Document omLmVqEab554Ej4811Y1QdvMo

REGULATORY PROGRAM OF THE UNITED STATES GOVERNMENT APRIL 1,1990 MARCH 31,1991 SL 062674 OVERVIEW 13 Supporter* of this system ergue that it would Developing a record of actual expenditures while wovide an incentive for better estimate* of the costs of minimirmy the recordkeeping burden on the faciilative proposals and a basis for an explicit private sector, ef the costs and tradeoffs ofsuch proposals. Bi|h ceilings would focus attention on the expected benefits of the program, and alternative approaches; cost ceilings that were too low would prevent agencies from issuing implementing regula te Such an approach would, needless to say, giva Identifying an appropriate "baseline," recognising that some costa would be incurred even in the absence of Federal regulation; and Estimating the coats of forgoing certain products where Federal regulation prohibits production or agencies an incentive to choose regulatory approaches distribution. | that would produce the greatest benefits at the lowest Frh of these raises difficult issues in designing an effective regulatory budget process. For example, the coats of banning a product are not directly measurable and can only be estimated using com ' statistical ISSUES AND AREAS FOR FURTHER UDY models. However, measuring only the *t compli ance coats for oversight purposes creates its toward While the fiscal budget process provides ratinueus record ofactus! expenditures, there is m jmparaUe record of the cost of meeting regulatory require h*"TMg substances and products inatas of controlling tliam Aa a first step in determining the feasibility of the ments* Members of Congress and the pest two regulatory budget concept, OMB has begun systemati Administrations have considered developing an ac cally to collect the costs of all significant published counting framework to record direct regulatory expen regulatory actions. Analysis ofthese data should aid in ditures, but more work needs to be done to solve the the development of ways to overcome the problems of practical accounting problems inherent in measuring regulatory budgeting, uncover unforeseen problems in the private expenditures that Federal regulations developing cost estimates, and more frilly refine e mandate. These include: workable regulatory budgeting process. Current Regulatory Issues in Risk Assessment and Risk Management Many Federal agency regulatory decisions are intended to reduce risks to human life and btalth. Government regulations control which agricultural chemicals mty be used to reduce insect damage, increase farm yields, and improve the quality of food products. Other rules govern hazards in the Nation's workplaces end emissions from its factories. There are regulations directing the way in which automobiles must be manufactured, commercial aircraft main tained, end trains operated. Hardly any widespread human activity that entails risk is free of lame degree ef social control, often achieved through government regulation. Regulatory decisions involving risk require agencies to address questions such as, How safe is `safe'?" and "How dean is 'clean1?" When government agencies promulgate regulations intended to reduce a risk or mitigate a hazard, they are engaging in what has become known as risk management. These policy choices inevitably involve consideration of both the risks entailed by the underlying activity and the social consequences of regulatory intervention. Thus, the first challenge of ride management is to set priorities to determine which risks are worth reducing and which ere not For government to cany out its risk-management responsibilities, there must be an extensive invest ment in the careful assessment and quantification of risks. The term risk assstamen* means the applica tion of ovdible scientific principles and statistical methods to develop estimates of the likely effects of natural phenomena and human activities. The need to keep risk assessment and risk manage ment separate has long been tire objective of responsi ble public officials. In 1983, the National Academy of Sciences (NAS) studied the process of managing risk * Retearehert, uiinc different method*. astumpticn*, and time period*, have formed incomplete eetimetei by adding ap the eoet ef individual regulation*. The** animate* accordingly thow ooneidcrabl* venation for current annuel coat* ranging from 160 billion to 1176 billion a year--6 to 15 percent of current Federal outlay*. I i i SL 062675 ajff 14 ULATOBY PROGRAM or THE UNITED STATES GOVERNMENT in the Federal Government and offered the following reeonnscndatioos, amongethen: Btcanmtndation J: JUfulatory spend** should take Mae* la Mtabliab asd maintain a daar emioaptuaJ disnneuoc htwtB aaaaaaaaat of hak* and th* astddortdoB if risk manafement alternative*, that is, the soantills findinf* and policy judgment* aabediad in riak aaaaaafnanta thould ha explicitly diatiafuiabad from th* political, eooomic, and eonaiderxtiim* that tafluannt tha daaign and ahoiat of rafulatoiy antagiaa.*1 Raeoenmendotion 1; Before aa agmcy daddaa whether a anbatanm ahould sr ahould not b* regulated a* a health buard, a detailad and tomprahanaiv* written riak aaaaaamant ahould ba prepared and mad* publicly available Thia wnttan aaatiarnmt ahould dearly diatispuiab be tween the aritntifie baaia and th* poW baaia fcr the aaucy't conclusion*. The belief that riak ueeument end riak manage ment ahould be kept separate enjoys widespread support among professional risk-assessment practi tioners and risk-management officials. "Other* have emphasized the importance of ensuring that policy biases do not distort the analysis of alternative risk-management choices.** The NAS principles have alio have been endorsed by a number of Federal agencies, including the Office of Science and Technol ogy Policy (OSTP), the Environmental Protection Agency (EPA), and the Department of Health and Human Services (HHS).* Unfortunately, risk-assessment practices continue to rely on conservative models and assumptions that effectively intermingle important policy judgments within the scientific assessment of risk. Policymakers must make derisions based on risk assessments in which scientific findings cannot be readily differenti ated from embedded policy judgments. This policy environment makes it difficult to discern serious hazards from trivial ones, and distorts the ordering of the Government's regulatory priorities. In some eases, the distortion of priorities may actually increase health and safety risks. This section explores some of the continuing difficul ties that plague the practice of risk assessment, end describes briefly their policy implications. It can be summarised in three observations: The continued reliance on conservative (worst-cast) assumptions distorts risk assessment, yielding esti mates that may overstate likely risks by several orders of magnitude. Many risk assessments are based on animal bioassays utilizing seasitivt rodent aperies dosed at extremely high levels. Conservative statisti cal models are used to predict low-dose human health risks, based on the assumption that human biological response mimics that observed in laboratory animals. Worst-case assumptions concerning actual human exposure are commonly used instead of empirical data, farther exaggerating predicted risk levels. Conservative biases embedded in risk assessment impart a substantial "margin of safety'. The choice of an appropriate margin of safety should remain the province of responsible risk-management officials, and should not be preempted through biased risk assess ments. Estimates of risk often foil to acknowledge the pretence of considerable uncertainty, nor do they present the extent to which conservative assumptions overstate likely risks. Analyses of risk-management alternatives routinely ignore these uncertainties and treat the resulting upper-bound estimates as reliable guides to the likely consequences of regulatory action. Decisionmakers and the general public often incor rectly infer a level of scientific precision and accuracy in the risk-assessment process that does not exist Conservatism in risk assessment distorts the regula tory priorities of the Federal Government, directing societal resources to reduce what art often trivial eardnogenic risks while failing to address more substantial threats to lift and health. Distortions are probably most severe in the area of cancer-risk assessment because many conservative models and assumptions were developed specifically far ettimst- * National Academy of Sdencea, Riak Aaeeeement in At federal Government: Managing the Avon, Washington. DC: National Academy Flw, 1M3 (hereinafter, NAS Risk Management Study), p. UL * Dad., p 183. "For rapmestativt view* of riak-aianrrnent practitioner* eee, eg., Letter B. Leva, The Strategy of BirW Regulation: Dedeion Framework* for Polity , Wuhingtcn. DC: Brooking*, 1981; Letter B. Lava, *><011104* of Riak Assessment," Chapter 3 is Quantitative Riak Ateeetment in Regulation, Letter B Lava, ad., Washington, DC: Brooking*, 1982, aap. pp. 82-84. For reprooestativo views ofriak-management affinal* eee, *.fWilliam D. Ruekelakaua, "Saeno*. Riak, and Public Policy* Vital Speeehee of the Day, Volume 49, No. 20, August 1, 1983, pp. 612-816. "Sea, eg., Howard Kuarouthtr and liaa Bendixen, *Bi*fit* Aaaaamant Sr Regulatory Problems,* and Barucb Fisdibaff and Louis Anthony Cox, Jr., "Conceptual Framework for Regulatory Benefit* Assessment,* Chapters 3 and 4, rwpoetively, is Beneftte Aaetument: The State of the An, Judith D. Be&tkovsr, Vincent T. Covello, and Jeryl Mumpower, eda., Dordrecht, Netherlands: D. Reidel, 1908, pp. 44--4i, 59-81. "See U.S Offioe of Science asd Taehnoloc Policy, "Chemical Caronopens: A Review of the ft------t sad It* Aseodated Phsctplaa,* Prisoplc 29 (50 FR 10376, March 14, 1985, hereinafter, OSTP Riak Aeteetment Guideline*y, U-S. Environmental Protection Agency, "Guideline* for Camnogvr, Riak Assessment," 61 FR 34001 (September 34,1986, hereinafter, SPA Carcinogen Riak Aeetument Guideline*); U-S Department of Health and Human Service*, Riak Aaaeaament and Riak Management of Tone Subetaneee, April 1985, p. 20. OVIKVlfiW 16 ing upper bounds for these risks. Riss-assessment aethods with similar conservative biases are less common elsewhere, particularly in those areas where ml-world data are available, or where the mechanism by which injury or illness occurs is better understood. A renewed commitment to the NAS recommenda tions is dearly warranted. As quantitative risk MHtpnnf pUyt an increasingly significant role in risk management, the seed to separate sdenee from policy becomes ever more important, if either process is to tpifrtiiw public confidence. As former EPA Administrator William D. RucksIshaus has noted: Biik f------ b* faeaad m tiSc aridaae KMStific eoaaantua only. Nothrnf rod* public dans turn than tha auipidon th. 7 eooaida kaoi baao aBowad to iafluaac* the a i*nt of m d & aa ALTERNATIVE RISK-ASSESSMENT METHODOLOGIES Risk assessments of chemical .^stances in general (and of possible carcinogens in particular) involve a mixture of facts, models, and assumptions. There is considerable debate concerning the scientific merits of the models and assumptions commonly used in risk assessments. In some cases, a scientific consensus has developed to support a particular model or assump tion. In other instances, however, certain models and assumptions ire relied upon because they reflect past practices rather than the leading edge of edenee. Furthermore, a scientific basis for several of tha most critical models and assumptions simply dots not exist. Most scientists agree that these models and as sumptions impart a conservative hiss: that is, they lead to risk projections that the actual (but unknown) risk is very unlikely to exceed. These "upper-bound" estimates are often useful as a screening device, to exclude from regulatory concern potential hasards that are insignificant even under worst-caie condi tions. Unfortunately, upper-bound risk estimates are routinely employed for altogether different purposes, such as estimating the likely benefits of regulatory actions. Policymakers are required to act on the basis of bused representations of both the magnitude of the underlying hazard and the extent to which Govern ment action wfl] ameliorate it Contemporary risk assessment relies heavily upon animal bioassay and epidemiology. Etch approach has theoretical advantages and disadvantages. In practice, both can be misused to bolster preestsblished conclu sions. The following discussion emphasises problems In carcinogenic risk assessment, because the preven tion and cure of cancer plays such a major rule in policy issues involving risks to life and health. Animal Bloaaaay Animal taaCfog enables scientists to estimate risks ex ante, bsftw human health affects materialise, whereas epidemiological studies can only detect such effects ax peat In addition,' animal teats can be conducted under tightly controlled laboratory condi tions, which provide more reliable aatimatee of exposure and avoid many of ths confounding factors that often plague epidemiological investigations. The relatively abort lifetimes of experimental mammals (such as rets and mice) allow scientists to ascertain the possible affects of leng-taxm exposure in just a few years. Animal tatting suffers serious limitations, however, arising from certain oitieal assumptions. Despite its^fo routine application, there is no accepted stientifioV basis for the assumption that results can be meaningfrilly extrapolated from teet animals to humans." Some scientists believe that animal data should not be used in assessing human health risks.* Another critical limitation is the reliance on very high doeca to generate adverse effects in test ani mals* A mathematical modal must be used to bridge the gap between these high-dose exposures and the low-dose exposures more typically freed by people. Many different mathematical models can be conftructsd to fit the data at high doses. These models often vary enormously, however, in their predictions of risk at low dooes. Beyond those unavoidable methodological con straints, the results of animal bioassays may be subject to conflicting scientific interpretation or strongly influenced by the choice of research method. William D. Ruckelzbaui, (op dt\ p. S14. r OSTP Guideline*. Guidelisa 8, p. 10878. " Saa. tf. Bruor Aaaa, Rasa* Macaw, sad Loia Swinky Gold, *Raakinj PoaaOl* Cardaopanic Hazard*,* Same*, Vat. 886, April 17, 1887; Gio Batu Gori, "Tha R*|ulatioc tf Cardaogaaic Hazard*,* Sdenet, Vol. 808, April 18, 1080. " OSTP Guidelines, Guidelis* 11, p. 10877. SL 062677 16 HEGUIATORY PROGRAM OF 1HE UNITED STATES GOVERNMENT Tissue preparation and histology present obvious opportunities for error, as experts may disagree as to how slides should be interpreted.10 This problem generally is not significant at high doses, where malignancies are often obvious. At low doses, however, pathologists often differ in how they distinguish tumors from hyperplasia. Subjectivity cannot be avoided where such interpretations of the data must be made.11 Epidemiology Epidemiology is attractive because it largely avoids these two problems. It focuses on observable human health effects instead of on hypothesized outcomes based on animal experimentation, and it relies upon real-world exposures to generate empirical data. Many ofthe serious problems associated with animal studies can be avoided, allowing researchers to develop risk estimates that are directly related to human health. Unfortunately, epidemiological research suffers from its own set of limitations. For example, retrospective studies often have difficulty correlating morbidity and mortality with exposure to specific substances. Expo sure data are commonly lacking, incomplete, impre cise, or affected by systematic recall or selection biases. Furthermore, the risks these studies seek to detect are often very small relative to background, thus making statistically significant effects difficult to observe. When health effects are latent, correlating exposures to illness is even harder. Besides these unavoidable methodological limitstier.;, epidemiological studies often suffer from out right bias. Many studies employ scientifically ques tionable procedures aimed at demonstrating positive relationships between specific substances end human illness.13 Some researchers use inappropriate statisti cal procedures to "mine" existing databases in search of associations. One result of these practices is that epidemiological studies often display contradictory results." Despite these constraints, properly conducted ani mal bioassays and epidemiological studies both hive useful teles to play in quantitative risk assessment Indeed, they are complementary. The usual weaknaasea of epidemiological investigations--unreliable exposure data, confounding affects--are readily voided in laboratory experiments on animals. The weaknesses of animal bioassays--high- to low-dose extrapolation, animal-toman conversion--do not arise in epidemiological studies. Careful risk assessment incorporates both types at analysis to ensure that the emerging picture of human health risk is as complete as possible, and that inferences derived from this picture are themselves internally consistent ISSUES IN RISK ASSESSMENTS DERIVED LARGELY FROM ANIMAL BIOASSAYS Animal trioassays tend to dominate current risk assessments. An important reason for this is that the derivation of dose-response relationships is a oitical regulatory motive for performing quantitative risk assessment Animal studies are ideally suited to serve this purpose by virtue cf the controlled conditions under which dose and response can be calibrated. Epidemiologies] studies often are relegated to provid ing merely a "reality check" to ensure that the implications of animal bioassays are plausibly consis tent with real-world experience. Because of this heavy emphasis on animal testing, the focus here is on several major problems that arise with respect to risk assessments primarily baaed on the results of animal bioassays. The Use of Sensitive That Animate Tb enhance the power of animal tests, scientists typically rely on genetically sensitive test animals. It "la the engine] analyeie of the ret bioaeeay need to derive the dnee-reeptmac ftmetien far dioxin, f of S5 control* wore eeid to develop liver turnon. An independent revitw of tfaie date moulted to 16 of the 86 control* being dmtaified ae having *och tuaon. Sec U.S. Environmental Protection Agency. A Conor RuA-Specific Dorn KttvneUt fori, XT, 9-TCDD, Appendix A, EPA/V0Q-6&*>07Ab, June 1988 (hereinafter, Dioxin Risk Autumtm Appendix A), pp. 2-3. " Coiin N. Perk end Ronald D, Snee, `Quantitative Ri*k Aeccecmcnt: Stete-cf-the-Art for CartinofeneaU." Chapter 4 in Rixk Management ofExixting Chenucah, Rockville, MD: Government Inetitntoe, 1963, p. 88. " Alvan R. Feinatmn, "Scientific Standard* to Epidemiological Studiee of the Menace <f Daily Iifia* Science Vol. 242 Daeembar 2 1968, pp. 1257-1263. " linda C. Mayct, Ralph I. Horowitz, and Alvan R- Petnetoin, "A Collection of 86 Topic* with Cantradicttoy lUauhe in faafTrmtml Raoearth,* International Journal of Epidemiolap, VoL 17, No. 9 (19881, pp. 660-665. SL 062678 OVERVIEW 17 ^ anckar whether theie spedea nccuretely mimic biological responses in humans. BffPt test spedes are extremely sensitive. For approximately one-third of all male B6C3F1 miff, a common test species, spontaneously develop Uver tumors.