I have a new R Street Institute policy study out this week doing a deep dive into the question: “Can We Predict the Jobs and Skills Needed for the AI Era?” There’s lots of hand-wringing going on today about AI and the future of employment, but that’s really nothing new. In fact, in light of past automation panics, we might want to step back and ask: Why isn’t everyone already unemployed due to technological innovation?
To get my answers, please read the paper! In the meantime, here’s the executive summary:
To better plan for the economy of the future, many academics and policymakers regularly attempt to forecast the jobs and worker skills that will be needed going forward. Driving these efforts are fears about how technological automation might disrupt workers, skills, professions, firms and entire industrial sectors. The continued growth of artificial intelligence (AI), robotics and other computational technologies exacerbate these anxieties.
Yet the limits of both our collective knowledge and our individual imaginations constrain well-intentioned efforts to plan for the workforce of the future. Past attempts to assist workers or industries have often failed for various reasons. However, dystopian predictions about mass technological unemployment persist, as do retraining or reskilling programs that typically fail to produce much of value for workers or society. As public efforts to assist or train workers move from general to more specific, the potential for policy missteps grows greater. While transitional-support mechanisms can help alleviate some of the pain associated with fast-moving technological disruption, the most important thing policymakers can do is clear away barriers to economic dynamism and new opportunities for workers.
I do discuss some things that government can do to address automation fears at the end of the paper, but it’s important that policymakers first understand all the mistakes we’ve made with past retraining and reskilling efforts. The easiest thing to do to help in the short-term is clear away barriers to labor mobility and economic dynamism, I argue. Again, read the study for details.
For more info on other AI policy developments, check out my running list of research on AI, ML robotics policy.
This is a compendium of readings on “progress studies,” or essays and books which generally make the case for technological innovation, dynamism, economic growth, and abundance. I will update this list as additional material of relevance is brought to my attention.
Discourse magazine recently published my essay on what “Industrial Policy Advocates Should Learn from Don Lavoie.” With industrial policy enjoying a major revival in the the U.S. — with several major federal proposals are pending or already set to go into effect — I argue that Lavoie’s work is worth revisiting, especially as this weekend was the 20th anniversary of his untimely passing. Jump over to Discourse to read the entire thing.
But one thing I wanted to just briefly highlight here is the useful tool Lavoie created that helped us think about the “planning spectrum,” or the range of different industrial policy planning motivations and proposals. On one axis, he plotted “futurist” versus “preservationist” advocates and proposals, with the futurists wanting to invest in new skills and technologies, while the preservationists seek to prop up existing sectors. On the other axis, he contrasted “left-wing or pro-labor” and “right-wing or pro-business” advocates and proposals.
Lavoie used this tool to help highlight the remarkable intellectual schizophrenia among industrial policy planners, who all claimed to have the One Big Plan to save the economy. The problem was, Lavoie noted, all their plans differed greatly. For example, he did a deep dive into the work of Robert Reich and Felix Rohatyn, who were both outspoken industrial policy advocates during the 80s. Reich as affiliated with the Harvard School of Government at that time, and Rohatyn was a well-known Wall Street financier. The industrial policy proposals set forth by Reich and Rohatyn received enormous media and academic attention at the time, yet no one except Lavoie seriously explored the many ways in which their proposals differed so fundamentally. Rohatyn was slotted on the lower right quadrant because of his desire to prop up old sectors and ensure the health of various private businesses. Reich fell into the upper quadrant of being more of futurist in his desire to have the government promote newer skills, sectors, and technologies. Continue reading →
Wishful thinking is a dangerous drug. Some pundits and policymakers believe that, if your intentions are pure and you have the “right” people in power, all government needs to do is sprinkle a little pixie dust (in the form of billions of taxpayer dollars) and magical things will happen.
Of course, reality has a funny way of throwing a wrench into the best-laid plans. Which brings me to the question I raise in a new 2-part series for
Discourse magazine: Can governments replicate Silicon Valley everywhere?
In the first installment, I explore the track record of federal and state attempts to build tech clusters, science parks & “regional innovation hubs” using state subsidies and industrial policy. This is highly relevant today because of the huge new industrial policy push at the federal level is building on top of growing state and local efforts to create tech hubs, science parks, or various other types of industrial “clusters.
