a—v‘

 

‘ ' : " - ; aw.
' ' ' " 3:51,
. ,MY,

 

."u'iw . .r i ‘ gun 5‘ id
”3* 3‘ ”it ’i‘u, Mm ;:
, g, ."sﬁgﬁfg
" ‘ ' "‘ x , ‘ waif: 13
Rm ‘ v. A - 1‘? -" ﬁng
8:? In" 3 R” ,- 3:1}. -" , ~. 3", érﬁ‘
‘ ix ‘ 33v ga'xg . \lev ﬂs '
’ﬁ . ' , ' I,

4

!
3;» u

 

kV-hlu’uiﬁ‘a‘ .ntm . o .. -

‘ 13-:

"I.

hut.»

V??-

.5 <4

1; ($3.

t; i ‘I a I‘
ﬁr. .

 

5%?

’ ‘ .. z #5?
‘ ,, 53‘s.. Lg.
' > 3"

‘34"!

{F'h‘ ’1' 9

.. - ". mfg-g! "7'\ W

if. 419-“ ‘

novulhld-

. u.

 

r
“43*“

- “ﬁrtf- ,5," “. (I.
‘ x- .‘ -_" -,'<
”.3“, A» v} m’éﬂ ‘ ‘ ,._ ' . .
h .& ..L ”at. ‘H 4* ._
‘ ”rgaggﬁ. . {w
’r x "P '3'? JIM-i . avg-.5- n ‘ \P
’5

. ‘5'
A -L¢s‘\
‘?:’f£y
3:; .‘ 3

7.,
‘5

a
n _ “5'
N} . rm

1;.“

33%"
. 4. ,6‘

R\ '.: "
r i ‘ 33“ ‘
fﬁager‘ch .
.1 i 9', 4 .3».

‘—
E '55:"

as 4.-
(-
k 3, ~

a.
1M.

[‘qu '
- aka? $53.3.
. y ' ‘7... ‘ ‘
. "£13; a :i . any}?
‘ -! 1*; f 3:17
“43"?

it?
is?

i v:
“I;

55’

K

ﬁt?

‘3 1%.
é:\
N

 

v' 53" _
(:2 fig). I? =:‘
a. ‘- ,,

. . - 1L
[‘1'

.‘- ;— '2";
513111“; '92!
u"J,f, 7?”;

r ;* ,' x

L n I:
$3.245},
..

:‘t

2»: N
‘ Sigma
_ r z.,_

'ﬁ'". :-

' c
.17! . ,~. A "1‘ '. v‘ . ~
"‘ ’L’.‘ “'3'?" ’5 '.’“:...';‘J.‘ “ J ‘
0-! “m:,;’:.".€'l&"$f"w"' .,
.J.,......-. r "‘
~..
‘ '

 

2:2? "* , ”ml. 3:.
f u t M

m. . . ,.A .
"way. at; my”
.

2:34:44”-

I‘ .
‘ wav'rw v
' IV". v’o—o r0. ' ‘ “
' '7 1'21” 7.7?
~..| " {I 2'..'.!.-1.‘.r.... .
" ' a—- '1:00
.1. an

6 .. Jx-Jr ....

' .gfémépfh 7;}.

4..“ z.“ .1 ,
”If

 

o
-

 

 

'J .‘I.‘
Lt...' J.
‘ M

 

. a}:

   
     
 

In
’

‘ThtSab‘

\llllllllllllllllll «

This is to certify that the
thesis entitled

A POLICY ANALYSIS MODEL FOR THE EVALUATION OF DIOXIN
DIOXIN REGULATIONS IN MICHIGAN
presented by

Erich Peter Ditschman

has been accepted towards fulﬁllment
of the requirements for

M.S.

degree in Resource Development

 

 

W Major professor
Date £7) 1 7; I ?q/ '

0-7639

MS U is an Afﬁrmative Action/Equal Opportunity Institution

 

I .

1 LIBRARY
iMichlgan “I
1 University :

 

 

PLACE IN REI'UFIN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.

DATE DEE DATE DUE DATE DUE—l

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

' ‘ g—u— ~—
MSU Is An Affirmative Action/Equal Opportunity Inuitution I
l QWMHJ

 

'_.——--"'—" , , _— 7 _._.--...._

 

 

 

A POLICY ANALYSIS MODEL FOR THE EVALUATION OF
DIOXIN REGULATIONS IN MICHIGAN

By

Erich Peter Ditschman

A THESIS

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Resource Development

1991

ABSTRACT

A POLICY ANALYSIS MODEL FOR THE
EVALUATION OF DIOXIN REGULATIONS IN MICHIGAN

By

Erich Peter Ditschman

A policy analysis model using the economic concept of "optimality," is used to
assess Michigan’s dioxin policy by evaluating potential private costs to the pulp and
paper industry and the social beneﬁts of decreased risk to adverse effects of dioxin
contaminated surface water. Mead Corporation’s Escanaba integrated mill and the
Delta County ﬁsh eating population are used to illustrate the model. Three different
bioconcentration factors are used to evaluate Michigan Department of Natural
Resource’s water quality based efﬂuent limits and Michigan Department of Public
Health’s health advisory triggers. Three willingness to pay values were used along
with mill abatement costs to derive a total cost of pollution. Results indicate that the
current efﬂuent limit of 0.022 parts per quadrillion for the Mead mill is not pareto

optimal; it places an undue burden upon the mill at a cost that society is unwilling to

P3)“

Copyright by
ERICH PETER DITSCI-IMAN

1991

ACKNOWLEDGEMENTS

Phew! The acknowledgement page. The thesis committee agrees, friends and
family are elated, and there may be time on the horizon to read a novel. I would not
be able to write an acknowledgement page if it were not for the support, suggestions
and criticisms provided by my circle of friends. The encouragement I received and
the fortitude I desired, emanated from those with whom I share space: my circle of
friends. There are a few within this circle that deserve more gratitude than this page
can provide. However, this page, along with a heartfelt thank you, will have to
sufﬁce. Eckhart Dersch, thank you for inspiring me to look critically at existing
policy in an effort to derive creative answers to societal problems. Darrell King,
thank you for your patience and for demonstrating to me on a regular basis that
everything is truly connected to everything else. Tom Edens, you were there from
inception to publication; thank you for your insight, oversight, and daily support.
And, to Andria Moore Ditschman, your critique of this thesis was invaluable.

Without your love, support, and critical mind, I’d probably be a burn, or at least less

one masters degree - thank you. .

iv

TABLE OF CONTENTS

LIST OF FIGLIRES
LIST OF TABLES
CHAPTER ONE: INTRODUCTION

1.1 The Policy Making Environment

1.2 The Environmental Concern of Dioxin

1.3 Dioxin in the Great Lakes

1.4 Risk Management Approach to Regulating Toxic Substances
1.5 Michigan Dioxin Policy

1.6 Study Objective

CHAPTER TWO: DIOXIN CONTAMINATION

2.1 Sources of Dioxin Contamination
2.2 Environmental Characteristics of Dioxin
Bioaccumulation
Health Effects
2.3 The Pulp and Paper Industry as a Source of Dioxin
Contamination

CHAPTER THREE: INSTITUTIONAL SETTING FOR DIOXIN POLICY

3.1 International Response to Toxic Substances in the Great
Lakes

3 .2 Federal Response to Dioxin Contamination
Environmental Protection Agency
Food and Drug Administration

3.3 Michigan Response to Dioxin Contamination
Department of Natural Resource
Department of Public Health

3.4 Inconsistencies in Michigan’s Dioxin Policy

CHAPTER FOUR: ECONOMIC CONCEPTS FOR EVALUATING
PERSISTENT TOXIC SUBSTANCE POLICY

41 Economics as a Means to Analyze Policy
The Concept of Value
Maximum Social Wellbeing
4 . 2 Using The Concept of Pareto Optimal Solution to Evaluate

Bags:
viii

xi

H

ONQOMv—s

hip—A

23
23
24
26

29

34

35
36
41
45
47
48
54
57

61

61
62

Policy 65

4. 3 Estimating Damage Caused by Persistent Toxic Substances 72
Willingness to Accept Values for Wildlife Damage 74
Willingness to Pay Values for Reducing Risk of Death 79
CHAPTER FIVE: DESIGN AND APPLICATION OF THE POLICY
ANALYSIS MODEL 83
5.1 Study Site 83
5.2 Developing a Dioxin Damage Cost Function 91
5. 3 Developing an Abatement Cost Function 104
CHAPTER SIX: RESULTS AND DISCUSSION 109
6. 1 Results 1 11
6.2 Discussion 120
In Search of a Pareto Optimum 120
Evaluation of Michigan’s Dioxin Policy 133
CHAPTER SEVEN: SUMMARY AND CONCLUSIONS 138
7.1 Summary 138
7 .2 Conclusions 144
Potential Damage Caused by Persistent Toxic
Substances 144
Using Risk Assessment/Management for Policy
Development 145
Costs Associated With Persistent Toxic Substance
Policy 148
Policy Implications 149
7.3 Recommendation for Future Research . 151
BIBLIOGRAPHY 154
APPENDIX A Calculation of a Regional Bioaccumulation Factor for
the Escanaba River for Dioxin (Anderson et a1. , 1990) 163
APPENDIX B Rule 57 (Mich. Admin. Code r.323. 1057) 164
APPENDIX C Michigan Department of Natural Rescurces Rule 57 (2)
and Water Quality Based Efﬂuent Limit for Mead,
Escanaba Calculations 165

APPENDIX D Average Fishing Population in Delta County 166

vi

APPENDIX E Mead Escanaba Pulp Mill Draft Dioxin Minimization
Plan for Wastewater Treatment and Kraft Pulp Mill
Modiﬁcations (Abbott, 1990). 167

APPENDIX F Base Year Value Calculations for Willingness to Pay
Values 4169

vii

Figure 1.1
Figure 2.1

Figure 2.3

Figure 4.1

Figure 4.2

Figure 4.3

Figure 4.4

Figure 4.5

Figure 5.1

Figure 5.2

Figure 5.3

Figure 5.4

Figure 5.5

Figure 5.6

LIST OF FIGURES

Risk assessment/risk management process
The 2,3,7,8 - TCDD molecule

Kraft pulp operation with a conventional
bleach plant

Market diagram - supply and demand as
determinants of economic value

The optimum level of pollution

The optimal level of pollution after
technological innovation

The optimal level of pollution after new

epidemiological evidence indicates a toxic
substance characterized by a threshold

Social damage due to the release of
emissions from an economic activity

Escanaba and the Escanaba River

Mean Dioxin Discharges from March 1987 —
June 1990

Estimation of social damage cost function
due to the release of dioxin from the Mead
pulp mill

Deriving damage costs for the release of
dioxin from the Mead pulp mill

Per day Mead abatement costs and
corresponding dioxin discharges

The Mead abatement cost curve extended and
corresponding dioxin discharges

viii

31

63

70

71

72

73

84

90

92

103

107

108

T

 

Figure 6.1

Figure 6.2

Figure 6.3

Figure 6.4

Figure 6.5

Figure 6.6

Figure 6.7

Figure 6.8

Mead total abatement cost, damage cost, and
total cost ,of pollution for MDPH risk
assessments, based on a willingness to pay
value of $6.8 million dollars and a
bioconcentration factor of 51,600, as a
function of dioxin discharge

Mead total abatement cost, damage cost, and
total cost of pollution for MDNR risk
assessments based on a willingness to pay
value of $6.8 million dollars and a BCF of
51,600, as a function of dioxin discharge

Mead total abatement cost, damage cost, and
total cost of pollution for FDA risk
assessments based on a willingness to pay
value of $6.8 million dollars and a BCF of
51,600, as a function of dioxin discharge

The log of the ratio abatement cost/
willingness to pay for MDNR, MDPH, and the
FDA, and corresponding levels of dioxin
discharge

Comparison of MDPH, MDNR, FDA, ﬁsh dioxin
body burdens based on willingness to pay
values and bioconcentration factors

The risks of contracting cancer for a
population dioxin contaminated ﬁsh based
for FDA, MDPH and MDNR cancer potency
factors and a willingness to pay value of
$0.5 million

The risk of contracting cancer for a
population dioxin contaminated ﬁsh based

for FDA, MDPH and MDNR cancer potency
factors and a willingness to pay value of
$1.4 million

The risk of contracting cancer for a
population dioxin contaminated ﬁsh based

for FDA, MDPH and MDNR cancer potency
factors and a willingness to pay value of

ix

112

112

113

114

125

127

128

Figure 6.9

$6.8 million

Comparison of MDPH, MDNR, FDA, pareto
optimal dioxin efﬂuent levels based on
willingness to pay values and

’ bioconcentration factors

128

132

Table 1.1

Table 3.1

Table 3.2

Table 3.3

Table 3.4

Table 3.5

Table 4.1

Table 4.2

Table 4.3

Table 5.1

Table 5.2

Table 5.3

LIST OF TABLES

I?
u.»
0

Comparison of trace concentration units

An overview of the US EPA national dioxin
strategy 42

Risks of activities used in the development
of Rule 57 50

One in a million cancer risks used in the
development of Rule 57 51

Evaluation of sport ﬁsh caught native to
the Tittabawasee River 56

State and federal agency dioxin potency
derivations 5 9

Gross estimate of ﬁsh eating birds and
mammals near the Escanaba pulp mill on
the Escanaba River 77

Gross estimate of the quantity of ﬁsh in
the Escanaba River downstream from the Mean
Escanaba pulp mill 77

Gross estimated willingness to accept
values for ﬁsh and wildlife in the
vicinity of the Mean Escanaba pulp mill 79

Dioxin concentration in ﬁsh collected from
the Escanaba River on October 4, 1988 85

Mead abatement processes, capital costs,
depreciated capital and operating costs,
and total costs 106

Mead abatement technologies, cumulative per
day costs, and resulting dioxin discharges 106

xi

Table 6.1

Table 6.2

Table 6.3

Table 6.4

Table 6.5

Table 6.6

Pareto optimal levels of dioxin discharge
based pm MDNR and MDPH risk assessment
assumptions and a BAF of 7,238 1/kg for the
Escanaba Mead pulp mill and the sport ﬁsh
eating Delta County, MI population (costs
per day - 1990)

Pareto optimal levels of dioxin discharge
on MDNR and MDPH risk assessment
assumptions and a BCF of 51,600 l/kg,
for the Escanaba Mead pulp mill and the
sport ﬁsh eating Delta County, MI
population (costs per day - 1990)

Pareto optimal levels of dioxin discharge
based on MDNR and MDPH risk assessment
assumptions and a BAF of 140,000 l/kg,

for the Escanaba Mead pulp mill and the
sport ﬁsh eating Delta County, MI
population (costs per day - 1990)

Pareto optimal level of dioxin discharge
based on FDA risk assessment assumptions
and a BAF of 7,238 l/kg, for the Escanaba
Mead pulp mill and the sport ﬁsh eating
Delta County, MI population

(costs per day - 1990)

Pareto optimal levels of dioxin discharge
based on FDA risk assessment assumptions
and a BCF of 51,600 1/kg, for the Escanaba
Mead pulp mill and the sport ﬁsh eating
Delta County, MI population

(costs per day - 1990)

Pareto optimal levels of dioxin discharge
based on FDA risk assessment assumptions
and a bioaccumulation factor of

140,000 1/kg, for the Escanaba Mead pulp
mill and the sport ﬁsh eating Delta
County, MI population

(costs per day - 1990)

xii

116

117

118

119

119

120

Table 6.7 Damage costs per day and associated cancer
risks, based on MDPH, MDNR and FDA cancer
potencies and a BCF of 51,600 l/kg, for a
dioxin efﬂuent limit of 0.02 ppq 134

xiii

CHAPTER ONE: INTRODUCTION

1.1 The Policy Making Environment

Regulating toxic substances is similar to spending a day at a three ring circus.
In one ring, scientiﬁc professionals measure, analyze and conjecture truth and justice
in an attempt to explain the risks associated with the toxins in people’s daily lives. At
the opposite end, in a dimly lit ring, federal and state regulatory agencies juggle
platitudes from industry, environmental interest groups, and the scientiﬁc community
such as "The discharges contribute to a pollution that leads to birth defects, cancer,
respiratory problems and death" (Port Huron Times Herald, 8 April 1990), "What we
are resisting is the attempt to impose discharge limits that are unrealistically low"
(Onstream, 1989), and "Cabbage and broccoli contain a chemical whose breakdown
products behave in the body in much the same way as dioxin, one of the most feared
industrial contaminants” (University of California, 1990).

In the center ring, the main attraction, surrounded by high intensity spotlights,
are the ace reporters who excite, bemuse, and befuddle the unsuspecting audience
with explosive headlines, "Paper Mill Dioxin Pose Threat, EPA: Eating ﬁsh from
downstream raises cancer risk," (Port Huron Times Herald, 1 May 1990) colorful
placards reading, "Dioxins: Dancing with Death," (Port Huron Times Herald, 8
April 1990) and ﬂaming editorials, "Dioxin: Oops Never Mind" (Detroit News, 10
June 1990).

This is the policy making environment in which government agency personnel

2

must develop policies to regulate toxic substances. Risk managers‘ base the creation
of policy on science, public perception and political climate. At any time during the
policy making process any one of these three factors may be weighted differently,
with the resulting policy having very real impacts. Risk managers must have
information concerning the magnitude of these impacts if they are to develop and
implement policies that maximize the health, welfare, and safety of society.

This research develops a policy analysis model that can be used to examine the
impacts of regulating toxic substances in surface water. The model is rooted in
pollution cost theory, which accounts for the costs to industry of abating pollution as
well as the costs society incurs from exposure to pollution. First, using the ﬂow rate
of a river and the concentration and discharge rate of an industrial efﬂuent, an amount
of toxin in ﬁsh is calculated. Second, estimates of damages suffered by a population
from consuming toxin tainted ﬁsh are made. Third, estimates are made of the
private costs associated with decreasing or eliminating the toxin in the industrial
efﬂuent. Finally, the damages to a population and the costs incurred by a polluting
ﬁrm are compared in an effortto evaluate one particular impact: total mst to society.

This policy analysis model is applied to a particular regulatory situation in
Michigan which involves the toxic substance 2,3,7 ,8-TCDD and the Mead

Corporation’s kraft pulp mill in Escanaba. The total cost to society of the dioxin

 

\no
‘ The term risk manager refers to the regulatory agency personnel Inst.
develop and implement rules and regulations in response to legislative action.
In developing rules and regulations risk managers attempt to minimize risks in
an effort to protect the health, welfare and safety of society.

3
regulations developed by the Michigan Department of Natural Resources (MDNR),

Michigan Department of Public Health, and US. Food and Drug Administration are

evaluated.

1. 2 The Environmental Concern of Dioxin

There has been signiﬁcant debate over the actual risks that dioxin poses to
human health and the environment. As the scientiﬁc community continues to assess
the risks, public perception of dioxin as one of the most carcinogenic substances
known to humankind prevails. Until the health risks of dioxin are more fully
understood, public perception of risk will be a dominant factor in shaping federal and
state policy. Governmental agencies can ill afford to risk public health, welfare and
safety, and as a result will err conservatively in policy development.

Dioxin is the unintended side effect of a number of industrial activities and is
also found naturally in the environment. Dioxin generically refers to a family of 75
similarly related compounds, polychlorinated dibenzo-pdioxins, of which 2,3,7 ,8-
TCDD is one. A class of related compounds, polychlorinated dibenzo-furans
(furans), are generally found in close association with dioxin, which, although they
share similar chemical characteristics, are less toxic than dioxins. This research is
devoted exclusively to the study of 2,3,7,8-TCDD (referred to in this study as
dioxin).

In the absence of human health data on the effects of dioxin exposure, the

results of animal tests are extrapolated to give some indication of potential risks to

4

human health. A number of toxicological studies have been conducted to assess the
impacts of dioxin on animals. Results of these studies indicate a variety of adverse
health effects, including the occurrence of cancer (Kociba et al. 1976, Kociba et a1.
1978, Murray et al., 1979, Allen et al., 1977), on a number of animal species at
various doses. Kociba et al. (1978) reported various cancers of the liver in female
rats at extremely low doses of dioxin, 0.01 microgram per kilogram (ug/kg). On the
basis of these and other studies, current federal and state regulation of dioxin is based
on the assumption that dioxin is carcinogenic. Whether or not dioxin is actually
carcinogenic in humans is widely debated. While the debate continues, regulatory
agencies have taken the conservative position of presuming dioxin to cause cancer in
humans. This presumption creates a policy environment based on imperfect
information which can result in signiﬁcant resource expenditures by industry to meet
a standard that provides an uncertain level of protection to society.

Because of dioxin’s intense toxicity there is concern for exposure at very small
doses. Scientiﬁc instrumentation has developed to the extent that toxic substances can
now be measured at extremely low concentrations. Dioxin has been found in the
parts per trillion (ppt) range in ﬁsh, and in the parts per quadrillion (ppq) range in
pulp mill efﬂuent. Current Michigan Department of Natural Resource regulations
require the discharge of dioxin from a new pulp mill to be below 0.01 ppq level,
which is below current scientiﬁc measurement capabilities (presently limited to the 3-
4 ppq range). Table 1.1 puts these measurements in perspective; they have been

adapted from comparisons made by Warren B. Crummett of the Dow Chemical

Company (Kagel).

Table 1.1 Comparison of Trace Concentration Units

 

Unit 1 part per trillion 1 part per quadrillion

 

Length 1 inch/16,000,000 miles 1 inch/16,000,000,000 miles
Time 1 second/320 centuries 1 second/320 millenniums
Money $.01/$10,000,000,000 $.01l$l0,000,000,000

Weight 1 pinch salt/ 10,000 1 pinch salt/10,000,000
potato chips potato chips

Action 1 bogey/3,500,000,000 l bogey in 3,500,000,000,000
golf tournaments golf tournaments

Quality 1 bad apple per 1 bad apple per
2,000,000,000 barrels 2,000,000,000,000 barrels

Rate 1 dented fender per 1 dented fender per

10,000,000 car lifetimes 10,000,000,000 car lifetimes

 

There are two primary properties that can increase the hazard potential of
dioxin to humans and the environment: persistence and bioaccumulation. Dioxins are
generally resistent to biological breakdown and can persist in the environment for
years. Dioxins are also lipid (fat) soluble and tend to be accumulated by living
organisms.

The bioaccumulation property of dioxin poses a potential threat to Great Lakes
anglers who consume their catch. Fish bioaccumulate dioxin, magnifying the
concentration in water from parts per quadrillion to parts per trillion in their ﬂesh.
This increased dioxin concentration makes the consumption of sport ﬁsh a higher risk
to the human population than that of exposure to the skin, or the drinking of dioxin

tainted surface water.

I . 3 Dioxin in the Great Lakes

A number of studies have detected dioxin throughout the Great Lakes Basin.
Kaczmar et a1. ( 1985) reported detectable residues of dioxin in bottom feeding ﬁsh
ranging from 17 to 586 nanogram per kilogram (a ng/kg is equivalent to a ppt) in
many“ Michigan watersheds. Analyses conducted on ﬁsh from Lake Ontario, Lake
Erie, and the Hudson River (Smith et al., 1983) and Lake Huron (Stalling et al. 1983)
indicated dioxin in a variety of ﬁsh species with concentrations ranging from non-
detectable (at 3.2 ng/kg) to 107 ng/kg.

A number of analyses have been conducted on Michigan rivers receiving
dioxin laden discharge from industry. Results of a 1988 analyses conducted on the
Menominee River downstream from Champion Paper Company’s Quinnesec pulp and
paper mill indicated the presence of dioxin in redhorse sucker, smallmouth bass,
northern pike, and walleye in the range of non-detectable (at 0.11 ng/kg) to 1.97
ng/kg (MDNR, 1988a). A 1988 sampling of walleye from the Tittabawassee River at
Midland near Dow Chemical Company, conﬁrmed that dioxin was present, ranging
from 1.3 to 5.61 ng/kg (MDNR, 1988b). The results of a 1989 analysis of ﬁsh in the
Escanaba River near the Mead Corporation’s Escanaba pulp and paper mill identiﬁed
northern pike, white sucker and smallmouth bass with detectable residues of dioxin
ranging from 2.86 to 23.4 ng/kg (MDPH, 1989).

The ﬂushing time of the Great Lakes, particularly the deep Upper Great
Lakes, plays an important part in characterizing the potential harm from persistent

toxic substances. The ﬂushing time to remove ﬁfty percent of a conservative

7

material, such as dioxin, from the Great Lakes ranges from 120 years for Lake
Superior, 50 years for Lake Michigan and Lake Huron, approximately 7 to 10 years
for Lake Ontario, and 2.5 years for Lake Erie, (Bennett, 1978, Rainey, 1967, Sly,
1967). Long ﬂushing times increase the amount of time a material stays in the lakes

which increases the chances for the material to be bioaccumulated in ﬁsh.

I . 4 Risk Management Approach to Regulating Toxic Substances

The regulation of toxic substances occurs at the federal, state, and local level.
These regulations are manifested in a variety of forms. There are federal laws that
affect the development, use, storage and disposal of toxic substances and the
protection of resources from the disposal of these substances. Examples of these laws
include the Federal Insecticide, Fungicide, and Rodenticide Act of 1972, Safe
Drinking Water Act of 1974, Toxic Substances Control Act of 1976, Clean Water
Act of 1972, and the Resource Conservation and Recovery Act of 1976. States
promulgate rules as a result of state and federal legislation. Michigan’s Water
Quality Standards are promulgated as rules (Part 4 of the General Rules of the Water
Resources Commission) under the Water Resources Commission Act, PA. 245, of
1929 as amended. Local ordinances can also play a role in regulating toxic
substances through the siting of chemical production facilities. The primary process
used by both federal and state governments in regulating toxic substances in surface
water is risk management.

Risk management agencies use a risk management approach to protect the

8

public health and environment from toxic substances such as dioxin. Risk
management refers to an optimal balance between uncertain beneﬁts and uncertain
costs (Haimes, 1990). According to Haimes, risk is the measurement of the
probability and severity of adverse effects. Probability refers to the likelihood that an
event will occur and severity is a measure of the magnitude of the event’s effects. In
the case of dioxin, risk includes the likelihood of a human to contract cancer as a
result of being exposed to the chemical.

Risk management is the second part of a two part process that also includes
risk assessment. Risk assessment is deﬁned by the US National Academy of
Sciences (1983) as:

...the scientiﬁc activity of evaluating the toxic properties of a chemical and the

conditions of human exposure to it in order both to ascertain the likelihood that

exposed humans will be adversely affected and to characterize the nature of the
effects they experience.
Wentz (1989) points out that ”risk assessment techniques include environmental
impact assessment, systems analysis, cost-beneﬁt analysis and probability analysis,"
and explains that "risk assessment is really based upon environmental impact
assessment in that it quantiﬁes the potential hazards of economic development and
technological change. "

In developing protection policies based on risk assessment and risk

management, risk managers must address four questions:
1. What can go wrong?

2. What is the likelihood that it will go wrong?

3. What are the consequences?

9
4. What can be done (Haimes, 1990)?

Besides risk, policies must also address the issue of safety. Measuring risk is an
empirical, quantitative, scientiﬁc activity, while judging safety is a normative,
qualitative, political exercise. Judging safety is judging the acceptability of the risks
(Haimes, 1990).

While risk assessment and risk management are independent activities, they are
often linked by risk managers to provide a system for devising risk based protection

policies (Figure 1.1).

Risk Assessment

 

Hazard Identification \

 

 

 

Dose Response _. RISK .—
Assessment Characterization

 

 

 

 

 

 

 

 

Exposure Assessment /

WM / Risk
: Communication

Development of
Regulatory Options

I

Evaluation of Options
3
Agency Decisions and Actions

 

  
  
 

 

 

 

 

 

 

 

 

 

 

 

Figure 1.1 The Risk Assessment/Risk Management Process2

 

2 Adapted from Bedford et al., 1990 and the National Academy of Science, 1983.

10

Risk assessment includes four steps:

Hazard Identiﬁcation: The gathering of information to determine whether a
particular chemical is or is not causally linked to particular health effects.

Dose Response Assessment: The determination of the relation between the
magnitude of exposure and the probability of occurrence of the proposed
health effects.

Exposure Assessment: The estimation of the number of people which will be
exposed and the characteristics of the exposure before or after application of
regulatory controls.

Risk Characterization: The integration of hazard identiﬁcation, dose response
assessment, and exposure assessment to describe the nature and often the
magnitude of human risk (Pollock et al., 1989).

