What Is a Supermarket?

On February 26 the Federal Trade Commission (FTC) moved to block the proposed union of two giants of food retailing, Kroger Company and Albertsons Companies, respectively the second- and fourth-largest supermarket operators in the United States. In 2023 Kroger took in $150 billion from 2,750 stores in thirty-five states and the District of Columbia; Albertsons posted revenue of $79 billion from 2,300 stores in thirty-five states and D.C. The antitrust complaint is to be heard initially by an FTC administrative law judge, who has the power to review the evidence and issue a cease-and-desist order against violations of the law. Separately, the FTC has asked a federal district court to issue a preliminary injunction that would delay the merger while the investigation moves forward. Whatever the result of these proceedings, the case will likely end up before a US Court of Appeals. 

The FTC’s complaint contends that “the destruction of competition between these two head-to-head rivals risks raising prices, worsening services, and lowering quality for the millions of consumers who rely on Kroger and Albertsons for their groceries and other everyday goods.” Unusually for an antitrust complaint, it also addresses the merger’s potential impact on workers, asserting that the merged company would have excessive power in the market for grocery-store labor, “likely leading to lower wages and reduced benefits, opportunities, and quality of workplace conditions and protections.” 

The FTC’s action is the latest battle in a longstanding war over market power in food retailing. It raises some perplexing questions about which neither courts nor economists who specialize in competition have had much to say. What if blocking the union of two giants benefits other, even larger giants? Should antitrust law be used to improve labor unions’ bargaining power? And what, exactly, is a supermarket?

Until the 1890s American grocery stores were all pretty much alike. Most were bare-bones shops, often no larger than the living room of a row house, occupied by independent grocers, who sold goods hardly different from those at the store down the block. Chains came into being when advances in canning and box-making gave manufacturers a simple way to put labels on products, allowing stores to distinguish themselves from one another by selling different brands. Many of the early grocery chains emerged from tea-store chains. A&P, formally the Great Atlantic and Pacific Tea Company, sold teas before turning to groceries around 1890; between 1920 and 1962 it was the world’s largest retailer of any sort. Both Kroger and Albertsons also began by selling tea: Kroger was founded in 1883 in Cincinnati as Great Western Tea Company, and Albertsons traces its origin to the Acme Tea Company, established a few years later in Philadelphia. 

At the turn of the twentieth century there were a few hundred chain grocery stores, mainly in the Northeast. Two decades later there were probably more than 10,000. These chains had enormous advantages over mom-and-pop stores. They could afford newspaper advertisements and fancier premises, brand their own products, and cart merchandise from their own warehouses in their own vehicles. Their greatest advantage, though, was the sheer magnitude of their sales volume, which they leveraged to demand that suppliers sell to them directly at a volume discount, circumventing wholesaler markups. A chain that saved a few cents on a crate of apples or a case of breakfast cereal could retail such products at prices mom-and-pop stores could never match. 

There is often a reasonable economic justification for volume discounts: for instance, the cost of making and delivering a can of tomato sauce is lower when the manufacturer is filling a firm monthly order for 10,000 cases than an occasional order for half a dozen. But volume discounts were controversial at a time when price-cutting was widely considered unfair. In 1897 the magazine American Grocer accused A&P of “supreme selfishness” for underpricing competitors. A decade and a half later, while running for president, Woodrow Wilson expressed much the same opinion, as did Louis Brandeis, the Massachusetts consumer advocate whom Wilson appointed to the Supreme Court in 1916. “The evil results of price-cutting are far-reaching,” Brandeis wrote in 1913. It would harm consumers in two ways. In the first place, retailer pressure would force manufacturers to lower prices, which in turn would lead them to reduce product quality, thereby driving high-quality goods from the market. Second, any benefits of discounting would be short-lived because the price-cutter would eventually drive competitors out of business and then hike prices like a monopolist. 

Brandeis’s views found expression in a powerful nationwide movement against retail chains. From the mid-1920s, states and cities across the country imposed taxes designed to disadvantage them. In 1929 and 1930 alone some 142 bills to tax chain stores were introduced in twenty-nine state legislatures. In May 1931 the Supreme Court, including Brandeis, upheld an Indiana tax that rose based on the number of retail stores under the same ownership. As early radio talk show hosts promoted the anti-chain movement, politicians from coast to coast clambered aboard. “I would rather have thieves and gangsters than chain stores in Louisiana,” the populist Democrat Huey Long, then a US senator, pronounced in 1934, shortly after his state adopted the nation’s highest chain-store tax. 

Taxation was not the only weapon against large grocers. Many states adopted “fair trade” laws, which typically required chains to maintain uniform prices at all stores or capped how much they could mark up merchandise above the wholesale price. The federal Robinson-Patman Act of 1936 prohibited sellers from charging different prices for the same commodity and selectively offering advertising or promotional allowances. These policies were aimed principally at A&P, which in 1938 had 12,000 stores, more than its four largest competitors combined. 