** The same phenomenon occurred in an important bioassay concerning dioxin using female Sprague-Dswley (Spartan) rets. Tumors observed in dosed animals were predominantly located in the liver. However, approximately one-fifth ofthe animels in the control group also developed liver turnon." TTts ralcvsnce of elevated liver tumors in hypersensitive pecies has been questioned bv scientists and is not universally considered probat evidence of eardnogenidty. Nevertheless, cancer ; aaaeasmento often proceed on the assumption .at theae data are sufficient to conclude that a substance is indeed a carcinogen.*4 The reliance on eensitive test animals also biases risk assessments in a more subtle way. It establishes powerful incentives to search for and develop increas ingly sensitive test spedes. As test animals become more sensitive, repeated testing using identical proto cols will tend to result in higher and higher estimates of risk even if all other factors are held constant Selective Uae of Alternative Studies In their respective risk-assessment guidelines, both OSTP and EPA recommend that relevant studies should be considered irrespective of whether they indicate a positive relationship." In practice, however, studies that demonstrate a statistically ignirtwit positive relationship routinely receive more weight than studies that indicate no relationship at aH** For example, the plant growth regulator daminosida OUar) and its metabolite unsymmetncal 1,1-dimethyIhydrmxiiM (UDMH) recently received B2 classifications ("probable human carcinogen"). Each of thase classifications was based on a single positive bioassay," Overcoming such a classifies* ->n requires, at a fffiwttniim, two "essentially identi studies showing no such relationship* In the cas of Alarind UDMH, however, a more stringent test was apparently applied: Three high-quality negative studlea showed no significant effects; these studies appear to have received little or no weight in the classification decision." Selective Interpretation of Sarahs fcisk-assessmont guidelines generally give the great est weight to the most sensitive test animals. Thus, if a substance has been found to cause cancer in onl "Ames tt el., (op. eiL), p. 276. "plena Pitk Attoumtnt Apptndiz A, pp. 2-3. * Sot Asm tt al., (op tit.), p. 276 (arguing that such data art trrevelast); OSTP OuUotinm Gttidelins 2, p. 10377 (mrhiding that ouch data "must be approached earahiUy*); and tPA Carcinogen Pith Autummt Guidttiim, p. S3S95 (nuking tba paUey Judgment that uch data are sufficient evidence at cardaegeneeis). Liver tumor* demiwatad in EPA'e dinon risk aaaaaamant. Baa Diem itiok Automate, appendix A, pp.2-3. rSee OSTP Guidtlintt, Guidahna 25, p. 10373; EPA Ceraisaftn Rial Auusmtte Guititlinu, p. 33225. "Sat SPA Carcinogen Pitk Autmmtte Guidelines, p. S3222-34000. A tingle animal taat that shows a positive result to an unsaual degree* ip 33999j u sufficient to warrant at laaat a B2 classification ("probable human cardnogen*), avan if this raauh occur* in a spades known to hava a high rata of tpoatanaeua tumor*. A atreng animal faioaasay or epidemiological study showing no evideno# of cardnogeme effect cannot overcoat thi* proaumptiao (p. 34000). "Sta Attend Poor Review tt Damineaida (Alar) and UDMH (Unsymmetrical 144tmetby]bjdntiae),* Memorandum from John A. Quaat to Mark Boodtt, U.S. Environmental Protection Agency, OPTS. May 15,12S2 (hereinafter, AtariUDMH Jtetmal Poor Ravine Ho. St Hus internal OPTS panel reviewed aeveral recant studies an Alar and UDMH. One Budy of Alar yielded a statistically ngnificnnt increase in common hmg tamer* in mice, but dy far ana of three doaaga level*. Ranih* war* not statistically aignificant at ana higher and two lower dotage*, and control* alao displayed unusually high turner incidence. S05 of the lung tumor* in doted mice were benign, verms (W in the commie. One stud) of UDMH yielded statistically significant increase* in common lung and unorntmon liver fniwnri in mitt, but only for the higher at two dosagaa. 97% of the lung tumor* in doted mice war* benign, venue 100% fa the eantral*. 29% ef the liver turnon in doted cue* war* benign; no tumor* were obeervad fa the condole. Pnor studies that purported to tho a carcinogenic raepoose had bean Judged inadequate by EPA*> Scientific Advisory Panel, an external peer rrv group The Office of Faatidda* and Toxic Subataneat (OPTS) panel noted that a different internal EPA riak-eseeecment panel (the Carcinogen Assessment Group) considered theee studies euffident to justify B2 classifications when H evaluated them far EPA`i Office af Solid Waste and Emergency Response Despite the scientific controversy, the OPTS panel interpreted these prior studiee as `supporting evidence* under EPAi hsk-aseettment guidelines. "See EPA Carcinogen Riok Assessment GuuULinoo, p. 33225 (establishing the need far mplicatc identical studies showing no effect), and p 33999 < establishing the '" requirement af two well-designad studiee showing no increased tumor incidence to warrant a *no evidence* deteraunauonl. 41 AlarIUDMH Internal Poor Btvittc No. 2, pp. S, 2, 2. EPA's scheme far enrdnagen classification is itself an issue among edentisu. Bee. eg.. U.S Environmental Protection Agency, Risk Assessment Forum, WaHuhep Popart an EPA Guidelines far Carcinogen Rukt Assessment. EPA/62M-69/015, Washington. DC: March 1269, pp. 21-26. SL 062679 18 REGULATORY PBOGRAM OF TOE UNITED STATES GOVERNMENT pedes or gender but ehcrwn to exhibit no effects elsewhere, the results pertaining to the sensitive species or gender typically will be need to develop estimates of human-health risks. For example, if male mice develop cancer from a substance but female mice and rats of both genders do not, then the results from the male mouse often will be used to derive estimates of cancer risks to humans.0 Once a positive result has been obtained in an animal bioassay, a substance often will be provision ally classified as a probable human carcinogen. The statistical burden of proof then shifts to the no-effect hypothesis. Because it is logically impossible to prove a negative, however, this practice establishes a virtually irrebuttable presumption in favor of the carcinogenesis hypothesis. Severe Testing Conditions Current risk-assessment protocols require the uae of very high doses. Unfortunately, high doaes art often toxic for reasons unrelated to their capacity to eauee cancer. A common procedure is to use whet is celled the maximum tolerated dose (MTD), which is the most that can be administered to a test animal without causing acute toxicity. At such exposure levels, substances often cause severe inflammation and chronic cell killing. For example, formaldehyde causes nasal tumors in rats when administered in high doees. However, MTD administration severely inflames nasal passage tissues. It is therefore unclear whether the cancers induced ere caused by formaldehyde per se or fay the toxic effects of high doaes. Results such as these have caused some adentista to question the validity of rodent tests performed at the MTD for estimating human health risks that arise from exposure at low doses.4* By combining very high doaes with highly sensitive teat subjects, tome bioassay? are predisposed to discover apparent carci nogenic effects. Belevtnce of Animal Bioassay Results An important reason why animals vary in their sensitivity is that they have different physiologies, metabolic processes, reproductive cycles, and a host of other epedee-epedfic characteristics that largely re sult from unique evolutionary paths. of these factors needs to be carefully considered in evaluating the significance of animal data with respect to human health. This is rocognixed in both the OSTP and EPA guidelines, but it is often neglected when the guide lines are applied to specific substances.4*4 The most important assumption in this regard is that animal test results can he meaningfully extrapo lated to humane. A recent study of chemicals tested under the auspices of the U.B. National Tbxicology Program shows that this assumption can lead to the erroneous classification ofmany chemicals aa probable human carcinogens.4* Positive associations have been obtained in either rwts or mice for half of 214 chemicals tested. However, results were consistent across these two genetically similar spedss only 70 percent of the time. If it is assumed that rodent bioasuys have the same sensitivity and selectivity with respect to human carcinogens as they do between rodent spades, and it is farther assumed that 10 percent of all chemicals are faa fact human carcino gens, then 27 of every 100 randomly selected chemi cals would be misclastifiad as probable human carcinogens. Only three chemicals would be misclasd- fied as noncardnogens. Thus, "false positives* would be 9 times more common than "false negatives."4* Of coune, this ratio of false positives to false negatives reflects highly conservative "upper-bound" assumptions concerning sensitivity and selectivity. Given the high degree of similarity between rats and mice end the limited resemblance between rodents and humans, the ecnaitivity of rodent bioassays with respect to human carrinogenidty is probably much lower than 70 percent Furthermore, other research indicates that selectivity may be as low at 5 percent Sm EPA Carcinogen RuM Assessment Guidelines, p. 23907 (4sU few kng-tam animal stndiM Growing Om greatest eensitrrity should generally be given the greatest emphasis). 44 Sm. *.g,, Amo tt al., fap. dlj, pp. 276-277. 41 OSTP Guidelines, Guideline 25, p. 10378; SPA Caranofen Risk Assessment Guidelines, p. 24003 (responding to comment* on the draft guideline* and affirming agreement with OSTP Guideline 26). 44 Letter B Lave, Fanny X. Ennevar, Herbert 6. Rossnbronr, and GUbmt 8. Oman, Tnformatitm Value of the Rodent Sioaeeay,* Nature, Val. 336 (December 15, 1988), pp. 631-633. mFalst negatives occur when a teet fails to detect effects when thsy are in bet present. Sensitivity retie la the capacity da test to minimise falee negatives. False positives occur whan a test appear* to detect affects that In fact are ebeent Selectivity refer* to a Met'* ability to minimise falee positive*. The 9 to 1 ratio of false positive* to false negative* calculated by lava et al. aaaumaa that both selectivity and sensitivity equal about 70%. SL 062680 Agusttec only for this tow selectivity suggests that ^1^ podtivec are almost 30 times mor** common than jMgatives. This raises serious questions concernin$ the practical utility of the current approach to fiwi Uoassays tor the purpose of quantitative risk assessment/' Other tocton should also be considered when ikying upon animal bioassay results as the primary i for quantitative risk assessments. Far example, certain substances are toxic or even carcinogenic by cne pathway but not by others. Nevertheless, animal bioassay protocols often emphasize the most sensitive pethway. As tong as human exposure is likely to arise the same way, then this choice may be reasonable. However, the pathway to which the teat species is sensitive sometimes reflects azrexposure route thstti implausible or irrelevant tor humans. For axampto, tonnaldehyde causes nasal turnon in rata at 12 times the rate observed is the next most sensitive animal species. This extreme sensitivity may be related to the tact that rats breathe only through the nose. then may be important differences between ani mals and humans that make specific turnon irrele vant. For example, some chemicals cause cancer in the symbal gland of the rat; because humans leek such e gland it is unclear whether these results metier in estimating human health risk. Other substances induce cancer through biochemical mechanisms not tound in humans. A greater controversy surrounds the 'nation whether the aarne weight should be given \ oign and malignant tumors. The srientific conaenr hot benign and malignant tumors should be eg, ted only when it is scientifically defensible to do In practice, however, benign and malignant tumm - era routinely aggregated unless a strong case can be made against the practice.4* The difference between those default assumptions ia significant: One approach counts only carcinomas that art present, whereas the ether counts tumors that might become carcinomas. In an extreme case, a substance that promotes benign tumors but never causes cancer could be classified as a probable human carcinogen simply because benign and malignant tumors are treated equally. In addition, tumor incidence to commonly pooled across sites to obtain a total estimate of carcinogenic effects.*0 This implititly assumes that cancer induction to independent across sites and not the result of either metastasis or the same biological mechanism. Given the extreme sensitivity of tost spetios and the regular nae of MTD administration, other explanations for tumors oceuning at multiple sitas appear just as This Choioe of Doee-Retponac Model No tingle mathematical model to accepted as generally superior for extrapolating from high to tow doee**1 Consequently, Federal agencies often nee a variety of different models. Bather than being a scientific footnote to the risk-assessment process, however, the choice of modal to actually an impotent policy issue. The multistage model appears to be the moot used method for estimating tow-doae risks from chemicals, and there are two major sources of bias embedded in this choioe: its inherent conserva tism at tow doses, and the routine use of the *linearixod* form in which the 95 percent upper bound to used instead of the unbiased estimate. The multistage model essentially involves fitting e polynomial to a data eat, with the number of "stages" identified by the number of terms in the polynomial. Since animal bioassays rarely have more than three dose levels, it to unusual to see applications of the multistage modal with more than two stages. Al though the multistage model enjoys some scientific support because it to compatible with multistage theories of carcinogenesis, in practice the model fails to include enough stages, due to the abeence of sufficient alternative exposure cohorts. The multistage model typically yields tow-dose risk estimates that are higher than moat other models. For example, when five different dose-response models were analysed in a recent risk assessment of cad mium, estimate* of cancer risks at moderate doaes varied by a factor of 100. This difference among *' lev* r at., (op. at.), p 631. Adju*ti&f alao tor 1m* aantitivity radueaa the ratio of folaa poCttw t* IUm nafmtfvaa. Tor example, if araaitivity it only 10 percent and *11 other parameters remain rmehaafad, than thi* ntio dartinm to 9.5 to 1. Hewevar, thia impUea that hath type* of autistic*! error* art rampant, which niaa* quastian* coaearainc the practical utility of animal faioeaaayc. Thia ia, in fact, paoatiy the concern niaad by lev* at oL, (op. eUJ, who conclude that *ucb taata an coat-affective invwtmante in information ly under extraordinary conditions. ^ *QSTP Guidelintt, p. 10376. ____ EPA Carcinogen ftuk Assessment Guidthntt, p. 33997. *li. 41OSTP Guidtlinet, Guideline 96, p- 10378; Amee et aL, (op. dL), p. 276. 