December 9, 2013
Transcript: 2013 Strategic Foresight Forum - The Challenges and Opportunities of the Third Industrial Revolution
|Welcome and Moderator:
U.S. Economics Editor,
Chief Economist and Executive Director of Global Market Insight,
Principal Research Scientist,
MIT Center for Digital Business
Daniel M. Price,
Rock Creek Global Advisors LLC
Location: Atlantic Council, Washington, D.C.
Federal News Service
We are encouraging a lot of questions from Twitter as well, so please, those of you in the room who tweet, please continue to do so. We have fantastic panels today. And so with that, I think instead of taking any more time away from this panel, we want to begin the very first panel discussion. So thanks very much again for coming.
ROBIN HARDING: OK. Well, thank you very much. Let me add my thanks to you all for making it through the snow. I'm Robin Harding of the Financial Times. I think it's quite nice to have snow just as we're having a discussion about technology, because it's a good reminder that we haven't solved all the problems yet. And a few wet flakes falling out of the sky are still enough to bring the city of Washington entirely to a standstill, apart from the Atlantic Council.
But I'm delighted to be joined by a fantastic panel, Marco Annunziata from GE, Andrew McAfee from MIT, and Daniel Price from Rock Creek Global Advisors to discuss some of these issues around technology and the future of the economy and the future of work. And it seems to me at the moment that we're a bit confused about whether technology is going to bring some hellish dystopian future where 90 percent of the population is unemployed because robots are (falling ?) around the streets clearing the snow and there's no work for anybody, or whether it opens the doors to a new wave of productivity, growth that will raise living standards for people in the U.S. and all around the world. And I hope our panel can provide some answers on these questions.
We're going to go to Andrew first, and then Marco and then Dan.
(Off-side conversation.) (Laughter.)
ANDREW MCAFEE: So thank you for that introduction and thank you all for joining us. As someone from Boston, though, I do want to point out that this is not snow.
And I'd like to start by making a very strong statement about what the role of a society's economic engine is, and it really only has one role, one thing it should be doing well, which is – two things actually, which is, first of all, to be turning out – the word I use is "bounty," in other words, a wide variety of goods and services at decent prices and high levels of quality.
The second role of an economic engine is to be producing more bounty over time, while consuming fewer resources to do so. So people say dismissive things about efficiency or productivity. That's absolutely the wrong way to look at it. What we want is more and more bounty out of time from our economic engine. The way to get that is technological progress, which tech progress has been called the only free lunch that economists believe in.
And as a side note, the role of an economic engine is not to provide full employment. That would be a very easy goal to accomplish. We'd just have everybody working the farms with hand implements. But you can see how short-sighted that would be. It would doom people to a static standard of living. It would also stifle innovation because then there would be no incentive to invent better farm equipment, for example.
So the role of the economic engine, in my mind, is quite clear in a society. The good news is that in the industrialized West, our engines have been doing a pretty good job at their main task, and the happy news that I can bring is that it's about to get a lot better, because we're living in a period where day after day science fiction technologies are becoming everyday business reality. So just in the past week we had Amazon talking about drone delivery of goods to our houses. The following day Google announced that – its stealthy acquisitions in the robotics base and the things they were going to do with it. I had a chance to ride in a Google autonomous car last summer. I lived to tell about it. I watched a team of our best students from MIT and a team of students from Harvard Business School get stomped in Jeopardy by Watson, and IBM-developed supercomputer. I could go on and on.
The point, though, is that these are not the crowning achievements of the digital age. These are the warm-up acts. And what we're heading into is a period that my colleague and co-author Eric Brynjolfsson and I are calling the second machine age, which will be an acceleration of technological achievement that I honestly believe we have not seen since the very first Industrial Revolution a bit over 200 years ago. This is phenomenally good news. I think it's the best economic news on the planet.
But Robin, to your other main point, there are going to be some consequences, some negative consequences that come along with the second machine age. I believe the biggest challenge that we're going to face is the fact that as technology races ahead, it can leave some people behind in the sense of them having the desire to offer their labor to the workforce, to the economic engine and the economic engine having less and less use for it, because as digital labor can do more, we need human labor. We might need human labor to do less.
So that's the specter that's confronting us. The data are fairly clear about anemic job growth and actually wage declines and inequality increases not just in America but around the world. My suspicion is that if we don't start intervening and thinking about it, those trends are likely to get worse instead of better. And the most active debates that I have with my smartest economist colleagues are around whether this time is actually different, because like most of us know, for 200 years people have been predicting the onset of mass technological unemployment. For 200 years they have been wrong, mainly because innovators and entrepreneurs keep finding new things for people to do in order to grow their new companies and new industries. The question is, is this time finally different? I have a sneaking suspicion that this time might actually be different. I could well be wrong about that. I would join a long litany of people who have been wrong about that.
