1. See Bernstein, Regulating Business by Independent Commission. For a criticism of the life-cycle theory, see Owen and Braeutigam, The Regulation Game, 11-12. [BACK]

2. See, generally, Paul J. Quirk, Industry Influence in Federal Regulatory Agencies (Princeton: Princeton University Press, 1981). [BACK]

3. Country elevators, estimated to number 10,400, receive grain from farmers for later delivery to terminal elevators or grain processors. Storage capacity ranges from one hundred thousand bushels to over a million, and the average number of workers is approximately four. Inland terminals, numbering 710 by one count, receive grain directly both from farmers and from country elevators, holding it for delivery mostly to export facilities and grain processors. There are approximately eighty export facilities. Inland terminals and export facilities have large capacities, in the millions of bushels, and employ an average of forty to fifty workers each. [BACK]

4. Bulk handling at storage facilities is more hazardous than grain handling at processing facilities (e.g., cereal plants) that have throughput. Feed milling involves a grinding and mixing process whereby grains are blended with protein concentrates, vitamins, drugs, and minerals. The receiving operations in feed mills are almost identical to those in grain elevators. Grain and feed handling is accomplished by bulk conveyors and bucket elevators, like in grain elevators, but the operations tend to be smaller and slower than in most grain elevators, resulting in less dust generation. On the other hand, the grinding and processing creates additional dust. A report prepared for OSHA by Arthur D. Little, Inc., estimates the following distribution of grain-handling facilities by type:

5. Country elevators generally have lower throughput than other facilities, but those acting more as transfer points than as storage areas have unusually high throughput. Differences in elevator technology and design further complicate any assessment of grain elevator safety. Dust-control technology has improved considerably in the past twenty years, so higher throughputs are much safer in facilities with relatively new technology. Basic elevator design has improved as well. Newer facilities are built with a minimum of ledges, rafters, and other features that allow layered dust to accumulate. Some older facilities, retrofitted with faster handling equipment, still lack the benefit of safer building design. Compared to others of equivalent throughput, these facilities are safer in some respects and less safe in others. [BACK]

6. For a discussion of the shortcomings of these data and a defense of the higher estimate, see U.S. Department of Agriculture, Office of the Special Coordinator for Grain Elevator Safety and Security, Prevention of Dust Explosions in Grain Elevators—An Achievable Goal (Washington, D.C.: Government Printing Office, 1980), 28. There are other hazards associated with grain handling as well. Workers have suffocated while inspecting bins, and the Environmental Protection Agency is concerned about outdoor dust emissions and fumigant use. Still, there is a consensus in government and industry that the explosion hazard poses the greatest danger. [BACK]

7. The uneven correlation between production and sales further compounds the difficulty of evaluating injury data over time; for example, 1985 was a banner year for production but a bust for exports and, hence, for grain handling (see William Robbins, "A Barren Prospect for Wheat: As Exports Dry Up, Another Big American Harvest Goes into Storage," New York Times, June 16, 1985, sec. F, p. 4). [BACK]

8. Most primary explosions occur because an ignition source comes in contact with a dense dust cloud. These "explosions" actually entail a rapid combustion in which flames spread through the dust cloud and heat is produced more rapidly than it is dissipated to the surroundings. [BACK]

9. USDA, Prevention of Dust Explosions, appendix C, table 44, p. 110. [BACK]

10. National Academy of Sciences, National Materials Advisory Board, Pneumatic Dust Control in Grain Elevators: Guidelines for Design Operation and Maintenance, NMAB 367-3 (Washington, D.C.: National Academy Press, 1982), 103-8 (Appendix C: "Results of an Experiment to Determine Whether Dust Suspensions in Bucket Elevator Legs Can be Kept Below the Lower Explosive Limit by Pneumatic Means"). [BACK]

11. Much of the uncertainty surrounding pneumatic dust control is due to its checkered past. A lot of ineffective systems have been installed and others have not been properly maintained. The dust control systems in many facilities were designed by the local air conditioning and furnace outfits. Heating and cooling principles do not lend themselves well to this application, leading the NAS to conclude that "contractors who design and install systems often are poorly qualified for the work." [BACK]

12. National Academy of Sciences, National Material Advisory Board, Prevention of Grain Elevator and Mill Explosions, NMAB 367-2 (Washington, D.C.: National Academy Press, 1982), 18. [BACK]

13. See note 6 above. [BACK]

14. Only three of the fourteen major explosions examined in a USDA study occurred in facilities less than fifteen years old. There are no reliable figures on the number of elevators built since 1980, but sources in the grain industry agree that at least 95 percent of all elevators are older. [BACK]

