The failure of government-sponsored cartels and development of federal farm policy.

Hoffman, Elizabeth ; Libecap, Gary D. ; Gilligan, Thomas W. 等

"Even today, most important enduring monopolies or near monopolies in the United States rest upon government policies. The government's support is responsible for fixing agricultural prices above competitive levels, for the exclusive franchises of public utilities and radio and TV channels, for the single postal service--the list goes on and on." George J. Stigler [1993, 399]

I. INTRODUCTION

It is often believed that the coercive power of government can overcome the negotiation and enforcement problems that plague private cartels.(1) This claim is made in standard microeconomics text books, such as Hirshleifer's [1984, 262], and it is an integral part of the economic theory of regulation (Stigler

[1971], Posner [1974]). Regulation can have the same effect as a successful cartel--it raises prices above competitive levels. Indeed, Posner [1974, 344-46] argues that it may be less costly for industries with large numbers of firms to obtain supportive government regulation than to privately form a carte.(2) By capturing regulatory agencies, industry can use the state for its own purposes.(3)

Although broadly correct, the view that government sponsorship is an effective mechanism for the successful formation of cartels neglects the costs of organizing the industry into an effective lobby group and the costs that politicians face as they attempt to define and enforce total output levels and individual quotas. Especially in those cases where the industry is large and heterogeneous, and hence, most needing of government assistance, are these costs likely to be significant. Under these circumstances, the industry will lack a consensus on total production targets and firm quotas. Where the cuts must be large in order to reach price-setting goals, quota definition and enforcement will be controversial and political opposition will be intense. These conditions provide incentives for politicians to modify cartel goals and to seek alternative ways of raising industry prices and incomes. Accordingly, classic government-sponsored cartels may not emerge, even though the industry has sought state intervention to augment private collective action.

This point is supported by research in other settings where private negotiations among heterogeneous parties to address common pool problems have failed. The usual problem is disagreement over the distribution of costs and benefits, and the parties often turn to the state for a solution. Yet, as Johnson and Libecap [1982] have shown for fisheries and Libecap and Wiggins [19851 for oil field unitization, the distributional problems encountered in private contracting spill over to the political arena, causing delays and modifications of regulations. In the case of fisheries, regulatory authorities adopt policies that maintain status quo rankings of heterogeneous fishermen and expand the stock (through fish hatcheries, for instance), and politically controversial quotas to reduce harvest typically are not introduced until the fishery is seriously depleted. In the case of oil field unitization, politicians also are unable to resolve conflicts over the definition of unit shares or quotas, the same issue that blocks private agreement. As a result, government-imposed unitization generally does not occur until late in the life of the field after many of the common-pool losses have already been inflicted. Even then, the unitization rules leave many margins for dissipation uncontrolled.(4)

This paper argues that the organization costs facing a heterogeneous industry and the political costs facing politicians in responding to its demands for regulation must be given more attention to explain the actual policies that emerge and to predict when agency capture and government-sponsored cartel formation will be successful. The argument is illustrated by examining the development of New Deal agricultural regulation, where initial cartel efforts were gradually modified in response to the political opposition of certain farm groups. Through this modification, modern agricultural regulation emerged, which is quite different from what originally was proposed. The case of orange marketing agreements is used to demonstrate the problems of agency capture and cartel formation when there were conflicts within the industry over firm quotas and other regulatory details, even though the federal legislation and the administrative agency explicitly favored a cartel.

Today, agriculture is one of the most regulated sectors of the American economy. As Theodore Lowi [1979, 68] has described: "Agriculture is that field of American government where the distinction between public and private has come closest to being completely eliminated." The production and sale of almost every commodity is affected by some government policy through a complex mix of programs. These include price supports and associated government purchases of "excess" supplies, export subsidies, tariffs and quotas on imports, output controls through acreage reductions and marketing orders, and subsidies for inputs, such as irrigation water, electricity, and research and development. Major commodities like cotton, wheat, rice, peanuts, tobacco, wool, mohair, honey, milk, corn, barley, oats, rye, sorghum, soybeans, and sugar, as well as, specialty crops like fruits, nuts, and vegetables, are affected.(5)

Although federal agricultural policy includes controls on inputs, such as land, and outputs through marketing orders, the major emphasis has been on demand enhancement through purchases by the Commodity Credit Corporation and related agencies. Under price support programs, the Commodity Credit Corporation (CCC) sets a price floor for designated commodities. The CCC offers farmers a policy-determined price (loan rate) for their prospective crops. If the market price exceeds the loan rate, farmers sell their crops and pay off the loan from the CCC. If the loan rate exceeds the market price, farmers default on their loans, and the CCC, in effect, purchases their output. Supplemental payments to farmers, called deficiency payments, may be made if the loan price is below a target parity price for the crop. Purchases of excess commodities by the Commodity Credit Corporation or other federal agencies lead to the accumulation of stockpiles by the federal government that are kept off the market. Some of these stocks are distributed through the school lunch program, food for the poor and elderly, and foreign food aid. Other stocks are stored and, occasionally, destroyed.

The myriad complex of current federal farm policies is generally attributed to the New Deal. But New Deal agricultural policies began more simply, principally as government-sponsorship of cartels. When the Agricultural Adjustment Act of May 12, 1933 (the AAA) was being considered in Congress, wholesale agricultural prices had fallen by 51 percent since 1929.(6) As with industry under the National Industrial Recovery Act (NIRA), the adopted solution for agriculture was surplus removal and production and marketing controls through the assignment of quotas to farmers and shippers.(7) There is no indication from congressional debates and hearings on the Agricultural Adjustment Act of serious concern that the farm problem would resist this remedy.(8) Yet, as the 1930s progressed, quota disagreements, cheating, expansion through new entry or new production along unregulated margins, and general opposition among farm groups to production controls signalled the breakdown of government-administered cartels. Well-organized farm groups lobbied Congress and the Department of Agriculture for less onerous regulations that did not rely on production controls.(9) As the political costs of cartel enforcement rose, increasingly the federal government turned to alternative methods of raising farm prices, chiefly demand enhancement through the purchase of excess supplies, using general tax revenues. These policies avoided the need to more tightly limit output and define individual quotas. Given the political influence of agricultural groups, it is not surprising that agricultural policy drifted away from strict, government-administered cartels.(10)

II. CARTELS UNDER THE AGRICULTURAL ADJUSTMENT ACT

Agriculture was particularly hard hit by the Great Depression. The prices of many commodities had not recovered from the sharp fall in 1921, so that between 1919 and 1933, wholesale farm prices had fallen by 67 percent, whereas, over the same period nonagricultural wholesale prices had fallen by 45 percent. Moreover, the decline in agricultural prices was particularly severe after 1929.(11) The agricultural crisis of the 1920s brought a rise in militancy among farm organizations, such as the Farmers' National Relief Conference, the Farmers' Union, the National Grange, and the Farm Bureau Federation.(12) They lobbied Congress and the President for government intervention to raise prices and farm incomes. Although the emphasis in the 1920s had been on obtaining government assistance for private cooperative efforts to store excess stocks, in 1933 the demand was for much more direct action--cartel controls on production and market supplies, negotiated and enforced by the federal government.(13) The policy of output control was strongly supported by the new Secretary of Agriculture in the Roosevelt Administration, Henry A. Wallace.(14)

Indeed, the Agricultural Adjustment Act was the outcome of well-organized farm group lobbying, and the statute largely was drafted by Frederick P. Lee, Legislative Counsel for the Farm Bureau Federation.(15) The aim of the law was to raise agricultural prices to re-establish the relative purchasing power of farmers that had prevailed from 1909 to 1914. The operating assumption was that the farm problem was primarily the result of overproduction.(16) The Agricultural Adjustment Act called for farmers to enter into agreements with the Secretary of Agriculture to reduce their acreage in seven basic commodities--wheat, cotton, corn, rice, tobacco, hogs, and milk--in return for federal benefit payments to be derived from taxes levied upon the processing of the commodities for consumption.(17) Those farmers who did not take part would be ineligible for the benefit payments.

