Final Examination -- Spring 1999 
Time Allowed -- Four Hours  

This examination consists of 5 pages, including this one. Please check now to make sure you have all the pages.

 

Instructions:

This is an open-book examination. The examination consists of three questions. The weight assigned each question corresponds to the time suggested for each. The suggested times are purely advisory, however. You can allocate your time any way you choose. Please start each question in a separate booklet, following the color scheme as indicated. Write on one side of the page only. Please mark the front of each booklet to indicate which question is answered therein.

Organization and conciseness count, so plan your answer before you write.

If for any question you think additional facts would be relevant to the issues discussed in your answer, state what facts you need and how they would affect your analysis.

Question I

[One and one-quarter hour]

The Patients’ Bill of Rights Act of 1999, recently introduced in the U.S. Senate, contains a provision, in full text below, that purports to prohibit certain discrimination on the basis of "predictive genetic information." Pick a role as a lawyer for either (i) the US Chamber of Commerce; (ii) NYPIRG (New York Public Interest Research Group); or (iii) a Senator. Your client wants you to analyze the legislation, including whether it is effective for its purpose, its impact on health insurance markets, its possible implications for the provision of health care, and whether it is desirable public policy taking into account your client’s interests as you understand them. Your analysis could include suggestions as to how the legislation should be modified in light of your analysis.

SEC. 714. (a) PROHIBITING PREMIUM DISCRIMINATION AGAINST GROUPS ON THE BASIS OF PREDICTIVE GENETIC INFORMATION. A group health plan, or a health insurance issuer offering group health insurance coverage in connection with a group health plan, shall not adjust premium or contribution amounts for a group on the basis of predictive genetic information concerning an individual in the group or a family member of the individual (including information about a request for or receipt of genetic services).

(b) LIMITATION ON COLLECTION OF PREDICTIVE GENETIC INFORMATION_ Section 702 of the Employee Retirement Income Security Act of 1974 (29 U.S.C. 1182) is amended by adding at the end the following:

`(c) COLLECTION OF PREDICTIVE GENETIC INFORMATION_

`(1) LIMITATION ON REQUESTING OR REQUIRING PREDICTIVE GENETIC INFORMATION_

Except as provided in paragraph (2), a group health plan, or a health insurance issuer offering health insurance coverage in connection with a group health plan, shall not request or require predictive genetic information concerning an individual or a family member of the individual (including information about a request for or receipt of genetic services).

`(2) INFORMATION NEEDED FOR DIAGNOSIS, TREATMENT, OR PAYMENT_

`(A) IN GENERAL_ Notwithstanding paragraph (1), a group health plan or health insurance issuer that provides health care items and services to an individual or dependent may request (but may not require) that such individual or dependent disclose, or authorize the collection or disclosure of, predictive genetic information for purposes of diagnosis, treatment, or payment relating to the provision of health care items and services to such individual or dependent.

(b) AMENDMENT RELATING TO THE INDIVIDUAL MARKET_ The Public Health Service Act (42 U.S.C. 300gg_11 et seq.) (relating to other requirements), as amended ... is amended ... by adding at the end the following:

`SEC. 2753. PROHIBITION OF HEALTH DISCRIMINATION ON THE BASIS OF

PREDICTIVE GENETIC INFORMATION.

`(a) PROHIBITION ON PREDICTIVE GENETIC INFORMATION AS A CONDITION OF ELIGIBILITY_ A health insurance issuer offering health insurance coverage in the individual market may not use predictive genetic information as a condition of eligibility of an individual to enroll in individual health insurance coverage (including information about a request for or receipt of genetic services).

`(b) PROHIBITION ON PREDICTIVE GENETIC INFORMATION IN SETTING PREMIUM RATES_ A health insurance issuer offering health insurance coverage in the individual market shall not adjust premium rates for individuals on the basis of predictive genetic information concerning such an enrollee or a family member of the enrollee (including

information about a request for or receipt of genetic services).

`(c) COLLECTION OF PREDICTIVE GENETIC INFORMATION_

[The proposed language tracks the proposed amendment of section 702 of ERISA described above.]

-----------------------------------------------------

Question II

[One and one quarter hour]

Section 313 of the Emergency Planning and Community Right-to-Know Act of 1986 ("EPRCRA") requires large manufacturing facilities to file annual reports on routine releases and transfers of several hundred toxic chemicals found in wastes from ordinary business operations. This has led, for each firm, to the production of company-specific "Toxics Release Inventory" ("TRI"). The reports are collected by the states and the EPA and disseminated through a variety of means. The EPA must make this information public in a computerized database.

Section 313 provides in relevant part:

(g) Form

(1) Information required

... the Administrator shall publish a uniform toxic chemical release form for facilities covered by this section.... Such form shall__

(A) provide for the name and location of, and principal business activities at, the facility;

(B) include an appropriate certification, signed by a senior official with management responsibility for the person or persons completing the report, regarding the accuracy and completeness of the report; and

(C) provide for submission of each of the following items of information for each listed toxic chemical known to be present at the facility:

(i) Whether the toxic chemical at the facility is manufactured, processed, or otherwise used, and the general category or categories of use of the chemical.

(ii) An estimate of the maximum amounts (in ranges) of the toxic chemical present at the facility at any time during the preceding calendar year.

(iii) For each wastestream, the waste treatment or disposal methods employed, and an estimate of the treatment efficiency typically achieved by such methods for that wastestream.

(iv) The annual quantity of the toxic chemical entering each environmental medium.

(2) Use of available data

In order to provide the information required under this section, the owner or operator of a facility may use readily available data (including monitoring data) collected pursuant to other provisions of law, or, where such data are not readily available, reasonable estimates of the amounts involved. Nothing in this section requires the monitoring or measurement of the quantities, concentration, or frequency of any toxic chemical released into the environment beyond that monitoring and measurement required under other provisions of law or regulation. In order to assure consistency, the Administrator shall require that data be expressed in common units.

(h) Use of release form

The release forms required under this section are intended to provide information to the Federal, State, and local governments and the public, including citizens of communities surrounding covered facilities. The release form shall be available ... to inform persons about releases of toxic chemicals to the environment; to assist governmental agencies, researchers, and other persons in the conduct of research and data gathering; to aid in the development of appropriate regulations, guidelines, and standards; and for other similar purposes.

-----------------------------------

A recent commentator writes: "Making the TRI data publicly available to any citizen has been described by an EPA Administrator as ‘among the most important weapons in efforts to combat pollution.’ The present Clinton EPA administration regards the TRI as ‘among our most potent environmental weapons.’" This is buttressed by evidence that over the 1988-94 period, members of the Chemical Manufacturers Association voluntarily reduced their emission of toxics by 52% while sales volume was up 10 percent.

