When business executives and policymakers attempt to forecast the long-term equilibrium of a market, they often rely on complex economic models. However, a recent honors thesis authored by Zhipeng Yu at Saint Mary's University highlights that the answers might lie in a mathematical framework established over a century ago.

The research explores the application of "nonnegative matrices" and stochastic processes, which are mathematical tools used to analyze systems that evolve probabilistically over time. While the thesis dives deep into complex algebraic proofs, its most striking revelation is a real-world application: accurately modeling the historical upheaval of the U.S. automotive market.

The Math Behind Brand Loyalty

To understand the market dynamics, the research utilizes a concept known as a "Markov chain.". In simple terms, a Markov chain is a mathematical system that models a sequence of events where the probability of each event depends only on the state attained in the previous event.

In the business world, this translates perfectly to consumer behavior and brand loyalty. The author examined a "transition matrix" representing the U.S. auto market trends over a one-year period. This matrix tracked the percentage of customers who repurchased a car from the same brand (market share retained) versus those who switched from one brand to another (market share gained or lost)

The 1970s Automotive Upheaval

During the 1950s, the U.S. automotive market was highly domestic, but this shifted dramatically when Japanese automakers introduced small, modestly priced cars. Following the 1973 Arab Oil Embargo, East Asian manufacturers rapidly captured market shares due to consumer demand for fuel efficiency and affordability.

The research grouped the market into four main competitors: GM, Ford, Chrysler, and Japanese/other brands. Using historical consumer switching data, the model established a matrix of transition probabilities. For example, the data showed Ford had a 70% retention rate, while losing 10% of its customers to GM, 10% to Japanese brands, and 10% to Chrysler.

A Staggering Prediction for the Year 2000

Here is where the pure mathematics takes over. Using a foundational rule called the Perron-Frobenius Theorem (originally developed in the early 1900s) mathematicians can prove that a system like this will eventually reach a "stationary state," or a point of long-term equilibrium. This means that after enough time passes, the constant churn of customers switching brands mathematically balances out into a final, stable market share.

Starting with the actual market share data from 1970 (where GM held 43.5%, Ford 26.9%, Chrysler 14.4%, and others 15.2%), the researcher ran the mathematical model to project the future.

The mathematical formula predicted that by the year 2000, the market would stabilize at:

• GM: 28.7%

• Ford: 23.2%

• Chrysler: 15.1%

• Japanese/Others: 33.0%

When comparing this purely mathematical prediction to the actual real-world market data recorded in the year 2000, the numbers were an exact match: GM was at 28.7%, Ford at 23.2%, Chrysler at 15.1%, and Japanese/others at 33.0%.

The Takeaway for Modern Industry

While the author notes that the real-world convergence rate was slower than the mathematical model, the incredible accuracy of the final destination proves a vital point. The mathematical principles of stochastic processes possess a profound ability to explain and forecast long-term market dynamics. For today's business leaders, integrating these rigorous mathematical theorems into data modeling could be the key to visualizing the future of highly competitive industries.