The close ties between early New Deal agricultural and industrial policies were stressed by a leading historian of agricultural policy, Murray Benedict [1953, 294]: "In both the NRA and the AAA emphasis was placed on the raising of prices through artificially induced scarcity." Indeed, 450 agricultural codes were transferred from the National Recovery Administration to the Agricultural Adjustment Administration between June 26, 1933 and December 6, 1933 for implementation.(18)

For major crops, output was to be reduced through acreage controls. Each year, the Secretary of Agriculture was to determine how much land should be removed from production for covered commodities to raise prices to target levels and to pay farmers for the idled acreage. A base acreage was to be established for each grower, and production quotas were to be determined by percentage reductions in the base.(19) For specialty crops, the Secretary of Agriculture could issue a marketing agreement if 50 percent of the shippers and two-thirds of the farmers in a region agreed to the provisions of the agreement. The marketing agreements authorized the Secretary to limit interstate shipments through weekly allotments to shippers that were enforced through revokable shipping licenses and fines for violation.(20) Between 1933 and 1935, sixty-one marketing agreements were approved by the Secretary of Agriculture for milk, oranges, grapefruit, dates, pecans, walnuts, olives, raisins, asparagus, and other commodities.(21)

Under the new cartel policies, the federal government aggressively began to reduce supplies. In June 1933, between 25 and 50 percent of the sown cotton crop was plowed up, and wheat and tobacco acreage was reduced.(22) A severe drought between 1933 and 1936 also helped to decrease supplies of cotton, corn, and wheat. But as output controls were implemented and further acreage reductions predicted, farmer unrest grew. Opposition mounted to production quotas, yields were increased on quota acreage through the substitution of capital and labor for land, and farmer participation in Agricultural Adjustment Act programs declined.(23) In addition, the dramatic actions taken by the Agricultural Adjustment Administration in 1933, such as the plow down of cotton acreage and the emergency hog slaughter, brought wide-spread criticism of the agency.(24) To more strictly enforce farm quotas, Congress passed the Bankhead Cotton Control Act, April 21, 1934, and the Kerr-Smith Act, June 28, 1934 for tobacco, taxing production from acreage beyond individual assignments.(25)

But tighter cartel controls were not the primary response of Congress and the Roosevelt administration to the political reaction to cartel enforcement. Farm groups lobbied the government to relax the quotas and to turn to alternative methods of raising agricultural prices.(26) A way that avoided the distributional problems of assigning, enforcing, and reducing farm quotas was to relax the cartel policy and have the federal government purchase surplus production. The Commodity Credit Corporation, which had been created by executive order on October 16, 1933, was already available to implement this policy shift.(27) Major government acquisitions to augment private demand would reduce the supply cutbacks needed to raise prices. This solution was new. Historically, the federal government had not bought agricultural goods to fix prices.

In September 1933, the Federal Emergency Relief Administration announced that it would buy $75,000,000 ($239 million in 1967 prices) of surplus commodities.(28) In October 1933, the Farm Credit Administration purchased sixteen million bushels of wheat, and the Commodity Credit Corporation raised the loan rates (prices it paid) for hogs, wheat, and cotton. The Federal Emergency Relief Administration followed with a purchase of hogs for relief purposes.(29) By the end of 1934, 48 percent of U.S. cotton production was either purchased by the CCC or pledged to the agency as collateral for loans under price-fixing arrangements. Stocks held by the CCC declined as production fell during the drought years of 1935 and 1936, but by 1938, the agency once again held 38 percent of the cotton crop and by 1939, had 12 percent of the corn crop and 23 percent of the wheat crop. The 1938 Agricultural Adjustment Act, which is the basis for much of current farm policy, specifically directed the Commodity Credit Corporation to make pricesupporting loans (purchases) whenever certain specified conditions arose. Tellingly, there were no such provisions in the 1933 law.(30) Political pressure forced CCC loan rates higher between 1934 and 1941, leading to the greater accumulation of wheat, cotton, and corn stocks.(31) The value of CCC loans to farmers, with crops held as collateral, rose from $260,000,000 in 1934 ($650 million in 1967 prices) to $457,000,000 in 1939 ($1.1 billion in 1967 prices).(32) By 1939, loans and other benefit payments from the federal government were as much as a quarter of total farm income.(33) These loans (purchases) to raise farm prices and incomes reflected the move away from the cartel policy adopted in 1933.

This summary of modifications of New Deal farm policy highlights some of the tensions encountered in government-sponsored cartels. It, however, is at too aggregate a level to make clear the distributional conflicts over quotas that forced changes in agricultural policy. To examine these issues, we provide a case study of orange marketing agreements in the 1930s, where disagreements within the industry prevented the establishment of the cartel authorized by the Agricultural Adjustment Act. One of the responses was government purchase of excess stocks for distribution through relief programs, although this practice was not as extensive as with other commodities.(34) The major response to an inability to establish a cartel to restrict orange shipments as planned, even with government support, was to adopt a less politically controversial regulatory arrangement.

Orange marketing agreements were put into place in California and Florida by the Secretary of Agriculture on December 18, 1933, among the first marketing agreements implemented under the Agricultural Adjustment Act. As with other commodity programs, there was optimism in the Department of Agriculture that shipping controls would succeed in raising prices. There was consensus within the industry for government regulation of shipments. Between 1930 and 1933, nominal orange prices had fallen by 75 percent, whereas the consumer price index had fallen by 22 percent, and growers saw intervention by the federal government as necessary for limiting the amounts placed on the market.(35) Unlike wheat, corn, or cotton, which were grown across vast areas of the country by millions of growers (1,208,368 wheat farms; 4,597,949 corn farms; and 1,986,726 cotton farms in 1930), oranges were grown in confined areas by a relatively smaller number of growers.(36) There were approximately 19,000 orange growers in both Florida and California.(37) Moreover, 90 percent of the California production was distributed through two existing cooperative organizations, the California Fruit Growers Exchange (75 percent) and the Mutual Orange Distributors (15 percent). About 25 percent of the Florida production was distributed through the Florida Citrus Exchange.(38) All three organizations were active proponents of federal marketing agreements, and the California Fruit Growers Exchange, in particular, developed close ties with the Agricultural Adjustment Administration.(39) Most production in California was concentrated within a radius of 90 miles around Los Angeles, and the Florida growing region was a rectangle of approximately 300 by 150 miles in the middle of the state.(40) Finally, since oranges were a perishable crop, inventories that could depress prices were less of a problem than for other commodities.

Despite these comparatively favorable conditions, a cohesive industry position in negotiations with the administrative agency did not emerge, nor did the marketing agreements succeed in restricting orange shipments as planned in December 1933. Although the 1933 marketing agreement was accepted in California, it was rejected in Florida because of conflicts over quotas. The marketing agreement was terminated in 1934. Between 1934 and 1937, two other marketing agreements were executed by the Secretary of Agriculture for Florida, but cancelled, before an acceptable arrangement could be devised in 1939. The final Florida marketing agreement did not involve weekly prorationing of orange shipments as used in California. Instead, it relied upon temporary shipping holidays and adjustable size and quality controls to limit interstate shipments. Additionally, national prorationing of orange shipments across the states, necessary for an interstate cartel, was not implemented as authorized by the Agricultural Adjustment Act. With these comparatively looser controls on shipments, orange prices did not rise to target parity levels.(41)

III. THE FAILURE TO CARTELIZE INTERSTATE ORANGE SHIPMENTS

The California and Florida Orange Industries

To appreciate the industry differences that led to conflicts over quotas and other aspects of the proposed cartel, it is necessary to summarize orange production conditions in the 1930s. California and Florida were by far the dominant producers of oranges, with California accounting for 67 percent of U.S. output in 1930-31 and Florida 32 percent.(42) Oranges from both regions competed as close substitutes in the fresh fruit market.(43) Until the late 1940s, there was no frozen concentrate or significant use of oranges in juice.(44) California produced two kinds of oranges: winter navels with a season of October to June and summer Valencias with a season from May through October.(45) Florida produced at least five varieties, all during the winter season: Parson Brown and Hamlin (October-December), Homosassa and Pineapple (January-March), and Valencia (April-June).(46) Florida growers tended to specialize in a certain variety, which often was determined by growing conditions. Storage possibilities at this time were limited, especially for Florida fruit. Because of climate conditions, Florida oranges did not store well on the tree and had to be harvested quickly in order to avoid fruit drop. In California because of relatively cool nights, oranges could be stored on the tree for up to two or three months.(47) Accordingly, all Florida oranges competed with California navels, whereas California Valencias generally did not compete directly with any other orange.