15. National Fire Protection Association, Standard for the Prevention of Fires and Explosions in Grain Elevators and Facilities Handling Bulk Raw Agricultural Commodities, NFPA 61B-1980 (Quincy, Mass.: NFPA, 1980) sec. 6-6.3, p. 20 (the various versions of this document are hereafter referred to as NFPA 61B, followed by the year of publication). [BACK]

16. Ibid., sec. 5-1.5, p. 14. [BACK]

17. See, generally, National Fire Protection Association, Guide for Explosion Venting, NFPA 68 (Quincy, Mass.: NFPA, 1984); NFPA, Standards on Explosion Prevention Systems, NFPA 69 (Quincy, Mass.: NFPA, 1984). [BACK]

18. The specifications for dust-control systems provided by UL in 1926 have long since been deleted from 61B, although it is difficult to ascertain why from the committee minutes. The first edition of what later came to be called NFPA 61B required that air aspiration systems be installed in conformance with the UL Research Bulletin no. 37 on controlling floating dust. See National Fire Protection Association, Regulations of the National Board of Fire Underwriters for the Prevention of Dust Explosions in Terminal Grain Elevators: Recommended by the National Fire Protection Association (Boston: NFPA, 1925), sec. 33, p. 6. [BACK]

19. NFPA 61B-1980, sec. 8-1.1, p. 23. [BACK]

20. See, for example, Secretary of Labor v. Valley Center Farmers' Elevator, Inc., 8 O.S.H.C. 1061 (1979) (violations for hazardous dust accumulations vacated for failure to establish the amount of dust necessary for the creation of an explosion); and Secretary of Labor v. Cargill, 7 O.S.H.C. 2114 (1979) (violations vacated for failure to establish that accumulations of one-eighth to one-fourth inch constitute a "recognized hazard" in the industry). [BACK]

21. Testimony of William Phillips, assistant vice president, loss prevention and security, Continental Grain Co., informal OSHA hearing, In the Matter of Grain Handling (Kansas City, Mo., June 21, 1984, vol. III, transcript), p. 49. [BACK]

22. NFPA 61B-1980, sec. 6-6.6, p. 20. Information on Continental's policy was obtained through interviews. [BACK]

23. Ibid., sec. 7-2.4, p. 22; sec. 2-4.1, p. 10; sec. A-11-6.5, p. 41. [BACK]

24. A one-page "divisional guideline" issued by the Continental Grain Co. states that dust accumulations exceeding one-eighth of an inch over a broad area constitute an unsatisfactory condition requiring "immediate attention." This guideline was not intentionally made public by Continental. It was introduced into evidence at an OSHA hearing, however, and a Continental official acknowledged its authenticity in an often hostile exchange with OSHA officials and a representative of organized labor. See informal OSHA hearing, In the Matter of Grain Handling (Kansas City, Mo., June 21, 1984, vol. III, transcript), pp. 47-59 and Exhibit 38. The one-eighth-inch standard is also incorporated into Canadian standards and a proprietary standard developed by the Factory Mutual Research Corp. See "Deadlock over Explosive Dust," Science, November 4, 1983, 485, 486. [BACK]

25. NFPA 61B-1973, foreword, p. 5. [BACK]

26. The committee's official response was that "a vast majority of elevators' legs require slow down switches [and it] would be difficult to identify those that could be run without such a device." See National Fire Protection Association, Technical Committee Reports: 1986 Fall Meeting (Quincy, Mass.: NFPA, 1986), "Report of Committee on Dust Explosion Hazards," Part I, Comment 61B-27-(5-1.9) (Log #71), 9. [BACK]

27. Cheryl Cook, "AGRI Damages Called Unfair," Council Bluffs Nonpareil, May 29, 1985, 1. [BACK]

28. Major explosions occur only when there is fuel to spread the explosion. A specific requirement of almost any level of stringency would, in all likelihood, form the basis for liability in those cases in which the fuel, for whatever reason, was ignited. [BACK]

29. NFPA 61B-1980, sec. 5-5.2, p. 16. This language was adopted at the July 1985 committee meeting after a proposal to add the italicized language in the following provision: "All bearings shall be properly maintained as prescribed by the bearing manufacturer and/or the lubricating instructions" (see attachment to Memorandum from Max Spencer to NFPA 61B Committee Members, June 20, 1985). The Cargill representative objected to "the bearing manufacturer dictating to the operator." "Put the responsibility on the person who owns it," he argued, "and he will maintain it well if he is running a good shop." The committee did not adopt the proposed change. [BACK]

30. For example, one commenter suggested a new paragraph "with stronger wording which clarifies that this is an advisory guide not an industry standard." NFPA, Technical Committee Reports: 1986 Fall Meeting, "Report of Committee on Dust Explosion Hazards," Part I, Comment 61B-1-(1-1) (Log #4), 6. For a sample of similar proposals, see ibid., Comment 61B-4-(1-1.4 New) (Log #5) and Comment 61B-9-(1-2.4 New) (Log #12). [BACK]

