New on FinoLib · Volume I: Global Economy
I have just published a new article on FinoLib, my free finance education library: Economic Impact of AI and the Geography of Innovation. It is part of Volume I: Global Economy. Below is a summary of the topic in plain, expert language, so you can follow the idea even before you open the full article.
7.8 The Long View — AI’s Economic Impact Over the Next Decade
The idea in plain words
AI numbers come in four kinds: observed data, estimates, scenarios and vendor forecasts, and only the first is a measurement. Forecasts of AI’s payoff differ several-fold.
Key points
■Goldman Sachs Research estimated in April 2023 that generative AI could raise global GDP by 7% (almost $7 trillion) and lift productivity growth by 1.5 percentage points over ten years.
■McKinsey put the potential value of generative AI at $2.6 to $4.4 trillion a year across 63 use cases it analyzed (June 2023).
■Daron Acemoglu (2024) finds a rise in total factor productivity of no more than 0.66% over ten years in total, and less than 0.53% once harder-to-learn tasks are allowed for.
A closer look
Forecasts of AI’s payoff differ several-fold, so the first skill is to know what kind of number you are reading. Goldman Sachs Research estimated in April 2023 that generative AI could raise global GDP by 7% (almost $7 trillion) and lift productivity growth by 1.5 percentage points over ten years. McKinsey put the potential value of generative AI at $2.6 to $4.4 trillion a year across 63 use cases it analyzed (June 2023). Both are scenarios: they assume how fast AI is adopted and then compute what follows. They are not measurements, and they do not agree with every economist (see the box below).
Where Economists Disagree: How Much Will AI Raise Productivity?
The high-gain view. Generative AI touches a large share of tasks in office work, so if firms adopt it widely, the gains add up: Goldman’s scenario lifts productivity growth by 1.5 percentage points over a ten-year period. The low-gain view.
Recap
AI numbers come in four kinds: observed data, estimates, scenarios and vendor forecasts, and only the first is a measurement. Scenarios differ several-fold: Goldman’s 7% rise in global GDP over ten years against Acemoglu’s gain of at most 0.66% in total factor productivity (about 1% in GDP). Edge cases such as vendor financing, power limits and useful-life choices show where the standard growth story bends.
7.9 The Geography of Innovation — Silicon Valley, Bengaluru, and the World’s Technology Capitals
The idea in plain words
Invention clusters in about a dozen technology cities because talent, capital and ideas compound when they share the same neighborhoods, an effect economists call agglomeration. Tech cities make the future; the financial cities price and fund it.
Key points
■Silicon Valley’s recipe combined Stanford pushing students toward industry, Cold War defense spending seeding the first electronics firms, the 1957 founding of Fairchild Semiconductor and the invention of modern venture capital on Sand Hill Road.
■Fairchild’s alumni went on to found Intel and dozens more firms, giving the valley its silicon and its name.
■Nvidia, OpenAI, Google, Meta and Apple all sit within an hour’s drive of each other, a concentration of market value with no historical precedent.
A closer look
Software travels at light speed, so innovation should theoretically happen anywhere. In practice it happens in about a dozen postcodes, because of what economists call the agglomeration effect: engineers switch jobs across the street carrying ideas with them, investors can meet five startups before lunch, universities feed talent in, one generation’s successful founders become the next generation’s angel investors (chapter 4.7: Venture Capital and Private Equity — The Pre-IPO World), and failure carries no stigma because the ecosystem instantly reabsorbs you. Each ingredient strengthens the others — which is why clusters, once formed, are extraordinarily hard to dislodge, and why governments worldwide keep trying (and mostly failing) to build “the next Silicon Valley” by decree.
A worked example
Trace the Chain — Part 7 Thought Experiments
1. Trace one rupee of a hyperscaler’s AI capex through the three layers of the AI economy — which companies capture it at each layer, and where does the margin concentrate? 2. A single Taiwanese company fabricates most of the world’s leading-edge chips. Connect this fact to three other chapters in this guide (hint: supply chains, geopolitics, sector concentration). 3. If AI automates 40% of junior reconciliation tasks, what happens to the career ladder that produces senior reconciliation experts — and what should firms do about it? 4. Why has every government-decreed “next Silicon Valley” mostly failed, while Bengaluru — never centrally planned as one — succeeded?
Recap
Invention clusters in about a dozen technology cities because talent, capital and ideas compound when they share neighborhoods, an effect called agglomeration. Silicon Valley grew from Stanford, Cold War defense spending, Fairchild Semiconductor and modern venture capital.
What the full article covers
In the full article I take you through these sections:
■7.8 The Long View — AI’s Economic Impact Over the Next Decade
■7.9 The Geography of Innovation — Silicon Valley, Bengaluru, and the World’s Technology Capitals
It includes worked examples, a recap, a reveal-on-click quiz and the official sources behind every figure.
Volume I: Global Economy
Read the full article on FinoLib
The complete article, Economic Impact of AI and the Geography of Innovation, with every worked example, diagram, quiz and source, is free to read in Volume I, Global Economy on finolib.com.
Explore the Finance Education Series: Vol. II — Capital Markets · Vol. I — Global Economy · Vol. III — Finance Encyclopedia · Vol. IV — Finance Careers · Vol. V — Personal Finance