Most Florida growers and shippers were independents, with only about 25 percent of the state's production pooled and marketed through the Florida Citrus Exchange. Cooperatives, such as the Florida Citrus Exchange, pooled fruit during the season, and growers received the seasonal average price. This practice served to spread the risk of seasonal price fluctuation among growers, lower shipping costs if there were economies of scale in shipping, and improve marketing since known quantities and qualities of fruit could be delivered to particular destinations throughout the season.(48) Independent shippers, however, engaged in spot purchases of fruit from growers whenever harvest and market conditions warranted. Neither these growers or shippers were members of formal cooperatives. In California, approximately 90 percent of the orange production was pooled and marketed through either the California Fruit Growers Exchange or the Mutual Orange Distributors.

The variation in membership in formal pooling cooperatives between California and Florida was a major distinction between the two regions that, as we show, had important implications for government-sponsored cartels. The differences in pooling were due to much more heterogeneous fruit in Florida, which raised the costs of pooling, and sharply different subseasons and corresponding price expectations among Florida growers, which reduced the incentive to engage in seasonal pools.(49) The quality of California oranges, in contrast, was uniformly high because of favorable and consistent growing conditions. Moreover, since California oranges stored well on the tree, the California Fruit Growers Exchange prorated harvests across growers throughout the season, picking only a portion of each grower's crop at any time. This practice ensured that each grower's fruit was sold throughout the season so that no grower would differentially benefit or suffer from temporary price swings. This practice also served to enforce the cooperative's shipping restrictions for pooling.

In Florida, fruit was much less uniform, and because of limited storage possibilities harvests could not be prorated across the season to even grower price expectations. Hence early fruit was harvested and shipped in October and December; midseason fruit was shipped from January through March; and late-season fruit was shipped from April through June. This meant that Florida growers had specific subseasons with much narrower ranges of price expectations than did growers in California, who produced for the entire season.

Generally, oranges prices followed a U-shaped pattern across the season, high early in the season, low during the mid season, and high again late in the season. The mean prices per box for the three Florida subseasons for 1925-26 through 1932-33 were: early oranges--$4.34; midseason oranges--$3.81; and late in the season $4.89.(50) Accordingly, producers who specialized in early-season varieties had little incentive to pool across sub-seasons. Because fruit did not store well on the tree, these producers knew that their fruit would be harvested and sold at a time when prices were expected to be higher than later in the season. Moreover, they had no incentive to engage in activities that would smooth price fluctuations across the entire season. Such activities would only serve to lower their expected returns.

These conditions help to explain why seasonal pooling of fruit through formal cooperatives was much less common in Florida than in California.(51) Cooperative pooling through the California Fruit Growers Exchange provided a basis for regulating shipments under the federal marketing agreements in California, but pooling through the Florida Citrus Exchange was not extensive enough in Florida to play that role. An established pooling cooperative, like the California Fruit Grocers Exchange, became a ready-made vehicle for regulatory controls on shipments under the Agriculture Adjustment Act, since restrictions on deliveries could be imposed on the pooling organization and then prorated across the contributing growers and their shippers. The assignment and management of individual grower/shipper cartel quotas could be accomplished within the existing structure of the pool. Policing involved insuring that the pooling organization adhered to the quantities authorized by the Agricultural Adjustment Administration. If a single or at least a small number of pooling organizations existed in each state, then nationwide shipping controls would have involved assigning quotas to each organization and monitoring compliance.

Because formal cooperatives reduced the transactions costs of implementing and monitoring the marketing agreements, the Agricultural Adjustment Administration sought to promote membership in them through the design of the marketing agreements. This goal was particularly aimed at increasing membership in the Florida Citrus Exchange, but the policy was opposed by independent growers and shippers.

The Orange Marketing Agreements

We now turn to the intraindustry conflicts over shipping quotas that led to the breakdown of federal efforts to form a cartel for oranges in the 1930s. This failure occurred despite the fact that growers in both California and Florida had similar objectives for federal marketing agreements. With increased output from plantings made in the 1920s and falling prices due to the depression, there was consensus in the industry that federal controls on shipments were necessary to raise prices.(52) The distributional consequences of the quota policies adopted by the Agricultural Adjustment Administration at the behest of the large cooperatives, however, led to conflict within the industry over the proposed cartel.

Meetings were held in Washington D.C. on July 20 and September 7-9, 1933 and January 6 and June 18, 1934 between the Department of Agriculture and the industry to draft the marketing agreements and to set up a national prorationing scheme. At the first meeting, California and Texas each had nine delegates, Arizona one, but Florida had thirty-seven because of differences in opinion within the state as to the nature of the proposed regulations.(53) The California delegates argued for national prorationing with fixed state quotas and a national price stabilization plan (a national cartel). They offered a draft marketing agreement for adoption by the Agricultural Adjustment Administration.(54) The California proposal was based upon a 1932 private arrangement between the California Fruit Growers Exchange and the Mutual Orange Distributors to prorate weekly shipments of Valencia oranges within the state and the California Prorate Act, enacted on June 5, 1933, which had provisions for marketing orders that were very similar to those proposed in the Agricultural Adjustment Act.(55) By contrast, the Florida industry presented at least two competing draft marketing agreements, one supported by the Florida Citrus Exchange and similar to that proposed by the California Fruit Growers Exchange, and one backed by the Florida Citrus Growers Clearing House Association, which represented many of the independent growers and shippers in Florida.

The Agricultural Adjustment Administration supported and ultimately adopted the draft marketing agreements proposed by the California Fruit Growers Exchange and the Florida Citrus Exchange. The Special Crops division of the agency, which was responsible for marketing agreements, was headed first by Howard Tolley and then by W. R. Wellman, both of whom had close ties to the California Fruit Growers Exchange.(56) The marketing agreements called for the weekly prorationing of orange shipments among shippers within each state whose quotas would be based on season-long contracts for fruit and set by administrative committees in each state.(57) Quotas were to be determined by a "prorate base" assigned to each shipper on the basis of the amount of fruit held under contract with growers at the beginning of the season.(58) The prorate base was the shipper's fraction of total seasonal orange shipments from the state, and multiplying it times the authorized weekly total fixed each shipper's weekly quota. Obtaining a prorate base would be no problem for shippers who were members of formal cooperatives, since season-long contracts were an integral part of the pooling agreements administered by these organizations. Independent shippers, who engaged in periodic spot purchases of fruit, however, would not have had fruit under contract at the beginning of the season when the prorate base was defined for each shipper. Accordingly, they would have had a zero prorate base and, hence, would not have qualified for a weekly quota under the provisions of the marketing agreement. The adoption of this quota arrangement was an effort to require growers and shippers in Florida to join the Florida Citrus Exchange(59) Officials of the Department of Agriculture argued that the success of the marketing agreement depended upon broad participation in cooperative shipping pools in Florida, as was practiced in California.(60)

Not only did the Department of Agriculture adopt a quota rule to encourage membership in the Florida Citrus Exchange, but the Florida Citrus Exchange was given a majority of the positions on the state administrative committee. Under the marketing agreement, Secretary of Agriculture Henry A. Wallace appointed the members of the Florida Control Committee that was set up to determine weekly shipping levels and to assign shipping quotas. Most of those selected were from the Florida Citrus Exchange. The California marketing agreement allowed for the election of members of the administrative committees for that region.(61)

Independent shippers and growers within the Florida Citrus Clearing House Association, who attended the Washington meetings to draft the marketing agreements, understood the effect of the prorationing rule in requiring membership in pooling cooperatives. The department recommended that growers who were worried that their shippers would not have quotas under the prorationing rule, link up with established shippers who did.(62) During negotiations in the fall of 1933, the Florida Citrus Growers Clearing House Association demanded that the Agricultural Adjustment Administration modify its proposed marketing agreement for Florida, because it would force independent shippers out of business. The agency refused, arguing that the agreement could be amended later if necessary. But, while ratification of the marketing agreement required concurrence of 50 percent of the shippers and two-thirds of the growers, amendments required two-third's concurrence of both groups. The Florida Citrus Growers Clearing House Association also circulated a competing marketing agreement, but it was not adopted by the Secretary of Agriculture.(63)

There was general agreement in Florida that some form of federal regulation was desirable. The issue was the form regulation would take. For example, James. C. Morton, Vice President of the Florida Citrus Growers Clearing House Association, wrote to Agricultural Secretary Henry A. Wallace, November 27, 1933 to protest "the inequitable restrictions of the prorate clauses in the Agreement." Nevertheless, he called for modification of the proposed agreement, not its abandonment.(64)

Instead of prorationing rules, the independents favored the use of shipping holidays and quality restrictions to regulate shipments. Shipping holidays could block all deliveries from the state for a specified period of time to alleviate temporary market gluts. Size and quality standards could be set to deny shipment of fruit that fell below the standard, and the standard could be adjusted from time to time to provide flexible restraints. Quality standards also provided some industry-wide public goods in maintaining product reputation.(65) Enforcement for both policies would involve inspection and monitoring of all deliveries across state lines, rather than insuring individual quota compliance, as was necessary under prorationing.