But I think my deeper point is that even if this time is different, that doesn't mean it's bad news. It's still phenomenally good news, and our goal should be trying to mitigate the negative consequences of the second machine age instead of trying to decelerate it or shape it or just engage in handwringing about its consequences. Technology is not destiny. We shape our destiny. So even if we are heading into some very tough times for portions of the labor force, there are interventions, there are things we can do. That should be the subject of our discussion, as opposed to, is technology some flavor of new evil in the world? That's exactly the wrong way to look at it.
MR. HARDING: Fantastic.
MARCO ANNUNZIATA: First of all, on the question of are we seeing a new Industrial Revolution, a new economic revolution, for me the answer is unambiguously yes. And the way I see it is, what you're seeing happening now actually is a third wave of what you've seen happening first with the Industrial Revolution, then with the Internet Revolution in the '90s, and now you're bringing the two together.
So there's a lot of skepticism and pessimism on technology. Lots of people have been going around saying that the new wave of technology is just silly games, social media and nothing else. I have a fundamentally different perspective, because what's happening now is the Internet Revolution that have first dominated the consumer sector is now getting into industry, which is the engine of economic growth. And you can see it transforming industry in two ways. One is what we have called the industrial Internet, or you can call it the Internet of things. It's basically bringing intelligence, interconnected intelligence into physical essence. You do it with sensors. You do it by connecting the machines to the Internet. You create an enormous amount of new possibilities just by bringing the power of data into the physical assets.
The second phenomenon which is happening is this confluence of advancement of factories, so 3D printing, new manufacturing techniques, new materials, advanced materials, robotics, a whole set of phenomenon which are transforming the way that industry is operating. This is bringing about new products with different features. It's also changing the way that industries are structuring their processes, their production techniques, the way they're organized, the speed at which they operate. These are all phenomena which have an extremely powerful economic impact. And I'm an economist. I work in a big industrial company that is strong with innovation. So I see it on the ground, and it is impressive. So I feel very strongly that what we are seeing is igniting a new economic revolution. And as you mentioned then, it is something that is just starting now. So this is early days. This will unfold over the next several decades.
Something that people tend to forget when they compare what's happening now to the Industrial Revolution is that the Industrial Revolution unfolded over a period of 150 years from the beginning to when the full effect played out. So we can't call the outcome of the game quite yet.
Then there are implications in terms of disruption and – (audio break) – for the workforce, and these are extremely important and have to be addressed with training at the level of companies, at the level of government. So the disruption onto the workforce in the short term has to be addressed. It's something that needs to be taken seriously.
But we should also not fall prey to the idea that we are building a new world where there will only be room for extremely smart people, extremely wealthy people. I often hear the argument that we're opening up a world where there is room only for engineers and data scientists. But what's happening is all these applications of new technologies are aimed at empowering and making work easier and more rewarding for people at all levels of the skill distribution. So if you think of how easy it is for a child to pick up an iPad and figure out how to play with it, we are now bringing out new generations of user devices for the industrial workforce which are equally intuitive. So part of what happens is the revolution gives workers at all skill levels greater access to information, collaboration, makes life easier for everybody.
Now, the disruption will be there, but I would also make two more points. One is, I wanted to quickly disagree with you and the – on the creation of new jobs. To me the reason why I actually don't think this time it will be different is because what creates new jobs is not innovators. It's not innovators trying to figure out new companies that can employ people. It's demand. It's the fact that innovation, by creating productivity, creates new wealth, and there is no end to the desires we have for new products and new services, and then the ingenuity kicks in at all levels for the skills distribution to give something that people want to buy. So this, I think, is where the job creation also comes from.
Finally, my last point, I think it's extremely important to look at this from a global perspective and not from a national perspective. Partly because of the global economic situation, high levels of unemployment in some countries, rising inequality in some countries, there's a strong emphasis on solving the problems at a national level.
And we're losing sight of two things. One is that innovation is helping to drive economic growth in the emerging markets, reducing income inequality if you look at it at the level of the world population. The second is that innovation moves faster if it is global Within our company, for example, we have scientific research centers across countries. Over 90 percent of our innovations come out of collaborations across centers. Almost no innovation is done nationally. If we surrender to erection which is nationalistic and protectionist, we're going to stop all this. We're going to slow down innovation and slow growth. So we really need to take a global perspective.
MR. HARDING: Great.
DANIEL PRICE: Thank you, Robin. I want to pick up where Marco left off, which is the identification of one of the risks to this new era of technological development and innovation. And I start from the proposition that what enables the developments that Andy and Marco were talking about is the ability to collect, process, analyze and distribute data on a massive scale.
So whether you call it big data or data analytics, this is critical to developing and refining all kinds of products. If you're an engine manufacturer, an automobile manufacturer, you need to collect data on brake performance in a variety of driving environments and a variety of driver behavior. In order to do that, you need to be able to collect and process warranty and maintenance records globally. If you're an airline aircraft engine manufacturer and you want to improve the performance of your turbines, you need to constantly be collecting data on their performance as they move around the globe. If you're an insurance company, in order to refine your pricing of policies for whether it's health or property or casualty, you need to collect and process that data. If you want to improve water distribution, energy efficiency, if you want to deal with base load shifting and intermittent sources of energy, you need data analytics, and you need to be able to move this data to places where it can be usefully deployed. And I think that's a given, and I don't think anyone would argue with that.
