Technology, innovation and strategic competition: A conversation with The Hon Dr Andrew Leigh and Dr Robert Atkinson

In an era of heightened economic and technological competition, the United States and Australia are rapidly recalibrating their strategic approach. The technology policy landscape is front-of-mind across all aspects of business and government. This conversation explored a range of technology issues facing Australia and the United States including, the role of AI and automation in jobs, innovation and inequality, data sovereignty, the CHIPS Act, decoupling and technology regulation. 

The United States Studies Centre hosted a discussion featuring The Hon Dr Andrew Leigh MP, Assistant Minister for Competition, Charities and Treasury and Dr Robert Atkinson, President of the Information Technology and Innovation Foundation (ITIF) moderated by Dr Miah Hammond-Errey, USSC Emerging Technology Program Director. 

Event transcript

Please check against delivery

Michael Green: [00:00:15] Evening. Thank you very much for joining us. I'm Michael Green. I'm the CEO of the US Studies Centre at the University of Sydney. We're gathered here today, as many of you know, on the lands of the Ngunnawal people. We pay our respects to their elders past, present and emerging and extend that respect to all Aboriginal and Torres Strait Islander peoples today. One of our speakers, Rob Atkinson and I, are both from Bethesda, Maryland, just north of Washington, D.C., which until English settlers arrived in the early 1600s, was the land of the Piscataway and Accohannock people. And often when Americans hear the acknowledgement of country, they come away and say, we should be doing that in the United States. And it's sort of in that spirit that we're delighted today to have Rob Atkinson from Washington here, together with other distinguished speakers, to talk about technology competition, emerging technology policy, where the United States and Australia are aligned, which is most areas, and when we're not, which is in some surprising areas. We at the US Studies Centre have started an emerging technology program to try to look for policy solutions that bring together stakeholders from universities, from big tech and the private sector, from banks, from government, from the national security space. Miah Hammond-Errey, who runs the program, is leading a series of dialogues and initiatives to try to draw on the expertise across different sectors.

Michael Green: [00:01:43] Because new technology is like GPT4, which OpenAI just released, Quantum, big data, biotech, hypersonics. These are all creating all-of-government, all-of-society, all-of-alliance questions about literacy, innovation, export controls, ethics, national security, privacy. There are issues that require, as I said, not just governments, but all of society and closely aligned countries like the United States, Australia, Japan and others. So it's in that spirit, we're having this discussion tonight. Dr Rob Atkinson is at the US Study Centre this week as a visiting fellow. He's the President and founder of the Information Technology and Innovation Foundation based in Washington, DC. When I was at the Centre for Strategic and International Studies before coming here in August, I did quite a bit of work with Rob and his team. They are the pre-eminent technology policy think tank in Washington. There's a lot of heat, a lot of noise about technology policy issues, and Rob's team always adds clarity backed by data, often grabbed by Congress on a bipartisan basis to formulate legislation like the Chips and Science Act. We're delighted that Rob is here. We had excellent discussions in Sydney and today in Canberra we're whisking back to Sydney tomorrow morning.

Michael Green: [00:03:05] And joining Rob is, I understand, a friend of decades, the Honourable Dr Andrew Leigh, a member of Parliament, the Assistant Minister for Competition, Charities and Treasury. He worked in government for more than a decade in various shadow ministry positions. He was the parliamentary secretary to the Prime Minister from 2013. Has a PhD from Harvard and has written numerous books on innovation, economics, inequality and Australian politics, including one I just saw in Japanese. But I assume that's a translation you didn't, despite your accomplishments, did not actually write it in Japanese. And then moderating the discussion is our own Dr Miah Hammond-Errey. She's the Director of our Emerging Technology program. She has extensive experience in technology, national security and intelligence, with more than 15 years leading tactical, operational and strategic activities for the Australian Government. She's an expert on emerging trends, advises on complex challenges and national security threats, in particular emerging tech. Her PhD at ANU focused on big data and its impact on national security and intelligence. So I'm going to turn it over to Miah. We'll have a discussion among our distinguished panellists. Then we'll open it up for your questions. I'll come back and do a short benediction at the end, and then we'll have drinks so you can talk more about the subject. So thank you very much. And Miah, over to you.

Miah Hammond-Errey: [00:04:43] Thanks for the introduction. Andrew welcome. Joining me on stage are, as Mike just introduced, two incredible voices in Australia and the United States. The current focus on threats to regional stability, particularly in these alliances are of extreme significance bilaterally as well as AUKUS, the Quad.

Miah Hammond-Errey: [00:05:15] Yeah, I'm totally technical.

Miah Hammond-Errey: [00:05:22] As nation states attempt to understand, innovate and regulate the digital economy, let alone emerging technologies, we see an incredibly busy tech policy landscape in Australia and the US, as well as globally. The conversations span cyber security, recent strategies in the US and Australia, released just last in the last month, are examples of this, but also privacy, e-safety data, antitrust or competitiveness, domestic competitiveness for the economics people in the room as well as decoupling and national security concerns. And it can be really confusing for policymakers and consumers to differentiate and understand this discussion because it spans consumer goods, critical infrastructure, emerging technologies, defence technologies, as Mike just mentioned, as well as digital and Internet policy. It's really rare to get two insightful and influential voices of this calibre in the same room willing to talk about such significant and big challenges. So you're in for a treat tonight, as Mike said, I'll moderate a discussion and then you'll have some substantial time for questions. So start thinking.

Miah Hammond-Errey: [00:06:38] These two have collectively written so many books that when I tried to research this, I actually lost count. So we'll just say many, many books between them. But you've both published a book a year apart with the same publisher who I am hoping to get as a sponsor. I'll switch my book report. I'd like to ask you if you could offer some insights into each of your books. I'll give a short overview. Rob, Your 2019 book with Michael Lind is called 'Big is Beautiful Debunking the Myth of Small Business'. Can you give us an overview of the book and some key insights for our audience?

Robert Atkinson: [00:07:15] Sure. Well, thank you Mike. It's really a pleasure to be here. My main impression of Australia, particularly after I went in the capital today is boy, people are so nice. It's unbelievable. The guards were super nice to me.

Miah Hammond-Errey: [00:07:30] This is coming from a Canadian.

Robert Atkinson: [00:07:31] Well, yeah, but I live in the US. So anyway, so Mike and I wrote this book because there is an emerging narrative in the US that small businesses are really, really good and everything, you know, motherhood, apple pie, sainthood, small, big, bad. We don't want it, you know big companies are inherently bad. So what we did in the book is we just said, what's the evidence show about firms size and social benefit? And what we found was that virtually on every single indicator you could find, diversity, wages, productivity, R&D, exporting cyber security, environmental protection, social investment, big companies, on average, not every big company, but on average were better than small companies. And what our message was was not, well, let's discriminate against small companies. It was to say we should have size neutrality. In the US, for example, it's legal for a small company of under 15 people to discriminate on the basis of race. Why would we allow that? Why would we allow that? If you're a small company, you pay fewer taxes than you do if you're a large company. So our whole point of the book was we should have what we call size neutrality.

Robert Atkinson: [00:08:53] If you're going to have a rule, it should apply to all companies. If you're going to favour certain companies you shouldn't favour based on size. That's all it was. And it wasn't saying, by the way, that some companies could become too big or that there aren't potential issues in antitrust, particularly around conduct. But what we were saying is that large companies, particularly in the US, where we look, but by the way, this data were true, the data were true in every country we looked at Europe, Japan, Korea, the same dynamic is average productivity per worker goes up linearly by the size of the firm. So that's not to say 'oh, small businesses have no role, they're they're bad.' But it is to say we shouldn't have favouritism. So in France, my last point. In France, they have a law. They have to have a set a law that says when you hit 50 employees, the government is coming for you essentially. New taxes, new regulations, this and this and this. And so these LSE economists did a study and they found that there is a large preponderance of firms in France that have 49 employees, as you would expect. So that's really what the book was about.

Miah Hammond-Errey: [00:10:02] Andrew, your 2020 book with Josh Gans called 'Innovation and Inequality' is a look at the two concepts in the US. Can you offer an overview and share how this might apply in Australia?