Why should a [voluntary] disclosure program like TRI have been so effective in reducing the release of toxics? Analyze the possible mechanisms of its success. Should disclosure lead to mandatory emission controls? What are the implications of the evidence about TRI for OSHA’s Hazard Communications Standard?

----------------------------------------------------

Question III

[One and one-quarter hour]

Although the matter is not free from doubt, most scientists believe that the earth’s average daily temperature of has increased over the past 100 years and will continue to increase because of the "global warming" effects of the emission of "greenhouse gases" such as CO2 through the burning of fossil fuels. This could lead to significant climate changes that could radically change the distribution of farmable land and raise the level of the ocean.

At a conference held December 1997 in Kyoto, Japan, the US and other developed countries agreed to a set of binding emissions reduction targets of approximately 7% below 1990 levels, beginning in 2008.

Among the other features of the Kyoto Protocol are these:

(1) The parties agreed to the outlines of a plan of emissions permits trading among nations with emissions targets.

(2) Countries with emissions targets may also get credit towards their targets via the Clean Development Mechanism" ("CDM") under which developed countries can enter into cooperative projects to reduce emissions in developing countries such as the construction of environmentally sound power plants or manufacturing facilities.

(3) Countries with emissions targets may also get credit by creating greenhouse gas "sinks"; e.g., reforestration that will absorb greenhouse gases.

(4) Certified emission reductions achieved starting in 2000 will count toward compliance with the first budget period.

(5) To enter into force, the Protocal must be ratified by at least 55 countries, accounting for at least 55% of the total 1990 carbon dioxide emissions of developed countries.

You are a senior staffer to Sen. Prolix, a member of the Foreign Relations Committee that will shortly hold hearings on whether the US should ratify the Kyoto Protocol. She wants you to prepare a memo that will analyze each of these features of the Kyoto Protocol, that will provide her with a set of questions to ask about implementation plans (and an explanation as to what turns on the answers), and your view about the impact of the Protocol on relations between the developed and developing world

Exam #701

Role: U.S. Senator

The gist of the plan is quite simple - it prevents insurers from using genetic information to raise premiums or deny eligibility in both the group and individual insurance market. While I will analyze the two markets separately, there are some problems common to both which I will discuss here.

1) Enforcement - there is an important loophole in both plans which allows the provider to request genetic information for the purposes of diagnoses etc. Given the extreme lack of protection for this kind of information (patient confidentiality has become something of a farce in the modern era) this could make for quite an enforcement problem. If insurers did use such information illegally, it would be very difficult for the patient to find out and, even if they did, to afford the necessary legal enforcement.

2) Regulation under uncertainty - nobody is quite sure at this point what exactly genetic science will eventually allow us to do entering with legislation at this point may well cause problems in that we may prevent the use of information which may well become absolutely essential to the practice of medicine. Of course the legislation can always be changed, but given the workload of Congress and the difficulties of garnering its attention the whole plan my be somewhat premature.

The Group Plan

The use of genetic information should definitely be banned in the area and the proposed legislation is probably an excellent idea. The information in question here is extremely relevant to an insurer’s potential adverse selection problem - the problem of taking on a disproportionate amount of bad risks because of information asymmetries. There are two ways around this problem - mandatory pooling and screening. Those already in the group health insurance market are engaged in mandatory pooling, thus although genetic information promises to be an unparalleled screening mechanism here, in this market it could only be abused - used for the purposes of excluding bad risks and creating artificially beneficial risk pools. In summary, given the fact that group based health plans are already subject to mandatory pooling, thus mitigating the adverse selection problem, there is little justification of allowing insurers to screen as well.

An additional problem that would occur in the absence of such regulation in the group markets is that employers would have a tremendous incentive to screen their employees -- to eliminate bad risks before they joined the group plan and raised premiums for everyone -- this legislation prevents that, which is good. Thus, the effects of this legislation on the group market are essentially to protect the status quo -- to allow the mandatory pooling already functioning in these markets to continue functioning as the insurers primary weapon against adverse selection -- by preventing the use of genetic information the legislation protects potential employees from screening by potential employers as well.

Individual Market

The benefits here are far from clear. By definition, in the individual market the insurer is subject to severe adverse selection problems -- there is no mandatory pooling thus leaving the insurer with screening as the only possible option to defend themselves against adverse selection. Genetic screening, at this point, promises to be the most accurate and effective form of screening ever available, and making its use illegal would be a tremendous loss to the insurer. In the absence of effective screening, insurance rates in the individual market will continue to be prohibitively high -- especially given that the types of people who NEED individual insurance are often those who cannot get it through work -- the poor.

The attraction of the legislation is that an individual with an expensive genetic predisposition would have a fighting chance of getting insurance, but this benefit may be dubious and illusory — and using this type of legislation to attain it even more so.

By preventing screening, the legislation effectively forces insurers to charge very high rates to everyone in an inordinately large and impressive risk pool. Allowing genetic screening would create the possibility of tailoring risk pools more accurately, thus reducing the cost of individual health insurance for most and thereby increasing the number of insured. IF the government really wanted to protect those with genetic predispositions, it may be better to do so directly -- through some sort of risk adjustment or subsidy to insurers willing to take on such patients. This would increase the availability of insurance for both those with genetic predispositions and the general market.

The harsh reality is that genetic screening promises to be the most effective means by which insurance companies can set up well tailored risk pools in the individual market. One of the unfortunate side effects of this will be prohibitively high rates for those with costly genetic predispositions. But the solution to this problem is a direct subsidy for that particular pool, not general legislation which makes individual insurance an unaffordable dream for the general population (not eligible for insurance through the workplace). Whether as a senator I would want to support the plan depends on my various constituencies - do a have a disproportionately large % of employed in my state in a group health plans (definite support - at least for the group part of the legislation), or rather, a large % of unemployed/uninsured. Do I have a great deal of insurance industry/money within my State etc. These considerations aside - assuming I have an average state (whatever that means) I think the group plan is definitely worth supporting, while the individual plan less so - especially if we can somehow convince the legislature to take a more direct approach to protecting the poor with bad genetic predispositions.

Exam #715

MEMORANDUM

TO: New York Public Interest Research Group (NYPIRG)

From: Association Counsel

RE: PROHIBITION OF DISCRIMINATION ON THE BASIS OF "PREDICTIVE GENETIC INFORMATION"

The primary goal of Sec. 714 (for the group market) and Sec 2753 (for the individual market) of the newly proposed Patients’ Bill of Rights is to expand the coverage of private health insurance by allowing individuals with certain genetic dispositions that reflect a high possibility of contracting certain diseases to enroll and not be discriminated in private health insurance eligibility and premium. While the objective of the bill is praiseworthy, there exists several problematic concerns that arise from the start. However, the first question is, is the bill effective in seeking to accomplish its purpose.