Now, what are we facing? We are facing the prospect of data localization and data nationalism resulting from, I would say, a confluence of three probably unrelated developments, but they have come together. Let's start with the obvious one: the Snowden revelations, and the realization that databases' communications are vulnerable to penetration by governments – and not just one government, although one government has attracted most of the attention. For those of you who are from other countries and work for governments in other countries, you know the United States is not alone. So that's one piece.
The other piece is the legitimate concern about personal privacy and whether there has been sufficient transparency in how personal data was going to be used.
The third strand is, for want of a better word, born of a kind of industrial policy mentality that says we may have missed a step in innovation in our country, and we don't have an X, Y or Z company. So what can we do to capture the value of the data we generate and produce local champions?
So we see initiatives around the globe that contemplate national-only clouds, or server location requirements, or restrictions on the cross-border flow of data that respond to one or another of these pressures. And what I think we need to do in order to keep the positive trends going, identified by Andy and Marco, is to separate out these issues, deal with them in their respective boxes, understanding that everything is connected, and make sure that in respect of the cross-border flow of commercial data on which innovation and a digital economy relies, is not siloed, is not walled off, because this kind of protectionism will, in fact, lead to intellectual isolationism, and thus thwarting innovation.
MR. HARDING: Fantastic. Well, thank you very much. Let me start by challenging this techno optimism a little bit. The economist Robert Gordon has a thought experiment, where he says – thumbs down already? (Laughter.) So what he says is.
MR. : (Off mic.) (Laughter)
MR. HARDING: I look forward to the answer. What he says is, you can swap all the things that have been invented in the last 10 or 20 years, but in exchange you have to give up, you know, a past innovation, like the indoor toilet, or the washing machine. And particularly on a day like this with the weather like this, I'm glad for the indoor toilets, and I think I might choose it over Facebook and Twitter. But are these innovations really matching up to those of the past in terms of the productivity they generate and the way they improve our lives?
MR. MCAFFE: Can I jump in on that? Because I've debated Bob about this. Bob Gordon is a phenomenally smart, first-rate economist who I disagree violently with on these things. And there's a flip answer and a deeper answer to Bob's objections. The flip one is, if you ask the 2 billion people around the world who have received mobile telephones in the past five or 10 years for the first time, to give those up and get a flush toilet instead, they would laugh you out of the room. So his thought experiment is really polluted by the endowment effect, what we already have. There are scads of people who find that a really uninteresting experiment in the opposite way that he intends.
His deeper point, though, is that we have seen huge waves of technological progress in the past – absolutely true. His point is the one that we are seeing right now doesn't measure up to the big ones of the past.
That's where I have a problem. I'm not saying the past ones weren't a big deal. They were huge developments for humanity. As we've heard, I think, consistently across this panel, we ain't seen nothing yet when it comes to digital progress, despite the fact that we've had computers around for a few decades.
So I have a very different view of the near-term and especially of the medium-term future than Bob does. As these innovations that we've been talking about get out there and diffuse through the economy, which they're going to do remarkably quickly, I think we're going to see stuff that will make mockery out of a lot of the innovations of the past, even though I like my toilet very much.
MR. ANNUNZIATA: So just quickly, it's exactly right. The question, the way Robert Gordon phrases in the thought experiment, is misleading, because if you had asked people some time ago, OK, would you give up fire to have the wheel, they know – they kind of liked fire. So we take some things for granted and then we move on and want more. So that's not the issue.
But I think more importantly, you know, the point that Gordon makes is that if you look at things like steam engine, electricity, they have transformed the way we live, the way we move around, and nothing else that's happening now does the same.
And here I strongly disagree, because what's happening now is, we have created a revolution in the way we access and use information. And to argue that that is not as revolutionary, as transformational as the way that we can now move from the U.K. to the U.S. faster, I think, is absolutely wrong. There's no doubt in my mind that the way in which we can access and use information has incredibly strongly increased the power we have as economic agents and the freedom we have as individuals, hence the risks that you were point out of what happens if we try to stop that.
MR. HARDING: This is probably another question for Andy, but could you talk a little bit about what the technologies are that are going to drive this increase in productivity and how they're going to do it? Because sometimes I hear, you know, robotics or biotechnology or whatever, and people have been saying for 20, 30 years without this point necessarily be reached.
MR. MCAFEE: That – right.
MR. HARDING: So explain to us –
MR. MCAFEE: So my community of technologists has been promising the moon to you-all for as long we've had digital technologies. So you could be forgiven for kind of turning a deaf ear to these hyperbolic claims that we all keep making over and over again.