Andrew Leigh: [00:10:13] Well, thanks very much, Miah, and can I too acknowledge the Ngunnawal people in whose lands we're meeting today and say what a hoot it is to have the US Studies Centre put me on the stage with Rob Atkinson. I first got to know Rob in the summer of 2001 when I was working at the Progressive Policy Institute, along with Lisa Middlebrook who is here in the audience as well. And Rob very generously had me on as a co-author for a report on digital governments and a report on the digital divide in which we talked about the growth of broadband and made a set of projections which completely fell apart once mobile internet came in and further proof, if any, were needed, that economists should stay out of the forecasting game. But Rob is an extraordinary thinker and iconoclastic one and somebody who is extremely ready to work with unusual collaborators. So. So, Rob, thank you for collaborating with me then, and it's a pleasure to collaborate now.

Robert Atkinson: [00:11:15] And Andrew is being modest because that report that he wrote was actually very influential in shaping US e-government policy in Congress.

Andrew Leigh: [00:11:23] Well, it's exciting to hear. So the, going back to your question there Miah, Innovation + Equality, which I have just for Mike sake in it's Japanese edition, was a book which Joshua and I wrote to try and address the idea that everything has to be a trade-off between growth and equity, that there's this so-called equity efficiency trade-off. And we wanted to point to a host of areas in which actually we can have faster growth and less inequality. For example, we point to the importance of getting superstar teachers in front of students in disadvantaged areas, and the benefits that can accrue are massive in terms of lifetime earnings. Indeed, if you look at the best research on the benefits of great teachers, it suggests that if you value a teacher in terms of the lifetime earnings boost they can deliver, they are literally worth their weight in solid gold. We looked too at the way in which entrepreneurial experience can sometimes not be evenly shared across a population, and that by providing access to mentors and money, we can ensure that we find more of those lost Einsteins and lost Marie Curie young people who have a brilliant idea but just can't implement it for lack of the systems that allow them to become innovators.

Andrew Leigh: [00:12:56] And we focus too, on the way in which our laws can impede people from engaging in start-ups. I've just finished Chris Miller's terrific book, 'Chip Wars', which talks about the way in which the so-called traitorous eight set up Fairbank Semiconductor in Silicon Valley and managed to kick-start the Valley's entrepreneurial culture. They could do it because in California, non-compete agreements are unenforceable, which allows people to leave their current firm and start up a new business with a new company. But right now, a fifth of American workers are covered by non-compete agreements, including gardeners, early childcare workers, and a whole range of people in frontline occupations. We think that a whole lot more innovation could be unlocked if non-compete agreements in the United States were not enforced in the way in which they are. So that gives you a little taste of what we were trying to do with Innovation + Inequality.

Miah Hammond-Errey: [00:13:58] Thank you. One of the policy suggestions from the book focuses on independent workers in what we might call the shared or gig economies. Can you explain the new relationships between customers, platforms and providers and what your proposal in that book outlined and then bring its applicability to the Australian employment market? And Rob I would welcome your thoughts coming after that as well.

Andrew Leigh: [00:14:23] Yeah, the gig economy has been terrific for consumers, but not always great for workers. We saw in Australia a very short period in which a handful of delivery drivers were killed on the job and the risk that for some of those drivers, they didn't get access to, their families, didn't get access to the sort of compensation that they would have been entitled to if that worker had been injured in a regular job. So part of this arises because we only have two categories in employment law. You're either an independent contractor or you're an employee. And employment law hasn't really kept up with the situation of a worker who is being told the circumstances in which they will work, the earnings that they will get if they work, but is not mandated to put in as particular number of hours. I think classically of delivery drivers, Uber drivers and the like. So we talk in the US context about ways in which you might broaden the categories of employment to have something called an independent worker. In Australia, Tony Burke is currently looking at the protections that are provided in the gig economy, and it's a similar challenge here. We see the huge benefits of the gig economy for consumers, but we don't think that should give you a leave pass from looking after the workers that are providing those services.

Miah Hammond-Errey: [00:15:45] Rob, obviously the employment situation in Australia is different in terms of healthcare relationships with employers. But do you have any insights from your research in the US?

Robert Atkinson: [00:15:56] Well, I think Andrew and I are going to agree on a lot of things and maybe disagree on some of the smaller things. So we also at ITIF, we proposed a third category. One of the problems with the first category where like, when when you've done surveys of Uber drivers or when they've seen referendums, they overwhelmingly do not want to be treated as regular, full-time employees. They want the flexibility. I mean, I met an Uber driver, a young black woman who was getting her PhD, and she said, I just can't focus on my PhD 40 hours a week. I need a little break and I need to pick up some money. I had another guy who was, he and his wife, and he was starting up a ministry. This was a big thing in the US. Independent ministry, but he needed to make some money on the side. So I think that flexible model is a really important one and we should respect that a lot of workers want that. One of the problems by putting this category of yes–no is if companies, again, I'm not speaking for Australia, which I don't know anything about, but in the US, if you wanted to start giving that independent contractor a little bit more benefits like I'm going to help you with retirement or I'm going to help you with a training program, you started automatically get locked into the full board and then you've got to do everything. And so the companies are incentivised to not do anything. And I think what you want is, you want this, as Andrew and I know from the old new Labour, new Democrat, you want the third way, the third way of being able to encourage companies and in some cases maybe mandate them to do certain things for the workers, but not take that whole category of full-time labour, which has its own restrictions and categories and things.

Miah Hammond-Errey: [00:17:35] Small- to medium-sized businesses are often described as in Australia, as the lifeblood of our economy, and they represent the vast majority of businesses, upwards of 95%. Rob, can you describe from your experiences in the US, and particularly in relation to the size neutrality concept you outlined earlier, what some of the opportunities and risks are from an innovation perspective?

Robert Atkinson: [00:17:57] Well, first of all to say small business is the lifeblood of Australia, it's kind of yeah, you're going to have businesses. So it's not like small businesses are the lifeblood. You have a bunch of businesses. And then the fact that and I'm not saying this is true, but if it were the case that you have policies that are helping companies to stay small or helping smaller companies at the expense of larger companies, you're distorting what would be an optimal firm size mix. And there's going to be no economy that should have. But by the way, guess which countries have the highest rate of small businesses? They're the countries that have the lowest per capita income. Countries in Africa and all the micro businesses. They have no money because they can't get scale. So I think what you want is you want to have that mix be set by sort of just how industry structures work. Some industries are going to be bigger than others. But on the innovation point, it's important to understand that. And this is good University of Chicago research on this by Chad Syverson, most small companies don't want to become big companies. They're lifestyle companies. You know, guy runs it with his wife or vice versa. They employ a few people and they're happy, which is fine. I'm a lifestyle entrepreneur. ITIF has 30 people, I have no desire to be Brookings. I don't want to grow to be a big think tank because why would I want to do that? I'm enjoying my smallness.

Robert Atkinson: [00:19:22] But from a public policy perspective, what you really want or you want those what are called gazelles. You want companies that start small and end up with a thousand workers or 5000 workers and become global players in Australia. And so I think that's really the key is not so much small but age. So one of the things we propose is we never thought it's going to happen, but we propose it was to get rid of the Small Business Administration in the US and create and replace it with the NBA, not the SBA, the NBA, not the National Basketball Association, but the New Business Administration. And the goal should be, in my view, much more about helping new businesses get off the ground. And so we'd be arguing a five-year exemption from regulations or things like that give you a safe space to get off the ground. But once you're off the ground, why would we exempt you from racial restrictions or things like that? So I really think the key in this small business, large business thing, is much more about how do you get high-growth, small companies that want to continue to grow. You know, it's amazing in Japan, an interesting story, I think it was Korea, sorry, you go over ten years period, they had less than 500 companies grow from small to mid-sized, 500. In California, we have like 500 a minute, you know. So that's really the key, is getting small companies to become at least mid-sized companies.

Miah Hammond-Errey: [00:20:48] And do you have any, any policy levers that you've seen to successfully grow those companies from small to medium size?

