profit, which can be measured quite differently to that concept of gross operating surplus I talked about before, and also the concept of depreciation, and the differences between economic depreciation and accounting depreciation. But, you know, there are a lot of very complex nuances about that. And so a lot of the economic activity surveys asking the questions that allow us to take the business’s accounting estimates and convert them into economic estimates. So that’s absolutely kind of a foundational piece of us understanding that business side of the economy. So at a quarterly level, then its equivalent will be the quarterly business indicators survey. So that’s a somewhat smaller survey and asks a much smaller set of questions was kind of trying to give us a bit of an idea about what’s happening quarter to quarter, as well. And that’s kind of a very important survey there but the range of inputs is enormous. I’m household final consumption expenditure quarterly uses something like 15 or 20 different input data series, including retail trade, data from APRA, data from various private sector organisations who provide information, data from the quarterly business indicators surveys. So there’s a, an absolute broad range, because things like those big economic activity survey or the quarterly business indicators survey don’t necessarily cover everything. And so we bring the other bits of information to kind of tell us about, particularly the expenditure side of GDP, but also thinking about the role of government, where we have a lot of data that comes directly from federal, state, territory and local governments and tells us about their activities. We have detailed trade information, probably some of our most detailed data sets on the trade side, which allows us to get a pretty good idea of what’s going on with exports and imports. We rely very heavily on the consumer price index and the Producer Price Index publications, because they give us an idea of prices in the economy. And they allow us to convert those current price numbers into volume estimates of what’s going into GDP. So the real GDP as it’s often called.

Gene Tunny  11:17

Right, okay, yep. And that’s, that’s what’s often reported, that’s the that’s seen, that’s the data set that are the item that is looked at to determine whether the economy is going into recession or not. It’s that real GDP number, that volume number where you’re trying to abstract or control for inflation that occurs? Okay, so you’ve got less data for quarterly GDP than annual, it seems, is that right? So you’ve got detailed estimates, annually, you’ve got this other survey, quarterly, you’ve got, I mean, you still got a whole bunch of data. And what are you doing? You’ve got all of these models? You’ve got I mean, Is it done in spreadsheets? Are we able to go into that, or is it in R or or Python? And then is there some way of describing what goes on?

Robert Ewing  12:10

Well, so it is, fundamentally, it’s a very large code base, in our language and computer stuff, piece of software called Fame. And if anybody listens here, knows Fame and doesn’t currently work at the ABS, please immediately make a job application, we’ll fast track you, we need as many people who know this fairly obscure computer language as possible. And so we have here that’s 1000s and 1000s, of lines of code across all sorts of different modules. The basic approach is we’re trying to build things up from the absolute base level estimates. So we’re trying to look at the lowest level of the information that comes in and when we build the GDP up from that, so it’s very much a bottoms up approach to estimating it. The approach that we take is quite different in different areas. So in some cases, we get pretty good information, quarter by quarter on what is happening in the economy. So exports and imports, as I mentioned, is a pretty good case. We know, not perfectly but you know, we have almost census level data, we know almost every transaction that’s happening with imports and exports, there are some complexities around imports and exports of services where we don’t have as strong information. But we have a pretty good idea about what’s happening there. And so all we really need to do is just make some conceptual and timing adjustments and add everything up at the other end of the spectrum would be something like imputed rent. This is a really interesting kind of concept here in GDP. So one of the really important parts of the overall economy is the service we all get from the houses that we live in. Now some of us pay for that to the rent that we send to our landlord, kind of once every week or fortnight or month. But if you own your own home, we still want to count that value, because it’s still an economic service that’s being provided in a market like environment. And so for that we have what’s called imputed rent. So we impute how much rent is there. The data set that we have for that is primarily the Australian census. So every five years as part of the census, we go out and measure every household in Australia. And that also gives us an estimate of how many households they are there are and in what type of houses and in what locations. And so putting that into a model allows us to estimate imputed rent. But between censuses, we don’t have a solid survey or a data point. And so we have to model that. So it’s really a form of nowcasting. We’re using the information that we have about changes in the population, information about household formation and information about kind of your building construction. estimates of how much demolitions are happening to us forget how the number of households are changing and we obviously have information about rents from other series. And so at that end of the spectrum, it’s really a model, which is using a bit of data. But it’s not a massive amount of data compared to like its base, which is set every five years with the census. And that will be something the team will be turning its mind to very shortly having just had the second set of census results released.

Gene Tunny  15:21

Yes, yes. Great, great work. Everyone’s excited about that. So I saw your boss, David Gruen who’s well, it was, I mean, we both worked for him and within the various times in Treasury when we were there, and you’re working for him again, at ABS. He’s been getting a bit of media on that on people working from home, which is good to see. So all very good. Rob, can I ask you about this quarterly data? So you’ve got all of these bits of data going into it? And you’ve got these different modules? And there’s all this code? Which was a revelation. So I’ll have to do some research on that. I found that fascinating. What’s the quality of the quarterly data like? I mean, I know you revise it in the future, and there’s a statistical discrepancy, isn’t there? Could you tell us a bit about what the quality of that is like first, and then we might go into the annual data and what you’re doing there? How you’re trying to, yep. If you can tell us about that, please. That’d be great.

