This is an independent, source-linked historical case study. It is not authored, endorsed or approved by Warren Buffett or Berkshire Hathaway. The original documents remain with their respective rights holders. Not investment advice.
Periods: 1951–1952, 1976–1980, 1995–1996. Primary record: Warren Buffett's GEICO history in Berkshire's 1995 shareholder letter.
Why the earliest research mattered
Buffett describes enrolling at Columbia to study with Benjamin Graham in 1950–51. He learned that Graham chaired Government Employees Insurance Company, visited its headquarters, and pursued substantial research into its operations. He subsequently acquired GEICO stock before selling his early position in 1952.
His retrospective illustrates an important behavioral error: he had identified an unusually promising enterprise but sold to pursue another apparently cheaper security. He later used GEICO as an example of the opportunity cost of abandoning a high-quality investment too early.
The 1970s underwriting crisis
GEICO later suffered from errors in claims-cost estimation and the pricing of insurance policies. According to Buffett, the business nearly failed; management changes under Jack Byrne in 1976 helped restore a path to survival. Berkshire bought a large stake in the second half of 1976. By the end of 1980 it had invested $45.7 million and held 33.3% of the stock. GEICO's repurchases later expanded Berkshire's ownership percentage without new share purchases by Berkshire.
There was no riskless turnaround. Insurance requires continual solvency, pricing, reserves and disciplined use of float. Favorable economics cannot offset unsound promises to policyholders.
The 1995 decision to own everything
In 1995 Berkshire agreed to pay about $2.3 billion for the half it did not already control, closing after the reporting year. Buffett acknowledged that the price was steep. His stated case centered on the company's cost position, growth, and exceptionally capable managers.
Testable mental model
The sequence joins accumulated domain knowledge, long-term franchise economics, serious near-term distress, managerial recovery, and a later controlled acquisition. It is not equivalent to buying any insurer after a stock-price decline. To study similar opportunities, evaluate underwriting and reserve adequacy, low-cost distribution, the source and cost of float, policyholder persistency and management integrity.
What the source does not establish: An independently audited full chronology of every statutory reserve estimate; contemporary documentation would be needed.
Original source: Berkshire 1995 shareholder letter, GEICO Corporation.