After the results of risk assessment have been clearly articulated risk managers

apply risk management techniques to the decision making process. Risk management

includes three steps (Bedford et al. 1990):

1.

2.

3.

Development of regulatory options.

Evaluation of public health, economic, social, and political consequences of
regulatory options.

Consequent agency decisions and actions.

An important distinction between risk assessment and risk management is that,

ideally, risk assessment consists of objective, quantiﬁable determinations. Risk

management generally involves a multitude of unquantiﬁable factors including

perceptual, economic, cultural, and political inﬂuences (Wentz, 1989). Risk

Communication is a separate component shared by both processes. It refers to the

communication of relevant information to pertinent audiences.

Bedford et al. (1990) suggest that risk communication is the most important

11

element in the risk assessment/management process. It is the element that either
"convinces the public that the system is working and that their health is being
protected or leads to distrust of the message and the involved parties. " For risk
communication to be effective it is necessary for the agencies involved to deliver the
same message to the same audience. Bedford et al. states/"it is important that the
results of the risk assessment process be as uniform as is reasonably possible given
the different purposes for its use within individual state agencies as well as balance
between agencies.” The concept of uniformity in risk assessment methods can also
be expanded to include consistency between states as well as between federal and state
agencies.

The current sophistication in measurement and the techniques used to
determine the effects of toxic substances on certain species are activities pursued in
the realm of scientiﬁc inquiry. However, the utilization of these data for risk
management and safety judgement is done outside the realm of science in the political
arena. It is in the political arena that federal and state agencies are left to deal with
the question of what is a safe level of exposure, or ”how safe is safe?"

The protection polices developed by agencies do not directly answer the
question of "how safe is safe?" rather they provide a range of risks. Agencies use
mathematical models to "estimate the upper boundary (95 percent) on risk of
increased incidence of cancer over background cancer rates for a population exposed
to certain concentrations of a chemical over a lifetime of exposure under assumed set

of conditions (MDNR, 1984)." The result of this cancer risk assessment is generally

12

expressed in terms of additional cases of cancer in a given number of individuals.
For example, the U.S. Environmental Protection Agency (EPA) bases its protection
policies on risk assessments relating risks to one excess case of cancer per one
million exposed population. The Michigan Department of Natural Resources and the
Michigan Department of Public Health both use measures of risk relative to one
excess case of cancer per 100,000 exposed population. Risk managers can compare
the risks associated with a certain toxic chemical to risks in people’s lives.

On a daily basis people are faced with numerous risks both unavoidable, such
has being .struck by lightening and being in an automobile accident, and avoidable,
including skiing, swimming, and hunting. By evaluating the range of avoidable and
unavoidable risks, a risk manager can choose a limit that relates the risk associated
with being exposed to a toxic chemical that provides some margin of safety that

exceeds the majority of risks that people face.

1.5 Michigan Dioxin Policy

The Clean Water Act of 1972 (33 U.S.C. 1251 et seq.) speciﬁes a number of
mandates to achieve its rigorous goal of maintaining the integrity of the Nation’s
waters. The Act calls for the elimination of pollution discharges into navigable
waters and declares a national policy "that the discharge of toxic pollutants in toxic
amounts be prohibited." In the development of the Act a number of non regulatory.
"market" mechanisms, including tax-subsidy approaches and pollution certiﬁcates, ’

were proposed to control pollution. However, in the end, Congress adopted a

l3
regulatory approach which directly addresses, the quantity of pollution discharged

(Goldfarb, 1988).

The Act provides for the development of water quality-based efﬂuent limits
and technology-based limits to control pollution. According to Goldfarb, the Act’s
principal control mechanism ”is uniform national technology-based efﬂuent
limitations, progressively tightened until a ’zero discharge’ goal is reached.“ As an
interim step, until these technology-based efﬂuent limitations are developed, water
quality-based efﬂuent limits are used by states to control the amount of toxic
substances entering surface waters. Federally authorized state agencies generally use
a risk management approach in developing state water quality-based efﬂuent limits
(referred to as efﬂuent limits).

The Michigan Department of Natural Resources (MDNR) uses risk
management to formulate policies to protect the population from the potential adverse
health effects of exposure to dioxin contaminated surface water. The MDNR
develops efﬂuent limits as part of an EPA authorized state National Pollutant
Discharge Elimination System (NPDES) program. ll'hese limits act as warnings that
no discharge will be tolerated if it disrupts the integrity of a body of water (Goldfarb,
1988). The MDNR established an efﬂuent limit of 0.01 parts per quadrillion (ppq)
for new discharges of dioxin into Great Lakes tributaries. The current NPDES permit

for the Escanaba Mead pulp and paper mill3 is 0.022 ppq dioxin for discharges to

 

3 There are 104 bleach kraft mills in the U.S. Each mill has its own speciﬁc process
conﬁguration, however, the mills generally share a number of similar characteristics
such as the use of a chlorine based bleaching process and kraft pulping technology.

14
the Escanaba River.

The Department of Public Health (MDPH) uses risk management to deveIOp
sport ﬁsh consumption advisories and corresponding health advisory triggers to
inform the public of the risks associated with eating sport ﬁsh burdened with toxic
substances. The MDPH has set 10 ppt as the health trigger for dioxin in ﬁsh. If a
body of water has been identiﬁed as having ﬁsh with a body burden of 10 ppt or
greater of dioxin, an advisory is issued warning individuals not to eat the
contaminated species. The question that must be addressed is what relationship
exists, if any, between water quality based efﬂuent limits and health advisory
triggers?

Both the MDNR and MDPH use risk management in developing dioxin
regulations. However, the differences in an agency’s application of risk management
techniques, in the development of efﬂuent limits or health triggers, may have
signiﬁcant effects on social welfare when they are implemented. Combs et al. (1989)
state:

. . .subjectivity is a factor in the interpretation of the data upon which the

standard is based. Two standard setting bodies, using the same toxicological

information, and equally qualiﬁed personnel to interpret that information, may
produce numerical standards that are markedly different. This subjectivity in
the interpretation of the data has led to many questions about the
appropriateness of various numerical standards and to the belief by many that

the numbers represent little more than ”black magic. "

Although the MDPH health trigger is not a numerical standard, but rather a

 

The Mead Corporation’s Escanaba Mill is representative of the characteristics found
in the majority of the U.S. bleach kraft mills.

15
mechanism for communicating risk, it is an integral part of Michigan’s policy to
control human exposure to dioxin. The MDNR’s efﬂuent limit is the target the
discharging ﬁrm must meet, while the health trigger is a value that indicates the level
at which a ﬁsh can be burdened by a toxic substance and still be safe to eat. Having
two agencies independently formulate dioxin policy can cause miscommunication,
provide for inconsistencies, and raise serious questions concerning the effectiveness of
these policies in protecting human health.

The term effectiveness, when used in the context of gauging policies that
protect human health, refers to whether or not a policy is protecting a population to a
given level of risk, or in some cases, a range of risks. The risk level deﬁnes the
region in which agencies have judged exposure to be safe.

One inconsistency between agency responses is their cancer potency factors.
Both agencies use the linearized multistage model4 for extrapolating animal data to
humans, an assumed risk of one excess case of cancer per 100,000 exposed
population, and the same animal study (Kociba et al. 1978). However, because of
different risk lillit'iieis'i{n‘é‘r‘it assumptions they derive potencies that are signiﬁcantly
different. The MDNR uses a potency factor of 1.51 x 10’ mg/kg/day and the MDPH
uses a factor of 3.57 x 10‘ mg/kg/day. Because the potency value is critical to

determining the cancer risk of a toxic substance, this magnitude of difference in the

 

The linearized multistage model is used to ﬁt animal laboratory data which was
generated in a high dose range to a dose response curve in order to predict responses
to a particular carcinogen at low doses. The model assumes that there is a risk of
cancer being developed at any dose.

16

factor alone will have a signiﬁcant effect on the safety margin of a particular
protection policy. Other inconsistencies between regulatory responses to dioxin

control are identiﬁed in Chapter 3.

1. 6 Study Objectives

The desirable approach to the assessment and management of any particular
chemical includes the analysis of all relevant information (Bedford et al. , 1990).
Bedford et al. suggest that ideally this evaluation would be conducted by a panel of
experts composed of representatives from the disciplines of medicine, toxicology,
ecology, sociology, economics, and other relevant ﬁelds. However, such a panel
would require tremendous expenses in terms of money, time, and effort, all of which
are in limited supply in state and federal agencies.

In the absence of such a panel, agency personnel are left with limited
information on which to base the development of toxic substance policy. To assist
risk managers in making decisions, a policy analysis model based on biological and
economic data is developed which provides information on the exposure and
associated risks of a toxic substance as well as the associated costs.

A model is merely an abstraction of reality. To gain utility from a policy
analysis model, several factors must be addressed which, when combined, lead to an
outcome which the policy was designed to effect. The ﬁrst key factor that must be

addressed by the policy analysis model is a ﬁrm’s choice of abatement methods. For

17

example, an efﬂuent limit for a toxic substance is developed in order to prevent the
further contamination of a water resource from a particular contamination source.
The efﬂuent limit adopted by the risk manager is determined to be at a level that
protects the health, welfare and safety of a particular population. This efﬂuent limit
sends a message to the source that it will face regulatory action if it does not comply
with the limit. The source, usually an industrial ﬁrm, in turn adopts technologies in
order to decrease the concentration of the toxic substance in the efﬂuent to the
prescribed level. Until the source is in compliance, and even after it reaches the
efﬂuent limit, the toxic substance will be released at some level into the environment.

Each ﬁrm is faced with a set of choices in determining which abatement
technology to adopt. These choices are limited by a number of constraints including
the cost and effectiveness of the technology. A ﬁrm will generally seek to just meet
the efﬂuent limit using the least cost technology. The effectiveness of the technology
will determine the amount of toxic substance that enters a water body.

The fate of the toxic substance is the second key factor that must be
addressed. Once the toxic substance is in the aquatic environment it may adsorb to
particulate matter and immediately settle on the bottom of the water body or it may be
transported some distance before it settles; it may also be consumed by aquatic
organisms. The substance could also dissolve in the water or volatilize into the
atmosphere.

If the toxic substance is incorporated into ﬁsh tissue then the fate of the

contaminated ﬁsh must be identiﬁed. If the ﬁsh is eaten by predatory species such as

18

herons, bears or humans, the third factor in the model is to identify the effects on the
species which consume the contaminated ﬁsh. If the consumption of contaminated
ﬁsh causes harm, a fourth factor is to determine the costs associated with this harm.
Each of these factors must be examined within the policy analysis model if a complete
picture of the effectiveness of the efﬂuent limit is to be developed.

The policy model developed in this research is based on the economic theory
of social welfare maximization which states that the most desirable condition for any
society is maximum social well-being (welfare) at a minimum costs. When applied
to the problem of assessing the impacts of toxic substances on public health, the key
component of the model is the amount of a toxic substance that ends up in ﬁsh tissue.

There are three primary factors governing the amount of toxins in ﬁsh tissue:
efﬂuent concentration of a toxic substance, rate in which efﬂuent enters a water body,
and rate of ﬂow of the water body. Other ancillary factors will be discussed in
Chapter 2. Once established, the amount of the contaminant in ﬁsh can be used to
identify potential exposure to a speciﬁc population. This exposure can then be used
to characterize the risk of contracting cancer for the population.

Economic valuing techniques are applied to a population’s risk of contracting
cancer in an effort to determine a damage cost for dioxin exposure. Economists have
recently begun to quantify what individuals would be willing to pay for reducing their

risk of death. Although the risk of death is one (everybody will die) there are

 

Maximum social well-being is deﬁned as a state "in which society is as well off as it
can possibly be, given its resource base, its productive technology, and the tastes and
preferences of its members" (Randall, 1987).

19

activities which, if undertaken, have been statistically shown to provide a greater
chance of risk related death than other activities. For example, based on a ﬁfty year
period a ﬁre ﬁghter has a 32 in 1,000 chance of dying on the job, while a service or
government worker has only a 40 in 10,000 chance of dying on the job in the same
period (Clark et al., 1987). The "value of life", as stated by Blomquist (1979), "is
based on changing the probability of survival by a small amount." Because risk is not
traded in markets economists use various methods to determine what people are
willing to pay to reduce their risk of dying from speciﬁc avoidable behaviors.

Economists estimate a willingness to pay value to indicate the change in well-
being that would result from changing the risk of death related to speciﬁc activities.
They derive this estimate by measuring "how much of other goods and services a
person is willing to give up to get a reduction in the risk of death (Fisher et al.,
1989). " This concept of valuing the risk of death from avoidable behavior is critical
to the development of this policy model. A more detailed discussion will be provided
in Chapter 4.

By associating willingness to pay values with a population’s risk of contracting
cancer, the implied costs that the population incurs due to the exposure of a toxic
substance can be calculated. These costs are referred to as "damage costs". In this
study, the damage costs are derived by applying values (found in the literature) that
people are willing to pay for reducing their risk of dying from speciﬁc activities to
the potential risk of death, from contracting cancer resulting from eating dioxin

burdened ﬁsh, faced by the Delta County sport ﬁsh eating population. Costs can

20

also be calculated to reﬂect what a discharging ﬁrm incurs to control the release of
the toxic substance to a regulated level. These costs are called ”abatement costs. "

This policy model is used to evaluate both current and potential policies in
order to determine whether or not the pollution control level maximizes social welfare
and minimizes the costs to society-~usually referred to as Pareto optimality”. This
model takes the earlier explained concept of effectiveness a step further by
incorporating the constraints of maximum protection at a minimum cost. The water
quality based efﬂuent limits and the health advisories are developed to provide an
adequate margin of protection to society without much concern to the costs involved
in achieving the protection. This model adds potential damage costs and abatement
costs to the protection equation and extends the evaluation of whether or not a policy
is protecting a population to a given level of risk to include the costs imposed on
society of achieving a particular level of risk.

This study will attempt to determine whether or not the inconsistencies in
applying risk assessment/ management to the development of standards and health
triggers by Michigan regulatory agencies provides a toxic substance policy that is
economically optimal (in the Pareto sense). A policy analysis model is developed and

applied to determine whether or not the application of the Michigan Department of

 

This concept of maximum welfare at a minimum cost is often referred by economists
as achieving a Pareto optimal solution. Named for the Italian economist Vilfredo
Pareto (1848-1923), the concept refers to a solution in which no one individual can be
made better off without necessarily making some other individual worse off. The
term Pareto efﬁciency is also used to describe such an allocation as "efﬁcient if
conditions cannot be made unambiguously better (Nicholson, 1985). "

21
Natural Resource’s 0.022 ppq dioxin efﬂuent limit for Mead’s Escanaba kraft pulp

mill and a Michigan Department of Public Health dioxin health triggercf 10 ppt in
ﬁsh for the sport ﬁsh eating population of Delta County leads to an efﬁcient (Pareto
optimal) allocation of resources. The resgrchable question for this study is: do the
private costs to the Mead Corporation’s Escanaba kraft pulp mill, incurred as a
result of the Michigan Department of Natural Resource’s dioxin efﬂuent limit,
exceed the benefits of the reduced risk of contracting cancer, which are based on
MDPH and MDNR risk management approaches, received by the Delta County
sport fish eating population?

The 9mm of this thesis is to develop a policy analysis model that can be
used to evaluate toxic substance regulations in Michigan, by identifying the private
costs to industry and the beneﬁts to society of decreased risk to adverse effects of
toxic contaminated surface water, and determining a range of Pareto optimal
solutions. A critical component of this study lies in translating a quantity of a
regulated efﬂuent into a measurable quantity in ﬁsh in order to evaluate whether the
MDNR efﬂuent limit provides the margin of safety required by the MDPH health
trigger without unnecessarily burdening the discharger. The U.S. Food and Drug
Administration risk management approach to dioxin regulation will also be evaluated
with this policy analysis model in order to compare the compatibility of federal and

state approaches to regulating dioxin.

CHAPTER TWO: DIOXIN CONTAMINATION

Dioxin has been at the forefront of the public’s interest since the early 19703
discoveries of dioxin contamination in New York’s infamous "Love Canal" and at
Times Beach, Missouri. Since its discovery, the compound has been at the center of
policy debates between citizens and regulators, regulators and industry, and amongst
various regulatory agencies. It has also been the subject of intense scientiﬁc scrutiny,
especially the 2,3,7,8 TCDD. And, on a regular basis has been a prominent feature
in the developed world’s newspapers.

Through their investigations, scientists have identiﬁed a number of sources,
both natural and human based, of dioxin. The fate and transport of the chemical,
along with its chemical characteristics, is now fairly well understood in the scientiﬁc
community. As a result, industry is developing process changes to reduce the ﬂow of
dioxin from industrial activities into the environment. However, even after signiﬁcant
resource expenditures and countless hours of research, one key piece of information
remains uncertain: the effects of dioxin on wildlife and human health.

This chapter provides a limited literature review in an effort to provide a brief
summary of information concerning, the sources, chemical characteristics, and health
effects associated with dioxin. The information will provide a foundation for

understanding the development of dioxin policy in the United States.

22

23

2.] Sources of Dioxin Contamination

Dioxins are by-products unintentionally created in the manufacturing of other
chemicals. Regulatory concern for dioxin originated from the 19705 detections of
dioxin in the herbicide 2,4,5-T. Other sources identiﬁed in the 19705 included the
manufacturing of 2,4,5-trichlorophenol (2,4,5-TCP) and hexachlorophene. The
discovery of 2,4,5-T in waste oil still bottoms used in horse arenas in Times Beach,
Missouri and the leaking of wastes from earlier chemical production, including 2,4,5-
T, disposed of in the "Love Canal" in New York, provided the impetus for federal
and state regulatory policies for the compound.

In the late 19703, reports surfaced of dioxin being released from combustion
sources, particularly municipal waste incinerators (Barnes, 1985). Dioxin is now
associated with metallurgical processes such as smelting, the manufacture of pulp and

paper and car exhausts (Rappe, 1988).

2.2 Environmental Characteristics of Dioxin

Speciﬁc polychlorinated dioxins are deﬁned by the number of chlorine atoms
attached to the basic molecules, and by the position of the chlorine molecules. The
2,3,7,8 TCDD molecule is identiﬁed by its four chlorines, and is considered the most

toxic (Figure 2.1) (Rappe, 1988 p.137).

24

Cl 0 Cl

0 0

CI 0 Cl

Figure 2.1 The 2,3,7,8-TCDD molecule.

Dioxin is a colorless crystalline solid at room temperature. It is nearly
insoluble in water and tends to travel on particles or accumulate in sediments and
organisms (Wentz, 1989). The compound binds strongly with organic carbon and
sediments which W opportunities for the compound to volatilize to the
atmosphere, to be degraded by the sun, or react with water. Once the compound

enters an aquatic ecosystem there is very little chance of it exiting.

Bioaccumulation Because of its persistent nature, dioxin is available to be
bioaccumulated by ﬁsh. Bioaccumulation is the process by which an organism
accumulates a substance as a result of ingestion of water and food in which the
substance is dissolved or to which the substance is bound. Dioxin is also highly
soluble in fat which increase its opportunities to accumulating in the fatty tissue of
ﬁsh. A number of studies have shown the potential for ﬁsh to bioaccumulate dioxin
(Metcalf and Liu, 1973; Isenee and Jones, 1975; and Tsushimoto et al., 1982).

Dioxin uptake in ﬁsh varies signiﬁcantly between species of ﬁsh and the organs

25
within a ﬁsh (Keuhl et al. 1987, Muir et al. 1985). Body burdens (an amount of

toxic substance incorporated in ﬁsh tissue) of dioxin in ﬁsh result from dioxin
concentrations in several sources including river water, organic carbon consumed
directly by the organism either passively or actively, and food; and by several
species-speciﬁc parameters including: uptake, assimilation and feeding rates,
deputation, and growth (Anderson et a1. 1990 p.8).

Bioaccumulation factors (BAF) and bioconcentration factors (BCF) are unitless
ratios of toxic substance concentration in water to the concentration of the toxic
substance in ﬁsh. Bioconcentration refers to the process by which a dissolved
substance can be taken up by an organism directly from water. The type of data and
its availability generally determines the method for estimating bioaccumulation or
bioconcentration factors.

It has been suggested that food chain models utilizing site-speciﬁc data with
their inherent kinetic properties are probably the best means of estimating body
burdens (Anderson et al. 1990 p. 18). In the absence of food chain models, Anderson
et al. (1990) suggested utilizing site-speciﬁc data for a kinetic approach as the next
best method. Using a mean dioxin concentration from the analysis of twenty—two ﬁsh
(northern pike, smallmouth bass, and white sucker) of 10.69 ppt and an Escanaba
River dioxin concentration of 1.48 ppq, Anderson et a1. calculated a site speciﬁc BAF
for dioxin of 7 ,238 (see Appendix A).

The BCF utilized by the MDNR is standardized to reﬂect the value for fresh

ﬁsh tissue having a lipid content of 9.6% , and is preferably determined using

26

standardized laboratory tests. If bioconcentration factors are not available from
laboratory studies, ﬁeld data may be used to determine a BCF. In the absence of
ﬁeld data MDNR guidelines require the BCF be calculated using the following
equation:

log BCFc = 0.847 log Kow - 0.628 (MDNR, 1985)

where: log Kow = the log (base 10) of the ratio of the octanol to
water equilibrium concentrations of a compound.

BCFc = calculated bioconcentration factor from log Kow
or other regression equations.

The current laboratory derived BCF for dioxin is 51,600 for developing water quality
based efﬂuent limits for dioxin in Michigan (Taft, 1990).

An EPA laboratory determination of lake trout dioxin bioaccumulation kinetics
speciﬁc to Lake Ontario ﬁsh exposure conditions concluded that 140,000 is a
reasonable BAF estimate (Cook, 1990). These three estimates of a BCF and BAFs
are a representative sample of the variability that exists in determining the potential
for ﬁsh to accumulate dioxin. The EPA (1990b) used similar BCFs of 5,000 and
50,000 in a recently published risk assessment study of dioxin contaminated receiving

waters from U.S. chlorine-bleaching pulp and paper mills.

Health effects There have been a number of accidental human exposures to dioxin.
These exposures have generally been workers who were accidently exposed to dioxin

during the production or handling of the 2.4.5-trichlorophenoxy acetic acid (2,4,5-T)

27

and products made from this chemical, or through the use of other herbicides
contaminated with dioxin. Studies of exposed individuals are at times contradictory
and are often inconclusive.

For example, a comparison study of 1,261 Agent Orange (a herbicide
containing the dioxin contaminated 2,4,5-T and 2,4-D used to defoliate jungle flora
during the Vietnam War) exposed personnel and 19,101 non exposed personnel
resulted in no statistical difference between cumulative mortality (Wolfe et al. 1989).
However, a Vietnamese study derived the tentative conclusion that the difference in
the infant mortality rates between the villages sprayed with agent orange and the non-
sprayed villages suggests that there may be increased health risk to infants consuming
dioxin-contaminated mother’s milk (Dai et al. 1989).

Bonsor et al. (1989) summarize the rampant uncertainty associated with
current dioxin epidemiological studies,

(1) People who' lived in a trailer park at Times Beach, Missouri, where
[dioxin] contaminated oil was used for dust control, show serious impairment
of the body’s immune system, which has the function of protecting from
infectious agents. The effect has been noted in many studies with animals, and
is associated with a loss of lymph tissue. Therefore it was surprising that the
impairment in Times Beach people was not accompanied by increased
morbidity or mortality, and has not been documented in other human studies.
(2) An association between increased soft-tissue cancer and degree of exposure
to herbicides was shown in Swedish workers. The results cannot be directly
attributed to PCDDs because of the great mixture of chemicals included in the
exposures. (3) Liver damage, as measured by various tests of liver enzyme
function, was seen in at least 3 locations among the workers who showed
chloracne, and was also documented following the Seveso explosion, and
among Americans who had sprayed agent orange. Liver damage also shows
up in most sublethal studies with laboratory animals. (4) Damage to the
nervous system has been evident in several of the more severe exposures of
humans, mostly in the peripheral nerves, with loss of sensation. For each of
these items, there were other studies which failed to ﬁnd such effects.

28

The following four deﬁciencies are found in most if not all of the current
epidemiology studies making it difﬁcult to derive a deﬁnitive conclusion on health
effects and the cancer risk of dioxin to humans: (1) lack of sufﬁciently measured
exposure, (2) lack of sufﬁcient time between exposure and the study for disease
development, (3) a population too small for one to expect to ﬁnd cases of soft tissue
sarcoma, and (4) possible lack of contamination of the commercial product with
dioxin (CDC, 1983 p.3).

In the absence of concrete human health data animal studies are used to make
extrapolations of adverse human health effects. The toxicity of a substance is
traditionally measured through the use of a bioassay. In a bioassay, a single dose of a
measured amount of a toxic substance is administered to rats or other species to
determine how many animals die. The dose at which 50 percent of the animals die is
considered as the LD50 (lethal dose). An oral LD50 is derived by dosing the animal
by delivering the material to the stomach. The oral LD50 of dioxin varies
signiﬁcantly among species, from 06-2 ug/kg in guinea pigs, and 44 ug/kg in rats to
about 5 mg/kg in hamsters (CDC, 1983 p.5). In spite of this range, among different
tested animals, dioxin is still considered extremely toxic (CDC, 1983).

Because of its accumulative property, after repeated dosing, dioxin is stored in
adipose tissue and to some extent, in the liver and other organs. Lifetime studies
conducted on rats and mice have concluded that dioxin affects reproduction in female
animals, depresses the cell-mediated immune response, is toxic to the liver, and

causes cancer (CDC, 1983). Through the evaluation of animal bioassays, the EPA

29
(1984) independently concluded that several rodent studies establish dioxin as an
animal carcinogen in multiple species and organs and is probably carcinogenic in
humans.
The MDNR, in developing its regulatory response to toxic substances, assumes
that all animal carcinogens are human carcinogens. The MDNR assumes the
following: J

since every known human carcinogen, with the exemption of arsenic, has also

been found to be carcinogenic in animals, prudent policy is to accept the use

of such data, rather then wait for proof of human carcinogenicity (MDNR,

1984 p.27).

In the Kociba et al. ( 1978) study animals were given dioxin in the doses of
0.001, 0.01, and 0.1 ug/kg/day. The animals in the highest dose group exhibited
various cancers of the liver in 11/49 female rats compared to 1/ 86 in the control
group. At the 0.01 ug/kg dose, hyperplasia of the liver and lung was observed. The
lowest dose group observed no adverse effects and was considered the no-observable-
adverse-effect-level (N OAEL"). The Kociba et al. (1978) study is the foundation for

quantitative risk assessments for dioxin made by the EPA, FDA, MDPH, and the

MDNR.

2. 3 The Pulp and Paper Industry as a Source of Dioxin Contamination
After ﬁnding dioxin in native ﬁsh collected downstream from a number of

pulp and paper mills, the EPA and the paper industry undertook a study in 1986 to

 

7 The N OAEL refers to the highest level of toxicant that results in no observable
adverse effects to exposed test organisms (MDNR, 1985).

3O

determine the occurrence and fate of dioxins and related furans in ﬁve bleached kraft
mills. Results of the EPA/Paper Industry Cooperative Dioxin Screening Study,
released in 1988, indicated that dioxin is formed as trace contaminants during the
bleaching of kraft hardwood and softwood pulps with chlorine and chlorine
derivatives. The study also indicated that dioxin was found in treated wastewater
efﬂuent from three of ﬁve of the kraft mills at levels ranging from 0.015 to 0.12 ppt
(USEPA, 1988a).

Kraft is both the Swedish and German word for strong. It refers to a
particular pulping process in which wood is separated into its individual ﬁbers by
cooking wood chips under pressure and elevated temperature in a digester with strong
alkali, Figure 2.3 (Smook, 1982). The lignin and non-ﬁbrous material are recovered
as black liquor at a 96 to 99.5 percent recovery rate. The remaining black liquor
becomes part of the mill efﬂuent and can negatively impact the downstream aquatic
environment (Bonsor et al., 1989). The efﬁciency and reliability of the black liquor
recovery system is directly related to all efﬂuent parameters except those related to
chlorinated organics such as dioxin (Bonsor et al., 1989).

The majority of the pulp is then bleached by chlorine and related chemicals
(Figure 2.3), and dried for sale or used on site to make paper. Traditional bleaching
processes cause up to 7 percent of the weight of the pulp to be discharged in the form
of a wide variety of compounds, including organochlorines. A minute amount of the
total organochlorine discharge consists of dioxins (Bonsor et al., 1989).

