U.S. Labor Productivity Is Rising at Its Fastest in 20 Years — and It's Not Because of AI: What It Means for Philippine Operations

U.S. Labor Productivity Is Rising at Its Fastest in 20 Years — and It's Not Because of AI: What It Means for Philippine Operations

In the United States, labor productivity is climbing at its fastest pace in two decades — yet AI is not the main driver, at least not yet. This article looks, from a practical standpoint, at the order in which Japanese companies with Philippine operations should approach digitization and AI adoption.

---

Part 1: Why This Matters

Step 1: The Philippine Business Context (3 min)

In the United States, the output produced by a single worker — labor productivity — has been rising at its fastest pace in at least 20 years. Yet according to the source article, the main driver is still not artificial intelligence. It is described as the cumulative result of a tight labor market where hiring is hard, the digitization of work, and the spread of remote work.

This is not someone else's story for Japanese companies with operations in the Philippines. The Philippine economy leans heavily on work outsourced from abroad — call centers, shared accounting service centers, IT help desks, and the like. The source article's point — that gains come not from a simple "we added AI, so productivity rose," but from digitizing the workflow itself and rethinking how people are deployed — is an important lens for running local operations.

More and more often, head office in Japan asks, "How much can we cut with AI?" In that moment, whether or not you know that even U.S. hard data shows AI's effect has yet to appear clearly in the statistics changes how persuasive your explanation is.

A Manila office. At the Monday-morning standing meeting, the Japanese manager turns his laptop screen toward the team. "Head office asked me how many people we can cut with AI. But if you look at the latest U.S. data, the reason productivity is rising isn't AI itself. Let's first rethink our own workflows. Then we'll decide where to place AI." Maria, a local staff member, nods, looking a little relieved.

Step 2: Organizing the Key Points of the Source Article (5 min)

Here are the facts stated in the source article, organized by theme.

Source: The New York Times — "U.S. Workers Are More Productive Than Ever. And That's Without A.I." (July 14, 2026)

This table was compiled from publicly available facts for educational purposes. For details, please see the original article linked above.

Related: see How AI Automation Helps Philippine SMEs Solve the Labor Shortage Without Hiring More Staff.

Step 3: Comprehension Check (5 min)

Q1. Does the source article explain that the main reason U.S. labor productivity is rising is AI? Or does it point to other factors?

Hint: Early in the article, AI is framed as "one of the ingredients."

Q2. Since when, and at what level, has the U.S. unemployment rate held? How unusual is a stretch that long?

Hint: The figure 4.5% and the 1960s are your clues.

Q3. How did oil and gas employment change from around 2013 to this summer? Over that same period, what happened to profits and output per person?

Hint: Recall the starting point of 200,000 and the endpoint of 115,000.

Q4. What events does the source article cite as reasons the productivity numbers can look lower in the short term?

Hint: Output is calculated after removing price changes. Two events pushed prices up, last year and this year.

Q5. What analysis does the source article introduce about the relationship between productivity growth and workers' wage growth?

Hint: The "double" figure in Jared Bernstein's analysis is the key point.

---

Related: see How AI Adoption Helps Philippine SMEs Stay Competitive in 2026.

Part 2: Putting It Into Practice

Step 4: Adoption Steps in the Philippines (10 min)

The core of the source article is that what produces results is not AI itself, but the digitization of work and rethinking how people are deployed. Applying that order to a Philippine operation gives you the following approach.

A few notes for each stage.

At Stage 1, it's important not to be greedy with your success metrics. Too many metrics wear the team out, and producing numbers becomes the goal itself.

Stage 3 is where the restaurant-chain example from the source article is instructive. The unglamorous improvement of managing inventory in the cloud actually led to results.

At Stage 4, do not start on the premise of cutting people. The source article reports that many companies delayed hiring on the grounds of AI-driven efficiency, but in the Philippines the procedures for workforce reductions are set out in detail by the Department of Labor and Employment (DOLE). Getting the order wrong can create legal problems.