And to back up a second, Bob Gordon is absolutely right that in recent years we have seen a productivity slowdown. Our economic engine is not doing as well as it was in some previous time periods. Is that a blip or is that the new normal?
Robin, to answer your question, the reason I think it's a blip and absolutely not the new normal is we've seen two fundamental developments in recent years that I think are going to put us on an entirely different trajectory. The first is the – not just the appearance, the demonstration of – I don't know what adjective to use. Let's call it real artificial intelligence. So the fact that Watson could beat the two best "Jeopardy!" players in the world – this was not some cute lab demo. This was not a constrained thing. This was an application of real artificial intelligence.
The reason we now have real artificial intelligence, as opposed to the very fragile and brittle and underwhelming stuff that we had before, is exactly to your point. It's data. It's these oceans of digital data that we're sitting on which are giving the geeks entirely new approaches to some long-standing problems in AI. Every – all the advances that we're seeing come out of fairly smart algorithms applied to oceans of data to do things that look like human intelligence. They're not. They're nowhere near the – nowhere close to the way our minds work, but they can mimic it very, very well. So that's the first one.
The second one is what we've been talking about as well, which is the fact that honestly, for the first time, we have billions of people interconnected on the same digital network. This has never, ever been the case before in human history, and it's very easy to underestimate how big a deal that is.
Of – the economist Julian Simon is one of my intellectual heroes, because he was a voice in the wilderness for a long time, when there was a lot of pessimism about the environment and about progress, and he said, no, you underestimate the single biggest engine of betterment and productivity that we have are the minds of the people in the world. So part – so population growth is good news, not bad news. And just to add a little bit to that, population growth, when we can interconnect those brains and have them work together – we have never, ever seen anything like this before. It's a recent phenomenon. It's only going to accelerate.
A.J. Liebling was a wonderful journalist. He said freedom of the press is limited to those who own one. You could say the same thing about the freedom to share information, the freedom to compute. That's being democratized at a rate that astonishes me, that we've never, ever seen before.
So it's only a tiny exaggeration to say that for hundreds of millions of people around the world right now, they have access to almost the same computing power that I have as a member of the MIT community. That's fantastic.
Again, we ain't seen nothing yet. The devices are getting cheaper. The access is going up. So the community of potential innovators out there in the world is increasing by leaps and bounds, and going to do so at ever accelerating rates. That gives me great optimism.
MR. HARDING: So we haven't got into the labor market side of this and the effects on work. I want to start taking questions in a moment – oh, start straightaway. So Peter –
Q: I'll wait. Go ahead.
MR. HARDING: OK. Well, I'll – so I'll do one more and then we'll start taking questions.
But I wanted to ask Dan, because it seems – always seems to me like Washington finds these issues totally incomprehensible and extremely hard to react to politically. So we saw President Obama give a speech about inequality and the economy last week, but beyond the sort of repeated mouthing that we need to upgrade skills, there doesn't really seem to be a coherent response, and Washington doesn't even seem able to think about how it might – going about mounting a coherent response. Did you have any thoughts on the politics of this and what the political agenda should be to try and help society adapt to these changing technologies?
MR. PRICE: It's a great question, and as you know from having spent time in this city yourself, the political impulse is never to do nothing. The political impulse – and there was a member of a panel yesterday who I think expressed this – is to think that government is the answer and can supply the answer and a superior answer to that which the private sector may supply.
So my concern is that whether it's through the tax code or whether it's through regulation of the Internet or the new pipes – and for the guy who said, well, would you trade it for a flush toilet, I mean, what we're talking about is a new set of pipes, right, and do you trade your toilet for running water or for – I mean, it – so my concern is that government will be tempted to intervene and intervene in ways that are not necessarily productive.
The last point I'll make on this is as follows. It is true that we have seen a fairly rapid concentration of wealth. We need to address that. A way to address that is to take and redistribute. Another way to address that is to lift up those who need help.
I am hopeful that the solutions offered not only by the U.S. government but by governments around the world will be some wise combination of those and lean more towards the latter than the former.
MR. HARDING: Great. Well, we will start taking questions now. If you could identify yourself and your affiliation when you ask a question – microphones. Great. So the gentleman here.
Q: Good morning. I think the panel did remarkably well. So I don't want to disagree. I just want to add something to what the panel has said.
Two points – well, actually three. First of all, in your newspaper this morning, there were two excellent articles, worth – relevant to the conversation today, one by Gideon Rachman on the loss of faith in the future by the West. And I think this conversation is part and parcel of that. I think this is part of the reason there is a loss of faith, is a kind of pessimism associated with the kind of technological transformations we're now talking about, by – because of misrepresentations and failures of imaginations by people like Bob Gordon.
