张五常《蜜蜂的神话》英文版

The Fable of the Bees: An Economic Investigation Author(s): Steven N. S. Cheung Source: Journal of Law and Economics, Vol. 16, No. 1 (Apr., 1973), pp. 11-33 Published by: The University of Chicago Press Stable URL: http://www.jstor.org/stable/724823 Accessed: 01/11/2008 07:20 Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showPublisher?publisherCode=ucpress. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We work with the scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that promotes the discovery and use of these resources. For more information about JSTOR, please contact support@jstor.org. The University of Chicago Press is collaborating with JSTOR to digitize, preserve and extend access to Journal of Law and Economics. http://www.jstor.org THE FABLE OF THE BEES: AN ECONOMIC INVESTIGATION* STEVEN N. S. CHEUNG University of Washington Economists possess their full share of the common ability to invent and commit errors .... Perhaps their most common error is to believe other economists. GEORGE J. STIGLER E VER since A. C. Pigou wrote his books on "welfare,"1 a divergence between private and social costs has provided the main argument for instituting government action to correct allegedly inefficient market activities. The analysis in such cases has been designed less to aid our understanding of how the economic system operates than to find flaws in it to justify policy recom- mendations. Both to illustrate the argument and to demonstrate the nature of the actual situation, the quest has been for real-world examples of such defects. Surprisingly enough, aside from Pigou's polluting factory and Sidgwick's lighthouse, convincing examples were hard to come by.2 It was not until 1952, more than thirty years after Pigou's initial analysis, that J. E. Meade pro- posed further examples and revitalized the argument for corrective govern- * Facts, like jade, are not only costly to obtain but also difficult to authenticate. I am therefore most grateful to the following beekeepers and farmers: Leonard Almquist, Nat Giacomini, Ancel Goolsbey, L. W. Groves, Rex Haueter, Harold Lange, Lavar Peterson, Elwood Sires, Clarence Smith, Ken Smith, John Steg, P. F. Thurber, and Mrs. Gerald Weddle. All of them provided me with valuable information; some of them made available to me their accounting records and contracts. R. H. Coase inspired the investiga- tion, Yoram Barzel saw that it was conducted thoroughly, and Mrs. Lina Tong rendered her assistance. The investigation is part of a proposed research in the general area of contracts, financially supported by the National Science Foundation. 1A. C. Pigou, Wealth and Welfare (1912); and The Economics of Welfare (1920). 2 Pigou had offered other examples. The example of two roads was deleted from later editions of The Economics of Welfare, presumably in an attempt to avoid the criticism by F. H. Knight in Some Fallacies in the Interpretation of Social Cost, 38 Q. J. Econ. 582 (1924). The railroad example has not enjoyed popularity. Most of Pigou's examples, how- ever, were drawn from land tenure arrangements in agriculture, but an exhaustive check of his source references has revealed no hard evidence at all to support his claim of in- efficient tenure arrangements. 11 12 THE JOURNAL OF LAW AND ECONOMICS ment actions.3 Meade's prime example, which soon became classic, concerned the case of the apple farmer and the beekeeper. In his own words: Suppose that in a given region there is a certain amount of apple-growing and a certain amount of bee-keeping and that the bees feed on the apple blossom. If the apple-farmers apply 10% more labour, land and capital to apple-farming they will increase the output of apples by 10%; but they will also provide more food for the bees. On the other hand, the bee-keepers will not increase the output of honey by 10% by increasing the amount of land, labour and capital to bee-keeping by 10% unless at the same time the apple-farmers also increase their output and so the food of the bees by 10% .... We call this a case of an unpaid factor, because the situation is due simply and solely to the fact that the apple-farmer cannot charge the beekeeper for the bees' food ... .4 And Meade applied a similar argument to a reciprocal situation: While the apples may provide the food of the bees, the bees may fertilize the apples.... By a process similar to that adopted in the previous case we can obtain formulae to show what subsidies and taxes must be imposed ... .5 In another well-known work, Francis M. Bator used Meade's example to infer "market failure": It is easy to show that if apple blossoms have a positive effect on honey production . . . any Pareto-efficient