Why Fundamentals Fail the New Economy | Jacob Pozharny on “Sentiment” Analysis’ Role in New Economy Stocks

Episode Summary

Executive Summary: Jacob Pajarni explains Bridgeway’s framework: traditional valuation and profitability models work well for low-intangible “old economy” stocks but break down for high-intangible “new economy” companies, where sentiment and textual analysis become more predictive. He details how the firm combines sell-side, buy-side borrow data, and earnings-call/regulatory text to find mispricings across countries, sectors, and individual stocks, emphasizing nimble, transparent, largely market-neutral stock selection.

Main Topics: Old economy vs. new economy valuation (Priority: 5/5): Pajarni argues that discounted cash flow, ROE, and price-to-book relationships remain useful for low-intangible industries but become unreliable for companies whose value is driven by intangibles like R&D, brand, or customer relationships. Bifurcated research process: fundamentals and sentiment (Priority: 5/5): Bridgeway uses a contextual stock selection process that leans on fundamentals for old-economy sectors and sentiment analysis for high-intangible sectors, with the mix varying by industry. Sentiment signals from sell-side and buy-side data (Priority: 5/5): The firm builds sentiment scores from analyst estimate leadership/followership and borrow availability/fees, using these to anticipate changes in consensus and short-interest behavior. Textual analysis of earnings calls and filings (Priority: 4/5): Bridgeway uses supervised-learning topic extraction from transcripts and filings to check assumptions and detect externalities, including corporate actions, meme-stock behavior, and region-specific shocks. Country and sector dislocations from geopolitical stress (Priority: 4/5): Charts on countries, sectors, and topic-specific text suggest mispricings around the Iran war, especially in energy, shipping, defense, and regional utilities. Market-neutral, global stock selection at scale (Priority: 4/5): The strategy is built around roughly 10,000 securities, emphasizing mid- and small-cap names outside the U.S., while avoiding large country, sector, or size bets. Risk management, holding periods, and technical analysis (Priority: 3/5): Pajarni says sentiment signals typically have a shorter horizon than fundamentals, shorts are monitored closely, and technicals are used mainly for risk management rather than as primary signal generators.

Key Arguments: Traditional valuation models underperform for high-intangible companies because earnings and book value fail to capture R&D, customer relationships, brand, and IP. For old-economy sectors, the classic link between profitability and valuation still holds and can be exploited systematically. Sell-side estimate leadership can identify where consensus EPS is headed before the crowd catches up. Borrow availability and borrow fees can reveal buy-side shorting intentions before trades appear in price. Textual analysis of earnings calls and regulatory filings can identify when a stock screen’s assumptions are invalid or when external events distort comparables. Sentiment shifts, not sentiment levels, are often what matter for detecting mispricing. The biggest opportunities exist in less efficient markets, especially mid- and small-cap stocks outside the U.S. A market-neutral process can still take modest country tilts, but alpha should mainly come from stock selection rather than macro bets. Technical analysis is less useful as a signal than as a risk-management overlay. Success does not require being right most of the time; a 53-55% batting average can be enough if risk/reward is favorable.

Data Points: Coverage universe: ~10,000 securities - Number of stocks Bridgeway covers in its global opportunity process. Sentiment metrics: ~7 metrics - Pajarni says the firm uses around seven sentiment measures, including buy-side and sell-side indicators. Topic extraction scale: ~2 million sentences - Volume of sentences ingested from earnings calls and filings since the start of last year. Document scale: ~250,000 documents - Approximate number of filings/transcripts processed in the textual analysis workflow. Country sentiment examples: Brazil, Taiwan, Korea, Indonesia highest; UAE, Australia, South Africa, China, India lowest - Aggregate median sentiment shift chart across countries. Country tilt: 1% to 3% - Typical country exposure tilt Bridgeway is willing to take based on bottom-up signals. Gross exposure outside the U.S.: Most of the portfolio - Bridgeway says its best opportunities are outside the U.S. in less efficient markets. S&P 500 exposure: as little as 2% of gross exposure - Illustrating the firm’s preference for less efficient markets over highly efficient large-cap U.S. names. Outperformance opportunity: up to 8x better - Emerging small-cap stock selection opportunities vs. S&P 500 names, according to the firm’s research. Holding period for sentiment model: about 3 months - Typical horizon for sentiment-driven positions. Holding period for aggregate model: 12 months - Forecast horizon for the broader combined model. Typical investment horizon: about 6 months on average - Pajarni’s summary of how stable signals are translated into holding periods. Target hit rate: 53%-55% - The batting average needed to make the process work, according to Pajarni. Short availability/timing: constant monitoring during market hours - The firm watches shorts closely and exits quickly when they move against it.

Pivotal Quotes: "For companies of high levels of intangible capital, fundamentals have become much less predictive of price action." — Jacob Pajarni: Core thesis on why old-school valuation breaks down for new-economy stocks. "We have completely bifurcated our stock selection approach." — Jacob Pajarni: Describing Bridgeway’s split between fundamentals for old economy and sentiment for new economy industries. "The bigger the noise, the more mispricing opportunities there are." — Jacob Pajarni: Explaining why noisy sentiment/return relationships create alpha opportunities.

Implications: Listeners should view intangibles-heavy stocks through a different lens: valuation alone is insufficient, and sentiment/text analysis can uncover mispricings. For investors, the edge comes from disciplined, assumption-checking stock selection, not broad market calls or pure technicals.