The question seems to be a YES. For both group plans (say employer health insurance) and individual plans, the bill makes sure that predictive genetic information will not work to deny coverage to a possibly high-risk individual (say carrying a genetic disposition that is 10x more likely to get a certain disease, or even HIV+?) Or raise her premium. What makes the bill truly effective is its inclusion of the phrase "including information about a request for or receipt of genetic services." Normally, an insurer does not specifically have to look at every genetic testing result to discern who might be a high-risk person. Rather, they can INFER from the fact some people purposefully failed to take a certain requested test, or they took a special genetic diagnostic test, etc. This bill, however, forbids the insurers to even make this indirect distinction. Thus it completely assures that a person with "bad" genetic disposition will have a chance to sign for an insurance without being discriminated.

However, is this necessarily good for the health insurance market? There seems to be several problematic aspects. First, from the perspective of the insurers, the bill severely undermines their ability to perform UNDERWRITING, and properly discern the high-risk from the low-risk people. The bill (sec. 714(c)(1)) basically forbids any insurer to ask for predictive genetic information that might reveal the true risk factor of an individual unless the request is limited for aiding in actual treatment, diagnosis, etc. Two significant impacts this will have on health insurance market are adverse selection and cross-subsidy. Adverse selection will kick in as more "high -risk" people with certain predictive genetic information are accepted/not thrown out, resulting in higher use of medical services by these people, thus need for higher premium, and low-risk people dropping out as they feel the premium is too high. The process continues, and ultimately, the high-risk people end up paying premium equal to their risk ( actually, more than their actual risk due to LOADING FEE). In order to prevent adverse selection, one can either conduct SCREENING to find risk, or enforce mandatory insurance on everyone. What the bill denies is the first option of preventing adverse selection. Furthermore, the problem of cross-subsidy will be enlarged, as the low-risk, healthy people must now subsidize for more high-risk, as the insurer is disallowed to deny access to some of these people by law. This is in a way, a strong example of forced redistribution, (Epstein argument) that may be even morally questionable (like forced redistribution for food, housing, etc.) However, there is a distinction between normal case of a forced redistribution and the case of predictive genetic information. In the former, whether it be cross-subsidizing for poverty, illness, disability, etc., one might assume that the harm ensued because of individual’s laziness, stupidity, carelessness, lack of education, etc. In short, they attribute the responsibility of harm to the individual, at least to significant degree. (Thus for Epstein, this becomes problematic). However, in the case of genetic predisposition, there is NO individual action or will involved. Rather, it is strictly determined by genes, and by nature. Thus, in terms of moral force, the bill stands in a much more favorable light than some other forced cross-subsidy. But, there may be another side to this issue. Normally, high-risk people are also very poor (lack of proper diet, unsafe car, etc.) Thus, it makes some sense to cross-subsidize the poor. However, in this case, a "bad" predictive genetic information has no correlation with wealth. A high-risk person under this standard may as well be a millionaire, who can and SHOULD pay for his high-risk. Thus, the argument can go both ways.

One typical problem of insuring high-risked people, that of moral hazard, does not seem to be relevant here however. Because this deals with predictive genetic information, and since people usually do not have control over their genetic makeup, the bill should not cause people to be less careful, etc.

How will this affect the provision of health care? In many ways, this bill is not too drastic as it might seem. There have been other bills, for example, Kassebaum-Kennedy Act, (HIPPA), that allowed those with pre-existing conditions to enroll in health coverage. This bill is not as drastic, since predictive genetic information does not necessarily mean having a certain disease as of now. Thus, the effect on the low-risk people is less. For group plan that is operated by HMO under managed care, the plan may negatively affect the actual health care services by doctors, if capitation fee are used. Under the employer group insurance, the employees would supposedly go to a same set of providers, whom will be forced to deal with more high-risk people. Assuming high-risk people require more services and treatment, that may put pressure on their capitated fee, resulting in doctors giving less than adequate treatment to other less risky people in order to make up for the cost and perhaps still earn some of those HMO provided "bonuses." However, this effect is not as problematic as adverse selection or cross-subsidy.

Overall, the desirability of the plan depends on how many previously denied people in New York will gain access to the plan. If the number is small, inclusion of them might be a good thing ( considering the moral argument that these people have no control over their genes). However, if the number is too large, from the perspective of NY insurance community, this will cause too many negative side effects. One modification of the bill to improve its effect might be to separate the group and the individual plan, and prohibit discrimination on the basis of genetic information only under the group plan. The reason for this is as follows. Because the government subsidizes the employer insurance plan through tax deductions, employees have less incentive to drop out. (For some it’s even mandatory). Thus, the effect of possible adverse selection will be dampened under the group plan. For the individual market, one might find a person dropping out rather quickly if the premium rises even a little bit. Considering they are already paying significantly higher premium for personal insurance without the benefit of tax subsidy, it might be better to not penalize this group. 

Exam # 735

For the purposes of this essay I will represent the New York Public Interest Research Group (NYPIRG). Their interests, as I understand them, are to advocate health care policies that improve the extent and scope of health insurance for the average American.

A long-time concern of health care advocacy groups has been the freezing-out of the market of individuals with pre-existing illnesses or predispositions to certain illnesses. Authors such as Stone have argued, in the context of HIV positive individuals, that failure to insure such people essentially posses the burden on to public institutions and, in turn, taxpayers. Those who are ill thus receive care, but it is care provided for them on behalf of the government. (i.e. taxpayers)

Screening for genetic risks differs, of course, from the situation regarding HIV positive individuals in that the illness has not manifested itself (and may be latent for quite some time) and, in normative terms, because behavior is not a factor in terms of genetic predisposition as it is in terms of AIDS or Hepatitis (acknowledging of course that behavior is only one factor among many). The concern with genetic testing then is that insurance companies will use it as a means to segment risk pools; that is they will use the tests to select low-risk individuals rather than high-risk and, in so doing, increase premium prices for the high risk group, potentially pricing them out of the insurance market (just as redlining affects high-risk groups in the lending market).

I will now evaluate the program along two axes: (1) how well does it protect individuals from genetic screening; (2) how does it affect access to coverage in terms of the general population; that is, what are its full equilibrium effects.

Examining the legislative proposal two questions occur in regard to its effectiveness in assisting those with genetic pre-dispositions to disease.

(1) The plan applies to group health plans, and issuers offering coverage in connection with such plans. While providers to individuals are also covered, it is unclear if there are any gaps in applicability. While this will seemingly apply to most plans, it is by no means clear that it will apply to all. Universality is important not only because we want to protect the most individuals possible, but also because we do not want to place certain firms or plans at a competitive disadvantage. If a certain segment of the industry can still screen individuals than this will lower their costs while increasing costs to other insurers – if individuals can transfer plans without screening, increasing even more the costs to those remaining in a far from virtuous cycle.