31. For example, compare NFPA 61B-1973, sec. 3011, p. 10 (explosion venting will "minimize damage to the building or equipment and injury to personnel"), with NFPA 61B-1980, sec. 4-1, p. 13 (explosion venting "so that structural or mechanical damage is minimized.) [BACK]

32. See NFPA 61B-1959, introduction, sec. 1, p. 1.; NFPA 61B-1973, sec. 701, p. 19. [BACK]

33. NFPA 61B-1980, sec. 2-2, p. 9, and sec. A-2-2, p. 36. [BACK]

34. See, for example, Secretary of Labor v. Cargill Inc. (Conrad, Montana), 8 O.S.H.C. 1745, 1747 (1980) (appendix sections concerning tramp metal detectors "wholly advisory" rather than mandatory, and private advisory standard cannot form the basis of a violation). [BACK]

35. Interbin vents reduce the danger of explosive dust clouds forming and igniting during bin loading. They are much cheaper, but less efficient, than attacking the problem through dust collection. The concern, particularly among insurance representatives on the committee, is that interbin vents can help spread a fire or explosion. Basically, punching holes in walls, for whatever reason, violates fundamental principles of fire protection. There are very few bin fires, however. The primary hazard in grain elevators is explosions, and experience indicates that explosions will get through a concrete wall with or without a vent. But the likelihood of explosion may actually be decreased by use of such vents. See NFPA 61B-1980, sec. 3-2.2, p. 12. [BACK]

36. Dust explosions are so powerful that it is difficult, some say impossible, to use venting successfully in most facilities. Ongoing studies of explosion venting funded by the NGFA were slowed down when an experimental silo was unexpectedly destroyed in an explosion. Comments about the practicality of venting a silo roof in accordance with 61B are raised periodically by those arguing that in many designs the roof is too small to provide an adequate vent. "The roof vent [in most cases] will be about as effective as a rifle with a cardboard barrel," notes an NAS panelist. The committee does not take issue with these arguments. It retains the requirements as an "ideal" that admittedly will not be effective in all circumstances. For an example of criticisms of the venting requirements, along with the official committee response, see NFPA, "Technical Committee Reports: 1986 Fall Meeting," Report of Committee on Dust Explosion Hazards," Part I, Comment 61B-22-(4-2) (Log #28), 8. [BACK]

37. NFPA, National Electric Code (ANSI/NFPA-70), Article 500. [BACK]

38. Division 2 is supposed to represent conditions where there are not explosive concentrations of dust in the air but where there could be dangerous accumulations of static dust in electrical equipment. This makes sense for machinery operating in the most active parts of the elevator. "It makes no sense," notes an insurance representative, "for equipment on the third floor of an elevator head house." [BACK]

39. NFPA 61B-1980, sec. 5-1.9, p. 14; sec. 5-2.2, p. 15; sec. 5-6.4, p. 16. [BACK]

40. For an analysis of this provision, see Donald L. Morgan and Mark N. Duvall, "OSHA's General Duty Clause: An Analysis of Its Use and Abuse," Industrial Relations Law Journal 5 (Spring 1983): 283 -321. [BACK]

41. See, generally, National Academy of Sciences, National Material Advisory Board, The Investigation of Grain Elevator Explosions, NMAB 367-1 (Washington, D.C.: National Academic Press, 1980). [BACK]

42. Federal Register 45 (February 15, 1980): 10732 (request for comments and information and notice of informal public meetings). [BACK]

43. NAS, Prevention of Grain Elevator Explosions,40. The panel also noted that collecting information about actual explosions "is difficult because of legal constraints" (35). [BACK]

44. Ibid., 9. [BACK]

45. U.S. House, Agriculture Subcommittee on Wheat, Soybeans, and Feed Grains, Review of Grain Elevator Safety, July 21, 1982 (Washington, D.C.: Government Printing Office, 1982.) [BACK]

46. It included provisions for (1) emergency action plans, (2) employee training, (3) permit systems for hot work, (4) procedures for bin entry, (5) briefing of contractors, (6) housekeeping, (7) grate openings, (8) filter collection specifications, (9) bulk grain driers, (10) grain stream processing equipment, (11) emergency escape requirements, (12) preventative maintenance, (13) inside bucket elevator design, and (14) partially inside bucket elevator design. [BACK]

47. Proponents and critics alike agreed that the housekeeping provision accounted for 70-75 percent of the initial cost of the rule and upward of 95 percent of the recurring costs. [BACK]