Robert Ewing  16:20

Yes, sure. So I think that’s so I think there’s two answers given to the quality of the quarterly data. So this is all in the context of we kind of spent a lot of time and a lot of resources and a lot of effort on making the quality, the highest it can be. But inevitably, a lot of the systems and the data in the economy is fundamentally on an annual, particularly financial year basis. So if you think about the financial accounts of the company, they will be producing kind of view monthly and quarterly financial information. But it’s only at the end of the year, that they really do that complete process of producing a full set of accounts, and making all of the adjustments and thinking about all of the different bits and pieces they need to take into account. And no matter how well you’ve gathered the quarterly data, the annual data is always going to be better. So you’re always going to have this view of the world that is a little bit sketchy. And yeah, and I think that if you look at the revisions, the revisions, they will move the numbers around, they very rarely change the story of what’s going on. So GDP might move up a little point one or point two, down to point one or point two, but very rarely does the story of what’s happening to the economy change very substantially. The other element to that is that even the quarterly estimates get a bit better with time. So we produce the estimates for GDP, roughly nine weeks after the end of the quarter. So our June quarter is published in the first week of September, on the first Wednesday in September. So that is a pretty good amount of time. But there’s still information that continues to come in, there will be businesses that we surveyed, who didn’t have time to respond, there will be data sources, for instance, ATO tax data, some of that won’t even arrive until after the national accounts has been published. So our picture of each quarter gets better with time as we gather more data, and you know, some late returns come in. And so that’s part of what contributes. But the big story is that really, GDP is fundamentally an annual measure that we can do on a finer time horizon using really a combination of different approaches. And for some things, such as household final consumption expenditure, there’s not a massive difference in the quality between the quarterly and the annual household final consumption expenditure has pretty much the same sets of information. But if you’re thinking about production, it’s quite different. And yeah, you mentioned statistical discrepancy. I think this is probably a good, a good way to talk about how we actually measure those three measures of GDP in practice, and how we actually make them equal to each other as they theoretically should be. So when we measure GDP every quarter, we can. It is very rare that the three measures come out exactly the same. Probably has happened once or twice. We were probably very suspicious when it did happen. As of now that can’t be right. There can’t be equal. We do two things in response there. Firstly, we use the differences between them as a way of looking at the different components and models and data sources and thinking about, are there gaps here? Are there adjustments that we should be making? There are other pieces of information that we should be seeking that would allow us to bring those measures closer together, but we’re not going to arbitrate them all together without some source of evidence for that. And so the three measures will be a little bit different from each other. So in order to get the measure of GDP that’s published on the front page of the webpage, what we do is we average the three of them. And so we have what’s called GDP A for average. And that is really just literally you add the three measures, and you divide by three. And you know, for those who are fans of the details, you’d notice, now we’re doing that in terms of the level of GDP, not the percentage change. Now, of course, it would be nice to get rid of that statistical discrepancy. And to do that, we need that information set that we have for the annual publication. And this brings us to, I suppose, what is the core of kind of producing like the benchmark GDP estimates, which is a process that’s called supply use balancing. And so I talked before about the idea that we can match production to income to expenditure at that top level. But in theory, we can also do that for every individual product that’s produced in the economy, that we should be able to say, for every products, let’s say, mushrooms, just picking on that one, because it’s very early in the list. We know, you know, in theory, we know we’ve produced you know, this many mushrooms. And you know, we know this many mushrooms got used by restaurants and other people and food manufacturers, this many mushrooms got sold to households, this many mushrooms got exported. And so in theory, we know well, those two numbers, the supply of the product and the use of the product have to be equal. We’ve defined as the way we’ve defined GDP, of course, the way that we’ve measured everything, through all these different surveys, we haven’t gotten a complete survey of every single person in the economy, we haven’t asked every single household what they consume. So we know there are some errors there and they don’t balance. And so every year, we go through this process of supply use balancing where we look at products, we don’t look at it at the level of every single kind of view, 375 mil Coke can, and 750 mil Coke can, and we have about 300 products that we look at so that we broad categories, like legal services, or metal, steel manufacturing products, or things like that. And we look for each of those products. We want to balance, how much was produced, how much got used by households, by government, how much got exported, how much went into inventories, and how much was used by other industries in their own production. And we make sure across all of those products, the supply and use is balanced. And once we’ve achieved that, then we have the three measures of GDP will be equal, because we’ve now brought those two halves of the measurement of the economy together. And so that’s, I suppose that’s the really, that’s the big work that kind of goes on, after a year is completed. So once we’ve finished the financial year, then the work starts on producing those supply use tables and doing that balancing. And when you’ve done that, the reason that’s important is I suppose it gives you that foundation that you can build on. So when policymakers are looking at GDP, they’re mainly going to be focused on what’s happening to GDP. Right now, they don’t care about what happened to GDP a year and a bit ago, which is about how long it takes us to produce those supply use tables. But through producing them, we make sure we’ve got an accurate representation of how much every different industry and sector and product matters. And so when we take that information, that’s you know, kind of you know, that lesser information set that we have quarterly to update the numbers and say, well, this is how much this bid has grown. We’ve given GDP the best base to sense, put everything together and give you the most accurate information out of the most recent quarter.

Gene Tunny  24:14

Okay, we’ll take a short break here for a word from our sponsor.

Female speaker  24:19

If you need to crunch the numbers, then get in touch with Adept Economics. We offer you Frank and fearless economic analysis and advice. We can help you with funding submissions, cost benefit analysis studies, and economic modelling of all sorts. Our head office is in Brisbane, Australia, but we work all over the world. You can get in touch via our website, http://www.adapteconomics.com.au. We’d love to hear from you.

Gene Tunny  24:48

Now back to the show. So this they supply use tables. That’s what you’d call an input-output table. Is that right Rob? This is an input-output table your developing, how these industries interact with each other, the flows of resources between them and then what goes to sales, what’s exported, etc?

Robert Ewing  25:09

It’s a very closely related cousin of an input-output table. So there are a couple of pretty technical differences between supply use tables and input-output tables. Importantly, in Australia, the input-output tables that we produce are a lot more detail. So I mentioned supply use, we’ve got about 60 industries times 300 products, when we get to the input-output tables, we’re able to expand that to about 115 industries and a bit over 900 progress. So we can give you more detail in the parameters as a couple of other technical details. But it’s broadly that same sort of concept. It’s trying to understand what is produced in the economy, and how it flows through the economy. And they’re a very powerful analytical tool.

Gene Tunny  25:53

Yeah, absolutely and Rob what I think’s interesting is you meant just, I mean, you’ve got all of this data and you’ve got these three measures that on a quarterly basis, you’re trying to get as close as possible. And you mentioned that, look, you have to make some adjustments. I mean, so there’s some judgement involved, but that’s going to be informed as much as possible by other data or whatever information you’re gathering. And you talked about he getting data from a huge range of sources, government and your own surveys, from the private sector. You probably can’t go into this. But what I’ve what I’ve been surprised at now is just how much real time data or up to date data that banks for instance, so the who else is Dun and Bradstreet, they’ve got whatever they’re called now, they’ve got more up to date data about how things are going into the economy, I presume the ABS is, is trying to get hold of some of that data.

Robert Ewing  26:52

Yeah, no, absolutely. And we have, for instance, on the bank data, we have a new publication, it’s been running since February this year, which is the monthly household spending indicator. And so that takes information from bank transactions, so credit card and debit card data. And it gives you a much broader picture of what’s happening to household consumption, then the retail trade survey does, the retail trade survey was a great survey. It’s one of the longest running surveys, and it’s very closely watched. But it only tells you about a third of consumption. Because today, in terms of things, you can walk into a store and buy or kind of buy through a retailer online, that’s only about a third of what households consume every day, when we’ve got the amount of the household spend indicator variable to expand out to more categories of consumption. And so that covers a little over two thirds of consumption of households. So that kind of finer grained data is definitely something we’re very interested in. Another example is the monthly business turnover indicator, which is another fairly recent publication. I think that started a little under 12 months ago, that takes ATO business activity statement data, and it gives you a month by month picture of how the turnover is moving in different industries. And so these are giving us that month by month picture of what’s happening in the economy. And that’s something which in the past, we’ve had to rely on the quarterly surveys and the quarterly national accounts for and I think that’s really exciting development with the data becoming more available. And, you know, the ABS showing its capability to transform that data into useful insight for people is definitely something we see is a growing kind of part of the business but going forward.

Gene Tunny  28:51

Yeah, absolutely. And it’s so important as we’re trying to monitor the economy as the RBA and the Treasury are monitoring the economy. And, and you know, that obviously feeds into policy. But yeah, that’s, that’s something for another another time. Right, on the supply use balancing, if you just got some a little bit, a few more minutes, while you’re trying to get these data internally consistent, it’s all about making sure everything adds up. All of these data are consistent. It’s the story makes sense. This balancing, I think I asked you about this on LinkedIn, what do you do, how do you do this, this a some sort of algorithm, is it..

Robert Ewing  29:31

So it’s a mix of things here? So for a lot of the large things, it is a human being looking at the data and looking at the whole picture and making a decision about all based on what I know, this is the bit here that must be wrong. So there’s a lot of it, which we’re not yet able to put into an algorithm. But if you think about I mean, I described that kind of 60 by 300 kind of matrix, you know, it’s an enormous number of cells. Once you’ve got to fill in, and so once we get past those things where you can reasonably make a human judgement. We use an algorithm, it’s a constrained optimization algorithm. And what we do is we tell that algorithm, the same things that we know from the national accounts framework. So we tell that algorithm, things like, Well, we know, you know, this supply has to equal this use. And we can also give it some other parameters, kind of like, it’s rare that the financial sector consumes many sheep directly, for instance, you know, so you can apply a bunch of constraints there, and it can then kind of go away, and it can algorithmically kind of find a sensible solution for you based on that. But the reality is that, you know, right, this second, there isn’t a good solution from a lot of this beyond a human being who can look at and just make a judgement about what makes sense here, you know, does it make more sense to kind of say that the excess rolled steel production is going to be going to export or to households? Because the algorithm is not going to have a view on that one, but the human is going to be able to say, no, no, we must have just mismeasured some exports there. Yeah, let’s have a look at that.