Robert Atkinson: [00:20:57] Yeah, I mean, I think there are lots of areas where US policy is suboptimal, but one policy I think that's quite good is to we have a rich ecosystem of venture capital and it's not just venture capital giving money, it's venture capital giving guidance and networking and linkage. We have what's called the SBIR program, Small Business Innovation Research Program. That was actually how, you know, the company, Qualcomm wireless chips, they were funded by SBIR. So we have state governments like you do, they have very active programs to help small companies. We have a really top-notch system of universities, public and private, that do a great job of helping small companies get off the ground. I mean, if you go to MIT, for example, look at MIT. MIT is created if you look at the companies that spun out of MIT and look at their total revenue, it would be the 12th largest country in the world. So you can create ecosystems that make that easier. And I know you're all looking at that here in Australia and you've done some of those things, but I think that's really what we should be thinking about.

Miah Hammond-Errey: [00:22:07] Do you have anything you'd like to add on that?

Andrew Leigh: [00:22:09] Well, I think Rob's point about focusing on new businesses is really important. You certainly don't want policies that encourage Bonzai businesses that encourage firms to forego growth opportunities, which would be good for productivity and good for their workers as well. So the focus on new business is important. A lot of the OECD research is pointing to this as a key priority. And in that sense, I'm interested in both the formal government structures, whether there's more that we can do as a government to encourage start-up innovation and also what the governance and venture capital ecosystem looks like. You had this notion for a while in Silicon Valley that going public was too much of a hassle. And so you saw for a while the growth of unicorns. A lot of start-up funding going to allow firms to stay private and to avoid all that excess regulation, red tape, reporting that can hamper a brilliant founder as they're aiming to start up a company. And then out of that ecosystem you saw emerge WeWork and Uber under Travis Kalanick and a range of companies where the lack of discipline from the public markets really did become a problem. And so now things have swung back a little and there's more of a recognition that there is actually a discipline that comes through an IPO and through getting public money in that sense. So we're very alive to that discussion. And the notion that where you get your money from can shape your growth trajectory, but also ensure that the firms are behaving in the way in which they need to is they graduate to mature organisations.

Robert Atkinson: [00:24:00] Yeah. One other point I think is important to understand, at least in the US ecosystem is because of Sarbanes-Oxley. Was a little bit harder to go public. So one avenue that a lot of entrepreneurs bet on and hope for when they're taking big risks to start a company is acquisition. And my son's 30-year-old kid in Silicon Valley, I'm hoping that's what happens because he has some stock options. Yeah, but and I had to loan him money, ridiculous. But anyway, I think we shouldn't underestimate that because in the US the narrative is somehow that we shouldn't really allow acquisitions, but acquisitions in the US actually can really drive innovation because they provide another exit. And sometimes if you're putting money into this, a lot of time what you're looking for is can I get a good exit, sometimes it's an IPO, sometimes it's just growth and you don't need an exit. But sometimes it is acquisition. And I think all three can play an important role.

Miah Hammond-Errey: [00:24:59] I want to just draw on one thread before quickly looking at inequality and then getting to the tech policy hot topics. One of the policy suggestions from your book, Andrew, is about building innovation training for everyone and kind of drawing out what Rob was saying there about the ecosystem of innovation. Where is Australia at now and can we improve and if so, how?

Andrew Leigh: [00:25:20] Yeah, I think definitely improve. If you look at a range of the innovation precincts, most cities have them and there's a lot of variation in terms of how well they create environments in which innovators are able to learn from one another and get the benefits of co-location. I'm interested in some of the innovative models. So University of Technology Sydney has an aim to provide a start-up experience to half of its students. So the idea here, Miah, isn't just that the engineering students will start up a business, but that the nursing students and the education students and the communication students will also have their own chance to build a start-up. That democratisation of start-up experiences is pretty exciting to me and I think has a lot of potential. Australian venture capital returns over the last couple of decades have underperformed the stock market as a whole. So one of the questions that you then ask is, well, how do we encourage investors into, to partner better, with firms that have more growth potential? Is it something about the way in which investors in a Silicon Valley context are able not just to provide cash, but also to effectively link up their firms with a better ecosystem? Joshua Gans' Creative Destruction Lab at the University of Toronto has spun out of it a whole wide range of successful firms, and he puts that down to the way in which they're able to link up in the same room, successful entrepreneurs and budding entrepreneurs build those partnerships and encourage the meeting of that raw energy and intellectual heft with judgement and wisdom from the more established business people in the room.

Miah Hammond-Errey: [00:27:20] Thank you. You've both talked about innovation, but also inequality. I sense you may have different meanings or focuses for your discussion on inequality, but where do you see the tensions lie and what can we do to reduce innovation inequality?

Andrew Leigh: [00:27:36] Well, to reduce inequality, I think one of the things in the United States that's proven very successful is the Earned Income Tax Credit. Topping up the wages of low-wage workers in who in families with kids, which provides direct cash injection to low-income households while encouraging work. All of the evaluations I've seen of the Earned Income Tax Credit suggests it has a big positive impact on labour supply, but also on the household well-being. And the proposals to extend that through to childless adults make a lot of sense to me. The expansion of the EITC has been on sort of Democrats wish list for many years, but the we haven't seen an expansion of the kind that Bill Clinton was able to pull off in 1996, which had a massive effect in pulling low-income Americans out of poverty.

Miah Hammond-Errey: [00:28:35] Do you want to add anything before we get to hot topics?

Robert Atkinson: [00:28:43] Raising taxes on the rich and basically getting rid of hedge funds managers who make $100 million dollars a year. I'm a little bit, but there was actually a very good study about a few years ago that looked at the top 15 hedge fund managers in the US. They made more than the top 500 CEOs made. So when everybody, and I think CEO compensation is too high, I don't get me wrong, but if you really want to look at inequality in the US, you have to look at the financial sector. They have doubled in size and I would argue probably not very much of a net benefit. I mean, I love venture capital, I love banking, but we've financialised the US economy in very dramatic ways and the returns have been massive. And so just that sector alone, you know, just make them pay a boatload of taxes and definancialise it. I support the Earned Income Tax Credit, but I would actually much rather have a higher minimum wage. One of the advantages of a minimum wage. First of all, we haven't raised our minimum wage in the US in a long, long time. So, because of inflation is going down. One of the advantages of the minimum wage, I wrote a piece for a democracy journal. And it was called Minimum Wage Maximum Growth. And the argument I made is that companies in the US, if you look at other countries that have sort of higher enforced minimum wages, they focus much more on what you can call high-performance work organisations. They use more technology, they train their workers more because they have to figure out a way to address these high costs.

Robert Atkinson: [00:30:21] The US has essentially has this easy way out. You get low wages, it's topped off by the government through the EITC and so you can basically be lazy and not automate, not use technology. And that's one of the reasons why US productivity has not gone up as much and there's pretty good evidence of that. So I'm not saying we should get rid of the EITC, but I think it should be complemented by a robust minimum wage, at least in the US. I think it's $7.50 an hour and it should be $12 an hour minimum and then indexed to inflation. You know, it's funny because when Trump tried to appoint a secretary of Labor who had been a fast food entrepreneur, I can't remember the guy's name, he said, you know, if you raise the minimum wage, you Democrats, we're going to be forced to automate. I'm like, cool, cool. That'd be cool. You want to work in a fast-food restaurant? Because I don't think most people do because they make terrible wages. So that's the other thing maybe we can get into is sort of how much does automation benefit workers? But if you look at if you look at Australian productivity, I'm just reading the Australian Productivity Commission report, it's pretty bad and the US is pretty bad. I mean we have not raised productivity and that is to me, it's equally if you want to raise standards of living for the bottom half, productivity growth is at the same level of importance as sort of more fairness and distribution and opportunity policy. So I don't think we can. To me, that's why I love the minimum wage is it's a double hit. You know, you get opportunity and you get growth out of it.

Andrew Leigh: [00:31:58] Yeah, certainly the I, think it's Dan Hammermesh who makes the point that if you go to Europe, you're much more likely to see a waiter using an electronic ordering system than you are in the US where there's much more, labour is cheaper. So the labour-capital ratio is higher. And Alison Booth has some nice stuff on monopsony power suggesting that if you want to encourage investment in training, then, then higher minimum wage is going to have that effect. So strongly supportive of that one too.

Miah Hammond-Errey: [00:32:28] You've both touched on automation. I wanted to ask you about AI. As you know, last night OpenAI released GPT4 with a number of improvements on the earlier version. For many people in society chat, ChatGPT has become the first touch point they've really had with AI. What are your thoughts on the technology? Not specifically GPT, but broader AI. Where do you think it will go? And to your point, Rob, where do you see this shifting productivity?