The crusade reached its peak in 1942, when Franklin Roosevelt’s administration charged A&P with violating antitrust law. US v. New York Great Atlantic & Pacific Tea Co. was not a merger case; over eight decades A&P had purchased few stores and never acquired another retail chain. The government’s complaint was rather that the chain strong-armed suppliers to lower their prices. “A&P sells food cheaply in its own stores because it is a gigantic blood sucker, taking its toll from all levels of the food industry,” one prosecutor told the court. The company was convicted and fined. To avoid further attacks, it abandoned aggressive discounting and turned to operating like just another grocery chain, gradually declining until it closed its doors for good in 2015.

The court’s decision in the A&P case ushered in half a century of relative stability in the grocery market. With price competition limited, consolidators such as Kroger and Albertsons expanded by building larger stores and acquiring other chains, often from family owners. Many regional markets featured a sizeable number of competitors, including independent grocers and local chains. In 1993, for example, seven supermarket operators had market shares above 10 percent in Phoenix and independent stores accounted for nearly half of supermarket sales in Milwaukee. Rabid price-cutting was rare.

This changed with the entry of Walmart. Then known for selling general merchandise at low prices in small towns across the South and Midwest, in 1987 Walmart joined a regional supermarket operator to open a store called Hypermarket USA in suburban Dallas. This joint venture, occupying around five times the space of the average supermarket, was intended to test whether shoppers would buy groceries in huge stores that sold other types of goods as well. It proved a success. The next year Walmart rolled out its first “supercenter,” offering meat, produce, and packaged foods alongside lamp shades, television sets, and children’s clothing. No other retailer could match such an assortment of products.

Walmart made warehousing, logistics, and advertising more efficient by clustering numerous supercenters around a grocery distribution center. As a result, its grocery prices, on average, were about 10 percent below those at nearby supermarkets. By 2003 it was the largest food retailer in the US. Today the company accounts for more than half of grocery sales in many parts of the country, from Bismarck, North Dakota, to Oklahoma City to every major market in Mississippi. 

The FTC looks closely at mergers involving supermarkets in the same geographic area. By building stores rather than buying them, however, Walmart has largely escaped this scrutiny—and, in a certain way, even benefited from it. In 2000, for example, when Kroger proposed buying seventy-four supermarkets in Oklahoma and Texas from the Winn-Dixie chain, the FTC determined that the deal “would create a new dominant firm in Fort Worth” and blocked it. Two years later Winn-Dixie unloaded the stores mainly to smaller chains that could hardly challenge Walmart, which built dozens more supercenters in the region and became its dominant food retailer. 

The proposed merger of Kroger and Albertsons raises similar issues. It would allow the combined firm to operate more stores in many local markets, triggering alarm signals from the Herfindahl-Hirschman Index (HHI)—the main gauge of competition on which antitrust authorities rely. The HHI is calculated by summing the squares of the market share of each company in the market. If, to take a simple example, ten supermarkets each control 10 percent of a local market, the index reading would be 10² x 10 = 1000. Under federal guidelines, a merger that drives the local index over 1800 and raises it by over one hundred points is presumed to “substantially lessen competition,” no matter how large the market shares of other firms. 

It is frequently the case that one merger triggers another and then yet another, as companies within an industry compete to achieve greater scale. Often, mergers in one industry lead to fusions among customers or suppliers; as health insurers have consolidated to increase their ability to hold down payments to hospitals, for instance, formerly independent hospitals have been rolled up into networks that wield greater bargaining power with insurers. Similarly, the use of the HHI as a guide rewards early movers: a deal that would be unproblematic when the index is low may be unacceptable once other firms in the same industry have merged, raising the index nearer to the 1800 level.

By relying on the HHI, the trustbusters effectively tilt the playing field in Walmart’s favor. Assume that Walmart has 30 percent of the local grocery market, which is the case in much of the US. That counts for 900 points (30²) on the HHI. If there will be seven other grocers in the area, each with a market share of 10 percent, each counts for 100 points on the index, making the overall index reading 1600. Suppose two of those grocers decide to merge. Their joint index reading will jump from 200 (the sum of the squares of each firm’s 10 percent market share) to 400 (the square of the combined firm’s 20 percent share). The overall index for the market has increased by more than 100 points and would now read 1800 (900 for Walmart + 400 for the newly merged firm + 500 for the five remaining small firms), thus rendering the merger suspect. The HHI doesn’t affect Walmart’s ability to retain a large market share, but it may inhibit smaller firms from merging in hopes of obtaining the local efficiencies that Walmart enjoys. 

In its complaint seeking to block the Kroger-Albertsons merger, the FTC contends that “the retail sale of food and other grocery products in traditional supermarkets and supercenters” represents the relevant market for antitrust authorities to scrutinize. This is almost identical to the market definition the FTC used back in 1999, when it demanded that Albertsons and American Stores sell 144 supermarkets to win approval of their merger: “The relevant line of commerce…is the retail sale of food and grocery products in supermarkets.” 