062&8X SL so REGULATORY PROGRAM OP THE UNITED STATES GOVERNMENT stimstes widened as doaea declined toward the very low levels within the range of regulatory concern. At eery low doses, two of the five models predicted excess lifetime cancer risks greater than one in one thousand (104), a risk oftentimes regarded by policymaker* as unacceptable. However, two other equally plausible models predicted essentially no excess cancer risk at alL Since none of the five models offers a scientifically superior basis for deriving low-dose risks, the choice of model is therefore a pivotal policy decision. The accepted practice under these circumstances is to develop a subjectively-derived "best" estimate while folly informing decisionmakers as to the extent of uncertainty surrounding it" In the cadmium case, as in most others, this practice was not followed: Estimates of the number of statistical cancers that would be prevented by regulation were presented based only on the multistage model." The linearized, multistage model (IMS) is a special version of the multistage model in which the 95 percent upper confidence limit of the linear term is used instead of the unbiased estimate. That is, the model identifies the largest value for the linear term that cannot be rejected at the 95 percent confidence level and uses it in place of the unbiased estimate. Assuming that the model has been correctly specified, there is only a 5 percent chance that the true risk exceeds this level. The LMS has become the preferred statistical approach because estimates derived from it appear to be more "stable* than estimates obtained from the ordinary multistage model. The "stability" issue origi nally arose because unbiased estimates of low-dose risks are very sensitive to the maximum-likelihood estimate (MLE) of the value of the linear term. When the MLE of the linear term is positive, it dominates estimated risks at low doses. In some instances, however, the MLE of the linear term is aero, and low-dose risk estimates decline precipitously. Using the 95 percent upper confidence limit ensures that the linear term is always positive, thus eliminating the inherent "instability" of low-dose risk estimates de rived from the multistage model. M Another often-cited advantage ofthe LMS procedure Is that it provides a "yardstick' for comparing [ potencies across chemicals.11 A uniform risk-assess- | ment procedure such as the IMS, it is argued, enables policymakers to better understand the relative signifi cance of a broad array of chemical hazards and set regulatory priorities accordingly. Finally, the IMS is often defended on the ground that it is prudent to err on the side of caution when dealing with potentially carcinogenic chemicals. Be cause the LMS generates upper-bound risk estimates, policymaker* ean be confident that actual risks an likely to be lower. None of these purported advantages of the LMS approach has a sound statistical basis. It is a fundamental axiom of statistics that unbiased esti- i mates are generally preferred to biased ones. Using * the upper confidence limit instead of the unbiased ! estimate exaggerates underlying specification erran ' instead of eliminating them. "Instability" is overcome, | but at the cost of greater errors in specification. The inherent instability of the multistage model reflects a generalized misspsdfication of doseresponse--that is, the real human dose-response relationship is often very different from what the multistage model constrains it to be. The model is extremely sensitive to small differences in observed tumor incidence, which ean cause dramatic changes in estimated low-dose risks. The LMS procedure elimi nates this sensitivity without remedying the underly ing specification error. Proper statistical procedure requires correcting model misspsdfication, not mask ing its symptoms behind biased parameter estimates. The IMS procedure inflates low-dose risk estimates by a factor of two or three when the MLE of the linear term is positive. However, it increases low-dose risk estimates by orders of magnitude when the MLE of the linear term is zero.1* This means that the degree of hidden conservative bias is substantially greater for what are demonstrably lower risks. By its very nature, the IMS cannot serve as a useful yardstick for comparing the relative risk of a variety of potential carcinogens. If a given statistical procedure generated identical biases across substances tested. *8ee, eg., OSTP Guidtlintt, Guideline* 27, 29, and 31, p. 1037S; EPA Cardnagcn Ruk Atttumeni OuuitUmt, pp. 33999, 94003. " Occupation*] Safety and Health Adminietjution, "OecupetianaJ Expaaure to Cadmium; IVopoaid Suit,* 66 FR 4076 (February 6, 1990). M Albert L Nichole and Richard J. Zeckhauaer, The Danper* of Caution: Conaarvatiam in Aaaeeamcnt and the Miemanagwnent of Kiek," Chapter 3 in Advancet in Applied Afurro-Economic*, Volume 4: Kish, Uncertainty, and the Valuation ofBenefit* and Cotit, V. Kerry Smith, ad., Greenwich, CT: JAJ Pme, 1966, pp. 65-62, aep. pp. 62-63. A nontechnical vereion of thie paper ie available by the eame authon aa The Perile of Prudenee: How Conaorvative Riek Aeaeaamente Distort Regulation," Regulation, Novambar/Dacember 1966, pp 13-24. "U.S Environmental Protection Agency, A Cancer Ktk-Sptdfie Date Stdmatt for SJ,7.S-TCDD, EPAA00/6-88O07A*, June 1966 (hereinafter, Diomn Ruk Atuument), pp. 45-46. * Nichole and Zeckhauaer, op. at., pp. 62-63. SL 062682 OVERVIEW 21 then it would still yield an accurate rank-ordering of statistical procedure. The "margin ofsafety" argument theoretical hazards. Similarly, if the procedure added tacitly usurps from policymakers the authority and s atochsstie bias from a uniformly distributed random responsibility for risk-management decisions. Finally, variable, the resulting rank-ordering would still be the statistical "instability" overcome by the LMS is an accurate on an expected-value basis. The problem with artifact of specification error, not any scientific theory the IJfS is that it generates biases that intensify with ofhuman carcinogenesis that warrants the intentional the degree to which the multistage model misspedfies use of biased parameter estimates. The habitual the true dose-response relationship. Even if the reliance upon either the multistage model or its LMS multistage model provided an accurate rank-ordering descendant cannot be supported by sound scientific ofhazards, the LMS could not do so, because it injects principles. biases that are aystematic with atatistical Alternative models ere available, ofcourse, and they misspetification. have been applied in many quantitative risk assess The LMS procedure (and the multistage model ments. Because proper model specification is the itself) is also fatally flawed as a yardstick for foundation of applied statistical methodology, altera.' regulatory priority setting because it fails to take tives to the multistage mode) should be expected ar account of human exposure in the calculation of unit encouraged. Indeed, innovation is the hallmark m' risks. Regardless of the procedure's capacity to scientific inquiry; policies that institutionalize any accurately rank-order hazards, failing to adjust unit particular model specification effectively stifle scien risks by relative human txposurt virtually guarantees tific advancement that regulatory priorities will be misordered. Re sources tend to be focused on reducing the greatest theoretical hazards rather than the most significant human health risks.** Unfortunately, models other than the multistage model are often discouraged in practice.** Agandas may require substantial scientific evidence in support of an alternative model before allowing it to be used. Finally, the "margin of safety argument in favor of Alternative models thus face a burden of demonstrat the LMS unequivocally contradicts the widely recog ing scientific plausibility that the multistage model nized need to distinguish science from policy." The cannot satisfy. Evan in the extraordinary case in LMS introduces into each risk assessment a con which this burden can be satisfied, estimates may be servative bias of varying but unknown magnitude. required from the linearized multistage model any This pr- *ice fundamentally alters regulatory way* derisionn ing, Instead of leaving policy dedsionc to policymakers, the LMS disguises fundamental policy decisions concerning the appropriate margin of safety behind the veil of science. The potential human health threat posed by dioxins provides an excellent example ofthe problem of model ** action. Using the same linearized multistage model, ERA, the Centers for Disease Control (CDC), and the : In summary, the LMS cannot be justified as a Food end Drug Administration (FDA) have arrived at I method of scientific risk assessment The "yardstick" upper-bound risk estimates that span an order of \ defense implicitly asserts that scientific advancements magnitude.*1 Depending on the data and assumptions in risk-assessment methodology should take a back used, the linearized multistage model predicts unit eat to the preservation of an outdated and misguided risk factorz that vary by as much as L200, with the Some KMSOfu have attempted to deriae alternate* tedaue of relative human health risk that aepbdtly mwt far variation* in haman exposure. Arne* *t at., (op. at.), pp 272-273, daaeiba ana each ahenmtire (the Homan ExpoourVRodent Potency index, ar HER?) and r,pv.. ij.Jti ,*io*t for 33 rebalance*. Baeauaa the HERP index ia baaed an a relative rather than ahaohtta acale, the distartinf efleet cf mntervativ* biaiet embedded in the underlying ritk aaaasameata hae bean significantly reduced. Many wbstsnoaa euepected of being environmental carewocent rank vary low an the HERP index, suggesting ^tat regulatory prioritie* have bean aarioualy taiadireeted. See. u . NAS RuA Uanagmant Study, p. 161; OSfP Risk Atmumant Quidstinm, Principle 20, p. 10373; and SPA Careuuqtn Risk dsmvntnt Gutdtlintt, p. 5*001. See, f| . Arne* at at., fop. at.), p. 276 (continued reliance an linear modal* despite the accumulation of evidence again* linearity); and Letter B. Lev*, *Health and Safety Ink Analyaia: Information for Batter Deottona,* Sdsnct, VoL 236, April 17,1M7, pp. 291-295, P- ^ tegeneie* often re*m modeL..j improvement* and data that yield lower riak estimates). SPA Corexnagtn Ritk Atttumsni Guidslinst, pp. 33997-33998. *la the abeanee of odequ'a information to the oentrery, the linearised muhtttafr procedure will be employed. ... Coanderabl* uncertainty will remain concern; *spon**t at low do***, therefor*, in moot ue*. an upper-bait riak animate uainf the lineariaad muhuUf* procedure obould tlao b vented* Dioan Ruk Asstamsru Apptndu A p- 13. Unbiased risk estimates vary by a waiter .or. t SL 062683 22 REGULATORY PROGRAM OF THE UNITED STATES GOVERNMENT three risk estimates mentioned earlier clustered at the high end of the range.*1 Risk assessments baaed on different models have led other governments to establish unit risk factors that are a thousand times leas stringent than the most commonly used of these three; one study suggests that this particular estimate overstates the most likely risk estimate by a factor of almost 5,000. Conversion from Animals to Humana Once risk has been extrapolated to low doses in rodents, scientists must convert them to human dose-equivalents. The two most common approaches involve the use of body-weight or surface-area conver sions, and there are scientific reasons for choosing either approach in individual cases. The surface-area approach leads to estimates of risk that are between 7 ud 12 times greater than those baaed on the body-weight method, depending upon the test species. Despite the ambiguity of the underlying science, the more conservative aurface-area method is often ap plied reflexively." multiplicative effect of overstating risk by a factor of two at five different points in an exposure assessment will overstate actual risk by a factor of thirty-two. Worst-Case Environmental Conditions When data are .available they often relate to unusually sensitive environments or highly contami nated conditions. When estimating regional or nation wide exposures, agendes often use data from these local *hot spots" in developing more general national astimates of health risks. However, such date are never representative and estimates extrapolated from them are generally unreliable and misleading. In addition, chemicals often degrade naturally after they have been released to the environment In some cases, degradation occurs very quickly, whereas in othan the process may take many yean or even decades. A common practice in exposure assessment modeling is to assume that exposures remain constant over time--that is, chemicals are assumed never to degrade, or degradation by-products are assumed to pose identical risks. ISSUES ARISING FROM HUMAN EXPOSURE ESTIMATES In addition to developing estimates of the doee-response function, agencies must estimate the likely level of human exposure. This section examines some of the issues and problems that arise in conducting an exposure assessment It is a generally accepted principle of exposure oceocUlfeBt that estimates should be based on the most likely scenario, with appropriate consideration of uncertainty" Nevertheless, agendes often use conser vative assumptions for exposure when real-world data are unavailable. When each of these assumptions tends to overstate likely human risks, the multiplica tive effect of even a small overstatement at each stage in an exposure assessment will yield a substantial overestimate of actual exposure. For example, the** Hie Maximum-Expo--d Individual In addition to estimating the amount of a substance that may actually be present in the environment a risk analysis must also consider the conditions under which humans may be exposed. Actual risks vary considerably depending on location, mobility, and a host of other factors. Neverthelass, estimates often are baaed on the upper-bound lifetime cancer risk to the maximum-exposed individual (MED, the hypothetical person whose exposure is greater then all others. Sometimes, risks to the entire population are esti mated by assuming that everyone is exposed at the MEI level. Because environmental regulations are often justified using MEI-based risk assessments, actual risks may be substantially lower than what decisionmakers and the general public perceive them to be. " Ditton Ritk Atmumtni, pp 46-49. 104 risk-specific do-- (RsDs) derived from the linearised multistage Bedel spun tbe rang* from : ::: *_ 1.2 pieogram/kg/day. The RsDe of EPA, CDC, aad FDA ere 0.006, 0.03, end 0 06 pg/kg/day, respectively. ** Dioxin Ritk Atttummi, p. 4. *" EPA Caranogrn Aixumxnt Guidtliti, p. 33996. *EPA wfl] continue to nee this [surface arte] --u-g factor onlw data on a epeafic agent euggeet that a different aealing factor it justified." "EPA guidance document* have historically called for unbiased estimates ef exposure. See, #4., U.S Environmental Protection Agency, Guideline* for Exposure Aaeessment,* 60 FR 34042-34054 (September 24,1966, hereinafter, EPA Expoturt AtttttmtnS Guidthntt). U.S. Environmental Protection Agency, Suptrfund Public Htolth Evaluation Manual, OSWER Directive 9235 4-1, October 1966, and U.S Environmental Protection Agency, Suptrfund Expoturt Axttttmtni Manual (Revised Draft}, OSWER Directive 9285.5-1, December 1966 EPA recently abandoned the calculation of unbiased exposure estimate* far Superfund site* an the ground that it was insufficiently conservative EPA'i new protocol require* the estimation ef `reasonable maximum exposure* instead ef the average and upper-bound aeumatei Reasonable maximum exposure constitutes * new term of art that EPA intends to be "well above the average rase' but not a* extreme as the upper-bound. It provides a new opportunity for embedding conservative assumptions into exposure assessment and exaggerating estimates of actual human-health risk at Superfund sites See Ritk Atttttmtnt Guidons for Suptrfund, Volume I. Human Htalth Evaluation Manual (Pan A), Inunm Final, EPA/540/1-89/002. December 1969, Chapter 6, pp. 5, 47-60. SL 062684 OVERVIEW 23 In developing the MEI risk level, analyses invari ably assume that the level of exposure is continuous aver > 70-year lifetime. This assumption overstates risks, because people are mobile, encounter a constantly ringing portfolio of daily risks to life and health, and can take actions that reduce risk. vs. Real-World Exposure Data The thread that connects these exposure assessment tasues is that simple constructs which overstate assure are typically used in lieu of real-world data, aften because such data are unavailable. The riak generated by these models depend on the validity of their assumptions; even small biases in exposure assessment assumptions can result in a frihetutial overstatement of riak. For example, regulatory agendas may not have statistically reliable real-world data on pestidde residues in agricultural products. They also may not know the proportion of a given crop that hat been treated with a particular pesticide. A common resolu tion of these uncertainties is to assume that raciduea are equal to the regulatory "tolerance"--the maximum level allowed to be present in food sold in interstate commerce--end that 100 percent of the relevant crop has been treated- Both assumptions overstate actual exposure, but are encouraged by agency guidance as a way to instill conservatism in risk assessment*1 When data are available, however, the extent of this conservative bias becomes evident In e recent spedal review for the pestidde Captan, for example, EPA reduced its earlier upper-bound lifetime cancer risk estimate by two orders of magnitude when it replaced the original conservative assumptions with real-world data. Even with these improvements, EPA still reported that upper-bound risks were probably over stated. For example, field testa were performed beaed on applications at the maximum legal rata and aa dose to harvest as the label permita. Similarly, feeding studies assumed that diets were dominated by feedstuff* that happened to contain high residues relative to other feedstuff's, such aa almond hulls and raisin waste. As EPA noted, even if these assumptions accurately represented typical animal diets, they would do so only for portions of California where these crept ere grown; nationwide extrapolations baaed on these `hot-spots* would very likely overstate expo sure.