Andrew Leigh: [00:32:59] I'll dive in first if you like, and then Rob can give you the more considered answer. It's going to be augmenting a huge number of jobs. I was just looking at Cade Metz's piece in The New York Times, which is worth a read because he identifies how GPT4 improves on 3.5. You can do things like you upload a picture of a fridge and say, 'What do you suggest I cook tonight?' And the AI recognises the items in the fridge and suggests a couple of different ideas for dinner. And its ability to, for example, summarise text in a pithy way or to take a summary and identify ways in which that summary is flawed. It can now do very well in most standardised tests, SAT, bar exams and the like. So it's going to augment a whole lot of jobs we do. And how do we assist people working with AI is going to be a major policy challenge. I, we suggest in Innovation + Equality that better-connecting MOOCs through to the formal training system may be a useful way of advancing, and that training people to continue to learn will be, will mark out the countries which are best able to use AI. This whole idea that you get a block of education at the start of your career and then you just spend down that investment over the course of the next 50 years. There's no way that can work in an environment in which technology is rapidly advancing. We want to be able to link up MOOCs and formal education and encourage workers to continue to acquire skills and so that the workers of the future are saying, Here's how I can use ChatGPT4 to be far more productive. And in a way, Rob highlighted a moment ago with the minimum wage to therefore earn a higher wage.

Robert Atkinson: [00:35:03] Yeah, I think there's, we did a study a few years ago where we were able to get a data set that University of Wisconsin had curated, which was from the US Bureau of Labor Statistics. And we keep track of like it's 500 occupations or something, 600. And so we had occupational data of the number of jobs for every decade from 1870 so 1870, 89, all the way to 2010 and then 2015. And we said what was, what decades had the most absolute value change some either growth or contraction of occupations. It was 1890s and so the biggest 1890s and the next the biggest sort of disruptive technology in the US economic history was the tractor by far just by far. Then the 1950s was another big one. It was basically there were all these people who were women largely doing 'what number can I connect you to?' Those jobs went away. There were also large numbers, mostly women. 'Oh, you would like that file? Let me go to the file cabinet and file it for you. Get it for you.' So this idea that we're in this phase of sort of massive occupational restructuring just hasn't been borne out. If you look at the most, if you, again, going back to 1995 using BLS data, the year before COVID. So you can't use COVID. So 2019 or 18 was the lowest rate of occupational disruption we've had since 1995 when we started collecting that data.

Robert Atkinson: [00:36:37] In other words, you're losing your job through no fault of your own. The lowest rate and the lowest rates by decades were the 90s, 2000s. So I think this is way over-exaggerated. And to be clear, I don't think at least in the US, we have a terrible system of helping good workers make transition. So this is not a thing to say we shouldn't have way better policy. I mean, we want to look at the best country in the world. It's Singapore in my opinion. They have a totally comprehensive, totally technology advanced program to help workers make transitions. We should adopt that program in the US, but I don't think we should be worried all that much about massive unemployment. Andrew mentioned complementarity or augmentation. I think one of the big debates in the US are, there are people in the US who say we should only support technology that augments worker skills but doesn't replace them. I completely disagree with that. If you look at the National Science Foundation Robotics Initiative, it only funds researchers who are looking at augmentation robotics, not replacement robotics. Now, imagine back in the 1930s when Brunswick the bowling ball company, you know, if they said, you know what, we're not going to make these bowling ball centre things that come down and pick up all the pins, we're going to have better technology so the ball boys know how to put the pins together faster. No, we just said, You know what? That's a really terrible job doing being a ball boy on a bowling alley. We're going to automate the thing.

Robert Atkinson: [00:38:09] And I think if we want to raise productivity, my goal in the US is, you know, in our median, ours is higher than yours, but our median income is $60,000 a year. I don't know how anybody lives on $60,000 a year. I don't want to sound obnoxious, but I live in DC. You can't live on $60,000 a year unless you want to live in an apartment. But you have a family. You can't do it. I'd like to see that be $150,000 a year. I'd like that to be the median. And the only way you're going to do that is a lot of automation. And so I think it's dangerous to think that we shouldn't embrace automation. Now, is it going to hurt workers? Yes. And there will be workers hurt. I totally get that. That's why I think we should have income support, training support. But I think we can't sort of say, well, we don't want to hurt any workers, so we can't embrace automation and productivity.

Andrew Leigh: [00:38:58] And this is, of course, something that comes up all the time.

Robert Atkinson: [00:39:00] And I'm not a politician, by the way, so I can say that, yeah.

Andrew Leigh: [00:39:04] This is something that comes up all the time with trade, right? So the argument is always, should we hold back trade if it's going to adversely affect some workers, although it's going to be beneficial to society as a whole. And generally economists answer on trade as it is technology, is let's take that increase in aggregate well economic output and redistribute it in a way in which everybody can be better off. It's harder to do in practice than it sounds in theory, but surely it's got to be the right answer if you want to have a race to the top.

Miah Hammond-Errey: [00:39:38] You've both spoken not so much tonight, actually, but in various other forums on antitrust and domestic competitiveness. Andrew, can you outline some of the major developments from the digital platform inquiry?

Andrew Leigh: [00:39:50] Yeah. So the work that the Australian Competition and Consumer Commission has been doing is looking at some of the ways in which digital platforms have attained dominant positions in the market. So just take a typical morning, you might wake up, pick up your iPhone, use it to check your Facebook feed to see what your friends are up to and then go on Instagram to check out your favourite celebrities. And then just before you head into work for the day, you might have a look at Google weather to work out whether it's going to rain today. Well, in just those four interactions, picking up an Apple phone, using Facebook, using Instagram and using Google Search, you've engaged with technologies which didn't exist 20 years ago when Rob and I co-authored our reports and which now have majority market share. And that majority market share means that those firms are potentially able to use their market power in ways that are problematic for the economy as a whole. You look, for example, at the way in which Amazon deployed $200 million in the space of a month to wipe out Diapers.com, an upstart that was threatening part of its business. Some of the challenges that our app developers face as they're looking at paying 30% of their revenues to Apple in order to do any in-app purchases. And the data challenges that emerged around Facebook and the Cambridge Analytica scandal owed, had their genesis in part in the highly dominant role that Facebook play.

Andrew Leigh: [00:41:39] So we're looking at platform-specific regulation, whether it would be appropriate, as we do in the national security context, to designate certain platforms as being particularly significant and therefore tailoring rules around those, as well as whether it would be appropriate to have a ban on unfair trading practices which went right across the economy to tech and to non-tech firms alike. But we're consulting on those and very happy to engage with the thoughts of those in the room as well as more broadly, the competition and consumer lobby groups.

Miah Hammond-Errey: [00:42:18] I sense Rob you might have some different perspectives on this.

Robert Atkinson: [00:42:21] Yeah, I'm so happy because Andrew and I were agreeing way too much. Yes. So I have to say here, I would respectfully have a different view. In the Obama White House, the National Economic Council came out with a very good report on competition in essentially the platform economy, the Internet economy. And they said something that I'm going to paraphrase, which is, when you have network based-industries, markets tend to concentrate to one or two players and what is a network-based industry. So this is not a network-based industry. It's like a pen and paper. But Facebook is a network-based industry because the more people on Facebook, the more valuable it is to the user. And so, imagine, I wrote this in our book, which was I had to say, imagine that you have now Andrew has succeeded, or you break up Facebook and I know you're not talking about break up, but let's just say hypothetically and now we have Facebook and we have Headbook. And every time you want to post a picture of your kid's birthday party, you have to post it to Headbook and to Facebook. Well, there's a reason why you have Facebook. It's because nobody wants to do that. It's just way too much hassle. Why is there really only one sort of microblogging thing called Twitter? Because I don't want to have to tweet on three different platforms.

