Persuasion in Finance

Sendhil Mullainathan, Andrei Shleifer

Abstract: Persuasion is a fundamental part of social activity, yet it is rarely studied by economists. We compare the traditional economic model, in which persuasion is communication of objectively valuable information, with a behavioral model, in which persuasion is an effort to fit the message into the audience's already held beliefs. We present a simple formalization of the behavioral model, and compare the two models using data on financial advertising in Money and Business Week magazines over the course of the internet bubble. The evidence on the content of the persuasive messages is broadly consistent with the behavioral model of persuasion.

Introduction

In a perfectly rational economic environment, financial advertisements should serve as a conduit for objective data. Investors, seeking to maximize utility, would require metrics such as expense ratios, risk-adjusted returns, and long-term performance benchmarks to make informed decisions. However, the reality of financial marketing often diverges sharply from this ideal. Advertisements are frequently populated with evocative imagery—such as serene landscapes or family milestones—and aggressive slogans rather than hard data.

Economists Sendhil Mullainathan and Andrei Shleifer address this discrepancy by contrasting the Traditional Model of persuasion (communication as information) with a Behavioral Model (communication as resonance). Their research investigates why the content of financial advertising fluctuates so dramatically over time and whether these shifts are driven by objective market conditions or a strategic desire to "pander" to the prevailing psychological state of the investing public.

Theoretical Framework: Growth vs. Protection

The authors develop a stylized behavioral model centered on the idea that investors oscillate between two competing belief systems, or "mental frames":

  1. Growth: A mindset where investing is viewed as a vehicle for wealth accumulation, typically activated by high past market returns.
  2. Protection: A mindset where investing is viewed as a means of securing one's future against uncertainty, typically activated by market volatility or declines.

In this model, firms do not necessarily attempt to change an investor's mind. Instead, they act as profit-maximizers that choose signals—whether emphasizing high returns or safety—that maximize the investor's perceived utility based on their current state of mind. This leads to the "Silk in a Bottle" effect: much like a shampoo advertisement might mention "silk" to evoke a feeling of smoothness without providing a chemical benefit, financial ads use "associational thinking" to link their brand with the audience's current emotional needs.

Methodology

To test these models, the researchers analyzed a hand-collected dataset of 6,440 advertisements from two major publications between 1994 and 2003:

  • Business Week: 1,469 ads (weekly).
  • Money Magazine: 4,971 ads (monthly).

This decade-long window is particularly significant as it encompasses the rise, peak, and eventual collapse of the late-1990s internet bubble. The authors manually coded these advertisements to track the presence of performance data, the types of products being promoted (e.g., growth funds vs. value funds), and the thematic messaging used by major financial institutions.

Key Findings and Case Studies

The Pandering Effect

The study found that firms do not simply report "good news" whenever it is available. Instead, they report the kind of news the market is currently primed to receive. During the height of the bubble, advertisements focused almost exclusively on absolute returns. However, after the market crash, even firms with funds that were technically outperforming the S&P 500 often ceased advertising their returns entirely. This suggests that once the "Growth" belief system was deactivated in the public consciousness, performance data—even positive data—lost its persuasive power.

The Merrill Lynch Trajectory

The evolution of Merrill Lynch’s branding provides a clear illustration of the behavioral model in action:

  • Pre-Bubble: The slogan "A Tradition of Trust" emphasized safety and stability.
  • Peak Bubble: The messaging shifted to "Be Bullish," directly pandering to the prevailing greed and growth-oriented mindset.
  • Post-Crash: The slogan became "Ask Merrill," pivoting to address the insecurity and need for guidance felt by investors in a bear market.

Strategic Silence: The T. Rowe Price Evidence

Perhaps the strongest evidence against the traditional informational model is the behavior of T. Rowe Price during the post-1999 period. While many of their funds were significantly outperforming the S&P 500 during the downturn, the firm chose not to highlight these "good" relative returns. Under a rational model, a firm should always advertise superior performance. Under the behavioral model, however, T. Rowe Price’s silence was strategic: because the general public had shifted to a "Protection" frame, advertising any returns was seen as less effective than focusing on stability and professional advice.

Empirical Results

The quantitative data strongly supports the behavioral hypothesis:

  • Market Correlation: The frequency of ads featuring fund returns was highly correlated with lagged S&P 500 performance. In Money Magazine, this correlation reached 76.2%.
  • Product Shifting: Growth fund advertisements nearly vanished during 2001–2002. Conversely, value fund advertisements—which align more closely with a "Protection" mindset—increased in frequency after the market collapse.

Conclusion and Policy Implications

The research concludes that financial persuasion is less about informing the public and more about echoing their existing biases. This "pandering" behavior suggests that financial institutions may actually exacerbate market volatility by reinforcing the prevailing sentiment of the crowd rather than acting as a stabilizing, rational counterweight.

While the study does not explicitly measure the capital inflows generated by these ads, it suggests a troubling cycle where marketing reinforces the very psychological biases that lead to bubbles and crashes. This raises significant questions for public policy: if financial rhetoric is designed to exploit cognitive frames rather than provide data, should there be stricter regulations on the style and themes of financial advertising, rather than just the accuracy of the numbers provided?

Summary updated

Abstract

Persuasion is a fundamental part of social activity, yet it is rarely studied by economists.We compare the traditional economic model, in which persuasion is communication of objectively valuable information, with a behavioral model, in which persuasion is an effort to fit the message into the audience's already held beliefs.We present a simple formalization of the behavioral model, and compare the two models using data on financial advertising in Money and Business Week magazines over the course of the internet bubble.The evidence on the content of the persuasive messages is broadly consistent with the behavioral model of persuasion.
BibTeX
@misc{mullainathan2005persuasion,
  author = {Sendhil Mullainathan and Andrei Shleifer},
  title = {{Persuasion in Finance}},
  year = {2005},
  doi = {10.3386/w11838},
}

Loading related content...

Figure