There was a second excellent article by Gillian Tett, who I think was meant to be here as well, on collaborative tools in organizations, including some provided by my own company, enabling enormous productivity, engagement, at a much higher level. That is, we have automated service at a very low level. What we're now enabling is new powerful tools for people like us. You know, basically the tool we've had so far is the cellphone and the – and the laptop. We are now able to use tools built on the kind of data and systems that will enable higher-level workers to achieve much higher productivity than they ever have before. So I think that is true.
I think the timing issue here is what is very much at issue, on the one hand. If you look at when quantum computing was discovered, it was roughly 1910 to 1925. The first proposal for a serious device using it was 1947, 30 years later, at Bell Labs. They said, we think we might be able to make a solid-state device using quantum physics.
The first practical device was 1974, the first Intel chip that came out using that technology – in other words, over 50 years from the science to the full realization of the technological potential. And I think that's the kind of time frames we're talking about. It may be a bit shorter now because knowledge is spread a bit faster and so on. But the fact is that it takes time to go from fundamental science, like biological sciences, like information sciences and so on, to the full realization of the technology.
So we are still, I think, in the relatively early stages of both IT and biotechnology, now enhanced by that connectivity around the world bringing many more minds to the task. So I think that's number one.
But my own company, Salesforce.com, which provides software to business – we are enabling Marco's company to connect his jet engines to his laboratories and so on, and understand how they are performing in the real world – is a new set of capabilities. Our company was the first cloud computing company. I can take no credit for any of this; I joined the company two years ago.
But since I joined two years ago, we went from 6,000 employees to 13,000 employees in two years. OK? And we're only slowing down because we're just having trouble absorbing all those people. It's the demand. The demand that we can see is almost infinite for our product and our services. And so what we're seeing is a growth rate of 30 percent a year both in revenue and in employment. We are now – we went from zero to the largest employer in San Francisco in a decade.
And I think that kind of dynamic is the reality of the way in which technologies advance. No one was employed in cloud computing 15 years ago. Now hundreds of thousands of people around the world are employed in cloud computing in ways they weren't before. And this is not the first story or a unique story. So it is, I think, typical, not atypical.
MR. HARDING: OK. And could I just ask you to give your name and affiliation.
Q: Oh. Peter Schwartz with Salesforce.com.
MR. HARDING: Great. Thanks.
And the gentleman behind you there. Yes. Either of you. (Chuckles.)
Q: (Off mic.)
MR. HARDING: Whoever wants to go first.
Q: (Off mic) – my friend's going to have an aortic seizure if he participates in – (inaudible). (Laughter.)
MR. HARDING: OK. And –
Q: (Chuckling.) Thank you. Rob Meagley from ONE-Nanotechnologies in Berkeley. We've bandied about now a few different kinds of revolution, and we've talked about fire and the wheel, and we've talked about the Industrial Revolution, and now there's this new wave.
I'd like to point out that there's good evidence that the way we are computing and manipulating data and networking each other is changing the way we think, changing the way our minds are organized internally. The connection in the brain of children that are growing up with the Internet is demonstrably different, subtly, than it was for previous generations. I submit to you that that's probably unprecedented since fire changed our teeth. That's huge, and I'd like to toss that out for discussion.
Q: Thank you. Richard Burge from Wilton Park in the U.K. And I'm anticipating getting back to Britain tonight, actually. So – (chuckling) – so I'd just like to ask about the role of government in a world where we're finding other governments are starting to close down their Internets and prevent access. What is the role of democracies in disrupting the attempts of nondemocracies to close down access to the world for their people and for us to get access to their people? Isn't there a case for saying that cyber is not about defending yourself; it's about actually going on the offensive, not to defend yourself but to actually open up the world for people for whom it is not being opened?
MR. HARDING: OK. Some great themes. Let's have a set of responses from the panel. Marco, do you want to go first?
MR. ANNUNZIATA: Yeah. I'll first start with the point on the Internet and the way we process data changing the way we think but also changing the brains and almost the physiology of the brains. So I think it's – it's an important point which goes to the issue of how innovation is made.
So I think it is – to me, that is one of the most important points, because it shows you, in almost a very concrete way, how information and data are as powerful as electricity. They're tremendously transformational. They change the way we think as individuals, but they also are training us to think in a way which is much more interconnected, much more open, enormously more so than it was ever the case before.
And to me, that ties in with the second point, the second question, which is, what is the role of democracy? I think the role of democracies has to be to continuously push for more openness and more open trade of information, the way we've been pushing for more open trade of goods and services.
MR. HARDING: And Andy.
MR. MCAFEE: Could I just pick up on that? On the – on the role of government, I've read really fantastic reports about how the State Department is deploying mobile Internet technologies in the sense that you can drop them out of a plane and provide Internet access in places where the government thinks they have shut it down. That's just fantastic, right? I'm just in favor of tools that let people express themselves and remain part of this community.
But if I could broaden that question out a bit, I talk to fairly right-leaning and fairly left-leaning economists about these issues, and you have to go a long way in either direction before you reach really serious disagreement about the role of government to facilitate this period that we're living through.