The Pulp and Paper Research Institute in Canada’s (PAPRICAN) 1988

31

laboratory studies concluded that a "boundary line" of dioxin formation depends upon
the proportion of chlorine and chlorine dioxide used and the amount of bleach
chemicals added (Fales, 1988). Non-chlorinated contaminants, especially in defoamer
made from recycled oil, were identiﬁed as a second major potential source of dioxin
(Fales, 1988). New defoamers without dioxin precursors are now on the market

making this a fairly negligible source of dioxin in current pulping operations.

 

Whitoliqua i
' i E

Filer

 

Pulp
ch08!

 

 

 

 

 

‘
Bleached
pulp
o a:

 

 

 

 

 

 

 

cut-w» ' Alkali Chlorine We Alkali Chm
dioxide” tom delirium m m
voam v 52mm "Omar _v Hm ., 5.919. v (2.9392
‘ Blcmryomuom ‘

 

 

 

Figure 2.3 Kraft pulp operation with a conventional bleach plant (USEPA, 1990a).
To reduce chlorine use the industry is examining a number of options. The

substitution of chlorine dioxide for molecular chlorine appears to be the most

promising approach for existing mills. A second option is the substitution of peroxide

as a ﬁnal polishing stage where added brightness is desired. Oxygen deligniﬁcation is

32

an option for new mills or existing mills with adequate chemical recovery capacity.
Other measures such as improved pitch control, brownstock washing, improved
process control for good uniformity, and kappa factor control8 are effective measures
which have generally been incorporated into current pulp mill processes (USEPA,
1990a).

The pulp and paper industry is examining several available and emerging
technologies to reduce dioxin formation in both the pulping and bleaching stages.
Technologies currently available for minimizing dioxin contamination through pulping
process changes include: extended deligniﬁcation, oxygen deligniﬁcation, polysulﬁde
cooking, and improved pulp washing. Current bleaching process technology changes
include: chlorine dioxide substitution, oxygen extraction, Monox-L substitution,
control of chemical dosage, improved mixing, split chlorine addition/pH control, and
monitoring of the chlorine multiple (USEPA, 1990a)9. Three wastewater treatment
technologies have shown some effectiveness in eliminating dioxin, these include
ultraﬁltration, chemically assisted clariﬁcation, and enhanced photooxidation,
however, they are not widely used.

The Michigan Department of Commerce (MDC) has been marketing the state

as an ideal site for pulp and paper investments (MDC, 1982). MDC maintains that

 

8 A kappa factor is a measure of lignin in pulp, according to a standard
laboratory procedure.

9 See USEPA 1990a for a comprehensive treatment of available and emerging
technology for decreasing and eliminating dioxin formation in pulp and paper
production.

33

Michigan can provide the required forest resources, freshwater, large land area and
accessibility to markets for the potential products. Bleached kraft pulp, sanitary paper
products, and printing/writing papers have been identiﬁed as the best use for the
forest resource (MDC, 1982).

Despite the current dioxin regulations pulp and paper mills continue to
investigate pulp and paper mill investment opportunities in Michigan. This is
exempliﬁed by James River Corporation’s recent analysis concerning the suitability
for siting a pulp and paper mill on Michigan’s Keweenaw Peninsula. Although,
James River ultimately decided against siting a mill on the Keweenaw, the potential
exists for future pulp and paper mill development in Michigan.

Because of Michigan’s attractiveness to pulp mill development and the ever '
present need of increased employment and tax opportunities it is critically important
that state regulations be objectively examined to ensure that the health, welfare and
safety of the people of Michigan is preserved. The use of the policy analysis model
developed in Chapter 4 is one tool that is available to assist state regulators in
determining what pollution control level will maximize social welfare and minimize
the costs to society. However, before the model is presented a brief examination of
the current institutional setting for dioxin policy is necessary. The following chapter
will address federal and state responses to dioxin contamination as well as illustrate

some of the inconsistencies within Michigan’s dioxin policy.

CHAPTER THREE: INSTITUTIONAL SETTING FOR DIOXIN POLICY

Policy is a course of action adopted by governments to make decisions and
inﬂuence the behavior of ﬁrms and individuals (Morris, 1979). Policies are
generally made in response to a perceived or real threat to society, such as exposure
to pollution. The policy making process is one marked by perpetual evolution; as
policy is developed and implemented it is continually assessed and reformulated. The
dynamic nature of society forces policy to be receptive to change. This constantly
changing nature of environmental policy is exempliﬁed by the recent passage of the
1990 Clean Air Act amendments which strengthened the 1977 amendments to the
1970 Clean Air Act.

Environmental policy is often differentiated by resource, (air, water, land)
with various agreements, laws and rules developed to address speciﬁc threats to the
quality of the resource. Pollutants that threaten human health or the aquatic
environment at relatively low concentrations are generally referred to as toxic
substances (Wentz, 1989). Three primary directives exist for restoring and protecting
the Great Lakes from toxic substances: (1) the Great Lakes Water Quality Agreement;
(2) the Federal Clean Water Act, and equivalent state laws; and (3) the Great Lakes
Toxic Substance Control Agreement. These agreements encompass international,
federal and state jurisdictions.

In the U.S., the Environmental Protection Agency and the Food and Drug

Administration (FDA) of the U.S. Department of Health and Human Services have

34

35

primary jurisdiction for enforcing international and federal policy for protecting the
population from exposure to toxic substances, such as dioxin, in surface water. Great
Lakes States also enforce international policy, in the form of the Great lakes Toxic

Substance Control Agreement, as well as implement state toxic substance regulations.

3.] International Response to Toxic Substances in the Great Lakes

The Great Lakes Water Quality Agreement (GLWQA), signed in 1972, revised
in 1978 and amended in 1987, is an international commitment signed by the U.S. and
Canada, in cooperation with Ontario and the eight Great Lakes states, " to restore and
maintain the chemical, physical and biological integrity of the waters of the Great
Lakes Basin ecosystem (IJC, 1989).” Article 2 of the agreement, commits Canada
and the U.S. to a policy requiring, that the discharge of toxic substances in toxic
amounts be prohibited and the discharge of any or all persistent toxic substances be
virtually eliminated (EPA, 1989).

The Great Lakes Toxic Substances Control Agreement, signed by the eight
Great Lakes Governors in 1986 and agreed to by Ontario and Quebec in 1988,
commits the states to actions congruent with the GLWQA. Speciﬁcally, Principle IV
states:

The signatory States commit to continue reducing toxics in the Great Lakes

Basin to the maximum extent possible. Such actions shall be consistent with

the Federal Clean Water Act goal of prohibiting the discharge of toxic

pollutants in toxic amounts, as well as the Great lakes Water Quality

Agreements’s aim to "virtually eliminate" the discharge of all persistent toxic

substances.

The governors further agreed that the permitting process is the "best means now

36

available” to regulatory agencies for controlling the release of toxic substances. The
Agreement also states that ”discharges, emissions or releases of toxic substances will
be controlled by a regulatory permit process in order to reduce or eliminate the
negative effects of toxics on human health and the environment. "

These international policies provide speciﬁc goals and objectives for
controlling the release of toxic substance. To achieve these goals agencies need to

implement and enforce federal and state protection policies.

3. 2 Federal Response to Dioxin Contamination

The Federal Clean Water Act of 1972 (33 U.S.C. 1251 et seq.) is the primary
water pollution control law in the United States. Section 101 refers to the elimination
of pollution discharges as a speciﬁc goal:

The objective of this Act is to restore and maintain the chemical, physical, and

biological integrity of the Nation’s waters. In order to achieve this objective,

it is hereby declared that, consistent with the provisions of this Act,

(1) it is the national goal that the discharge of pollutants into
navigable waters be eliminated by 1985 ;

(2) it is the national goal that wherever attainable, an interim goal
of water quality provides for the protection and propagation of ﬁsh,
shellﬁsh, and wildlife, and provides for recreation in and on the water
be achieved by July 1, 1983;

(3) it is the national policy that the discharge of toxic pollutants in
toxic amounts be prohibited;

In an attempt to achieve these rigorous goals, the act "combines two approaches to

water pollution control: a water quality-based approach and a technology-based

37
approach" (Goldfarb, 1988 p. 167).

Water quality standards provide the foundation for water quality-based
pollution control, which up until 1972 characterized U.S. water pollution control
policy. Water quality standard are deﬁned by Goldfarb Ias having two parts:

(1) a designation of the desired use for a given body of water, and

(2) the water quality criteria appropriate for that use.

The water quality criteria "are speciﬁc levels of water quality that, if not exceeded,
are expected to render a body of water suitable for its designated use. " In order to
control the discharge of particular substances from an industrial ﬁrm, water quality
standards are translated into water quality-based efﬂuent limitations. These efﬂuent
limitations are ”restrictions on quantities, rates, and concentrations in waste water
discharges measured at the discharger’s outfall pipe" (Goldfarb, 1988).

Although water pollution control policy was traditionally based on the water
quality-based approach, it has been generally deemed a failure due to the following
eight deﬁciencies identiﬁed by Goldfarb:

(1) There is not yet an adequate scientiﬁc basis for tying water quality
criteria to designated uses.

(2) Assigning wasteload allocations to discharges based on mathematical
models is still an uncertain enterprise because of the relatively primitive
nature of even the best advanced water quality models.

(3) Modeling and wasteload allocation difﬁculties are compounded by the
concept of ”mixing zones.” A mixing zone is an area around a
discharge point in which a discharger is permitted to mix its wastes
without liability for violating water quality standards.

(4) Even if waterbodies could be modeled precisely there would be thorny
problems of distributional equity in attempting to allocate wasteloads.

38

(5) The variability of water quality-based efﬂuent limitations is also a
major obstacle to enforcement.

(6) Because water quality—based efﬂuent limitations are so difﬁcult to set
and enforce, there are many waterbodies or segments of waterbodies
where state agencies have not established them, especially for toxic
pollutants.

(7) Many peOple consider the water quality-based approach to be morally
intolerable. It assumes that, to some extent,"pollution is the price of
progress. "

(8) The bioassay is a highly useful tool for determining the toxicity of
mixed waste streams, but it has limitations with regard to water
pollution control in general. Bioassays are insensitive to long-term
effects, bioaccumulative effects, and synergistic or antagonistic effects
of multiple discharges.

In 1972, Congress redirected U.S. water pollution control policy to include
uniform national technology-based efﬂuent limitations which are to be "progressively
tightened until a ’zero discharge’ goal is reached" (Goldfarb, 1988). The technology-
based approach is focused on preventing the causes of water pollution rather than
relying on a tolerable level of pollution. According to Goldfarb, the current CWA
debate ”now involves whether further water quality improvement is worth the cost of
installing expensive control technology, and whether these more sophisticated control
mechanisms are necessary to restore or maintain desired waterbody uses. "

While the move to water pollution policy is towards the prevention oriented
technology-based approach, until these limits are developed, water quality standards
will remain the primary state control mechanism for water pollution. The National

Pollutant Discharge Elimination System (NPDES) is the program in which states and

regional EPA ofﬁces translate efﬂuent limitations into enforceable permit conditions.

39
An NPDES permit contains three parts: efﬂuent limitations and monitoring

requirements, monitoring and reporting, and compliance schedules. In general, each
regulated pollutant in a permit has efﬂuent limitations expressed in terms of load and
concentration with each pollutant having a maximum daily discharge limitation and an
average monthly discharge limitation (MDNR, 1989c). For example, the Mead
Escanaba pulp mill’s NPDES permit requires a monthly average of 0.022 ppq dioxin,
however, it does not contain a daily maximum for dioxin. The permit also speciﬁes
that for monitoring purposes samples must be made twice monthly using a 24 hour
composite efﬂuent sample (MDNR, 1989c).

Reasserting the goals of both the 1972 CWA and the GLWQA, section 118 of
the 1987 CWA amendments states that the U.S. should "seek to attain the goals
embodied in the Great Lakes Water Quality Agreement of 1978 with particular
emphasis on goals related to toxic substances." In Section 118(c)(1)(E) of the CWA,
Congress speciﬁes the EPA as the lead agency to coordinate its actions with other
federal, state and local authorities to ensure their input and support in achieving the
objectives of the GLWQA (EPA, 1989a). Section 304(a) of the CWA requires the
U.S. EPA to establish criteria which enable states to adopt water quality-based
efﬂuent standards, anti-degradation policies”, and implementation procedures

necessary to achieve the goals of the CWA in the Great Lakes (EPA, 1989a).

 

‘0 The ”restore and maintain” language of section 101 of the CWA provides a
national goal that not only requires the clean up of polluted waters but also
forestalls the degradation (referred to as anti-degradation) of current high-
quality waters (Goldfarb, 1988 p. 174).

4O

Discharge under the NPDES must meet the more stringent of either water
quality-based efﬂuent standards or technology based limitations. Water quality-based
efﬂuent limits are derived by states to meet state water quality standards. The EPA is
responsible for developing treatment based standards (technology-based approach).
The agency is expected to have proposed best available, economically achievable
technology limits by 1993, with ﬁnal regulations by 1995 (USEPA, 1990b). These
efﬂuent guidelines will focus on process changes designed to prevent pollution from
initially occurring (U SEPA, 1990b).

The NPDES program requires all dischargers, industrial and municipal, to
obtain a discharge permit. The permit is "either a temporary privilege to use a
waterbody for waste disposal until improved pollution control technology is developed
or, where an ambient based variance is available, a warning that discharges that
disrupt the integrity of the water body are unacceptable" (Goldfarb, 1988, p. 208).
The program is administered by either an EPA regional ofﬁce or by a state that has
received EPA permission to issue permits. The CWA sets provisions for the U.S.
EPA to delegate a NPDES program to a state, provided it possess the resources and

statutory authority to implement it.

41
W In 1983, the EPA issued its "National Dioxin

Strategy," which provided a systematic framework for investigating the nature of
dioxin contamination throughout the U.S.ll This strategy was a ﬁrst attempt at
comprehensively collecting information to assess the risks posed by dioxin in the
environment and to subsequently minimize (manage) any risks to human health or the
environment.

The strategy consisted of seven tiers of activity focusing on known and
suspected sources of dioxin contamination (Table 3.1). The ﬁrst part of the strategy
dealt with production facilities of 2,4,5-T or its derivatives. The second part formed
the ”National Dioxin Study,” which included sampling air, water, soil and ﬁsh and
selected pesticide formulators, combustion sources, sites of commercial pesticide use,
and chemical manufacturing sites (Barnes, 1985).

Among the results from the National Dioxin Study, dioxin was identiﬁed in
native ﬁsh collected downstream from pulp and paper mills. The subsequent ﬁnding
of dioxin in bleached kraft pulp and paper mill wastewater sludge lead to the
development of a detailed process evaluation study conducted by the EPA and the
paper industry (EPA, 1988a). The ﬁndings of the cooperative study conﬁrmed that

bleach kraft pulp mills were a source of dioxin contamination of surface waters.

 

” See Van Strum and Merrell for a comprehensive overview of early dioxin
regulatory issues.

42

 

 

 

 

 

 

 

 

 

 

Table 3.1 An overview of the U.S. EPA National Dioxin Strategy12
Tier 1 2,4,5-Trichlorophenol (2,4,5, TCP) production sites
Tier 1A Waste sites associated with Tier 1 sites
Tier 2 2,4,5-TCP used as a precursor
Tier 2A Waste sites associated with Tier 2 sites
Tier 3 Formulators/Blenders of 2,4,5-TCP and derivatives
Tier 3A Waste sites associated with Tier 3 sites
Tier 4 Combustion Sources
Tier 5 Present! past use sites
Tier 6 Other manufacturing facilities where poor quality control might

result in formation of dioxin
Tier 7 "Background” sites

 

 

 

In response to these ﬁndings the EPA established an agency goal to eliminate

the presence of dioxin in discharges from pulp and paper mills to U.S. waters (EPA,

1988b). In support of this goal the EPA developed an interim strategy for the

regulation of dioxin discharges from pulp and paper mills. The strategy consists of :

(1)

(2)

(3)

aggressive action to fully implement or, where necessary, to develop
State water quality standards for dioxin applicable to water bodies
where mills using chlorine bleach processes are discharging;

collection of data on each of the 104 affected mills, including dioxin
levels in their pulp, efﬂuent and sludge;

detailed technical evaluation of wastewater treatment technologies
and/or in-process changes to reduce or eliminate the presence of dioxin
in wastewater discharges; and

 

‘2 Adapted from Barnes, 1985.

 

(4)

43

issuance of NPDES permits that regulate and require monitoring for
dioxin, examine efﬂuent toxicity and provide for monitoring to tighten
controls consistent with the ﬁnal strategy and requirements of the CWA
(EPA, 1988b).

In 1989, the EPA Region 5 Ofﬁce, which has jurisdiction over the Great

Lakes States, developed a detailed approach for the regulation of pulp and paper mill

dioxin discharges as a follow up to the interim strategy. This approach was proffered

to states as a framework for setting water quality and technology based limits in paper

mill efﬂuent. The major elements of the approach include:

(1)

(2)

(3)

Placement in all bleaching mill permits of Best Management Practices
requirements, dioxin control plans, and interim limits for dioxin
designed to allow ﬂexibility from mill to mill, yet require steady
progress toward the lowest possible levels of dioxin discharge. These
provisions would not take effect unless "trigger " by existing or future
conﬁrmation of dioxin at the mills.

Placement of water quality based efﬂuent limits for dioxin in permits
for all bleached kraft mills and all other bleaching mills where dioxin is
found. Compliance required by June 1992 if listed on the CWA 304 (1)
list. Non detection of dioxin with good quality assurance would be
deﬁned as compliance.

Monitoring requirements for wastewater, sludge and ﬁsh tissue
(USEPA, 1989b).

Independent of the National Dioxin Study, EPA undertook risk

assessment/management activities to assess the health risks associated with dioxin and -

develop a regulatory response (USEPA, 1985). Based on long-term animal studies

(Kociba et al., 1978; NTP, 1982a; NTP, 1982b), EPA concluded that dioxin should

be regarded as a probable human carcinogen (USEPA, 1985). In making this

determination the agency used a linearized multistage model to estimate human

carcinogenic risk from dioxin.

44

The EPA model assumed a linear dose response between dioxin and cancer,
suggesting that any exposure to dioxin potentially poses some level of risk of
contracting cancer. The model is exponential "approaching 100% at high doses, with
a shape at low doses that is described by a polynomial function" (Anderson et al.,
1990). This statistical calculation identiﬁes an upper bound below which one can be
95% certain the actual dose-response curve will fall. "The slope of this upper 95 %
bound of the dose-response curve is called the cancer potency factor and has units of
(mg/kg/day)" (Anderson et al., 1990).

Using female rat liver data from Kociba et al. (1978) and the linearized
multistage model, EPA developed an upper bound carcinogenic potency factor for
dioxin of 1.56 x 10’ (mg/kg/day). In 1984, the agency published an Ambient Water
Quality Criteria for dioxin of 0.013 ppq, under its CWA authority (USEPA 1984). In
developing this criteria, EPA assumed the carcinogenic potency factor of 1.56 x 105
(mg/kg/day), a one excess case of cancer per one million exposed population, a dose
of 2.88 x 10“" (mg/kg/day), a bioconcentration factor of 5,000, and a 6.5 g/day
consumption of ﬁsh by a 70 kilogram individual.

This criteria was offered as a model to assist states in the development of
water quality-based efﬂuent limitations for dioxin. While the EPA has primary
jurisdiction over protecting the quality of U.S. waters and is responsible for
developing water quality criteria, the FDA has responsibility for administering laws

that ensure the purity and safety of food, such as commercial ﬁsh.

45
W The FDA ﬁrst developed contaminant action levels in

the late sixties and early seventies to regulate interstate commerce of ﬁsh. These
action levels have since been used by a number of Great Lakes states to establish
sport ﬁsh consumption advisories (Clark et al. , 1987).

State agencies have jurisdiction over sport ﬁsheries. The FDA consumption
advisory is provided to state health departments as additional information that can be
used in determining state health advisories. It is the philosophy of the FDA that the
degree of control necessary to protect the sport ﬁsh consuming public "is best
determined on a local basis by local authorities" (Hile, 1985). An FDA advisory
provides states with several options including: the option to close a ﬁshery
completely, advise people to limit their consumption of ﬁsh from a ﬁshery, or take
actions entirely different from those described in the FDA’s advisory (Hile, 1985).

Starting in 1980, the FDA initiated bilateral scientiﬁc consultations with the
Canadian Health Protection Branch to share and evaluate the information available to
both countries concerning dioxin. A joint laboratory effort by the agencies was
undertaken to survey the extent of dioxin contamination of ﬁsh in the Great Lakes.

In deriving consumption advice, the FDA relied on a risk
assessment/ management approach that included the laboratory results of dioxin
residues in ﬁsh, ﬁsh consumption data, analytical variability and toxicity studies. An
extrapolation from animal to human using the same rodent data from the 2-year
chronic feeding study as the EPA, provided the basis of the risk based advice (Kociba

et al., 1978).

46

The FDA animal-to-human extrapolation of the no-observable-adverse effect
level (N OAEL) for dioxin exposure from the rodent data indicated an intake of 1
ng/kg body wt/day as the no effect level, or based on a 70 kg individual, a total daily
intake of 70 ng as the level in which no adverse effect is identiﬁed. Based on 36. 8
g/day consumption of a ﬁsh (which is signiﬁcantly higher than the EPA, MDNR and
MDPH 6.5 g/day consumption value) containing average dioxin residue levels of 25
ppt, the total daily intake of dioxin would be 0.92 ng or 13 picogram per kilogram
(pg/kg) body weight per day which is less than 1/70th of the no-effect level, less than
1/700th of the lowest-effect level, and less than 1/7000th of the carcinogenic level
(Cordel, 1981 p.386).

The use of the NOAEL combined with a safety factor in calculating an
acceptable daily intake for carcinogens assumes the existence of a threshold. At the
time the FDA derived these initial health risks, carcinogenesis was generally believed
to have a linear dose-response relationship and not a threshold. In response to
criticism over the initial calculations the FDA recalculated the dioxin cancer risk
using extrapolation models (MDPH, 1986). With this information the FDA concluded
that there did not appear to be cause for concern for ﬁsh distributed in interstate
commerce (FDA, 1981). Recent evidence now indicates that dioxin’s mechanism of
carcinogenesis does have a threshold (CanTox, 1989). The FDA currently basis their
health risk advice on the linearized multistage extrapolation model which is based on
the linear dose-response relationship.

Based on the extrapolation model, the FDA recommended advice to Great

47

Lakes States having a problem concerning consumption of dioxin contaminated ﬁsh by
anglers and consumers is as follows:
if the TCDD levels found in ﬁsh average less than 25 ppt, FDA believes that
there is little cause for concern. On the other hand, if the average values
exceed 50 ppt, the State should seriously consider more stringent methods to
limit the taking of ﬁsh from these areas. For those values between 25 and 50
ppt, sport ﬁshermen who generally consume ﬁsh only a few times a year,
should restrict their intake to no more than one meal a week. Permanent
residents of these areas who might consume the ﬁsh over the entire year,
should restrict their intake to no more than 1-2 times a month (FDA, 1981).
This advisory is similar to the EPA’s water quality criteria; they both provide
information to state agencies. The states are the front line forces for implementing
toxic substance policy. They have primary jurisdiction over their natural resources

and are generalljharged by their constitution to protect the health, welfare, and safety

of the public, which includes the protection of natural resources.

3. 3 Michigan Response to Dioxin Contamination

In Michigan, there are two bleached kraft mills discharging efﬂuent to Great
Lakes tributaries and one mill discharging to a municipal system. The Champion
Corporation located in Quinnesec, Michigan discharges to the Menominee River
which empties into Lake Michigan. The Mead Corporation located in Escanaba
discharges to the Escanaba River which is also tributary to Lake Michigan. The Scott
Paper Company discharges into the Muskegon municipal system. Dioxins and furans
have been detected at levels of concern in efﬂuent, sludge and ﬁsh captured adjacent
to the Champion and Mead mills (IJC, 1989). Analytical results of the U. S.

EPA/Paper Industry Cooperative Dioxin Study indicated Champion mill efﬂuent

48

contained 9.0 ppq of dioxin and that dioxin was not detected at 17 ppql3 for the
Mead efﬂuent (Staniec, 1989).

Michigan policy to protect human life from the potential adverse effects of
dioxin in surface waters was established in the late 19703, in response to the Dow
Chemical Company’s discharge of dioxin-laden efﬂuent to the Tittabawasee River.
Both the Michigan Department of Natural Resources (MDNR) and the Department of
Public Health (MDPH) developed independent regulatory responses to this

contamination.

W The MDNR is statutorily and constitutionally
charged with responsibility for the control of the discharge of pollutants into the air,

soils, and waters of the state, and the protection of natural resources of the state from
pollution, impairment or destruction. In response to dioxin contamination of the
Tittabawasee River from Dow Chemical Company, the MDNR revised Rule 57 of the
Michigan Water Quality Standards (Mich. Admin. Code. r. 323.105 7).

Rule 57 is considered a narrative water quality standard which contrasts to a
numerical rule which creates absolute values speciﬁed for a list of toxic substances.
As required by the Michigan Constitution, Rule 57 is an attempt at balancing natural

resource protection and the maintenance of a viable economy. The philosophy behind

 

‘3 The dioxin efﬂuent analysis for the Mead Escanaba mill indicated a 17.0 ppq
ND (non detect) which indicates that no dioxin was found above 17.0 ppq
(which was the lowest limit of the analysis). Dioxin may have been present
below 17.0 ppq.

49

the development of the rule is best characterized by the following quote from an
MDNR (1984) support document:

If we wish to continue to enjoy our current lifestyle, we must accept chemicals

as part of our daily lives, accept some level of risk associated with these

chemicals, and expect some additional cost of living associated with improved
treatment of wastes to remove chemicals. Most chemicals, when manufactured
or used under the appropriate conditions, can be controlled so that they
represent little risk of adverse impacts on human health or the environment.

The goal of the proposed regulation is to assure that discharge to toxic

substances is properly regulated and controlled.

The rule consists of two subrules (see Appendix B). Subrule (1) is a general
statement prohibiting levels of toxic substances in the waters of the state that are
harmful to human health and the environment.

Toxic substances shall not be present in the waters of the state at levels which

are or may become injurious to the public health, safety, or welfare; plant and

animal life; or the designated uses of those waters. Allowable levels of toxic
substances shall be determined by the commission [Natural Resources

Commission] using appropriate scientiﬁc data (Mich. Admin. Code.

r.323.105 7(1)).

Subrule (2) outlines the method of developing allowable toxicant levels in the
waters of the state applicable to point source discharges. Rule 57(2) guidelines detail
procedures for calculating levels of toxic substances necessary to protect aquatic life,
wildlife, and public health from threshold effects of toxic substances; and toxic
substance concentrations which provide an acceptable degree of protection to public
health from cancer (MDNR, 1984).

Rule 57(2) establishes a maximum concentration of a substance to protect

humans from adverse effects resulting from contact with or ingestion of surface

waters and from ingestion of ﬁsh taken from surface waters. The subrule places an

50

upper boundary of one excess case of cancer per 100,000 exposed population for
carcinogens not determined to cause cancer by a threshold mechanism.

In developing Rule 57 the MDNR convened an expert committee to advise
staff on appropriate methodology. The Rule 57 Advisory Committee used the
commonly accepted risks presented in Tables 3.2 and 3.3 to derive the Rule 57 risk

level.

