For six decades, I’ve worked on product design and manufacturing for companies such as Polaroid, Seiko, and Apple, as well as startups. When I worked at Polaroid and Apple, they still had their own factories and chose to go to Asia to build products. As noted in Patrick McGee’s bestseller Apple in China, I was responsible for bringing Apple’s first product to Taiwan in 1994. As a product design engineer and program manager, I was always involved in where my products were manufactured, because of the impact manufacturing has on a product’s success.
In my jobs at Polaroid and Apple, I turned to Asia for different reasons. I brought manufacturing to Asia for Polaroid in the 1970s, when it was rarely done. By the time I did it at Apple in the 1990s, Apple was late to the game.
Today, as pressure mounts for Apple to build iPhones in the U.S., I know from all my experiences that it is an impossible demand. We don’t have the attitude, the infrastructure, or the resources. We lost that battle decades ago, and cannot will it back.
The U.S. never invested in building consumer tech manufacturing at the scale of China or Taiwan. We prioritized R&D, design, and innovation, and didn’t value low-margin manufacturing work. During the Cold War, U.S. government investment and incentives were directed toward aerospace and defense, not consumer electronics. Our manufacturing base became dominated by contractors focused on military specifications, not consumer mass production.
Manufacturing consumer electronic products, in contrast, requires an infrastructure of suppliers, competitive labor costs, large workforces, and flexibility. Resources in China and Taiwan have allowed many U.S. companies to dominate the world in technology. While China became the manufacturer to the world, it helped us become the innovators to the world.
Going offshore: the early days
When I joined Polaroid in 1966, I designed consumer cameras that sold millions per year. Once the design was done, we’d hand it off to our internal manufacturing group, only to have them come back to us months later with requests for small design tweaks to save a few cents here and there.
It was a slow, iterative process with little concern for time to market. I often wondered why I was working so hard to meet an aggressive design schedule and then had to wait around for the product to finally ship. Analysts and industry leaders emphasized the importance of time to market as a critical factor for a product’s success, especially in competitive and fast-evolving markets.
I wondered why we couldn’t be as creative in the rest of the process as we were in the design stages. So, I began to look for ways to get products to market more quickly.
Outsourcing manufacturing to companies building similar products seemed like a good approach. They already had the experience and knowledge, so they could move quickly. That allowed the manufacturing and design teams to collaborate earlier in the process.
It wasn’t that I didn’t try to do manufacturing in the United States. I reached out to companies with manufacturing capabilities that looked to be synergistic, but it often took weeks to hear back. Many of the companies were focused on defense-related products and showed little interest in building consumer tech products with tighter schedules and lower margins.
I wondered why this was the case. Could it be that these U.S. resources were focused on the defense industry during the Cold War and never invested in consumer manufacturing? Or perhaps the defense business was just more lucrative. It was disappointing, but, fortunately, I found another solution.
At Polaroid, I turned to a few companies in Japan that had been supplying some of our components and wanted a chance to build the entire product. They offered to take our design early in the process, improve it for manufacturing, and deliver the product in high volume much more quickly than we could on our own. We were being offered the opportunity to tap into their product expertise and quality-obsessed manufacturing that Japanese companies were known for, and to learn from them in the process.
It turned out that these companies were so good that we could almost drop off a drawing package and pick up the product months later. They brought speed, experience, and predictability, and enabled us to be much more effective designers with smaller teams. Every time I’d return from a visit to Japan, I’d feel a sense of accomplishment and excitement. It was like night and day compared with using our internal manufacturing organization.
This pattern repeated itself over the next few decades as I worked at other companies. When Japan became too costly and less efficient, I went to Taiwan and later China, as each of these countries developed their manufacturing capabilities.
The U.S. could have created similar capacity over the years, especially after the Cold War, but never did. Manufacturing was just never considered an important need. It was not glamorous and never considered to be strategic.
In the 1990s, when a company I cofounded, ThinkOutside, invented the Stowaway folding keyboard, we found a Taiwan keyboard company that was experienced with manufacturing tiny key mechanisms, and partnered with them. Like many other companies working to get products to market quickly, we were doing what we did best and letting others do what they did best.
At this time, there were a few contract manufacturers in the U.S., including Solectron and Flextronics, that could manufacture others’ products in their plants in the U.S. and elsewhere in the world. They started with printed circuit boards and had expanded into complete product manufacturing; they also had some design capabilities.
This evolution mirrored the tech industry’s growing reliance on outsourced manufacturing, driven by cost, scale, and time-to-market pressures. But I found that the U.S.-based options were bureaucratic lumbering giants, reminding me of my Polaroid manufacturing experiences. They wanted only to take on large-volume projects from well-established companies, and rarely took the kind of risks that Asian companies did.
Asia becomes the world’s manufacturer
Meanwhile, Taiwan and China were becoming destinations for companies to build technically complex products because they were so good at it. They were aided by huge government investments in office parks, infrastructure, and free-trade zones.
Taiwan quickly moved from building calculators in the 1970s to notebook computers in the 1980s. James Fallows chronicled the rise of China in his famous three-part 2007 article in The Atlantic, “China Makes, The World Takes,” as China became the manufacturer for the world.
Scores of Chinese companies became available to manufacture a customer’s original design. They were known as original equipment manufacturers, or OEMs. If they offered their own designs that a customer could sell under its name, they were called original design manufacturers, or ODMs.
It was a win-win partnership. We used China to build our products, and they relied on us keep their factories running by marketing and selling products they designed.
Companies often specialized in specific product categories. If you wanted to build a particular type of product, you’d go to a company that was already building something similar. You tapped into their expertise, supplier relationships, and buying power, and everyone would benefit from the economies of scale and relevant experience.
While some first-time customers worried about two competing products being made by the same OEM, each company’s areas were usually secured from the other. When I worked on the Barnes & Noble Nook e-reader, we used Foxconn, the same manufacturer that built the Amazon Kindle. That allowed us to go from an initial concept to market in less than a year.
Product designers large and small now had a destination where they could bring their product designs and get them manufactured quickly and efficiently. You no longer needed to be a Sony to have access to manufacturers. Chinese factories enabled thousands of companies across the United States and the world to turn their ideas into viable products.
As these OEMs and ODMs grew in number, component manufacturers and other suppliers opened their own factories nearby, close to the facilities using their parts, providing even greater efficiency and faster time to market.
As an example of how effective this was, I was about to ship the initial lot of a new consumer electronic product to the U.S. from Shenzhen, the Chinese technology hub adjacent to Hong Kong, and it was to arrive just before going on sale. At the last moment, I found an error in the instruction manual that needed to be fixed before the product could ship. We called the printer, which was a few blocks away. It made the correction, and delivered correctly printed manuals two hours later.
During this time, we never considered that we were taking jobs from Americans. Nothing like this ever existed in the U.S.—not the industry, the factories, nor the workers. What China offered was unique.
Apple was late to the game
When I joined Apple in 1994 to work on the Newton MessagePad personal digital assistant, the company was manufacturing most of its computers in its own factories in California, Colorado, Ireland, and Singapore, even though other computer companies had been outsourcing most of their notebooks to Taiwan for years.
The first version of Newton, the MessagePad 100, was developed with Sharp Electronics in Japan because of that company’s expertise in building handheld products such as its Wizard organizer. But its cost came in way above plan, and I was asked to quickly develop a new model at a more affordable cost.
Apple’s internal manufacturing organization had neither the interest nor experience to master products like the Newton. Its leaders considered our product to be a nuisance and encouraged us to find our own manufacturer.
Taiwan was the obvious answer. After all, that’s where Sharp was now building its second-generation Wizards, and where most of the world’s calculators and notebook computers were being built, other than Apple’s. The new Newton MessagePad 110 was the first time Apple developed a product to be manufactured in Taiwan.
I flew to Taipei and spent two weeks visiting potential manufacturers, mostly notebook and calculator manufacturers. We selected Inventec, a $320 million manufacturer at the time. It was busy producing calculators for Texas Instruments, phones and fax machines for telecom companies, and notebook computers for Compaq and Zenith.
But we ran into an immediate challenge. Inventec would need an $800,000 machine to assemble the processor chip powering the Newton, machinery not yet used in Taiwan. The company’s CEO, Richard Lee, insisted on buying it at Inventec’s expense. He never considered it a burden, explaining it would give them new capabilities.
Inventec assigned a group that began working with our Apple design team. We began with a handshake deal while Lee suggested we negotiate a contract in parallel. We all knew drawing up contracts could take many months, particularly at Apple, and he didn’t want to lose that time.
Our joint team of about a half-dozen engineers met monthly in Taipei. The industrial designer on the team, Jony Ive, had recently joined Apple. This would be one of his first projects, and his first time he’d been to Taiwan. (I later learned that he complained to his boss about how hard I made him work.)
The product was completed in eight months and went on to fail almost as badly as the first model. Sales volumes were about 10% of plan. A few months later when I apologized to Lee, he told me not be concerned. He was happy to be working with Apple and considered this to be a win. He was thinking of the long term, and was proud of the Apple relationship. I was embarrassed.
Inventec went on to develop and build notebook computers, servers, and a range of other products for most of the world’s technology companies, including Apple. Its revenues for 2025 were $22 billion.
How design impacts manufacturing
The Newton was constructed like most other consumer tech products at the time. It consisted of a molded plastic clamshell enclosure, a large printed circuit board populated with electrical components, a pressure sensitive touch screen, a stylus, and a battery compartment that held four AA cells. Ive’s design contribution was creating a beautiful fluid-looking package that made the product look smaller than it was. It was nicely done, but its fit and finish were not dissimilar from other products being produced at the time, such as Nokia phones.
Years later, when Ive became Apple’s senior VP of industrial design, he elevated its design to a level never seen in consumer products. I recall seeing the first iPhone and marveling at how precisely the parts fit together, the flawless finishes of the parts, and the use of new materials never employed at mass scale. His designs set a new standard for all consumer tech products. Other products suddenly seemed crude by comparison.
Apple’s design and manufacturing engineers were responsible for putting these new designs into mass production. It certainly could not be done in the U.S., which didn’t even have the ability to build much simpler products. The new design standards with the ultraprecise fit and finish meant that the products needed to be built China.
When I hear politicians demanding that iPhones be built in the U.S., it’s clear they have no understanding of how products are created and built. You can’t just wish it to happen. You would need to create much of what China and Taiwan have built over the past 30 years.
It’s not just setting up factory buildings filled with assembly machinery and workers. You would need the entire infrastructure of suppliers and manufacturers of the parts that go into the products: the factories to build displays, batteries, and electronic components. You’d require the experience, and the right attitude.
Shipping the components from China and doing final assembly in the U.S. would also cause huge issues. Let’s suppose you bring displays in from China, and they have a defect. If you were building the product in Shenzhen, you’d call the display company that’s an hour away and they would have engineers in your factory in a few hours to address the problem or provide new parts. Try that from the United States, and you would likely shut down your assembly line for several weeks.
Now consider what would happen once you multiply this by the thousands of parts and processes that go into a product. Not only is it inefficient, it’s also just not practical to build these products here. It makes as much sense as building an automobile in the Arctic from parts made in Detroit.
The Mac Mini isn’t an iPhone
With all this said, Apple, under pressure to bring products back to the U.S., has agreed to build Mac minis in Houston. They have asked Foxconn, the same company that has built, staffed, and managed much of Apple’s China manufacturing, to create and operate the assembly operation.
The Mac mini is a much simpler and lower volume product than the iPhone. Essentially, it’s a printed circuit board assembly housed in an aluminum box. But it will still require many of its components to be shipped from China. Politically, it may be a good move, but practically, it means very little.
It will be interesting to see if Apple’s new CEO, John Ternus, makes any fundamental changes to Tim Cook’s manufacturing strategy. As a product designer, I’m guessing that Ternus—who is also a product designer—wants to get his new products to market in the fastest and most efficient way possible and will continue to rely on Asia. That approach has helped make so many of his past products huge successes.