solution . . . will associate with apple blossoms a positive Lagrangean shadow-price. If, then, apple producers are unable to protect their equity in apple-nectar and markets do not impute to apple blossoms their correct shadow value, profit-maximizing decisions will fail correctly to allocate resources ... at the margin. There will be failure "by enforcement." This is what I would call an ownership externality.6 It is easy to understand why the "apples and bees" example has enjoyed widespread popularity. It has freshness and charm: the pastoral scene, with its elfin image of bees collecting nectar from apple blossoms, has captured the imagination of economists and students alike. However, the universal credence given to the lighthearted fable is surprising; for in the United States, at least, contractual arrangements between farmers and beekeepers have long been routine. This paper investigates the pricing and contractual arrangements of the beekeeping industry in the state of Washington, the location having been 3 See J. E. Meade, External Economies and Diseconomies in a Competitive Situation, 52 Econ. J. 54 (1952). 4Id. at 56-57. 5 Id. at 58. 6 Francis M. Bator, The Anatomy of Market Failure, 72 Q. J. Econ. 351, 364 (1958). 13 THE FABLE OF THE BEES selected because the Pacific Northwest is one of the largest apple-growing areas in the world. Contrary to what most of us have thought, apple blossoms yield little or no honey.7 But it is true that bees provide valuable pollination services for apples and other plants, and that many other plants do yield lucrative honey crops. In any event, it will be shown that the observed pricing and con- tractual arrangements governing nectar and pollination services are consistent with efficient allocation of resources. I. SOME RELEVANT FACTS OF BEEKEEPING Although various types of bees pollinate plants, beekeeping is confined almost exclusively to honeybees.8 The hive used by beekeepers in the state of Washington is of the Langstroth design which consists of one or two brood chambers, a queen excluder, and from zero to six supers. A brood chamber is a wooden box large enough to contain eight or ten movable frames, each measuring 9-~ by 17-5/ by 1-3% inches. Within each frame is a wax honeycomb built by the bees. In the hexagonal cells of this comb the queen lays her eggs and the young bees, or "brood," are raised. It is here also that the bees store the nectar and pollen which they use for food. Honey is not usually extracted from this chamber but from the frames of a shallower box, called a super, placed above the brood chamber. The queen excluder, placed between the super and the brood chamber, prevents the laying of eggs in the upper section.9 The bees, and consequently the beekeepers, work according to a yearly cycle. Around the beginning of March, a Washington beekeeper will decide whether he wants to prepare for the pollination season by ordering booster packages of bees from California to strengthen his colonies, depleted and 7 The presence of apple honey in the market is therefore somewhat mysterious. While occasionally apple orchards in the Northwest do yield negligible amounts of nectar, bee- keepers are frank to point out that the dandelion and other wild plants in the orchard are often the sources of "apple" honey, so called. Elsewhere, as in New York, it was re- ported that apple orchards yielded slightly more nectar. See, for example, A. I. & E. R. Root, The ABC and XYZ of Bee Culture 386 (1923). The explanation for this divergence of facts, to my mind, lies in the different lengths of time in which the hives are placed in the apple orchards: in Root's day the hives were probably left in the orchards for longer periods than today. 8 See George E. Bohart, Management of Wild Bees, in U. S. Dep't of Agriculture, Bee- keeping in the United States 109 (Ag. Handbook No. 335, 1971). [Hereinafter cited as Beekeeping. . . ]. Leafcutters, for example, have recently been introduced for the pollina- tion of alfalfa and clover seeds. But these bees yield no honey crop and are seldom kept. 9 For further details see Spencer M. Riedel, Jr., Development of American Beehive, in Beekeeping. .. 8-9; A. I. & E. R. Root, supra note 7, at 440-58; Carl Johansen, Bee- keeping (PNW Bulletin No. 79, rev. ed. March 1970). 