Gene Tunny  31:15

Yeah, gotcha. Okay. Yeah. So some constraint optimizations and judgement. Very good. And right so Rob, this is all fascinating. Is there any, any other points you think would be important for, for us to understand in detail to appreciate just the magnitude of this task? So I think I’m hoping we people understand how important these data are. And I’ve covered that in a previous episode. But is there anything else we should know, in terms of just the, the, you know, what, what are the challenges for ABS in doing this? Because it does take several months, doesn’t it, you do this on a quarterly basis, and it comes out, it comes out two months after the end of the quarter, doesn’t it? So there’s obviously a lot of work involved.

Robert Ewing  32:02

Yeah. And I think I divide that into two halves. So there’s about a month of data gathering and so this is kind of you the various surveys going out and measuring things. So it’s kind of, you know, working with partners, such as the banks and private sector, and governments, and so on to get the information streams. And then there’s about a month of what we call compilation, sitting down, running the pros, running the computer code, seeing how the numbers make sense or not. And then ultimately, you know, writing media releases and producing a publication and getting it onto the website. And just to put it in a plug, also, for our new product, which is, we are each quarter producing a nice little list of, you know, 10 to 15 things you should know about the national accounts. It’s nice and easy to digest. And you can just find that on the ABS web page. So if you find the idea of clicking on a GDP release a little intimidating, we’ve got some much more user friendly products available now. But yeah, it’s as you say, it’s about a couple of months to put it together. It’s the work of hundreds of people across the ABS, once you count the people in the survey divisions, the data acquisition divisions, all the other publications such as the balance of payments, the capital expenditure, the business indicators publication, which all feed into the national accounts, the communications team, the compilation team it’s hundreds of people across the ABS contributing to every quarter. And they are not to forget the 1000s of businesses and households answering surveys out there. I mean, the most fundamental thing is just how grateful the ABS is for the people who take the time out of their busy business and personal lives to kind of give us this data that we need to tell Australia about what’s happening.

Gene Tunny  33:57

Yeah, yeah, that’s a good point. Yeah, absolutely.

Robert Ewing  34:00

And one final point I’d make is, GDP is the figure that gets all the fun headlines. But the national accounts are a very rich publication. They tell you about a lot of different parts of the economy, they can take a bit of expertise to understand. But there’s a lot of information in the national accounts beyond GDP. It tells you about how each state is tracking, it tells you about the consumption of different goods and services in the economy. It tells you about the balance sheets of households and governments and businesses. There’s a massive amount of information beyond that top level GDP figure in there and I think a lot of the criticisms you sometimes hear about GDP not taking into account depreciation or not taking into account other things. An awful lot of that is in there somewhere. But it is a very big publication. I know it can be a bit intimidating to try and find anything in it.

Gene Tunny  34:58

Yes, yeah, but you’re right there, Rob. And, I know that there is work going on. And it may not be in the national accounts, the core national accounts, we do have satellite accounts. I know you’ve, you’ve tried to estimate the contribution of different sectors like tourism. And I’m trying to remember if you’ve done natural, natural capital estimates. So I think there are some other estimates where you’ve tried to estimate them, I’ll have to check. I remember, this is one of the issues or one of the things people are concerned about is that GDP doesn’t take into account environmental degradation. And so but there are estimates, there are people who have looked at those sorts of estimates, I’ll have a look offline and just add something in the show notes if I need to. But I know that that’s one of the issues people are concerned about.

Robert Ewing  35:49

And it looks, it’s there. It’s not my area of expertise. But there’s a whole range of work that the ABS has done on environmental economic accounts, which is really bringing together those two data sources. And I think that’s one of the unique advantages of the ABS is because we kind of sit in the middle of this data architecture, we can bring together the data that we have from these different domains and put them next to each other. And we can see how the story of the environment and particular aspects such of water align with the story of the economy. And you know, it’s something which is still like an ongoing piece of work to fully develop all the products there, but something we’ve been actively developing over the past.

Gene Tunny  36:32

Okay, very good. Rob, in from the ABS. Thanks so much for your time. I really enjoyed that. And, yeah, I look forward to seeing more of your contributions on LinkedIn. I think that I’ve been learning a lot about the work that you’ve that you do over at ABS and I’ve got a renewed appreciation for that work. So thanks again, Rob. Very good.

Robert Ewing  36:53

Thank you very much.

Gene Tunny  36:55

Okay, that’s the end of this episode of Economics Explored. I hope you enjoyed it. If so, please tell your family and friends and leave a comment or give us a rating on your podcast app. If you have any comments, questions, suggestions, you can feel free to send them to contact@economicsexplored.com And we’ll aim to address them in a future episode. Thanks for listening. Until next week, goodbye

Thanks to Josh Crotts for mixing the episode and to the show’s sponsor, Gene’s consultancy business www.adepteconomics.com.au

Please consider signing up to receive our email updates and to access our e-book Top Ten Insights from Economics at www.economicsexplored.com. Also, please get in touch with any questions, comments and suggestions by emailing us at contact@economicsexplored.com or sending a voice message via https://www.speakpipe.com/economicsexplored. Economics Explored is available via Apple PodcastsGoogle Podcast, and other podcasting platforms.

Categories
Podcast episode

GDP & the National Accounts: What they are and why they matter w/ Brendan Markey-Towler – EP153

The National Accounts are a huge intellectual achievement and an incredibly useful set of data, including GDP and its components. Chatting about the National Accounts with Economics Explored host Gene Tunny is fellow economist Dr Brendan Markey-Towler, author of the Substack newsletter Australian Economy Tracker. Brendan explains how the National Accounts help us track the current state of the economy as well as longer-term trends, such as shrinking manufacturing sectors and growing services sectors in many advanced economies.

You can listen to the episode via the embedded player below or via podcasting apps including Google PodcastsApple PodcastsSpotify, and Stitcher.

Links relevant to the conversation

Brendan’s Australian Economy Tracker Newsletter

Brendan’s post discussed in this episode

Planet Money episode on Simon Kuznets

Australian Financial Review article (pay-walled, alas) which reported “Federal government business generated $1.7 billion in revenue for the big four accounting and consulting firms over the past five years – though the government has a different take on the contract value of that business.”

Transcript: ROI of education: how economists estimate it + US economic update – EP152

N.B. This is a lightly edited version of a transcript originally created using the AI application otter.ai. It may not be 100 percent accurate, but should be pretty close. If you’d like to quote from it, please check the quoted segment in the recording.

Gene Tunny  00:01

Coming up on Economics Explored.

Brendan Markey-Towler  00:04

So, that’s where we get the view that Australia is less and less a country that makes things and builds things. Construction, manufacturing declining as a share of GDP.

Gene Tunny  00:16

Welcome to the Economics Explored podcast, a frank and fearless exploration of important economic issues. I’m your host, Gene Tunny. I’m a professional economist based in Brisbane, Australia, and I’m a former Australian Treasury official. 