Because shipping holidays and quality standards generally applied across the board, the distributional consequences were less severe than those associated with the proposed allocation of quotas under the marketing order proposed by the Agricultural Adjustment Administration. Quality constraints did harm marginal growers of low-quality fruit, but those growers appeared not to be sufficiently influential to block them. Shipping holidays typically were short enough so as not to cause serious losses, but since lengthy storage was not possible, these shipping interruptions could raise orange prices. Moreover, these alternatives did not require membership in organized cooperatives. An example of broad-based support for shipping holidays in Florida is the February 6, 1933 call by the Florida Citrus Exchange, the Florida Citrus Growers Clearing House Association, and other shippers for a six-day shipping holiday in order to raise prices.(66)

Although the California marketing agreement was implemented routinely in December 1933, opposition in Florida to the prorationing rule and to the Florida Control Committee appointed by the Secretary of Agriculture was immediate. The Florida marketing agreement was challenged in Federal District Court by two shippers, Hillsborough Packing and Lake Fern Groves (Yarnell v Hillsborough Packing Co., 70 F.2nd 435). An injunction was issued against prorationing on January 18, 1934 by Judge Alexander Akerman in the southern district in Tampa, who ruled that the marketing order under the Secretary of Agriculture was unconstitutional. Prorationing controls by the Florida Control Committee were temporarily halted. Although the injunction was removed in February 10, 1934 by an appellate court and the ruling was reversed by the Fifth U.S. Circuit Court of Appeals, April 14, 1934, the injunction was applied at the height of the Florida orange season, and it raised uncertainty about the future of prorationing.(67)

Both shippers objected to the design of the prorationing rule, but for different reasons. Lake Fern Groves shipped very high quality fruit, and hence preferred reliance on grade and size restrictions to control shipments instead of volume restrictions through prorationing. Hillsborough, on the other hand, engaged in periodic cash purchases under short-term contracts with growers rather than participating in a pool. It was precisely this kind of shipper that would be disadvantaged by a quota rule that assigned shipments based upon long-term contracts struck at the start of the season.(68) The prorationing rule remained so controversial that the first marketing agreement for Florida oranges was terminated in August 1934.

Throughout the summer and fall of 1934, members of the Florida Citrus Exchange and the Florida Citrus Growers Clearing House Association corresponded with officials of the Agricultural Adjustment Administration regarding the redrafting of the marketing agreement. Each side wanted its position considered and to be assured of adequate representation on the drafting committee.(69) A second marketing agreement was initiated December 1934. There were two minor modifications in the order, but the Department of Agriculture continued to maintain the basic prorationing framework.(70) Past shipments were to be given greater emphasis in designing quotas, but the weights assigned to fruit controlled through long-term contracts and past shipments were left to the Control Committee. This naturally became a point of contention given the makeup of the Control Committee.(71) Independent shippers and growers objected to the lack of information made available on the calculation and assignment of quotas.(72) Throughout 1934 and 1935 there were conflicts over the membership of the Control Committee and demands for access to its records in prorationing allocations.(73) In the face of continued opposition, the second marketing agreement for Florida oranges was terminated July 15, 1935.(74)

A third marketing order was not put into place until May 1936, ten months after the termination of the second order and after the 1935-36 shipping season had passed. As before the Department of Agriculture maintained prorationing of orange shipments as the primary method of regulation. The proration rule continued to emphasize fruit contracted for or purchased at the beginning of the season, but it placed more weight on past shipments, extending the number of years of past shipments to be considered from two to three. Grade and size restrictions using federal specifications also were continued. Nevertheless, as with the earlier marketing orders, conflicts continued over the assignment of quotas and Department efforts to force membership in cooperative pools. Court challenges of the prorationing rules brought conflicting opinions by Federal District Judge Holland in Miami, who sustained the marketing agreement in February 1937, and Judge Akerman in Tampa, who issued an injunction against it in March 1937.(75) The third marketing order for Florida oranges was terminated July 31, 1937.

Over a year of negotiations between the Agricultural Adjustment Administration and the Florida industry was necessary before a final and successful marketing order was implemented February 22,1939. The new marketing order contained no quota rules or prorationing provisions. Regulation, instead, focused upon uniform grade and size restrictions and shipping holidays, the framework originally demanded by independents in the Florida Citrus Growers Clearing House Association. Neither of these regulations required individual quotas or membership in agricultural cooperatives. Hence by 1939, the regulation of orange shipments through formal agricultural cooperatives as envisioned by enthusiastic officials of the Agricultural Adjustment Administration in 1933 had been discarded. Given this history, the marketing agreements had little chance of raising orange prices to their target levels during the 1930s.(76)

IV. CONCLUDING REMARKS

Although economists have long recognized that private cartels are difficult to sustain, they generally have been too sanguine in their assessment of the potential for government-built or assisted cartels. The coercive power of the state has seemed to be a natural remedy for forcing agreement and compliance with output reductions and individual quotas when no private consensus can be reached. This view, however' neglects the costs faced by politicians and bureaucrats when the industry is heterogeneous and there is disagreement over quota policies.(77) Yet, these are precisely the conditions under which government assistance is asserted to be most necessary. Under these circumstances, a government-sponsored cartel may be no more successful in achieving production restrictions than were private cartels. The advantage of government efforts, as federal agricultural programs make clear, is that there is a much longer menu of alternatives for raising prices, such as government purchases to enhance demand.

As noted in the beginning of the paper, agriculture is perhaps the most heavily regulated sector of the American economy, a process that largely began with the Agricultural Adjustment Act of 1933. Although the focus of that law was on production control and marketing restrictions, political opposition to output controls by various farm groups brought a shift in emphasis to demand enhancement with government acquisitions of "excess" stocks. Gradually in the 1930s, through the purchase of commodities by the Commodity Credit Corporation and other similar agencies and their distribution through relief, and later, through subsidized exports, food aid, and school lunch programs, the modern character of federal agricultural programs took shape. Government purchases were much more acceptable to influential farm groups than were production and shipping controls in the effort to raise farm prices and incomes. As a broad, generally unorganized group, taxpayers increasingly absorbed many of the costs of federal farm policy.

The case of orange marketing agreements illustrates the distributional conflicts over quotas that can be encountered in attempting to establish government-sponsored cartels. Competing views regarding quota design in the proposed cartel prevented a cohesive industry position in negotiating the marketing agreements with the Agricultural Adjustment Administration. Further, the strong political reaction in Florida to the quotas that were adopted by the agency forced repeated modification of the marketing agreements over six years until an acceptable arrangement could be devised. The final marketing agreement, however, had little resemblance to the nationwide cartel outlined in 1933 under the Agricultural Adjustment Act.

(1.) Kolko [1965, 2-29], for example, argues that the desire for government cartel enforcement was behind the support of the railroads for the Interstate Commerce Act of 1887. Tests of interest group support for the act are provided by Gilligan, Marshall, and Weingast 11989,19901. Stigler [1964, 46 8] describes the policing problem facing cartels and how government can assist in cartel stability. The role of legal cartels in international trade is discussed by Jacquemin, Nambu, and Dewez [1981], Davidson [1984], and Audretsch [1989].

(2.) For the most straightforward discussions of the role of government in the economic theory of regulation, see Stigler [1971, 5; 1974] and Posner [1974, 34446].