14 THE JOURNAL OF LAW AND ECONOMICS weakened during the winter and early spring. Alternatively, he may decide to build up the colony by transporting the hives to farms or pastures in warmer areas, such as Oregon and California. The colony hatches continuously from spring to fall, and the growth rate is rapid. Reared on pollen, the infant bees remain in the brood stage for about three weeks before entering the produc- tive life of the colony for five or six weeks. Active workers spend three weeks cleaning and repairing the brood cells and nursing the young, then live out the remainder of their short lives foraging for pollen and nectar.x? Because of the bees' quick growth, the working "strength" of a colony includes both brood and workers, and increases from about five frames in early spring to about twelve by late summer. Spring is the primary season for fruit pollination, and beekeepers usually market a standard colony strength of roughly four frames of bees and two to three frames of brood for pollina- tion services. But since empty frames are needed to accommodate the expand- ing colony, two-story hives, with 16 or 20 frames, are used. The swarming period, beginning in mid-summer and lasting until early fall, is the peak honey season, and the yield per hive will vary positively with the colony strength. Because the maximization of honey yield requires that the colonies be of equal strength, they are usually reassorted in preparation for the major honey season, so that the number of colonies at the "peak" is generally larger than the number in spring.n1 When pollen fails in late fall, the hives become broodless and the bee population begins to decline. During the idle winter months adult bees live considerably longer than in the active season, and they can survive the winter if about 60 pounds of nectar are left in the hive. But in the northern part of the state and in Canada, where cold weather makes the overwintering of bees more costly, the common practice is to eliminate the bees and extract the remaining honey. It should be noted here that bees can be captured, and that they can be easily eliminated by any of a large number of pesticide sprays.J2 The cost of enforcing property rights in nectar is therefore much lower than economists have been led to believe. 10 For further details see Carl Johansen, supra note 9; F. E. Moeller, Managing Colonies for High Honey Yields, in Beekeeping. . . 23; E. Oertel, Nectar and Pollen Plants, in Beekeeping. . . 10. 11 According to a survey conducted by Robert K. Lesser in 1968, based on a sample of 30 out of 60 commercial beekeepers in the state of Washington, the total number of peak colonies is 14.6% higher than that of spring colonies. See Robert K. Lesser, An Investi- gation of the Elements of Income from Beekeeping in the State of Washington 74 (un- published thesis, Sch. of Bus. Admin., Gonzaga Univ., 1969). 12 See, for example, A. I. & E. R. Root, supra note 9, at 97-103; Eugene Keyarts, Bee Hunting, Gleanings in Bee Culture 329-33 (June 1960); U.S. Dep't of Agriculture, Protecting Honey Bees from Pesticides (Leaflet 544, 1972); Carl A. Johansen, How to Reduce Poisoning of Bees from Pesticides (Pamphlet EM 3473, Wash. St. Univ., Col- lege of Ag., May 1971); Philip F. Torchio, Pesticides, in Beekeeping. . . 97. 15 THE FABLE OF THE BEES Few agricultural crops, to my knowledge, exhibit a higher year-to-year variance of yield than does the honey crop. Several natural factors contribute. Cold weather and rain discourage the bees from working, and winds alter their direction of flight. Also, the nectar flows of plants are susceptible to shocks of heat and cold.?3 The plants yielding most honey are mint, fire- weed, and the legumes, such as alfalfa and the clovers. Fruit trees usually have low nectar flows, although orange blossoms (in California) are excel- lent. Indeed, the pollination of fruits, especially the cherry in early spring, may actually detract from the yield of honey: less honey may be in the hive after pollination than was there initially, owing to the bees' own consumption. Another reason for the low honey yield from fruit trees is the relatively short time that the hives are left in the orchards. Cross-pollination is accidentally effected as the bees forage for nectar and pollen. Pollination services were not marketed before World War I, primarily because small farms had enough flowering plants and trees to attract wild insects. It was not until 1910 and the advent of modern orcharding, with its large acreage and orderly planting, that markets for pollination services be- gan to grow rapidly.24 Today, the services are demanded not only for pro- duction of fruits but also for the setting (fertilizing) of seeds for legumes and vegetables. Evidence is incontrovertible that the setting of fruits and seeds increases with the number of hives per acre, that the pollination pro- ductivity of bees is subject to diminishing returns, and, despite some bee- keepers' claims to the contrary, beyond some point the marginal productivity may even be negative.?5 There is also strong evidence that