At the federal level, this summer, the Senate passed a 2,300-page industrial policy bill, the “United States Innovation and Competition Act of 2021,” that included almost $10 billion over four years for a Department of Commerce-led effort to fund 20 new regional technology hubs, “in a manner that ensures geographic diversity and representation from communities of differing populations.” A similar proposal that is moving in the House, the “Regional Innovation Act of 2021,” proposes almost $7 billion over five years for 10 regional tech hubs. Meanwhile, the Biden administration also is pitching ideas for new high-tech hubs. In late July, the Commerce Department’s Economic Development Administration announced plans to allocate $1 billion in pandemic recovery funds to create or expand “regional industry clusters” as part of the administration’s new Build Back Better Regional Challenge. Among the possible ideas the agency said might win funding are an “artificial intelligence corridor,” an “agriculture-technology cluster” in rural coal counties, a “blue economy cluster” in coastal regions, and a “climate-friendly electric vehicle cluster.”
In my essay, I note that the economic literature on these efforts has been fairly negative, to put it mildly. Continue reading →
Here’s a new animated explainer video that I narrated for the Federalist Society’s Regulatory Transparency Project. The 3-minute video discusses how earlier “tech giants” rose and fell as technological innovation and new competition sent them off to what the New York Times once appropriately called “The Hall of Fallen Giants.” It’s a continuing testament to the power of “creative destruction” to upend and reorder markets, even as many pundits insist that there’s no possibility change can happen.
Industrial Policy is a red-hot topic once again with many policymakers and pundits of different ideological leanings lining up to support ambitious new state planning for various sectors — especially 5G, artificial intelligence, and semiconductors. A remarkably bipartisan array of people and organizations are advocating for government to flex its muscle and begin directing more spending and decision-making in various technological areas. They all suggest some sort of big plan is needed, and it is not uncommon for these industrial policy advocates to suggest that hundreds of billions will need to be spent in pursuit of those plans.
Others disagree, however, and I’ll be using this post to catalog some of their concerns on an ongoing basis. Some of the criticisms listed here are portions of longer essays, many of which highlight other types of steps that governments can take to spur innovative activities. Industrial policy is an amorphous term with many definitions of a broad spectrum of possible proposals. Almost everyone believes in
some form of industrial policy if you define the term broadly enough. But, as I argued in a September 2020 essay “On Defining ‘Industrial Policy‘,” I believe it is important to narrow the focus of the term such that we can continue to use the term in a rational way. Toward that end, I believe a proper understanding of industrial policy refers to targeted and directed efforts to plan for specific future industrial outputs and outcomes.
The collection of essays below is merely an attempt to highlight some of the general concerns about the most ambitious calls for expansive industrial policy, many of which harken back to debates I was covering in the late 1980s and early 1990s, when I first started a career in policy analysis. During that time, Japan and South Korea were the primary countries of concern cited by industrial policy advocates. Today, it is China’s growing economic standing that is fueling calls for ambitious state-led targeted investments in “strategic” sectors and technologies. To a lesser extent, grandiose European industrial policy proposals are also prompting new US counter-proposals.
All this activity is what has given rise to many of the critiques listed below. If you have suggestions for other essays I might add to this list, please feel free to pass them along. FYI: There’s no particular order here.
A decade ago, a heated debate raged over the benefits of “a la carte” (or “unbundling”) mandates for cable and satellite TV operators. Regulatory advocates said consumers wanted to buy all TV channels individually to lower costs. The FCC under former Republican Chairman Kevin Martin got close to mandating a la carte regulation.
But the math just didn’t add up. A la carte mandates, many economists noted, would actually cost consumers just as much (or even more) once they repurchased all the individual channels they desired. And it wasn’t clear people really wanted a completely atomized one-by-one content shopping experience anyway.