(3.) The capture model of regulation and the broader arguments of interest group politics behind regulation are outlined in Stigler [1971], Peltzman [1976], Becker [1983], Wilson [1980], Joskow and Noll [1981], Kalt and Zupan [1984], and Gilligan, Marshall, and Weingast [1989; 1990].

(4.) In both fisheries and oil fields, delaying regulation until rent dissipation is extensive raises the returns to agreement and reduces political opposition. By waiting until dissipation is serious, information asymmetries regarding the valuation of individual shares and other distributional issues become less significant, compared with the costs of not reaching agreement.

(5.) For an excellent discussion of federal agricultural policy, see Del Gardner [1993]. Bruce Gardner [1981, 21] outlines the three main types of intervention--government purchases of excess stocks at policy-determined prices; supply controls through acreage reductions or controls on supplies placed in the market; and income payments equal to the difference between the income from the target price and the market price, called deficiency payments. Agricultural policy and some of the politics involved are discussed by Knutson, Penn, and Boehm [1983].

(6.) By contrast, wholesale industrial prices had fallen by 22 percent (U.S. Department of Commerce [1975,199]). The relative fall in agricultural prices and the rise of lobby pressure for government intervention is discussed by Benedict [1953, 2771.

(7.) 48 U.S. Stat. 31. For discussion of the Agricultural Adjustment Act, see Murphy [19551, Shover [1965], and Perkins [1965, 1969]. Sunstein [1987, 439] discusses cartel formation and enforcement under the National Industrial Recovery Act and the Agricultural Adjustment Act.

(8.) Perkins [1969, 53] notes some skepticism during congressional debate in the House about the ability of the act to reduce production, but these concerns were quite limited.

(9.) The unusual political influence of farm groups in molding government policy is discussed by Lowi [1979, 69-76]. Many of the same groups that originally supported the government cartel policy came to oppose it. This opposition reflects two conditions. One is that direct government administration of cartels had not existed before and farm organizations probably did not foresee the distributional consequences of defining and enforcing quotas. Second, given falling demand because of the depression, they also did not see how severe the cutbacks had to be if cartels alone were to raise commodity prices.

(10.) Peltzman [1976] and Becker 11983] describe the role of politicians as brokers responding to the demands of influential interest groups. Political influence is relative, and no group gets all that it wants. Farm groups, such as the Farmers' Union and the Farm Bureau Federation, have been effective lobbyists, and given the over-representation of rural states in the Senate, agriculture has been successful m obtaining transfers from general taxpayers. See Gardner [1993] for discussion.

(11.) U.S. Department of Commerce [1975, 199-200], Perkins [1969, 11].

(12.) For discussion of the agricultural crisis of the 1920s, see Perkins [1969, 10-48] and Hoffman and Libecap [1991].

(13.) Perkins [1969, 27-32] describes the emphasis in the 1920s on cooperative output reductions with some government help. The development in the late 1920s of the Domestic Allotment Plan by the USDA represented a move toward more explicit federal actions to limit output. These concepts were incorporated into the Agricultural Adjustment Act.

(14.) Nourse, Davis, and Black [1937, 20]. The literature of the time is clear on production control as the solution to the farm problem. For example, the American Institute of Cooperation, which published American Cooperation, formed a roundtable committee m 1932 on production control, and the journal published articles in the early 1930s on the legality, necessity, and need for production control (Hulbert [1932], Ezekiel [1933]). Tellingly, by 1938, the association was publishing articles on government purchases and the problem of cooperative solutions (Brands [1938], Stedman [1938]). Perkins [1969, 43, 81-6] discusses production control as the major tool for farm relief and describes Secretary Wallace's strong commitment to it. This policy led to conflict with George Peek, the first administrator of the Agricultural Adjustment Administration, who preferred controls on marketing and export stimulation. Peek was replaced by Chester Davis, who was committed to production control. See also, Irons [1982 111-55], Schultz [1949, 41], and B. Gardner [1987, 55].

(15.) Murphy [1955, 160], Shover [1965], and Perkins [1969, 37-44].

(16.) Cochrane and Ryan [1976, 12] describe the farm problem of the 1920s and early 1930s as one of chronic, excess productive capacity.

(17.) These seven were later augmented by beef, dairy cattle, peanuts, barley, flax, grain sorghum, sugar beets, sugar cane, and potatoes under the Jones-Connally Act of 1934 (48 U.S. Stat. 528). Breimeyer 11983, 3431 discusses the paid acreage reductions under the Agricultural Adjustment Act.

(18.) Nourse 11935, 311. Nourse [1935, 24-49] also makes clear that the various aspects of the Agricultural Adjustment Act were designed to reduce market supplies in order to raise prices and farm incomes.

(19.) Benedict [1953, 303].

(20.) Marketing agreements also took other forms for different specialty crops, such as quality controls and shipping holidays. The original agreements for a variety of fruits, nuts and vegetables, were voluntary. In the face of noncompliance, they were supplemented with marketing orders issued by the Secretary of Agriculture as authorized by amendments to the Agricultural Adjustment Act, August 24, 1935. These marketing orders were binding on all growers and interstate shippers of the commodity covered by the agreement (Nourse, Davis, and Black [1937, 231-34]).

(21.) Nourse [1935, 53].

(22.) Perkins [1969, 103, 124].

(23.) Benedict [1953, 311] points out that farmer participation in acreage controls under the Agricultural Adjustment Act fell after the first planting season.

(24.) Perkins [1969, 103, 140].

(25.) Benedict [1953, 304]. 48 U.S. Stat, 31; 48 U.S. Stat. 598.

(26.) The literature is uniform in concluding that the output and market controls of the Agricultural Adjustment Act were unsuccessful. Schultz [1949, 143] points out that although corn acreage fell by 8 percent between 1937 and 1939, output grew by 17 percent. A severe drought in 1933 helped to reduce wheat production that year. For assessments, see Nourse, Davis and Black [1937, 289-320], Benedict [1953, 313], and Benedict [1955, 443-44]. Stricter production controls were achieved only in tobacco and peanuts (B. Gardner [1987, 21].

(27.) Perkins [1969, 168, 224].

(28.) Perkins [1965, 221]; U.S. Department of Commerce [1975, 199].

(29.) Perkins [1965, 226-28]. The October 1933 wheat purchases were about 3 percent of total U.S. production in 1933 (U.S. Department of Commerce [1975, 5111, a small beginning that was to grow.

(30.) The Agricultural Adjustment Act of February 16, 1938, 52 U.S. Stat. 31, followed the 1933 act. The Commodity Credit Corporation, for example, was to make price-supporting loans (purchases) for cotton and wheat; if the market price were below 52 percent of the parity price or if production was expected to exceed domestic consumption and export demand (apparently at the parity price), then the corporation was to make non-recourse loans to farmers that could be defaulted on if the market price remained below the loan rate. The buildup of stocks and repeated purchases by the Commodity Credit Corporation were justified by the Ever-Normal Granary policy adopted by Secretary Wallace in June 1934. See Breimeyer [1983, 346]. For discussion of other late New Deal programs, see Cochrane and Ryan [1976, 132-64].

(31.) Benedict [1953, 333, 376-78].

(32.) The quantities pledged to or purchased by the CCC are from U.S. Department of Agriculture [1941, 20] and CCC loan amounts are from U.S. Department of Commerce [1975, 488]. Price indices from U.S. Department of Commerce [1975, 199]. Selected inventories as a share of that year's production are:

 cotton corn wheat
1956 51% 20% 95%
1957 46 24 86
1958 9 28 57
1959 7 25 103
1960 35 27 88

Inventory data are from The Report of the President of the Community Credit Corporation U.S. Dept. of Agriculture [1956, 3; 1957, 2-3;1958, 3-4; 1959, 3-4; 1960, 3]. Annual production data are from U.S. Department of Commerce [1975. 510-511].

