My wife and I are huge Mad Men fans – we’re waiting for the 5th season to make it to Netflix so we can catch up. It resonates in particular for us since we grew up in the 50s and 60s. But is advertising still like that, we wonder? Fortunately, the HBS Working Knowledge Blog has the answer for us:
Fans of the television show Mad Men are well acquainted with the mystique of the advertising business, circa 1960s, where relationships were consummated over martinis and campaigns fashioned through the wizardry of creative director Don Draper, swooping into the pitch meeting at the eleventh hour with his latest brilliant inspiration. For decades, advertising agencies have thrived on, and in some ways fostered, the idea of advertising as a creative black box.
“How you get a breakthrough campaign rather than a run-of-the-mill one has long been the burning question for clients and their agencies,” says Alvin J. Silk, the Lincoln Filene Professor of Business Administration, Emeritus, at Harvard Business School. “Hence the famous saying attributed in US advertising circles to legendary retailer John Wanamaker: ‘Half the money I spend on advertising is wasted; the trouble is I don’t know which half.’ “
For the past several decades Silk has devoted himself to prying open the black box, understanding the economics underlying the work advertising agencies do: the development and production of advertising campaigns. As he’s done so, Silk has charted fundamental changes in the advertising business—some of them well known, others that address long-standing puzzles or that challenge conventional wisdom.
In a trio of papers, Silk has laid out his research on three key developments that have shaped the industry in recent years: unbundling of agency services, the problem of competing clients sharing a common agency, and concern over consolidation with the growth of holding companies. To understand them, it’s first important to appreciate how advertising has traditionally operated.
Shaping the industry
For much of the twentieth century, companies relied on a “full-service” agency for most or all of their advertising service needs, including both creative development and media planning and buying. Agencies were compensated primarily by commissions related to the volume of media space and time they purchased for clients’ campaigns.
This practice was institutionalized by an arrangement known as the “recognition system” between agencies and publishers, and administered by a set of trade associations. Among other things, the system supported standards for granting of credit to agencies and established a fixed commission rate paid by magazine and newspaper publishers to agencies. These practices served to discourage price competition among agencies and facilitated the bundling of services by full-service advertising firms.
After complaints from advertisers and a series of antitrust investigations, the recognition system was essentially dismantled in 1956 with a consent decree signed by the trade associations involved in administering the system.
But nothing much changed.
As Silk, Mohammad Arzaghi, Ernst R. Berndt, and James C. Davis detail in The Unbundling of Advertising Agency Services: An Economic Analysis, agencies continued to bundle their advertising services and collect media commissions for decades. By analyzing US Census Bureau data, the researchers demonstrated that unbundled services accounted for only 13 percent of business in 1977, 20 years after the consent decree.
In part, the report suggests that while the change dismantled the institutional structure used to administer the recognition system industry-wide, it did not prevent agencies and media suppliers from continuing the same practices individually.
That practice finally started to change in the 1980s. Unbundling began to increase significantly, with income from fees and other direct compensations growing from 23 percent in 1987 to 65 percent by 2007.
What happened? The paper presents an economic model of an agency’s decision to unbundle its services that accounts for both the slow pace of change and the variable preferences for unbundling among agencies and their clients. Using microdata collected by the US Census Bureau from 1982 to 2007, the researchers found that agencies were more likely to unbundle if they were large and diversified in their service offerings, as media prices grew higher.
As agencies were unbundling, growing, and diversifying, they were also becoming more globalized.
When clients compete
In combination, these developments greatly exacerbated an old problem: the potential for conflicts of interest to arise when competing companies hire the same advertising agency for their business.
Historically, advertising agencies in the United States and Europe did not simultaneously serve accounts or clients that were competitors—a practice that also limited an agency’s growth. The holding company was introduced as a means of addressing this issue.
The essential idea was to operate two or more agencies (or “networks” of specialized service agencies) under common ownership as quasi-independent entities. But since the mid-1980s, waves of mergers and acquisitions have created potential conflicts of interest as independent agencies that were separately serving competing clients suddenly came under common ownership. In addition, mergers between advertisers could be a source of conflict-of-interest problems: for example, when the ad agency of an acquired company also provided services to a firm that competed with the client in another product category.