This is episode 153 on GDP and the National Accounts. What they are and why they matter. 

Chatting about the national accounts with me this episode, is my good friend and fellow economist, Dr. Brendan Markey-Towler, who started a new sub stack newsletter, Australian Economy Tracker. Brendan explains how the national accounts help us track the current state of the economy, as well as longer term trends, such as shrinking manufacturing sectors and growing services sectors in many advanced economies. 

In the show notes, you can find relevant links and any clarifications. Please send any comments or questions to contact@economicsexplored.com. I’d love to hear from you. I’ve been very grateful for all the comments on recent episodes. Your comments really helped me figure out the issues that you’re interested in, and the types of guests that you’re interested in hearing from. So, please keep the comments coming to me.

Right oh! Now for my conversation with Brendan Markey-Towler on the national accounts. Thanks to my audio engineer, Josh Crotts for his assistance in producing this episode. I hope you enjoy it. Brendan Markey-Towler, welcome back to the program.

Brendan Markey-Towler  01:43

Gene, it’s always a pleasure to be here. Sorry, I’m a bit husky today, but I’ve bruised my throat. I’d like to pretend that it was under heroic circumstances, but it was not.

Gene Tunny  01:52

Okay, well, thanks for participating. I understand it’s not damaging your throat, you’re able to talk, you’ve been talking all day. And you’re still happy to talk.

Brendan Markey-Towler  02:01

I could talk under wet cement, mate. So, a bruised throat isn’t going to stop me.

Gene Tunny  02:07

Well, you know, now, you can get a job as a rugby league commentator, possibly?

Brendan Markey-Towler 02:14

That’s true. I’m more of a union man. Yeah, but I will go with league. That’s good. 

Gene Tunny  02:18

Right oh, okay. So, the topic of today, national accounts, what it is, why it matters? You’ve started a sub stack and one of your first pieces that came out on the sub stack was on the national accounts. And you displayed a level of enthusiasm for the national accounts that is very rare. And it actually reminded me of just how marvelous the set of data – the national accounts are, and what a superb intellectual achievement. 

So, going back to the work of Simon Kuznets, and Colin Clark, who, was it Stone as well, Richard Stone, who formulated the methodology financial accounts, and then it was like a system a toss by the UN. So, I think, what your note did was it really helped us; well, it really reminded me of just how impressive those national accounts are. So, could you just tell us first, what you were trying to do in that note? And what’s your sort of general take on the national accounts, please, Brendan? Why do you think they’re so important?

Brendan Markey-Towler  03:28

Partly to justify why I had no friends at school. Because I get excited about nerdy stuff like this. But look, when you actually know what the national accounts are, they’re extremely interesting. And what they really do is they aim to provide a snapshot of the activity within an economy over a set period of time. So, in Australia, and throughout almost the world, I’m not sure of any country that doesn’t do it this way. It gives you a snapshot of all the activity that went on in an economy over the previous quarter. And the central number that depicts that activity is the number that we call gross domestic product. And gross domestic product is a measure of how much wealth was added to the economy, how much production, how much activity, and under the three great categories production, exchange, and income, or earning. That’s what the national accounts do. And they add that up into a single number, GDP. And that tells you how much activity went on in the economy over that quarter. 

Now, where it gets really interesting, is that number not alone would be kind of cool. And we talk about the GDP growth rate. That’s what we mean when you hear on the news that people say economic growth or the economy grew by, that’s what they meant that GDP number increasing or decreasing. But where it gets really interesting is that we approach GDP in three ways. And you can think of this as looking at the economy as the same thing, but from three different directions. And that changes the way that you interpret that number. So, we call these GDP I, or at least I call them GDP I, GDP O, and GDP E. That is, GDP expenditure, GDP income and GDP output. 

And what those numbers are doing are adding up GDP, the activity in the economy, looking at that activity from one to three ways: as a production, as an expenditure, and as an income, right. So, if you think about it this way, when you go down and you buy something that’s dear to our heart, here in Queensland, you go down into buy your coffee, there’s three things going on, there’s three ways that they get that same transaction gets measured and add to GDP. From the expenditure side, the expenditure that you make, when you buy that coffee goes into GDP E, and we add all of those up together, and we get GDP. That expenditure becomes income from the perspective of the person behind the bar. And that gets added up into GDP income. 

And there’s also an interesting concept of gross value add, which is how much value has been produced by that transaction. The way that we measure that in GDP O, is we take the value of the output that was sold and subtract the value of the inputs that went into it. And that by definition, that’s the value that was added. 

So, that’s the three ways that we add up GDP and we get an interesting view of the economy from that. A little bit further breaking that down, obviously, you can break that down to the level of the individual transaction. But the you know, you don’t get a huge amount of information that you get so much information, you have no information. So, we categorize at a high level, these different activities to get a sense of what’s driving GDP. So, within GDP E, the expenditure, which is the most popular and most focused on of the national accounts measures of GDP, we break down expenditure by consumption, investment; in Australia, we break down by housing, as well, government expenditure, both consumption and investment, and net exports.

Gene Tunny  07:34

And by investment, we mean capital investment, we mean expenditure on capital goods. So, we mean, new housing developments, or we mean, new, non-residential buildings, new schools, new factories, new capital equipment that’s purchase.

Brendan Markey-Towler  07:55

That’s right. Yeah. So, in Australia, we call it gross fixed capital investment, which is at the addition to the capital stock of the country in the capital stock of the country is; in Australia, again, we trade a little, perhaps, oddly, that we add housing into that. But factories, equipment; we actually add intellectual property as well. So, science and technology research get added into that figure. And so that’s what we that’s, that’s the way that we break down the economy. 

So, when we break down GDP E that way consumption, investment, government spending net exports, we get a sense of which sector of the demand side of the economy is pulling the economy along. Is it household consumption? Is it buying new houses or building new houses? Is it businesses investing? Is it government consuming, spending money? Or is it government investing? Or is it coming from the international sector? And that gives us a lot of information about the activity within a country, it also gives us information about what might be dragging economic growth as well. So, that’s expenditure. 

Another really interesting measure, well, I mean they’re all interesting, but the second measure GDP O – GDP output, sometimes called GDP gross value add, gives us a sense more of the supply side of the economy. 

So, expenditure gives us a view of what’s driving the economy on the demand side. GDP O gives us a view of what’s driving the supply side. So, we get GDP in Australia, broken down by industry. And that’s where it gets really interesting because we can see which industries are adding the most to GDP. So, that’s cool. We can say, oh, mining adding more? Or how much is mining adding to GDP and how much is it driving or dragging on GDP? Ditto for professional scientific and technical services is another one that we use, agriculture and fishing, public administration safety; how much are these sectors adding to GDP and how much are they dragging or driving GDP. And then finally, the GDP I number. This is typically not quite as informative as the others, which is kind of ironic because it’s the easiest to add up because we just look at the tax returns. GDP I, breaks down GDP by income. And in Australia, we do it by what we’d call the greatest states of Australian society. So, wage earners, non-financial corporations, financial corporations, and government. And we can get a view of who’s earning the income within GDP. How what of that GDP that’s expended and outputted. Where is the income from that activity accruing to? Is it accruing to wages? Is it accruing to company profits? If it’s an accruing company profits, is it occurring to financial or non-financial companies? So, that’s some of the really interesting stuff that we get from GDP, it gives us this, really, especially in Australia, because our accounts are quite amazing.

Gene Tunny  11:05

Yeah, we’ve got some of the best in the world for sure. 

