Also called: The learning curve or the cumulate average model or experience curve effects Pioneered by Theodore Wright in 1936. While studying production costs during the 1920s, Theodore Wright determined that for every cumulative doubling in the number of airplanes produced, manufacturers realized a consistent cost decline in percentage terms. For example, the cost to produce the 2,000th plane was 15% less than that to produce the 1,000th plane, and the cost to produce the 4,000th plane, 15% less than that to produce the 2,000th. > Wright’s Law aims to provide a reliable framework for **forecasting cost declines as a function of cumulative production ** > Can be used to forecast cost declines associated w/ any technology > Specifically, it states that for **every cumulative doubling of units produced, costs will fall by a constant percentage.** > Cost decreases as a power law of cumulative production ![[Pasted image 20210919201458.png]] ![[Pasted image 20210919201721.png]] ### Takeaways - We learn by doing - The cost of each unit produced decreases as a function of the cumulative number of units produced - [[Moore's Law]] forecasts costs as a function time vs Wright's Law which forecasts costs as a function of units. - Wright’s Law must be evaluated according to a metric that incorporates improvements in performance - It does not imply that costs go to zero as: cumulatively doubling the units produced is exponentially hard and each doubling takes more time as the no. of units produced increases.--> Line terminates at a fixed point. - A mature technology **appears to stop declining in price **because it takes so long to double production off of a large cumulative base. ### Example for [[Batteries 101]] ![[Pasted image 20210919202727.png]] ### Example for [[Electric Vehicles]] Since it began taking shape in the early 1900’s, the auto industry has enjoyed an** 85% learning curve** – which has translated into a** 15% cost decline with every cumulative doubling of units produced.** Ford’s Model T conformed exceedingly well to Wright’s Law from 1909 to 1923, as shown below. ![[Pasted image 20210919203438.png]] ![[Pasted image 20210919203451.png]] ![[Pasted image 20210919203745.png]] ### EV vs ICE: Wright's Law effect - Globally, automakers have manufactured only 3.4 million electric vehicles (EVs) over time, less than 1% of the 2.5 billion internal combustion engine (ICE) vehicles ever made - As a result, even if the production of ICE vehicles were to stabilize at ~90 million vehicles annually, its** cumulative doubling would take 29 years**, compared to roughly one year for EVs. - Wright’s Law applies equally to both technologies but the **difference in time to achieve a cumulative doubling explains the difference in annual cost declines**: the cost of the average **ICE vehicle will edge down by ~0.5%** in the coming year while that of its **EV counterpart costs will drop by ~12%**. ![[Pasted image 20210919204003.png]] ### Shortcomings 1. Requires long history of data for all production to date. 2. Involves anticipated demand: requires an understanding of the price elasticity of demand, or how much demand will respond to future price declines - Learning Rate: How much costs will fall with every doubling of production - Doubling Rate: Rate at which production would double cumulatively - Watch out for demand changing discontinuously as costs + capabilities fall at tipping points that trigger mass market adoptions - When a seemingly mature technology crosses critical demand thresholds, the decline in its cost trajectory will re-accelerate. Source: - https://ark-invest.com/wrights-law/ - https://ark-invest.com/articles/analyst-research/wrights-law-predicts-teslas-gross-margin/ #kp