Silk explores this problem in his 2012 paper, Conflict Policy and Advertising Agency-Client Relations: The Problem of Competing Clients Sharing a Common Agency.
According to Silk, industry opposition to an agency serving competing accounts is driven by two concerns: first, proprietary information might be leaked (even though there have been no recorded cases of that happening) and second, an agency serving competing accounts will give rise to “divided loyalty.” What assurance does a client have that it, rather than a competitor, will be assigned the agency’s best talent and be the beneficiary of breakthrough ideas they may develop?
“There is this general sense that if you are going to come up with a great creative idea that moves the market in some sector, you better do it for me rather than my competitor,” says Silk.
While “exclusivity”—one agency, one client—has been the norm in the United States for decades, Silk found as he examined newspaper articles and court filings that two significant changes in conflict policies adopted by agencies and their clients have occurred in recent years.
First, safeguards to protect proprietary client information that function as organizational, locational, and personnel mobility barriers have become essential components of conflict policies. Second, a family of hybrid conflict policies has evolved that feature elements of the “split account” system long practiced in Japan.
“What surprised me,” Silk says, “was that these developments have received relatively little attention. Exclusivity is still the norm, but the consensus around it is slowly shrinking.”
The consolidation myth
Choices made in conflict policies can influence the size and distribution of agencies that the industry can support. This has led Silk to investigate how the structure of the industry in the United States has changed over the past few decades following several M&A waves. Concern has been raised that the industry has become highly concentrated, dominated by a handful of large holding companies.
In the paper “How Concentrated Is the U.S. Advertising and Marketing Services Industry? Myth vs. Reality” (forthcoming in the Journal of Current Issues and Research in Advertising), Silk and coauthor Kip King take aim at the conventional wisdom that the ad industry has become highly consolidated.
In measuring degrees of industry concentration, the trade press has counted only billings of the largest 500 firms or so. By contrast, Silk and King analyzed data from the US Census Bureau that includes revenue from some 10,000 agencies of all sizes, over half of which are small and serve local rather than national clients.
In the case of traditional advertising agencies, they found concentration increased slightly from 1977 to 2002, but then declined somewhat by 2007, falling well within the range that economists use to judge industries as “competitive.”
Moreover, the share of total revenue from the entire US industry (traditional advertising agencies plus firms specializing in PR, media buying, marketing research, etc.) captured by firms controlled by the four largest holding companies has remained a stable 25 percent for the past decade. This figure is roughly half the estimates often reported in the trade press.
That makes sense, says Silk, since a range of factors favor a highly diverse and geographically dispersed industry, especially low-entry costs and limited economies of scale. The advertising business continues to attract creative talent who form and staff small, nimble firms, all the while perpetuating a diverse and highly competitive industry structure.
Since Don Draper’s time, the advertising industry has gone through—and might not be done yet with—a major restructuring. Technology, globalization, government intervention, and a shift away from reliance on mass markets have changed the environment in which advertisers, media, and agencies operate.
As Silk’s research demonstrates, agencies have shifted historical industry policies, starting with a transformation from full-service and media commission-based compensation to unbundling and cost-based fee compensation. The traditional norm of strict exclusivity has been at least partially relaxed, and agencies have begun adopting hybrid conflict-of-interest policies wherein competing accounts are served by a common agency—augmented by safeguards to protect confidential information and avoid conflicts of interest.
Despite these changes (and probably because of them), the result is an industry that “remains relatively unconcentrated at both firm and holding company levels, diverse in scale and scope, geographically dispersed, and highly competitive.” This structure, Silk says, is well suited to serve the heterogeneous demands of large national advertisers and smaller local ones.
“I think the story here is how the industry has evolved and adapted” to the growing demands of advertisers and to the increasing supply of media available, says Silk, adding that the industry continues “to deliver advertising messages and remain viable as intermediaries in the process of creating and producing advertising campaigns.”