* Since two of the highest product-specific risks ware attributed to milk and meat, these remaining conservative biases can be expected to be significant. IMPLICATIONS OF CONSERVATIVE RISE ASSESSMENT FOR RISK MANAGEMENT AND REGULATORY DECISIONMAKING The primary purpose of risk assessment is to provide data as a basis for risk management decisions. Providing useful data requires the synthesis of information concerning risks and exposure levels into a coherent package that can be used to develop regulatory options. Decisionmakers then can use these risk estimates in evaluating regulatory alternatives. Unfortunately, the way in which risk information is characterised tends to overstate risks, soaking them appear mudr greater than they are likely to be. As a result, decisionmakers may make regulatory choices that are very different from the ones they would make if they were frilly informed. Quantification of Uncertainty In accordance with the recommendations of the National Academy of Sciences, the OSTP Guidelines explicitly call for the quantification of uncertainty, particularly as it arises in the selection of dose-re sponse models and exposure assumptions.** Unfortu nately, Federal regulatory proposals that utilize risk assessment rarely provide this information, nor do they analyze the implications of uncertainty for decisionmaking. Instead, many risk assessments only identity a lifetime upper-bound level of risk." The differences between upper-bound and expectedvalue estimates may be considerable. As we indicated earlier, the upper-bound riak estimate for dioxin may be 5,000 timea greater than the moat Hktly estimate. Plausible riak estimates for perchlorocthylene (the primary solvent used in dry cleaning) vary by a factor of about 85,000.* In some instances, decisionmakers may not be informed that riak estimates differ because of policy choices hidden in the risk-assessment methodology. In EFA's proposed rule limiting emissions from coke m EPA Expoturt Ateeumtnt Guideline*. p. >4063. *Wboa there ie uncertainty is the edentific tacts. It ie Agency policy to err on the M* of public safety * rSec, e.g., US Environment*] Protection Agency, "Captan: Intent to Coned Ragietrutionr, Condueiim of Special Review* M FR 127-1126 (February 24, 1988). m OSTP Guideline*, (Guideline 27), p. 10378. ^Sec. t-x , EPA Cardnattn Risk Ateeument Guideline*, p. 33908. NichoU and Zeckhauaer, (ep. diJ, pp. 64-85. SL 062685 24 regulatory program or the united states government evens, for example, cancer risks were estimated based on the LMS model--a model that is designed to yield upper-bound estimates of risk. In previous rules involving similar types of risks, however, EPA used the unbiased maximum likelihood estimate. 1b the extent that decisionmakers were not informed that the higher estimate of risk was largely due to a different low-dose extrapolation procedure, regulatory decisions based on this risk assessment were likely to reflect misunderstanding rather than science.71 Plausible estimates of likely cancer risk can often be found buried in regulatory background documents. However, Federal Register rulemaking notices seldom present such estimates alongside upper-bound esti mates. This practice overstates baseline human health threats, as well as the amount of risk reduction that may be accomplished by regulation. Policymakers and the public are misled because they typically see only the upper-bound estimates of the threat The prevalent Federal ageney practice is to calcu late the benefits of Federal regulatory initiatives based solely on upper-bound estimates of risk and exposure. In a recent proposal to reduce occupational exposure to cadmium, for example, the Occupational Safety and Health Administration (OSHA) developed risk estimates based on five alternative models for animal data, and two alternative models for human data. Across these seven dtta/model combinations, estimated excess lifetime cancer risk at the least stringent of the two proposed exposure standards varied from 0 to 153 cases per' 10,000 workers occupationally exposed for 45 years. OSHA based its proposed exposure standards on one of these dati/model combinations--the multistage model ap plied to animal data. This data/model combination predicted an excess lifetime cancer risk of 106 per 10,000 exposed workers, and was used to estimate aggregate cancer incidence and the risk-reduction benefits attributable to the new standard. Uncertain ties in the underlying risk assessment, which span several orders of magnitude, were not carried forward through the exposure assessment and benefit calcula tion stages. This analytic error effectively obscured the uncertainty surrounding the true incidence of cadmium-induced lung cancer, and resulted in benefit estimates that may exceed actual reductions is occupational illness by several orders of magnitude.1* \ Miaordarwd Prioritie*, Parverae Outcomes Logically, one would expect that the routine over statement of likely risks would lead to inefficient regulatory choices. Decisionmakers, convinced that s certain substance or activity poses a significant threat to public health, might well take actions that they would otherwise resist Alternatively, they might take actions that address the wrong real-life risks. 1b the extant that risk assessments differ in the degree to which they adopt conservative assumptions, it is difficult to determine which activities pose the greetest risks and hard to establish reasonable priorities far regulatory action. Because conservatism in risk assessment is especially severe with respect to carcinogens, it is reasonable to expect that other health and safety risks tend to receive relatively less attention and weight As a result society may actually incur greater total risk, because of misordered priori ties caused by conservative biases in cancer risk assessmentn A perverse and unfortunate outcome of using upper-bound estimates based on compounded conser vative assumptions is that the practice may actually increase risk, even in situations where cancer is the only concern. Regulatory actions taken to address what are in fact insignificant threats may implicitly tolerate or ignore better known, documented risks that are far more serious. For example, before it was banned, ethylene dibromide (EDB) was used as a grain and toil fumigant to combat vermin and molds. Vermin transmit disease, and mold* harbor the natural and potent carcinogen aflatoxin B. The astimated human cancer risk from the aflatnrin contained in one peanut butter sandwich is about 75 timas greater than a ftill day's dietary risk from EDB exposure. On this basis alone, it might have been appropriate to accept a small increase in cancer risk from EDB to reduce the much larger cancer risk from aflatoxin. By eliminating the relatively small hazard from EDB, Federal risk managers may have intensi- "knur from Wendy Gramm (Administrator of tbs OSes tt Informsties and Regulatory Affairs) to Lae Thomas (Administrator of the Environment*] Protection Agency), August 12, 1M6, p. S. ''Occupational Safety and Health Administration, "Occupational Exposure toTiiiiiiasil Rule* UTtderolJUgitur 4076, 4060, 4093. n This is precisely the policy issue raised by Nichols and Zackhauaar, (op dt.), pp. 69-71, who note that EPA's 1985 decision to limit laad in gasoline was threatened by concents about potential increases in bansene exposure. Any tradeoff between lead and benzene neks would have been biased against lead, as estimates of benzene risks are more conservative limply becauae H is a carcinogen, whereas laad is not. SL 062686 fftS S 9 V I ovixvitw % tied the relative]}' potent threat of afletoxin aaaoeiated in technologically advanced society. Unfortu with an increase in the prevalence of mold contamina nately, ft also susceptible to hidden biases that may tion.1* The emphasis on riaks faced by the maximum* exposed individual may also cause a perverse result by increasing overall population riaks. For example, CPA's proposed regulation of the disposal of sewage sludge would probably create more public health risk than it eliminates. The proposal outlines a regulatory that would shift disposal from generally safe tmdenni. ; its atientific integrity and the basis for policymakers' reliance on such information in risk management decisions. For policymakers and the public to continue to rely on risk assessment in the development efregulatory initiatives, a renewed effort must be made to separate science from policy and provide risk information that is both meaningful and tmliokU practices to relatively risky alternatives. Thus, setting sludge quality standards to achieve an MEI uppers Expected Value Estimates hound lifetime eaneer risk of one in 100,000 CIO-*) Perhapv most important current need in regula would prevent 0.2 statistical eaneer cases resulting tory deci -taking is ftr carefully prepered and from monofilling and land application. However, it identifies, credible estimates of the likely risks would cause 2.0 additional statistical cancers by fercing a shift away from these disposal approaches involved. Keiying on worst<aae anal' i baaed on extremely conservative risk assessment id exposure toward incineration.1* models leads to widespread misundertu ding on the These problems can be addressed by providing part ef both Government officials and individual decisionmakers with the fill! range of information on dtizens. Decisionmakers at all levels need unbiased the risks of a substance or an activity. Thus, and impartial risk information ao they can fbcus their decisionmakers should be given the likely risks as well attention on significant problems and avoid hsfag as estimates of uncertainty and the outer ranges of the distracted by minutiae." tl potential risk. Then, if regulatory decisionmakers want to choose t very cautious risk management Weight-of-Evidenoe Determinations T- iy * strategy, they can do so and a margin of safety can be applied explicitly in the final decision. This approach is superior to one in which the expected risk and an Similar procedures are needad for assigning weight^ to each relevant study in the risk-assessment litera unknown margin of safety are hidden behind the veil ture. Current practice gives undue weight to studies 7 of e succession of upper-bound estimates adopted at that show positive relationships. Resulting risk classi key points in the risk-assessment process. fications are thus conservatively biased estimates S ; The public and affected parties also benefit from derived from samples of similarly biased bservations. 1 knowing both the expected risk and the margin of safety rather than being given upper-bound estimates Full Disclosure that are probably very different from actual risks. People are likely to have a better intuitive under Efficient and responsible decisionmaking requires that policymakers and the public be fiilly informed ) t standing of the significance of averages than they have of unlikely extremes. Tb the extent thet e margin ef safety is appropriate--perhaps to protect unusually about the implications of the regulatory alternatives among which they must choose. Meeting this require ment demands a csrefiil discrimination between i sensitive subpopulations--the magnitude of this mar science and policy. Whan risk estimates depend on gin can be more readily communicated if made assumptions and Judgments *"** of data, the t explicit. In addition, providing information in this way meaning and implications of these nonscientific pa should help improve public confidence in quantitative rameters must be clearly articulated. risk assessment as the basis for decisionmaking. AVOIDING CONSERVATIVE BIASES IN RISK ASSESSMENT Risk assessment remains a powerful and useful scientific tool for estimating many of the riaks that Avoiding Perverse Outoomee Csrefiil attention needs to be paid to the likely results of regulatory alternatives, with an eye toward avoiding choices that have the perverse effect of increasing net risk. All human activity involves risk. ``Amo tt aL (op. tit), p. 273. *U.S. Environaents] Protection Aaoney, "Standard* for tb* Ditpoaal of Sewage Sludga; Propoaad Kola," M Pit 8746-6902 CFabraary. 1 1%989,>. * Nitbolt and Zaekhautar, op at., pp. 72-76. SL 062687 S6 REGULATORY PROGRAM OP THE UNITED STATES GOVERNMENT Decisionmakers need to be rare that specific actions taken in the name of risk-reduction in one area do not make matters worse elsewhere. Quantitative risk assessment can help in this regard so long as the methods applied are not inherently biased in a way that undermines comparisons across alternatives, each rf which entails some degree of risk. Our discussion has covered only the highlights of risk-assessment methods, yet we have identified several independent places at which conservative assumptions are commonly used. Individually, each of tb*et assumptions might appear to be prudent responses to scientific uncertainty. In combination, however, they result in a distortion equal to the product of the individual conservative biases. 1b Illustrate, suppose that there are ten independent steps in a risk assessment and prudence dictates assumptions that in each instance result in risk estimates two times the expected value. Such a process would yield a summary risk estimate that is more than 1,000 times higher than the most likely risk estimate. Because there are usually many more than tan steps, and many of them will incorporate conservative biases that exceed an order of magnitude, riak estimates based on such practices will often exceed the most likely value by a factor of one million or more. Whan risk assessments contain hidden value judg ments, their sdentific credibility is inevitably compro mised. 1b the extent that policymakers and the public fail to understand the magnitude of the margin of safety embedded in quantitative riak assessments, policy choices are distorted from the course that would have been selected if decisionmakers had been better informed of the actual risks. Ironically, these policy decisions may actually increase total societal risk. Tbo much attention is focused on relatively small hazards that have been exaggerated by conservative risk assessments, leaving alone larger risks that have been estimated using unbiased procedures. Information as an Alternative Regulatory Strategy Federal regulation was initiated to deal with economic problems caused by monopoly and so-called "excess competition.* Subsequent events hive shown that, in general, economic regulation--fixing prices, establishing restrictive terms of trade, and erecting barriers to entry--is usually inefficient end detri mental to innovation. In response to these lessons, Federal regulation of this type has been under increasing criticism. As indicated above, however, much more needs to be done to reform economic regulation and restore competition. Federal regulation has more recently been initiated to deal with what economists call externalities, situations in which participants in voluntary market transactions do not bear the full costs or capture all of the benefits of these exchanges. Common examples of externalities include environmental pollution and traffic congestion, common property resources such as fisheries and public forests, and "public goods" such as basic scientific research. In each of these instances, regulation may be an appropriate mechanism to modify or restore distorted market processes, or to establish markets where heretofore they have not existed, to maximize net social benefits (including environmental, health, and safety benefits). The key ingredient is the determination that existing markets are, in some significant manner, failing to perform efficiently. The traditional regulatory approach to externalities has been the promulgation of standards. Because this approach often remedies existing externalities by creating new ones, economic incentive instruments are becoming an increasingly popular alternative to stand ards. The principal attraction of economic incentives is that they rely on market forces rather than attempt to suppress them. Ibis section explores another alternative regulatory strategy--the production, provision, or mandated disclosure of information. The first subsection briefly summarises the economies of information as it relates to regulatory decisionmaking. Three points stand out in this discussion. First, because information is costly to acquire and the capacity to process it is limited, there is an optimal level of information for every market transaction. Second, differences in the amount and quality of information between buyers end sellers are normal and do not necessarily indicate market failure. Rather, these difference* generally reflect variations in the costs and benefit! that are ettributable to information. Third, competitive markets pro vide powerful incentives for buyers and sellers to meal relevant information. Market processes, not government regulations, provide the dominant motiva tion for generating, acquiring, and disclosing informa tion. The role ofgovernment regulation thus should be to supplement these processes when they prove to be inadequate, not to supplant them when they work well The second subsection identifies