Thus, without definition of the terms I would want to examine if any insurers are excluded. If they are an important amendment would be to universalize the coverage.

(2) With our focus on the concerns of patients another important aspect of the plan is the exception to (c)(1) contained in (c)(2)(A).

First, the scope of the exception "for purposes of diagnosis treatment or payment" is very broad; so broad as to threaten to swallow the rule. Concern as to the breadth of the exception is heightened by the following problem: although an insurer may request, but not require, information for the reasons cited above there is a strong possibility that the insurer will treat refusal to comply with a request as though it were equal to a negative test result. This could then lead to the very problem we are trying to avoid.

I would therefore clarify this language to ensure that refusal to comply with a request cannot lead to increased premiums, and I would tighten up the definition of the exceptions. For instance, "payment" seems incredibly broad and potentially all inclusive.

We want to ensure that the legislation allows people at-risk to get tested without fear of being priced out of the market, otherwise we would discourage testing and all the possibilities of early detection and treatment that come with it.

(2) The next broad area to address is how the legislative changes made here will affect coverage in the wider population. Any such question must begin with an emphasis on the central insurance tension in this area.

Broad pooling will cover more people but will also drive out those low-risk insurees: e.i. those who see the costs of the group rate being greater than the risks they face as individuals (especially give the "floor" provided by public institutions).

Narrow pooling, by more exactly correlating risk and premium eliminates low-risk flight but threatens to price high-risk groups out of the market.

Here, in the group market, the risk is that the prevention of genetic screening will cause entry of certain high-risk individuals that would otherwise be screened out of the group: this, in turn, will raise the "rate" for the group and lead to drop-outs amongst those at lowest risk.

While the plan clearly prevents direct rate-raising there exists the possibility that insurers will recoup their increased risk through higher "loading costs," which it is unclear that his plan covers or prevents.*

[* But note discussion about individuals; does the legislation prevent

I) Insurer from raising group A’s rates or does it also preclude

II) Insurer from raising all group rates to reflect the increased risk.

If (I) but not (II) then overall rates will rise to reflect increased risk to insurers and low-risk consumers of insurance will be priced out. If (I) and (II) then the discussion illustrates that increased costs to insurers will still be reflected through the system].

Second, if insurers cannot pass on costs through direct premium increases they may will seek to cut costs in other areas, as through increased use of managed care organizations. This increased emphasis on efficiency gains in this area could lead to problems of under-treatment or increased reliance on utilization reviews, which will drive up loading costs. In essence then the decision to make insurers accept greater unknown risk will result in increased costs which will then be reflected back through the system.

On the face of the legislation the private individual market is treated differently. Here an insurer may not use predictive genetic testing as a criteria for eligibility or for setting premium rates.

But the problem with this is that while insurers cannot say: "I am raising your premium X because you tested..." they can easily say: "I am raising the average cost of insurance to individuals to reflect my increased risk." This will create a perverse incentive of sorts, for those who have no genetic predisposition the increased average cost will potentially drive them out of the market, while for those with a genetic predisposition the average rate rise will still provide for insurance at below their expected risk. So in a way this is a perfect illustration of adverse selection as those most at risk will be given insurance at a rate below their proper pool (i.e. below what it would be if premiums were correlated to risk precisely).

Thus, the problem, as with any program that increases benefits and/or eligibility is that you will have concurrent cost increases throughout the system that may cause the rate of uninsured to increase.

Pooling may be enforced by either making it mandatory or by providing incentives to make it appealing enough for low-risk to join (i.e. employer-based health plans not taxed on income used for health benefits).

If implemented then the ripple effect of increased coverage for a high-risk group will have to be watched closely, and if there is an increase in the uninsured then incentives may have to be provided to maintain pooling.

-----------------------------------------

Exam #824

The purpose of the two provisions is the same: to stop layering of insurance pools by insurance companies. Senate does not want groups or individuals to be discriminated against due to their genetic makeup, so they designed two similar statutes see 714 and see 2753 to deal with discrimination. However, it seems that the Senate has failed to realize that the market effect will be different. I will try to explain to the Senators the market effect of both group and individual discrimination prohibitions.

The group prohibition of request or receipt of genetic services is a great idea. Some coverage is bought by groups you can assume to get uncorrelated risks, making insurance much more feasible. Also since it’s doubtful that the genetic makeup of one group is much different from genetic makeup of other group, you will get very little adverse relation among groups. Of course this only case if there is forced pooling within the group, so that low-risk won’t drop out. You can achieve this coercively by law or by providing tax break to make it stupid not to join pool even for low risk. Of course tax break may be regressively set up due to progressive tax % and fact that wealthier people buy more insurance since tolerate less risk. But let’s assume legislature sets up incentive in a way that is distributionally fair.

Since pools accumulate in way that has little difference in cost for each group, this avoids cherry-picking by insurance companies. More importantly, an employer won’t feel the need to screen employees for genetic makeup prior to hiring. In addition, this avoids the problem of lock-in from adverse genetic makeup, since otherwise no new company will hire somone once it knows of the threat . Low-risk individuals might feel unfair to have to cross-subsidize high risk individuals, but as long as we provide them with enough tax incentive they won’t be too mad.

The same problem of pooling due to work conditions which is not uncorrelated within groups still exists, but these 2 statutes not meant to deal with that problem. However, if for some reason genetic pools within worker pools are very correlated, it is a big problem. Let’s say more workers have not only high risk of injury due to work conditions but also mine worker because genetically inferior and more susceptible to disease and sickness, as compared to lawyer. This makes cross-subsidizing bigger and more likely discrimination will occur indirectly. HMO’s can look to other factors such as work or nationality to get around prohibition of discrimination on the basis oi predictive genetic info.

Forgot to mention: you have to prohibit both request and receipt of genetic information. This is case because HMO’s can assume that those who don’t publicize information have inferior genes than those who do.

The treatment, diagnosis and payment provision for groups has problems that Senators may not be aware of. By giving individual choice of disclosure for treatment goal is to provide health care provider with information that will keep costs down and provide optimal level of care. Additionally, patient can keep privacy and pay for treatment instead of disclosing private information. The problem is that the statute is very vague. Why does insurer get to see information; can’t we just limit it to the doctor? Since doctor usually making decision about treatment, in most cases we can. But what if have utilization review? Now insurer gets information, since need it as part of decision-making process. Is this ethically right? Most people have problem changing someone’s care due to genetic makeup and likelihood of successful treatment. Remember this is identified life not statistical. Senator should maybe add provision that no utilization review can take into consideration genetic info. Don’t want someone denied health care just because half as likely as another individual with different genetic makeup to be cured. Very bad public relations in egalitarian society. Problem though we live in a society with scarce resources, have to distribute them somehow, to get best overall welfare. But very bad practical move to distribute them based on genes.