Step 5: Common Mistakes and How to Avoid Them (5 min)

Failure pattern 1: Believing "if we add AI, productivity will rise"

Bad example: Head office told us to "raise productivity with AI," so we first signed up every employee for a paid AI service. Three months on, the number of cases handled had barely changed.

Good example: First, write the workflow out on paper and pick two steps with a lot of rework. Then trial AI on just those steps. Expand the scope only after confirming results. As the source article explains, what has pushed productivity up is the digitization of work and rethinking staffing; AI is an ingredient that will take effect from here.

Failure pattern 2: Immediately turning efficiency gains into layoffs

Bad example: AI cut working time by 20%, so we decided to stop renewing contract staff. Distrust spread among local staff, and our best people started leaving.

Good example: Put the freed-up time toward quality checks or new work you couldn't get to before. If a staffing review does become necessary, follow the procedures set out by the Department of Labor and Employment (DOLE) and explain things fully in advance. As the source article notes, when efficiency gains keep failing to reach workers' wages, resentment builds inside the organization.

Failure pattern 3: Entering customer information into an outside AI service without checking

Bad example: To speed up the work, we pasted a list containing customers' names and contact details straight into a free AI service and had it summarized.

Good example: Before handling customer information, make a list of what may and may not be entered. Check the terms of your contract, set the service so that what you enter is not used for training, and only then use it. Put your internal criteria in writing, measured against the rules of the National Privacy Commission (NPC).

---

Part 3: Going Deeper

Step 6: Related Technical Terms (5 min)

Labor productivity is a figure showing how much output a worker produces in one hour. In the source article, it is calculated by dividing output by hours worked, with the output side found after removing price changes. When you measure "how many inquiries one person handled per hour" at a Manila call center, you are using exactly this idea.

Generative AI is AI that creates things such as text and images on its own. The source article notes that, as of March, the then-chair of the U.S. Federal Reserve said its effects had not even begun to appear. At a Philippine operation, a familiar use is drafting replies to inquiry emails that mix English and Tagalog.

Machine learning is a mechanism by which a computer finds rules in large amounts of data and becomes able to predict what will happen next. The source article lists it as one of the elements that supported the productivity gains since COVID-19. In Philippine retail, it is used to forecast demand store by store from past sales data and set order quantities for inventory.

Cloud computing is a way of working that borrows another company's equipment over the internet instead of keeping servers in-house. The source article introduces the example of a restaurant chain using it to manage inventory and raise efficiency. In the Philippines, where typhoons and power outages are common, the fact that your data stays safe even when the office machines go down is a major practical advantage.

Labor share is the portion of the earnings generated across the whole economy that goes to workers' pay. The source article introduces the analysis that productivity growth over the past decade was double the inflation-adjusted growth in wages. It serves as a yardstick when you consider how much of the gains from AI-driven efficiency at a Philippine operation to return to local staff's compensation.

Step 7: Thinking About How to Apply This to Your Company (10 min)

Is the productivity gain really thanks to AI?

Something to think about: List the tasks at your Philippine operation where the number of cases handled or the turnaround time improved in the past year or two. Was that improvement due to adopting AI? Or was it due to rethinking the workflow, swapping out who does the work, or digitizing documents? Check whether it has the same structure as the U.S. examples in the source article.

Whose share are the hours freed up by efficiency?

Something to think about: The source article points out that productivity growth has far outpaced wage growth. If working time at your operation dropped by 20%, what would you put that 20% toward? Discuss which choice — cutting headcount, taking on new work, or improving compensation — is effective for retaining local talent.

Where is the digitization you should tackle before AI?

Something to think about: Count how many paper documents, personal files kept by individuals, and procedures passed on by word of mouth remain in your operation. As with the restaurant chain and medical records in the source article, there are plenty of cases where unglamorous digitization takes effect first.