TABLE 3.2 Risks of activities used in the development of Rule 57 (MDNR,

1984).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Everyday Risks Living in the Time to Accumulate a I in Average Annual Risk per Capitn Exltapolulcd In Risk/70 )r
United States l00,000 Risk of Death lifetime
Motor vehicle accident 15 days 2 x IO‘ 1.4 x 102
Falls 60 days 6 x 10’ 4.2 x 10’
Drowning 100 days 4 x 10’ 2.8 x 10’
Fires I30 days 3 x 10’ 2.0 x 10’
Firearms 360 days 1 x 10’ 7.0 x lO‘
Electrocution 20 months 5 x 10‘ 3.5 x 10‘
Tornados 200 months 6 x 10’ 4.0 x 10’
Floods 200 months 6 x 10’ 4.0 x 10’
Lightening 20 years 5 x 10’ 3.5 x 10’
Animal bite or sting 40 years 2 x 10’ 1.4 x 10’
OCCUPATIONAL RISKS
General
manufacturing 45 days 8 x 10’ 5.6 x 10’
trade 70 days 5 x l0’ 3.5 x IO’
service/government 35 days i x 10‘ 7.0 x 10’
transpon/publicutilitics 10 days 4 x 10‘ 3.0 x 10’
agriculture 150 hours 6 x 10‘ 4.0 x 102
construction 140 hours 6 x 10‘
mining/quarrying 90 hours 1 x 10’ 4.0 x 10’
Speciﬁc 7.0 x 102
coal mining (accidents) I40 hours 6 x 10‘
police duty 15 days 2 x 10‘ 3.0 x 10’
railroad employment 15 days 2 x 10‘ 1.4 x 10’
ﬁre lighting "0 hours . 8 x 10‘ L4 2: 10’

 

 

 

 

 

51

Table 3.3 One in a million cancer risks used in the development of Rule
57 (MDNR, 1984).

 

Source of Risk

 

Cosmic Rays one transcontinental round trip by air; living 1.5 months
in Colorado compared to New York; camping at 15,000
feet for 6 days compared to sea level

 

 

Other radiation 20 days of sea level natural background radiation; 2.5
months in masonry rather than wood building; 1/7 of a
chest X ray

Eating and drinking 40 diet soda (saccharin)

6 pounds of peanut butter (aﬂatoxin)

180 pints of milk (aﬂatoxin)

200 gallons of drinking water from Miami or
New Orleans

90 pounds of broiled steak (cancer risk only)

 

Smoking 2 cigarettes

 

 

 

 

The committee "felt that the risk associated with exposure to these chemicals in
ambient water should generally be below that of common everyday risks and
recommended that an estimated risk level of 10’ (1 in 100,000) be used as the upper
boundary on risk for establishing allowable levels of carcinogens in the waters of the
state applicable to point discharges" (MDNR, 1984).

The rule requires that a point source discharge not develop an estimated risk to
public health greater than 1 in 100,000 above background (the level of existing or
dioxin associated with "natural" sources in the environment) in the surface water after
mixing with an allowable receiving stream as speciﬁed by rule.

The DNR staff feels that the actual risk to the public health associated with

exposure to these chemicals [carcinogenic substances] in most surface water of

the state under these conditions, will be considerably less than 1 in 100,000

and will be well below that of common everyday risks since the background
rate of a person contacting [sic] cancer is 1 in 3 (MDNR, 1984).

52

The concentration, established by Rule 57, is used in calculating water quality
based efﬂuent limits to control the discharge of toxic substances in accordance with
the State’s National Pollutant Discharge Elimination System (NPDES) authority. The
Rule 57 concentration is calculated using the formula for estimating excess lifetime
cancer risk (risk = potency x close).

The MDNR derived a cancer potency factor for dioxin from the same female
rat liver data (Kociba et al., 1978) used by the EPA and the FDA. MDNR
justiﬁcation for the use of this data was, "because the study quality is unsurpassed and
because the data yielded the highest potency value (MDNR, 1989a)." Using the
multistage model (Global 79) to estimate the upper bounds of risk at low doses, the
MDNR derived the potency of dioxin of 1.51 x 105 mg/kg/day (MDNR, 1989a).
This potency value is then used in calculating a Rule 57(2) value of 1.4 x 105 ng/l

using the following formula (see Appendix C for calculations):

 

D x Wh
C = WC + (F x BCF)
where: C = Concentration of carcinogen (mg/1)
D = Dose” (6.62 x 10" ng/kg/day)
Wh = Weight based on a 70 kg individual"

 

‘4 MDNR, 1989a

‘5 The Rule 57 Advisory Committee recommended the 70 kg value since it has
been widely accepted in the scientiﬁc community for dealing with diversiﬁed

populations of people. The EPA uses this number in the development of water
quality criteria (MDNR, 1984).

53

WC = 0.01 l/day for surface water and 2 l/day for drinking
water16

F = Fish consumption17 (0.0065 kg/day)

BCF = Bioconcentration factor (51,600)

The Rule 57(2) value is then used to calculate a monthly average water quality based
efﬂuent limit (WQBEL) using the following relationship:

Vl/4R+E-RR

WQBEL = QE
where: V = Rule 57 (2) value (1.4 x 10'5 ng/l)
QE = Outfall design maximum ﬂow (mgd)
OR = 95% exceedance ﬂow for river‘8 (mgd)
CR = Ambient river dioxin concentration (zero assumed)

The efﬂuent limit formula is applied to individual industrial and municipal discharges

which require NPDES permits. The values used in the development of the efﬂuent

 

‘6 The 0.01 1/day value for surface waters not protected for drinking water is to
account for incidental exposure such as absorption through skin or ingestion of
small quantities while swimming (MDNR, 1984). The value of 2.0 liters for
surface waters protected as a drinking water source is recommended by the
EPA for establishing drinking water standards (MDNR, 1984).

‘7 The 0.0065 kg of ﬁsh per day is also used by the EPA. It is based on average
national ﬁsh consumption, including marine and shell ﬁsh. An estimate of
consumption of inland ﬁsh by Michigan anglers was calculated at 0.0040
kg/day, however, MDNR decided to use the larger EPA number "because it
was calculated on better data and was slightly more conservative" (MDNR,
1984).

‘3 "The 95 percent exceedance ﬂow is the ﬂow which is exceeded in the river 95

percent of the time" (MDNR, 1984).

54

limit will vary based on river ﬂow characteristics and maximum discharge rate for

each discharge source. The water quality based efﬂuent limit for the Mead Escanaba

mill is 0.022 ppq.
W The MDPH has the responsibility of protecting public

health and in that capacity to provide the public consumption advice related to sport-
caught ﬁsh. This is primarily a risk communication function rather than a regulatory
mandate. The MDPH provides health advisory information in a booklet which is
distributed to anglers who obtain a ﬁshing license. The agency provides the advice,
however, it is up to individual ﬁsherpersons to comply. The MDPH does not police
Michigan lakes and streams to ensure that ﬁsherpersons are complying with the
advice, however, they do monitor areas of contamination in an effort to maintain
timely advisories. Fish consumption advisories are based on tissue concentration of
contaminants in species and size classes of ﬁsh in relation to some trigger level
(Foran and VanderPloeg, 1989). When tissue concentrations excwd a health advisory
trigger, consumption advise ranging from restrict consumption for certain groups,
such as pregnant women and children, to do-not-eat is issued.

Although U.S. Food and Drug Administration or U.S. Environmental
Protection Agency enforcement guidelines exist for some contaminants, the MDPH
has historically established speciﬁc sport ﬁsh consumption advisories for Michigan
anglers. The MDPH uses a risk assessment/ management approach in developing

health advisory triggers. The MDPH’s selection of an "acceptable contaminant

55

concentration" is an evolving process which is guided by the following factors: ﬁsh
consumption patterns, contaminant level trends, vulnerable population, history, choice
of risk models and adoption of various assumptions used in mathematical
extrapolations (MDPH, 1986).

In 1986, the MDPH’s Center for Environmental Health Services selected 10
part per trillion (ppt) as a trigger level for establishing an advisory for dioxin
contaminated ﬁsh in the Tittabawasee River. The advisory, including similar
conservative assumptions, was latter extended to include other Great lakes tributaries.
The MDPH dioxin risk assessment, like that of the EPA, FDA, and MDNR, is based
on the Kociba et al. (1978) female rat liver data. The MDPH also used a multistage
model for carcinogenesis, because it is "generally regarded as the most biologically
defensible of the low dose extrapolation models (MDPH, 1986 p.6). "

The MDPH risk modeling process included the following conservative
exposure assumptions:

1. That the level of TCDD contamination would remain constant at the
present level for the lifetime of the ﬁsh consumers.

2. That persons consume uncooked skin-on ﬁllets for a lifetime and that
all of their ﬁsh come from the Tittabawasee River (MDPH, 1986 p. 10).

The uncooked skin-on ﬁllet signiﬁcantly increases the conservative nature of their
health advisory trigger. Stachiw et al. (1988) demonstrated that restructured carp
ﬁllets lost between 35 and 67 percent (with an average of 54.6 percent) of dioxin by
cooking the ﬁsh in a manner which allows the ﬁsh oils to drip away from the edible

portion of the ﬁsh. Dioxin may also be reduced by trimming the fatty areas of the

56
ﬁllet prior to cooking (MDPH, 1986).

The MDPH used the Center for Disease Control (CDC) risk ranges derived
for sport ﬁsh collected in the Tittabawasee River (Table 3.4) and associated doses in
deriving their health advisory trigger. The CDC ranges were based on a one excess
case of cancer per 100,000 exposed population and calculated using a cancer potency
factor of 3.57 x 10‘ mg/kg/day derived from a multistage model (MDPH, 1986).
Through their risk assessment process the MDPH concluded that "TCDD
concentrations at or below 10 ppt would result in an insigniﬁcant increased cancer

risk to the Michigan sport-ﬁshing community (MDPH, 1986 p. 10)."

Table 3.4 Evaluation of sport ﬁsh caught native to the Tittabawasee River.l

 

 

 

 

Size Range 2,3,7,8 TCDD Quantitation
Length Weight Number (parts per trillion)
Number (inches) (pounds) Tested Range Mean 95% C.I.
White Bass 9.6 - 13.8 0.4 - 1.1 13 5.7 -15.0 8.0 4.24< X (11.76
Crappie 7.9 - 10.4 0.2 - 0.6 9 2.8 - 4.5 3.9 3.02< )t <4.78
Walleye 18.1 - 27.6 2.0 - 6.9 14 2.5 - 14.0 5.28 3.36< X (6.64
Smallmouth
Bass 14.6 - 15.4 1.6 - 1.8 3 2.8 - 6.4 -
Pike 21.3 - 28.0 2.0 - 4.3 3 6.1 - 15.0 -
All species - - 262 - 6.76 5.12< x <3.4’
‘ MDPH, 1986

14 walleye, 1 composite of 3 crappie, 1 composite of 3 white bass, 3 pike and 3
smallmouth bass

Assuming the samples analyzed are representative of fish of these species in the
Tittabawasee River, the mean 2378 TCDD concentration in part per trillion would
fall within this confidence interval 95% of the time.

57

3. 4 Inconsistencies in Michigan ’s Dioxin Policy

The risk assessment/ management process is employed by a number of
regulatory agencies for a variety of regulatory programs. The purpose behind
individual regulatory programs inﬂuences the assumptions used in the risk
assessment/ management process. The Michigan Environmental Council has identiﬁed
three purposes for which risk assessments might be conducted (Bedford et al. , 1990):

1. 9193mm purposes--where the purpose of the assessment is to
develop controls on pollution sources that will W;

2. cgrrgtive purposes--where the purpose of the assessment is to
determine the appropriate response by either a responsible party or
government agencies to an existing environmental exposure; and

3. communication purposes--where the purpose is to provide potentially
exposed individuals with usable information concerning the hazards
associated with toxic chemicals.

A water quality based efﬂuent limit is an example of a preventative based
regulatory program, whereas, a health advisory trigger is communication based. The
efﬂuent limit is a level that is prescribed to a discharging ﬁrm, and in effect states,
"that at a certain level the toxic substance in your efﬂuent will provide no harm to
human health or the environment." Similarly, the health advisory trigger
communicates to the angler that ﬁsh below a certain level are safe to eat. Because
an efﬂuent limit is issued on a three year cycle, the limit must be conservative enough
to provide for risk and uncertainty over a three year period. In contrast to the
prudent efﬂuent limit, a health advisory trigger is based on the best currently

available data and can be readily adjusted to account for new information.

Both of these regulatory activities encompass a substantial degree of

58

uncertainty. Uncertainty exists among the epidemiology studies which have been used
as a basis for assuming dioxin is carcinogenic in humans, as well as in interpreting
which tumors to count and how to count them. Uncertainty exists in the "acceptable"
risk level that has been assumed to provide society with an "adequate" margin of
safety. There is uncertainty in the determination of how much ﬁsh an individual
actually eats. Risk managers are faced with uncertainty at nearly every stage of the
risk assessment/management process, yet it is still a valuable tool in protecting the
health, welfare, and safety of society as long as the purposes are well deﬁned and the
assumptions are justiﬁed.

Table 3.5 illustrates some speciﬁc differences between EPA, FDA, MDNR,
and MDPH risk assessments for dioxin. Each agency relies on a cancer potency
factor for dioxin which is based on one particular rat study, Kociba et al. (1978), and
all use a multistage model. However, they derive different cancer potency factors.
The cancer potency factor is the foundation for the risk assessment of dioxin. A
difference in magnitude as exhibited between the environmental agencies, EPA and
MDNR, and the public health agencies, FDA and MDPH, can create a signiﬁcant gap
in the safety margin provided by the risk management process. The average margin
of safety provided by the environmental agencies is 4.55 x 105 compared to the 3.0 x

10“ risk level of the public health agencies.

59

Table 3.5 State and federal agency dioxin potency derivations (MDPH, 1986).

 

Study
Used for Tumor Potency 10‘ Risk
Agency Extrapolation Type Model (mg/kg/day) (pg/person/day)
EPA Kociba Liver Multistage 1.56 x 105 4.5
Lung
Nasal
MDNR Kociba Liver Multistage 1.51 x 105 4.6
CDC/MDPH Kociba Liver Multistage 3.57 x 10‘ 20.0
FDA Kociba Liver Multistage 1.75 x 10‘ 40.0

 

One signiﬁcant difference between the MDPH and FDA values and the MDN R

and EPA values is the surface area species conversion factor used by MDNR and

EPA in their risk assessments. Both the EPA and MDNR assume different species

are equally sensitive to the effects of dioxin if they absorb the same dose per unit

body of surface area (Anderson et al., 1990). As a result, the EPA and MDN R

extrapolate the rat data on the basis of body surface area. The MDPH believes the

factor, which multiplies the risk by approximately 5.4, is difﬁcult to justify and basis

its extrapolation on a body weight scaling factor (MDPH, 1986). Another

discrepancy in the risk assessment of dioxin is EPA’s inclusion of nasal and lung

tumors which have the potential to over estimate cancer risk (MDPH, 1986).

60
While the program purposes of the MDNR efﬂuent limit and the MDPH health

advisory trigger are different, combined, they form a policy with the speciﬁc purpose
of protecting Michigan citizens from the adverse effects of dioxin exposure. To
answer the researchable question presented in Chapter 1, two initial inquiries must
ﬁrst be addressed. The ﬁrst question, is: does the level of dioxin permitted into the
environment, by water quality-based effluent limits, when bioaccumulated by
fish, provide the same margin of safety as the MDPH health advisory triggers?
The policy analysis model presented in the next chapter will provide insight into the
effectiveness of the Michigan policy as well as pose the second question which

concerns the costs of a policy.

CHAPTER FOUR: ECONONIIC CONCEPTS FOR EVALUATING
PERSISTENT TOXIC SUBSTANCE POLICY

Persistent toxic substance policy has often relied on regulations that are based
primarily on risk assessment and on the notion of some acceptable risk (e. g. one
excess case of cancer per one million exposed individuals or one excess case of
cancer per one hundred thousand exposed individuals) with little regard for the actual
costs involved. The science of economics provides the risk manager with a number
of tools which can add the dimension of personal values to the policy development
process. As risk managers debate the issue of "how safe is safe?" economic analysis
can provide a means of measuring an individual’s acceptance of risk. This chapter 5
will provide a discussion of how economics can be used to quantify levels of
acceptable risks and how they can be used to evaluate the human impacts of dioxin

policy decisions.

4.] Economics as a Means to Analyze Policy ‘
Economics is the study of how individuals and society make choices
(Nicholson, 1985). Individuals express preferences by placing value, often in

monetary terms,19 on objects of choice. Because individuals exhibit different tastes

 

‘9 Three basic value relationships which inﬂuence policy are identiﬁed in the
literature: values expressed by individual preferences; public preference value
which is expressed through social norms; and functional physical ecosystem
value. Individual preferences are assigned values in terms of willingness to
pay and willingness to accept compensation, public preferences are held values
which are aimed at inﬂuencing individual preferences, and physical ecosystem

61

62

and preferences, any object of choice, often referred to as a good (which implicitly
includes services as well), may have a number of assigned values. For example, the
choice to develop a pulp mill may be perceived as beneﬁcial to those who are seeking
employment. However, the development may seem detrimental to those who fearm
it will have a negative impact on the environment. Those individuals who are seeking
employment will place a higher value on the pulp mill and favor the decision to
develop the mill. The individuals who are fearful that a pulp mill will cause
environmental degradation will place a lower value on the pulp mill and oppose the

development decision.

W Economic values are determined by: (1) individual demand for
an object, (2).a seller’s ability to supply the object, and (3) the technical capacity to
produce the object (Talhelm, 1990). Demand expresses an individual’s willingness to
pay to buy an object, while supply expresses willingness to accept payment for selling
an object. Figure 4.1 illustrates this concept.

The demand function is an expression of an individual’s demand (which
includes factors inﬂuencing demand such as preferences, price, and budget) for a
good. It shows the maximum quantity individuals would be willing to buy at each
given price. The supply function expresses the seller’s preferences (which includes

the seller’s ability to produce goods from various inputs) for accepting compensation

 

value is a non-preference value that is measured in natural sciences (Pearce
and Turner, 1990).

63

for a good, and illustrates the minimum price at which a seller is willing to sell a
quantity of the good. Where demand and supply intersect (A) a particular price (b)

and quantity sold (c) is observed in the market (Randall, 1987).

 

 

 

Price
Supply
(willingness to accept
compensation)
b
Demand
(willingness to pay)
0 Quantity
Figure 4.1 Market diagram - supply and demand as determinants of

economic value.

In the neoclassical sense, the market deﬁnes a set of economic laws which
govern economic activity. In the market individuals are thought to maximize their
utility. The amount of personal utility yielded determines the economic value of
marketable commodities, unpriced environmental goods and services, or sympathy for

posterity (Pearce and Turner, 1990).

64
Maximum Social Wellbeing Welfare economics is based on the concept that

individual preferences are revealed by the choices they make, and "efﬁciency and
consistency of choice reﬂect rational behaviour" (Pearce and Turner, 1990). Pearce
and Turner state that "’the basic theorem of welfare economics’ seeks to legitimise
rational behaviour as being socially desirable and also to justify some government
intervention to improve the conditions under which individuals make choices.“ At
the heart of welfare economics is the theory of social welfare maximization at a
minimum cost. Randall (1987) deﬁnes maximum social well being as a "condition in
which society is as well off as it can possibly be, given its resource base, its
productive technology, and the tastes and preferences of its members."

The Pareto criterion is used as a measure of maximum social wellbeing. The
criterion, named for the Italian economist Vilfredo Pareto, refers to a Pareto optimal
solution as one in which no one individual can be made better off without necessarily
making some other individual worse off. The term Pareto efﬁciency is often used to
describe an allocation as "efﬁcient if conditions cannot be made unambiguously
better" (Nicholson, 1985).

The materials balance approach is one market model of environmental

 

2° There is much debate among economists concerning the actual mechanism of
individual human behavior. Neoclassical economists accept the ’rational and
egotistic person’ as the model of human behavior. Other economists support
the humanistic paradigm which rejects the ’rational person’ and accepts a
behavioral psychology approach emphasizing a "hierarchy of needs in place of
a ﬂat plane of suitable wants" (Pearce and Turner, 1990).

65

management adopted by economists to measure Pareto efﬁciency.21 This approach
assumes an economic optimum (Pareto efﬁcient) level of pollution can be deﬁned,
given certain assumptions. The level of pollution at which the costs of abating the
pollution are equal to the costs of the damage caused by the pollution, is often
referred to as the "economic optimum level of pollution" (Freeman et al., 1973;
Randall, 1987; Pearce and Turner, 1990). However, Pearce and Turner point out
that due to deﬁciencies in data the optimum situation is not a practical policy
objective. They state that "instead society sets ’acceptable’ levels of ambient
environmental quality, and policy instruments [such as efﬂuent limits and health
advisory triggers] are directed at these standards. The analytical task is to seek out

the least-cost policy package sufﬁcient to meet acceptable ambient quality standards."

4. 2 Using the Concept of Pareto Optimal Solution to Evaluate Policy

A variety of criteria have been developed to identify the "least cost policy
package." Bro et al. (1987) suggest that a rational response to controlling public
exposure to toxic substances is to consider the magnitude of the threat relative to the

costs and beneﬁts of pursuing corrective action or otherwise avoiding risks. The

 

2‘ The property rights approach is another widely used market model.
According to Pearce and Turner, (1990 p. 17) in the market model,
"environmental pollution is a form of market failure, usually because of the
over-exploitation of resources held in common property or not owned at all.
The market fails therefore when property rights are inadequately speciﬁed or
are not controlled by those who can beneﬁt personally by putting the resources
to their most highly valued use."

66

costs of abating dioxin contamination are primarily the private costs a ﬁrm incurs
through process changes. These costs are generally capital costs associated with
adopting new abatement process equipment, operational costs associated with
changing a process requirement such as adding a speciﬁc chemical to the process, or
a combination of both. The beneﬁts of dioxin abatement are primarily social in
nature, realized in terms of decreased risk to society from adverse effects associated
with swimming in, or drinking, or eating ﬁsh from surface waters.

In general, economists view zero discharge as a near impossible pollution
abatement goal. They argue that the total prohibition of emissions would eliminate
the production of a signiﬁcant amount of necessary goods as well as luxuries (Dewees
et a1, 1975 p. 2). They criticize policy goals for not accounting for total or marginal
beneﬁts and costs. Randall (1987 p.372) has suggested that the 1972 Clean Water
Act, with its target of zero discharge of pollutants into the nation’s waterways by
1985, is an extreme example of an "economically unrealistic" and inﬂexible goal for
water pollution abatement.

Following traditional abatement theory, the primary question addressed by
environmental economists then becomes not whether to allow pollution but how much
pollution to allow (Dewees et al., 1975 p.2). In a society that must meet its needs
from limited resources, the answer lies in the distribution of those resources among
competing needs.

The search for the optimum pollution level is often deﬁned in terms of costs or

beneﬁts. Dewees et al. (1975) list the following objectives for environmental policies

67

that can be used to evaluate alternative policy choices: Pareto efﬁciency, abatement
efﬁciency, administrative efﬁciency, promotion of technological progress, and the
distribution of control costs among polluters.

The concept of maximizing social welfare refers to achieving a pollution
control level such that any further control would impose abatement costs greater than
savings in pollution damage or welfare beneﬁts that would result (Dewees et al, 1975
p.16). Pareto efﬁciency is achieved at the point at which no person could be made
better off by more or less investment in pollution control without making others worse
off.

According to Dewees et al. (1975) the empirical determination of optimal
pollution control levels is difﬁcult to achieve and the search for abatement efﬁciency
frequently replaces the search for social efﬁciency. Abatement efﬁciency takes the
environmental quality standard or goal as given and examines only the cost of meeting
that goal.

Administrative efﬁciency is the search to achieve a given level of control at
the least possible administrative cost. This objective is sometimes viewed as striking
a balance between the costs of abatement, the beneﬁts of reduced emissions, and the
costs of administering a program (Dewees et al., 1975 p. 16).

The promotion of technological progress is a fourth objective. If control costs
can be reduced by new technology then it is important that policies foster more rapid
technological progress. A ﬁnal objective deals with the distribution of control costs

among polluters and of the beneﬁts received by those exposed to the effects of

68

pollution or its control. Abatement efﬁciency stresses the minimization of total costs.
Depending on speciﬁc operational conﬁgurations, procedures and costs, under this
criteria, ﬁrms producing similar products might suffer extreme variations in
abatement costs or percentages of abatement.

Dewees et al. (1975) suggest that one problem implicit in numerous
environmental policy considerations, but not explicitly addressed in their criteria, is
that of uncertainty. They point out that the information about costs of pollution
control or consequences of continued emissions is not well documented and suggest
that policy must be made without a deﬁnite understanding of the comparative
consequences for human health or economic growth of alternative policies (Dewees et
al, 1975 p. 17). The policy analysis model developed in Chapter 5 addresses this
information deﬁcit by incorporating the costs associated with policy decisions into the
policy decision making process.

The policy analysis model is built upon the fundamental pollution control
model which determines a Pareto optimal solution (Freeman et a1. 1973). Although
in reality an actual optimal solution is impossible, a theoretical optimal solution can
provide valuable information concerning the Pareto efﬁciency of a policy. The
pollution control model as described by Freeman et al. (1973) encompasses the
materials balance approach to environmental management. This fundamental beneﬁt-
cost analytic approach is a traditional choice among economists in determining the
economically efﬁcient level of persistent toxic substances (Moore, 1984).

The concept of zero discharge, as advocated by the U.S. Clean Water Act and

69

the Great Lakes Water Quality Agreement, is an example of setting an acceptable
level of ambient environmental quality. Rather than a policy endpoint, zero discharge
is a goal designed to promote technological progress. Pearce and Turner (1990)
suggest that zero discharge can only be a goal, rather than an actual endpoint, stating
that the laws of thermodynamics suggest that a nonpolluting product can not exist.22
They state that in order to achieve zero pollution there would have to be zero
economic activity. Therefore, in the absence of zero discharge an optimal level of
pollution is sought to maximize social wellbeing.

The pollution control model is often referred to as a cost minimization
model23 because economic theory considers social welfare to be maximized when the
total cost of pollution is minimized (Dewees et al., 1975, Freeman et al., 1973).
Figure 4.2 illustrates this concept. The curve sloping upward to the left from Q,
represents the total cost of decreasing human risks from a persistent toxic substance
such as dioxin, and is labeled "abatement costs." Abatement costs are those costs that
a ﬁrm incurs to control the release of a regulated toxic substance. The curve sloping
upward to the right from Q, represents the total cost of increasing human risks from

dioxin, and is labeled "damage costs." Damage costs are the costs associated with

"adverse impacts on human health, outdoor recreation and aesthetic opportunities,

 

22 The second law of thermodynamics implies that "any system and its
surroundings as whole spontaneously tend towards increasing randomness or
disorder" (Miller, 1982).

23 It is this concept of cost minimization, whether it is referred to as the pollution
control model or the cost minimization model, that is the foundation for the
policy analysis model developed in this paper.

7O

ecosystem productivity, and industrial productivity" (Moore, 1984).

The total cost of pollution is the sum of the damage and abatement costs,
shown by the heavy line. Q2 indicates the minimum of this total cost curve and is
labeled the optimum level of pollution (Freeman et al., 1973; Randall, 1987; Pearce
and Turner, 1990). In this paper optimality is used as a guide in determining the
efﬁciency of existing policy, illustrating a range of values and not one speciﬁed

"economically optimal" level of pollution.

. total cost of pollution

 

 

concentration of dioxin

Figure 4.2 The optimum level of pollution.

71

This optimum level of pollution, Q, is not static. A technological innovation
could be discovered that could drastically increase abatement efﬁciency, shifting the
abatement cost curve to the left thereby indicating a new optimal level of pollution,
Q, (Figure 4.3). Similarly, if new epidemiological data indicated that a toxic
substance which was once thought to be characterized by a non threshold mechanism
was found to actual have a threshold, the damage curve would shift to the right and a

new optimal level of pollution would be identiﬁed, Q, (Figure 4.4).

 

 

 

 

concentration of dioxin

Figure 4.3 The optimal level of pollution after technological innovation.

72

 

 

 

Figure 4.4 The optimal level of pollution after new epidemiological evidence
indicates a toxic substance characterized by a threshold.

4. 3 Estimating Damage Caused By Persistent Toxic Substances

Depending on the concentration in efﬂuent discharge, persistent toxic
substances can cause substantial damage to the environment. If released in large
enough quantity and high enough concentration, toxic substances may prove to be
acutely toxic resulting in immediate severe damage to aquatic organisms. However,
if a toxic substance, such as dioxin, is released at a very low concentration the

environmental damage is less certain.

Due to dioxin’s persistent nature small amounts tend to accumulate in the

environment. If the substance is taken up by an organism it will bioaccumulate in its

tissue. As other organisms prey upon the contaminated organism, the dioxin

undergoes biomagniﬁcation as it is cycled up the food chain. As a result, the release

of a minute amount of dioxin at a level below any health concern can be magniﬁed to

a quantity above levels deemed to be safe, such as the MDPH’s 10 ppt health

advisory trigger.

Because deriving abatement costs is a relatively straight forward exercise

which consists of valuing the application of a particular pollution control technology,

the critical component of a useful policy analysis model lies in deriving damage cost.