Still making information to doctor very helpful. Now doctor canavoid giving patient certain drugs and procedures that may be harmful. Save life and costs. Nobody will have problem with this. Yet, patient may still not want to fearing later discrimination. What if loose job. Now need insurance and fear word get around, even though doctor patient confidentiality. Fear lock-in, even though other statute take care of this patient still may be worried. Therefore, add that once a person is insured, he or she may to keep the insurance indefinitely (but now you may need to charge the person a fee if unemployed because of the loss of the tax subsidy). Also, patient may fear giving information to insurer. What if company refuses to treat, knowing the patient will die or will just not benefit sufficiently. Individual need guarantees against it.

Benefits for sec. 2753 far less. You have no possibility of employment discrimination here which was big thing feared in sec. 714. Also, no forced pooling here so tremendous problem of adverse selection. Not feasible to try forced pooling through taxes, since individual may be unemployed or poor and very little tax advantage to join pools. Back to problem of adverse selection which is great here. Adverse selection problem of asymmetrical information. Now individual knows about genetic makeup and insurance companies prohibited from knowing about this potentially very important piece of information. Insurance companies will correctly assume that there is something wrong with people who buy insurance. Now people can better predict expected health costs (through genetic testing perhaps) and only those with higher than average risk will buy insurance. Insurance rates go up, further pricing out low risk individuals. This spiral will continue till pool shrinks and get what we call unraveling.

Nothing you can do about it. You can get rid of 2753 (b). This now allows layering by insurance companies. Groups pooled together based on risk. Apply different premiums to each, much more likely to prevent unraveling. However now high risk individuals have to carry own weight, no cross-subsidization. Big problem, as many might not be able to afford it. What may happen is a large percentage of those by not buying insurance decide society will insure them since have little wealth to protect or just not possible to protect wealth and pay insurance fee. Many people also scared of not having insurance may decide to quit job and be eligible for Medicaid. Not good for the economy.

Best alternative for individual coverage is allow layering but subsidize those with high risk. Now low risk insured get benefit of not having to deal with risk adversion on own since in insurance pool, and high risk also insured and not great burden to society. Win election!!!

But we must remember very important thing about partial equilibrium: change one piece of puzzle, other pieces do not stay the same. Behavior of insurance companies and HMO’s change to reflect prohibition of genetic info. This increases adverse selection problem as previously stated since information asymmetry increased: HMO’s and insurance companies will increase studies looking to links between genetic info and behavior or traits they can discriminate against. Although efforts only limited success, may want to pass legislature dealing with those issues.

Finally, main reason prohibit genetic discrimination is not just efficiency. Reason is that society don’t like to hold people responsible for things they are not in control of or traits not responsible for. "Sink or swim together." Great campaign slogan. Win election!

Exam # 717

Introduction

Disclosure programs such as TRI provide multiple mechanisms that all work in the same direction as a more hard-line command and control approach to a problem, and also allow regulators to make more informed baseline judgments about the possible need for future command and control or other regulation. While many of the apparent benefits of the TRI plan were probably real, some may be the products of imaginative, self-interest-based number-crunching and "magic tricks" on the part of the regulated industry. This essay will examine the reasons why TRI was apparently so effective in reducing the release of toxins, and will then go on to explore whether the program’s results should lead to mandatory emissions controls, and what its implications are for OSHA’s Hazard Communication Standard.

Legitimate reasons for effectiveness

One of the primary, most obvious, and also most effective reasons for the success of reporting programs such as TRI is the fact that it promotes organization and accountability, both within the firm and in its relationships with the industry, the public, and the government. Before TRI, many of the affected firms may have been exceedingly sloppy in their treatment of toxic chemicals, and eve sloppier in their internal monitoring of that treatment. A pre- TRI company-stated goal of reducing emissions would have been ineffective if lower-level employees did not act out the noble aspirations of the board of directors. TRI, then immediately encourages internal organization and accountability in this realm. Instead of simply proclaiming a goal of reducing or closely monitoring emissions, firm management is forced to develop a concrete plan that will communicate chemical treatment up the rungs of the company ladder. This increased internal accountability, if effective, is likely to have a positive mutating effect on other aspects of the firms operations, thereby increasing internal communication, accountability, and productivity such that other areas of the firm’s operations also benefit. Reporting, though seemingly a burden, may actually increase the firm’s efficiency by forcing it to identify and reconcile internal trouble spots, whether they be in the emissions area or elsewhere. This development may explain, in part, the ability of the members of the Chemical Manufacturers Association increase sales volume while reducing their emissions dramatically.

As well as increasing internal accountability, disclosure programs positively affect accountability to the public and to the political arena. Whether or not it was explicitly designed as such, TRI necessarily must have served as a mechanism to increase public and political awareness of the implications of emissions. This phenomenon increases overall access to infromation - to firms, the industry, the public, and politicians - such that the risk of market failure due to lack of information is reduced. It also reduces the presence of cognitive errors throughout all of these parties. For example, a concrete knowledge of real emissions information will provide consumers of that information with a "typicality" or availability" not previously present. This availability will increase their inferences of negative consequences, and generate public and political action to cut down on emissions. The very threat of this public and political action, even before it occurs, will encourage any firm with reasonable foresight to reduce the visibility of emissions problems so that they are not "jawboned" by regulators into changing their behavior as a result of the new activism.

TRI also increased accountability to insurers. Before TRI, a significant element of moral hazard may have been present in that firms may have emitted hazardous chemicals without regard to negative consequences. Their attitude was likely to be, "If we get sued, the insurance company can pick up the lawyer’s bills and damages awards, and we’ll just take them out for a nice dinner at Le Cirque when it’s all done with." With mandated reporting, however, insurers are able to identify high-risk firms, and pool them such that they pay increased premiums directly based upon the volume and danger of their emissions. In order to keep their premiums down in the face of increased information flow to insurers that allows the insurers to identify real risks, then, TRI provides a powerful financial incentive to reduce emissions. If they do so, they will be able to avoid the possibly higher premiums associated with a negative experience rating, and reap the benefits of the lower premiums associated with a positive one.

In order to reduce emissions to ward off the threat of potential command and control regulation, firms also receive an incentive from TRI to make innovations that will reduce emissions. Such incentives would not have existed to the same extent before TRI. In addition, TRI prevents a "race to the bottom" scenario from occurring, by which firms would spend a minimum amount of resources to address emissions problems in order to maintain the highest possible level of competitiveness within their industry. Indeed, programs like TRI might encourage increased cooperation among firms, since the entire industry would be negatively affected if the voluntary reporting plans indicated to the government that extensive command and control regulation was necessary.