(33.) Schultz [1949, 154]. Nourse, Davis, and Black [1937, 285] suggest that one-fourth of the increase in farm income in 1933 was due to transfer payments two-thirds in 1934, and one-half in 1935. See also Rucker and Alston [1987]. As the CCC loan program became the centerpiece of the federal farm program, Murray Benedict [1953, 389] commented: "The Commodity Credit Corporation's activities came to have a second purpose which was not compatible with its stabilization function. This was the function of maintaining prices continuously above the free-market levels, rather. than merely that of ironing out the effects of ups and downs of production and control." He also noted: "That production control would not be highly effective,...was not, of course, in their [USDA officials] thinking m the early years." Even so, government policies failed to bring agricultural prices to their parity levels. By 1940 wholesale prices for nonfarm goods reached 91 percent of their 1929 levels, however, agricultural prices remained at 65 percent of those in 1929. Further, through 1940 the ratio of agricultural prices to general prices remained well below those reached during the parity period 1909 to 1914. u.s. Department of Commerce [1975, 200]. For 1909-1914 the ratio of farm wholesale prices to all wholesale prices averaged 1.04; in 1929, the ratio was 1.10; m 1933, it was .78 and in 1940, it was .86.

(34.) Major commodities like wheat could be stored for some time and justified through the Ever-Normal Granary policy. This kept supplies off of the market. Oranges could not be stored for long periods, and government distribution through relief programs potentially depressed market prices. For discussion of purchases of Florida oranges, for example, by the Federal Surplus Commodities Corporation, see the Florida Citrus Inspection Bureau [1938, 157, 169].

(35.) Manthy [1978, 47-52], u.s. Department of Commerce [1975, 211].

(36.) U.S. Department of Commerce [1930, Agriculture General Report, Vol. 4, 730, 738, 817].

(37.) U.S. Department of Commerce [1930, 561-65, 720-25].

(38.) Hopkins [1960, 5], Spurlock [1943, 4].

(39.) The close ties between the California Fault Grocers Exchange and the Agricultural Adjustment Administration are outlined in Hoffman and Libecap [1994]. For discussion of the active role of the California Fruit Grocers Exchange and the Mutual Orange Distributors in lobbying for federal marketing agreements, see Citrograph [April 1933, 161, 167].

(40.) Citrus Industry [May 1934, 5].

(41.) Tighter prorationing limits in California and the use of shipping holidays in Florida appear to have moderated price fluctuations in the 1930s compared to those that existed in the 1920s. See Hoffman and Libecap [1994]. They point out that cartel success would have been difficult in any event, even had Florida responded in the same way as California to the marketing agreements. There were other problems caused by falling incomes and entry that would have plagued the orange cartel. As reported in the U.S. Department of Commerce [1975, 225] real personal income in the U.S. fell by 28 percent between 1929 and 1933, and such shifts in demand would have forced recalculation of individual shipper and state quotas. Quota negotiations and enforcement are difficult enough as it is with out having to deal with demand shifts. For discussion of quota problems in another context, see Johnson and Libecap [1982].

(42.) Shuler and Townsend [1948, 7].

(43.) See U.S. Department of Agriculture [1938, 180-81, 244, 245] for New York and Chicago orange prices and for shipment data to various markets. See also Thompson [1938, 3, 26-7] for discussion of the intense competition between the two states and their relative shipments to particular markets. Hoffman and Libecap [1994] report differences in the log of weekly Florida and California orange prices in New York City for the 1926-D and 1927-28 seasons from the New York Times. The differences trend toward zero, as would be the case if the oranges were close substitutes.

(44.) Thompson [1938, 28-9], Reuther, Webber, and Batchelor [1967, 36].

(45.) In 1936, 60 percent of California acreage was in Valencias and 40 percent was in navels. Thompson 11938, 3-7]

(46.) For discussion of orange types, their seasons and production, see Reuther, Webber, and Batchelor [1967, 66, 74], Shuler and Townsend [1948, 9-11], and Thompson [1938, 71.

(47.) Reuther, Webber, and Batchelor [1967, 437-84] and Webber and Batchelor [1943, 82].

(48.) There were no futures markets in fresh oranges at this time, so pooling provided a means of spreading the risk of price fluctuation. See Hoffman [1932, 54-5].

(49.) Ziegler and Wolfe [1975, 219-29] discuss differences in orange qualities in Florida.

(50.) The mean prices were calculated from monthly data from the New York auction market as reported in United States Department of Agriculture [1934, 516 517; 1940, 215, 216]. They are for the leading months in each sub-season to avoid transition months between sub-seasons.

(51.) Cooperatives also had higher enforcement costs in Florida than in California, since truck shipments increasingly were more of an option for Florida growers than those in California, where most shipments were by rail. With truck shipments, cooperative rules could more easily be violated, whereas rail shipments could be monitored at relatively lower cost. While 11 percent of the Florida crop was shipped in small lots by truck in 1931, by the 1940-41 season some 24 percent went by truck. See Citrus Industry [January 1933. 6] and Joubert [1943. 31].

(52.) Citrus Industry [September 1933, 25].

(53.) Citrus Leaves [August 1933, 20], Citrus Industry [March 1934, 26].

(54.) Nourse [1935, 133, 159], Citrus Industry [August 1933, 10, 14; October 1933, 10], Citrus Leaves [February 1934, 4].

(55.) Thompson [1938, 39]. With cheating by some growers and shippers and the onslaught of the Great Depression, the private Valencia agreement did not succeed in raising orange prices, but it provided a prototype for the marketing agreements adopted by the Department of Agriculture (Citrograph [September 1933, 301]). The California Prorate Act included provisions for industry committees to determine weekly prorationing quotas, voting procedures to implement regulation, and revokable shipping certificates for shippers (Citrus Leaves [April 1933, 5-7; July 1933, 3, 4, 14-20]).

(56.) both Howard Tolley and H. R. Wellman were affiliated with the Giannini Foundation at the University of California at Berkeley that worked closely with the California Fruit Growers Exchange and other California agricultural cooperatives.

(57.) Citrograph [September 1933, 301].

(58.) Shippers generally paid 20 percent down to secure the contract (Ockey [1936, 34, 37], Citrus Leaves [October 1933, 3, 4, 11-20; January 1, 1934, 1, 2, 16].

(59.) U.S. National Archives, Record Group 145, Agricultural Adjustment Administration, Central Correspondence File, Box 362, letters from James C. Morton, Florida Citrus Growers Clearing House Association to Henry A. Wallace, November 27 1933, December 8 1933; telegraph, December 10, 1933 from James C. Morton, Florida Citrus Growers Clearing House Association to J. W. Tapp, Agricultural Adjustment Administration, letter, December 19, 1933 from A. E. Fowler, Florida Control Committee to W. G. Meal, Agricultural Adjustment Administration, December 19, 1933, with the Florida Marketing Agreement attached.

(60.) Since the 1920s, the Department of Agriculture had assisted cooperatives in marketing their crops and in controlling supplies through stockpiles and exports. These actions to promote farmer cooperatives and raise prices were promoted by a series of laws enacted or considered during the 1920s: the Capper-Volstead Act of 1922, the Cooperative Marketing Act of 1926, the Agricultural Marketing Act of 1929, and the McNary-Haugen bills of 1924-1928 (42 U.S. Stat. 388; 44 U.S. Stat. 802; 46 U.S. Stat. 11). See Hoffman and Libecap [1991] for discussion.

(61.) Citrus Industry [December 1933, 7, 10], Citrus Leaves [[October 1933, 3, 4, 11-20; January 1934, 1-2, 16].

(62.) U.S. National Archives, Record Group 145, Agricultural Adjustment Administration, Central Correspondence File, box 362, telegrams, December 10, 1933 and letters December 12, 1933 from James C. Morton, Florida Citrus Growers Clearing House Association to J. W. Tapp, Agricultural Adjustment Administration, and R. G. Tugwell, USDA; letter December 27, 1933 from thirteen growers to Henry Wallace, USDA; letter December 28,1933 from A. M. Prevatt, a Florida grower to Henry Wallace, USDA; letter from O. G. Strauss of the Florida Control Committee to Jasper Wolfe, a Florida shipper, March 22, 1934.

(63.) U.S. National Archives, Agricultural Adjustment Administration, Central Correspondence File, "Proposed Amendments, California Arizona Agreement," November 9,1933. U.S. National Archives, Record Group 145, Agricultural Adjustment Administration, Central Correspondence File, Box 362, telegrams, December 10,1933 and letters December 12,1933 from James C. Morton, Florida Citrus Growers Clearing House Association to J. W. Tapp, Agricultural Adjustment Administration, and R.G. Tugwell, U.S. Department of Agriculture; letter December 27,1933 from thirteen growers to Secretary Henry Wallace, U.S. Department of Agriculture; letter December 28,1933 from A. M. Prevatt, a Florida grower to Secretary Henry Wallace, U.S. Department of Agriculture; letter from 0. G. Strauss of the Florida Control Committee to Jasper Wolfe, a Florida shipper, March 22,1934.

