A flight with Wright's Law
Perfecting the art of Bengali doodh cha on an induction oven is, for a task so seemingly humble, a surprisingly herculean feat. My first attempts were an insult to my heritage. The delicate alchemy of turning Amul milk powder and water into a creamy base often resulted in a lumpy, gritty liquid. Taming the fierce, binary nature of the induction cooktop felt impossible; the concoction would either sit there stubbornly inert or boil over in a furious, sticky wave. Add loose-leaf tea and a dash of monk fruit sweetener to this chaotic process, and the result was frequently a sad, beige tragedy, destined for the sink.
Fast forward a hundred cups or so. The process is now a fluid, almost meditative morning ritual. The powder whisks into a perfect vortex, the induction hob hums at the precise temperature, and the tea infuses to a rich, terracotta hue. It’s faster, wastes far less of my precious ingredients, and produces a cup of cha that is a genuine comfort. I had, without knowing it, stumbled upon a small-scale, domestic demonstration of a rather profound economic principle: Wright’s Law.
Long before baristas were fussing over blooming times, an aeronautical engineer named Theodore Paul Wright was watching men build airplanes. This was 1936, a time when aviation was shedding its skin of wood and fabric for the sleek promise of aluminum and rivets. And no, despite the auspicious surname and his chosen field, T.P. Wright was no relation to Orville and Wilbur, though he was very much a titan in the industry they had pioneered. It was a laborious, sprawling, and fantastically expensive enterprise. But amidst the cacophony of the factory floor, Wright discerned a subtle, rhythmic pattern. He observed that for every cumulative doubling of aircraft produced, the number of labor hours required to build the next one decreased by a consistent percentage, typically around 10-15%.
This wasn’t a forecast or a theoretical model; it was a brutally empirical observation. It became known as Wright’s Law, or more broadly, the experience curve. The mathematical formulation is almost elegant in its simplicity. If is the cost of producing the -th unit, the law states that:
Here, is the cost of the very first unit, is the cumulative number of units produced, and is the “learning factor”—the percentage of the cost that remains after production doubles. For a 15% cost reduction, would be 0.85. The upshot is that costs don’t fall with the passage of time, but with the accumulation of experience.
This is a fundamentally different beast from its more famous cousin, Moore’s Law. Gordon Moore’s prediction was about the physics of miniaturization, a relentless two-year cadence of cramming more transistors onto silicon. It’s a law of technology, dictated by the cleverness of engineers pushing against the frontiers of the very small. Wright’s Law is something else entirely. It’s a law about us. It describes how people, and the organizations they form, get better at doing things. The “how” is almost irrelevant to the law’s predictive power. The improvement could stem from a worker’s muscle memory on an assembly line, a clever new jig designed by a foreman, a more efficient supply chain, or the slow, accretionary wisdom of an entire industry. It simply states that we learn by doing.
And whence has this law taken us? Far beyond the airplane hangars of the 1930s. Look at the technologies shaping our century. The price of solar photovoltaic cells has plummeted in a way that seems almost miraculous, not just because of breakthroughs in labs, but because we have been manufacturing them at a truly colossal scale. Each doubling of installed capacity has driven down the cost, turning solar from a niche curiosity into a global energy powerhouse. The same story can be told for the lithium-ion batteries that power our phones and electric cars, and for the powerful algorithms that are learning to fold proteins.
Herein lies the profound “wherefore” of Wright’s Law: it provides a potent argument for action. If cost is a direct function of cumulative production, then policies that spur early and widespread adoption—like public subsidies for renewable energy or electric vehicles—aren’t just handouts. They are investments in accelerating the journey down the experience curve for everyone. By buying our way to scale, we are effectively buying our way to a cheaper, more accessible future.
Yet, it would be a mistake to view Wright’s Law as some immutable law of nature, a guarantee of perpetual, effortless progress. Like all observations about the messy world of human affairs, it comes with a litany of caveats. The learning curve, for one, eventually flattens. My tea-making technique is now quite good, but I’m unlikely to halve the time it takes again without resorting to a fully automated machine, which is a different thing entirely.
Furthermore, the law only governs the parts of a process that can be learned. It has nothing to say about the stubborn, non-negotiable costs of raw materials. The price of lithium or cobalt does not bend to the will of our manufacturing experience. And there’s a more insidious danger: the temptation to fit a curve to past data retrospectively and present it as a prophetic vision of the future, ignoring potential bottlenecks like regulatory hurdles, land availability, or a shortage of skilled labor.
In the end, Wright’s Law is less a law and more a piece of institutional memory, writ large in economic data. It’s the universe’s way of telling us that practice, if not making perfect, certainly makes things cheaper and more efficient. It is the unseen hand that guides our own, a quiet hum in the background of progress, connecting my fumbling attempts to brew a decent cup of cha to the grand, industrial ballet that builds our modern world. It’s the simple, powerful truth that if you do something enough times, you just might get the hang of it. And when a whole civilization gets the hang of something, it can change everything.
And by the by, just so that prude nationalist Indians can cut out the Chinese hate talk, the Hindi word for tea — “chai” (चाय) — comes from Mandarin Chinese, specifically from Mandarin “chá” (茶), just like Bengali “cha” (চা). Anyway toodles—I have an urge to make another cup.