48. NAS, Prevention of Grain Elevator Explosions, Appendix D: "Report of the Subpanel on Recommended Standards and Regulations," p. 130. [BACK]

49. "False Sense of Security for Workers Should Not Be Created by Rule, OSHA Told," O.S.H. Rep. (BNA) (March 29,1984), p. 1170. [BACK]

50. National Grain and Feed Association, Petition for a Partial Stay of the Grain Handling Facilities Standard, In the Matter of Grain Handling Facilities Standard, March 14, 1988, 2. [BACK]

51. For example, most industrywide estimates were based on a nonrandom survey of seventeen facilities. And the analysis of pneumatic dust control, according to a grain elevator insurer, was "uninformed and inaccurate." [BACK]

52. Estimating the cost of the "grate opening" requirement, for example, depended on assumptions about (1) the number of facilities that currently have either pit magnets or perpendicular grate bars with appropriate spacing, (2) the percentage of those not in compliance that would choose magnets instead of grates, (3) the average product, installation, and maintenance costs for magnet systems, and (4) similar estimates for constructing metal overgrates. [BACK]

53. On the grate opening requirement, for example, the OSHA study assumed that most facilities would add low-cost overgrates. The NGFA figured that most operators would add more expensive magnets. [BACK]

54. For example, Booz, Allen estimated the total area of country elevators that would have to be swept or vacuumed at 49.4 million square feet. Industry estimated 77.5 million. Since there are not even reliable figures on the exact number of country elevators, both estimates are within a reasonable margin of error given existing data. [BACK]

55. OSHA estimated that 10 percent of existing facilities already complied with the housekeeping requirements. The figure may be higher. Most of those familar with the industry agree that Cargill, for example, maintains comparatively safe facilities. The estimates offered by industry seem unrealistically gloomy in predicting that no facilities are in compliance. [BACK]

56. Booz, Allen & Hamilton, Inc., "Some Impacts of a Proposed OSHA Standard on Grain Handling Facilities" (Report prepared for Office of Regulatory Analysis, OSHA, April 1984), II-4-II-35. [BACK]

58. Federal Register 49 (January 6, 1984): 996. The alternatives added by OMB are puzzling. They certainly did not add flexibility to the standard. The housekeeping requirements were already stated in performance terms that allowed sweeping, pneumatic dust control, or any other effective method. The "once per shift" proposal, as industry representatives privately admitted, meant almost nothing. All facilities do some sort of housekeeping once per shift. "The language here is so vague that you could sweep with a rake for five minutes and be in compliance," argued an AFL-CIO lobbyist. The pneumatic dust-control alternative was equally puzzling, as nobody argued seriously that all facilities should be required to install such equipment. [BACK]

59. These "worksheets," intended to demonstrate the high cost of the standard, provided simple formulas with which operators could calculate their own compliance costs and send the results to their congressman. [BACK]

60. U.S. Senate, Subcommittee on Labor of the Committee on Labor and Human Resources, Hearing (S. Hrg. 98-946) (Yukon, Okla., July 13, 1984, testimony of Joe Hampton, executive vice president, Oklahoma Grain and Feed Association), 20. [BACK]

61. Joann S. Lublin, "Grain-Elevator Rule Forges Unusual Link as AFL-CIO Backs Labor Agency's Plan," Wall Street Journal, August 24, 1983, 50. [BACK]

62. Federal Register 52 (December 31, 1987): 49592. For a report on Senator Harkin's efforts, see O.S.H. Rep. (BNA) (December 2, 1987), p. 999. [BACK]

63. These areas are (1) floor areas within thirty-five feet of inside bucket elevators, (2) floors of enclosed areas containing grinding equipment, and (3) floors of enclosed areas containing grinding equipment. The standard is not limited to these areas, however. "Priority housekeeping shall include at least [these three areas]" (29 CFR 1910.272 [i] [A]; emphasis in original). [BACK]

64. 29 CFR 1910.272 (p) (7). [BACK]

65. Federal Register 52 (December 31, 1987): 49598. [BACK]

66. Ibid. [BACK]

67. Editorial, Feedstuffs, April 4, 1988. [BACK]

68. National Grain and Feed Association v. OSHA, 858 F.2d 1019 (5th Cir. 1988). The court expressed "no opinion as to whether the agency underestimated the compliance costs of its standard" but agreed with industry challenges that the record was inadequate. Approximately one year later, OSHA "redetermined" that the one-eighth-inch action level is feasible, promulgating the "final rule" again with a "supplemental statement of reasons" (Federal Register 54 [December 4, 1989]: 49971). [BACK]

69. Informal OSHA hearing, In the Matter of Grain Handling (Washington, D.C., June 14, 1984, transcript), 696. [BACK]