The great pyramids of America: A revised history of U.S. business groups, corporate ownership, and regulation, 1926–1950

Abstract

Research Summary: Control‐magnifying (pyramidal) business groups—multiple tiers of partially‐owned listed affiliates and fully‐owned private affiliates, a dominant organizational form around the world—are virtually absent in America today. Using newly‐assembled historical data, we show that such groups were ubiquitous in the U.S. economy in the 1930s. They came under attack because of their economic and political sway: Some New Deal reforms—proscriptions against public utilities pyramids, intercorporate dividend taxes, and rules governing investment companies—were explicitly aimed at deterring existing groups and preventing new ones from forming. Others, for example, estate taxes and securities law reforms, may have also worked against them. No single reform triggered the immediate dissolution of groups; they broke up under an ongoing anti‐big business sentiment. These events offer lessons for policymakers today. Managerial Summary: Most listed U.S. firms are “standalone”—they do not control, and are not controlled by, other listed firms. Control‐magnifying (pyramidal) business groups—multiple tiers of partially‐owned listed affiliates and fully‐owned private affiliates (e.g., Samsung or Tata)—are ubiquitous around the world but virtually absent in United States. In such groups, the combination of multiple tiers and partial ownership enables control over vast corporate empires. Using newly‐assembled historical data, we show that the U.S. corporate ownership as we know it today is a recent phenomenon: pyramidal groups were common in United States of the 1930s. President Roosevelt, who regarded their economic and political power as excessive, initiated a sequence of reforms which appears to have worn these groups down and probably also kept new ones from forming. JEL CLASSIFICATION: G3, G34, G38, N22.

A trade-off theory of ownership and capital structure

“…So far the model only considers income taxes, for consistency with prior capital structure literature. However, regulatory authorities restrict group activities because of either governance (Kandel et al, 2015), or tax (OECD, 2016) or financial stability (Herring and Carmassi, 2009) motives. Our structural model allows to infer how group-specific rules impact ownership, leverage, default costs and taxes.…”

Technological progress and the future of the corporation

“…3 The nascent limitations on anti-competitive practices introduced around the turn of the 20th century did not preclude the formation of a new type of corporate empire: pyramidal groups controlled through holding companies to which existing antitrust tools did not apply. Kandel et al (2018) describe this process in detail; holding companies first became legal in New Jersey in 1888 and then other 'competing' States followed and legalised the holding company organisational form. Of special interest for the discussion here are public utilities (PU) groups/holding companies: Much like the rise of railroads and others monopolies in the 19th century, the rise of PU groups was related to the rise of new (network) technologies, telephone, telegraph, electricity, etc.…”

“…Once again, much like in the late 19th century, the existing regulatory tools were deemed inadequate or insufficient. Antitrust regulation, in particular, was underfunded and not effective against the new threats in the 1930s (Kandel et al 2018). Hence, new regulatory tools had to be invented to deal with the business group/ holding company form of monopoly power.…”

“…The Public Utilities Holding Company Act (PUHCA) targeted PU groups and, together with other new regulatory tools (the intercorporate dividend tax, increased estate taxes etc. (see Kandel et al 2018)), eventually brought about the virtual elimination of the business group corporate form in the US by 1950.…”

“…Wu (2010), who describes the history of Bell/AT&T during the 20th century, also alludes to cycles of innovation in telephony and subsequent monopoly power, some of which took place during the 1920s and 1930s. 5 The distinction between the business groups and holding companies and their affiliates that do not constitute a group-which is based on the number public companies under common control (see Kandel et al 2018)-does not matter for the present discussion. 6 In a recent article, Khan (2016) describes in considerable detail why the standard ('Chicago') tools of antitrust analysis are insufficient for addressing the market power of Amazon and other tech giants.…”