Brendan Markey-Towler  11:09

They really are and we get a really rich view of what’s driving and dragging the Australian economy. What’s creating the wealth in our economy and what’s potentially dragging on the wealth of our economy. And kind of, we get a sense as well, where it’s going.

Gene Tunny  11:26

Okay, so the few things I want to talk about there, Brendan. Okay, so you mentioned that GDP; well, is it an approximation of the addition to wealth? Let me think about this. I mean, part of it is in addition to wealth, to the extent that you’re increasing the capital stock, but then part of it is consumed, and then part of the investment is consumption of fixed capital. So, I mean, it’s national income really, isn’t it? I mean, it’s related to wealth. Yes. So, it’s certainly related to that. It gives us a picture of our national income. I think national income was the original term for it, wasn’t it?

Brendan Markey-Towler  12:11

Yes, although national income gets a little trickier because the we focus on GDP, because it’s really limited to the geographical definition of the country. And that distinction was made early on in the development of the methodology, because national income is a bit fuzzier because it’s typically added up by nationals, rather than by where the activity occurred. So, that’s why the classic example that we give in an economics course, is that national income for a country like Luxembourg is, I think, Ireland, sorry. National income for a country like Ireland is actually much higher than its GDP, because a lot of its nationals live overseas. So, there’s few distinctions that we make within it. But really, what it’s giving you is a view of the activity that’s occurred in the economy, the economy being that system of human behavior, why we produce and exchange stuff that we need for everyday life. And so obviously, that adds to the stock of wealth in the economy, because some of that gets consumed and taken out and other elements of it gets allocated to the national wealth. 

So, yeah, it’s a flow metric in the classic distinction between stocks and flows. It a reflection of the consumption and investment activity in an economy during a particular period.

Gene Tunny  13:40

Yes, it was developed during, well; the need for it became obvious during the 30s, when they were trying to quantify the extent of the Great Depression, I think Kuznets produced a report for the US federal government that strangely became a best seller. I mean, it was the first time someone had produced numbers like this. There’s a great planet money episode on that. I’ll try and find it and link to it in the show notes.

Brendan Markey-Towler  14:09

Well, that’s a good point, right? Because before then everyone kind of knew when times were good, or times were bad. And so, you could tell there were panics and manias and crashes as Charles Kindleberger famously said, but before the national accounts were developed, we never really were able to quantify what that was. And a lot of this was crystallized by John Maynard Keynes, his famous book, The General Theory of Interest, money and employment. I’ve got that wrong, interest money I think I got three. I’m one of the few in my in my generation, I think who actually read the book, which is, which is why it’s embarrassing I can’t remember the name because we always refer to it as the general theory.  And what Keynes was trying to do there was give a theory of why we experienced these manias, panics and crashes, you know, boom and bust. And the problem was that when he wrote it, he was dealing with a lot of abstract thoughts and that needed to be measured. And I’ll actually give a little plug here for our home state of Queensland because Queensland was at the forefront of this, currently the building out at UQ, which houses the School of Economics, the University of Queensland, the School of Economics there is housed in the Colin Clark building, which is kind of ironic because Colin Clark didn’t become an academic at UQ until much later in life, I think around the 1980s. But Colin Clark was at the forefront of developing the methodology, not only for what the national accounts are, but how you actually design the surveys that add up those numbers and find out what the numbers are. 

Gene Tunny  15:49

And he’s quoted in Keynes’s book because Keynes used his estimates of consumption spending for Great Britain, if I remember correctly, in the general theory. 

Brendan Markey-Towler  16:01

And it’s kind of funny. So, Colin Clark who came out here to Australia and did a tour of Australia and he was the hotshot wizkid political economist from Cambridge. And he met with all of the premiers because back in those days, we understood the constitution. So, the premiers were much more powerful than the prime minister. And when he came up here to Queensland, the premier at the time William Forgan Smith, which the alumni of UQ will know, is that is the main building at the University of Queensland. Kind of, a nice little coincidence. Forgan Smith basically said to him, look, do you want to come and be my adviser on all things economics? As Forgan Smith was a great reformer and trying to develop the Queensland economy, he needed to be able to measure the size of the Queensland economy: what was driving, what was dragging, what was causing development, what was dragging on development. And there’s a famous letter that Colin Clark writes back to Keynes to say, I’ve been offered a job to basically become the shadow premier of Queensland. I’m not going to turn that down. And Keynes, I think said something to the effect of where is Queensland. So, then, Colin Clark came out, join the Queensland Statistical Bureau and, he was instrumental in the development of the national accounts and as a point to why the national accounts are so important. While Colin Clark was doing that, he’s obviously thinking about what goes into an economy? What is an economy? What exactly does it mean to say an economy? Because when you actually; we all kind of know what it is, is the economy stupid?

Gene Tunny  17:44

It’s an abstraction, isn’t it? 

Brendan Markey-Towler  17:47

But it is an abstraction. And so, he had to think about, Okay, what does it actually mean? What is an economy, what counts as economic activity? And this is becoming very pertinent again, in these days, where we’re talking about things like Facebook and Amazon and Google where a lot of the activity that goes on there, we sort of think of as economic but it doesn’t measure it. But what happens as a result of Colin Clark thinking through these questions, is he’s starting to develop views of how economic development occurs. So, he ends up writing a large book, which sort of became a classic and development economics on how economies develop, what the basis for economic development are, what the settings for economic policy should be to encourage development. Particularly important question here in Queensland, which was a quite underdeveloped economy at the time.

And as a result, he became a very close adviser to Bob Santamaria, who those diehard fans of Australian politics will know was instrumental in the foundation of the Democratic Labor Party. So, this is the guy who invented a lot of the methodology behind the national accounts. So, when you understand something at that level, when you understand what an economy is, when you know how to measure it, imperfect as that measure may be, you get really rich insights into how an economy is tracking over time. And you get really rich insights as a result that develop over a long period of time of working with these things of what drives economic growth. You can situate those numbers in a history that tells you why the economy is growing, or why it’s not.

Gene Tunny  19:32

Yeah. Where do you get that Colin Clark story from? Is that in that book you keep talking about by, was it Millmow?. 

Brendan Markey-Towler  19:38

Yeah. Alex Millmow, A History of Australasian Economics Thought. I think that’s where I got it from. Yes, it is where I got it from. It’s a really good book because Alex points out that a lot of Australia’s economic contributions to economic thought came from really practical questions like this. How do we measure?

Gene Tunny  19:57

Absolutely. Yeah, yeah, absolutely. Okay, we’ll take a short break here for a word from our sponsor.

Female speaker  20:07

If you need to crunch the numbers, then get in touch with Adept Economics. We offer you frank and fearless economic analysis and advice. We can help you with funding submissions, cost benefit analysis studies, and economic modelling of all sorts. Our head office is in Brisbane, Australia, but we work all over the world. 

You can get in touch via our website, http://www.adepteconomics.com.au. We’d love to hear from you.

