the area of food retailing. Of particular importance were economic reports completed during the 1960's: Economic Inquiry Into Food Marketing Concentration and Integration in Retailing (1960), The Structure and Competitive Behavior of Food Retailing (1966), The Use of Trading Stamps by Food Retailers (1966), and Use of Games of Chance in Food and Gasoline Retailing (1968).

I. SUMMARY

This study confirms previous Government findings that the distribution system performs less satisfactorily in low-income areas of our inner cities than in suburban areas. Many foodstores serving low-income, inner-city areas are small, less efficient, and have higher prices. Consumers in these areas are frequently sold lower quality merchandise and are provided fewer services than in other areas. Moreover, the retail facilities of low-income areas are often old and in a shabby state of upkeep.

On the basis of special investigational surveys and hearings the staff found no evidence that leading chainstore operators in the District of Columbia and San Francisco employ discriminatory policies designed to exploit low-income customers. Each of the largest food chain operators had an official policy of price and quality uniformity. To a significant degree, systematic departures from store-to-store price uniformity were discovered. However, for the most part, these involved responses to special competitive situations and could not be interpreted as reflecting an effort to discriminate against low-income customers although that was generally the result.

These findings, however, should not lead one to conclude that food. distribution in low-income areas is free of problems, or that the lowincome customer does not pay more for food. On the contrary, the reverse is quite likely to be true. Food marketing is not as well organized in low-income areas as in newly developed suburban areas. There simply are not as many modern, efficient supermarkets in low-income areas as there are elsewhere. Thus, the low-income consumer is more likely to do his shopping at a small, independent mom-and-pop store. Such stores generally charge higher prices, whether located in low- or high-income areas. At the same time those supermarkets which are operating in low-income areas generally face less intense competition than they would elsewhere. The lack of competition means there is less pressure to maintain tight managerial control, to improve quality and service, or to lower prices.

Although departures from official company price lists are occasionally authorized to dispose of distressed perishable merchandise, differences in competitive conditions clearly outweigh this as a reason for deviations from official areawide prices. Deviations from areawide prices of chain organizations most frequently occur where stores of one chain meet those of other chains in strong "head-to-head" competition. In these areas special price reductions are frequently authorized, more concern is shown for the appearance of physical facilities and the quality of services provided, and there are more extensive promotional efforts, including the giving away of things of value, such as prizes in promotional games. In many instances authorized price reductions amount to several cents per dollar of total store sales. It is highly significant that not once during the period

Retail Food Prices in Low and Higher Income Areas, Bureau of Labor Statistics, U.S. Department of Labor, February 1969.

investigated in this study did a "special competitive situation" occur in an inner-city poverty area in either of the two cities studied.

The structural characteristic that is primarily responsible for ineffective competition in low-income inner-city areas is an inadequate number of supermarket competitors. To a great extent, the condition of entry into the low-income central city is related to entry conditions and the degree of competition in the metropolitan area as a whole. In these aspects of competitive structure there are important differences between the two cities studied-Washington, D.C., and San Francisco-that may in part explain the performance differences found between the two cities. Concentration of grocery store sales. is very high in the Washington, D.C., metropolitan area-the highest of all major U.S. cities. The four largest chains of the Washington metropolitan area accounted for 67 percent of total grocery store sales. This percentage was half again higher than the average for other cities ranking among the 20 largest. San Francisco, on the other hand, ranked low in sales concentration among major cities. The four largest chains in the San Francisco metropolitan area accounted for only 33 percent of total grocery store sales. Concentration of supermarket sales in the inner city is much higher. In the District of Columbia, which is somewhat larger than the "inner city" of the metropolitan area, the four largest food chains accounted for 83 percent of supermarket sales.

Entry barriers into the Washington market also appear to be high. Within the last decade two chains have attempted to break into the Washington market. One of these has abandoned its attempt and the other has achieved only marginal success. The Kroger Co., the Nation's third largest grocery chain, entered the market in 1960 with the purchase of a small local chain. After making a substantial effort to expand its market share, Kroger sold its Washington area stores in 1966 to another grocery chain operating in the area.

The second chain that attempted to enter was an aggressive lowmargin food retailer from the New Jersey area, which had a history of successful entry into several markets before attempting to enter the Washington market. This chain came into the Washington market by opening three stores. It has since closed two of them. Just prior to this chain's entry into the Washington market, the stores of two leading Washington area chains located near the stores of the new entrant cut their prices substantially below those charged in the rest of the metropolitan area. In doing so, these stores operated on abnormally low margins and-for those stores for which data were available-sustained substantial losses.

Special competitive pricing situations were not the only variations in food chain conduct due to competition. Pricing surveys conducted by the Federal Trade Commission staff revealed that food chain promotional activities also varied according to the competitive setting of the store. Specifically, advertised special items were more frequently unavailable in less competitive, low-income area stores than in more competitive, higher income area and suburban stores.

