RALEIGH — When my first book, “The Heroic Enterprise,” was published by Simon & Schuster in 1996, I asked about releasing it on audio. “Let’s see how it goes,” my editor replied, warning me that while there was an audiobook market by then — primarily in the form of cassette tapes — it wasn’t very large. If my book on corporate social responsibility turned out to be a bestseller, there might be enough demand to justify a cassette version.

John Hood

There wasn’t. To salvage my ego, I concluded that writing a bestseller about corporate social responsibility was a metaphysical impossibility.

Now, my editor and I had this conversation in 1996. It never occurred to me to ask if consumers would have the option of reading “The Heroic Enterprise” in digital format on their desktop or laptop computers, though of course the book had been written in digital format. I didn’t ask about tablets or smartphones, either, for obvious reasons. Nor did I wonder what might happen to the audiobook market once listeners no longer had to purchase physical tapes or CDs and lug them around.

Predicting the path, pace, and implications of technological change can be extraordinarily difficult. Relevant expertise may boost your accuracy a bit — but then again, acquiring such expertise may embed you so deeply in the past and present of your field that it’s hard to imagine a radically different future.

The list of famously wrong predictions by seemingly reputable experts is long. “The Americans have need of the telephone, but we do not,” said the chief engineer of the British Post Office in 1876.

“We have plenty of messenger boys.”

In 1943, the chairman of IBM forecast a world market for “maybe” five computers. “Television won’t be able to hold on to any market it captures after the first six months,” movie mogul Darryl Zanuck insisted in 1946. “People will soon get tired of staring at a plywood box every night.”

Not all faulty predictions err on the side of skepticism. Many of the techno-optimists who predicted flying cars by the 1970s or moon bases by the 1990s were scientists and engineers. They made straight-line extrapolations from limited data, or failed to consider the real-world costs and benefits of theoretical advances.

Some of the thorniest issues in public policy today arise from present or forecasted changes in technology. At a mid-May committee hearing in the U.S. Senate, three specialists in artificial intelligence presented three very different takes on what the federal government should do to help Americans reap the potential economic benefits of AI while protecting against threats to privacy, academic rigor, national security, and social cohesion. One argued for a new federal agency to regulate the technology, for example.

Another said that was a bad idea.

Here in North Carolina, lawmakers are grappling with the effects of automotive innovation — higher fuel efficiency for gas-powered cars and the proliferation of hybrids and electric vehicles — on highway revenues. And at the local level, COVID-19 acted as a catalyst for remote work by giving employers, employees and customers a crash course in videoconferencing.

Have we seen a permanent shift in how many people work and where they live? If so, that has implications for housing policy, transportation spending, the vitality of downtown business districts, and the operation of city and county governments.

No one knows for certain how these trends will play out. My point is that no one can ever know for certain. Good public policy doesn’t require certainty. It requires the application of a discrete number of clearly articulated rules to a large number of fuzzy, ever-evolving cases. It also requires healthy doses of humility and flexibility.

I’ll end where I started: with publishing. It may surprise you to learn that in 2021, e-books accounted for only 10.5% of total sales. That’s down from 13% in 2017. Paper books still reign supreme at 80%, while audiobooks (in digital form) are at 9.5% and growing rapidly.

Predicting tech trends is hard.

John Hood is a John Locke Foundation board member.