McKinsey Global Institute: AI Critical for Productivity, Economic Growth
At a time when productivity is stagnating around the world, artificial intelligence and automation offer a lifeline. These advanced technologies bring the promise of lower costs, more efficient business processes, and innovative new business models. But adopting these technologies will disrupt work as we know it, requiring a markedly different set of skills and the ability to adapt to working alongside increasingly capable machines. Businesses will have important choices to make in response to these challenges.
These are the views of James Manyika, Chairman and Director of the McKinsey Global Institute, the business and economics research arm of McKinsey & Company. In the following interview, James discussed the impact of AI, the future of work, and what companies should be doing to prepare for the future.
Q. In a recent Harvard Business Review article on automation and the future of work, you mentioned that we are living in a time when “we need productivity growth more than ever.” Could you explain why that is? What’s causing productivity to stall in advanced economies?
The past 50 years have been a remarkable period for global GDP growth. It has been driven in equal measures by two things: expanding the labor pool and increasing the productivity of workers. But now, one of those drivers is running out of steam. As the workforce ages in many of the world’s largest economies, including in the U.S., we’re facing a serious slowdown in labor supply growth. That means if GDP growth is to be sustained in the U.S., Europe, Japan, and other advanced economies, productivity growth must carry a lot more weight. To put it another way, we need much more productivity growth, because if productivity growth holds at historical averages while labor supply growth stagnates, we could see a potential 40% decline in global growth rates.
Q. Technology has been widely hailed as having the potential to boost productivity. We’ve seen some tremendous advances over the last decade or so, with breakthroughs from cloud computing to smartphones and the Internet of Things.
However, in the U.S., productivity growth during this period has been dismal. In fact, some researchers are claiming we’re seeing the slowest growth rate since records began in the late 1800s. What’s your take on this apparent contradiction?
I think we could be seeing round two of the Solow Paradox. In 1987, MIT economist Bob Solow famously quipped to the New York Times that “you can see the computer age everywhere but in the productivity statistics.” That paradox was only resolved in the late 1990s, when the full benefits of the computing revolution started to show up in the economic measurements. By then, leading players in major sectors like wholesale and retail had adopted the latest technologies to transform their supply-chain and business operations. Other players followed suit and productivity rose across the board.
Something similar could be happening right now. Digitization has not yet reached scale. While the information and communication technologies (ICT), media, financial services, and professional services sectors are rapidly adopting digital technologies, larger sectors such as healthcare and construction are not digitizing as quickly, and are probably less digitized than is generally acknowledged. The most impressive capabilities of AI and machine learning are even further away from widespread adoption.
Once business innovation in using these technologies catches up with the opportunities that these technologies offer, productivity numbers are likely to improve. That’s the point at which adoption moves beyond process optimization to fundamentally transform business models and alter value chains.
We believe the potential for outsized gains could be even greater this time around because of the scale and network effects associated with automation, AI, and other intelligent technologies.
Q. You mention AI and its potential to deliver outsized benefits. We recently spoke with AI expert Dr. Kai-Fu Lee, who has suggested that “focusing on incremental gains like using AI to improve customer service is missing the bigger opportunity … the real opportunity is for companies to rethink their entire business models.” Can you explain how AI can drive new business opportunities in this way?
The McKinsey Global Institute recently analyzed more than 400 use cases across 19 industries and nine business functions to highlight the broad and significant business potential of advanced AI techniques. We found that as AI applications start to reach maturity, companies that are swapping in AI and rethinking their business stand to gain a serious competitive advantage.
For example, AI deployment is helping companies discover new ways to serve customers. Take the pharmaceutical sector — an AI algorithm can comb through vast sets of data, finding hidden patterns, and help to identify promising molecules earlier on, that can then more quickly develop into new drugs. Other benefits include significantly boosting performance, raising throughput, improving predictions, outcomes, accuracy, and optimization — and yes, enhancing productivity.
As Kai-Fu Lee and others have pointed out, AI-driven models and capabilities are also disrupting entire business models, pushing incumbents to revisit core business expectations.
Q. What should companies be doing now to take advantage of these opportunities?
Companies should focus on AI use cases where there are proven technology solutions today that can be adopted at scale, such as robotic process automation and some applications of machine learning.
Building a data ecosystem is also vital to deploying AI. Data is at the heart of the disruptions occurring across economies and without it, getting the AI engine started is impossible.
Apart from organizational changes, new skills are also needed for successful AI. Aside from the need to reskill workers to collaborate with machines, specific talent is required to develop and deploy AI solutions. One challenge here for companies is the shortage of AI and machine learning experts who can apply these techniques.
Q. You mention the need to reskill workers. With the adoption of AI and automation, we’ll see machines carrying out more of the tasks done by humans, complementing the work that they do, and even performing some tasks that go beyond what they can do. You’ve written that you believe there will be enough work to go around. Can you explain this a little more?
Even as some workers are displaced, there will be growth in demand for work and consequently jobs.
While it is difficult to make predictions, at MGI, we developed scenarios for labor demand to 2030, based on various catalysts of demand for work such as productivity growth and rising incomes, as well as factors such as demographic trends leading to increased spending on healthcare. These scenarios showed a range of additional labor demand of between 21% to 33% of the global workforce (555 million and 890 million jobs) to 2030, more than offsetting the numbers of jobs lost, which our midpoint scenario was 15% of the global workforce, or 400 million workers. So there will be “jobs gained” as well as “jobs lost.”
Another important point is that there will be “jobs changed” as well. When we looked at the tasks within occupations that could be technically automated, we found that only about 5% of occupations can be nearly fully automated, from a technical standpoint. But 60% of occupations had about a third of their tasks that could be technically automated. So many more jobs will change than be automated away.
Q. McKinsey research suggests that many new occupations will emerge and may account for as many as 10% of jobs created by 2030, if history is a guide. What are the implications of that for the private sector?
Workers will need different skills to thrive in the workplace of the future. We’ll see a growing demand for advanced technological skills such as programming. Social, emotional, and higher cognitive skills — such as creativity, critical thinking, and complex information processing — will also be in demand. Growing occupations, meanwhile, will include those with difficult to automate activities, such as managers and doctors, as well as care workers and teachers. Training and retraining mid-career workers and new generations for the coming challenges will be another imperative.
To meet these demands, business leaders must substantially increase their investment in human capital. They will need to work together with governments to better coordinate public and private initiatives in this regard, including ensuring the right incentives to invest more in human capital.
Businesses have the opportunity to play a major role when it comes to growing the educational ecosystem in order to provide solutions as scale. That may involve working with colleges and universities to develop apprenticeships and other types of visiting programs, or even simply providing better information signals about the skills they are seeking.
Businesses will be on the front lines of the workplace as it changes. Many companies are finding it is not just in their self-interest, but part of their societal responsibility, to train and prepare workers for a new world of work.