three rationales for government intervention in the production or man dated disclosure of information. Two of these are economic--the public-good character of some types of SL 62688 APPENDIX V Regulatory Impact Analysis Guidance A Bsgulstory Impact Analysis QUA) should demon* grate that a proposed regulatory action Mtiafias the nquimnents of Section 2 of Executive Order No. 12291. Tb do so, it should show that: Ihers is adequate information concerning the used for and consequences of the proposed action; The potential benefits to society outweigh the potential cost*; end Of ell the alternative approaches to the given regulatory objective, the poeed action will wriTT"T net benefits to so . y. The fundamental test of e satisfactory RIA it whether it enables independent reviewers to make an formed judgment that the objectives of Executive Order No. 12291 are satisfied. An RIA that indudti U the elements described below is liktly to fulfill dus requirement Although variations consistent with At spirit and intent of the Executive Order may be warranted for some rules, most RIAs should include these elements. The guidance in this document is not in the form of acchanistic blueprint for a good RIA cannot be written according to a formula. Competent profws* SanaJ judgment is indispensable for the preparation rf s high-quality analysis. Different regulations may aS for very different emphases in analysis. For one imposed regulation, the crucial issue may be the question cf whether a market failure exists, and such of the analysis may need to be devoted to that key question. In another case, the existence of e irket failure may be obvious from the outset, but tensive analysis might be necessary to estimate the aptitude of benefits to be expected from proposed ngulstory alternatives. The amount of analysis toether scientific, statistical, or economic) that reticular issue requires depends on how crucial that ww is to determine the best alternative and cm the waplexity of the issue. kcgulatory analyfis inevitably involves uncertain* to and require* informed professional judgments. Vhmever an agency has questions about such issues the appropriate analytical techniques to use or the tomatives that should be considered, it should eon* to with the Office of Management and Budget as toy is the analysis stage as possible. Ibis document is written primarily in terms of toposed regulatory changes. However, it is squally Wplicable to the review of existing regulations. In the toer ease, the regulation under review should be compared to a beseline case of no regulation and to reasonable alternatives. Dements of Regulatory Impact Analysis Preliminary and final Regulatory Impact Analyses of major rules should contain five elements. They are: (1) a statement of the potential need for the proposal, (2) an examination of alternative approaches, (3) an analysis of bener'ts and coats, (4) the rationale for choc - z the pr ^^eed regulatory action, and (5) a etatr -Qt of statutory authority. These elements are expired in Sections I-V below. I. STATEMENT OF POTENTIAL NEED FOR THE PROPOSAL In order to establish the potential need for tike proposal, the analysis should demonstrate that (a) market failure exists that is (b) not adequately re* solved by measures other than Federal regulation. A. Market Failure The analysis should determine whether there exists e market failure that is likely to be significant Ones such market failure has been identified, the analysis should show how adequately the regulatory alterna tives to be considered address the specified market failure. The three major typas of market failure are externality, natural monopoly, and inadequate infor mation. 1. Externality. An externality occurs whan one party's actions impose uncompensated benefits or costs on another outside the marketplace. Environ mental problems are a classic case of externality. Another example is tin ease of common property resources that may become congested or overused, such as fisheries or the broadcast spectrum. A third cxampla is a "public good," such as defense or scien tific research, whose distinguishing characteristic is that it is iaeffidant, or impossible, to exclude individ uals from its benefits. 2. Natural monopoly. Natural monopoly exists whre a market can be served at lowest cost only if prr-'-vrtion is limited to a single producer. Local tele `'one, gas, and electricity services ere examples. 8. ladtquate information. The optimum, or ideal, leva] of information is not necessarily the maximum possible amount, because information, like other goods, should not be produced when the costs of doing so exceed the benefits. The free market does not 653 SL 062689 654 REGULATORY PROGRAM OF THE UNITED STATES OOVBUOffiNT Mcwaifly fupply an optimal level of information, because information, once generated, can be die* aeminated at little or no marginal cost, and because it is commonly infeasible to exclude nonpayers from reaping benefits from the provision of information by others. Where market failure due to inadequate in formation is the rationale for government inter vention, a regulatory action to improve the availabil ity of information will ordinarily be the preferred alternative. The current state of knowledge about the econom ics of information is not highly developed. Therefore, regulatory intervention to address an information problem should only be undertaken where there is substantial reason to believe that private incentives to provide information are seriously inadequate and that the specific regulatory intervention proposed will provide net benefits for society. In many circumstances, the availability of informa tion, while perhaps not optimal, is reasonably ade quate, so that attempts to regulate information are as likely to make things worse as to make them better. Information about a particular characteristic of a product, for example, would be reasonably adequate if buyers could determine the existence of the charac teristic by inspection of the product before purchase or (in the case of a frequently purchased product) by use of the product Even if the characteristic could not be determined by buyers, government interven tion would not be warranted where sellere have incentives to reveal the existence of the characteristic to buyers. Sellers will have substantial incentives to supply information about any characteristic that is important to buyers and valued positively by them, particularly if the level of the characteristic varies between the products of one seller and another. In these circumstances, sellers whose products rank highly in the valued characteristic can increase their sales by informing buyers of the superiority of their products. If the level of the characteristic does not vary between the products of one seller and another, individual sellers have lew incentive to inform buyers about the characteristic. Even so, the incentives of individual sellers or of a trade association to supply information may be subatantiaL Sellere are least likely to supply adequate informa tion about a particular characteristic of their product where the characteristic is negatively valued by con sumers and the level of the characterstic does not vary between the products of one seller and those of another (e.g., cholesterol in eggs). Even in such cir cumstances, substantial information about the char acteristic may be available to buyers. For example, sellers of rival products may supply the information (e.g., while sellers of butter may have no incentive to tell buyers about cholesterol in butter and its consequences, sellers of margarine do have incentive). Where the negative characteristic unv-" a health or safety hazard, the threat of foturv JZ net liability lawsuits may give sellers adequate tivss to reveal information about the potential hu ard. News media, consumer groups, public h*ha agencies, and similar services may supply inJiZ tion not supplied by sellers. In summary, while it possible to identity situations in which market fadun due to inadequate information is more likely to war rant regulatory intervention, each situation must be examined on a case-by-case basis. There should be a presumption against the need for certain types of regulatory actions, except in special circumstances. A particularly demanding burden f proof is required to demonstrate the potential nasd for any of the following types of regulations: Price controls in competitive markets Controls on production or salts in competitive markets Mandatory uniform quality standards for goods services, unless they have hidden safety or other defects and the problem cannot be adequately dealt with fay voluntary standards or information disclosing tbs hazard to potential buyers or users Controls on entry into employment or production, except (a) where indispensable to protect health and safety (e.g., FAA tests for commercial pilots) or (b) to manage the use of common property resources (eg., fisheries, airwaves, Federal lands, and offshore areas). B. Alternatives to Federal Regulation Even where a market failure exists, there may be no need for Federal regulatory intervention if other means of dealing with the market failure resolve the problem adequately or better than the proposed Fed eral regulation would. Among the alternative means that may be applicable are the judicial system (particularly liability cases to deal with health and safety), antitrust enforcement, and workers' compen sation systems. An important alternative that may often be rele vant is regulation at the State or fool level In determining whether there exists a potential need for a proposed Federal regulation, the analysis should examine whether regulation at the Federal level is more appropriate than regulation at the State or local level. This analysis may support regulation at the Federal level when rights of national citizenship (such as legal equality among the races) or considera tions of interstate commerce are involved. If inter state commerce is involved the analysis should at tempt to determine whether the burdens on sL 062690 APPENDIX V 655 ggrtUte commerce anting from different State and tal regulations are eo great that they outweigh the ^taatagee of diversity and local political choke. In me cases, the nature of the market failure may gielf suggest the most appropriate governmental b*e! of regulatien^For example, pollution that spills mu state lines (such as add rain whose precursors at transported widely in the atmosphere) is probably tat controlled fay Federal regulation, while localised jdhition (such as garbage truck noise) is probably an efficiently handled by local government rogule- b general, because demands among localities for fffirent governmental aarr differ and bee at anpetition among governor units for taxp* ~s ad citizens may encourage sect ragulatior e maDest unit of government ole of correctir e market failure should be cha : This must, bo* r, fat balanced against the poesibility of higher u tasose national firms would be required to cc oly with more than one set of regulations and bat. use administering similar regulations in more than one governmental unit involves some coats of duplication, lfaui, some analysis may be necessary to determine which level of government can most efficiently regu late s specific market failure. If the analysis does suggest e potential nsed for a Federal action, it should also consider alternatives of soaregulstory Federal measures. For axample, as an shemative to requiring an action or the use of a particular product, it may be more efficient to subeiie a Similarly, a fee or charge may be a preferable alternative to banning or restricting a product or action. An axample would be an effluent discharge it, which has been recommended as an efficient way to limit pollution, because it causes pollution sources with different marginal coats of abatement to control (ffiuesu in an efficient manner, ho addition, legisla tor* measures that make use of economic incentives, nth u changes in insurance provisions or changes in property rights, should be considered. IL AN EXAMINATION OF ALTERNATIVE APPROACHES lbs R1A should show that the agency has considtd the most important alternative approaches to the problem and must provide the agent's reasoning hr selecting the proposed regulatory change over mch alternatives. Ordinarily, it will be possible to Juninste tome alternatives by a preliminary analytot, leaving a manageable number of alternatives to k evaluated by quantitative benefit-cost analysis rowding to the principles t< ye described in Section OL The number and choice of alternatives to be elected for detailed benefitieost analysis is unavoid ably A matter of judgment There must be balance between thoroughness of analysis and prac tical limits to the agency's capacity to carry out analysis. Alternative regulatory actions that should be ex plored include the following: L Afore performance-oriented etandardt for health, paftty, and environmental regulation*. Performance standards are generally to be preferred to engineer ing or design standards because they allow the regu lated parties to achieve the regulatory objective in the most cost-effective way. In general, a performance standard should be preferred wherever that perform nance can be measured or reasonably imputed. Per formance standards should also be applied as broadly as possible without creating too much variation in regulatory benefits; for example, by setting emission standards on a plant-wide or firm-wide basis rather source by source. It is misleading and inappro priate, however, to characterize a standard as a performance standard if It is sat so that them la only on* feasible way to meet it; as a practical matter, such a standard k a design standard. 2. Different requirement$ far different tegmenta of the regulated population. For example, them might be different requirements for large and small firms. If such a differentiation is mads, H should be baaed on perceptible differences in the costa of compliance or in the benefits to be expected from compliance. For example, some worker safety measures may exhibit economies of scale, that is, lower costa per worker protected in large firms than in small firms. A heav ier burden should not be placed on on* segment of the regulated population on the grounds that it is better able to afford the higher coat; this is a turn formula for loading disproportionate coats on the most productive sectors of the economy. 8. Alternative level* of etringency. In general, both tha benefits and coats associated with a regulation will increase with tha level of stringency (although costa w(D eventually increase more rapidly than bene fits). It is important to consider alternative levels of stringency to better understand the relationship between stringency and benefits and costa. This approach will increase the information available to the decisionmaker on the option that maximizes net benefits. 4. Alternative effective date* of compliance. The timing of a regulation may also have an important affect on he net benefits. For axample, costs of a regulation may vary substantially over different com pliance dates for an industry that requires a year or more to plan its production runs efficiently. In this instance, a regulation whose requirements provide. SL 062691 656 REGULATORY PROGRAM OF THE UNITED 8TATES GOVERNMENT sufficient lead tine is likely to sdiieve its goals at a much lower overall cost than a regulation that is effective immediately. 0. Alternative method* of ensuring compliance. Compliance alternatives include the appropriate en tity (local, State, or Federal) enfordng compliance, whether compliance is enforced by on-site inspection or periodic reporting, and structuring compliance penalties so that they provide the most appropriate incentives. 