(64.) National Archives, Record Group 145, Agricultural Adjustment Administration, Central Correspondence Files, Box 362.

(65.) with more heterogeneous fruit, reputation was a particular concern for Florida growers with respect to their California competitors. Because Florida oranges often had traces of green in their skins, unlike the more uniformly golden California Navels, fruit was often dyed in Florida. See Florida Citrus Inspection Bureau [1938, 157] for data on "color-added" oranges. As with any restriction, controls based on shipping holidays and quality standards would have distributional effects. Those growers who had planned to ship their crops at the time of a shipping holiday would suffer. Nevertheless, shipping holidays had much broader support among Florida growers and shippers than did prorationing.

(66.) Citrus Industry [February 1933, 5]. Growers in both California and Florida also pushed for marketing programs to expand total demand for oranges and upon purchases by the Federal Surplus Commodities Corporation to help reduce total supplies (Citrus Industry [November 1936, 5]). These programs were popular because neither required industry agreement on quota allocations, which had important distributional implications.

(67.) The constitutional issues raised by Judge Akerman and the hostility to the Agricultural Adjustment Act are discussed in Irons [1982, 142-49].

(68.) Citrus Industry [February 1934, 10], and U.S. National Archives, Record Group 145, Agricultural Adjustment Administration Central Correspondence File, Box 362, letter, January 22, 1934 from J. A. Yarnell of the Florida Control Commission to P.R. Porter, USDA; letter, January 24, 1934 from W. G. Meal, USDA, to O. G. Strauss, Florida Control Commission; letter, January 29, 1934 from P. R. Taylor, USDA, to J. H. Treadwell, a Florida grower-letter from Rex Tugwell, USDA, to U.S. Attorney General, February 2, 1934, memo for Mr. Arthur Bachrach from W. G. Meal, USDA, February 15, 1934.

(69.) For example, see May 15, 1934 letter to Porter R. Taylor, General Crops Section, Agricultural Adjustment Administration, from James Harrison of the Clearing House Association and May 14, 1934 letter to W. G. Meal and A. W. McKay, General Crops Section, Agricultural Adjustment Administration from O. G. Strauss, Secretary of the Florida Control Committee and aligned with the Florida Citrus Exchange. National Archives, Record Group 145, Agricultural Adjustment Administration, Central Correspondence Files, Box 362.

(70.) See the draft, Florida Citrus Agreement, March 10,1936, National Archives, Record Group 145, Agricultural Adjustment Administration, Central Correspondence Files, Box 362.

(71.) Citrus Leaves [November 1934, 6].

(72.) Citrus Industry [September 1934, 25; January 1935, 8], Citrus Leaves [November 1934, 6]. U.S. National Archives, Record Group 145, Agricultural Adjustment Administration, Central Correspondence File, Box 12, letters to Henry A. Wallace from James C. Morton, Florida Citrus Growers Clearing House Association, November 10, 1934 and November 27, 1934; letter from P. R. Taylor, USDA to C. M. Brown, a California grower, November 14, 1934; letter from Donald J. Nicholson, a Florida grower to Henry Wallace, November 20, 1934.

(73.) U.S. National Archives, Record Group 145, Agricultural Adjustment Administration, Central Correspondence File, Box 362, letter from A. W. McKay, Agricultural Adjustment Administration, to C. L. Bundy, a Florida grower, November 1,1934; Box 12, letter November 10, 1934 from James C. Morton, Florida Citrus Growers Clearing House Association, to Henry Wallace, November 10,1934, Box 363 November 27 1934 letter to Henry Wallace from James C. Morton, Florida Citrus Growers Clearing House Association; letter to C. M. Brown, California grower from P. R. Taylor, November 14, 1934.

(74.) In the meantime, state legislation creating a Florida Citrus Commission and authorizing shipment regulation based on quality and size standards was implemented. Florida Citrus Inspection Bureau [1936, 5-53]. The Florida Citrus Commission, named by the Governor, was created to take the place of the controversial federal Control Commission, named by the Secretary of Agriculture (Citrus Leaves [April 1936, 1; June 1936, 3]).

(75.) Citrus Leaves [May 1937, 9]. U.S. National Archives, Record Group 145, Agricultural Adjustment Administration, Central Correspondence File, Box 257, Florida Citrus Exchange Bulletin, January 29, 1937 to all district and association managers; letter, March 27, 1937, from Henry A. Wallace, U.S. Department of Agriculture, to L. P. Kirkland, Florida Citrus Control Committee; press release, U.S. Department of Agriculture, March 27, 1937.

(76.) As Hoffman and Libecap 11994] show, through the continued prorationing of California orange shipments and periodic prorationing of Florida shipments, along with the use of shipping holidays, the path of prices smoothed after 1933. The close relationship between the Agricultural Adjustment Administration and the California Fruit Growers Exchange helps to explain why California continued to comply with federal cartel restrictions in the face of repeated noncompliance by many Florida shippers. The marketing agreements provided federal enforcement of California regulations, and the California industry expected that the department would eventually force Florida into compliance.

(77.) Political influence, of course, is key. For example, as described by Libecap and Johnson [1980], members of the Navajo Tribe objected to forced livestock reductions on the reservation under the New Deal Indian policies administered by John Collier. They lacked, however, the political influence to block the reductions.

REFERENCES

Audretsch, David B. "Legalized Cartels in West Germany." The Antitrust Bulletin, Fall 1989, 579-600.

Becker, Gary. "A Theory of Competition among Pressure Groups for Political Influence." Quarterly Journal of Economics, 98(3), August 1983, 371-400.

Benedict, Murray R. Farm Policies of the United States, 1790-1950: A Study of their Origins and Development. New York: The Twentieth Century Fund, 1953.

--. Can We Solve the Farm Problem: An Analysis of Federal Aid to Agriculture. New York: The Twentieth Century Fund, 1955.

Brandt, J. "A National Program to Deal with Surplus Farm Programs," in American Cooperation. Washington, D.C.: American Institute of Cooperation, 1938, 221-23.

Breimeyer, Harold E "Agricultural Philosophies and Policies in the New Deal." Minnesota Law Review, 68(2), 1983, 333-52.

Citrograph. Western Agricultural Publishing Co., Fresno, Calf, selected issues.

Citrus Industry. Associated Publications Corp., Bartow, Fla., selected issues.

Citrus Leaves. Mutual Orange Distributors, Redlands, Calif., selected issues.

Cochrane, William, and Mary Ryan. American Farm Policy, 1948-1973. Minneapolis: University of Minnesota Press, 1976.

Davidson, C. "Cartel Stability and Tariff Policy." Journal of International Economics, 17(3/4), November 1984, 219-37.

Ezekiel, M. "The Cooperative Approach to Production Control," in American Cooperation. Washington, D.C.: American Institute of Cooperation, 1934, 178-83.

Florida Citrus Inspection Bureau. Annual Reports. Lakeland, Florida: 1936-8.

Gardner, Bruce. The Governing of Agriculture. International Center for Economic Policy Studies and the Institute for the Study of Market Agriculture Lawrence, Iowa: Regents Press, 1981.

--. The Economics of Agricultural Policies. New York: Macmillan, 1987.

Gardner, B. Delworth. Plowing Ground in Washington. San Francisco: Pacific Research Institute for Public Policy Research, 1993.

Gilligan, Thomas, William Marshall, and Barry Weingast. "Regulation and the Theory of Legislative Choice: The Interstate Commerce Act of 1887." Journal of Law and Economics, 32(1), 1989, 35-61.

--. "The Economic Incidence of the Interstate Commerce Act of 1887: A Theoretical and Empirical Analysis of the Short-Haul Pricing Constraint." Rand Journal of Economics, 21(2), 1990, 189-210.

Hirshleifer, Jack. Price Theory and Applications, 3rd edition. Englewood Cliffs, N.J.: Prentice-Hall, 1984.

Hoffman, G. W. Future Trading Upon Organized Commodity Markets in the United States. Philadelphia: University of Pennsylvania Press, 1932.