Gene Tunny  20:36

Now back to the show. Okay, now, I did want to go back to the point you made about the difficulty of well, the issues around the modern economy and the India head, etcetera. There was a great lecture that John Quiggin, who’s a professor at UQ. And if any Australian economist is going to win a Nobel Prize, it’d be John. I mean, he’s one of the most cited academic economists that Australia has. I mean, maybe, Warrick McKibben could win one. So, but yeah, certainly, John is;

Brendan Markey-Towler  21:11

I always like for John Foster personally.,

Gene Tunny  21:15

Well, John Quiggin, is incredibly distinguished economist and his view at the this lecture he gave was that the problem with GDP is that it’s gross, its domestic and its product. Okay, so we’ve already talked about the domestic issue. So, the fact that you could have a lot of production, but if all your incomes remitted overseas, okay, because it’s just foreign mining companies producing and sending profits home, and then you may not see all of that benefit. But the point he was making is it because its product, and it’s measured at market prices, what you could be missing out on is consumer surplus, you’re not necessarily measuring the benefit to consumers, because all of these products are provided for, well, a lot of them for free. But yet, the foreign company makes money out of you in some other ways, because it’s monetizing your attention, isn’t it?

Brendan Markey-Towler  22:11

Yeah. And so, this is a debate that’s been really reopened, it’s been a perennial debate in economics, and there’s a lot of interesting ideas floating around, inspired by it, which is that when we talked about, you know, how GDP is added up, we talked about the exchange, okay. But the only way that we really observe and exchange is by the exchange of money, right? So, the price multiplied by the quantity of goods or services sold. Now, the problem merges; what happens in a world full of freemium models? What happens in a world where the price of a Facebook membership is zero? That sort of kind of, well, I don’t particularly like Facebook. So, you know, I would challenge just how much consumer surplus is creating, but there’s, you know, many people would argue that there is a value added.

Gene Tunny  23:11

I think TikTok is creating the most at the moment. Especially among the younger generation..

Brendan Markey-Towler  23:16

Massively, yeah. the only thing that shows up in the national accounts from Facebook, Google, TikTok, Instagram, is the data sales. That’s the only thing that shows up in the national accounts. I mean, apart from the marketplace exchanges that go on as well in the Facebook marketplace, and so on like that. But really, it’s ultimately the advertising for Google the sales of data from all of them. That’s the only thing that shows up in the national accounts. So, but there’s more than that, as well. Another problem, And Peter Thiel has recently raised this issue.

Gene Tunny  23:53

Oh, the billionaire? Right.

Brendan Markey-Towler  23:57

The chap who founded PayPal, he thinks that we’ve actually had no economic growth or very little economic growth in the past 70 years. And the reason he says that is because he contends that what is observed as economic growth in the past 70 years, is actually just us bringing production and exchange; valuable production exchange that used to happen in the home, into markets. So, cooking, cleaning, keeping the house in order, gardening; all this stuff gets done on marketplaces, rather than in the home. And that’s a bias in GDP. It doesn’t measure that stuff because it’s not on a marketplace. It can’t be observed. So, that’s another argument. 

You know that GDP doesn’t measure the actual value that’s being created. Now, the problem ultimately is, this goes back to a problem of micro economic theory, which is what is utility? And what is consumer surplus? And actually, from my perspective, why I ultimately say, look, let’s stick with GDP. It’s the worst measure we have, except for all the other things. Some countries have toyed with measuring gross national happiness. You know, New Zealand is toying with that at the moment, Bhutan famously measured it. The UN uses the Human Development Index, which is a weighting of GDP per capita literacy rates and life expectancy, I think.

Gene Tunny  25:31

All of which are highly correlated, aren’t those?

Brendan Markey-Towler  25:33

Yeah, and so, that was a March Ascends Brainchild, Jagdish Bhagwati famously said, well, yeah, they’re correlated. So, what are we talking about here? So, all those debates over replacing GDP ultimately, were reduced to a deep, deep philosophical problem, which economists are not well placed to solve, which is, what is value? What is good, what is true, what is beautiful? And I got some views on that. But as an economist, I ain’t got nothing to say about that. And so, when economists start dabbling in it, you kind of go, I used to be a fan of the happiness literature. But now I read and go, ah, this is, you know, it’s very simplistic. We’re going to use subjective wellbeing measures to add up Gross National Happiness. Okay, fine, that’s a really subjective and not very tangible measure. Whereas I can look out the window and see the cranes on the skyline here in Brisbane and see that’s an objective, measurable thing.

Gene Tunny  26:37

Well, it stood the test of time, hasn’t it? So, we’ve been using it for decades now. And there’s a general feeling that it does capture the state of the economy reasonably well. I mean, there are going to be people who grumble about it from time to time, but generally well, in Australia, at least when we had the recession, I mean, I always remember the 91 recession, because I was in high school at the time. And like, things just look bleak for anyone who was in high school and wanted to get a job. But then that was the period when retention rates at high school really ramped up. So, it was it was telling us something important there and it tends to; like it could give false signals, there’s a big debate at the moment over what’s happening in the US. But then look, the economy’s looks like it is slowing to an extent. There’s the impact of the Federal Reserve hikes. So, let’s wait and see how it all plays out. I mean, my feeling is, it’s generally a pretty good indicator of the state of the economy. 

Brendan Markey-Towler  27:38

I look bad, I’m a Queenslander first, Australian second, and as a result, I do have a bias which is towards tangible reality. Right, feelings are very ephemeral. And feelings are important, right? They are very important, but they’re really difficult to measure. And they’re very subjective, and they can be easily manipulated. Now, GDP can be manipulated as well, depending on how you count things up. But at the end of the day, it’s stuff that’s being produced stuff that’s being consumed. And it’s tangible, observable goods and services. So, insofar as I really have a criticism of GDP, my major criticism is that it really; I agree with Peter Thiel largely, biases us away from realizing the value that is produced in a house. 

And look, I’ve got a young, I’ve got a four-month-old son now so and my wife is at home, taking care of that. And I tell you what, that is incredibly mind blowing valuable work that she’s doing; doesn’t show up anywhere in GDP. Now, that doesn’t negate GDP. Because I think the solution to that is really, let’s just realize what GDP is actually measuring. Now, that does work in a political debate, because in politics and the way that the media works, you need a number and you need that number to be growing, otherwise, elections get lost, and so on and so forth. But when you’re, you know, when you’re doing grown up analysis instead of politics, I think the solution is to look at what GDP is actually measuring. It’s not a measure of value and if you think of it that way, then you’re wrong. Stop thinking of it like that. Think of it as it’s a measure of the production of stuff and the exchange of stuff within the economy, within the market that we can observe. Don’t try and start thinking about as a measure of all of the economic activity that ever happens in an economy. Just recognize the limitations, it doesn’t measure this stuff that goes on the household and that’s incredibly important.

Gene Tunny  29:51

Yeah, fair enough. That’s a good point. I’ll have to come in another episode to this issue of what’s in GDP? What’s out? What does it all mean? I’ll try and have that discussion in a future episode because there is a couple of other things I wanted to pick up on from your note; your note reminded me of a couple of things. And it’s the fact that this system is so beautiful, I mean, we end up getting from two different directions, possibly two different sets of data. I mean, we can look at what spend on consumption goods, final consumption goods, now, we have to be careful, we’re talking about final consumption goods and final investment goods, because what we’re trying to do is avoid double counting, we’re trying to get; because there are a lot of business to business transactions, businesses selling to other businesses inputs, so you have to take care of all that and make sure you’re not double counting title output, you want the expenditure on final goods and services. 

So, if you look at that, that ends up telling you what GDP is, once you add exports, subtract imports, because, well, if you import something, then you don’t have to produce it here. So, there could be stuff that shows up a consumption spending or an investment spending that’s imported, and we didn’t produce it here. So, you have to subtract it. And likewise, if we’re exporting something, well, we produced it here, we know we produced it here, then that adds to our output. But then, you look at spending data, on the other hand, you can look at income data. So, you are saying, look at the wages data, look at the profits data. And yeah, I guess it is coming from the ITR. I’m not sure exactly where the IBS gets it from. But I mean, that’s a likely source. I do surveys of businesses.