Robert Atkinson: [00:43:41] So what you had is specialisation in LinkedIn, which is sort of a professionalised network. You have YouTube, which is more of the video platform. You've got TikTok, which is a different kind. So number one, that's just the way these markets work. And the idea that somehow you would use structural antitrust policy to change that to me is problematic. Secondly, how many of you pay for your Google searches? Anybody, no? These are free. These are free. And so normally what antitrust policy looks at is they look at evidence of monopoly to rent-seeking for consumer harm through higher prices. These companies are providing these amazing things. Erik Brynjolfsson at Stanford has done these studies. How much would you pay to give up Google Maps? And it's like 150 bucks a year. How much would you? So the consumer value of what's called the consumer surplus is very, very high with these things. So I'm like, wow. The other thing about these is I totally agree that there are some industries maybe where there's too much concentration and you have rent-seeking monopolies, you get lazy. The five top American tech companies invest more in R&D every year than the entire UK economy. Five companies. So they're not sitting with their profits and going, Oh yeah, we're going to. What they're doing with their profits is they're investing in the next thing. Why are they investing in the next thing? Because they know they could die at any moment.

Robert Atkinson: [00:45:06] So look at Facebook two years ago, both Facebook's dominant and who thinks Facebook is dominant today? Facebook has had some real problems and some real challenges they're facing. Who thinks Google is dominant? Well, I was talking to Microsoft last week and they rolled out the latest ChatGPT on Bing. So there is, I think, Schumpeterian risk of creative destruction in this space. The last thing I'll say is, oh, I have to say also with Cambridge Analytica, the problem there was Cambridge Analytica. That's the problem. It's not Facebook. And by the way, just you remember on Europe has this thing called the GDPR. Well, Cambridge Analytica was subject to the GDPR when they did this terrible, stupid, awful thing. So here we are saying we need regulation. So you have this really strong, overly strong regulatory regime in Europe and it didn't work.

Robert Atkinson: [00:46:05] And then lastly, I'll just say on the Diapers.com, and there's a piece that my colleague and I wrote on a lot of this. This all came, by the way, from this woman called Lina Khan, who is now head of the Federal Trade Commission, when she was a Yale law student, and she wrote a Yale Law Review article. Everybody thinks, oh, my God. Yale Law Review. Well, it's only by students. Let's remember that. It wasn't like a real law review article that a real antitrust attorney or a lawyer would do. So she was a student when she wrote it and she wrote The Amazon Paradox. The antitrust Amazon paradox. And so we did an analysis of that. And when you go in in detail and look at all of her assertions about what Amazon did, including Diapers.com, it doesn't hold up for scrutiny. The DIapers.com, the amount of market share that Amazon had in diapers was minuscule. It was less than 8%, I believe. So there was no market power when they did that thing with. It wasn't even Diapers.com. It was someone who made the company. So what I would argue we need to do in antitrust when it comes to platforms. And I would add one other thing. I think platforms are really important business model or organisational model, just like the rise of the industrial corporation Alfred Chandler talked about during the beginning of the 20th century. I think what we should be focused on is anti-competitive behaviour. If Amazon or some other company are doing something that's purely anti-competitive. Yeah, let's go after them. But selling their own batteries, having a really great logistics system, I just don't see what the consumer harm is there. And yeah, it's like a.

Miah Hammond-Errey: [00:47:41] Right of reply. Andrew.

Andrew Leigh: [00:47:42] Yeah. So as you said, Rob, we were agreeing way too much. So this is, this is good. I would push back slightly on the Brynjolfsson figures and the suggestion that that captures everything, everything of value. If you ask poker machine addicts how much you would have to pay to stop playing the pokies, they'll give you large numbers. But that doesn't tell you that the social value of poker machines is very high. In the period since smartphones and social media came out, we've seen a substantial worsening of youth mental health. Jonathan Haidt's work, which looks at self-harm, at suicide, at body image from American teens, suggests that the period since 2007, 2009 has coincided with a significant worsening. We've also seen the same trends here in Australia and so the distraction economy has served to make it more difficult for many workers to stay focused on their task because there is this amazingly exciting, addictive machine, sort of the virtual slot machine sitting next to you and that can drag you away from either working or in-person socialising. So we've seen it coincide with a decline in in-person socialising. I don't think it's a coincidence that over this period young teens around the advanced countries around the world are less likely to have in-person conversations, less likely to go on dates, less likely to have a job, less likely to have a driver's license. The so-called backseat generation has coincided with the rise of these technologies.

Andrew Leigh: [00:49:29] What has it done to politics? Well, I think it's fairly clear that Donald Trump wouldn't have become president if Facebook and Twitter didn't exist. And I think it's very clear that he is now one of the front runners for winning the presidential election in 2024. And if these platforms didn't exist, that would be much less likely. Political disinformation is spread largely through social media. So when we're thinking about the very real benefits that these platforms deliver, you know, linking up diverse communities, allowing people who are uncomfortable with face-to-face interactions to have their say, we also want to put on the scales the significant harm to youth mental health and to the political system.

Andrew Leigh: [00:50:14] In terms of acquisitions, I mean, the Instagram acquisition did represent Facebook promising not to merge the back-of-house operations, but ultimately going ahead and using its market power to turn that into a behemoth which crowded out another potential competitor. And the pace of acquisitions is remarkably high. So dialling back the acquisitions I think would be a useful way of ensuring that consumers had a little bit more choice.

Robert Atkinson: [00:50:54] We could spend the rest of the night on this because we could just keep going.

Andrew Leigh: [00:50:57] If Miah gives us permission.

Miah Hammond-Errey: [00:51:00] Do you want a few minutes? And then I wanted to ask a question about innovation and then we might open up to the room.

Robert Atkinson: [00:51:04] A couple of things. One, we have there are different debates on this in the IT tech Internet world. And one debate is about social platforms. So Amazon's not really a social platform. They're a buying platform. So I don't think that's an antitrust issue. If we broke up the major social platforms into five companies each, all this would be way worse, way worse, because you would have companies who would be you know what I love? Hard, right? Nazi neo-Nazi stuff. Come on. The bigger platforms. They just have to sort of they have to sort of serve everybody in a way. So they have to kind of go down the middle of the road. And I'm not saying they're perfect, but I also think this is super hard. I'm actually, I was asked to be on the content advisory commission for Twitter, for TikTok, sorry, a great group of folks who know way more than I do. And I got to tell you, we're in meetings where it's like they think really, really hard, what are we doing with this body image issue? So the question then is, I don't know what you do about those questions other than ban this technology, because my impression is these companies are really, really working on this question. They know there's social, there's there's drug issues, there's body image issues, there's sexism issues, all that. And they are really, really working hard on it.

Robert Atkinson: [00:52:26] Secondly, the WhatsApp thing, I don't know how many people remember before, before Facebook took over or bought up WhatsApp, you had to pay for WhatsApp. It was a monthly subscription to pay for it and it had very few users, so you couldn't really use it with that many people. Now it has a gazillion users and it's totally free. I've been in business trips all over the world and I go on WhatsApp and I have a free video call with my daughter, who's a teenager. Free video call from around the world. And so to me, hey, buy them up, man. I don't. That's beautiful. The other thing is I think that the real reason for the, real reason for all of this stuff is for this political bifurcation. I hope you're not saying, by the way, the Russian stuff, because I that to me there is no evidence that the Russian collusion influenced the election. I think that was a myth. That's not to say there isn't a lot of misinformation and sort of weird stuff on that, but you really want to know where this came from. Go look at MSN and Fox News. I mean, the bifurcation of US politics isn't from social media. Social media is reflecting the bifurcation. The bifurcation happened with Rush Limbaugh and then the counter super left networks.

Robert Atkinson: [00:53:46] So we have the US suffers from this incredible political conflict, but social media didn't cause it in my view. It helped maybe amplify it, but it didn't cause it. So again. And then the last thing I'll just say is what's the solution? Very last thing. And you look at that Facebook study that where 20% of teens, I think it might have been girls, had problems with the image because of that. But it didn't, what the media didn't report is they found that 50% of the teens felt better about it. They felt better about themselves. And one of the reasons they felt better about themselves is we all remember when we were a teenager, we all think we're freaks and nobody knows anything because they could find people on Facebook who had similar attitudes or similar views and they could communicate with them and they could feel better about them. They're not the only one who may be questioning their sexual identity. So Andrew, I'm not saying that we should sit back and everything's, you know, Nirvana. I think these are super hard questions and we should collectively think about how to make them better. But I'm not sort of, to me, the net plus of social media is, so much vastly over weighs the harms. That's my take.