Throughout media history, bundles of all different sorts had been used across many different sectors (books, newspapers, music, etc.). This was because consumers often enjoyed the benefits of getting a package of diverse content delivered to them in an all-in-one package. Bundling also helped media operators create and sustain a diversity of content using creative cross-subsidization schemes. The traditional newspaper format and business is perhaps the greatest example of media bundling. The classifieds and sports sections helped cross-subsidize hard news (especially local reporting). See this 2008 essay by Jeff Eisenach and me for details for more details on the economics of a la carte.
Yet, with the rise of cable and satellite television, some critics protested the use of bundles for delivering content. Even though it was clear that the incredible diversity of 500+ channels on pay TV was directly attributable to strong channels cross-subsidizing weaker ones, many regulatory advocates said we would be better off without bundles. Moreover, they said, online video markets could show us the path forward in the form of radically atomized content options and cheaper prices.
Imagine visiting Amazon’s website to buy a Kindle. The product description shows a price of $120. You purchase it, only for a co-worker to tell you he bought the same device for just $100. What happened? Amazon’s algorithm predicted that you would be more willing to pay for the same device. Amazon and other companies before it, such as Orbitz, have experimented with dynamic pricing models that feed personal data collected on users to machine learning algorithms to try and predict how much different individuals are willing to pay. Instead of a fixed price point, now users could see different prices according to the profile that the company has built up of them. This has led the U.S. Federal Trade Commission, among other researchers, to explore fears that AI, in combination with big datasets, will harm consumer welfare through company manipulation of consumers to increase their profits.
The promise of personalized shopping and the threat of consumer exploitation, however, first supposes that AI will be able to predict our future preferences. By gathering data on our past purchases, our almost-purchases, our search histories, and more, some fear that advanced AI will build a detailed profile that it can then use to estimate our future preference for a certain good under particular circumstances. This will escalate until companies are able to anticipate our preferences, and pressure us at exactly the right moments to ‘persuade’ us into buying something we ordinarily would not.
Such a scenario cannot come to pass. No matter how much data companies can gather from individuals, and no matter how sophisticated AI becomes, the data to predict our future choices do not exist in a complete or capturable way. Treating consumer preferences as discoverable through enough sophisticated search technology ignores a critical distinction between information and knowledge. Information is objective, searchable, and gatherable. When we talk about ‘data’, we are usually referring to information: particular observations of specific actions, conditions or choices that we can see in the world. An individual’s salary, geographic location, and purchases are data with an objective, concrete existence that a company can gather and include in their algorithms.
In recent months, my colleagues and I at the Mercatus Center at George Mason University have published a flurry of essays about the importance of innovation, entrepreneurialism, and “moonshots,” as well as the future of technological governance more generally. A flood of additional material is coming, but I figured I’d pause for a moment to track our progress so far. Much of this work is leading up to my next on the freedom to innovate, which I am finishing up currently.
[originally published on
The Bridgeon August 30, 2018.]
What is an entrepreneur?
While it may seem straightforward, this question is deceptively complex. The term can be used in many different ways to describe a variety of individuals who engage in economic, political, or even social activities. Entrepreneurs affect almost every aspect of modern society. While most people probably have a general sense of what is meant when they hear the term entrepreneur, it can be difficult to provide a precise definition. This is due in no small part to the fact that some of the primary thinkers who have given substance to the term have placed their focus on different aspects of entrepreneurialism.
How Economists Talk About Entrepreneurs
Austrian economist Joseph Schumpeter thought that the purpose of an entrepreneur was “to reform or revolutionize the pattern of production by exploiting an invention.”
Schumpeterian entrepreneurs are highly creative, disruptive innovators who challenge the status quo in order to bring about new economic opportunities. American economist Israel Kirzner viewed the defining characteristic of entrepreneurs as “alertness.” Kirznerian entrepreneurs are individuals who are able to identify the ways in which a market could be moved closer to its equilibrium, such as recognizing a gap in knowledge between different economic actors.
In the time since Schumpeter and Kirzner helped lay the groundwork, a number of George Mason University-affiliated scholars have made major contributions to our understanding of entrepreneurialism. Continue reading →
The Technology Liberation Front is the tech policy blog dedicated to keeping politicians' hands off the 'net and everything else related to technology. Learn more about TLF →
The Technology Liberation Front is the tech policy blog dedicated to keeping politicians' hands off the 'net and everything else related to technology.