Week 24 Day 5: Narrative vs Numbers: Trust the Data

WeWork had an incredible story: revolutionize how the world works. It was valued at $47 billion. The spreadsheet said it burned cash faster than it earned it. It went bankrupt. Theranos had a legendary founder and a transformative narrative. The numbers said the technology did not work. The company was fraud. Always check the math, even when (especially when) the story is incredible.

Examples of narrative overriding numbers: Dot-com stocks (2000): 'The internet changes everything!' True, but not every company was worth $10 billion with no revenue. Pets.com, Webvan, and hundreds of others went to zero. Housing (2007): 'Real estate never goes down.' It did. Nationally. For the first time ever. Because the math of subprime mortgages was unsustainable. Tesla (2020-2021): 'Elon Musk will save the world.' Perhaps, but at $1.2 trillion market cap, Tesla was priced at $15 million per car sold -- a number that only works if Tesla becomes a dominant AI or energy company. The stock fell 65%. SPACs (2021): 'Blank check companies democratize investing!' The average SPAC lost 40% within 2 years of merger. NFTs (2021-2022): 'Digital ownership is the future!' $41 billion in NFT sales in 2022. By 2024, 95% of NFTs were estimated to be worthless. The antidote: whenever someone pitches an investment, ask two questions. (1) 'What are the expected cash flows, and at what discount rate are they priced?' (2) 'What has to go RIGHT for this to be a good investment at the current price?' If the answer requires heroic assumptions, the narrative is doing the work, not the numbers.

The tension between narrative and numbers in investment decision-making is studied in behavioral finance under the 'narrative fallacy' (Taleb, 2007) and the representativeness heuristic (Tversky and Kahneman, 1974). Shiller (2019) devoted an entire book ('Narrative Economics') to the role of popular stories in driving economic events, arguing that narratives go 'viral' and drive market outcomes independently of fundamentals. The quantitative evidence: Baker and Wurgler (2006) constructed a sentiment index and showed that when sentiment is high, stocks with high narrative appeal (non-dividend-paying, young, volatile) dramatically underperform subsequently. De Long, Shleifer, Summers, and Waldmann (1990) modeled 'noise traders' who trade on narratives rather than fundamentals and showed that they can influence prices for extended periods, creating both risk and opportunity for fundamental investors. The defense is disciplined valuation: John Burr Williams (1938) established that the value of any asset is the present value of its future cash flows. This framework forces narrative claims through a numeric filter. When the narrative says 'this company will dominate AI' but the current stock price already implies a market value greater than the entire AI market, the narrative has been fully priced. The margin of safety (Graham, 1949) provides additional protection: only invest when the price is significantly below the conservatively estimated intrinsic value, regardless of how compelling the story.