6. Informational measures. Measures to improve the availability of information include government establishment of a standardised testing and rating system (the use of which could be made mandatory or left voluntary), mandatory disclosure requirements (e.g., by advertising, labeling, or enclosures), and government provision of information (e.g., by govern ment publications, telephone hot-lines, or public in terest broadcast announcements). If intervention is necessary to address a market failure arising from inadequate information, informational remedies will generally be the preferred approaches. As an alterna tive to a mandatory standard, a regulatory measure to improve the availability of information has the advantage of being a more market-oriented approach. Thus, providing consumers information about con cealed characteristics of consumer products gives con sumers a greater choice than banning these products (for example, consumers are likely to benefit more from information on energy efficiency than from a prohibition 0ri tale of appliances or automobiles foil ing below a specified standard of energy efficiency). Except for prohibiting indisputably folse state ments (whose banning can be presumed beneficial), specific informational measures must be evaluated in terms of their benefits and costs. Paradoxically, the current state of knowledge does not generally permit the benefits and costs at informational remedies to be measured very accurately. Nonetheless, it is essential to consider carefully the costs and benefits of alterna tive informational measures, even if they cannot be qU^wiL_eu very precisely. Some effects of informa tional measures can easily be overlooked. For exam ple, the costs of a mandatory disclosure requirement for a consumer product include not only the obvious cost of gathering and communicating the required information, but also the lots of any net benefits of information displaced by the mandated information, the cost of any inaccurate consumer interpretation of the mandated information, and any inefficiencies arising from the incentive that mandatory disclosure of a particular characteristic gives to producers to overinvest in improving that specific characteristic of their products. Where information on the benefits and costs of alternative informational measures is insufficient to provide a dear choice between them, as will often be the case, the least intrusive alternative, sufficient to eccomplish the regulatory objective, should be chosen. Far example, it will often be sufficient for government to establish standardised testing and rating system without mandating its use, because firms that score well according to the system will have ample incen tive to publicise the foct 7. More marketorisnted approaches. In general, alternatives that provide for more market-oriented approachee, with the use of economic incentives re placing command-and-control requirements, should be explored. Market-oriented alternatives that may bt considered include fees, subsidies, penalties, marketable rights or offsets, changes in liabilities or property rights, and required bonds, insurance or warranties (in many instances, implementing these alternatives will require legislation). m. ANALYSIS OF BENEFITS AND COSTS A. General Principles Hie preliminary analysis called for by Sections I and II should have narrowed the number of alterna tives to be considered by quantitative benefit-cost analysis to s workable number. Ordinarily, one of the alternatives will be to promulgate no regulation at all, and this alternative will commonly serve as the base from which increments in benefits and costs sit calculated for the other alternatives. Even if alterna tives such as no regulation are not permissible statu torily, it is often desirable to evaluate the benefits and costs of such alternatives to determine if statu tory change would be desirable. Departments and agencies bear a similar burden when they perform environmental impact statements in which alterna tives that lit outside their statutory authority must be considered. In some cases, the desirability of specific alterna tives outside the scope of the agency's regulatory authority may be determined by use of basic eco nomic concepts in light of the principles enumerated in Section L In other instances, however, only a quantitative benefit-cost analysis fS" resolve the question, and such alternatives wifi need to be in cluded in the analysis of section. In addition, alternative forms of agency regulation wifi need to be evaluated by quantitative benefit-cost analysis. 1. Evaluation ofAlternatives. Except where prohib ited by law, the primary criterion for choice among alternative* is ejected net benefit (benefits minus costa). Other criteria may sometimes produce equiva lent results, but they must be used with care to avoid SL 062692 APPENDIX V 667 potf"flny serious pitfalls to be explained in Part I of this section and in Section IV. Both benefit* and 0ti should be expressed in discounted constant feUin. Appropriate discounting procedures are die- in the following section. 1fce distinction between benefits and costs In bene* fbeost analysis is somewhat arbitrary, since a posign benefit say be considered a negative cost, and wee versa, without affecting the net benefit (benefit! wjnus costs) decision criterion. This implies that the asuiderations applicable to benefit estimates also apply to costs and vice versa. The different issues are mtodcred separately under benefit! or cocta in Sectoms B end C below according to where they most dnir^. IT the 'opoted regulation ia composed of a number to distinct provisions, it is important to evaluate the benefits and costs of the different provisions sepaistely. The interaction effects between separate provitoms (such that the existence of one provision affects toe benefits or coats arising from another provision) nay complicate the analysis but does not eliminate toe need to examine provisions separately. In such a cue, the desirability of a specific provision may be tppraised by determining the net benefits of the proposed regulation with and without the provision in question. Where the number of provisions to large and interaction effects are pervasive, it to obviously topraetieal to analyze all possible combinations of provisions in this way. Some judgment must be used to (elect the most significant or suspect provisions for such analysis. 1 Discounting. The monetary values of benefits ami cuks occurring in different years should be discounted to their present values so that they are comparable. This to not the same as correcting for mfiSLuu. An inflation adjustment to mede with a price index, whereas discounting to present value to dene with discount rate. Benefite and coats axpressed in constant (i.e., unaffected by inflation) dol lars must further be discounted to present values before benefits and costs in different yean can be added together to determine overall net benefits. As a equivalent alternative to discounting non* ontu.ro benefits, the RIA may use the discount rate to annualize (amortize) costs over a period that (responds to the occurrence of the benefits. Regardlw of the discounting procedure selected, the RIA ust contain a schedule indicating *n the benefits sad costa occur. Discounting takes account of the f.. that resources ((nods or services) in a given year are worth more than identical resources in a later year. The underly ing reason for this is that resources can be invested * as to return more resources later. Partly because of productivity of investment, individuals value consumption in earlier yarn higher then consump tion in later yean. Modern analysis of discounting for public programs stresses the distinction between two rates of return: The btfora4ox rate, also known as the opportunity coat of capital. This to the real rate of return to marginal private investments. Eatimatoc of the opportunity coat of capital in the U.8. economy vary substantially. The 10 percent discount rate pacified by OMB Circular A-94 for use in evalu ating government programs to intended to repre sent the opportunity cost of capital. The q/teMox rata, also known as the consumption rate of interest This represents the rate at which consumers would be willing to exchange present for future consumption, that to, the rate at which consumers must be compensated for postponing their consumption. As with the opportunity cost of capital, alternative estimates of toe consumption rate of interest vary significantly. A rate of 4 percent to reasonably representative of toe range of alternative estimates end consistent with a 10 percent before-tax rate ofreturn. The basic concept underlying the academic litera ture on public-sector discounting to that economic welfare to ultimately determined by consumption and only indirectly by investment Therefore, toe value of investment must be measured by toe value of toe subsequent inowaae in consumption it permits. Any effect that a government program has on investment must be converted to an equivalent time-stream of consumption before being discounted. In practice, this results in a complex procedure that usee the before tax and after-tax discount rates, a "shadow price of capital," and toe impacts of benefits and costs on investment It to recommended that agendas continue to use toe well-understood procedure of discounting by a single rate (as specified by OMB Circular A-84) and, when appropriate, perform additional analysis using tos more complex shadow-price-af-capital meth odology. There are two circumstances whan ft to important to perform sensitivity analysis using the shadow price of capital approach: (a) Where the costa of tos regulation ere almost entirely current costs borne by consumers, to such circumstances, s low rate dose to 4 percent to called for. (This assumes, as to normally toe case, that the benefits are all in the form of disposable income or other benefits directly to individuals.) (b) Where some of the costs are capital costa financed out of saving and there to e long period between the time when most costs are incurred and toe time when most benefits accrue, to general, the SL 062693 656 REGULATORY PROGRAM OP THE UNITED STATES GOVERNMENT tmalltr the fraction of costs that are capital costs financed out of saving and the longer the time period between costs and benefits, the greater the likelihood that the shadow price of capital approach will be correct It is conceptually incorrect to adjust the discount rats as a device to account for the uncertainty of expected fixture benefits and costs. This procedure will virtually never lead to a correct adjustment of benefits and costs. Therefore, risk and uncertainty should be dealt with according to the principles in Section 3 below and not by changing the discount rate. 3. Haatmmt ofRisk and Unetrtainty. Where uncar* tainties exist about important parameters affecting the expected benefits or costs of an alternative under consideration, it is essential to carry out a acnsitivity analytu to determine the effect on net benefits of plausible variations in the value of the parameters. One form of sensitivity analysis involves calculation of the `switch-point' value of the parameter under examination, that is, the value of the parameter at the break-even point at which the net-benefit decision criterion switches over from favoring one alternative to favoring another. When this break-even point of the parameter value is determined, the analysis may then consider the probability that the true parameter value is above or below the break-even value. For example, if the major uncertainty about a proposed regulation were its cost, the analysis could calculate how high the cost would need to be in order to reduce the net benefit of the proposal to zero. If it is judged to be highly unlikely that the actual cost would be that high or higher, it may be concluded that the choice of the proposed alternative is not sensitive to uncertainties about its cost A primary objective of sensitivity analysis is to identity where additional analysia may be moat needed. If the choice of a specific regulatory action is sensitive to alternative parameter values that are about squally likely to be true, more research to better determine thus true parameter value could bo vary valuable. Wherever parameter astimatee are uncertain, for either benefits or costs, expected-value estimates should bs presented. Hypothetical best-case or wontease estimates may be presented as alternatives for ensitivity analysia. Where possible, information about the probability distribution of the parameter estimate should be presented. A common situation that arises in estimating both benefits and costs is that a number of different studies may exist which together provide e range of different estimate# for a particular parameter. In general, it is not appropriate to use the midpoint of the range of extreme values provided by the studies. Such a technique ignores the information provided fay all studies except those providing the extreme valum, which may be the least reliable. The preferred ap proach to deriving an expected-value estimate of a particular parameter in this situation would be to derive it aa a weighted average of the estimates of the individual studies, with the weight of each esti mate being based an the reliability (in the best judgment of the agency) of the study that produced it Where expected fixture benefits or costs ara un certain, their value to those who receive them may be different from their value if they were certain. (Often, but not always, a certain fixture benefit is worth more to people than an uncertain fixture benefit with the same expected value.) As noted in the previous section, it is incorrect to adjust the discount rate as a device to account for the riskiness of fixture benefits or ooste. Any allowance for risk should be made by adjusting the monetary values (for the year in which they occur) of the uncertain benefits and costs so that they are expressed in terms of their `certaintyequivalents." Far an uncertain benefit in fixture year X the certainty-equivalent is the number of certain dollars in year X that the uncertain benefit is worth to its recipient For example, suppose that a particular regulation reduces the probability of fire in a particu lar type of facility. As part of a benefit-cost analysis for this regulation, the dollar value of the expected reduction in fire lose would be calculated. Hie owners of the protected facilities place a higher dollar value on tfao risk of a fire than the expected dollar value of the Iocs. This is demonstrated by their wfllingneas-to* pay for firs insurancs. Therefore, their relative net coat (the percentage difference between insurance premiums and insurance company payments) for fire insurancs can be used to increase the ex pected dollar value of the reduction in fire loos to its certainty-equivalent value. In the example of the preceding paragraph, the adjustment for risk would involve an inoease in the value of the benefit, whereas uncertainty of a benefit is normally thought to reduce its certainty-equivalent value. The reason it that even though this benefit fay itself is uncertain, it acts to reduce the overall level of risk that would prevail in the absence of the regulation. This illustrate* the important principle that what matters is not the variability or riskiness of a regulation's not benefits by themselves but the regulation's effect on risk and uncertainty overall. While an adjustment to account for risk may be called fbr in the fire-risk example givan, a similar adjustment for the value of reductions in fatalities and injuries would not be appropriate. Assuming that SL 062694 APPtNDHV 669 UK tilue* of fatalities tad injuries have been derived ^ the willingneas-to-pay methodology recommended to Section B.2 below, they would already repreaent toe certainty-equivalent value of the uncertain risk., fhis i because the estimated dollar values represent toe certain dollar amounts that individuals would office to reduce these risks. Probably, in most cases, it will not be advisable to adjust for risk and uncertainty. As a theoretical Bitter, no adjustment for risk is necessary wherever toe net benefits are widely dispersed among many individuals and are not correlated with disposable tome. And in cases where this does not apply, risk ay be relatively unimportant or may already be taken into account by use of the wfllingnees-to-pay methodology. In onr.er cases, there may be no practi cal way to quantity the value of changes in risk. 4. Auumption*. Where benefit or cost estimates are heavily dependent on certain assumptions, it is atfwtiai to make these assumptions explicit and, where alternative assumptions are plausible, to carry sot sensitivity analyses based on plausible alterna tive assumptions. If the decision criterion proves to he sensitive to alternative plausible assumptions, this any necessitate further research to develop more evidence on which of the alternative assumptions is toe most appropriate. Because the adoption of a particular estimation methodology sometimes implies major hidden assumptions, it is important to analyse estimation methodologies carefully to make hidden assumptions explicit 6. International Trade Effect!. In calculating the benefits and costs of a proposed regulatory action, generally no explicit distinction needs to be made between domestic and foreign resources. If, for example, compliance with a proposed regulation re quires the purchase of specific equipment the oppor tunity cost of that equipment is ordinarily best repre sented by its domestic cost in dollars, regardless of whether the equipment is produced domestically or imported. The relative value of domestic and foreign resources is correctly represented by their respective dollar values, as long as the foreign exchange value of file dollar is determined by a free exchange market Nonetheless, an awareness of the role of international trade may be quite useful for assessing the benefits and costs of a proposed regulatory action. For exam ple, the existence of foreign competition usually makes the demand curve facing a domestic industry more elastic than it would be othervr Elasticities of demand and supply frequently c* significantly affect tbe magnitude of the benefits or costs of a regulation. A regulation that discriminates unjustifiably gainst foreign exporters is a form of economic pro tectionism. The economic loee to the U.S. due to tbe fact that protectionism is economically inefficient will be reflected in the net benefit eetimate of any prop erly conducted benefit-cost analysis. However, a bene fit-cost analysis wQl generally not be able to measure the potential UJ3. loaf from the threat of fature retaliation fay foreign governments. Therefor*, special attention should be given to any possibility tost a regulation would unjustifiably discriminate between domestic and foreign producers and consumers both discrimination against tonigners and tefavor of foreigners. The fact that a regulation has a differential effect on foreigners as compared to Americans docs not necessarily constitute discrimination. Ifc for example, an automobile safety standard could be complied with less expansively by large cars than by small ears, such a standard would be more favorable to American car producers, who produce relatively more large cars compared to toe fleet mix of foreign producers. None theless, such a differential efibet would not fca dis criminatory if the difference in compliance cost between large and email cars was necessary to achieve legitimate regulatory objectives in toe most efficient way. If a regulation has an advent differential effect on foreign producers or consumers relative to domestic producers and consumers that is not necessary to realize regulatory goals efficiently, then a discrimina tory affect on foreign trade exists. The RIA should identity any substantial differential efibet on interna tional trade and explain why it is necessary to achieve legitimate regulatory goals in the moat effi cient way. One means for reducing toe likelihood of international discrimination would be for a U.S. prod uct standard for an internationally traded good to be based on an international standard, wherever an international standard exists and is compatible with the health, safety, or environmental needs of the Uf>. International harmonization can be beneficial for reg ulations directly setting standards for internationally traded goods or services. For example, it would be appropriate to consider international harmonization to setting safety standards for automobiles. There is no similar advantage to international harmonization where a regulation does not directly affect the quality of an internationally traded good or service, even if it indirectly affects its costs ^s-g., environmental con trols for automobile plants) 6. Distributional Effeete. e who bear toe costs of a regulation and those are not toe same perec ajoy its benefits often Jenefit* and costs of regulation may also be d^cributed unevenly over time, perhape spanning several generations. There is no generally accepted way to monetize potential SL 062695 860 REGULATORY PROGRAM OF THE UNTIED STATES GOVERNMENT distributional effects. Attempts to incorporate dis tributional concerns in benefit-cost analysis require the establishment of unequal weights for different groups in society. Because positive economics treats equally the willingness-to-pay of all individuals, any alternative weighting would undermine the objective character of the analysis. Policymakers may wish, however, to take account of the distributional effects of various regulatory alternatives. Therefore, where there are potentially important differences between those who stand to gain and those who stand to lose under alternative regulatory options, the RIA should identify these groups and indicate the nature of the differential effects. The RIA should also present infor mation on the streams of benefits and costs over time as well as present value estimates, particularly where intergenerational effects are concerned. B. Benefit Estimates The RIA should state the beneficial effects of the proposed regulatory change and its principal alterna tives. In each case, there should be an explanation of tee mechanism by which tee proposed action is ex pected to yield the anticipated benefits. An attempt should be made to quantify all potential real incre mental benefits to society in monetary terms to tee maximum extent possible. A schedule of monetised benefits should be included that would show the type of benefit and when it would accrue; the numbers in this table should be expressed in constant, undiseounted dollars. Any expected incremental benefits teat cannot be monetized should be explained. The RIA should identify and explain in detail the data or studies on which benefit estimates are based. Where benefit estimates are derived from a statistical study, tee RIA must provide sufficient information so that an independent observer can determine the rep resentativeness of the sample, whether it was extrap olated from properly in developing aggregate esti mates, and whether the results are statistically significant. For regulations addressing health and safety risks, the calculation of potential benefits should derive from the agency's estimate of the mean expected value of the reduction in risk attributable to the standard. Estimates of the prevailing level of risk and of the reduction in risk to be anticipated from a# proposed standard should be unbiased expected-value estimates rather than hypothetical worst-case este/ mates, Extreme safety or health results should be weighted (along with intermediate results) by the probability of their occurrence to estimate the ex pected result implied by the available evidence. In addition, to the extent poeaible, tee distribution of probabilities for various possible results should be preeented separately, so as to allow for an explicit margin of safety, where required, in final decisions. If a margin of safety is to be provided, the proper p--t for it is the final stage of tee decision-making pr oass, not by adjusting the risk or benefit estimates in a conservative direction at the information-gatherug or analytical stages of tee process Conservative esti mates should be presetted as alternatives to best estimates for sensitivity analysis but should not sub stitute for than. It is important to guard against double-counting of benefits. For example, if a regulation improved the quality of the environment in a community, the value of real estate in the community might rise, reflecting tee greater attractiveness of living in tea improved environment It would ordinarily be incorrect to in clude the rise in property values among the benefits of the regulation. Ordinarily, the value of environ mental benefits (e.g., reduced health risks, scenic improvements) will already be included among the benefits. The rise in property values reflects the capitalized value of these improvements. Therefore, to count as benefits both the value of the environ mental improvements and tea corresponding increase in property values is to count the same benefits twice. Only where a direct estimate of the benefits has not been included would it be appropriate to include the increase in property values among the benafita. 1. General Considerations. The concept of "opportu nity cost" is tee appropriate construct for valuing both benfits and costa. The principle of "willingness- to-pay" captures the notion of opportunity cost by providing an aggregate measure of what individuals are willing to forgo so as to enjoy a particular benefit Market transactions provide tee richest database for estimating benefits based on willingness-to-pay, so long as the goods end services affected by a potential regulation an traded in markets. Estimation prob lems arise in a variety of '"^n--, of course, where prices or market transactions are difficult to monitor. Markets may not even exist in some for- ring regulatory analysts to develop appropriate prox ies that simulate market exchange. Indeed, the ana lytical process of deriving benefit estimates by simulating markets may suggest alternative regula tory strategies that create such markets. Willingness to pay always provides the preferred measure of benefits. Estimates of willingness-to-pay based on observable end replicable behavior deserve the greatest level of confidence. Considerably lea confidence should be conferred on benefit estimate* that are neither derived from market transactions nor based on behavior that is observable or replia* ble. Of course, innovative benefit estimation method SL 062696 APPENDIX V 661 ologies may be necessary in eome cams, and should be encouraged. However, reliance upon such methods intensifies the need for quality control to ensure that estimates derived conform as closely as possible to what would be observed if markets existed. 2. Principles for Valuing Dirtctly Observable Bene fits. Ordinarily, goods and services are to be valued at their market prices. However, in eome instances, the market value of a good or service may not reflect its true value to society. If a regulatory alternative involves changes in such a good or servios, its mone tary value for purposes of benefit-cost analysis should be derived using an estimate of its true value to society (often called its "shadow price"). For example. Suppose a particular air pollutant damages crops. One of the benefits of c oiling that pollutant will be the value of the cr sved as a result of the controls. If the price of .t crop is held above the five-market equilibrium * ce by a government price- support program it will overstate the value of the benefit of controlling the pollutant if the crop saved were valued at the market price established by the support program. The social value of the benefit should be calculated using a shadow price for crops subject to price supports. The estimated shadow price should reflect the value to society of marginal uses of the crop (e.g., the world price if the marginal use is for exports). If the marginal use is to add to vary large surplus stockpiles, the shadow price would be the value of the lest units released from storage minus storage cost. Therefore, where stockpiles are large and growing, the shadow price is likely to be low end could well be negative. S. Principles for Valuing Benefits that art Indirectly Traded in Markets. In eome important instances, a benefit corresponds to a good or service that is indirectly traded in the marketplace. Important ex amples include reductions in the health-and-safety risks, the use-value of environmental amenities and scenic vistas, and savings in time, lb estimate the monetary value of such an indirectly traded good, the willingness-to-pey valuation methodology is still con ceptually superior, because the amount that people are willing to pay for a good or service is the best measure of ita_valu* to them. As noted in Sections 4 and 6 immediately following, alternative methods may be used where there are practical obstacles to the accurate application of direct willingness-to-pey methodologies. A variety of methods have been developed for estimating indirect benef apply statistical techniqi market transactions the Generally, these methods distill firom observable ;n of willingness-to-pey that can be attributed :e benefit in question. Examples include estimates of the value of environ mental amenities derived fr i travel-cost atudies^fe hedonic price models that r. assure differences changes in the value of land, and statistical studies of occupational-risk premiums in wage rates. Contingent-valuation methods have become in* craaaingly popular for estimating indirect benefits, hut they suffer from the fret that survey instruments have a capacity to simulate real-world market behavior. Benefit estimate* derived from contingent- valuation studies thus have a greater burden of analytical ears to nsura that they rapraaent in an unbiased what actually occurs in the market place. .4 Principles and Methods for Using Benefits that are Not Traded Dirtctly or Indirectly in Markets. Bears types of goods, such as the social benefit of preserving environmental amenities apart from their ua* and direct enjoyment by people, are not traded directly or indirectly in markets. The practical obsta cles to accurate measurement an similar to (but genorally more severe than) those arising with re spect to infract benefits, principally because there are not market transactions to provide data for will- ingnesa-to-pay estimates. Contingent-valuation methods provide the only an alytical approaches currently available for estimating the benefits of such untraded goods. The absence of_ observable and replicable behavior with respect to benefit in question, combined with the difficulties avoiding bias in contingent-valuation studios, argues for great can and circumspection in the use of such methods. This means, for example, that estimates of willingness-to-pay must incorporate the variety of alternative means individuals have of expressing value for untraded goods. Moreover, analyses must fcithftiHy capture individuals' budget constraints, which restrict their willingness-to-pay for untraded as wall as traded goods and services. Benefit analyses derived from contingent valuation and similar meth ods thus require considerable analytic rigor in design and carafol execution. Absent such efforts, analyses based heavily on the benefits of untraded goods and services ordinarily would foil the test of a satisfactory BXA. 6. Methods far \bluing Health and Safety Benefits. For health and safety benefits, a distinction should be made between risks of Donfetel illness or injury and fotality risks. (a) Nonfatal illness and injury. Although the will- ingneas-tc-pay approach is conceptually superior, the currant state of empirical research in the area is not sufficiently advanced to assure that estimates derived by this method are necessarily superior to direct-cost valuations of reductions in risks of nonfatal illness or injury. Any injury-value estimate from a willingnes SL 062697 662 REGULATORY PROGRAM OP THI UNITED STATX8 GOVERNMENT to-pey study i necessarily an average over a apadfie appropriate to value reductions In risk from that high cambinatioD of injuries of varying severity. If the level more highly than equal risk reductions at lower average injury severity in such a study is greatly risk levels. In general, the greater the risk that an different from that for the regulatory action under individual bears, the higher will be the value the study, then the study's estimated injury value may individual places on marginal changas in risk. On the not be appropriate for evaluating that action. Accord* other hand, where a high risk is chosen voluntarily ingly, the agency ahould use whichever approach it those assuming the risk tend to be persons who piece considers most appropriate for the decision at hand. a relatively low value on averting safety risks. Empir The primary components cf the direct-oast approach ical studies of risk premiums in high-risk occupations are medical costs and the value of lost production. suggest that reductions in voluntarily assumed high Possibly important costs that may be emitted by the risks should be valued lees than equal risk reductions use of tbc direct-cost approach are the value of pein at ordinary risk levels. and suffering and the value of time lost from leisure and other activities that are not economically directly productive. Estimates of the value of fatality risks refer only to changes in an uncertain risk of death. They have no application to the certain prevention of the death of (b) Fatality. Reductions in fatality risks are bast an identifiable individual. monetized according to the willingneee-to-pay ap proach. The value of changes in fatality risk is sometimes expressed in terms of the "value cf life." Ibis is something cf a misnomer since the value of a life really refers to the sum of many small reductions in fatality risk. For example, if the annual risk of death is reduced by on* in a million for each of two million people, that represents two "statistical Uvea" saved per year (two million x one millionth two). If the annual risk of death is reduced by one in 10 million for each of 20 million people, that also repre sents two statistical lives saved. The conclusion that the fatality risk reductions in these two cases an equivalent implies an assumption. The implicit as sumption--that equal increments in risk are valued equally--allows different risk increments to 1m added together and compared directly. As a different exam ple, suppose there are two alternative reductions in the annual risk faced by an individual: 6. Alternative Methodological Frameworks far Esti mating Health and Safety Benefits. Several alterna tive ways of incorporating fatality risks into the framework of benefit-coet analysis may be appropri ate. Those may involve either explicit or implicit valuation of fatality risks. Ono acceptable explidt valuation approach would bo for tha agency to select a single value for reduc tions in fatality risk at ordinary risk levels (below Iff4 annually) and use this value consistently for evaluating all its programs that affect ordinary fatal ity risks. Another acceptable explidt valuation ap proach would be to use a range of values for reduc tions in fatality risk and apply sensitivity analysis as with other parameters that have alternative plausible values. The range of alternative values should bs s reasonable one, not one that indudes the moat ex treme upper and lower values of fatality risk reduc tion that have boon estimated. Extreme values srs A* from .10 x 10** to .09 x 10"* .01 x Iff4 B: from 1.00 x Iff4 to .99 x Iff4 .01 x Iff4 mom appropriate for faatawo-- <*f extraordinarily high risks (above Iff4 annually), with the extreme lew values being appropriate wham voluntary assumption of high risk leads to aelf-eelection and the extreme Since in both easea the reduction in annual risk is high values being appropriate wham the high risk is tiie same (.01 x Iff4), the value of A end B should be involuntarily assumed. considered the same. Th# sumption that equal increments in fatality risk are of equal value is a legitimate one, so long as the level of fatality risk is below Iff4 annually. Thera is evidence that the willingness-to-pay value for Wham the analysis usee a range of alternative values for reductions in fatality risk, it may be useful to calculate break-even values, as in othar sensitivity analyses. This requires calculating the borderline value of reductions in fatality risk at which the net increments in fatality risk does not change signifi cantly over a wide range of risk exposure below Iff4 annually. benefit decision criterion would switch over from favoring one alternative to favoring another (Le., the value of fatality risk at which the net benefits cf tbs For levels of annual risk exposure of 104 and above two alternatives am equal). This method wiD fre it cannot be assumed that equal increments of risk quently be infeasible because of its computational are valued equally. At these higher risk levels, it is demands or because alternatives are continuous particularly important to distinguish between situa rather then discrete (e.g., alternative stringendes for tions of voluntary risk assumption and those of invol exposure levels), hut wham appropriate, it is a useful untary risk. Where the high risk is involuntary, it is supplement to the sensitivity analysis. SL 062698 T r wW p |'l'r APPENDIX V 663 As implicit valuation approach could entail calcula- producer is paid for a unit of a good and the mini of the coat per unit of reduction is filtality tick mum amount the producer would accept to supply (coat per "statistical life saved*), with costs defined as that unit It is measured by the distanoe between the costs minus monetised benefits. This must be used price and the supply curve far that unit Consumers' with care since there is a serious potential pitfall: It surplus is the difference between what a consumer fc mot correct to choose between two mutually axdu- pays for a unit of a good and the amount farve alternatives by selecting the alternative with the consumer would be willing to pay for that unit It lowest cost par statistical lift saved. Hie alternative to measured by the dtotanoe between the price and aritb higher cost per lift aaved may nonethalaea be the nA curve fin- that unit) Aa another example, the alternative with the higher net benefit to society. even if a resource required fay regulation does not ; The way to avoid this pitfall while retaining the have to be paid tor because it to already owned fay the implicit valuation approach is to make ell calcula regulated firm, nonethaleas, the use of that resource tions of cost per lift saved in terms of increments to meet the regulatory requirement has an opportu \ between alternatives. Alternatives should be arrayed nity cost equal to the net benefit it would have , in order of their total reduction in expected fatalities provided In the absence of the requirement Any such and the incremental cost per lift saved calculated foregone benefits tor an alternative should be mone between each adjacent pair of alternatives. In con tized wherever possible and either added to the costs trast to explicit valuation approaches, this avoids the or subtracted from the benefits of that alternative. * necessity of specifying in advance a value for reduc Any costs that art availed as a mult of an alterna tions in fatality risks. However, a range of values will tive should be monetised wherever possible and ei i be implied by the final selection of an alternative. ther added to the benefits or subtracted from the ; This range should be consistent with estimated val coats of that alternative. ues of reductions in fatality risks calculated according AH costs calculated should be incremental, that to, to the willingness-to-pay methodology. they should represent changes in costs that would Another way of expressing reductions In fatality occur if the regulatory alternative to chosen compared risks is in terms oflife-years saved. For example, if a to coats in the base case (ordinarily no regulation or regulation protected individuals whose average re the existing regulation). Future eoets that would be maining lift expectancy was 40 years, than a risk incurred even if the regulation to not promulgated, aj^to reduction of one totality would be expressed as 40 well as costs that have already been incurred (sunEV life-years saved. Such s refinement may be desirable eoets), are not port of incremental eoets. If marginal fin- regulations that disproportionately protect young coat to not constant tor any component of costs, people (e.g., motor vehicle safety regulations) or el- incremental costs should be calculated as the tree derly people (e.g., regulations controlling cardno- under the marginal oost curve ever the relevant a*) Tw derive the value of a life-year saved from an range. estimate of the value of lift, first determine the average remaining lift expectancy of the sample pop ulation in the study from which the estimate was drawn. Assuming that the average age of the sample population is known, the average remaining lift ex pectancy may be derived from actuarial tablas giving lift expectancy in relation to age. Using standard compound interest tables, the virius of a lift-year aaved can then be determined as the eetimated value of lift annualized over s period equal to the number <M uf remaining average lift expectancy. Coats indude private-sector compliance costa, gov ernment administrative costs, and costs of reallocat ing worker* displaced as a reeult of the regulation. Costa that are not monetary outlays must be included and should be attributed a monetary value wherever possible. Such costs may indude toe value (opportu nity oost) of benefits foregone, losses in consumers' or producers' surpluses, discomfort or inconvenience, and loss of time. A schedule of monetised costs should be included that would ahow tbs type of oost and when H would occur; toe numbers in this table C. Coat Estimates should be expressed in constant, undiscounted dol lars. Any expected incremental costs that cannot be 1. General Considerations. The opportunity cost of monetised should be explained. An important type of an alternative is the value of the benefits foregone as ooet that often cannot be quantified to a slowing in a consequence of that alternative. For example, the tha rata of innovation or of adoption of new technol opportunity cost of banning a product (eg., a drug, ogy. For example, regulations requiring a oostly and food additive, or hazardous chemical) is the foregone time-consuming approval process tor new products or net benefit of that product. It is measured by changes new facilities may have such costs, as may regula in producers' and consumers' surpluses. (Producers' tions setting much more stringent standards for new surplus is the difference between the amount a facilities than existing coat. ~ SL 062699 It 664 irOULATORY PROGRAM Of THE UNITED STATES OOVtRNtfENT Two accounting cost concepts that should not ba countad as costs in benefit-cost analysis art intarast and dpradation..Tha tima valus of monsy is alraady aceountad for by the discounting of benefits and coats. Depredation is already taken into account by the time distribution of benefits and coats; the only legiti mate use for depredation calculations in benefit-cost analysis is to estimate the salvage value of a capital investment. 8. Beal Costs oereus Transfer Payments. An impor tant, but sometimes difficult, problem in cost estima tion is to distinguish between real costs and transfer payments. Transfer payments are not genuine costs but payments for which no real good or service is received in return. Several examples of problems that may arise from the confusion between transfer pay ments and real coots (or benefits) may help to iden tify situations in which forther analysis of the prob lem may be warranted. Monopoly profits, insurance payments, government subsidies and taxes, and dis tribution expenses are four potential problem areas. (a) Monopoly profitt. If, for example, sales of a competitively produced product were restricted by a government regulation so as to raise prices to con sumers, the resulting monopoly profits are not a benefit of the rule, nor is their payment by consumers a cost The real benefit-cost effects of the regulation would be represented by changes in producers' and consumers' surpluses. (b) Inturanet payments. Potential pitfalls in bene fit-cost analysis may also arise in the case of insur ance payments, which are transfers. Suppose, for example, a worker safety regulation, by decreasing employee injuries, led to reductions in firms' insur ance premium payments. It would be incorrect to count the amount of the reduction in insurance pre miums as a benefit of the rule. The proper measure of benefits is the value of the reduction in worker injuries, monetised as described previously, plus any reduction in real costs of administering insurance (such as the time of insurance company employees needed to process claims) due to the reduction in worker insurance, claims. Reductions in insurance premiums that are matched by reductions in insur ance claim payments are changes in transfer pay ments, not benefits. (c) Indirect taxes and subsidies. A third instance where special treatment may be needed to deal with transfer payments is the case of indirect taxes (tariffs or exdse taxes) or subsidies on specific goods or services. Suppose a regulation requires firms to pur chase a $10,000 piece of imported equipment, on which there is a $1,000 customs duty. For purposes of benefit-cost analysis the cost of the regulation for each firm ordinarily would be $10,000, not $11,000, since the $1,000 customs duty is a transfer payment from the firm to the Treasury, not a real resource cost This approach, which implicitly assumes that the equipment is supplied at constant costs, should be used except in spedal circumstances. Where the taxed equipment is not supplied at constant cost the technically correct treatment is to wiHtte how many of the units purchased as a result of the regulation an supplied from increased production and how many from decreased purchases by other buyers. The former units would be valued at the pries without the tax and the latter units would be valued at the price including tax. This calculation is usually difficult and imprecise because it requires of supply and demand elasticities, which an often difficult to obtain and inexact Therefore, this treat ment ahould only be used where the benefit-cost conclusions an likely to be sensitive to the treatment of the indirect tax. While costs ordinarily ahould be adjusted to remove indirect taxes on specific goods or services as described here, similar treatment is not warranted for other taxes, such as general sales taxes applying equally to most goods and services or in come taxes. (d) Distribution expenses. The treatment of distri bution expenses is also a source of potential error. For example, suppose a particular regulation raises the cost of a product by $100 and that wholesale and retail distribution expenses are on average SO percent of the factory-level cost It would ordinarily be incor rect to add a $60 distribution markup to the $100 cost increase to derive a $160 incremental cost per product for benefit-cost analysis. Most real resource costa of distribution do not inoesse with the price of the product being distributed. In that case, either distribution expenses would be unchanged or, if they increased, the increase would represent distributor monopoly profits. Since the latter are transfer pay ments, not real resource costs, in neither case should additional distribution expenses be included in the benefit-cost analysis. However, increased distribution expenses should be counted as costs to the extent that they correspond to increased real resource costs of the distribution sector as a result of the change in the price or characteristics of the product D. Expenditure Rules Regulations establishing terms or conditions of Federal grants, contracts, or financial assistance call for a different form of regulatory analysis than do other types of regulation. In some inrtiiTV--. a full blown benefit-cost analysis may be appropriate to inform Congress and the President more folly about the desirability of the program, but this would not ordinarily be required in a Regulatory Impact Analy se 062700 , " n rrn iw oai* APPENDIX V 666 gif. Die primary function of the RIA for this type of probably is not sensitive to likely alternative param adulation ahould be to verity that the tenoa or eter values. If the ratio is only slightly greater than conditions are the --neceaaary to achieve the porpoaea for which the hinds were appropriated. They ahould not contain conditions in pursuit of goals one, the conclusion probably is sensitive. The benefitcost ratio may sometimes be acceptable as a rough substitute for genuine sensitivity analysis where it is that art not germane to the purpose for which the not feasible to carry out a foil aanaitivity analysis fceji were authorized and appropriated. Beyond con* (a*., if the number of regulatory parameters to be trola to prevent abuse and to ensure that fimds lasted fay aanaitivity analysis is large). When so used, i appropriated to achieve a apedfic purpose are chan- the benefit-cost ratio should bo recognized as only a | neled efficiently toward that end, maximum diaa** crude approximation to a genuine aanaitivity analysis tion ahould be allowed in the use of Federal fonda, and tbs analyst ahould be aware of its limitations particularly whan the recipient is a State or local (04., the benefit-cost ratio is sensitive to the arbi government trary classification of an Ham as a benefit or an averted eoatX IV. RATIONALE FOR CHOOSING THE PROPOSED REGULATORY ACTION Whara the benefits of proposed regulatory alterna tives include reductions in fetality risks, an accept able alternative to direct calculation of not benefits is The RIA should include an explanation of the the indirect approach of calculating incremental costa reasons for choosing the selected regulation. Ordinar- par life saved between adjacent alternatives. This is r fly, the regulatory alternative selected should be the dons fay ranking all the alternatives according to the coe that achieves the greatest net benefits. If legal number of fives they save and than calculating the constraints prevent this choice, they should be identi change in costa and the change in fivee saved be fied and explained, and their net cost ahould be tween each alternative and the one with the next estimated. Where uncertainties are substantial or a large proportion of benefits cannot be monetized, other methods of summarizing the benefit-coct analysis may sometimes be appropriate. When alternative highest number of fives saved. If the alternative elected is the one whose inaemental cost per fife saved is closest to the wfllingness-to-pey value of fife, this decision criterion ia analytically equivalent to that of maximiriag not benefit forma of presentation are used, the objective must In ease* whara important banafite cannot be as continue to be the maximization of net benefits (ex signed monetary values, cost-effectiveness analysis cept where prohibited by law). Alternative criteria should bo used where possible to evaluate alterna must be used with care because of the potential for tives that generate equivalent nonmonetizablc bene errors or misinterpretation. Agencies need not calculate the internal rate of return for a regulation. The internal rate of return is often difficult to compute and is problematical when multiple rates exist. It must not be used as a crite rion for choosing between mutually exclusive alterna tives. As a criterion for choosing between alternatives fits. Coate should bo calculated not of monetized benefits. Between two alternatives with equivalent nonmonetizablc benefits, the alternative with the lower net eoete should be selected. Cost-effoctivenoss analysis ahould also be used to compare regulatory alternatives in eases when the level of benefits is specified fay statute. that are not mutually exclusive, it has no advantages ever the criterion of maximizing the present value of V. STATUTORY AUTHORITY net benefits. Benefit-cost ratios, if used at all, must be used with care to avoid a common pitfoU. It is a mistake to choose among mutually exclusive alternatives by se Tbs RIA should include a statement of determina tion and explanation that the proposed regulatory action is within the agency's statutory authority. lecting the alternative with the highest ratio of bene fits to costs. An alternative with a lower benefit-cost Further Reading ratio than another may have the higher net benefits. Whether a regulation's benefits are greater (or less) than its costs can be determined by whether its benefit-cost ratio is greater (or less) than one. The Edith Stokey and Richard Zeckhauser, A Primer for Policy Analytic. Chapter* 9 and 10 provide a good introduction to basic concepts. benefit-cost ratio may be used as a very simplified E. J. Miahan, Economics for Social Decisions: Blc indicator of the likely sensitivity of the result: If the menu of Coet-BenefU Analytic. Assumes some knowl benefit-cost ratio is much greater than one, the con edge of economics. Chapter* 5-8 should be helpful on clusion that the regulation's benefits exceed its costa the important subjects of producers' and consumers' SL 062701 666 EEOTIJtTOBYraOGEAM OF TOE UNITED STATES OOVEENMENT surpluses (not dis'iir--'1 extensively in this guidance document). W. Kip Viacusi, Risk By Choice. Chapter 6 is a good starting point for tha topic of valuing haalth and safety benefits. Other more technical sources are given in the bibliography. Robert Cameron Mitchell and Richard C. Using Surveys to \hlue Public Goode: The Contingent Wuedon Method. Provides a valuable discussion on the potential pitfalls associated with the use of con* tingent-valuation methods. V. Kerry 8mith, Ed., Advances in Applied Micro economics; Risk, Uncertainty, and the Mduation of Benefits and Caste. Judith D. Bentkover, Vincent T. CoveDo, and Jeryl Mumpower, Eds., Benefits Assessment: The State of the Art. SL 062702