I’d have to check exactly how much they’re using ATO data, but I know they do surveys of businesses to get that information. They’ve got a household expenditure survey, they’ve got surveys of, well I guess they got their business server; I’d be looking at what they spending on capital goods. Looking at what they’re earning. And so, they build up this picture of earnings that way, and also the gross value added in the business. Which as you described, is their revenue less their production costs, and wages are part of the value added to. So, wages plus the gross operating surplus, is your value added in the business?

Brendan Markey-Towler  32:21

Yeah, it’s a very slippery definition, because it’s not quite profits. But it’s, you know, the value of inputs minus the value of outputs. And that by definition has to be the value that is added by that business to the economy, insofar as we can measure it.

Gene Tunny  32:35

This is because we’re talking about gross domestic product. So, we haven’t subtracted for the depreciation of capital stock, because some of the investment that occurs is just replacing existing capital stock. So, the building wears out and we have to replace it.

Brendan Markey-Towler  32:52

Too hard. We set that aside. Depreciation is very funny thing to talk about.

Gene Tunny  32:56

Right? Yeah. Well, we’ll leave that for now. You got time just to chat about your great quote? I should have brought it in earlier. You use these different perspectives on GDP to provide a really nice summary of what’s been happening in Australia. I thought this was very good. Exactly. Okay, so after you analyze where the growth has occurred, and you know, it’d be good if you could explain this at the moment. You concluded this; to put it somewhat tribally, Australia is less and less a country that derives its wealth from making and building things. Still a country that makes its wealth by digging stuff out of the ground and renting houses, and more and more a country that consults and cares. Could you please explain how you came to that conclusion, Brendan?

Brendan Markey-Towler  33:53

Well, you so what I did there, this is one of the most informative aspects of the national accounts I’m very interested; everyone focuses on the demand side of the economy, because we’re all Keynesian.

Gene Tunny  34:07

What we’ve been heavily influenced by Keynes, yes. There’s no doubt about that, whether we’re Keynesian. So, that’s another question. You can go ahead. Yes.  

Brendan Markey-Towler  34:13

We are all Keynesians. But the supply side of the economy is super interesting. See which sectors of the economy are generating the wealth. Now, the way that you can do that is by looking at gross value add, right. So, then you take the gross value added by each industry divided by the total GDP and you get the share of GDP, economic activity, economic value that is being created by that industry. And you can track that over time. Now, the problem with that data why almost no one really uses it? Some people do, but almost no one does. And you’ve used it, Gene, is that there’s a lot there, the ABS breaks the economy down by I think its 20 sectors. possibly 25. So, you’ve got to kind of cut it down to get some useful insights from it. 

So, the way I did it was alright, let’s cut out everything that’s less than 5% of the economy and look only at things that produce more than 5% of Australian GDP. Now, no sector really produces more than about 15. But there’s a clear standout. And there are clear standout trends once you do that, and you clean the graph up by eliminating all the Martin “minor sectors”. And you see some very strong trends. 

Trend number one that’s quite striking, and I should emphasize, this is all by real data. So, we hold prices constant to see what’s going on at the volumetric level in each of these sectors. So, we hold P constant, and we look at what’s changing in Q. Q is for quantity. And so, there’s benefits and costs to doing that. But it’s valuable as an exercise as long as you’re aware of the limitations of doing that. First interesting thing, manufacturing and construction are in decline in Australia. They’re not producing as much value add. In volumetric terms, they’re not producing as much value add anymore. They’ve been declining for the past 10 years as a share of GDP. So, that’s where we get the view that Australia is less and less a country that makes things and builds things; construction, manufacturing declining as a share of GDP.

Gene Tunny  36:30

So, with manufacturing, we had a car industry once, we subsidized a car industry, we tried to buy ourselves a car industry, and it just could not be viable on its own. And there wasn’t any more money we could throw at it to keep it open. 

Brendan Markey-Towler  36:48

And you look at somewhere like Maroubra or Ipswich. Which would you know, once kind of manufacturing ish areas in Queensland. Maroubra main manufacturing now is government contracts, building bullets for the Australian Army.

Gene Tunny  37:03

And do they build trains, still?

Brendan Markey-Towler  37:06

They do now. Yes, Maroubra now has a trains contract to build trains for the Queensland Government as well. And I think Ipswich still has a little bit of a train industry as well. But really not too much, by the way of price manufacturers. It’s not to say that it doesn’t exist, and it’s not to say that it’s very valuable. Queensland, for instance, has very vibrant medical manufacturing sector. That’s kind of grown up on the back of our extremely good hospitals and medical research. But generally, across Australia, the story is one of the car industries; we don’t really make stuff anymore. It’s just not competitive to build stuff. And so, that number is reflecting something that you see a lot when you go down to Fortitude Valley here, which, you know, the state would like to think Silicon Valley. Yes. Anyway, it’s Fortitude Valley, Queensland Silicon Valley, you see that a lot of the companies there just want to grow big enough that they can afford to offshore their manufacturing elsewhere. And the classic one is, I think Trivium, the electric car battery manufacturer, which is, as soon as they got big enough, they got a loan from the Queensland Government and then went to build factories in Tennessee.

Gene Tunny  38:17

Is that right? Is that a good use of taxpayers’ money?

Brendan Markey-Towler  38:21

Well, I’m completely agnostic on that. So, that’s what’s that number is reflecting. Similarly, construction,  this runs a bit counter to the crane index that we’re seeing in the city at the moment, but construction has been adding less and less to the economy. It’s not just large construction projects, but construction is declining as a share of GDP. 

Gene Tunny  38:48

Well, I’ll have to look at this. But I think what could be explained is 10 years ago, we had that massive project up in Gladstone at Curtis Island where we built the three LNG terminals or what are they? Refrigeration or liquification facilities. They turn the methane that comes from the coal field, the coal seams to liquefy it so, they can put it on a boat economically and ship it to Japan or Korea. And that was like $70 billion.

And it basically doubled the level of capital expenditure in Queensland at the time. It’s absolutely extraordinary.

Brendan Markey-Towler  39:31

There’s a huge effort on part of government corporations to get that going. 

Gene Tunny  39:35

And then in the southern states, maybe a few years later, I can’t remember the time; we had that big apartment construction boom. So, that could be explained. I’ll have to look at the data but go on. 

Brendan Markey-Towler  39:48

And that’s what’s really good about the national accounts is kind of counter to what you’re seeing if you’re walking around, particularly, Brisbane at the moment. The number of cranes in the sky is astounding, but this is why statistics are important because what’s local loss to a particular area is not necessarily true of the entire country. And what’s even true of a particular sector of construction, residential construction, government construction is not necessarily true, it might mean that we’re not building that many mines, which ties into the second point, which is, although it has declined in volumetric terms, the mining sector is still the single biggest contributor to Australian real GDP. And it’s not close, it’s way up; I forget the exact number, but it’s well up towards 10% of the entire Australian economy value added is produced by the mining sector. 

So, that’s, you know, digging stuff out of the ground, selling it to various countries around the world.. Behind that really interesting sector is, is the rental sector. So, a lot of value added in the Australian economy. It’s the only sector that holds candle to mining is the rental sector where people are building houses and renting them.