Hufschmidt et al. (1989) provide a simpliﬁed schematic for understanding the steps in

deriving a point estimate of the damage cost caused by the discharge of a toxic

substance (Figure 4.5).

 

 

Pulp Mill ,,.__..

 

 

Figure 4.5

 

Efﬂuent
Discharge
ol

Dioxin
(ppq)

 

 

 

 

 

 

 

Dioxin
(one)

in
Surface
Water

 

 

Cancer

Diox n in
Humans

Dioxin
(opt)
in
Fish

 

 

Individual
Willingness
to Pay

to Reduce

‘ Risk

of Death
from
Contracting
Cancer

 

 

Society
Willingness
to Pay

to Reduce
Each

. Member's

Rlak

 

 

 

 

 

Estimation of social damage due to the release of emissions from an

economic activity (Hufschmidt et a1, 1989).

 

74

The economic activity in this study has been identiﬁed as pulp production
which contributes an efﬂuent discharge of dioxin into surface water. The primary
chemical characteristics that affect the fate and transport of dioxin in the environment
are persistence and bioaccumulation. The actual health effects on individuals and
aquatic organisms are uncertain, however, regulatory consensus has identiﬁed dioxin
as carcinogenic to mammals.

In addressing the researchable question the ﬁrst inquiry (described in Chapter
3) to be addressed by the policy analysis model is: does the level of dioxin permitted
into the environment, by water quality-based efﬂuent limits, when bioaccumulated by
ﬁsh, provide the same margin of safety as the MDPH health advisory triggers?
However, the answer to this inquiry presents only one half of the answer to the
researchable question. Assuming that over a period of time some level of dioxin (to
be determined by a regulatory agency) will cause cancer in exposed organisms, to
answer the question completely and fully evaluate Michigan’s dioxin policy a second
inquiry must be addressed by the policy model: what is the value of the damage

caused when organisms are exposed to dioxin in the environment?

Willingness to Accept Values fgr Wildlife Damage The majority of environmental

goods and services fall outside the boundaries of traditional markets. Angling is an
example of such an environmental good. Anglers do not normally pay anyone for the

privilege to extract wild trout from streams, nor pay for the actual ﬁsh (Talhelm,

75

1990).“ Talhelm points out that "the private market does not make available
angling for wild trout at a going market price." While trout ﬁshing may not have
speciﬁc market costs it certainly does posses some value or anglers would not pursue
the activity. There are also transportation and other costs which are not normally
accounted for in the market. In the absence of deﬁned markets for environmental
goods and services, economists have relied on non-market valuation techniques to
place values on recreation, the existence of wildlife, aesthetics, and other
environmental amenities.

The three primary tools for estimating non-market values are: (1) the travel
cost and (2) hedonic valuation methods which derive willingness to pay and
willingness to accept from "careful observations of how users exchange the cost of
use for the amount of use, " and (3) contingent valuation which "carefully asks pe0p1e
how much they would pay or accept if they had to, as if there were a market"
(Talhelm, 1990).

A signiﬁcant difference between contingent valuation and market valuation is
the resulting willingness to pay and willingness to accept values. In deﬁned markets
willingness to pay always equals willingness to accept, whereas, in contingent

valuation an individual’s willingness to accept may be substantially greater than

 

2‘ The term "pay" here refers to transactions in the market for the cost of ﬁshing
on state water bodies. It does not refer to the regulatory costs of obtaining a
permit which provides permission to capture ﬁsh in state waterbodies.

76
willingness to pay.”

One simpliﬁed attempt to value the damage dioxin exposure has on wildlife is
to ask people what they would be willing to pay to ensure the existence of the
quantity of each species currently living in an affected area; or, what they would be
willing to accept for the loss of each individual animal. Given that dioxin is
carcinogenic to various species, one assumption a willingness to pay/accept study
could hold would be that a speciﬁc amount of dioxin in a river over a period of time
would cause death to those species currently existing in the ecosystem. A census of
the species and the amount of individual animals would be taken and the values
provided by the respondents would be multiplied by the quantity of each species.
Talhelm suggests that a corrective factor of four be applied when using willingness to
pay to provide a value comparable to one derived using willingness to accept.

The following is an example of how the willingness to pay/accept concept
could be used to provide an estimate of wildlife damage from dioxin in the Escanaba
River. A simpliﬁed estimate of ﬁsh eating birds and mammals existing near the
Mead Corporation’s Escanaba pulp mill, on the Escanaba River, is provided in Table
4.1. Similarly, an estimate of ﬁsh living in the Escanaba River near the Mead pulp

mill is provided in Table 4.2

 

2’ Economists have found that when people are randomly asked to sell a
possession their willingness to accept cash for the item is about four times
greater than their willingness to pay for the same item (Talhelm, 1990).

77

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table 4.1 Gross estimate of ﬁsh eating birds and mammals near the
Escanaba pulp mill on the Escanaba River (Aartila, 1990).
Species Estimated Quantity
Brown Bear 1
Raccoon 10
Muskrat 15
Opossum 10
Great Blue Heron 12
Green Heron 6
Eagles 8
Herring/Ring Billed Gulls 9,000
Mergansers 50
Table 4.2 Gross estimate of the quantity of ﬁsh in the Escanaba River
downstream from the Mead Escanaba pulp mill (Wheelan,
1990).
Species Estimated Estimated Weight
Quantity (lbs)
Smallmouth 85 7.5
Rock Bass 140 7.5
Walleye 18 15.9
Northern Pike 14 4.7
Bluegill 31 0.7
Yellow Perch 41 1.0
Forage ﬁsh 3,031 57.0

 

 

 

 

In his 1990 study, Recommended Values for Computing Fair Restitution to the

Citizens of Minnesota for Fish and Wildlife Illegally Killed, Injured or Possessed,

 

78

Talhelm applied contingent valuation techniques to derive willingness to accept values
for ﬁsh and wildlife. Applying these values to the estimated numbers of ﬁsh, birds,
and mammals provides a gross assessment of the value of the estimated ﬁsh and
wildlife populations in or near the Escanaba River in the vicinity of the Mead pulp
mill (see Table 4.3).

This $102,519 value is an estimate provided merely to illustrate the
complexities involved in valuing ﬁsh and wildlife populations. This example clearly
neglects to account for the persistent nature of dioxin and the potential for future
harm to wildlife in the area. It fails to account for transport of the compound to other
areas within the Great Lakes Basin and any resulting negative impacts. The
population count is merely an educated guess and by no means an accurate census.
The value in this example, although certainly underestimating wildlife damage, does
provide a comparison for potential human damage. To provide a more accurate
assessment of wildlife damage values from dioxin an actual willingness to pay/accept

study would need to be undertaken in the affected area.

79

Table 4.3 Gross estimated willingness to accept values for ﬁsh and wildlife
in the vicinity of the Mead Escanaba pulp mill.”5

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Species Quantity Total Pounds Base Value Total Value
(5) (S)

Brown Bear 1 - 400.00 400.00

Raccoon 10 - 30.00 300.00

Muskrat 15 - 30.00 450.00

Opossum 10 - 10.00 100.00

Great Blue 12 - 100.00 1,200.00

Heron

Green Heron 6 - 100.00 600.00

Eagles 8 - 500.00 4,000.00

Herring! - 10.00 90,000.00

Ring Billed 9,000

Gulls

Mergansers 50 - 40.00 2,000.00

Smallmouth 85 7.5 20.00 1,700.00

(Black Baas)

Rock Bass 140 7.5 3.00 420.00

Walleye 18 15.9 30.00 $40.00

Northern 14 4.7 30.00 420.00

Pike

Bluegills 31 0.7 5.00 ' 155.00

Yellow 41 1.0 5 .00 205 .00

Perch

Forage 3,031 57.0 .50/lb 29.00

ﬁsh

Total Value 102,519.00
———4——— — ===l=_= r

 

 

 

 

Willingness tg Pay Vﬂugs fgr Reducing ﬁsk gt path Economists point out that "in

a world of scarcity difﬁcult decisions concerning tradeoffs between health and other

desirables are unavoidable," and suggest that "essential to efﬁcient provision of

 

2‘ Base values taken from Talhelm (1990), wildlife population estimates provided
by Aartila (1990) and ﬁsh population estimates provided by Wheelan (1990).

80

safety, and the environment is correct valuation of risks to human health" (Berger et
al., 1984). To evaluate policies that reduce the risks individuals experience in their
daily lives economists must ﬁrst determine a "value of life." Blomquist (1979)
suggests that the "value of life" is’ based on "changing the probability of survival by a
small amount." Economists have ad0pted the "willingness to pay" framework in their
attempt to place value on life and risk reduction (Blomquist, 1979; Smith and
Desvousges, 1987; and Fisher et al., 1989).

Economists are quick to point out that although the risk of death is one
(everybody will die) there are activities which, if undertaken, have been statistically
shown to provide a greater chance of non natural death than other activities. These
are the types of statistics the MDNR Rule 57 committee used in their determination of
1 in 100,000 acceptable risk level (refer back to Tables 3.2 and 3.3). For example,
based on a 70 year working lifetime, MDNR Table 3.2 indicates a farmer has a 4 in
100 chance of dying on the job, while a government worker has 7 in 100 chance of
dying on the job in the same period (MDNR, 1984).

According to Fisher et al. (1989), "a willingness-to-pay estimate values the
change in well-being that would result from changing the risk of death; it is measured
by how much of other goods and services a person is willing to give up to get that
reduction in the risk of death." Because risk is not directly traded in traditional
markets, economists rely on willingness to pay studies to place value on risk. Three
techniques generally used are: (1) wage-risk studies, (2) contingent market studies,

and (3) consumer market studies. Fisher et al. (1989) deﬁne the techniques as

81
follows:

Wage-risk studies estimate the wage premium associated with greater risks of

death on the job. The contingent valuation approach poses a hypothetical

market situation to survey respondents who are asked about their willingness to
pay for alternative levels of safety. [And] consumer market studies examine
the observable tradeoffs people make between risks and beneﬁts in their
consumption decisions.

These willingness to pay studies focus primarily on avoidable or voluntary
risks. Since voluntary risks provide options, whether or not to take the more risky
job for example, values can be readily obtained through wage-risk or contingent
valuation studies. Individual willingness-to-pay values can be summed in an effort to
determine what a group would be willing to pay for reducing each member’s risk by a
small amount. The value a group is willing to pay to reduce a members risk is
referred to as the value of a statistical life (Fisher et al., 1989).”

The following example using data from Delta County28 will illustrate this
concept. If each of the 8,242 ﬁshing license holders in Delta County were willing to
pay $20 for a reduction in risk from 3 deaths per 8,242 to 1 death per 8,242, the total
willingness to pay would be $164,840 ($20 x 8,242 people) and the value per
statistical life would be $82,420 (with two lives saved - the value for each life is
$82,420, which when summed equals the total willingness to pay: $164,840).

The concept of reducing the risk of death from avoidable behavior can be

 

’7 For clariﬁcation purposes, the value of a statistical life is the value a group is
willing to pay to reduce a members risk. Both of the bolded terms refer to
social damage and will be used interchangeably.

2‘ The Mead pulp mill in Escanaba is located in Delta County, Michigan.

82

extended, using the fundamental risk formula: cancer risk = potency x dose, to
determine the damage costs associated with human exposure to dioxin. First, a dose
is derived by assuming a given population (for this research the Delta County licensed
ﬁshing population will be used) is exposed to a quantity of toxic substance, through
eating contaminated sport ﬁsh.

Second, multiplying this dose by a cancer potency factor provides the level of
risk in which the Delta County licensed ﬁshing population may contract cancer. This
risk will be directly related to the amount of toxic substance released in the
environment. As more toxic substance is released ﬁsh will bioaccumulate it, thereby
increasing the dose. If individuals continue to eat the ﬁsh, their risk of contracting
cancer will similarly increase.

The third and ﬁnal step is to apply a value of a statistical life to the risk of
contracting cancer. This results in a damage cost directly related to risk: as risk
increases the value of a statistical life, otherwise known as the societal willingness to
pay to reduce risk, increases.

The concept of willingness to pay for reducing risk of death is the fundamental
component in determining damage costs in the policy analysis model. In Chapter 5,
within the context of the Michigan dioxin regulatory environment, this concept of
damage cost will be added to abatement costs in a cost minimization model in order to

illustrate the design and application of the policy analysis model.

CHAPTER 5: DESIGN AND APPLICATION OF THE POLICY
ANALYSIS MODEL

The policy analysis model employs the concept of achieving a Pareto optimal
solution through the use of a cost minimization model to evaluate policy. The model
is based on the ﬂow of a toxic substance from a point source into the environment.
The key components of the model include: (1) the resulting damage costs from the
release and (2) the abatement costs associated with the technology employed to
minimize the release. These costs are summed and provide a total cost of pollution
for any particular quantity of toxic substance release. The minimum total cost
indicates the optimum level of pollution (in the Pareto sense), see Figure 4.2. The
model is developed using a Lotus 123 spreadsheet. The development and application
of the model will be illustrated using Michigan’s dual agency approach to dioxin
regulation, the Mead Corporation’s pulp and paper mill in Escanaba, Michigan, and

the Delta County Michigan registered ﬁshing population.

5.1 Study Site

The Escanaba River, located in Michigan’s Upper Peninsula (Figure 5.1), is
currently the only identiﬁed stream with ﬁsh exceeding the MDPH health advisory
trigger of 10 ppt for dioxin. Table 5.1 illustrates the results of a Michigan
Department of Public Health ﬁsh monitoring program on the Escanaba River. The

number of ﬁsh sampled from the Escanaba River on October 4, 1989 with dioxin

concentrations are listed; a positive sign indicates a concentration above the 10 ppt

83

 

 

O Escanaba g
l/ (‘r 08
\
lo ““
/ 0% 3 It“
/
I

 

 

 

 

Figure 5.1

Escanaba and the Escanaba River.

85

Table 5.1 Dioxin concentration in ﬁsh collected from the Escanaba River on
October 4, 198829

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Species Length Weight Dioxin M DPH
(inches) (lbs.) concentration health advisory
(ppt) trigger
Northern Pike 29.1 6.1 3.31 -
Northern Pike 29.9 6.3 23.4 4-
Northern Pike 30.1 5.7 10.2 +
Northern Pike - 5.7 12.0 .' 4-
Northern Pike 27.6 5.0 13.6 +
Northern Pike 28.5 5.2 24.4 +
Northern Pike 26.0 4.0 7.15 -
Northern Pike 26.2 3.7 4.9 -
Northern Pike 26.4 4.6 9.81 -
Northern Pike 24.0 3.3 ' 2.86 -
Northern Pike 22.4 2.5 6.62 -
White Sucker 18.5 3.7 8.44 -
White Sucker 18.1 3.0 8.29 —
White Sucker 21.9 4.4 17.6 +
White Sucker 20.7 3.7 9.06 -
White Sucker 19.1 3.0 13.4 +
White Sucker 18.9 2.9 15.3 +
White Sucker 19.5 3.1 12.4 +
White Sucker 20.5 3.9 8.8 -
White Sucker 19.7 3.5 8.55 -
White Sucker 22.0 4.1 9.61 -
Smallmouth Bass 15.7 2.0 5.51
Average dioxin 10.69 +
concentration

 

 

 

 

 

 

 

29 MDPH, 1989

86
MDPH health advisory trigger.

Of the ﬁsh sampled in the Escanaba River, 43 percent Were above the current
MDPH health advisory trigger for dioxin. The source of dioxin has been traced to
the wastewater efﬂuent from the Mead Corporation pulp and paper mill in Escanaba,
located near the mouth of the Escanaba River in Delta County (Figure 5.1). Mead’s
1990 efﬂuent dioxin levels were in the 7-8 ppq range which is signiﬁcantly above the
MDNR’s water quality based efﬂuent limit of 0.022 ppq.

A MDNR survey indicated that the Escanaba River supplies a good ﬁshery for
bass, northern pike, smelt, and walleye both at the Mead site and downstream (Long,
1973). Above the Mead site the Escanaba River provides a trout ﬁshery. There are
two dams at the site which virtually cut off the migration of ﬁsh upstream from the
site (Long, 1973).

The Delta County population was 38,800 in 1985 (Verway, 1987). A review
of ﬁshing license data from 1986-1989 indicates a year round average ﬁshing
population of 8,242 based on resident and senior ﬁshing licenses only (Appendix D).
This number gives an indication of the anglers in Delta County. For the purposes of
this research it is assumed that these anglers ﬁsh in the Escanaba River at or below
the Mead pulp mill and consume their ﬁsh. This assumption does not take into
consideration that a number of anglers release their catch or share their catch with
friends and family. Non-resident and daily permits were not included in the average
ﬁshing population because they represent a limited exposure for the angler to dioxin

in ﬁsh.

87

River access is a factor which contributes to the conservative nature of this
value. The only access to the stretch of river were dioxin contaminated ﬁsh have
been identiﬁed is by way of Mead property. Although access to the public is
available, it is not generally taken advantage of. As a result, it is highly unlikely that
8,242 Delta County anglers will ﬁsh the stretch of river with contaminated ﬁsh and
catch a contaminated ﬁsh. While this value may be conservative, it provides a value
to illustrate potential cancer risk. It can certainly be adjusted to reﬂect a number of
ﬁshing circumstances.

The Mead mill is integrated; containing both pulp and
paper production processes. The mill produces book publishing paper. For the last
ten years the mill has been producing 60% hardwood and 40 % softwood pulp.
Utilizing the kraft pulping process the mill averages 1000 tons/ day when producing
hardwood pulp and 850 tons/day for softwood pulp.

The mill uses a standard ﬁve stage bleaching sequence, CEDED (Abbott,
1990a):

1. chlorination (C) where the pulp is subjected to a reaction with elemental
chlorine in acidic medium;

2. alkaline extraction (E) where the pulp is subjected to a dissolution of reaction
products with N aOH;

3. chlorine dioxide (D) where the pulp is subjected to a reaction with C102 in
acidic medium;

4. alkaline extraction; and
5 . chlorine dioxide (Smook, 1982).

A four year National Pollutant Discharge Elimination System (NPDES) permit,

88

issued to Mead Corporation on March 29, 1990, set a dioxin discharge limitation of
0.022 ppq from June 1, 1992 until permit expiration in 1994. This permit level is
less than the current detection technology which is around 3 ppq for dioxin.
Although the dioxin limit is 0.022 ppq, until measuring capabilities allow for
detection limits to be accurately measured, the MDNR permit indicates any discharge
of dioxin at or above the current level of detection is considered by the MDNR
(1989c) to be a speciﬁc violation of the permit. In effect, there are two dioxin limits,
( 1) a goal of 0.022 ppq and (2) an actual limit which corresponds with the level of
detection. As long as dioxin efﬂuent levels are below detection it is impossible to
determine whether it is in fact below 0.022 ppq.

The permit included the special condition of requiring Mead Corporation to
submit a Dioxin Minimization Program. The program is intended to show that the
mill is proceeding towards compliance with the ﬁnal efﬂuent limitations for dioxin
and towards the " goal of eliminating all detectable levels of 2,3,7 ,8-TCDD in any
wastewater stream discharged to the wastewater collection facility at this facility [the
pulp mill] (MDNR, 1989c)." The Dioxin Minimiutioanah‘iskrequirw to include
the following (MDNR, 1989c):

l. A review of all potential 2,3,7,8-TCDD sources, and provision for an annual
update of this review.

2. WasteWater treatment or process modiﬁcations, currently underway and/or
proposed, intended to minimize chlorine usage and reduce the formation and
discharge of 2,3,7,8-TCDD, including but not limited to increased substitution
of chlorine dioxide for chlorine, and the use of defoamers which contain
reduced concentrations of precursors which could form 2,3,7,8-TCDD; along
with a proposed timetable for implementation.

89

3. Wastewater treatment modiﬁcations incorporating the recommendations for
total suspended solids minimization contained in the Preliminary Report on the
USEPA Bench Scale Wastewater Treatability Study: Pulp and Paper Mill
Discharges of 2,3,7,8-TCDD and 2,3,7,8-TCDF / Proposed Interim Control
Measures / Interim NPDES Permit Strategy, October 1988, along with a
proposed timetable for implementation.

4. Any additional measures necessary to proceed towards compliance with ﬁnal
efﬂuent limitations for 2,3,7,8-TCDD and towards the goal of eliminating all
detectable sources of 2,3,7 ,8-TCDD into the Wastewater collection system,
along with a proposed timetable for implementation.

5. Provision for a quarterly monitoring program to acquire data for 2,3,7,8-
TCDD at the following indicated locations: individual process wastewater
lines, inﬂuent to wastewater treatment facility, and sludge.

The permit also requires Mead to collect ten northern pike and white sucker
downstream from the wastewater outfall during the months between August and
October. The standard edible portions of each individual ﬁsh will be analyzed for
2,3,7,8-TCDD using an analytical detection limit of 1 ppt for ﬁsh tissue (MDNR,
1989c).

A 1988 analyses of the mill wastewater treatment inﬂuent sources, indicated
that 94 percent of the dioxin input into the wastewater treatment system is associated
with the kraft mill bleach plant efﬂuent (Abbott, 1990b). Figure 5.2 illustrates dioxin
discharges, based on a 24 hour composite sample, from March 1987 through June
1990. The variability in dioxin concentration from March, 1987 to September, 1989

has not been accounted for by mill staff. Components of the Dioxin Minimization

Program were instituted after September, 1989.

50

 

2378 TCDD (ppq) 9O

 

    

 

        

 

 

                 

Mar Dec Jun Sep Dec Mar Jun Sep Dec Mar Jun
1987 l 1988 | 1989 | 1990
Month

Figure 5.2 Mead Dioxin Discharges from March 1987 - June 1990.

Mead submitted a draft Dioxin Minimization Program to the MDNR in March,
1990 (Appendix E). The program includes the actual and potential implementation of
the following four technologies to limit the milk discharge of dioxin: changing to a
puriﬁed oil based defoamer, chlorine dioxide substitution in the bleaching stage, the
addition of a peroxide extraction stage to its bleaching sequence; and adding oxygen
deligniﬁcation.

The mill switched to a puriﬁed oil based defoamer after September, 1989
which decreased dioxin discharge from the 40 ppq range to 12 ppq without any
signiﬁcant added costs (Figure 5 .3). Adding chlorine substitution in the bleaching
stage and implementing a new chip screening process decreased dioxin levels to below
8.5 ppq (this was the non detect level) in June of 1990 (Figure 5.3) (Abbott, 1990a).

Future process changes may include the addition of a peroxide extraction stage to the

91

bleaching sequence with an estimated 50% dioxin reduction, and the addition of

oxygen deligniﬁcation with an estimated additional 50% reduction in dioxin.

5.2 Developing a Dioxin Damage Cost Function

The social damage cost schematic provided by Hufschmidt et al. (1989),
Figure 4.5, has been altered to reﬂect the steps in determining potential damage costs
resulting from dioxin discharge from the Mead pulp mill (Figure 5.3). While the
model is designed to be applicable to point sources discharging persistent toxic
substances into surface waters in general, the Mead pulp mill in Escanaba is used to
illustrate model components.

There are two speciﬁc types of potential damage caused by the release of
dioxin into the environment: (1) damage to wildlife populations (including birds,
mammals, ﬁsh, etc.), and (2) damage to human populations. While dioxin has been
detected in a variety of species of wildlife, (Young et al., 1987, Elliott et al., 1989)
the actual effects on species’ health is still under investigation. The $102,519 value,
derived in Chapter 4 from the simpliﬁed attempt to value dioxin damage to certain
species of ﬁsh, mammals, and birds living in the vicinity of the Mead, provides only

a gross estimate of potential wildlife damage costs.

 

 

 

 

Ellluent
Pulp Mill Discharge
oi

Dioxin
(ppq)

 

 

 

 

 

 

Dioxin
(ppq)

in
Surface
Water

 

 

 

 

 

 

Cancer

Dioxin in
Humans

Dioxin
(pp!)
in
Fish

 

 

Individual
Willingness
to Pay

to Reduce

‘ Risk

oi Death
from
Contracting
Cancer

 

 

 

Society
Willingness
to Pay

to Reduce
Each

. Member's

Risk

 

 

 

 

 

 

Figure 5.3 Estimation of social damage cost function due to the release of dioxin

from the Mead pulp mill.

Although a signiﬁcant amount of uncertainty exists in determining the actual

damage to animal life and human populations, the Michigan dual agency dioxin

regulatory framework speciﬁcally addresses the protection of human health.30 The

effects of dioxin on animal life are no less important than the effects on human health,

and clearly more research is needed to deﬁne these effects. However due to the lack

of any reasonable species census and the limited number of willingness to accept/pay

studies concerning wildlife values, this study is speciﬁcally concerned with human

exposure to dioxin from the ingestion of dioxin contaminated ﬁsh. The gross estimate

 

3° The MDNR Rule 57 states that "Toxic substances shall not be present in the
waters of the state at levels which are or may become injurious to the public
health, safety, or welfare; plant and animal life. . ." The MDPH health
advisory trigger is designed to provide information on the human health risks
of consuming contaminated ﬁsh. Separately, the MDNR and the MDPH have
their speciﬁc mandates as discussed in Chapter 3. Combined, the MDNR
water quality based efﬂuent limits and the MDPH health advisory triggers

formulate state policy designed speciﬁcally to protect public health.

 

93

of animal damage from dioxin exposure is relatively small and is essentially included
in the study by default as a result of the magnitude and ranges of the willingness to
pay values for reducing risk of death.

Figure 5 .3 illustrates the sequential development of the damage cost function.
The Mead pulp and paper mill in Escanaba has been identiﬁed as the "economic
activity" discharging efﬂuent containing a ppq concentration of dioxin into the
Escanaba River. One pathway available to dioxin in the river is uptake by aquatic
organisms.31 Assuming that 100% of the dioxin dissolved and adsorbed to suspended
solids is bioavailable to ﬁsh a potential dosage relating dioxin in the efﬂuent to dioxin
in humans can be determined”. Once a human dose of dioxin resulting from the
efﬂuent is determined, a risk of cancer can be projected. Applying willingness to pay
values for reducing risk from avoidable behavior to the projected cancer risks (as
illustrated in Chapter 4) will provide the basis for estimating a damage cost function.

A series of six equations are used to deve10p the damage cost function. The
following sequence will be taken for each equation to illustrate their use in developing
the damage cost function : (l) the equation will be described, (2) listed, and (3) an
example of the use of the equation will be provided.

Equations 1 and 2 (combined) provide the foundation from which the damage

 

3‘ Other pathways available to dioxin include immediate settling on the river
bottom, being transported some distance from the source and then settling, and
being consumed by aquatic organisms which are consequently consumed by
predatory species other than humans.

32 Considering the various pathways available to dioxin in a river system,
assuming 100% bioavailability to ﬁsh is rather conservative.

94

cost function is deve10ped. They transform the concentration of dioxin in the efﬂuent
discharge into a level of dioxin in ﬁsh. Equation 1 converts projected dioxin efﬂuent

levels into a river water concentration of dioxin in parts per quadrillion (ppq).

(Equation 1) Z = Ec x Efx l/(Rf + Ef)
where: Z = river water concentration of dioxin (ppq)
Ec = efﬂuent concentration (ppq)
Ef = efﬂuent ﬂow (mgd)

Rf = 95% exceedance ﬂow for Escanaba River (mgd)

(Example 1) Z = Ec x Ef x l/(Rf + 130

where: Z = river water concentration of dioxin @pq)
Ec = efﬂuent concentration = 10 ppq
Ef = efﬂuent ﬂow = 36.6 mgd
Rf = 95% exceedance ﬂow for Escanaba River = 122.8 mgd
Z = 10 ppq x 36.6 mgd x l/(122.8 mgd + 36.6 mgd) =

2.296 ppq

Equation 2 converts the river water concentration of dioxin into a ﬁsh body
burden of dioxin using a factor or a bioaccumulation factor (BCF and BAF
respectively). The river water concentration of dioxin from Equation 1 is multiplied
by a BCF or BAF. The resulting value is then divided by a conversion factor of 1000
which converts the value to ng/kg.

As identiﬁed in Chapter 3, the Anderson et al. (1990) site speciﬁc

95
As identiﬁed in Chapter 3, the Anderson et al. (1990) site speciﬁc

bioaccumulation factor of 7 ,238, the MDNR factor of 51,600 (Taft, 1990), and the
bioaccumulation factor of 140,000 derived by Cook (1990) are used to provide a

range of potential dioxin concentrations in ﬁsh. These three factors are representative
of the differences and ranges for species and encompass all pathway variations

reported in the literature.”