Reasons for apparent effectiveness that may be more illusion than reality

Unfortunately, some of the TRI’s apparent effectiveness, lauded by the EPA as a "potent environmental weapon," may represent efforts by firms to present their emissions as lower than they actually are because of self-interest, or may mask certain emissions that are hidden for one reason or another, intentionally or otherwise.

First, the statutory language of Section 313 of EPRCRA allows firms to use either "readily available data" or "reasonable estimates of the amounts [of emissions] involved." It explicitly avoids mandating any additional monitoring techniques beyond those already "required under other provisions or law or regulation." This language would appear to afford firms quite a bit of leeway in establishing internal monitoring and reporting techniques. If the methods already required under other laws are weak or ineffective, the Section 313 does nothing to address this issue, and instead only exacerbates it.

Related to this issue is the problem of enforcement by the EPA. EPA is unlikely to have a large enough enforcement staff to conduct effective checks on the reports of affected firms. Firms would realize this point, and might be encouraged to be dishonest in their reporting procedures, due to competitive pressures and the desire to avoid regulation. Firms, and even an entire industry acting in conclusion, might deliberately minimize the emissions problem in order to convince the government that its regulatory dollars would be better spent elsewhere, since the brunt of the chemical emissions problem had already received apparently effective attention, and any further dollars spent on it by the government might receive diminishing or zero returns. "Please, take those dollars elsewhere, Uncle Sam, to a problem that really needs attention," the chemical firms might say.

In addition, situations may be present where firms are simply unaware that they are polluting, as in the case of New Castle v. Hartford Accident (Economic module p. 54), where New Castle was discharging contaminants without realizing it. The scienter issue in that case created a potential moral hazard situation, since New Castle would have had no incentive to improve their landfill operations if they were not legally responsible for emissions that they did not know were harmful. This is a problem that does not appear to be addressed by TRI.

Should disclosure lead to mandatory emissions controls?

The next logical step to take for the EPA is a determination of the action called for by the results of TRI. If it appears that the emissions problem is under control, the perhaps EPA’s budget could be more effectively spent enforcing other, more pressing problems. However, for some of the reasons mentioned above, the apparent magnitude of the emissions issue may be much greater than is facially apparent. At the minimum, EPA would be able to set a baseline of minimum standards for emissions, probably based upon the volume of manufacturing activity of a given firm. However, they might also see fit to institute a tradeable emissions permit system, under which firms could buy and sell permits amongst themselves, provided that the total global emissions were not greater than the chosen baseline level. Regardless of the approach taken, the information culled from TRI (discounting for now the potential problems already discussed) would provide the EPA with the ability to avoid, to some extent, the hazards of regulating under uncertainty.

Implications for OSHA’s Hazard Communication Standard

The historically problematic OSHA could apply some of the results and implications of TRI to increase the effectiveness of their Hazard Communication Standards (HCS). A period of health and safety reporting would provide OSHA with a better idea of the extent of the problems that they face. This would, at least in part, address the difficulty of standard setting, for it would provide OSHA with more information on which hazards in the workplace are most in need of abatement. It would not, however, address the more fundamental stumbling block faced by OSHA, which is the nearly impossible task of prescribing how every hazard in every one of the nation’s workplaces is to be reduced or eliminated, and creating standards that will work well in an almost infinitely-wide variety of employment settings.

Broadened reporting as applied to HCS could also have the very positive effect of increasing overall awareness of hazards, and in turn increasing employee demand for safety. Perhaps labor unions would even be encouraged to negotiate over safety issues, something that they have historically resisted. The increased awareness would serve to increase public and political pressure on HCS issues, since it would address the information imbalance that currently exists largely at the expense of the labor market. It would also tend to minimize some of the cognitive dissonance and heuristic biases that might cause the labor force to assume that workplaces are safer than they are in reality. In any event, once again discounting the aforementioned dangers and limitations of voluntary reporting programs, they would at the minimum make employees more knowledgeable consumers, and create a more equitable relationship between employer and employee. They would also have the crucial effect, discussed above in the TRI framework, of increasing overall accountabilty.

--------------------------------------------

Part I To examine why a disclosure program like TRI has been so effective in reducing the output of toxins a hypothetical comparison of two firms may help:

Firm B (for bad) merrily pollutes the environment and releases a TRI reflective of its disdain for the trees and fishes around its facility. Now such a firm will face a number of potential costs:

1) Due to the public release of information, transparency is increased and along with it the threat of public pressure. Information disclosure is empowering to concerned groups in that it allows for increased lobbying and even litigation efforts. Thus environmental groups could use TRIs to pressure firms as could community activists. This could take the form of community protests, consumer boycotts or divisive annual meetings (and what director wants to get questions regarding such issues?) Also information disclosure programs may enjoy wide political support as release of information cures an alleged inefficiency of the market while also satisfying advocacy groups by putting information into hands of the public.

2) While the firm need not collect data outside of that already mandated, the provision of such data to government regulatory bodies raises the possibility of further regulatory action by outside agencies which can be very expensive.

Firm A (for altruistic) see that its TRI could prove quite costly, given recent pollution problems at its plant. But Firm A also recognizes that almost its emission problems are probably so-called ‘low-hanging’ fruit; that is problems addressed on a low-cost basis. [Moreover, Firm A will also gain a competitive advantage with B by being able to trumpet its environmental concern to its consumers.] The transparency of self-disclosure and the potential costs associated with it if the results are bad encourage firms to take/find low-cost alternatives rather than face regulatory pressures that could result in high-cost centrally mandated solutions to publicly perceived problems. Providing information on risks may also reduce the psosibility of future litigation and regulatory costs. Think of doctors: studies show that one key determinant in who gets sued is the level of patient/provider trust. Here firms who disclose increase trust and customer loyalty whih maywell increase sales of its products, thus obviating any of the costs of information release.

Part II The mechanisms of success can be analyzed in comparison to normal problems of information gathering.

Ashford & Coldart noted that information on risk is attained in three stages.

(1) Must exist or be created – Here the firms face no cost burden in generating the information needed for the TRIs. This is because the information is a) readily known to or available to the firm, or b) the firm may use data collected pursuant to other provisions of law. This means there is no extra cost to the firm, especially since where no such data is available the section 313 only requires that the firm use "reasonable estimates." Reasonable must be widely denied as section 313(2) goes on to say that nothing requires monitoring or measuring beyond that already required by law.

(2) Must be obtained by the affected parties – Here the EPA and presumably state and local government ensure the distribution of the relevant data. Government involvement in the collection and dissemination of data ensures that no private groups or individuals must bear the costs of collection and dissemination (for which there might not be a market as Breyer notes).