And my silly little mnemonic for what the proper role of government is, is a pickup of the old "Old McDonald" nursery – "Old McDonald" nursery rhyme. I just repeating to myself, E-I-E-I-O, and that means education, infrastructure, entrepreneurship, immigration, and O is original research, your basic research. You've got to go a long way before you'll find people who disagree with those things.
How are we doing? This is – this is a time of pretty serious pessimism about government's ability to get anything done, and a lot of that is warranted. But on my E-I-E-I-O, I think we're about 2-for-5 these days. We came very close to comprehensive immigration reform. We might well get there the next time Congress comes together. And there is pretty – a widely shared understanding about how important entrepreneurship is, and we might be doing things like putting a startup visa program in place or facilitating the environment for entrepreneurs. So I'd rather have us be 4-for-5 or 5-for-5, but 2-for-5, batting .400, got Ted Williams into the Hall of Fame. Let's not – let's not lowball that.
MR. HARDING: Dan.
MR. PRICE: Well, I certainly y agree with you, and I think U.S. policy has for generations put a lot of store in the principle that the cure for bad speech is more speech. I mean, the heart of our First Amendment – when you look at radio broadcasts in the earlier part of this century, during times of war, whether hot war or cold war, I think you see a pretty consistent theme there.
The modern-day incarnation of this, which is worth watching for those of you who may not follow trade negotiations closely, is the digital economy provisions and the cross-border data flow provisions now being negotiated I the Trans-Pacific Partnership agreement, which include countries of varying levels of development – Malaysia, Vietnam, United States, Canada, Mexico, Australia, Japan – and will those countries get it right?
And to put a finer point on it, the same issues are coming up in the trans-Atlantic Trade and Investment Partnership negotiations, the FTA, the regulatory coordination exercise, between the EU and the United States. And these strands of whether it's Snowden and privacy or industrial policy, it'll be very interesting to see how those play out in those negotiations, because if the U.S. and the EU can't get it right on data flows, we will have a hard time persuading China that we've got the right balance.
MR. HARDING: So another round of questions. This gentleman here first.
Q: Thank you. Olof Ehrenkrona, Swedish minister for foreign affairs. Thank the panel for fantastic contributions.
I would say one of the big differences between the revolution we see today and the one we saw in the – in the 18th century and 19th century is the fact that we have a number of technological breakthroughs that actually goes together. And what combines them is the software revolution – ICT and biotech, ICT materials, ICT and robotics.
And in fact the software has – capacity has developed faster than the hardware capacity for quite a number of years now, actually. And the way that we through mathematical algorithms can change the physical – this is the big thing. This is really the big thing. And what also goes with it is the universal language for mathematics, which is for everyone to pick up. So it's not only that everyone is connected through the same Internet. It's also de facto that there is a potential for everyone to speak to everyone else in a way that everyone understands.
And that gives me to draw the conclusion that what we should do is actually to introduce programming as a – on a broad scale in the curriculum in the schools because this is the tool for our kids and their kids to communicate, and in a universal language. But it's also the tool which opens the space of open software and open source – this enormous space where you can develop what others have done before and create small wonders every day. So I would say this is – this is one of the things that I think we should actually advocate.
With regard to the labor market, I'm brought up in a farm in Sweden during the mechanization of agriculture. And I can tell you that it was not the smartest that stayed on the farm and used the new mechanical tools. Many of those who would have biggest problem in school with the theoretical part if the curriculum stayed on and could manage the new mechanized agriculture. And being in India – outside India Business School in Hyderabad, I saw them mowing a lawn. One was pushing the lawnmower. One was dragging the lawn mower. And the third person was supervising the two others.
And the big tragedy in India is that the politicians believe that the way to create jobs and to preserve jobs is to have – to have these kinds of operations. In the U.S. today, I would say that you mostly mow the lawn without anyone doing it. So I would say it is not a threat manual laborers that – if we mechanize that and increase productivity. It's a big opportunity for them as well.
MR. HARDING: Thank you. So, the gentleman in the third row and then the gentleman in the sixth row then.
Q: My name is Walter, W & L Global Enterprise. I just observed that United States once was the industrial power. Now we switch to service economy. Not everybody can go and study the computer science. Many of them are the hamburger flippers. So my question to you, when the industrial part can wake up finally and realize the best way is to mix between industry and service economy – not shipping the jobs overseas in other countries, especially in China where the wages is impossible to complete. So can we figure out how to mix between industrial and service economy?
MR. HARDING: Thank you. And the gentleman in – by the aisle – just back there on the sixth row.
Q: Yes. Randall Fort with Raytheon. We've talked about several of these threads or areas of progress – computation, there's biology, there's robotics, manufacturing. And it seems like we kind of focus on these things each in their own – each on their own little area. I'm curious, as we look forward these things seem to be increasingly interconnected and merging somehow. So it's sort of like sick building syndrome, you know, when they say, oh, well, we've looked at the carpeting and it's fine and we've looked at the paint and it's fine and we've looked at the coatings on the HVAC and it's fine. But nobody ever looked at how all those things interacted.
