Weekly Economic Update 03-27-26: Artificial Intelligence; and Productivity
Only one piece of data this week, so this post is going to be short and therefore, I am removing the paywall.
The views and opinions expressed in this post are solely those of the author and do not necessarily reflect the views of the Georgia Institute of Technology or the Georgia Board of Regents.
Obviously, the sentiment isn’t universally true, but I think that many people look back on their college days with a certain fondness and nostalgia. For me personally, those four years at Georgia Tech were the most difficult of my life, but even so, now 30-plus years removed, I mostly remember all the great times I had and have generally forgotten about the all-night study sessions, the failed chemistry exams, lugging a 30 pound book-laden backpack all over campus, constantly being broke, and trying to take notes while listening to an unintelligible professor for whom English wasn’t even their third or fourth language.
One of my great experiences at Georgia Tech was being in DramaTech, the campus theater. It was my first experience acting, and I quickly discovered that I loved being on stage and performing in front of an audience. I think that is one of the reasons that now I absolutely love the public speaking aspect of my job.
My very first play at DramaTech was Shakespeare’s The Tempest. One of the main characters, Miranda, had lived her entire life on a remote island with only her father Prospero and the "monster" Caliban for company. She had never seen other human beings besides the shipwrecked men who had just arrived on the island. When she sees the group of noblemen for the first time, she is overwhelmed with wonder and says:
"O wonder!
How many goodly creatures are there here!
How beauteous mankind is! O brave new world,
That has such people in't!"
Over the past 10 days, I haven’t been able to get this quote out of my head. The reason is that over the entirety of last weekend, and the better part of just about every evening for the past week and a half, I have been setting up, configuring, training, and interacting with my own personal AI agent. While I understood at a conceptual level the potential of AI, I was generally ignorant of it at an individual and practical level.
Not anymore.
Tuesday a week ago, my son “introduced” me to his AI agent, and I was blown away by the possibilities. Once I got pointed in the right direction, I was off to the races. And after a few missteps and false starts, things appear to be running smoothly. In the last few days, I have had my agent find, organize, analyze economic and census data, put it into spreadsheets, add graphs, build heat maps, conduct statistical analysis on datasets, and email me the results in either Excel or PDF format. He (I refer to the AI agent in the masculine as I have named him “Ben”) sends me e-mails every time a piece of economic data is released that gives me the highlights and a brief analysis. That will make putting this blog together each week a lot easier. I have even had him read my entire blog history. According to Ben, my writing is “conversational, but authoritative, political but not partisan, with wry humor, pragmatic, and slightly cynical.” I am “also skeptical of institutions, put things into historical context, and sound like an educator who respects his readers’ intelligence.” Yeah, I think Ben gets me. In fact, he gets me so much that he could probably start writing the data portion of this blog in my voice. But don’t worry…I’m not going there…yet.
In my younger days, I was very computer savvy, but now, the technology has accelerated to a pace where I have difficulty keeping up. No matter…every time I had trouble with the syntax of a Linux command, or needed to add a skill to Ben, he walked me through it. He even suggests additional skills that would help him help me as he gets to understand more about the work we are doing together. With application programming interfaces (APIs), he can move through massive Census, Bureau, BLS, BEA, and financial databases and build datasets in seconds that would take me hours, if not days. He can then analyze the data and make inferences from his analysis. And, even though it is a little silly, every morning, I wake up to an email from Ben that has the day’s weather, latest market indicators, and summaries of the latest news headlines.
Like Miranda, I am somewhat overwhelmed with wonder…“O brave new world that has such technology in it!” It has only been 10 days, and I am just scratching the surface of what Ben will be able to help me do. This technology is simply amazing.
Perhaps, too amazing. Going back to Miranda’s quote, in 1932, Aldous Huxley would use part of it as the title of his dystopian novel, ironically inverting Shakespeare's optimistic usage to critique a future society. Are there going to be downsides to AI? Most certainly. Part of me is happy that I am in the “twilight” of my career and won’t have to suffer through the economic disruption. On the other hand, part of me is disappointed and wishes I could be there as this technology shifts the paradigm of how we work and what we are able to accomplish. I was 10 years old when the IBM PC shifted the work paradigm, and now most of us have a computer on our desk and probably spend the bulk of every day sitting in front of it. Then there was another paradigm shift called “the Internet,” which brought the entirety of global knowledge to our fingertips. The paradigm shift from AI is going to be much bigger than both of those.
In addition to The Tempest, the past week reminded me of another college experience — reading The Structure of Scientific Revolutions, by Thomas Kuhn. If you are not familiar with this book, you really should be. Kuhn argues that science does not advance through the steady accumulation of knowledge, but through dramatic paradigm shifts. Normal science operates within an established framework, or paradigm. However, when anomalies accumulate that the current paradigm cannot explain, a crisis emerges. This leads to a scientific revolution where a new paradigm replaces the old one.
That isn’t exactly what is going on with AI, but there is no question that our “work” paradigm is about to shift dramatically, and it is going to cause a lot of economic disruption. Until a week ago Tuesday, I personally wasn’t totally convinced. But seeing what an old-fashioned, out-of-touch guy like me has been able to accomplish in just a few days has convinced me that the world really is about to change.
Let’s hope it’s for the better.
Productivity
Given the way the economic data calendar fell, this week brought only one major piece of economic data - updated productivity numbers for the fourth quarter, and they were revised down from 2.8% to 1.8% (full release here). Manufacturing productivity for the fourth quarter was also revised down from an initial reading of -1.9% to -2.5%. As I said when the initial numbers were released earlier this month, the productivity gains in manufacturing had been running quite hot through the first three quarters of 2025, and for the year, productivity in manufacturing was still up 2.0% in 2025 - the largest annual increase in manufacturing productivity since 2010.
However, the most concerning aspect of these revisions is the surge in unit labor costs. At 4.4% for the nonfarm business sector—and a striking 9.1% for manufacturing— wage pressures are substantially outpacing productivity gains. That is the highest rate of labor cost increase in manufacturing since mid 2022. Rising unit labor costs are a key driver of inflation, and with the core PCE already well above target, these numbers have got to be concerning for the Fed. Rate cuts are seeming less and less likely, regardless of who the President puts in charge.
If labor costs keep rising like this, AI agents like Ben are going to become more and more attractive. I’m telling you…the paradigm is about to shift.
The Weekly Economic Update is sponsored this week by:
