Marcus Weldon ran Bell Labs and served as Nokia’s chief technology officer until 2021, so when he loses his temper about Bell Labs in public, the industry should pay attention.
In a LinkedIn post this week, he estimated that Nokia has cut roughly half of the research jobs at Bell Labs since he left, from more than 1,200 people to perhaps 600, and called the contraction shocking and unprecedented. Nokia replied that Bell Labs remains a deeply important part of the company and is entering a new chapter under Guru Parulkar, with its research teams unified around breakthrough science for the AI era. Both statements can be true at the same time. Bell Labs is smaller, and Nokia is reorganizing what remains.
I share Weldon’s sadness about what has been lost. But I don’t think Nokia is the problem. Nokia picked up Bell Labs a decade ago in its €15.6 billion takeover of Alcatel-Lucent, and it inherited an institution whose economic engine had disappeared decades earlier. Expecting a company with €19.9 billion in annual sales to recreate the Bell Labs of the AT&T monopoly ignores the financial logic of today´s telecom, but it surely exposes the biggest weakness in our industry.
It is worth remembering our glorious past. Bell Labs invented the transistor in 1947 because vacuum tubes were becoming a maintenance problem for the telephone network, and in trying to fix telephony it accidentally launched the semiconductor industry. Claude Shannon wrote information theory there. Its scientists made foundational contributions to the laser, built Telstar for its 1962 launch, developed the CCD image sensor, and wrote Unix and the C programming language.
Decades later, Yann LeCun’s work at AT&T Bell Labs showed that convolutional neural networks could read handwritten digits on real cheques, a direct ancestor of modern AI. Work associated with Bell Labs has earned ten Nobel Prizes and five Turing Awards. In the 1960s, NASA created Bellcomm to borrow systems engineering talent from AT&T, Bell Labs, and Western Electric for the Apollo program.
Please don’t get me wrong here. AT&T’s engineers were no smarter than today’s, but the difference was the economic machine underneath them.
The Bell System controlled most of the American communications value chain, from research and manufacturing to the network and the customer, and a small slice of every telephone dollar flowed continuously into research. In 1974, AT&T booked $26 billion in revenue, about 1.4% of US GDP, and Bell Labs alone spent more than $500 million on nonmilitary R&D, roughly 2% of that revenue, while Western Electric spent even more on engineering and development. More than four cents of every AT&T dollar went into R&D that year.
The stability of the money and its vast amount meant that a researcher knew the budget would still exist in fifteen years and could work on problems that would not produce a product for fifteen years.
Telcos Outsourced the R&D of a $1.3 Trillion Industry
Telecom deliberately dismantled that structure, for the most part, for good reasons.
The 1984 breakup of the Bell System, liberalization in Europe and Asia, and the arrival of competition pushed operators to concentrate on customers, spectrum, deployment and operations, while Ericsson, Nokia, Huawei, Qualcomm, ZTE, Samsung and a long tail of smaller suppliers took over most of the underlying technology development.