So where do you see these things as they go out and everything increasing at Moore's law progressions and we get to bioinformatics and we get to biological manufacturing and we get to these computation rates which affect all that. How – what will be the impact of these various areas of technology coming together in the out years at those astonishing rates of exponential change?
MR. HARDING: OK, so we'll take another round of comments from the panel. Maybe we'll start with Dan and go this way.
MR. PRICE: I think, to pick up your last point, I will let those who are the real technologists address that aspect of it. But one element kind of underexplored, which I think eventually we'll get around to exploring, is the ethical dimension of having more perfect information. How do we allocate organs for organ transplant in the face of more perfect information on patient survivability or span of life?
How do we make decisions on allocation – I picked a colorful one – but how do we make resource allocation decisions once we have closer to perfect information? And do we do it solely on the basis of what the data would suggest would yield the highest return for the smallest allocation or resources. And the last question I'll pose is who makes those decisions?
MR. MCAFEE: And if I could pick up on that, you asked the excellent question, where is all this headed? I don't know. Nobody knows. I mean that in all seriousness. If you – if you look at the track record of deep domain experts at doing predictions anywhere beyond about 12 months out, in there are of expertise, it's essentially random. Any of us would do about as well. So anyone who can – who will tell you they know where all this is headed in five or 10 years, is lying to themselves or lying to you.
One thing I can say with a lot of confidence is that Moore's law, this astonishing improvement in computing capacity, is going to continue for another human generation, anyway – for 20 years anyway. I asked a really good computer scientist once if Moore's law was about to run out of steam. And he said yep, in about five years – same as for the past 30 years. (Laughter.)
So what I have great faith in is the innovators of the technology industries keeping this run going of amazing computing bounty. How is that going to be combined? How is that going to deployed? Nobody knows. You know, fasten your seatbelts, we're in for a deeply weird and, I think, a good time.
You bring up the question about the right mix of industrial and service activities in an economy, there's a very big misconception that the U.S. is no longer an industrial powerhouse. It is just categorically false. Our manufacturing industry grows year after year in terms of real value added. It grows every non-recession year, year after year. The year of peak manufacturing employment in the United States was right around – it's either 1979 or 1980.
So the story of manufacturing is incredibly clear: It's output going up and employment going down, both in raw levels. It's a productivity story. That manufacturing story, that output up and employment down, is now happening around the world. Global manufacturing employment is going down. Global manufacturing output is clearly going up.
Now, if you take that same phenomenon and expand it to other industries, actually which I think is not completely unlikely, then we find ourselves again with these labor market challenges. I don't know the economic law that says that there will always be another industry there to pick up the slack, to pick up the people made redundant by productivity increases over there in that industry. I just don't know of that economic law.
MR. ANNUNZIATA: On the last – on the point of services and industry, innovation is already blurring the lines between industry and services, between manufacturing and services – especially as the industrialized – as the machines become more intelligent. As you sell a machine, you're selling a set of solutions and services which come with it. So the distinction is really becoming more artificial and less useful than it used to be.
What's important also to recognize is that there are good jobs and bad jobs, both in manufacturing and in services. And what we really want, and this links to the point that the minister made, is we want to increase education to make sure that a larger set of people can get good jobs. Now, the point that the minister made I thought was interesting because there is the argument that everybody should learn basic coding, but you were saying something very interesting, that the people who stayed in the farms and benefited the most from the technology, from mechanization, were not the ones who would have learned out to build the structure of this new farm equipment.
So I think it's important to – as we think of reforming education – to first of all treat scientific literacy the way we've treated basic literacy. So we need to give our students better training in STEMs – math, science – the bar needs to be raised. I also recognize that not everybody needs to be computer scientist.
And finally, please, let's step away from the idea of shipping jobs overseas, taking jobs back, because I'll tell you what happens in my company – for example, in GE very quickly. We put jobs in India to innovate. Because our scientists are in India, they come up with new, cheaper, better medical devices that save lives in India, can then be reverse-engineered into the U.S. to help the U.S. health care system, but also contribute to the creation of jobs in the U.S. The global economy is not a zero-sum game.
MR. HARDING: OK, another very quick round of questions. You get a maximum 30 seconds now so we can get everyone in. The gentleman here first.
Q: Sergio Bitar from Chile, former minister of education, at the – at the Inter-American Dialogue now. From your projections, are there important issues on education – on the way we educate and the way the institutions of education are going to be reorganized, or the contents of education or the use of technology for improving teachers training – that you consider that are not well-taken into account now by the educational sector, that you will emphasize?
MR. HARDING: Thank you. Did you want – yes, one here.