Miah Hammond-Errey: [00:55:04] When does the next report from the agency come out, the digital services, I think the last one was received by government this March. Is that correct?

Andrew Leigh: [00:55:14] Yes. I'm not aware of what the timetable is for the next report.

Miah Hammond-Errey: [00:55:19] I think we might leave that those tensions there and take questions from the audience.

John Kunkel: [00:55:32] I think a big downside is Andrew is a small C conservatism, but we'll leave that to one side on his social media. My question is about the concept. And I'm trying to, I guess, again, maybe engender a bit of disagreement. Is international competitiveness a useful concept? Conscious that 30, almost 30 years ago, Paul Krugman with impeccable progressive credentials said it was a dangerous concept, it was that effect in foreign affairs. So I guess, with the elephant in the room being China. Is international competitiveness a useful concept? Please discuss.

Andrew Leigh: [00:56:09] I think it's a fabulous concept for sport where there is only one gold medal in every event and your competitor nation doing better means your chance of winning gold goes down. But we're talking about economic prosperity in general. The success of other countries benefits you. So when French pharmaceutical entrepreneurs develop a better treatment for an Australian disease, then we benefit from that from that growth. Clearly, if you're talking about military contestation, then you're in a realm that looks a little bit more like sport. You know, typically wars only have one winner. But in the area where, in areas where you can have technology transfer, there's huge benefits to productivity rising in other countries. Australia wants US productivity to go up. We benefited massively from COVID-19 vaccines being developed overseas and there's no sense in which we were losers because their vaccine was proved to be a successful cure for COVID-19, where the Australian vaccine gave false positives for HIV. Australia is a, is a beneficiary in most contexts, so in general I fall on the Krugman side of the debate, although you can identify exceptions to that rule.

Robert Atkinson: [00:57:39] Good. We get to disagree again. I would not call Paul Krugman's credentials impeccable. I would call his ideology impeccable. Look, the reality and part of the reason. Andrew's right, in my opinion, I agree with Andrew on certain things. Certain part of this is you have to think about economies divided into two components. There's the traded sector or the non-traded sector. So we're better off if Australia's non-traded sector gets higher productivity, because we can sell more to you and it raises global income. So for countries like, yeah, you fundamentally want your non-traded sector, your garbage collection your banks, things like that, your government services, you're not competing with the Chinese on that. Then there's the traded sector. And I fundamentally would if you look at sort of if you look at realist foreign policy versus idealist foreign policy, in the realist foreign policy, economic policy is that countries are in a race to kill each other. Okay. So China wants to destroy our industries. That is their goal. They don't want to have sort of, oh, well, you guys invent this. You. They want to take over the civil aviation market, Boeing and Airbus. They've already taken over the solar panel market. They want to take they've taken over a lot of the market in 5G equipment. So we're just we've got to acknowledge we're just straight up in a competition.

Robert Atkinson: [00:59:12] Somebody's going to win, somebody's going to lose. There's only a limited amount of. This report we did recently called the Hamilton Index of Advanced Industry Competitors. There's only a certain amount of trillions of dollars of probably $8 trillion of global output of advanced industries. Countries are in competition for whether you want that or not, because it's not going up to 20 trillion. It's like it'll grow gradually, but there's $8 trillion of stuff to be competed over. Electric vehicle production. And the question I think every country has to ask is, do you want it? And if you don't, other countries are more than happy to take it. So I completely think countries are in competition with each other. And the most important part of that is we should not allow countries to cheat in that competition. So I used to play basketball and I, you know, we had a team and we were super competitive, but we never played against a team that had brass knuckles that was paying off the ref. And that's what China is. And so I think fair competition between countries spurs us to do better. It makes us all work harder. Unfair competition with China is like brass knuckles. And so absolutely, countries are in competition for their traded sector, particularly their higher value-added trade sectors.

Jenny Gordon (ANU): [01:00:30] It's Jenny Gordon, ANU. Really interesting. I think the only one who wins from wars are the defence material industries. So that's the winner. Now, I've got a question. I was listening to Future Tense today, and they were talking about R&D, particularly digital R&D, but also just R&D generally and where the productivity, you know, Moore's Law is proving to kind of still hold. But they're saying that the investment that's going in has risen dramatically 16-fold to get sort of what we had in the past. And so do you see that we're on diminishing returns in these sort of digital technology areas because we're still not seeing productivity gains that we thought we'd see from it. So it's Robert Gordon versus some of the others who are sort of the tech enthusiasts versus Gordon, who said we've done it all. We don't have big gains lying on the table. So I'd love to get your views.

Andrew Leigh: [01:01:27] So GPT doesn't stand for general-purpose technology. But it could, because this is a classic example of a technology which has great potential once it is integrated into the economy. And as you know very well, Jenny, if you look at the steam engine, the productivity gains don't flow for the better part of half a century. If you look at electrification, it takes a while before factories are rejigged to take advantage of electrification. If you look at computerisation, the initial uses of computers look farcical in retrospect until offices are remodelled to take advantage of computers and artificial intelligence could well be something like that. In the short term, a kind of curio and wonder how we integrate it into the economy in the long run outperforming expectations. So I'm optimistic that that will change and shift and provide more opportunity. And I think when you envisage general purpose technologies, it should also make you sceptical of anyone who tells you that there is some fixed budget of technological innovation to be done in the world. Technological innovation, I think, opens up the possibility that actually the, the world's industrial base can be larger and can be very different over decades to come.

Robert Atkinson: [01:02:53] So I think there is pretty clear evidence that it is harder to extract from nature insights and actionable things than it was in the past. There was a lot more low-hanging fruit. I think one of the studies you're referring to is Nick Bloom and his colleagues who found 70 engineers today to get one advancement in Moore's Law than you needed one back then. I think that's true. And I think that is one of the reasons why innovation is harder than it was. It's like it's really hard to do. You look at drug development, actually, drug development was really easy. It's not easy, but going on. And then it went into this dip. And now what's interesting in drug development is you have this confluence of big data, AI analytics for drug development, you have nanotechnology and you have genomics together. And those three seem to be making us come up on this. So I think it's a combination of kind of these new technologies that are emerging. Are they going to overcome just the difficulty of extracting insights out of nature? So I'm generally an optimist. I mean, I've been an optimist my whole life. I think when everybody says, oh, my kids are not going to live a better life than I am, like, no way. That would imply negative productivity growth, which the world has never seen. Negative productivity growth over a longer period than one or two years. And so I'm kind of with Andrew.

Robert Atkinson: [01:04:17] I think one of the things is if you look at a lot of the technologies today, there's a famous line by Bob Solow, Robert Solow, and we see computers everywhere. But in the productivity statistics, we said this in 1989, you might not be old enough to remember what your computer was in 1989. It was a piece of trash. It was a piece of trash. It wasn't connected to the Internet. I had when I wrote my dissertation. I had two floppy drives because I couldn't afford a hard drive. It was like too much money for me, you know, computers got and the Internet. And so we did have this big productivity advance from like 95 to 2006, 2007. So I think these technologies like robotics and AI and vision, they're getting better and better and better. But people think they're so enamoured like, 'Oh wow', I can go on TikTok and you can see people have their facial image changed, you know, like, 'Oh, wow'. And we all get enamoured. But it's still it's a ways to go. I think before this tech, you know, we've got another five or six years to go maybe I don't know how many, but at some point the technology will be so good that it will be it will have GPT characteristics and it doesn't quite yet. I think so. But I would encourage us to just keep working at it and keep pushing it, and I think it will get there.

Michael Davies: [01:05:38] Michael Davies from. Firstly, Rob, thanks very much for your kind words about multinationals, but I have a question more about the debate that's started up in the United States about sort of fundamentally decoupling the United States economy from China particularly because you've written a bit about some of the dangers of innovation and mercantilism. But I was just wondering if for those of you know, in Australia where our debate about engagement with China sort of fundamentally different, what you can say about, you know, the risks of, well, both how do we balance the risks of having China as part of a technology stack, particularly in terms of IP theft and cybersecurity, versus the downside risks of not being able to access technologies or not choosing not to access technologies that are competitively priced?