Hoffman, Elizabeth, and Gary D. Libecap. "Institutional Choice and the Development of U.S. Agricultural Policy in the 1920s." Journal of Economic History, 51(2), June 1991, 397-411.

--. "Political Bargaining and Cartelization in the New Deal: Orange Marketing Orders," in The Political Economy of Regulation: An Historical Analysis of Government and the Economy, edited by Claudia Goldin and Gary D. Libecap. Chicago: University of Chicago Press and National Bureau of Economic Research, 1994.

Hopkins, J. T. Fifty Years of Citrus: The Florida Citrus Exchange: Z109-1959. Gainesville, University of Florida Press, 1960.

Hulbert, L. S. "Legal Status of Plans for Production Control" in American Cooperation. Washington, D.C.: American Institute of Cooperation, 1932, 504-15.

Irons, Peter H. The New Deal Lawyers. Princeton: Princeton University Press, 1982.

Jacquemin, A., T. Nambu, and I. Dewez. "A Dynamic Analysis of Export Cartels: The Japanese Case." Economic Journal, September 1981, 685-96.

Johnson, Ronald N., and Gary D. Libecap. "Contracting Problems and Regulation: The Case of the Fishery." American Economic Review, 72(5), December 1982, 1005-22.

Joskow, Paul, and Roger Noll. "Regulation m Theory and Practice: An Overview," in Studies in Public Regulation, edited by Gary Fromm. Cambridge: MIT Press, 1981.

Joubert, W. H. "Freight Rates on Florida Citrus." Economic Leaflets, 2(9), August, 1943, Gainesville: Bureau of Economic and Business Research, College of Business Administration, University of Florida.

Kalt, Joe, and Mark Zupan. "Capture and Ideology in the Economic Theory of Politics." American Economic Review, 74(3), June 1984, 279-300.

Knutson, R. D., J. B. Penn, and W. T. Boehm. Agriculture and Food Policy. Englewood Cliffs: Prentice Hall, 1983.

Kolko, Gabriel. Railroads and Regulation: 1877-1916. New York: W. W. Norton, 1965.

Libecap, Gary D., and Ronald N. Johnson. "Legislating Commons: The Navajo Tribal Council and the Navajo Range." Economic Inquiry, 18(1), January 1980, 69-86.

Libecap, Gary D., and Steven N. Wiggins. "The Influence of Private Contractual Failure on Regulation: The Case of Oil Field Unitization." Journal of Political Economy, 93(4), 1985, 690-714.

Lowi, Theodore J. The End of Liberalism, 2nd ed. New York: W. W. Norton, 1979.

Manthy, Ray S. Natural Resource Commodities-A Century of Statistics. Baltimore: Johns Hopkins University Press, 1978.

Murphy, P. L. "The New Deal Agricultural Program and the Constitution." Agricultural History, October 1955, 160-68.

Nourse, E. D. Marketing Agreements Under the AAA. Washington, D.C.: Brookings Institution, 1935.

Nourse, E. D., J. S. Davis, and J. D. Black. Three Years of the Agricultural Adjustment Administration. Washington, D.C.: Brookings Institution, 1937.

Ockey, W. C. Outlines of Marketing Agreements and Licenses Under the Supervision of the General Crops Section, Agricultural Adjustment Administration. Washington, D.C.: U.S. Department of Agriculture, Agricultural Economics Section, 1936.

Osborne, Dan. "Cartel Problems." American Economic Review, 66(5), December 1976, 835-44.

Peltzman, Sam. "Toward a More General Theory of Regulation." The Journal of Law and Economics, 19(2), 1976, 211-40.

Perkins, Van L. "The AAA and Politics of Agriculture: Agricultural Policy Formation in the Fall of 1933." Agricultural History, October 1965, 220-29.

--. Crisis in Agriculture: The Agricultural Adjustment Administration and the New Deal, 1933. Berkeley: University of California Press, 1969.

Posner, Richard A. "Theories of Economic Regulation." Bell Journal of Economics and Management Science, 5(2), 1974, 335-58.

Reuther, W., H. J. Webber, and L. D. Batchelor. The Citrus Industry, vol. 1, 2nd ed. Berkeley: University of California Press, 1967.

Rucker, Randy, and Lee J. Alston. "Farm Failures and Government Intervention: A Case Study of the 1930s." American Economic Review, 77(4), September 1987, 724-30.

Schultz, Ted W. Production and Welfare of Agriculture. New York: Macmillan, 1949.

Shepard, Lawrence. "Cartelization of the California Arizona Orange Industry, 1934-1981." Journal of Law and Economics, 29(1), April 1986, 83-124.

Shover, John L. "Populism in the Nineteen-Thirties: The Battle for the AAA." Agricultural History, 3940(1), January 1965, 17-24.

Shuler, P. E., and J. C. Townsend, Jr. Florida Citrus Fruit: Acreage, Production, Utilization, Prices and Tree Numbers. Orlando: Florida Crop and Livestock Reporting Service, 1(1), 1948.

Spurlock, A. H. "Florida Citrus Cooperatives," in Economic Leaflets, 3(1), December 1943. Garnesville: Bureau of Economic and Business Research, College of Business Administration, University of Florida.

Stedman, A. D. "The Philosophy and Practical Operations of the New Farm Bill," in American Cooperation. Washington, D.C.: American Institute of Cooperation, 1938, 178-83.

Stigler, George. "A Theory of Oligopoly" journal of Political Economy, February 1964, 44-61.

--. "The Theory of Economic Regulation." The Bell journal of Economics and Management Science, 2(1), 1971, 3-21.

--. "Free Riders and Collective Action: An Appendix to Theories of Economic Regulation." Bell Journal of Economics and Management Science, 5(2), 1974, 359-65.

--. The Theory of Price, 4th ed. New York: Macmillan Publishing Co. 1987.

--. "Monopoly," in The Fortune Encyclopedia of Economics, edited by D. R. Henderson. New York: Warner Books, 1993.

Sunstein, Cass R. "Constitutionalism after the New Deal." Harvard Law Review, 101(2), 1987, 421-510.

Thompson, James M. "The Orange Industry: An Economic Study." Bulletin 622. Berkeley: University of California Agricultural Experiment Station, 1938.

U.S. Department of Agriculture. Yearbook of Agriculture. Washington, D.C.: Government Printing Office, 1934.

--. Agricultural Statistics. Washington, D.C.: Government Printing Office, 1938.

--. Agricultural Statistics. Washington, D.C.: Government Printing Office, 1940.

--. Annual Report of the Department of Agriculture, Report of the President of the Commodity Credit Corporation. Washington, D.C.: Government Printing Office, 1941, 1956-1960.

U.S. Department of Commerce. Census of Agriculture. Washington, D.C.: Government Printing Office, 1930, 1940, 1954.

--. Historical Statistics of the United States. Washington, D.C.: Government Printing Office, 1975.

Webber, H. J., and L. D. Batchelor. The Citrus Industry, vol. 1, 1st ed. Berkeley: University of California Press, 1943.

Wiggins, Steven N., and Gary D. Libecap. "Firm Heterogeneities and Cartelization Efforts m Domestic Crude Oil." Journal of Law, Economics and Organization, 3(1), 1987, 1-25.

Wilson, James Q. The Politics of Regulation. New York: Basic Books, 1980.

Ziegler, L. W., and H. S. Wolfe. Citrus Growing in Florida, revised ed. Gainesville: University Press of Florida, 1975.

ELIZABETH HOFFMAN and GARY D. LIBECAP, Professor of Economic and Dean, College of Liberal and Sciences, Iowa State University, and Professor of Economics, University of Arizona and National Bureau of Economic Research, respectively. Financial support was provided by National Science Foundation grant SES-8920965. A related version of this paper that examines orange marketing order in detail, "Political Bargaining and Cartelization in the New Deall: Orange Marketing Orders," is in Claudia Goldin and Gary D. Libecap, eds., The Political Economy of Regulation: An Historical Analysis of Government and the Economy, Chicago: University of Chicago Press and NBER, 1994. The authors thank Andrew Dick, Tom Gilligan, Shawn Kantor, Barbara Sands, and participants in the sessions on Empirical Advances in Political Economy at the Western Economics Association Meetings, Lake Tahoe, June 1993, for their helpful