Next action: This week, pick three of your Philippine operation's main tasks and record the "handling time per case" for each for just one week. That number becomes the starting point for judging the effect of anything you adopt from here.

---

Part 4: FAQ

Q1. Head office is asking me, "By what percentage can AI cut our labor costs?" How should I answer?

First, share the point that even the latest U.S. data describes AI as not yet the center of productivity growth. The source article cites the digitization of work, remote work, and a tight labor market as the main factors. Then, the realistic proposal is to do the work of rethinking your own workflows first, and expand AI from the steps where you have confirmed an effect. Not promising a number up front is what, in the end, builds head office's trust.

Q2. Labor costs in the Philippines are lower than in Japan, so isn't there little point in adding AI?

Judging by the level of labor costs alone leads you astray. In Manila and Cebu, competition for experienced talent is fierce and turnover happens easily. When AI or better workflows lighten the work, you can put the remaining time toward quality checks and new tasks. The source article, too, explains that a hard-to-hire environment made companies look for efficiency. The same dynamic works in the Philippine talent market.

Q3. If we reduce staff through AI adoption, what should we watch out for in the Philippines?

Be aware that the formality of the procedure is valued even more than in Japan. The Department of Labor and Employment (DOLE) sets out procedures that include advance notice and written explanation. Handling it verbally invites disputes later. Also, because Philippine workplaces have a culture of avoiding direct disagreement with a superior, staff may appear satisfied on the surface while resentment remains. Always make time to explain carefully in small-group meetings.

Q4. I'm worried local staff will enter customer information into AI.

Put on a single sheet of paper what may and may not be entered, and hand it to everyone. As a rule, decide that customer names, contact details, and government-issued numbers are not entered into outside services. Check the scope of information covered by the rules of the National Privacy Commission (NPC), and keep it in writing as an internal standard. Since telling people only what is forbidden drives them to start using it in secret, it's important to also show approved ways to do the work.

Q5. Is there any point in measuring productivity even at a small operation?

Yes. If anything, changes in the numbers show up faster at a small operation. As the source article notes, productivity figures wobble in the short term and are affected by price movements. That is exactly why you should keep the metric you use internally simple — something the team accepts, like "handling time per case." Keep recording for about three months, and you'll be able to judge for yourselves whether an improvement is real.

---

Tips for Making the Most of This (3 Tips)

Start recording the "handling time per case" for your main tasks this week. As the source article shows, productivity data is hard to interpret and wobbles in the short term. If you get swept along by outside headlines without numbers of your own, you'll misjudge. Three tasks over one week is plenty to start.

Before AI, move the work that runs on paper and word of mouth to digital. What pushed productivity up in the United States were unglamorous improvements like a restaurant chain's inventory management and the digitization of medical records. Your Philippine operation surely has the same kind of room to grow.

Decide how you'll use the time freed up by efficiency before you adopt anything. The source article points out that productivity growth has far outpaced wage growth. Deciding in advance what the freed-up time goes toward — and telling local staff — reduces anxiety and makes it easier to win their cooperation.

---

Bonus: How to Use PH AI Works

PH AI Works supports the adoption of AI and technology in the Philippines. For Japanese companies with Philippine operations, we work alongside you — from organizing your workflows, to designing where in the process to place AI, to explaining it to local staff.

As a next step, you can consult us on things like the following.

• Organizing your Philippine operation's workflows and identifying candidates for digitization to tackle before AI
• Building internal criteria for entering customer information into AI services, and preparing explanatory materials for local staff
• An AI adoption plan for trialing small, and designing the metrics to confirm its effect

Feel free to get in touch. Consultations are free.

---

References & Sources

• The New York Times — "U.S. Workers Are More Productive Than Ever. And That's Without A.I." (July 14, 2026)
• National Privacy Commission (NPC)
• Department of Labor and Employment (DOLE)