Robert Atkinson: [01:06:40] Yeah. So I think that one of the things that China wants. So the old model where you were going to be a military hegemonic and I don't know enough about military, you know, they certainly are building up their defence capability stupid quick. But I think what China wants is they want to be the global economic and technology hedge. They want to be able they want the yuan and the RMB to be the dominant currency. They want all their technologies to be the dominant technologies and I think that would be bad for the world. I just do. I don't think we want China calling the shots. And so part of the whole notion of decoupling is how do you do things in a way that limit China's rise and slow it down and at the same time help speed us up so. There are people in the US who argue that we should have you know, we should tell Kentucky Fried Chicken, no more restaurants in China. It's like, what does it have to do with anything? So I think the argument should be we need to be thinking about decoupling from a US perspective with China on things that where we benefit and don't get hurt. So there are people who argue in the US we should not sell them sort of commodity computer chips because we want to decouple. Absolutely. We should sell them as much as possible because every dollar that we get from China is a dollar that our firms get to reinvest in R&D and production and a dollar less that their firms get.

Robert Atkinson: [01:08:00] So I think we don't want to decouple. I think what we probably what we should do is we should be thinking about a sort of gradual, slightly faster than gradual second sourcing program, which is why I think we should really be focused on India. India would be, we need India to be the new China for manufacturing. I think that would be better for everybody. And then we need to be thinking carefully about what kinds of limits we have on tech transfer in China. One of the problems is that China puts a gun to companies heads. If you want Chinese access to market, you've got to give us your technology. And the reason they're able to do that is because of their monopsony power. You should bring your antitrust case against the Chinese, their monopsonist, and they can do that. And so I think that's where the West or the allies have to come together and confront that collectively and say, we are not going to allow that to happen anymore. So it's a very difficult conversation in the US because we're sort of early on and with lots of emotional voices going back and forth at each other. But there's a lot of people in the US who I think are thinking pretty carefully and strategically about it. So I'm not, there'll be a few things here and there that bounce, like banning TikTok. It's like, that's the best you got. I mean, really. Do you think TikTok is a national security? I would be much more worried about civil-military fusion than I would be about TikTok.

Andrew Leigh: [01:09:26] Okay. Just a quick point on China. I think it is always important that we distinguish the Chinese people from the Chinese leadership, from the Chinese Australians, because there is a sense in which in these conversations using China shorthand can take away from the billion people, the benefit to a billion people of rapid economic growth and the ability to move hundreds of millions of people out of poverty and the desire among many people in, many Chinese citizens to have more personal freedoms than they do at the moment. I also would like to see a United States, which is a little more self-confident about its standing in the world, you know. Vis a vis China, the United States massively dominates in soft power. There is no Chinese equivalent of Hollywood. It dominates in the military sphere. It dominates in terms of friends and allies around the world. And if US China policy could embody a little bit more of that self-confidence rather than a sort of timidity that the world is about to fall apart, I think we'd all be better for it.

Miah Hammond-Errey: [01:10:48] I mean, there seems to be a growing tension there.

Robert Atkinson: [01:10:52] First of all, I think I've you know, I've advised and I think it's the wrong word. I've had lots of Chinese government delegations, CCP-declared delegations come through and they want to talk. And I tell them the same thing every time, which is if you really want to grow your economy, the way to do it is not what you're doing. So they were spending trillions of dollars to advance these high-value-added sectors. And when you look at I won't go through the boring details and look at how much productivity they've gained from it. It's minuscule. It would be much better off through some of the stuff where you increase expenditures by the middle class, you get higher productivity in your agricultural sector and your logistics sector, all this stuff sort of going back to Krugman and thinking about their non-traded sector. So what I worry about with China is I have no desire to limit the Chinese people's growth. I think what they're doing now is, is intentionally limiting the Chinese people's growth. Xi Jinping doesn't fundamentally care about the Chinese people's well-being. He cares about their power, his power and the CCP's power. So I think we should be telling the Chinese economy, the Chinese leadership, that you're not really trying you're not helping your population get a better standard of living. They would have a much higher standard of living now if they hadn't done all this stuff.

Robert Atkinson: [01:12:09] I was talking to a high-level Chinese official in DC one time and he was telling me about 'Rob, we have we have these tragedies in China'. And I said, Well, what do you mean? I'm thinking earthquakes. No, no, no. Right before the DVD player came out, we invested billions of dollars in VCR factories and we wrote most of it off. And that was a tragedy. I call that sort of yeah, a tragedy. Sort of stupid. I don't call it what it is. It was kind of dumb. So you take all that money that they could have put into health care, they could have put into education, and they just wasted it. And they waste so much money. I mean, you look at all these semiconductor execs that were going to jail because they just blew through $20, $30, $40, $50 billion dollars. So I don't think the Chinese leadership is really trying to help the Chinese people. I think they're trying to gain power technologically. So we should all be saying that because I agree Andrew that it can come across sometimes as looking that we're against the Chinese people and the advancement of the Chinese people.

Robert Atkinson: [01:13:17] Ultimately, there was a wonderful foreign affairs or foreign policy study a few years ago that said when you, and we shouldn't overstate it, but it was like when you get to some level of like $13,000, you know, there's $13,000 or $14,000 in per capita income, you sort of switch over to being a democracy. That's what Japan did the after World War two. Korea did in the 60s, Taiwan. To one level, China's going to get rich enough where they're going to go, we don't want this anymore. And so I think the Communist Party knows that. The question is, can you avoid all these problematic things in that period when they're rich and powerful but aren't rich enough to have democracy. So but I agree. The whole look, my daughter is adopted from China, so I'm sensitive to this issue. But I also think it's important to make it clear that we're not, this is not an anti-Asian policy. We're not against the Taiwanese. We want to have partnerships with the Taiwanese, with the Koreans, with the Indians, with the Japanese. It really is about a particular country that's doing things that are against the interests of the West.

Helen Mitchell: [01:14:25] Hello. Helen Mitchell from Australian National University. So I had a question. I was interested in your views on what kind or what type of innovation you think that the US might want to incentivise or not. So I'm not just talking here necessarily about subsidies, which are very topical at the moment. I mean, for example, you know, migration policy settings can influence some of the innovations that are incentivised. I'm thinking here about Rob's points about labour-saving automation, for example, becomes more of a commercial imperative when migration settings, you know, are sort of maintained rather than opened up, right? Disincentivising that type of innovation at the expense of other times or, for example, maybe more on the subsidy side., you know Chris Miller's insight that, you just read his book, Andrew, that, you know, the US has tended to do well on sort of the big science innovations, but potentially at the expense of some of the manufacturing and all the engineering type innovations. So I'm just interested in if you do have any views on what the US might want to be focusing on in terms of types of innovations, this is based on the assumptions that they're not just kind of, it's not just a way that inexorably flows without it being shaped by government intervention.

Robert Atkinson: [01:15:49] So I hope I quite got your question. I think the US tends to be good at science-based innovation and we tend to not be as good at engineering-based innovation. So not to have these broad generalisations of national innovation systems, but the Germans and the Japanese are much better at engineering-based innovation than we are, and the Chinese are better at engineering based, they're good at engineering based. So I think one of the big challenges for the US is how do we beef up engineering innovation? And that was a lot of what the Chips and Science Act was about and some of these other things is how can we, I mean, there's this notion in the US, you know, we, we invent these things and then they go offshore to be built. And then finally in the US, the elite class of policymakers have understood that that really is not sustainable anymore, that we have to, not that we have to build everything in America. We shouldn't want to have that. That's not the right goal. So I do think that's one. I think that the issue about immigration and low-wage labour and automation, I think the Democratic Party, they just want more low-wage immigration. They don't care if they legal or illegal. It's not a big deal. They think they're eventually going to be Democratic voters. The business class in the Republican Party wants more, particularly the small businesses and the dependents. They want more illegal. They want more immigration because it's cheap workers for them. It's really only sort of now this Trumpian wing of the Republican Party, which is much bigger on immigration, that's now dominant in the Republican Party, that wants to limit immigration. But if you sort of tied it to automation, I think there just isn't, there's, there's a really good economist in used to be an economist with the Bank of Toronto I think and any he uses this word that productivity is like Voldemort. It's the word that cannot be named. And so I don't know if that's true in Australian politics, but you never hear the word productivity in American politics, you just never hear it. And it's always about innovation and competitiveness, but never productivity. So it's hard to imagine our politics where you could have a presidential candidate say, you know, our productivity is terrible. We need to figure out how to automate it. Just, people don't want to hear it. They don't want to talk about it. So we're just, I think, a big mistake, obviously.