Q: Hans Christian Hagman from the Swedish Ministry of Foreign Affairs. Thanks for an excellent panel. It's tempting to ask the question, when we're talking about global or government intervention on – is government part of the solution or part of the problem when it comes to continued technology development, but I twist that around rather to ask how can and should technology change government and governing? If you could just – (inaudible) – a bit on that? Thanks.
MR. HARDING: Thank you. I saw some hands back there. Yes, the gentleman there, and I think someone else behind him.
Q: Just – Byron Calla, Capital Alpha Partners. Just a question on how this changes optimum corporate organization. Does it really still favor the global corporate multinational firm? Does it fragment businesses? Some thoughts on that, please.
MR. HARDING: The gentleman behind you.
Q: My basic question was how do we talk to – I think this is a very elite-led discussion. I think all these points are really interesting, but my sense is it's very much something that people read about Scientific American, the FT, The Economist or talk about at conferences like this. So how do we – but these things really shape people around the world outside of cities and really have a – quite a significant economic impact on everyone. In terms of getting to the political part, how do we even talk about this in a way that people understand how this impacts them in a daily life and in terms of how to engage our democracies on this?
MR. HARDING: OK, any more questions or comments? Yes, gentleman over there.
Q: Hi, Jason Healey with the Cyber Statecraft Initiative at the Atlantic Council. Jason Healey with the Cyber Statecraft Initiative here. And for Internet and cyberspace, the attackers have always had the advantage for decades. If we ever hit a tipping point – I mean, it's difficult to imagine that the attackers will always have the advantage and that's a stable situation – does that change the technological future that you see coming if we go past that tipping point and attackers just continue to have more and more advantage? Thank you.
MR. HARDING: Oh, yes.
Q: Stan Schmidt, Scientific American –
MR. HARDING: Please, wait for the microphone.
Q: Stan Schmidt, Scientific American. Question for you actually, Robin. The point was made yesterday and it – I think might attach to some of these points, but the notion that much innovation is essentially being financialized, and thus there is all sorts of sort of false incentives and financial, for whatever you want to call it, gamesmanship that really is underpinning. And I was just wondering if you could sort of touch on that a little bit.
MR. HARDING: All right. Well, while I think about that, the panel would like to make a final comment. Andy, you go first since you –
MR. MCAFEE: Can I pick up two other questions – first of all about education and second of all about how to organize companies in this world. There's a really interesting question – open question these days about what the role of all these amazing digital technologies is for education.
And there are two things – two possibilities. One is that it's a rising tide, that it can take all the school kids in a district or in a country and make them smarter. So far, our experience has been disappointing in that regard. When we've tried to deploy cutting-edge technologies and raise up, especially lower-performing students, the results were a little bit underwhelming. Now, it's an open question because this is still so new. Will we get better at using technology as a – as a tide that floats all boats upward? Open question.
What we do know is that technology for education is what I call a diamond detector. In other words, if there's a kid out there, almost anywhere in the world, who has the intelligence and the thirst and the tenacity to educate herself about cutting edge topics, she has access to some of the best teachers on the planet, some of the best learning materials on the planet. So what we're observing is all these little diamonds becoming evident around the world. And the global community of data scientists, for example, is very young, very global and very, very self-taught. I think that's amazing development. So I hope it becomes a rising tide. It's absolutely a diamond detector.
And then your question about organizing big companies, I think we are still increasingly heading into a world of big global companies, simply because of the premium on being a well-managed company is going up in the world we're heading into, instead of going down. As far as I can tell, a well-managed company in the world we're creating is one that's a little bit less hierarchical and a little bit more open to good ideas coming in from outside.
MR. HARDING: Outstanding. Dan, very quick comments to finish.
MR. PRICE: Let me pick up on the point of how do you bring others into this. And I think the way you bring others into this is by seizing some of the existing negotiating or consultative forums with a set of these issues. The G-20, I think, did remarkably well immediately post-crisis in designing reforms to the international financial system, many of which are still being implemented through various bits of domestic legislation around the world, one of which is being signed today here.
Through negotiating forums, whether it's TPP or TTIP, through the G-20, I would say there's an opportunity to put in place standards and rules governing data flow, governing privacy, creating this balance between competing interests so that we don't descend into unilateralism and eventually isolationism.
MR. HARDING: Thanks. And Marco?
MR. ANNUNZIATA: My last point would be that we really need to – (coughs) – sorry – we need to tell a better and more accessible technology story to bring everybody on board. And we need to separate it from the economic crisis over the last five, six years. And we need to make it vivid for people that, before the financial crisis, track what we had seen were waves of technological innovation that made life better, made products cheaper and created new jobs.
And we have to give the examples. We know it happened. We have to bring out the examples and explain how this is set to happen again. Otherwise, we get stuck into this set – mindset of technophobia, which is unhelpful and might stop the progress. So telling a better, more accessible technology story is essential.
MR. HARDING: OK. We're out of time so, Stan, I'll give you an answer directly. But please join me in thanking Marco, Andy and Dan for their very interesting – (applause).