Prices of advertised special items present substantial savings to purchasers. Sales of advertised special items during a typical week commonly reduce the average price level of a store by 5 percent. Those customers that take full advantage of the specials can save 10

percent or more on their weekly food bills. The availability or unavailability of specials therefore is an important aspect of pricing conduct. Federal Trade Commission price surveys of stores of Washington area chains found that 23 percent of advertised special items were not available in low-income area stores as compared to only 11 percent not available in higher income area stores. In San Francisco. the percentages were 7 and 5 percent, respectively. In addition to unavailability, advertised special items were also frequently found to be mispriced. For both cities, an average of 7 to 8 percent of advertised special items available in stores was found to be mispriced. There were three chances in four that the incorrect price was higher than the advertised price.

In addition to advertised specials, variations in many nonprice dimensions of chainstore conduct affecting the values received by customers were also related to competitive conditions. Although the FTC price surveys were primarily designed to record price and availability information, the checkers also observed various quality characteristics. An analysis of these reveals substantial differences in the appearance of perishable products such as produce and meat items packaged in the store.

The policy of the largest chain in the Washington, D.C., metropolitan area in distributing $1,000 winners in its games-of-chance promotion further illustrates the manner in which merchandising and promotion policies may be varied depending upon competitive conditions. During 1966, this chain awarded 48 $1,000 winners in the area served by its Washington division. Yet, whereas the central city represented by the District of Columbia contained 30 percent of the division's stores, only two of the 48 $1,000 winners were awarded at stores located in the District. Both prizes were awarded by stores in upper-income areas located at the far Northwest fringe of the District. The winners themselves lived in suburban Maryland. Available information indicates that the distribution of the balance of prize money-85 percent of the total-approximated the distribution of store locations. Some other leading chains in the Washington area also systematically allocated their major winners to certain stores; however, data were not available to determine how this allocation affected different areas.

During investigational hearings conducted over a 9-month period, the major food chains in the Washington, D.C., and San Francisco metropolitan areas explained their pricing systems and submitted various documents relevant to understanding them. Hundreds of pages of testimony were taken. Although some conflicting evidence was presented, none of the testimony indicated that any of the chains engaged in discrimination aimed specifically at low-income areas. The price and nonprice policies of chains appeared to be attuned to the competitive circumstances in which individual stores operated.

Finally, the evidence indicated that most chains had insufficient central office control over the operations of individual stores to prevent extensive mispricing of individual items. Price surveys indicated that between 5 and 10 percent of the prices found in the stores of a chain differed from those officially authorized by the chain for the dates of the surveys. For items not offered as "specials," there appeared to be a near equal likelihood that the unauthorized prices would be either high or low.

POLICY ALTERNATIVES

There are two general approaches to improving the quality of lowincome area food retailing. One focuses on conduct, with a view to preventing excessive mispricing and unavailability of advertised items. The other focuses on changing the structure of the market to make it more competitive. Performance of low-income area supermarkets would likely improve if more supermarket competitors entered the low-income area market.

Before turning to policy alternatives, a word of caution should be raised concerning oversimplified solutions to complicated problems. A frequently proposed reform for lowering food costs to low-income area residents would require food chains to charge the same prices and provide the same product quality and arrays of services in their lowincome area stores as in their suburban area stores. This is a simple approach, but like so many simple approaches it may overlook essential facts. The primary reason low-income inner-city residents pay more for food is that there are too few supermarkets in these areas. Most chains serving our major cities avoid operating in the lowincome inner city and many chains that once operated in those areas withdrew to the suburbs as the inner city became a less desirable place to operate. Were it not for the few chain and independent supermarkets remaining in low-income inner-city areas, food prices to the poor would be higher. Therefore ill-founded price regulations could hasten the day when many of the remaining supermarket companies will have abandoned the inner city. In situations where costs are higher in the inner city they must be covered. Short of a Government subsidy of some sort, such higher costs must ultimately be reflected in prices charged and services offered. This fact must be recognized in efforts to increase the availability of supermarkets in the inner city.

REGULATION OF CONDUCT

This study found that in a great many instances advertised items were out of stock or that prices had not been marked down in accordance with the advertisements. For all chainstores surveyed in Washington, 14 percent of items advertised were unavailable; in San Francisco, 6 percent of items advertised could not be found on the shelves. In both cities unavailability rates were substantially higher in lowincome areas.

The use of advertised "specials" is a key element in the competitive strategy of supermarkets. While many consumers trade at the same supermarket week after week, others comparison shop or can be persuaded by lower advertised prices to switch their loyalties to another store in order to save money. Since consumers cannot make price comparisons for the thousands of items in a supermarket, their attention is most likely to be drawn by specials advertised in newspapers or featured through displays within the store. A study of food store pricing states:

There are a few items such as coffee, flour, sugar, soap powders, and the like which each store feels must be in a majority of its "ads," since consumers are See below, pp. 26-32.