Andrew Leigh: [01:18:06] Yeah, Helen, I think it's a great question and certainly goes to the extent to which governments are able to identify sectors of the future. The track record of picking winners is not great for most countries around the world. And so my general philosophy as an economist is you are better to focus on general skills, broad-based education, which allows people to continue to learn on the job, opportunities for entrepreneurs, right, right across the spectrum, ecosystems that encourage start-ups, but without identifying the particular sort of narrow form of start-ups. And then you want to allow your manufacturing to be in areas where you've got a strong comparative advantage. You know, Rob spoke before about Chinese production of solar panels. That has been the rapid drop in the price of solar panels has been an important part of carbon abatement across advanced countries around the world. So there is, if we were to cut off all imports of Chinese-made electric vehicles, cut off imports of Chinese-made solar panels, we would get less carbon abatement in Australia than if we take advantage of those lower, lower priced manufacturing goods. Clearly countries want to look at niches, but in the context of Australia and the US, I think the US benefits when the tradable sector in Australia becomes more productive. So if our wool sector was to become twice as productive tomorrow, that would be a direct benefit to American consumers right there as well as consumers around the world. At the same time, it would potentially take away jobs in other countries, wool growing sectors such as such as New Zealand. But the world as a whole would benefit from a productivity improvement in a tradable sector of the Australian economy.

Robert Atkinson: [01:20:07] And just very quickly, we would love Australia to improve their wool productivity because we would be happy to have Australia make wool and you buy all our advanced stuff and we buy your wool. So I do think it's important if you want to go up the value chain, it becomes competition. I also think for a country like Australia, I agree with Andrew, you don't know what you're going to do, but for the US, we are so big. We know there are technologies we cannot lose. We cannot like if you're Denmark, you're not going to make everything. So you're like, Yeah, okay, we happen to be good at things. The US has to be good at a whole suite of things we have. We cannot lose aerospace capability. We cannot lose computing and chip capability. We cannot lose biotech capability, we can't lose quantum. And so in the Chips and Science Act, there are ten, there are ten technologies that are identified as key areas that we want to. Batteries are another one and materials. So I think where I would maybe build on what Andrew said is, government shouldn't pick individual narrow technologies. Like there's a big debate now about is it going to be lithium or there's a new technology called lithium air. Who the heck knows? Government can't know that. But governments can know is the battery technology. Battery technology is a critical technology for the US to be able to have some competitors in and it shouldn't pick a winner and firm. It shouldn't pick a winning technology, but it can pick these broad technologies like AI and bio that we should be good at.

Robert Atkinson: [01:21:37] I just, I can't resist on the solar thing because we did a big study for the Smith Richardson Foundation on the effect of Chinese innovation and mercantilist policies on global innovation. Anybody want to guess what year all the growth of solar panels? What year global solar patenting was? Peaked? It peaked at 2009, the year that Chinese production peaked. So what China did is they put out of business way more innovative solar panel companies, the Germans, the US and other countries. They destroyed them. The China solar panel companies are not innovative. They're innovative in production, in process capability. They know how to produce the same solar panel really cheaply. But that same solar panel is not good enough to decarbonise the world. We need new innovations like solar flexible things that could go on your office windows where you can cover an entire office with solar, with electrical generating thing. That is a cool innovation. So my point is, I don't think we should, I think we need to be careful with China that there are certain things China does that look good, but they're actually innovation destroying. So I just to me, maybe China could develop some cool innovation. But if they do it fairly, then I'm more optimistic than it would be good for the world than if they do it unfairly. Like if the market had been working, they wouldn't have destroyed really innovative high-patent solar companies. Because it would have been a mix, and that's not the case anymore.

Miah Hammond-Errey: [01:23:09] Thank you. We're running short on time.

Miguel: [01:23:16] Hello. Good evening, Miguel. Subset of TG Public Affairs. Uh, apologies if this question isn't as high calibre as the previous ones, but, um, I wanted to shift from technology innovation towards the workforce. Um, I just want to get your thoughts on sort of what policies can illustrated, sort of build the workforce. That tectonic shifts, though obviously you mentioned the big innovation technology. Um, and, you know, sort of we have sort of a skill shortage at the moment. And so the, the big issues that lack focus and are sort of screaming for, um, how do we have the right policy settings to attract people to the workforce, especially cybersecurity and Pillar II AUKUS in particular?

Miah Hammond-Errey: [01:23:55] Excellent. Thank you for bringing up AUKUS. I wouldn't have been happy if we left without talking about it.

Andrew Leigh: [01:24:01] Well, thanks, Miguel. Certainly one of the big aspects of AUKUS is the skills training element. And so Brendan O'Connor, the Skills Minister, was in Adelaide today, along with the Defence Minister, Foreign Minister and the Premier of South Australia, talking about the skills component of AUKUS, which gives you a sense as to how critical that will be. In terms of what we do. More broadly, investment in general skills pays off in the over a career better than investment in narrow skills. David Deming does the best work on this, showing that if you have invest, if you get narrowcast education, say in a very narrow apprenticeship field, then you're very likely to find work early on and you earn more in your 20s. But by the time we look to people in their 50s, that narrow set of skills has often become obsolete and people's earnings are tending to flatline. So if you want earnings, you want to maximise earnings over a career, then general skills and learning how to learn really matters. David takes it further and says something like a narrow set of scientific skills can pay off at young ages, but the payoffs to humanities and social studies, which are often a framework of thinking that allow people to learn, can be more beneficial over the course of a career. So we know we're going to need more education as the technologies advanced. We should keep that technology, the education as broad as possible. And since this is our last question, Miah, thank you for a wonderful set of questions and a really stimulating discussion. It takes me back to the halcyon days of summer 2001. I really enjoyed the conversation today.

Robert Atkinson: [01:25:55] No, I agree. Thank you. It's really enjoyable. I enjoyed your questions. I enjoyed being with you and of course enjoyed being with Andrew. So it doesn't seem like 22 years ago. It really doesn't. Weird. So thank you. Thank you, Miah.

Miah Hammond-Errey: [01:26:09] Thank you for coming.

Michael Green: [01:26:11] Let me just close by thanking all of you. It was a terrific discussion, really smart, really respectful, really nice dynamic, a nice dialectic, really made us all smarter. The first, I promise to you of many coming from Miah and the emerging technology program at the US Studies Centre. The overlaps were greater than the differences in your views. In some ways, the differences in your views may be smaller than the differences within the American debate, as you know. I just note a few of them. I think, Andrew, you're right about the American debate on China right now. The tone is a little too hyperbolic. We should sort of remember Theodore Roosevelt, speak softly and carry a big stick. But Rob is right, too. Chinese want a bigger stick. It's not just about joining technology. It's about the fusion of military in technology. I was fascinated by the discussion of scale. I'm glad you started with the two books, which approached that question from a very different perspective, which really it's culture in some ways, the Australian and American perspectives. This has been true for a long time in US–Australia relations. In the early 20th century. Herbert Hoover, a future American president, was sent to Western Australia with his Stanford training to use American economies of scale. He was driven out of WA by the Welsh miners.

Michael Green: [01:27:27] So if we don't always agree on these between the US and Australia, we do have to be careful about the divergences, small though they are. I am told and I think this is accurate, that half of new tech start-ups in Australia are started by people who got their career trajectory in American big time. And so it would be a loss for Australia if these divergences on things like data localisation disincentivise US firms to invest in innovating this economy. So be careful though. I think the US has to be careful as well because we don't have market dominance in determining the rules and there are big differences with us in Europe, much bigger than with Australia. And there are chasms with China where data localisation is a real problem. So getting these right will be important for both of us. But I think today's discussion is an example of how we kind of get close together, right? Not just for the American or Australian economies and societies, but for this whole region of the world, because if we're not on the same page, we really won't have a prayer. So this was excellent. Thank you all for coming. More to come from Miah. Please join me in thanking an excellent panel.