Gene Tunny  41:03

Okay. So, when you analyzed that, did you look at the industry, is it rental services? Or did you look at what’s in the national accounts as; there’s rental income, isn’t there? What do they call it? Trying to remember what the label is in the national accounts, but they impute rent for owner occupied dwellings as well, in that sector. If I remember correctly.

Brendan Markey-Towler  41:29

Rental services. I’m pretty sure is the exact name of the sector.

Gene Tunny  41:33

Looking at it by industry. Okay. Yeah.

Brendan Markey-Towler  41:36

So, that’s an important point, right? Because rent to also shows up as an income segment as well. Not nearly as big there. But the value add is quite large. And so that’s saying, you know, the Australian economy is very much one that is dominated at the moment, by digging stuff up out of the ground, and then sending it offshore, and providing housing for people. Those are the two biggest sectors of the Australian economy. And then, finally, the very long-term trend, we come to the third part of that bond ma that you so ably quaffed, which is, surprisingly, the sectors that are growing fastest as a share of the Australian economy are; you’ll have to double check me on this, but I’m pretty sure it’s called health care and social assistance.. And professional scientific and technical services. Those have gone quite strongly over the last few years as a share of GDP. 

Scientific and Technical Services is obvious enough, right? That’s the IT department and you know, the lab.

Gene Tunny  42:45

There’s professional too. 

Brendan Markey-Towler  42:49

Yeah. Professional Services is the big one. So, this is your consultancy lawyers. So on and so forth, right. It’s Eagle street, the consulting firms along Eagle street.

Gene Tunny  42:58

Where we are in Brisbane, in the top end of town, would you call it the big end of town? You’re sitting in water from place to the moment and the offices of Hopko Gannon, thanks, again for allowing us to use.

Brendan Markey-Towler  43:13

And so this area is growing really strong. I forget where the legal services are counted among professional service.

Gene Tunny  43:18

But I think I would be Yeah, sure.

Brendan Markey-Towler  43:21

They might be under administration, administrative services. But professional, scientific and technical services, basically, scientific and technical can kind of be in house. But a huge majority of that professional services is consulting, right? So, Australia is doing a lot more consulting as a share of GDP.

Gene Tunny  43:40

And this is business to business, typically? Business-to-business consulting services or business to government.

Brendan Markey-Towler  43:47

Business to government is the big one, especially here in Queensland right now. That’s not backed by a number. But that’s you know, that’s kind of;

Gene Tunny  43:58

There are numbers for the Australian Government. I’ll put them in the show notes, because I looked at what the Australian government has spent on the Big Four consulting firms like KPMG and PwC. And it’s hundreds of millions a year, right? It’s big money. 

Brendan Markey-Towler  44:12

And then, you go step below and the state governments will probably be even bigger again, because every consulting project by the Department of Public Works now gets a cut benefit cost analysis written by one of the big firms, right. So, just because of the procurement rules around that, so professional, scientific and technical services really growing as a segment of GDP, but also health care and social assistance. And so that I would posit is really a reflection of the ageing population. Ageing population, you need more health care and social assistance, certainly. That sector is growing very strongly – aged care.

Gene Tunny  44:49

Yeah. Which is NDIS too, the National Disability Insurance Scheme.

Brendan Markey-Towler  44:53

Absolutely massive, huge boom. You throw a stone in Brisbane and you hit NDIS provider, which is really not good, you shouldn’t do that because that’s naughty. And that getting on the back of Yeah, health departments are in Queensland; Queensland Health is the largest single employer in the state. That’s a massive sector. It’s a $20 billion in the state budget. That’s a big number, right? And we’re always trying to spend more on it. So, very big sector that. So, those are the two real growth sectors in the Australian economy. And again, I should stress by volumetric measures, right? So, notice that that kind of cuts against the mining booms like us, and that goes to the difference between real and nominal GDP. Real being a volumetric thing where we’re trying to hold prices constant, and the reason we do that is because nominal GDP could be growing because the actual underlying productive capacity of the economy is growing, or because inflation is growing. And real GDP tries to say, what’s the underlying volumetric productive capacity of the economy? How’s that growing and contracting. And in that measure, you really see the big growth sectors, mining is actually declining as a volumetric share of GDP as a share of real GDP, but it’s still the biggest by far professional, scientific and technical services, and healthcare and social assistance really, really growing. Yeah, that’s where the saying, that’s where my little trite way of putting it came from. Australia is less and less a country that makes things and build things. It’s still very much a country that digs stuff out of the ground and provides housing, but it’s more and more something of a white collar economy.

Gene Tunny  46:43

Oh, yeah. It’s postindustrial. We’re moving more to services. Yeah.

Brendan Markey-Towler  46:49

Natural I mean, with the natural resources sector.

Gene Tunny  46:52

Yeah. that’s right. And I mean, because the world wants to buy our resources. And for the last year or so, they’ve been paying ridiculously high prices for them. It’s an open question over whether we want to sell it. Right. Well, yes. I mean, there’s the big issues there of course that we don’t have time for.

You’ve been very generous with your time, Brendan

Brendan Markey-Towler  47:22

You are very generous letting me on the podcast to talk to people again, Gene.

Gene Tunny  47:27

You’re a great talker. Always enjoy having you on.

Brendan Markey-Towler  47:30

Even with the bruised throat? Like I told you, I could talk through a wet cement.

Gene Tunny  47:35

Very good. So, any final points before we wrap up?

Brendan Markey-Towler  47:39

No, it just ends up on I ended up with the note of circling back to where we started, which is don’t underestimate the national accounts. They’re a really, really, really interesting data set. They give us such a rich view. We didn’t even talk tonight about how in Australia, they break down by state as well, so, we can get an even richer view of how the different states are doing because you know, Australian economy tracker – my blog.

Gene Tunny  48:06

Okay, right. On Sub stack, is it?.

Brendan Markey-Towler  48:09

Yeah, on Sub stack. Please subscribe and contribute to the Markey-Towler retirement fund. It’s founded on two points, which is that one, the perfect graph says more than a doctoral thesis and two, there’s no such thing as an Australian economy. There’s actually six different city state economies and two territories. So, the national accounts in Australia are amazing, not just because of the depth of analysis, they allow us on the supply side of the economy, but on the demand side as well. We get some really, really rich version. So, a plug to remember has to diehard nerds who didn’t have friends at school, but now we have the national accounts.

Gene Tunny  48:53

I’m sure you had friends at school, Brendan. Brendan Markey-Towler, that’s been terrific. I really enjoyed talking to you about the national accounts. 

Brendan Markey-Towler  

I really enjoyed talking to you, Gene. Thanks for having me. 

Gene Tunny  

Okay, that’s the end of this episode of Economics Explored. I hope you enjoyed it. If so, please tell your family and friends and leave a comment or give us a rating on your podcast app. If you have any comments, questions, suggestions, you can feel free to send them to contact@economicsexplored.com and we’ll aim to address them in a future episode. Thanks for listening. Till next week, goodbye.

Credits

Thanks to Josh Crotts for mixing the episode and to the show’s sponsor, Gene’s consultancy business www.adepteconomics.com.auPlease consider signing up to receive our email updates and to access our e-book Top Ten Insights from Economics at www.economicsexplored.com. Also, please get in touch with any questions, comments and suggestions by emailing us at contact@economicsexplored.com or sending a voice message via https://www.speakpipe.com/economicsexplored. Economics Explored is available via Apple PodcastsGoogle Podcast, and other podcasting platforms.