(Equation 2) Y = Z x b/lOOO

where: Y = ﬁsh body burden of dioxin (ng/kg, which is equivalent to
P90
Z = river water concentration of dioxin (ppq)

b = BCForBAF

(Example 2) Y = Z x b/lOOO

where: Y = ﬁsh body burden of dioxin (ppt)
Z = river water concentration of dioxin = 2.296 ppq
b = BCF = 51,600
Y = 2.296 ppq x 51,600 x l ng/lOOO pg = 118.5 ng/kg

The amount of dioxin in ﬁsh from Equation 2 provides a means for

comparing MDNR water quality based efﬂuent limits and MDPH health advisory

 

’3 The EPA (1990b) used similar BCFs of 5,000 and 50,000 llkg in their risk
assessment study of dioxin contaminated receiving water from chlorine-

bleaching pulp and paper mills.

96
triggers. Using the MDNR water quality based efﬂuent limit of 0.022 ppq as the

efﬂuent concentration (Ec) in Equation 1, will result in a river water concentration of
dioxin in compliance with the Mead mill’s NPDES permit. This river water
concentration can then be used in Equation 2 to determine the ﬁsh body burden of
dioxin based on each of the lbioaccumulation factors. The resulting ﬁsh dioxin body
burdens can then be compared with the MDPH 10 ppt health advisory to determine if
the MDNR water quality based efﬂuent limit is protecting human health to the margin
of safety identiﬁed by MDPH. The results of this comparison are presented in
Chapter 6.

Once converted to a body burden of dioxin in ﬁsh, a human dose of dioxin
based on the consumption of contaminated ﬁsh can be derived. Equation 3 assumes
human consumption of 6.5 grams/day of ﬁsh by a 70 kg human.34 This calculation
also employs the relatively conservative assumption of 100 % bioavailability of dioxin

to humans.”

(Equation 3) D = (Y x F,)/W
where: D = dose (mg/kg/day)

Y = ﬁsh body burden of dioxin (ng/kg)

 

3‘ As identiﬁed in Chapter 3, these assumptions are used by the MDNR in
developing water quality based efﬂuent limits and the EPA in developing water
quality criteria.

3’ Boyer (1989) concluded that 85-95 % absorption can be expected of dioxin
bioavailability in humans from the ingestion of fatty or oily foods. EPA
(1990) assumed 95% bioavailability for their pulp mill risk assessment study.

97
F = ﬁsh consumption (g/day)

W = weight (kg)

(Example 3) D = (Y x F,)/W

where: D = dose (mg/kg/day)
Y = ﬁsh body burden of dioxin = 118.5 ng/kg
F, = ﬁsh consumption = 6.5 g/day
W = weight = 70 kg person
D = (118.5 ng/kg x 6.5 g/day)/70 kg = 1.10 x 10"

mg/kg/day
The dose value derived from Equation 3 can then be inserted into the
fundamental risk formula: risk = dose x potency. In Equation 4, the resulting doses
are each multiplied by the cancer potencies of either MDNR (1.51 x 10’), MDPH
(3.57 x 10‘), or FDA (1.75 x 10‘), to determine a predicted upper limit of excess
lifetime cancer risk“. The actual risk could be lower than this upper bound

estimate.

(Equation4) P = DxX
where: P = cancer risk (incidence of cancer per exposed population)
D = dose (mg/kg/day)

X = potency (mg/kg/day‘)

 

3‘ Because the difference between the EPA and MDNR cancer potency factors is
0.05, (EPA - 1.56 x 10’, MDNR 1.51 x 10’) the MDNR cancer potency factor
is used to represent the potential cancer risks of both of the environmental
agencies.

98
(Example4) P=DxX

where: P = . cancer risk (incidence of cancer per exposed population)
D = dose = 1.10 x 10'8 mg/kg/day
X = potency = 1.51 x 105 mg/kg/day‘1
P = 1.10 x 10‘8 mg/kg/day x 1.51 x 105 mg/kg/day" = 1.66

x 10?3 = 1.66 excess cases of cancer per 1000 exposed
individuals”

Multiplying the risk estimate from Equation 4 by the Delta County, Michigan sport
ﬁsh eating population results in an estimate of the excess lifetime cancer risk for the

population (Equation 5).

(Equation 5) F, = P x F,

where: F, = risk of contracting cancer of the Delta County sport ﬁsh
consuming population

P = cancer risk (incidence of cancer per exposed population)
F = Delta County sport ﬁsh consuming population (number of

individuals)

(ExampleS) Fg=PxF

'0

where: F: = risk of the Delta County ﬁsh consuming population
P = cancer risk = 1.65 x 103 = 1.65 excess cases of cancer
per 1000 exposed individuals
F, = Delta County ﬁsh consuming population (8,242)

 

37 1.65 cases per cancer is misleading since one either has cancer or does not
have cancer, however it is merely one value in the development of the total
damage cost function.

99

F, = 1.66 x 10“3 x 8,242 = 13.6 excess cases of cancer per
8,242 Delta County sport ﬁsh consumers

Through the use of these ﬁve equations, a relationship between the amount of
dioxin discharged from the Mead pulp mill and the incidence of cancer among the
Delta County sport ﬁsh eating population is developed. Changes in efﬂuent
discharges and cancer risk incidence are directly related (this relationship reﬂects the
total damage cost curve in Figure 4.2). To complete the damage cost function
economic values are applied to each risk of contracting cancer faced by the Delta
County sport ﬁsh consuming population (derived using Equation 5).

Equation 6 multiplies a value of statistical life (also referred to as the society’s
willingness to pay io reduce each members risk of death from avoidable behavior) by
the Delta County sport ﬁsh consuming population’s risk of contracting cancer, which
is represented by the ratio of the population’s risk of contracting cancer to the total
Delta County sport ﬁsh consuming population (8,242). To derive an equalized
willingness to pay value, the resulting value is divided by the mean risk level of the
sample used to derive the initial statistical value life.

In the absence of any Delta County willingness to pay studies to determine a
population-speciﬁc statistical value of life, estimates from the literature are used. A
review of this literature identiﬁed a number of studies.38 Three studies have been

selected which represent a range of judgmental best estimates for the marginal

 

3‘ Fisher et al. (1989) provide an excellent summary of early estimates of values
of reducing the risk of avoidable death.

100

willingness to pay and mean risk levels for reducing the risk of death from avoidable
behavior. These include: Olsen (1981), Gegax et al. (1985), and Blomquist (1979).
Fisher et al. (1989) reviewed each of these studies and determined a judgmental best
estimate for the value of a statistical life for each. These values are adjusted to 1990
dollars (see Appendix F) and are used in Equation 6.

Olsen (1981) examined the wage differentials received by workers on
hazardous jobs. He concluded that "compared to nonunion workers, union members
received substantially higher fatal accident premiums." Olsen suggests that these
increased premiums indicate that union members collectively place a higher value on
life than nonunion workers. The judgmental best value for a statistical life for the
Olsen (1981) study is $6.8 million, which is based on a mean risk level for the
sample of 1 x 10‘4 (one excess case of cancer per 10,000 individuals).

The Gegax et al. (1985) wage-risk study used a "mail study to collect
information on annual labor earnings, the perceived risk of fatal accidents at work,
the individual’s human capital, work environment, and personal characteristics"
(Fisher et al., 1989). The study provided a statistical value of life of $1.36 million
for union blue collar workers based on a mean risk level for the sample of 10.1 x 10“
(10.1 excess cases of cancer per 10,000 individuals).

Blomquist (1979) studied the "typical individual’s value of a small change in
the probability of his survival." The statistical value of life of $518,624 was
determined through the analysis of automobile seat-belt use. This value is based on a

mean risk level for the sample of 3 x 10" (3 excess cases of cancer per 10,000

101

individuals) .

(Equation 6) Cd = (WTP x (FJ8,242))/(S,)

where: Cd = damage cost ($)
F, = risk of the Delta County ﬁsh consuming population
(excess cases of cancer per exposed population)
WTP = society’s willingness to pay to reduce each
members risk of death from avoidable behavior
($)
Sr = mean risk level of sample (excess case if cancer per
mmma

(Example 6) Cd = (WTP x (F,/8,242))/(S,)

where: Cd = damage cost ($)
F, = risk of the Delta County ﬁsh consuming population =
13.6/8,242 (13.6 excess cases of cancer per 8,242 Delta

County sport ﬁsh consumers)
WTP = society’s willingness to pay to reduce each
members risk of death from avoidable behavior =

$6,801,622
Sr = mean risk level of sample =
l x 10"

c, = ($6,801,622x(13.6/8,242))/0.0001=
$112,232,549

A separate damage cost curve is developed for the twenty-seven combinations

of bioconcentration/bioaccumulation factors (7,2368, 51,600, and 140,000), cancer

102

potency factors (1.51 x 105, 3.57 x 10", and 1.75 x 10“) and values of statistical life
($6,801,622 million, $1,360,323, $518,624). Figure 5.4 expands the conceptual
diagram shown earlier (Figure 5 .3) by adding the variables for each equation used in
the deveIOpment of a damage cost curve. The damage cost functions are based on a
range of efﬂuent concentrations of 10 - 0.01 ppq. Each efﬂuent concentration in the
range provides one point, for a total of 30 points, on each of twenty-seven damage

cost functions.

103

.wou

.moo

_wOU

.mOU

ngU

.wOU

unOu

.nou

umOU

300

10.00

“”00

—wOu

unoo

:00

300

.uOU

unoo

unOu

—uOU

.900

:00

.000

.uOu

anon.

uoou

«$00

.5956 i eo.__.E n on
325.6 i :95... v.3
oaneu ill :25... 06»
00553 i €9.26 mam
oonanollili e235 v.5
0025... i 8....E new
eannu i .65.... «Om
eonseei c255 v.5

90253 i :25... 06w

conﬁne i co._..E wrom
036...: i co._..E v.3

009.:an CREE 0.0»

@362. ill 5:26 ndm
oannui :25... «.3
oooE-villl c256 06m
some: i 8.26 new
oomEaui :25... I»

838i 8.55 93

002.30 i co.=.E mdm
oanaui c055. v.5
cone-0i :95... new
conﬁne i :25... mdm
oonEaui :25... 3.0.
oooEnui eo....E ed»
0925:. i co.=.E n on
oousuui 295:. v;

002:an co:=E 9o»

.0 cozaaao.

i z»: oo 2.00 i a»: coon-oi 7.9 x n:

.I .3: .00 2.3

 

i .3: .oo «:60

 

 

i a: .8 2.3

i .3: .oo 2.00

 

Oi 1m: .00 @200

 

 

i .3: .00 23a

 

i .3: .00 3.00

i In: .00 3.00

 

3 29.253

x»: .eucaui «.0. x .m—

xm: .oucaui v.9 x a:

x»: 3053i 0.0. n .m—

xo: .oocooolllll v.9 x a:

..a: .oocsoiv.o_ .- 3.0

10.: ~00ca04l'l m. 0— x —0.—

.v c9330!

 

8 see-ace.

3 eo:2..ao.

cocci c092. 5..

.n c9330!

 

..uc .oocuui v.0. x 3 mi :2. i :35.» 231.3 000.0!

a cozcaao.

.w Caz-503

9:. 83.

.m c9333

.55 Eng v.82 05 EB... 5x06 mo ammo—8 2: 8m 88o camps—e mEZeoQ

an 03mm

: cos-:03

.23 6.0 . one o:

no: coucuoivkx x 5060'. Onouoll c0059 50:] Oz: oceans-Ii .ocoo uttolllll 6:00 2.02:0

104

5.3 Developing an Abatement Cost Function

The abatement cost function is related directly to the costs of the technologies
employed to reduce the concentration of dioxin in mill efﬂuent. Each of the four
technologies outlined in Mead’s Dioxin Minimization Program has a corresponding
cost. As these technologies are added to the pulp bleaching process, abatement costs
increase and dioxin in the efﬂuent decreases.

There were no signiﬁcant costs associated with Mead’s switch to a puriﬁed oil
based defoamer. However, the remaining technologies did have signiﬁcant capital
costs. The capital cost of adding chlorine substitution in the bleaching stage was $4.1
million and the new chip screening process cost $12.8 million for a combined cost of
$16,900,000 (Abbott, 1990a). If implemented, the addition of a peroxide extraction
stage to the bleaching sequence would have an estimated cost of $250,000, and the
addition of oxygen deligniﬁcation would have an estimated cost of $140 million. In
deriving an abatement cost it is assumed that both peroxide extraction and oxygen
deligniﬁcation will be implemented.

The total cost for dioxin abatement is the sum of the depreciated capital cost
and the depreciated operating cost. The depreciated operating cost is calculated based
on the following assumptions (HOppe, 1990):

1. Hardwood pulp price = $600/air dry metric ton of pulp (admt),
2. A 5% proﬁt margin,
3. A 30% tax margin,

4. 15% for overhead costs,

105

5. No local property taxes, and
6. 15 years for straight line depreciation.
Operating costs are back calculated from the hardwood pulp price using the

percentages for proﬁt margin, tax margin and the overhead costs:

Operating costs = ((($600/admt x 95 %) x 70%) x 85%) = $339.15/admt
The mill produces 1000 tons of hardwood pulp a day. Converting this value to air
dry metric tons and multiplying by 365 days will provide the annual hardwood pulp

production in air dry metric tons:

Hardwood pulp production = 1000 tons/day x .907 admt/ton x 365 days =
331,055 admt/yr

The hardwood pulp production multiplied by the operating costs divided by a 15 year

straight line depreciation will provide a depreciated operating cost (Hoppe 1990):

Depreciated
operating = (331,055 admt/yr x $339.15/admt)/15 yrs
costs = $7,485,153.55

Table 5.2 illustrates the total cost per year for the three technologies. Table 5.3

provides the cumulative per day equivalent abatement costs and the associated dioxin
efﬂuent concentration. The dioxin efﬂuent concentration for the addition of peroxide
extraction stage and oxygen deligniﬁcation is projected based on a 50% reduction for

each added technology.

106

 

 

 

 

Table 5 .2 Mead abatement processes, capital costs, depreciated capital and
operating costs, and total costs.
Present Depreciated Depreciated
Capital Capital Operating Total
Abatement Cost Cost/yr Cost/yr Cost/yr
Process ($) ($) ($) (Si
C102 substitution
Chip screening 16,900,000 1,126,667 7,481,843 8,608,510
Extraction stage 250,000 16,667 7,481,843 7,498,510
Oxygen
delignification 140,000,000 9,333,333 7,481,843 16,815,176
Table 5 . 3 Mead Abatement Technologies, Cumulative Per Day Costs, and
Resulting Dioxin Discharges
Dioxin
Abatement Cost Discharge
Technology $/day (ppq)
ClO2 substitution
Chip screening 23,585 7.5
Extraction stage 44,129 3.5
Oxygen
deligniﬁcation 90,198 1.5

 

107

Figure 5.5 is a plot of the per day abatement costs and the corresponding
dioxin discharges. The following power function for a dollar per day equivalent

dioxin abatement value can be ﬁtted to total abatement costs in Table 5 .3 (r2 = 1.00):

$/day = 126125 (dioxin concentration)"83

Using this equation, an abatement cost point is calculated for the 10 - 0.01 ppq range
of dioxin concentrations. When combined, these points suggest a total abatement
cost curve that extends beyond the current abatement range (Figure 5.6) and

resembles the hypothesized abatement cost curve illustrated earlier in Figure 4.2.

Dollars (x 1,000)
140 ——

 

120
100
80
60

4O

 

20

 

 

 

1.5 3.5 7.5
Dioxin Effluent Concentration (DDCI)

Figure 5.5 Per day Mead abatement costs and corresponding dioxin discharges.

108

Dollars (x 1,000,000) Per Day

 

 

 

 

 

 

 

 

1.llll
lixrrrﬁ‘tlveggiilll

(3.01 0.04 0.07 0.1 0.4 0.7 1 3 5 7.5 10
Dioxin Effluent Concentration (DDQ)

 

Figure 5.6 The Mead abatement cost curve extended and corresponding dioxin
discharges.

The abatement cost curve for the Mead mill is added to each of the twenty-
seven damage cost functions to derive twenty-seven separate total costs of pollution.
The minimum of each of these total cost functions provides twenty-seven separate
optimum levels of pollution (in the Pareto sense). A comparison of the range of
"optimal levels of pollution" provides a means for determining whether or not the
application of the MDNR’s 0.022 ppq dioxin efﬂuent limit for Mead’s Escanaba kraft
pulp mill and a MDPH dioxin health trigger of 10 ppt in ﬁsh for the sport ﬁsh
consuming population of Delta County leads to an efﬁcient (Pareto optimal) allocation
of resources. The results of this comparison and discussion are presented in Chapter

6.

CHAPTER SIX: RESULTS AND DISCUSSION

The policy analysis model uses the concept of cost minimization to determine a
Pareto optimal level of pollution (dioxin discharge), which has been described as the
level that maximizes social wellbeing and minimizes costs. In the absence of a true
optimal level of pollution, Pearce and Turner (1990) suggest, "the analytical task is to
seek out the least-cost policy package sufﬁcient to meet acceptable quality standards."
Michigan dioxin policy, as applied to the Mead pulp mill in Escanaba, has two
acceptable standards which both must be met: (1) an MDNR water quality based
efﬂuent limit for dioxin which is 0.022 for the Mead mill in Escanaba”, and (2) a
MDPH health advisory trigger for dioxin in ﬁsh which is 10 ppt for ﬁsh in the
Escanaba River.

Based on the set of assumptions for the policy analysis model presented in
Chapter 5, the ”least cost policy package" (referred to in this research as the Pareto
optimal dioxin discharge level) for Mead Corporation’s Escanaba kraft pulp mill has
been derived. An attempt will then be made to determine whether or not the "policy
package" matches the current standards of Michigan’s dual agency dioxin policy. If
the policy package matches the standards of Michigan’s dioxin policy, then the
Michigan policy will be deemed Pareto optimal. To determine whether such a match

exists the issue of dual agency coherence was addressed: Does the level of dioxin

 

’9 As pointed out in Chapter 5, this limit is at present purely hmthetical as
current detection levels with best available technology range between 1 and 4
PP‘l-

109

110

permitted into the environment, by MDNR’s water quality—based efﬂuent limit, when
bioaccumulated by ﬁsh, provides the same margin of safety as the MDPH health
advisory trigger? Identifying the value of the damage caused when humans are
exposed to dioxin in sport ﬁsh is a key component in making this determination.

To determine the Pareto optimal dioxin discharge level for the Mead mill, the
policy analysis model is used to evaluate the 27 separate policy scenarios which
encompass combinations of three different bioconcentration/bioaccumulation factors
(7,238, 51,600, and 140,000), three different cancer potency factors (1.51 x 105, 3.57
x 10“, and 1.75 x 10“) and three values of societal willingness to pay to reduce a
members risk of death from avoidable behavior ($6,801,622 million, $1,360,323,
$518,624). Although the primary purpose of this research is to evaluate Michigan
dioxin policy, a secondary purpose is to compare FDA’s dioxin ﬁsh advisory
development to the development and application of Michigan’s dioxin policy.
Therefore, the FDA’s dioxin in ﬁsh consumption advisory along with the agency’s

risk assessment assumptions, is included in the 27 policy analysis scenarios“.

 

4° A further option of including a range of consumption levels was not

taken due to timing constraints. However, EPA used a range of
consumption levels in their pulp mill dioxin risk study (USEPA,
1990b).

l l 1
6.1 Results

Each of the 27 policy scenario’s include an abatement cost curve, a damage
cost curve, and a total cost of pollution curve (similar to those illustrated by Figure
4.2). The series of Figures 6.1-6.3, graphically represent the logarithms of curves
for the policy scenario that includes a bioconcentration factor of 51,600 and the
willingness to pay value of $6,801,622 ($6.8 million). Each ﬁgure depicts the curves
based on a different cancer potency factor. Figure 6.1 illustrates the abatement cost,
damage cost, and total cost using the MDPH cancer potency of 3.57 x 10“. Figure
6.2 illustrates the abatement cost, damage cost and total cost using the MDNR cancer
potency factor of 1.51 x 105, and Figure 6.3 illustrates the abatement cost, damage
cost and total cost using the FDA cancer potency factor of 1.75 x 10“. At the
minimum of the total cost curve on each of the ﬁgures, a line has been drawn to
indicate the Pareto optimal solution (the point indicated on the x-axis by the line is
representative of Q, in Figure 4.2). The Pareto optimal level of dioxin discharge (the
level of dioxin discharge which maximizes social wellbeing and minimizes costs under
the speciﬁed set of assumptions) indicated by Figures 6.1-6.3 is: 0.2 ppq for MDPH,

0.08 ppq for MDNR, and 0.3 ppq for FDA.

112
5 Dollars (log)

 

l

7
l .

l

6.5

6 r- "3‘..¢~.‘_ ’g”

5.5

l

5.—

 

4.5

 

 

L l l l 1 1 l l l 1 l 1 1 l 1 1 l l 1 l l 1 1 l l l l 1

3.01 0.04 0.07 0.1 0.4 0.7 1 3.5 7 10
Dioxin Effluent Concentration (DDQ)

 

MDPH

4‘ Abatement Cost (log) + Damage Cost (log) ‘9" Total Cost (log)

Figure 6.1 Mead total abatement cost, damage cost, and total cost of pollution for
MDPH risk assessments, based on a willingness to pay value of $6.8
million dollars and a bioconcentration factor of 51,600, as a function of

dioxin discharge.
Dollars (log)

 

 

 

 

 

8»- W‘
MI
7 >— .
1%
6 W
5“.—
4 1 l l l l l l l l l l l 1 l l 1 1 #1 L 1 I l l l 1 l 1 ‘0
0.01 0.04 0.07 0.1 0.4 0.7 1 3.5 7 10
Dioxin Effluent Concentration (ppq)

MDNR
—“— Abatement Cost (log) —*— Damage Cost (log) + Total Cost (log)
Figure 6.2 Mead total abatement cost, damage cost, and total cost of pollution for

MDNR risk assessments based on a willingness to pay value of $6.8
million dollars and a BCF of 51,600, as a function of dioxin discharge.

113

Dollars (log)

 

7.5

7 ._ / 4"me
'1 /"// M

6.5

X
t\
t

6+—

1

5.5
5 .—

4.5 /

 

 

 

l 1 l 1 1 1 1 l l l l l l 1 l 1 1 1 l l l 1 l 1 l 1 l 1

0.01 0.04 0.07 0.1 0.4 0.7 1 3 5 7.5 10
Dioxin Effluent Concentration (ppq)

 

USFDA

—‘— Damage Cost (log) "l— Abatement Cost (log) + Total Cost (log)

Figure 6.3 Mead total abatement cost, damage cost, and total cost of pollution for
FDA risk assessments based on a willingness to pay value of $6.8
million dollars and a BCF of 51,600, as a function of dioxin discharge.

An alternative to measure the Pareto optimal level of dioxin discharge is the
total abatement cost to willingness to pay ratio (TAC/WTP). A ratio of one indicates
the break even point. This point is the level of protection (ppq) that people are
willing to pay for. Beyond this point society is unwilling to pay for additional
protection. Figure 6.4 graphically represents the concept of total abatement cost to
willingness to pay for the policy scenario that includes a bioconcentration factor of

51,600 and the willingness to pay value of $6.8 million, and the risk assessment

assumptions for each agency. The line drawn from the ratio curves to the x-axis

114

indicates the optimal levels of dioxin discharges based on what pe0ple would be

willing to pay for: 0.2 ppq for MDPH, 0.09 ppq for MDNR, and 0.3 ppq for FDA.

 

Log Total Abatement Cost/WTP

 

 

do \:

_4 l l 1 l l l 1 1 l l 1 l l l l l l l l 1 l 1 l 1 1 1

 

 

 

 

 

 

0.01 0.03 0.05 0.07 0.09 0.2 0.4 0.6 0.8 1 2 3.5 5 7 8 10
Dioxin Effluent Concentration (990)

+ MDNR TAG/WTP (log) ”*— MDPH TAO/WTP (log)
‘7‘“ FDA TAG/WTP (log)

Figure 6.4 The log of the ratio total abatement cost/willingness to pay for MDNR,
MDPH, and the FDA, and corresponding levels of dioxin discharge.

If compared, the total abatement costs to willingness to pay ratio closely
resembles the values obtained from the minimum of the total cost of pollution curve
(Figures 6.1 - 6.3). The slight difference exhibited among the MDNR values, 0.08
ppq (minimum of the total cost of pollution curve) compared to 0.09 ppq (total
abatement cost to willingness to pay ratio), is most likely a result of an error in

rounding the value.

115

Both types of graphic representations of the Pareto optimal level of dioxin

discharge could be depicted for each of the 27 different policy scenarios, however to

conserve space, Tables 6.1-6.6 have been constructed to present the results from all

twenty-seven policy scenarios. The values presented in Tables 6.1-6.6 are based on

the following assumptions: a 6.5 g/day consumption of sport ﬁsh from the Escanaba

River by a 70 kg individual, 100% bioavailability of dioxin in water to ﬁsh, and

100% bioavailability of dioxin in ﬁsh to human. Tables 6.1 - 6.3 present the

following information for both the MDPH and the MDNR risk assessment

assumptions:"1

1.

2.

total damage cost,
total abatement cost,
total cost of pollution,

the optimal level of dioxin discharge that corresponds with the minimum total
damage costs,

the ﬁsh body burden of dioxin that corresponds with the optimal level of
dioxin discharge given a speciﬁc bioaccumulation factor (BAF) or
bioconcentration factor (BCF),

the total abatement cost to willingness to pay ratio,

the Delta County sport ﬁsh consumer risk of contracting cancer, and

 

41

These assumptions include for both agencies, an acceptable risk level of one
excess case of cancer per 100,000 exposed individuals (1 x 10’), the Kociba et
al. (1978) rat study, and the use of a linearized multistage model. For the
MDNR, a cancer potency factor for dioxin of 1.51 x 105 mg/kg/day based on
a body surface area extrapolation of the rat data, and for the MDPH, a cancer
potency factor for dioxin of 3.57 x 104 based on a body weight scaling factor
is assumed.

8.

116

an extrapolated risk per 100,000 individuals.

Tables 6.4-6.6 present the same information for the FDA risk assessment assumptions

which include a cancer potency factor of 1.75 x 104 based on a body weight scaling

factor used to extrapolate rat data to humans.

 

 

 

 

 

 

 

 

 

 

Table 6.1 Pareto optimal levels of dioxin discharge based on MDNR and MDPH
risk assessment assumptions and a BAF of 7,238, for the Escanaba
Mead pulp mill and the sport ﬁsh eating Delta County, MI p0pulation
(costs per day - 1990).
MDPH MDNR

Willingness to pay $6,801,622 $1,360,323 $518,624 $6,801,622 $1,360,323 $518,624

Total

Damage 187.360 29.681 33.335 316,990 62.770 60.426

Cost ($)

Total

Am 224.21 1 39.91 1 44.589 479,674 70,949 90.0“

(201(8)

Total

Cost of 411,571 69,592 77,923 796.664 133.719 150.510

Pollraion (8)

Optirll

Pollution 0.5 4.0 3.5 0.2 2.0 1.5

mm

Fish

Body 0.8 7.0 6.0 0.3 3.0 2.0

aumrm

TAC/‘VI’P (ppq) 0.6 5.0 4.0 0.3 3.0 2.0

Delta County

sport ﬁsh consumer 0.02318242 0.182/8242 0.159/8242 0038/8242 0384/8242 0.288/8242

risk

risk per 100.000 0.28 2.20 1.93 0.46 4.66 3.49

 

 

 

 

 

 

 

 

117

 

 

 

 

 

 

 

 

 

 

Table 6.2 Pareto optimal levels of dioxin discharge based on MDNR and MDPH
risk assessment assumptions and a BCF of 51,600, for the Escanaba
Mead pulp mill and the sport ﬁsh eating Delta County, MI population
(costs per day - 1990).
MDPH MDNR

Willingness to pay 86,801,622 51,360,323 8518.624 86.801.622 81.360.323 8518.624

Total

Damage 534.279 79.348 101.847 897.949 155.585 171.172

Cost ($)

Total

Abaternera 479.674 90.084 90.084 1 £26,210 169.578 192.724

Cost (8)

Total

Cost of 1.013.954 169.432 191,931 1,930,145 326.21!) 365.038

Pollution (8)

Optimal

Pollution 0.2 1.5 1.5 0.1! 0.7 0.6

Level (ppq)

Fish

Body 2.0 18 18 1.0 8.0 7.0

Burden (W)

TAO/WTP (ppq) 0.2 2.0 1.5 0.09 0.1 0.7

Delta County

sport fish consumer 0.067/8242 0.486/8242 0.486/8242 0.110/8242 0358/8242 0823/8242

risk

risk per 100.000 0.81 5.6 5.6 1.3 11.6 9.97

 

 

 

 

 

 

 

 

118

Table 6.3 Pareto optimal levels of dioxin discharge based on MDNR and MDPH
risk assessment assumptions and a BAF of 140,000, for the Escanaba
Mead pulp mill and the sport ﬁsh eating Delta County, MI population

(costs per day - 1990).