pollination yield will improve if the hives are placed strategically throughout the farm rather 13 See E. Oertel, supra note 10; C. R. Ribbands, The Behaviour and Social Life of Honeybees 69-75 (1953); Roger A. Morse, Placing Bees in Apple Orchards, Gleanings in Bee Culture 230-33 (April 1960). Owing to its weather, Washington is not one of the better honey yielding states in the Union. Data made available to me by the U. S. Dep't of Agriculture indicates that over the years (1955-1971) Washington ranks 24th among 48 states in yield per colony and 20th in the total number of colonies. The U.S. Dep't of Agriculture data, like those obtained by Lesser, provide no information on the dif- ferent honey yields and pollination requirements of various plants and are therefore of little use for our present purpose. It should be noted that the U.S. Dep't of Agriculture overall yield data are significantly lower than those obtained by Lesser and by me. See Robert K. Lesser, supra note 11. 14 See M. D. Levin, Pollination, in Beekeeping. . . 77. 15 Id.; 9th Pollination Conference, Report, The Indispensable Pollinators (Ag. Exten- sion Serv., Hot Springs, Ark., October 12-15, 1970); G. E. Bohart, Insect Pollination of Forage Legumes, 41 Bee World 57-64, 85-97 (1960); J. B. Free, Pollination of Fruit Trees, 41 Bee World 141-51, 169-86 (1960); U.S. Dep't of Agriculture, Using Honey Bees to Pollinate Crops (Leaflet 549, 1968); Get More Fruit with Honey Bee Pollinators (Pamphlet EM 2922, Wash. St. Univ., March 1968); Protect Berry Pollinating Bees (Pamphlet EM 3341, Wash. St. Univ., February 1970); Increase Clover Seed Yields with Adequate Pollination (Pamphlet EM 3444, Wash. St. Univ., April 1971); Honey Bees Increase Cranberry Production (Pamphlet EM 3468, Wash. St. Univ., April 1971). 16 THE JOURNAL OF LAW AND ECONOMICS than set in one spot.16 The closer a particular area is to a hive, the more effective will be the pollination within that area. Although each individual bee will forage only a few square yards, the bees from one hive will collectively pollinate a large circular area,17 and this gives rise to a problem: given a high cost to control fully the foraging behavior of bees, if similar orchards are located close to one another, one who hires bees to pollinate his own orchard will in some degree benefit his neighbors. This complication will be further discussed in the next section. In the state of Washington, about 60 beekeepers each own 100 colonies or more; at the peak season the state's grand total of colonies is about 90,000. My investigation, conducted in the spring of 1972, covered a sample of nine beekeepers and a total of approximately 10,000 spring colonies. (One of these beekeepers specialized in cut-comb honey and he will be treated separately in a footnote.) Table 1 lists the bee-related plants covered by my investigation. As seen from Columns (3) and (4), some plants (such as cherry trees) re- quire pollination services for fruit setting but yield no honey; some (such as mint) yield honey while requiring no pollination service; and some (such as alfalfa) are of a reciprocal nature. Note that when alfalfa and the clovers are grown only for hay, pollination services are not required, although these plants yield honey. The practice of relocating hives from farm to farm, by truck, enables the beekeeper to obtain multiple crops a year, either in rendering pollination service or in extracting honey. However, while the maximum observed number of crops per hive per year is four and the minimum is two, my estimate is that a hive averages only 2.2 crops a year. More frequent rotation not only involves greater costs of moving and of standardizing hives, but abbreviates the honey yield per crop. In the southern part of the state, where the rela- tively warm climate permits an early working season, beekeepers usually begin by pollinating either cherry or almond (in California) in early spring. The hives may or may not then be moved northward in late spring, when apple and soft fruits (and some late cherry) begin to bloom.18 The lease period for effective pollination during spring bloom is no more than a week. But then, for a month or two between the end of fruit pollina- 16 See, for example, Douglas Oldershaw, The Pollination of High Bush Blueberries, in The Indispensable Pollinators, supra note 15, at 171-76; Roger A. Morse, supra note 13. 17 There is, however, little agreement as to how far a bee could fly: estimated range is from one to three miles. For general foraging behavior, see M. D. Levin, supra note 14, at 79; O. W. Park, Activities of Honeybees, in The Hive and the Honeybee 125, 149-206 (Roy A. Grout ed., 1946); C. R. Ribbands, supra note 13.