This essay is based on the working paper “What Purpose Do Corporations Purport? Evidence from Letters to Shareholders” by Raghuram Rajan, Pietro Ramella, and Luigi Zingales.
Since the 1932 Harvard Law Review debate between Merrick Dodd and Adolf Berle, myriad articles have been written on the goals corporations should pursue. Less attention has been devoted to the question of what goals corporations actually end up pursuing, and even less to the questions of what goals managers declare they want to pursue, why they do so, and with what consequences. These are the questions we try to address in this paper.
Analyzing Annual Reports to Shareholders
To achieve this goal, we exploit an important source of information that has been largely ignored by the finance and accounting literature: the letter that the chief executive officer or the chairman of a company writes to investors to introduce the annual report. The practice of an introductory letter predates the SEC mandatory disclosure rules and, by and large, has survived to this date, although some companies (most notably Apple) have recently stopped producing it. Until the introduction of social media, this letter represented the most important piece of direct communication between the CEO and a company’s various stakeholders.
To analyze the historical changes in companies’ stated goals, we follow a sample of firms over time: the 120 largest industrial firms by sales and the 30 largest financial firms by assets from 1955 to 2020. To read these letters and identify the goals, we make use of natural language processing (NLP)—a subfield of artificial intelligence that enables computers to comprehend and analyze text—which allows us to identify paragraphs in a letter that express goals and what goals they are. The paper's current version uses the BERT NLP program, though we are working on a new version that will use ChatGPT4.
Corporations’ Goals Have Increased over Time
Our analysis shows an explosion in the number of goals from 1955 to 2020. In 1955, barely 33 percent of the letters contained an explicit goal. Conditional on having a goal, the average number of goals was two. By 1980, almost 80 percent of the letters contained a goal, and, conditional on having one, the average number of goals was three. By 2020, all letters contained a goal, and the average number of goals was about seven. Throughout the sample period, maximizing the return to equity holders was the most popular goal, even though the formulation in terms of “shareholder value” only appeared in the 1980s. Societal goals had a temporary moment of popularity in the late 1960s and then exploded in importance in the last ten years.
Trends in Corporations’ Stated Goals
To explain the trends in letters’ goals, we look at changes in the importance of various stakeholders and the preferences of these stakeholders. Starting in the 1980s, an increase in institutional ownership and the rise of hostile takeovers made shareholders a more powerful constituency. Not surprisingly, shareholder value became a more frequent and explicit goal in shareholder letters. Similarly, increased foreign competition made CEOs more sensitive to customers’ needs, starting in the mid-1980s. Changes in the preferences of key constituencies also matter. For instance, corporations scrambled to announce an ethics goal as the public’s concerns with corporate ethics grew with the Enron, Tyco, and WorldCom scandals in the early 2000s. As more of a corporation’s institutional investors signed the UN Principles for Responsible Investments (PRI) in the last decade, the emphasis on social and environmental goals also increased proportionately to PRI signatories' weight among a company’s shareholders.
The variation in goals across firms can be explained by the commitment role of announcements. CEOs use the letter to signal their commitment to addressing a problem (lack of profitability or bad environmental record) or to sticking to a strategy like innovation.
Do CEOs Mean What They Say?
One important question is whether the stated goals are just “mere pufferies” empty of any real content or whether managers mean what they say. To answer this question, we look at the correlation between the goals stated in the latter and the target embedded in CEOs’ compensation. CEOs who proclaim shareholder maximization as a goal are more likely to be paid in stock, although it is unclear which is the cause and which is the effect. Interestingly, CEOs who announce environmental goals are also more likely to have a component of their bonus linked to environmental goals, but this component is small and the environmental target easily achievable.
Finally, we test whether these stated goals impact performance over the medium to long term (five to ten years). We find very little evidence that a focus on shareholder value is associated with higher (or lower) shareholder value, whether this is in the form of higher stock returns, a larger increase in dividends, or profitability. A focus on shareholder value does, however, tend to be associated with lower asset and sales growth.
Our findings suggest that the proliferation of goals between 1955 and 2020 results from the pressures CEOs receive from different stakeholders. The fact that these goals proliferate in number and complexity suggests that the CEO’s role has become significantly more complex over time.
Read the full paper here.
Luigi Zingales is the Robert C. McCormack Distinguished Service Professor of Entrepreneurship and Finance at the University of Chicago Booth School of Business.
This essay is part of the Corporate Governance Research Brief Series. Research briefs highlight research with policy implications for the regulatory systems that impact corporations.