 

 

 

 

 

 

 

 

 

 

MDPH MDNR
Willingness to pay 86.801.622 81.360.323 8518.624 86.801.323 81.360.323 5518.624
Total
Damage 724.798 1 14.819 128.954 1.532.835 242.825 233.758
Cost ($)
Total
Abatement 852.710 151.788 169.578 1.515.849 269.831 432.“)3
Cost (5)
Total
Cost of 1.577.507 266.607 298.532 3.048.684 512.656 576.361
Pollution (5)
Optimal
Pollution 0.1 0.8 0.7 0.05 0.4 0.3
Level (ppq)
Fish
Body 3 26 23 1.6 13 10
Burden (ppt)
TAC/WI'P (ppq) 0.2 1.0 0.9 0.05 0.5 0.4
Delta County
sport fish consumer 0.088/8242 0.703/8242 0.615/8242 0186/8242 1.486/8242 1.11/8242
risk
risk per 100.000 1.07 8.50 7.46 2.30 18.0 13.5

 

 

 

 

 

 

 

 

119

Table 6.4 Pareto optimal levels of dioxin discharge based on FDA risk assessment
assumptions and a BAF of 7,238, for the Escanaba Mead pulp mill and

the sport ﬁsh eating Delta County, MI population (costs per day -

1990).

 

Willinpreaa to pay

56.801 .622

81.360.323

5518.624

 

Total
Damage
Cost ($)

128.581

21.824

23.334

 

Total
Abatemera
Cost ($)

169.578

28.506

33.163

 

Total
Cost of
Pollution ($)

298.157

50.330

56.507

 

Optimal
Pollution
W (rm)

0.7

 

Fish
Body
Burden (ppt)

1.2

9.97

8.3

 

TAG/WTP (ppq)

0.9

 

Delta County
sport ﬁsh eonsurrrer
risk

0.016/8242

0.134/8242

0.111/8242

 

risk per roo.ooo

 

 

0.19

 

1.63

 

 

 

Table 6.5 Pareto optimal levels of dioxin discharge based on FDA risk assessment
assumptions and a BCF of 51,600, for the Escanaba Mead pulp mill

and the sport ﬁsh eating Delta County, MI population (costs per day -

1990).

 

Willingness to pay

$6,801,622

51.360.323

5518.624

 

Total
Damage
Cost ($)

392.853

51.862

66.567

 

Total
Abatemera
Cost ($)

342.603

70.949

70 .949

 

Total
Cost of
Pollution (5)

735.455

122.811

137.516

 

Optimal
Pollution
Level (ppq)

0.3

I)

 

Fish
Body
Burden (ppt)

3.5

24

 

TAG/WTP (ppq)

0.3

 

Delta County
sport ﬁsh consumer
risk

0.048/8242

0.317/8242

0.317/8242

 

risk per 100.000

 

 

 

3.85

 

 

 

120

Table 6.6 Pareto optimal levels of dioxin discharge based on FDA risk assessment
assumptions and a bioaccumulation factor of 140,000, for the Escanaba
Mead pulp mill and the sport ﬁsh eating Delta County, MI population
(costs per day - 1990).

 

Willingness to pay $6,801.622 $1,360,323 $518,624

 

Total °
Damage 479.674 90.084 126. 125
Cost ($)

 

Total
Abatement 710.586 105.532 90.30.:
Cost ($)

 

Total
Cost of 1.190.260 195.616 216.429
Pollution (5)
Optimll

Pollution 0.2 1.5 1
Level (ppq)

Fish -
Body 6.4 48.2 32 1
Burden (ppt)

 

 

 

rxcxwrr (ppq) 0.2 1.5 1.5

 

Delta County
sport ﬁsh consumer 0086/8242 0.646/8242 0432/8242
risk

 

risk per 100.000 1.04 7.34 5.24

 

 

 

 

 

 

6.2 Discussion

In Search of a Pareto Optimum The Pareto optimal levels of dioxin discharge
concentrations in Tables 6.1 - 6.6 indicate a range of optimal pollution levels,
corresponding ﬁsh body burdens, and cancer risks per 100,000 exposed individuals
for the MDNR, MDPH, and FDA risk assessment assumptions. The information
presented in the tables places value on the damage costs of human exposure to dioxin,
in an attempt to address the issues of whether or not Michigan state agency standards,
which make up the state’s dioxin policy, provide a consistent margin of safety.

The policy analysis model is designed to provide a "Pareto optimal pollution"

121

for every set of variables under consideration. However, in determining whether a
particular efﬂuent dioxin concentration provides an adequate margin of safety the
model results for the dioxin concentration (in ﬁsh) must meet two objectives: (1) be
below the 10 ppt MDPH health advisory, and (2) not exceed a cancer risk of one
excess case of cancer per 100,000 exposed individuals.

The following is an explanation of Tables 6.1-6.6. Each table is based on a
different bioconcentration/bioaccumulation factor. Tables 6.1 and 6.4 are based on a
bioaccumulation factor of 7,328. Tables 6.2 and 6.5 are based on a bioconcentration
factor of 51,600, and Tables 6.3 and 6.6 on a bioaccumulation factor of 140,000.
Each table presents different Pareto optimal levels of dioxin discharge, dioxin ﬁsh
body burdens, cancer risk for the Delta County sport ﬁsh consuming population, and
cancer risk per 100,000 exposed individuals for each of the three willingness to pay
values. Each of the following bullets are organized to present the
bioconcentration/bioaccumulation factor for which the information is based on, the
three willingness to pay values, and for each willingness to pay value the (1) pareto
optimal level of dioxin discharge, (2) ﬁsh dioxin body burden, and (3) cancer risk per
100,000 exposed individuals respectively. A Pareto optimal dioxin discharge
concentration which meets both of the objectives (below 10 ppt and a 1 x 105 cancer
risk) will be bolded.

The Pareto optimal level of dioxin discharge, the dioxin body burden in ﬁsh,
and the cancer risk per 100,000 exposed individuals based on a bioaccumulation

factor of 7 .238 and corresponding to the following societal willingness to pay factors

122

for reducing a member’s risk of death from avoidable behavior (the value of a

statistical life) are:

For 5681111111911: 0.5 ppq, 0.8 ppt, 0.28 x 10" for MDPH; 0.2 ppq, 0.3 ppt,
0.46 x 10‘ for NHDNR; and 0.7 ppq, 1.2 ppt, 0.19 x 10“ for FDA.

Based on a $1,360,323 ($1.4 millign) willingness to pay value the range
includes: 4.0 ppq, 7.0 ppt, 2.20 x 10" for MDPH; 2.0 ppq, 3.0 ppt, 4.66 x
105 for MDNR; and 6.0 ppq, 9.97 ppt, 1.63 x 10“ for FDA.

And, based on a $518,624 (89.5 million) willingness to pay value the range
includes: 3.5 ppq, 6.0 ppt, 1.93 x 10" for MDPH; 1.5 ppq, 2.0 ppt, 3.49 x
10“ for MDNR, and 5.0 ppq, 8.3 ppt, 1.35 x 10" for FDA.

Based on a bioconcentration factor of m and corresponding societal

willingness to pay factors for reducing a member’s risk of death from avoidable

behavior the Pareto optimal level of dioxin discharge, the dioxin body burden in ﬁsh,

and the cancer risk per 100,000 exposed individuals are:

Based on 86.8 million, the Pareto optimal level of dioxin discharge, the dioxin
body burden in ﬁsh, and the cancer risk per 100,000 individuals are as
follows: 0.2 ppq, 2.0 ppt, 0.81 x 10'5 for MDPH; 0.08 ppq, 0.3 ppt, 1.3 x
105 for MDNR; and 0.3 ppq, 3.5 ppt, 0.58 x 10“ for FDA.

Based on a $1.4 millign willingness to pay value the range includes: 1.5 ppq,
18 ppt, 5.6 x 105 for MDPH; 0.7 ppq, 8.0 ppt, 11.6 x 10" for MDNR; and
2.0 ppq, 24 ppt, 3.85 x 10" for FDA.

And, based on a $0.5 milligg willingness to pay value the range includes: 1.5

123
ppq, 18 ppt, 5.6 x 10" for MDPH; 0.6 ppq, 7.0 ppt, 9.97 x 10-5 for MDNR,

and 2.0 ppq, 24 ppt, 3.85 x 10" for FDA.

Based on a bioaccumulation factor of 1510.!!!) and corresponding societal
willingness to pay factors for reducing a member’s risk of death from avoidable
behavior the Pareto optimal level of dioxin discharge, the dioxin body burden in ﬁsh,
and the cancer risk per 100,000 exposed individuals are:

0 Based on 86.8 million, the Pareto optimal level of dioxin discharge, the dioxin
body burden in ﬁsh, and the cancer risk per 100,000 individuals are as
follows: 0.1 ppq, 3.0 ppt, 1.07 x 10" for MDPH; 0.05 ppq, 1.6 ppt, 2.3 x
105 for MDNR; and 0.2 ppq, 6.4 ppt, 1.04 x 10" for FDA.

0 Based on a W willingness to pay value the range includes: 0.8 ppq,
26 ppt, 8.50 x 10’ for MDPH; 0.4 ppq, 13.0 ppt, 18.0 x 10‘5 for MDNR; and
1.5 ppq, 48.2 ppt, 7.84 x 10" for FDA.

0 And, based on a M1211 willingness to pay value the range includes: 0.7
ppq, 23 ppt, 7.46 x 10" for MDPH; 0.3 ppq, 10.0 ppt, 13.5 x 10" for
MDNR, and 1.0 ppq, 32.1 ppt, 5.24 x 10‘5for FDA.

If considering only the ﬁrst objective, providing a dioxin concentration which
when bioaccumulated in ﬁsh provides a contamination value below the 10 ppt MDPH
health advisory trigger, 15 of the 27 policy combinations would provide a sufﬁcient
margin of safety. Figure 6.5 is a series of three graphs, one for each of the three
bioconcentration/bioaccumulation factors, which illustrates the variability among the

ﬁsh dioxin body burdens for the MDNR (DNR), MDPH, and FDA for each

124

willingness to pay value.

The ﬁsh body burdens of dioxin associated with the willingness to pay value of
$6.8 million dollars all fall signiﬁcantly below the MDPH health advisory trigger.
While a number of the ﬁsh body burdens of dioxin fall below the MDPH 10 ppt
health advisory for both the $1.4 million and the $0.5 million willingness to pay
values, it appears that these values closely parallel each other. For example, the ﬁsh
body burdens of dioxin for a BCF of 51,600 and willingness to pay values of $1.4
million and $0.5 million respectively are: 18 ppt and 18 ppt for MDPH, 8.0 ppt and
7.0 ppt for MDNR, and 24 and 24 ppt for FDA.

One explanation for the similar values is that although the $1.4 million
willingness to pay value is more than twice the amount of the $0.5 million willingness
to pay value, the $0.5 million value has a mean risk level threefold smaller than that
of the $1.4 million value, 3 x 10“ and 10.1 x 10" respectively. In developing the
damage cost function, the Delta County sport ﬁsh eating population’s risk of
contracting cancer is ﬁrst multiplied by the willingness to pay value and is secondly
divided by the mean risk level of the sample used to determine the willingness to pay
value. Dividing by the mean risk of the study sample equalizes the willingness to pay
values. As a result, once equalized, the $1.4 million and $0.5 million willingness to

pay values are relatively similar.

125

set-7.2.

 

 

 

 

 

 

 

 

 

 

 

L.

 

 

 

 

 

 

 

rlsu clam soot strata ("1)
O
r;
O
3
aeucr

 

 

o) 13: ab
ultuucuus ro nw (Millions)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

aer - smoo
e a __
g 1.
so 8
B
S
. 1;; .. t
5 re
.-. 5;? ..
if ° ' 0.3 1.5 a
urttlucusaa ro Mv (Uttttaaa)
DC! I 140.000
‘3 so
g I f a
so 5
z 1;) ..... e
<
' so
a . .
3 l D
g o ' 0.3 1.0 a)

 

HILUNGNC. ro PAY (U1111000)

Figure 6.5 Comparison of MDPH, MDNR, FDA, ﬁsh dioxin body burdens based
on willingness to pay values and bioconcentration factors.

126
The 15 body burdens of dioxin values (as listed in Tables 6.1-6.6) that are

below 10 ppt, from low to high are: 0.3 ppt, 0.8 ppt, 1.0 ppt, 1.2 ppt, 2.0 ppt (listed
twice), 3.0 ppt (listed twice), 3.5 ppt, 6.0 ppt, 6.4 ppt, 7.0 ppt (listed twice), 8.3 ppt,
and 9.97 ppt. The lowest ﬁsh body burden (0.3 ppt) of dioxin results from the policy
combination which includes an MDNR cancer potency factor, bioaccumulation factor
of 7,238, and a willingness to pay value of $6.8 million (Table 6.1). The highest
concentration, which is still below the 10 ppt MDPH health advisory, (9.97 ppt)
results from the policy combination which includes an FDA cancer potency factor, a
BCF of 7 ,238, and a willingness to pay value of $1.4 million. The signiﬁcance of
these two values is that although there is a difference of 9.67 ppt of dioxin, yet they
are both under the MDPH health advisory and are therefore both deemed to be safe to
eat.

The differences in the ﬁsh body burdens of dioxin are associated with each of
the different bioconcentration/bioaccumulation factors. The lower the value of the
bioconcentration/bioaccumulation factor the lower the value will be for dioxin in ﬁsh.
Based on the BCF used by the MDNR in making dioxin regulatory decisions (51,600)
in Figure 6.5 and Tables 6.2 and 6.5, ﬁsh body burdens of dioxin span the range of a
low of 1.0 ppt for a MDNR cancer potency factor and $6.8 million willingness to pay
value to a high of 24 ppt for a FDA cancer potency factor and willingness to pay
values of $1.4 million and $0.5 million. Clearly, the different agency risk assessment
assumptions, when the BCF is held constant at a current regulatory level, provide an

inconsistent margin of safety. The ﬁsh body burdens of dioxin estimates indicate that

127

the different potency values used for estimating excess lifetime cancer risk result in
different levels of protection. However, meeting the MDPH health advisory for
dioxin of 10 ppt is only one of the policy objectives, the cancer risk level must also
be evaluated.

The second objective in determining whether a particular efﬂuent dioxin
concentration provides an adequate margin of safety is to evaluate whether it provides
the same cancer risk used by the MDPH, MDNR, and FDA in their risk assessment
processes: a risk of one excess case of cancer per 100,000 exposed individuals.
Figures 6.6 - 6.8 illustrate the cancer risks .per 100,000 exposed individuals associated
with ﬁsh dioxin body burdens. The cancer risks were extrapolated from the potential
excess cases of cancer for the Delta County sport ﬁsh consuming population of 8,242

individuals (Tables 6.1 —6.6).

14 Risk of Contracting Cancer Per 100,000 _

 

12‘

10a '

8“ .,

61 .

4-

o T f ;
7 83 10 18

. _ 23 24 32.1
Fish Body Burden of Dioxin (ppt)

\hh

   

 

 

é

. .

 

 

 

 

 

- FDA

Figure 6.6 The risk of contracting cancer for a population dioxin contaminated ﬁsh
based for FDA, MDPH and MDNR cancer potency factors and a
willingness to pay value of $0.5 million.

128

0 Risk of Contracting Cancer Per 100,000

 

  

    
    
       

15“
10"
5‘ 2'

   
 

\\‘
\\

 

 

‘~‘\ “194111)
:Fi:at\§\i1\‘h\\s:ss§

:\\

,5

3.0 7.0 8.0 9.97 13 18 24 26 48.2
Fish Body Burden of Dioxin (ppt)

- FDA MDPH

 
 

 

Figure 6.7 The risk of contracting cancer for a population dioxin contaminated ﬁsh
based for FDA, MDPH and MDNR cancer potency factors and a
willingness to pay value of $1.4 million.

5 Risk of Contracting Cancer Per 100,000

 

      
       
       
 

 
 

W

22 2

       
 
      
       
 

\

 

xx
‘\‘1\

    
 

. \

:\
\

   

     

\\\
NR

\\

 

 

T

0.3 0.8 1.0 1.2 1.6 2.0 3.0 3155 6.4
Fish Body Burden of Dioxin (ppt)

- FDA

 

Figure 6.8 The risk of contracting cancer for a population dioxin contaminated ﬁsh
based for FDA, MDPH and MDNR cancer potency factors and a
willingness to pay value of $6.8 million.

129

One trend that is particularly evident from Figures 6.6-6.8 is that the MDNR
risk of contracting cancer values far exceeds the same values for MDPH and FDA.
This is directly related to the cancer potency factor, which for the MDNR (1.51 x
105) is an order of magnitude less than either MDPH (3.57 x 10“) or FDA (1.75 x
10“).

Only ﬁve of the policy combinations provide the l x 105 margin of safety used
by the MDPH, MDNR and FDA in their dioxin regulatory programs. An evaluation
of the ﬁsh dioxin body burdens and the cancer risks levels enables the range of
potential dioxin discharge concentrations to be narrowed to only those that meet the
objectives of Michigan’s dioxin policy: (1) be below the 10 ppt MDPH health
advisory, and (2) not exceed a cancer risk of one excess case of cancer per 100,000
exposed individuals.

The following ﬁve Pareto optimal dioxin discharge concentrations meet both
criteria: (1) 0.5 ppq for MDPH, (2) 0.2 ppq for MDNR, and (3) 0.7 ppq for FDA all
with a bioaccumulation factor of 7,238 and a willingness to pay value of $6.8 million;
and (4) 0.2 ppq, for MDPH, and (5) 0.3 ppq for FDA with a 51,600 bioconcentration
factor and a willingness to pay value of $6.8 million. In effect, based on the
assumptions provided in Chapter 5, these dioxin efﬂuent concentrations would provide
the range of Pareto optimal solutions. Each of the dioxin efﬂuent concentrations are
tenfold above the current 0.022 ppq MDNR efﬂuent limit for the Mead pulp mill.

If the second criteria is relaxed to accommodate cancer risks below 2 excess

cases of cancer per 100,000 exposed individuals, the number of Pareto optimal dioxin

130

discharge concentrations increases by ﬁve to include: 6.0 ppq for FDA with a 7,238
bioaccumulation and a willingness to pay of $1.4 million; 3.5 ppq for MDPH with a
7,238 bioaccumulation and a willingness to pay of $0.5 million; 0.08 ppq for MDNR
with a 51,600 bioconcentration factor and willingness to pay of $6.8 million; 0.1 ppq
for MDPH and 0.2 ppq for FDA with a 140,000 bioaccumulation factor and
willingness to pay of $6.8 million. In examining these additional values only one,
0.08 ppq for MDNR with a 51,600 bioconcentration factor and a willingness to pay of
$6.8 million, comes close to the current 0.022 ppq MDNR efﬂuent limit for dioxin in
the Mead mill discharge.

Returning to Table 6.2, this 0.08 ppq Pareto Optimal level of dioxin discharge
can be further investigated. The total damage costs, which are based on a societal
willingness to pay to reduce a members risk of contracting cancer, associated with
this 0.08 ppq value is $897,949/day. The abatement cost to the Mead pulp mill is
$1,026,210 a day. The mill currently spends $23,585 to reach a 7.5 ppq value for
dioxin in its efﬂuent. Using the same assumptions which provided the 0.08 ppq
efﬂuent limit, a value of 7.5 ppq has the potential for resulting in a ﬁsh dioxin body
burden of 88.9 ppt. The following cancer risks are associated with a 7.5 ppq value
are: 125 x 10'5 (MDNR assumptions), 14 x 105 (FDA assumptions), and 30 x 10’5
(MDPH assumptions). Each of these risks, in particular the 125 x 10’, which is
based on the MDNR risk assessment assumptions, exceeds the 1.3 x 10’5 risk
associated with a 0.08 ppq value.

Clearly under the study assumptions the current dioxin discharge level of 7.5

131

ppq comes no where near the margin of safety which the Michigan dioxin policy
demands. However, the margin of safety provided by a 0.08 ppq efﬂuent value may
be excessive. The resulting potential dioxin ﬁsh body burden is 1.0 ppt which is far
below the 10 ppt MDPH health advisory. This 0.08 ppq level would force the Mead
mill to spend $1,026,210/day to meet a margin of safety for dioxin in ﬁsh which is
ten times less than MDPH requires. It is difﬁcult to imagine the mill operating under
such a scenario. These two efﬂuent extremes of 7.5 ppq and 0.08 ppq suggest that
perhaps a true Pareto optimal solution lies somewhere in this range.

Figure 6.9 is a series of three graphs, one for each of the three
bioconcentration/bioaccumulation factors, which illustrate the Pareto optimal dioxin
efﬂuent concentrations. In comparing the ﬁsh dioxin body burdens from Figure 6.5
and the Pareto optimal level of dioxin in Figure 6.9 two observations stand out.

First, because ﬁsh dioxin body burdens are directly related to efﬂuent
concentration of dioxin the shapes of the bars are similar. Second, there is a
signiﬁcant difference between the $6.8 million willingness to pay values and the $1.4
million and $0.5 million values. This difference is directly related to the value placed
on a statistical life. The value of a statistical life which is higher ($6.8 million)
indicates that people are willing to pay more for more protection. This greater
protection is manifested in a lower Pareto optimal level of dioxin discharge.

After evaluating Tables 6.1 - 6.6 and determining the Pareto optimal levels of
dioxin concentration which meet the Michigan dioxin policy criteria, it is evident that

the MDNR water quality based efﬂuent limit of 0.022 ppq for the Mead pulp mill

132

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

aes . 7.2a-
; /m
:3
‘3 s l—
is
4 C
s / a
U
. /' <
a s
.3
a
:'
é 0 07a 1 's
2 murmurs: re u v (trillion)
act - sum
‘3
5
d
5
.0
'3
2 2
h
= Q E
-
a
O
8 Q
J
<
2 1.
3 wirtlueusas to ear (Hutton!)
act . 140.000
—1
‘o. P
8
3
9
‘

 

 

 

 

 

 

 

oh 1 's s.'a
ulumeuns TO MY (U1111000)

earnest aroma (nuns! mu ("4)

Figure 6.9 Comparison of MDPH, MDNR, FDA, Pareto optimal dioxin efﬂuent
levels based on willingness to pay values and bioconcentration factors.

133

does not fall into the range of possible Pareto optimal solutions.

Evaluation of Michigan’s Dioxin Policy The current MDNR National Pollutant
Discharge Elimination System permit authorizes the Mead pulp mill to meet a water
quality based efﬂuent limit of 0.022 ppq of dioxin from June, 1992 until the permit
expiries in 1994. Until June, 1992, any discharge of dioxin at or above the level of
detection, which currently ranges 3-4 ppq, is considered in violation of the permit
(Zugger, 1990). In the absence of a match between a Pareto optimal level of dioxin
discharge and the MDNR efﬂuent limit of 0.022 ppq, the policy analysis model can
be used to evaluate the speciﬁc MDNR dioxin efﬂuent limit.

The potential costs associated with meeting an efﬂuent of 0.02 ppq for the
Mead pulp mill is $3,242,995/day. The damage costs for a 0.02 ppq dioxin efﬂuent
concentration, based on the MDNR bioconcentration factor of 51,600, are presented

in Table 6.7

134

Table 6.7 Damage Costs per day and associated cancer risks,based on
MDPH, MDNR and FDA cancer potencies and a BCF of
51,600, for a dioxin efﬂuent limit of 0.02 ppq.

 

 

 

 

Damage Cost MDPH MDNR FDA
Based on a:

$6.8 million WTP $ 53,428 $225,984 $ 26,190
$1.4 million WTP $ 1,058 $ 4,475 $ 519
$0.5 million WTP $ 1,358 $ 5,744 $ 666

 

Delta County ﬁsh
consumer 0.0065/8,242 0.027/8,242 0.00317/8,242

risk level

 

General risks

 

per 100,000 .08 0.3 .04

 

per 1,000,000 0.8 3.0 0.4

 

per 10,000,000 8.0 30 4.0

 

 

 

 

 

 

135
At ﬁrst glance it appears that the abatement cost incurred by the Mead mill to

meet a 0.02 ppq dioxin level far exceeds any of the potential damage costs identiﬁed
in Table 6.7. In fact, it is 14 times greater than the highest potential damage cost
($225 ,984), which is based on an MDNR risk assessment assumptions and a $6.8
million willingness to pay, and 6,249 times higher than the lowest potential damage
cots ($519), based on FDA risk assessment assumptions and a $1.4 million
willingness to pay value. However, if the corresponding cancer risk is signiﬁcantly
high then these costs may well be warranted. In the case of the 0.02 ppq dioxin
discharge level, however, the cancer risks are far below the 1 x 105 risk level,
ranging from a low of 0.4 x 10‘ to 3.0 x 10‘, which indicates that perhaps that the
0.022 ppq value is excessive.

Fish body burdens of dioxin that correspond with a dioxin efﬂuent limit of
0.02 ppq are: 0.033 ppt for the 7,238 bioaccumulation factor, 0.237 for the 51,600
bioconcentration factor, and 0.643 for the 140,000 bioaccumulation factor. These are
two to three orders of magnitude below the health advisory trigger. This standard
exceeds the 10 ppt MDPH health advisory trigger by a signiﬁcant margin placing the
potential burden of a $3,242,995/day total abatement cost on the mill. Table 6.3
(which is based on the 140,000 bioaccumulation factor, MDNR cancer potency factor,
and a willingness to pay of $6. 8 million) indicates that the greatest amount people
would be willing to pay to reduce associated risks for dioxin is 0.05 ppq, based on
the ratio of total abatement cost to willingness to pay. This level indicates that

individuals are willing to accept a risk of contracting cancer greater than 1 x 105

136

(Table 6.3). The corresponding abatement cost for a 0.05 ppq dioxin efﬂuent
discharge ($1,515,849, Table 6.3) is over 50 percent less that the 0.02 ppq level.

This evaluation addresses the issues of meeting the MDPH health advisory
trigger and the agency cancer risk level, and provides a bases for addressing the
researchable question posed in Chapter 1: do the private costs to the Mead
Corporation’s Escanaba kraft pulp mill, incurred as a result of the MDNR’s dioxin
efﬂuent limit, exceed the beneﬁts of the reduced risk of contracting cancer, which are
based on the MDPH and MDNR risk management approaches, received by the Delta
County sport ﬁsh eating population?

Based on the above analysis it can be stated that the permitted MDNR water
quality based efﬂuent limit is not Pareto optimal; the private costs to the Mead
pulp mill, incurred as a result of the MDNR regulations, exceeds the benefits to
the Delta County sport fish consuming population of the reduced risk of
contracting cancer from dioxin contaminated fish. The MDNR efﬂuent limit
places an undue burden upon the mill at a cost that society is unwilling to pay.

Table 6.7 also illustrates the range of cancer risks using the MDNR
bioconcentration factor for the MDNR, MDPH, and FDA cancer potency factors. All
three potency factors indicate that the level of protection is a magnitude below that
which the MDNR and MDPH base their policies. As indicated by the results, a dual
formulation of policy is not efﬁcient and affords more protection than society is
willing to pay based on the 10 ppt health advisory trigger.

As indicated in Chapter 3, there is a difference between the purpose of the

137
values derived by MDPH and MDNR dioxin policies. The MDPH health advisory

trigger is based on communicating the risk of consuming dioxin contaminated ﬁsh to
potential consumers. The MDNR water quality based efﬂuent limit is designed to
prevent future harm (Sills, 1990). Until the risks of dioxin exposure are fully
understood it is difﬁcult to determine the level of control for dioxin sources.
Although the MDPH and the MDNR are working under different regulatory
philosophies, societal willingness to pay indicates that the current MDNR efﬂuent

limit for dioxin of 0.022 ppq is excessive.