This definition made sense when grocery stores sold only groceries, pharmacies sold only drugs and toiletries, and clothing stores sold only apparel. Back then, determining a grocer’s share of retail food sales was straightforward. But the retailing business isn’t what it used to be. In the face of Walmart’s competition, several supermarket operators—such as Kroger, H-E-B in Texas, and Meijer in the Midwest—have built large stores carrying garden chairs, auto parts, and footwear in addition to consumables like food and cleaning products. Conversely, many companies that do not claim to be supermarkets sell food and other traditional supermarket items. The trade publication Progressive Grocer lists CVS and Walgreens, both generally considered drug-store chains, as respectively the nation’s fifth- and sixth-largest sellers of food and consumables. Dollar General, better known for bare-bones stores that peddle cheap imports, says that “consumables” make up about 80 percent of its sales. Last year Target sold $22 billion of food and beverages. Club stores such as Costco move mountains of groceries. Amazon has a sizeable online grocery business in addition to its physical grocery stores. And deep discounters are quietly nibbling away at supermarkets’ food sales, notably Aldi, whose outlets average about a third of the floor area of a typical supermarket and offer only a fraction as many products.

In essence, the FTC’s complaint declares that Walgreens, CVS, and even Trader Joe’s, which offers a limited assortment of food in stores far smaller than a typical supermarket, don’t compete with Kroger and Albertsons in the grocery market. In some locations or for some types of consumers, that may be accurate; Dollar General’s target buyers for frozen fish sticks are households earning less than $40,000 a year, not shoppers with six-figure incomes. By ignoring the role of nontraditional food retailers, however, the FTC risks overestimating the giants’ market power. 

Undoubtedly there are some local markets where, in the absence of a Target, Costco, or Aldi nearby, Kroger and Albertsons combined would dominate grocery sales. This is a not a unusual issue. Kroger and Albertsons have proposed lists of stores they would sell off to maintain local competition, if allowed to merge. The FTC has rejected their lists, and is rightly concerned that C&S Wholesale Grocers, the main buyer they have identified, does not have a track record of running supermarkets with success. It will also want Kroger and Albertsons to sell enough stores in individual localities so that the buyers have a sufficient store base to compete effectively. These do not seem insuperable obstacles. If the two retailers are truly keen on joining forces, they should be able to come up with a list of divestitures that satisfies the FTC. 

Why then is the FTC so concerned about this case? The answer may lie near the end of the complaint, which asserts that the merger could lessen demand for labor, specifically union labor. One in seven grocery workers is represented by a union today, compared to one in three forty years ago. Both Kroger and Albertsons are islands of union strength. Between them, they employ over 700,000 workers, most of them represented by either the United Food and Commercial Workers or the Teamsters. Both unions have called for the FTC to block the merger on the grounds that it would eliminate their ability to play one company against the other in places where both presently have stores. Unstated in the complaint is an equally pressing concern: over the past several decades unions have found little organizing success in food retailing. If they lose members as Kroger and Albertsons combine stores, they are unlikely to find new ones elsewhere. 

Neither courts nor economists have devoted much attention to whether the possible weakening of union bargaining power is a legitimate antitrust issue. But the FTC’s attention to the subject is consistent with the broader concern expressed by some current commissioners about monopsony, the ability of a buyer to exert market power over sellers—in this case, sellers of labor. Antitrust thinkers have ignored monopsony for decades, mainly on the assumption that in a large, complex economy there are enough potential buyers of almost any good or service to give sellers multiple options. Since the 1980s antirust policy has been focused near-totally on consumer welfare, rendering monopsony irrelevant, since a buyer’s power to force a seller’s price down is unlikely to make consumers worse off. 

The current chair of the FTC, Lina Khan, is an adherent of the “New Brandeis” school of thinking about market power, which rejects the hoary claim that mergers are inherently good if they lower prices for consumers and seeks to inject matters like economic power and the concentration of wealth into the discussion; the New Brandeisians worry about how bigness endangers democratic institutions as well as economic freedoms. Khan’s FTC has brought novel antitrust cases during her three-year tenure as chair. At the end of 2021 it voted to keep the semiconductor giant Nvidia from purchasing the British company Arm, which supplies chip design tools to Nvidia and its competitors; that action looks prescient given the eightfold increase in Nvidia’s shares since the artificial intelligence boom began a few months later. In 2022 the commission sought to block Meta’s acquisition of Within, a virtual reality studio; a judge tossed the suit, which made the claim that Meta, the owner of Facebook, would dominate the incipient market for fitness virtual reality apps. In September 2023 the FTC joined seventeen state attorneys general in the most sweeping antitrust suit in decades, accusing Amazon.com of a wide range of illegal practices intended “to keep rivals from gaining the scale needed to compete effectively.” Now, in the Kroger-Albertson case, the FTC is pushing courts to address the argument that antitrust law can be used to protect workers.

Khan’s agenda to encourage judges to return to the roots of US antitrust policy is worthwhile. But the potential marriage of Kroger and Albertsons raises an unusual challenge: How can the FTC address the merger of two retailing giants without even larger giants coming out as winners? While the FTC enforces antitrust law against Kroger and Albertsons, it’s hard not to imagine Walmart and Amazon, both of which are notoriously anti-union, cheering from the sidelines, rooting for a resolution that hinders the expansion plans of their highly unionized competitors. This is the paradox the New Brandeisians will have to resolve.