(3)Those parties must be able to interpret the data – Again here the government requires uniformity so comparisons may be made, and thus presents the data in a manageable form understandable to the public.

Thus, the normal informational hurdles are overcome in the case of the TRIs.

Now several caveats. The hypothetical concerning Firm A and B illustrates the incentives that operate on a firm to produce TRIs indicating good behavior. This pressure, combined with the emphasis on self-reporting, could produce pressure to skew reports, i.e. by using unreasonable "reasonable" estimates. It is unclear whether the 52% drop in emitted toxins is from an independent study or from the firms’ own reports. If the latter a significant possibility of over-reporting exists.

This caveat also illustrates one reason why mandatory regulation in light of TRIs is unlikely to be beneficial. Such mandatory regulation will increase the incentives on firms to either doctor the numbers or restrict the reports to information garnered from measurements required by other programs.

If firms begin releasing only information already required, the regulatory benefits will be few. Thus, implementation of mandatory emission controls on the basis of the TRIs would raise the costs to firms significantly while reducing the benefits. Also, mandatory regulation would lead to problems of enforcement similar to those suffered under HCS. For instance, how could you enforce a standard as vague as "reasonable estimate?"

The system works well now because it places an emphasis on self-regulation; that is allowing firms to use their own knowledge to implement those changes that are most economical and efficient to them. In this sense transparency and external pressures must be high enough to cause firms to make low-cost changes but not so high as to make the risk of disclosure too high in relation to the benefit. We can also hope, of course, that transparency leads to internalization of outside norms regarding polluting behavior.

HCS was implemented by OSHA in response to the market failures existent in the workplace risk information market. The market failure in information was that:

1) such information did not exist at least not in an easily accessible form.

2) could not be obtained by workers, due to firm reluctance to publish such information (i.e. could reveal trade secrets, could be used against it in trade negotiations).

3)information that was obtained not easily understood by workers as their unions due to the typical questions of workplace disease (how much exposure, how long a latency period) and heuristic biases.

Because information is crucial to any market based approach to workplace safety its absence severely handicapped both workers and theorists (i.e. Viscusi).

But HCS requires mandatory disclosure in accordance with a two-tiered set of guidelines. OSHA created a list of approximately 2400 chemicals that were deemed hazardous and then said firms were to determine if other workplace chemicals were harmful on the basis of "available scientific evidence."

But despite fairly vigorous enforcement attempts it is widely conceded that the HCS mandatory disclosure system has not worked very well.

Success of TRIs allows us to draw several distinctions:

(1)Information collection costs under TRI are very low as noted earlier. Firms very much allowed to collect their own data. By contrast the HCS system has, in an attempt at flexibility, created hard to understand standards (i.e. "Available scientific evidence") that are reliant on value judgments and other fairly subjective criteria. In a mandatory disclosure scheme the presence of standards lead to problems of interpretations and litigation. This vagueness makes enforcement and compliance very difficult.

(2)Class to whom information is given differs widely. HCS mandates disclosure to workers themselves, which is quite distinct form release to the general public.

Firms can see some benefit from release of information to workers, but mainly see the problems that initially prevented them from releasing information. Firms and labor are locked in an adversarial bargaining posture that dilutes possible benefits. This divergence in costs/benefits meant TRIs are far more aligned in accordance with firm interests then the disclosure requirements under HCS are. This lack of incentive leads to mandating disclosure.

By contrast release to the public carries risks but also advantages in increased sales (indeed sales increased 10%) and decreased litigation costs.

(3) One benefit of TRIs that could be applied to HCS is the requirement of "common units" in 313(2). One of the chief problems with HCS is absence of standardized reporting, i.e. forms and education programs may take a shape determined by the company. This leads to difficulty in comparison, and the pressure of visible comparison would be a highly effective means to compel performance. Under TRIs advocacy groups can clearly compare Firm A and Firm B and deploy pressure accordingly.

Thus, TRIs require lower costs on the part of firms and carry greater benefits; HCS imposes more costs on firms for benefits primarily directed at the workers. By the same token it is unlikely firms would release HCS information if there were no mandatory requirements. So the different environments within which disclosure is being generated affect the processes by which that information may be generated.

Exam # 747

 The effectiveness of the TRI program may be due to intricacies of the market. Famed economist Adam Smith posited that the invisible hand led to the most efficient outcomes. For example those who produced undesirable products are soon kicked out of the market. The laws of supply and demand allocate resources efficiently. In this instance the laws of supply and demand are at work here as well.

The EPA administration has shown that members of the Chemical Manufacturers Association has reduced their emission of toxins by 52% while sales volume went up 10%.

Increase public awareness of the levels of toxic chemicals could influence the choices consumers make. Those companies that are more environmentally friendly receive a greater demand of their products by consumers. Thus those companies seeking to either capture a larger market share or retain their current percentage would induce programs that would reduce their toxic emissions as the lowest possible costs.

Thus competition may have ensued between companies to be more environmentally friendly. The invisible hand will lead those who can reduce emissions to do so until MC=MB while those who cannot compete in this market will either leave or develop ways in which they can. Thus TRI may lead to new innovations in the industry as well.

The ability of the market to work in this case assumes that there were no market distortions or other instigating factors. For example, this hypothesis assumes that there were substitutes for those environmentally unfriendly products. Thus in a market where there are no substitutes consumer tastes will make no difference. In addition the EPA’s statement does not reveal if their was a 52% on all toxic emissions or among a select group of toxins. This information would suggest that for those toxins while the market does not impact regulation for those substances may be necessary.

The 52% reduction may be due to other causes not necessarily related to increased reporting. For example, current technology used in the industry may have been reducing emissions during this period independent of the reporting. Also one needs to consider if regulations already in place such as the HCS may not have also had an impact in the reduction of emissions. The current regulation does not require any compliance with a standard save those already in place. Thus those limitations imposed elsewhere such as the NAAQ’s may be behind the 52% reduction in toxic admissions.

Without further study it is difficult to know what is actually the cause of the reduction in emissions of toxic substances. It may be that companies are subject to greater accountability that has led facilities to decrease their emissions level. 1(B) requires certification by a senior official. Companies may have seen an increased risk of tort liability through the TRI system in that the public is acutely aware of the dangers posed by their operations. Thus those living near a facility may now have the necessary information that will demonstrate causal links, without requiring expensive fact finding. Part of the large prohibitive costs of legislation are incurred through discovery. Here the government has required the facilities to provide that information to every possible future claimant. Voluntary reduction of emissions may have been a result of a cost benefit analysis. It is cheaper for the companies to pay to reduce emissions now than expensive litigation later.

It is difficult to know whether or not the TRI information should lead to mandatory emissions controls. It must first be determined what are the causes for this reduction. If it is indeed market forces that have led to reduction in emission then it should follow that those substances that don’t respond to market forces should be subject to some form of regulation.

However the types of regulation should be predicated on the costs and benefit produced by the regulation, as well as some risk-risk analysis to determine if the risks posed by the specific toxin actually support large investment capital needed to regulate it. The Clean Air Act of 1970 demonstrates how expensive command and control legislation can be.

In addition one should be wary of consumer preferences in that they are subject to change. Thus consumer choice for environmental friendly companies today may be supplanted by other needs or preferences tomorrow. Thus one should be wary of relying solely on the market to allocate resources where the market is concerned. The Coase theorem of efficient allocation assumes that transaction costs are zero. As transaction costs increase, the economically efficient outcome may not be the most environmentally sound.

Emissions controls may still be necessary in the case where increased accountability has led to reductions. This may be so because the mere fact that a decrease has occurred does not mean that the decrease has occurred at the safe level. It may still be necessary to reduce levels to increase safety. In addition studies have shown that public perception of risks differ from that of experts. Thus although companies are reducing emissions today due to a misperception of public knowledge about the attendant risks associated with their products, tomorrow they may realize that the public has no knowledge of the danger of the substances that they are being exposed to. A similar organization led unions to establish right to know laws. They lacked the necessary expertise to effectively bargain with employers.

The public may not be able to fully use the given information, significantly hampering their abilities to pressure companies to be more environmentally friendly through their purchases or through threats of regulation. In sum mandatory minimum standards may be needed to ensure a minimums level of safety in the TRI’s do not currently effect emissions levels or if these effects will not be long term.

The implications of TRI for OSHA’s HCS are unclear. On the one hand they may signal that HCS are having a significant impact. (If in effect reduction is a result of compliance with the earlier governmental regulation) Or it may illustrate that the type of regulation achieved through HCS may be obtained much more efficiently and cheaply through voluntary means rather that through expensive command and control regulation. Without isolating the cause of the reduction, which may be possible through regression analysis of the type used by Viscusi, it is difficult to know what impact if any TRI has on current or future EPA programs in regulating Health and Safety.

Environmental Groups (EG): Again, the release of the TRI data will reduce collective action problems among environmental groups as well as the community. Here, however, the TRI data may be put to use more effectively, as EG’s likely have greater access to interpretive devices/experts. As such, they are likely to have more accurate analysis/perception of risk created by toxic chemicals. Firms, who are aware of this, may view the reduction in emissions as preferable to release of data to EG’s who are likely to put the data through every conceivable test and may be tempted to even "massage" the data to produce the results they desire. Note as well that although the community speaks politically through its representatives, the environmental groups have better, more direct access to political process via individual politicians and agencies.

Agencies: The release of data by firms to agencies is perhaps the biggest reason for the reduction in the emissions. Previously, agency research-arms such as OSHA’s have been forced to do original research into chemical release amounts and effects. They have also been forced, frequently, to defend their findings in court. The TRI is likely to have reduced both of these problems. First, firms will have difficulty disputing their own data. Second, it signals agencies quite strongly as to where they should direct their research (before, agencies may have had to wait until adverse effects sprung up before they began snooping around firms).

Apparently, firms have decided that reducing emissions is better than either a) lying to the agencies or b) trying to defend themselves form the agencies. Also, the reductions allow them to set their own pace of reduction, as efficiently as possible. Firms that find innovative methods of emission reduction are rewarded by avoiding regulatory scrutiny while those that don’t are investigated and research is done into the effects of the emissions or the neighboring environment. Here, again, the PR issue comes up: listeners to Bloomberg Radio are usually not too excited about a stock when they hear that it is under EPA investigation (CEO stock options as an environmental tool!! ). Overall, the TRI functions as a quasi-negotiated regulation tool wherein firms self-regulate to avoid scrutiny.

Given all the hoopla about the effects of public disclosure on pressuring firms to reduce emissions, we may be led to believe that we should take the disclosure data and create a command-and-control type regulation to "beef up" the improvements realized under TRI. Why (I ask)? The measures to date have been effective; they spawn efficiency and productivity (-52% emission w/ +10% output) while promoting continued technological growth in emission reduction mechanisms. Using the emission data to set limits would restrain the need for technological innovation by allowing firms that are in compliance to stop searching for new emission reduction mechanisms. Furthermore, if the new emission caps are adopted instead of the TRI’s, we create new monitoring and enforcement costs for the agencies. Also, the command and control system would fail to signal the agencies as to how changes in market structure are leading to changes in the valuation of pollution "risk" (e.g., an increase in profit for a given good may make it worthwhile for firms to produce X even if it creates a lot of pollution. The TRI system allows firms to make this balancing test; they may choose to pollute if it is worthwhile profit-wise despite the likely public backlash; They should be allowed to make this decision, especially given the weight that firms have been giving to the public opinion/knowledge in making emission decisions).

Overall, then, we see that creating command-and-control measures based on TRI data might reduce the effectiveness of the program and reduce significant incentives to firms.

What are the implications of the effectiveness of the TRI on the OSHA’s HCS? Quite encouraging, I would assume. Workers, like average people, have problems with interpretation of data. But, as we’ve seen, this interpretative problem has not stopped the public from pressuring firms into self-monitoring. Also, workers may be more organized, thus akin to environmental groups, in that they force company union membership (sometimes) and eliminate free-rider collective action problems in the interpretation of data. Also, unions frequently have access to legislature and other political ends, as the environmental groups might. So, we see that a lot of the structures that made the TRI’s effective are in place in the workplace as well.

However, a problem may arise in that the OSHA standards may impose costs on workers in the form of reduced wage that TRI doesn’t impose on people (esp. if elasticity of demand for the product is low for price) or agencies. If we believe that wages are reduced for increased safety measures, we may have "paternalism" questions; can OSHA really make the safety decision/trade-off for workers (I say yes, but it is a question). Also, if we have a work force that is heterogeneous as to productivity, the OSHA HCS may push some workers out of work – (if costs go from X to (X+4), all the once-profitable workers at productivity (X+2) will be laid off if they can be identified).

So, although there are numerous parallels to HCS and TRI, there is always the problem of imposing costs onto the workers who are supposed to benefit from the regulation. (This HCS stuff also may not take into account the inability of workers to exit their jobs if chemical risks are too high; as Anderson says, workers may be locked-in due to firm-specific skills or deferred compensation. To what extent does "lock-in" occur within communities; e.g. is it hard for people to move if the TRI of the local firm indicates high pollution?) But, overall the TRI success probably bodes well for HCS. Also, the moral condemnation of pollution itself may, to some, be an immeasurable benefit of both schemes. Perhaps HCS and TRI both will lead to changes in the normative assessment of risk by pollution, thus reducing the need for later regulation.

------------------------------------------------------