CHAPTER SEVEN: SUNHVIARY AND CONCLUSIONS

The ﬁnal chapter of this thesis presents a summary of the important issues and
ﬁndings, offers conclusions for the development of persistent toxic substance policy

and suggests additional research.

7.] Summary

As pointed out in Chapter 1, a model is merely an abstraction of reality. The
information that it provides is only as good as the information used in its
development. In this research a simple model linking the fate of a persistent toxic
chemical (dioxin) with potential economic impacts was developed in an effort to
provide risk managers with information concerning the economic impacts of the
persistent toxic chemical policies that they develop. The general objective of this
study has been to shed some light on the risk assessment/management policy
development process and, in particular, to illustrate that environmental standards,
regardless of how they are developed, have real economic impacts on society.

To address this general thesis a speciﬁc objective was to develop a policy
analysis model that could be used to evaluate toxic substance regulation in Michigan
by identifying the private costs to industry, and the beneﬁts to society of decreased
risk to adverse effects of toxic contaminated surface water, through the determination
of a range of Pareto optimal solutions.

In Michigan, the goal for persistent toxic chemical policy is to protect the

138

139

public health, safety, and welfare from exposure to toxic compounds. Depending on
particular state agency mandates this goal can be expanded to include the protection
of plant and animal life as well as the designated uses of a water body from the
presence of toxic substances. This research focused speciﬁcally on the persistent
toxic chemical 2,3,7,8-TCDD (referred to as dioxin) and the policies designed by the
Departments of Natural Resources and Public Health to protect the citizens of
Michigan from exposure.

To evaluate this dual agency approach the following researchable question
was posed: do the private costs to the Mead Corporation’s Escanaba kraft pulp
mill, incurred as a result of the Michigan Department of Natural Resource’s
dioxin efﬂuent limit, exceed the benefits of the reduced risk of contracting
cancer, which are based on MDPH and MDNR risk management approaches,
received by the Delta County sport fish eating population?

Before this question could be addressed a general presentation of the sources
of dioxin and its chemical characteristics was provided. Since the initiation of the
National Dioxin Strategy the generation of dioxin has been associated with a number
of sources, including: chemical production, municipal incinerators, smelting
processes, automobile exhaust and the bleaching sequence in kraft pulp mill
operations. Persistence and bioaccumulation are the two primary properties of dioxin
that increase the hazard potential to humans and the environment. Combined, these
factors pose a signiﬁcant threat to Great Lakes sport ﬁsherpersons who consume their

catch.

140

The health effects of dioxin were brieﬂy examined with the general conclusion
that due to a number of deﬁciencies in current epidemiological studies, no speciﬁc
cause and effect relationships can be determined for dioxin exposure and damage to
human health. However, a number of animal studies have linked dioxin exposure
with the development of cancer. As most regulatory agencies assume that all animal
carcinogens are human carcinogens, for regulatory purposes, human exposure to
dioxin is assumed to pose the potential for cancer development in humans.

The regulatory policies for dioxin were examined for four agencies: the
Environmental Protection Agency, the Food and Drug Administration, the Michigan
Department of Natural Resources, and the Michigan Department of Public Health.
Each agency was identiﬁed has using a risk assessment /management process for
developing persistent toxic substance policy. The following three speciﬁc similarities
are shared by the agencies in their risk assessment policies: (1) the use of the Kociba
(1978) cancer rat study, (2) a l x 10" cancer risk, and (3) a linearized multistage
model for extrapolation. However, a number of inconsistencies also exist between the
agencies in their policy development approaches. The following list summarizes the
differences in agency risk assessment/management processes:
type of tumors counted in the Kociba study;
potency factors;

species conversion factor; and
regulatory mandate.

PPM“

The environmental protection agencies, the EPA and MDNR, use risk
assessment\management to develop efﬂuent limits and technology based standards to

control the amount of dioxin which is discharged by a polluting ﬁrm. The health

141

protection agencies use risk assessment/management to provide consumption advice
concerning either the interstate commerce of ﬁsh (FDA) or the in state consumption
of sport caught ﬁsh (MDPH).

In Michigan, the MDNR has developed a water quality based efﬂuent limit for
dioxin of 0.01 ppq for new mills sited tributary to the Great Lakes and 0.022 ppq for
the Mead pulp mill in Escanaba, as part of an EPA authorized state National Pollutant
Discharge Elimination System. The efﬂuent limit communicates to the pulp mill that
discharges above the limit will not be tolerated. The MDPH has set a health advisory
trigger for dioxin of 10 ppt in ﬁsh. The health advisory trigger communicates to
ﬁsherpersons that the? is potentially a substantial risk involved in eating ﬁsh above
the health advisory trigger.

To address the researchable question of determining optimal mill costs and
societal beneﬁts two issues had to be addressed. The ﬁrst issue was whether the level
of dioxin permitted into the environment, by water quality-based efﬂuent limits, when
bioaccumulated by ﬁsh, provides the same margin of safety as the MDPH health
advisory trigger. A second issue necessary in determining whether the costs to the
Mead mill exceed the beneﬁts to society (referred to as searching for a Pareto optimal
solution), was to determine the value of damage caused when organisms are exposed
to dioxin in the environment.

The policy model designed in Chapter ﬁve was used to address both of these
issues. The foundation for the policy model is the traditional pollution control model,

as explained in Chapter 4, which seeks to maximize social welfare and minimize

142

costs. The model depends on the development of a damage function, abatement
function, and a total cost of pollution, which is the sum of the damage and abatement
functions.

A series of six equations were used to link efﬂuent concentration of dioxin to a
speciﬁc point on a damage cost curve. The dioxin efﬂuent concentration was
translated to a concentration of dioxin in the river. Through the use of
bioaccumulation factors the river concentration of dioxin was transformed to a
concentration of dioxin in ﬁsh. The value of dioxin in ﬁsh was then used as a human
dose of dioxin and through the use of the risk equation: risk = potency x dose, was
translated to a risk of contracting cancer as a result of dioxin exposure. The general
cancer risk was multiplied by a value of the sport ﬁsh consuming population in
Michigan to derive a speciﬁc cancer risk for Delta County sport ﬁsh eaters. And,
ﬁnally, this speciﬁc cancer risk was multiplied by a willingness to pay value to reduce
a member of society’s risk of death from avoidable behavior.

The total costs per day of three abatement technologies scheduled for addition
to the Mead pulping and bleaching process were plotted against the associated
decrease in dioxin discharge, and a power function was ﬁtted to these values in an
attempt to derive an extended pollution abatement cost curve. The abatement cost
curve was added to twenty-seven different damage costs curves and the resulting
minimum values were compared to the current dual agency dioxin policy to determine

whether the Michigan policy was Pareto optimal.

Five dioxin discharges which met the policy objectives of having associated

143

dioxin ﬁsh burdens below 10 ppt and cancer risk levels below 1 x 10?5 were identiﬁed
as Pareto optimal: (l) 0.5 ppq for MDPH, (2) 0.2 ppq for MDNR, and (3) 0.7 ppq
for FDA all with bioaccumulation factor of 7,238 llkg and a willingness to pay value
of $6.8 million; and (4) 0.2 ppq, for MDPH, and (5) 0.3 ppq for FDA with a 51,600
bioconcentration factor and willingness to pay value of $6.8 million. Each of these
dioxin efﬂuent concentrations are tenfold above the current 0.022 ppq MDNR efﬂuent
limit for the Mead pulp mill.

A subsequent analysis of the 0.022 ppq efﬂuent limit indicated a cancer risk
range from 0.4 x 10‘5 to 3.0 x 10“, associated ﬁsh body burdens of dioxin of 0.033
ppt (7 .238 bioaccumulation factor), 0.237 ppt (51,600 bioconcentration factor), and
0.643 ppt (140,000 bioconcentration factor), potential abatement costs of
$3,242,995/day and damage costs range from a low of $519 to $225 ,984/day. Based
on the lack of a match with a Pareto optimal solution, as well as the ﬁsh body burden
of dioxin and the cancer risk values corresponding with a 0.022 ppq dioxin efﬂuent
limit, it was concluded that the permitted MDNR water quality based efﬂuent limit
is not Pareto optimal; the private costs to the Mead pulp mill, incurred as a
result of the MDNR regulations, exceeds the benefits to the Delta County sport
fish consuming population of the reduced risk of contracting cancer from dioxin

contaminated fish.

144

7.2 Conclusions

The conclusions are addressed in the following ﬁve categories: (1) potential
damage caused by persistent toxic substances, (2) using risk assessment/ management
to develop persistent toxic substance policy, (3) damage costs associated with

persistent toxic substance policy, (4) policy implications

Potential Damage Caused by Persistent Toxic Substances This research addressed one

speciﬁc potential damage related to the discharge of dioxin in the environment: the
potential of a speciﬁc population contracting cancer. There are, however, a number
of other potential damages associated with the discharge of a persistent toxic
substance such as dioxin into the environment. The same laboratory data that is used
to extrapolate risks to humans can be used to extrapolate risks to animals which live
in environments contaminated by dioxin or other persistent toxic substances. Further,
as was earlier indicated, studies have been conducted which indicate potential links
between health hazards to various animal species and dioxin exposure.

Since dioxin is persistent there is potential for damage to animals in the
immediate vicinity of a discharge as well as downstream or perhaps elsewhere within
the Great Lakes ecosystem. The bioaccumulation characteristic of dioxin provides a
pathway for dioxin to contaminate species in the aquatic community as well as the
terrestrial community. Fish can bioaccumulate dioxin, and predators both in aquatic
and terrestrial environments can be exposed to dioxin by consuming contaminated

ﬁsh. This research did not account for potential dioxin damage beyond that of

145

contracting caner by humans due to dioxin exposure.

Using Risk Assessment/Management for Policy Development The risk
assessment/management process is based on a number of variables including: the
determination of whether or not a particular chemical is causally linked to a particular
health effect, the determination of the relation between the magnitude of exposure
and the probability of occurrence of the proposed effects, and the estimation of the
number of people who will be exposed and the characteristics of the exposure before
and after application of regulatory controls.

For each of these variables risk managers must assess the appropriateness of
data provided from industry, public interest groups, the scientiﬁc community and the
regulatory community. The submission of new data from any reputable source
requires thorough analysis and justiﬁcation of the assumptions used by risk managers
before regulatory controls can be adjusted. In a process wrought by uncertainty this
can be a herculean task. However, the regulatory standards that are based on these
data do provide real environmental and economic impacts and deserve intensive
scrutiny.

A Michigan newspaper editorial exempliﬁes the task before the Michigan risk
managers. In response to the release of reanalysis of dioxin dose response data, the
Detroit News (10 June 1990) wrote:

The " good news" about dioxin comes from many sources, but most notable is

Robert Squire, the John Hopkins University scientist who analyzed rodent

studies on the poison for the Environmental Protection Agency (EPA) in the
19705. He wrote the report on which today’s excessively stringent standards

146

are based. Newer technology, he says, has shown that the safe daily dosage

he originally calculated for humans could be increased as much as 30 times.

”I do not believe that dioxin poses a cancer risk to humans at any anticipated

levels of exposure, " Mr Squire said in a devastating letter to EPA.
Based on the information provided by the Squire letter the Detroit News editorial
board concluded:

The evidence is piling up. The scientists were wrong and the dioxin scare was

another environmental ripoff. Michigan’s standards should be cut drastically

to reﬂect the real risks of this ”weak carcinogen".

If additional evidence provides support for Squire’s conclusion then the Detroit
News editorial board’s request may prove necessary. However, if risk managers
change current regulatory standards based on the assertion of one scientist they would
not be adequately protecting the health, welfare, and safety of society. Current
standards are based on one particular dioxin rat study because the study is "generally
regarded as the most biologically defensible of the low dose extrapolation models"
(MDPH, 1986 p.6. It has withstood the test of time. In time, perhaps the Squire re-
evaluation will prove to be the most "biologically defensible” .

Another matter that complicates the risk manager’s policy making is deﬁning '
" how safe is safe" . Persistent toxic substance policy is based on an acceptable level
of risk. This acceptable level of risk is the value the risk manager takes to reﬂect the
society’s acceptable level of protection. In the case of MDNR’s Rule 5 7 an expert
panel deﬁned acceptable risk to be one excess case of cancer per 100,000 exposed
individuals. Once the risk level is established the risk manager’s concerns are shifted

to other facets of the risk assessment/ management process. However, individuals are

left to contend with the impacts of the resulting regulatory polices.

147

People have different risk tolerances. One person may only be willing to
accept a one excess case of cancer per one billion exposed individuals. No matter
how unreasonable this risk seems to the risk manager an individual’s perception of
risk deﬁnes an individual’s reality. For an individual only willing to accept a 1 x 10'7
risk, accepting any greater could make life unbearable. Conversely, another
individual may be willing to accept a 1 x 10" risk and be bothered by a health
advisory trigger which recommends limiting ﬁsh consumption.

The risks that people accept are also related to whether the risk is voluntary or
involuntary. In the case of dioxin tainted ﬁsh, a ﬁsherperson may be greatly affected
by the closure of a ﬁshery due to high levels of dioxin while a non ﬁsherperson may
not be affected at all. A general societal solution is to reduce the risk of exposure to
persistent toxic substances by reducing the release of the compound in the
environment, thereby reducing the potential for exposure. In this manner the
voluntary risk of eating sportﬁsh would be reduced. This is precisely the aim of the
efﬂuent limit.

While science must drive the risk assessment/management process, public
perception and political/economic climate will continue to inﬂuence policy outcomes.
The policy model developed in this research allows risk managers to identify potential

economic impacts of risk management options and as a result be better equipped to

predict reactions to risk assessment/management based environmental standards.

148
Cgsts Associated with Persistent Toxic Substances Policy Economist suggest that zero

discharge requires zero economic activity. However, consumer behavior can be
altered to demand a product whose production process does not damage the
environment to the same severity of current production processes. For example, if
consumers preferred near-white paper to bleached white paper there would be no
cause for dioxin concern from pulp mills. The near white paper could be produced
by a process that does not require chlorine bleaching, the main cause of dioxin in pulp
production.

In addition to the cost to society of contracting cancer, and the potential cost
associated with the loss of species, there is the economic costs related to a closed
ﬁshery. If a ﬁshery is closed, demand for related goods and services such as gas,
ﬁshing gear, and food may decline. While these costs are related to not meeting
environmental standards there are also potential costs associated with meeting
standards. For example, if consumer preference remains steadfast to bleached white
paper, abatement costs, incurred as a result of meeting environmental standards, may
increase to a level that is prohibitive of producing paper. This would result in
damage costs of a different sort, that of the community economic costs associated
with a mill shutdown such as the loss of a percentage of the tax base or lost

employment opportunities.

149
Polioy Implications Each Michigan agency is bestowed with a speciﬁc protection

mandate. The MDNR has the responsibility to protect the natural resources of the
state from pollution, impairment or destruction. The MDPH has the responsibility of
protecting public health. Although these are fairly these broad mandates, basing
dioxin policy solely on efﬂuent limits and health triggers restricts state policy options.
The MDNR and MDPH programs provide a dioxin policy which is based speciﬁcally
on limiting exposure through end of the discharge pipe limitations (efﬂuent limits) and
as secondary protection, through ﬁsh consumption advise. Given the results of the
policy analysis model indicate the MDNR standards may be excessive, especially
when compared to MDPH health advisory triggers, there are alternative policy
recourses other than simply raising the efﬂuent limit.

Instituting educational program on toxic substance is one policy option. The
jury is still out on the human health effects of dioxin, yet public perception that it
presents a signiﬁcant danger to humans remains widespread. To address this
perception Michigan’s regulatory agencies could embark on an education campaign.
Two major issues could be stressed.

First, the risks associated with dioxin exposure should be adequately
communicated. Public perception is driven by the news media which provides
conﬂicting information. For example, one month journalists report, "Fishing
downstream of a paper mill can be dangerous to your health (Port Huron Times
Herald, 1 May 1990)," and the next, "Michigan’s standards should be cut drastically

to reﬂect the real risks of this ’weak carcinogen’ (Detroit News, 10 June 1990)."

150

Second, if dioxin continues to be a major concern the then society must be
provided with information concerning viable alternatives to bleached white paper. Of
course, as a State sponsored activity the Department of Commerce may have a
difﬁcult time excepting such an educational objective as decreasing economic
opportunity in the state.

At a minimum, consistency in risk assumption/management processes would
provide consistent information to society. Attempts at intrastate coordination of risk
assessment/management approaches, such as the Michigan Council of Environmental
Quality risk assessment guidelines, are needed to ensure that scarce state ﬁnancial
resources are more efﬁciently allocated. Similarly, the continuation of interstate
attempts at coordinating risk assessment/management approaches to developing
persistent toxic substance policy is also needed. The EPA Great lakes Water Quality
Initiative and the Great Lakes Fish Consumption Advisory Task Force are two
examples of this type of coordination (USEPA, 1989; Hesse, 1990). Inconsistencies
among state approaches can fuel interstate competition for pulp mill development.
The result of this competition could ultimately be environmental degradation rather
than the coveted and desired economic development. The EPA recently approved
Maryland’s 1.2 ppq efﬂuent standard for dioxin (Inside EPA, 1990). Echoing the
uncertainty addressed in this research, an EPA ofﬁcial commented that "a standard
one hundred times weaker than EPA’s ’isn’t surprising given the degree of dispute
within the scientiﬁc community’ about the potency of dioxin (Inside EPA, 1990).

This decision could have a signiﬁcant impact on where pulp mills locate. Although a

151

readily available source of wood ﬁber is critical to pulp mill siting, a sufﬁciently
lower standard in one state, given the same wood ﬁber supply, could entice a pulp
mill to relocate.

The policy analysis model presented in this research provides a framework by
which a number 4 of variables can be linked and analyzed to determine the economic
impacts of regulatory approaches. These variables include: the ﬂow of receiving
water, the amount of ﬁsh consumed, the weight of a ﬁsh consumer, cancer potency
factors, bioaccumulation factors/bioconcentration factors, site speciﬁc willingness to

pay values, and variable abatement costs.

7. 3 Recommendation for Future Research

Although there are a tremendous amount of research possibilities which could
improve the general understanding of dioxin, of critical importance to policy makers
is that of determining what the health consequences are of human exposure to dioxin.
If dioxin is determined to no longer be as toxic as it once was thought to be, perhaps
current efﬂuent standards could be relaxed. However, as long as there is a signiﬁcant
degree of uncertainty, prudent policy dictates a conservative standard. If a standard is
relaxed under uncertainty only time will prove whether the decision to relax the
standard was appropriate.

Other research possibilities include:

(1) The transport and fate of dioxin in aquatic and terrestrial ecosystems could

be further explored. A better understanding of dioxin pathways in the environment

152

could provide information that would allow economists to utilize valuation techniques
to attempt to quantify damage costs. $

(2) The use of non-market valuation techniques such as willingness to pay is
still considered quite controversial among economists. Continued research into the
validation of non-market valuation techniques could ensure the appropriateness of
using these techniques to evaluate policy.

(3) Studies have been conducted in Michigan to assess the attitudes of
ﬁsherpersons to the presence of toxic substances in Michigan rivers (U dd, 1985 and
Smith and Enger, 1988). However, speciﬁc studies to assess what risk level the ﬁsh
consuming public would be willing to pay for have yet to be conducted. To get a
better understanding of the Delta County ﬁsh consuming public’s attitudes studies
could be performed to determine the County population’s willingness to accept risk
and to pay for risk.

(4) Research into cost effective non dioxin generating technologies would
provide pulp mills alternatives in selecting abatement strategies.

(5) Behavioral research may provide insight into individual preferences and
result in educational programs that empower individuals to make choices based on
health considerations rather than only aesthetic considerations.

The variables evaluated in this research, cancer potencies, bioconcentration
factors, and willingness to pay, are plagued with uncertainty and at times are
. controversial. Instead of providing a speciﬁc defensible policy endpoint, this research

provides a framework in which policy makers can evaluated a number of

153

combinations of potential policy variables. Through careful evaluation of
scientiﬁcally defensible “variables the risk manager can predict potential outcomes and
develop programs that ensure the proper communication to society of policy

objectives.

BIBLIOGRAPHY

Aartila, D. 1990. Wildlife Biologist, District Ofﬁce, Michigan Department of
Natural Resources, Escanaba, MI. Personal Communication.

Abbott, R. 1990a. Environmental Engineer, Mead Corporation, Publishing Paper
Division, Escanaba, MI. Personnel Communication.

Abbott, R. 1990b. Draft Dioxin Minimization Program. Mead Corporation,
Publishing Paper Division, Escanaba, MI.

Allen, J.R., D.A. Barsotti, J .P. Van Miller, L.J,. Abrahamson, and J .J . Lalich.
1977. Morphological changes in monkeys consuming a diet containing low levels of
2,3,7,8—tetrachlorodibenzo-p-dioxin. Food and Cosmetic Toxicology 15:401-410.

Anderson, R, W.R. Aslop, D.G. Galya. 1990. Comments on the Michigan Water
Resource Commission’s Proposed Dioxin Efﬂuent Limit for Mead Escanaba,
Michigan Mill. ENSR Consulting and Engineering.

Barnes, D.G. 1985. Recent USEPA actions related to dioxins. Chemosphere,
14(6/7):755-766.

Bedford, J .W., H.E.B. Humphrey, G.K. Hurlburt, C.N. Park, M. Van Putten, D.
Wade, K. Wurzel. 1990. Draft State of Michigan Risk Assessment Guidelines.
Council on Environmental Quality, Iansing, MI. 22 pp.

Bennett, EB. 1978. Water budgets for Lake Superior and Whiteﬁsh Bay. Journal of
Great Lakes Research 4:3-4 pp. 331-342.

Berger, M.C., G.C. Blomquist, D. Kenkel, G.S. Tolley. 1985. Valuing changes in
health risks: a comparison of alternative measures. Southern Economic Journal
56(4):967-984.

Blomquist, G. 1979. Value of Life Savings: Implications of Consumption Activity.
Journal of Political Economy 87(3):540-558.

Bonsor, N ., N. McCubbin, and J Sprague. 1989. Municipal-industrial strategy for
abatement (MISA) kraft mill efﬂuents in Ontario. Prepared for the Technical
Advisory Committee Pulp and Paper Sector of MISA, Queen’s Printer for Ontario.
260 pp.

Boyer, 1.]. 1989. Bioavailability of ingested 2,3,7 ,8-TCDD and related Substances

(Draft). Prepared for EPAs Working Group on the Bioavailability of Dioxins in
Paper Products.

154

155

Bro, K.M., W.C. Sonzogni, and ME. Hanson. 1987. Relative cancer risks of
chemical contaminants in the Great Lakes. Environmental Management 11(4):495-
505.

Center for Disease Control. 1983. Detailed responses to subcommittee questions on
dioxin. Center for Environmental Health, Washington, DC.

Clark, J .M., L. Fink and D. DeVault. 1987. A new approach for the establishment
of ﬁsh consumption advisories. Journal of Great Lakes Research 13(3):367-374.

Cook. P. 1990. Alternatives for Predicting TCDD Bioaccumulation: The Lake
Ontario Experience. Paper presented at Dioxin and PCBs: National Conference on
Approaches to Address Human Health Risks and Aquatic Life Impacts. Washington,
DC.

Cordle, F. 1981. The use of epidemiology in the regulation of dioxins in the food
supply. Regulatory Toxicology and Pharmocology 12379-387.

Dai, L.C., N.T.N. Phuong, L.H. Thom, T.T. Nguyen, T.Van, L.H. Cam, N. T.
Thu, H.K. Chi, L.B.Thuy, and H.D. Can. 1989. A survey of infant mortality rate
in areas sprayed by phenoxyherbicides contaminated with 2378-TCDD. Poster
presented at the Ninth International Symposium on Chlorinated Dioxins and Related
Compounds, Toronto.

Dewees, D.N., C.K. Everson, and W.A. Sims. 1975. Economic Analysis of
Environmental Policies, University of Toronto Press, Toronto, Canada, 175 pp.

Dioxin: Oops, Never Mind. Detroit News 10 June 1990, Editorial, p. B2.

Eaton, BC, and DP Eaton. Microeconomics. W.H. Freeman and Company, New
York. 471 pp.

Elliott, J.E., R.J. Norstrom, P.E. Whitehead and R.W. Butler. 1989. PCDDs,
PCDFs and reproductive success of ﬁsh-eating birds from British Columbia. Paper
presented at the Ninth International Symposium on Chlorinated Dioxins and Related
Compounds, Toronto.

Fales, G. 1988. A progress report of PAPRICAN Dioxin Research, Paper Age,
December, 1988. p. 20.

Fisher, A., L.G. Chestnut, D.M. Violette. 1989. The Value of Reducing Risks of
Death: A note on New Evidence. Journal of Policy Analysis and Management,
8(1):88-100.

156

Food and Drug Administration. 1981. FDA Review of Dioxin Contaminated Fish.
Department of Health and Human Services, Rockville, MD.

Foran, J.A., D. VanderPloeg. 1989. Consumption Advisories for Spot Fish in the
Great lakes Basin: Jurisdictional Inconsistencies. Journal of Great Lakes Research
15(3):476-485.

Freeman, A.M., R.H. Haveman, A.V. Kneese. 1973. The Economic of
Environmental Policy. John Wiley & Sons, Inc., Santa Barbara, CA.

Gegax, D., S. Gerking, and W. Schulze. 1985. Perceived Risk and the Marginal
Value of Safety. Working paper prepared for the USEPA.

Goldfarb, W. 1988. Water Law, 2nd edition. Lewis Publishers, Chelsea, Michigan.
284 pp.

Greenpeace rallies outside paper mill. Port Huron Times Harold, 8 April 1990 p. B2.

Haimes, Y.Y. 1990. Risk of extreme events. Luncheon address, Pesticides in the
Next Decade: The Challenges Ahead, November 8-9, Richmond, VA.

Harpster, LS. and S.W. Jones (eds.). 1989. How James River is confronting the
dioxin issue. Onstream 8(9):5.

Hesse, J .L. 1990. Summary and Analysis of Existing Sportﬁsh Consumption
Advisory Programs in the Great Lakes Basin. Michigan Department of Health,
Lansing, MI. 14 pp.

Hile, J.P., Associate Commissioner for Regulatory Affairs, U.S. Department of
Health & Human Services. Letter to Mark Van Putten, National Wildlife Service,
August 14, 1985.

Hoehn, J.P. and D. Krieger. 1986. Review of techniques for weighing or valuing
the beneﬁts associated with environmental quality improvements - economics.
Contract #OSE 84-00321, 163 pp.

Hoppe, P. 1990. Plant Manager, James River Corporation, Port Huron, MI.
Personnal Communication.

Hufschmidt, M.M., D.E. James, A.D. Meister, B.T. Bower, J .A. Dioxin. 1989.
Environment, natural systems, and development: an economic value guide. The
Johns Hopkins University Press, Baltimore, MD.

157

Inside EPA. 1990. EPA okays state dioxin standard for water signiﬁcantly weaker
than own. An inside Washington Publication 11(38): 1.

International Joint Commission, 1988, Pulp and Paper Sector in the Great Lakes
Basin, Draft Report of Water Quality Board. 12 pp.

Isensee, A.R. and GE. Jones. 1975. Distribution of TCDD in aquatic organisms.
Environmental Science and Technology 9:688-672.

Kaczmar, S.W., Matthew J. Zabik, and Frank M. D’Itri. 1985. Part per trillion
residues of 2,3,7,8-tetrachlorodibenzo-p-dioxin in Michigan Fish. In Chlorinated
Dioxins and Dibenzofurans in the Total Environment 11, Keith H. Lawerence,
Christoffer Rappe, and Gangadhar Choudhary (eds.), Boston: Butterworth Publishers,
pp. 103-110.

Kagel, R.O. Undated. Analytical problems with plant efﬂuents. Dow Chemical
Company, Midland, MI.

Kociba, R.J., Keeler, P.A., Park, C.N., and Gehring, R]. 1976. 2,3,7,8-
Tetrachlorodibenzo-p-dioxin (TCDD): Results of a l3-week oral toxicity study in rats.
Toxicology and Applied Pharmocology 35:553-574.

Kociba, R.J., D.G. Keyes, J .E. Beyer, R.M. Carreon, C.E. Wade, D.A. Henber,
R.P. Kalnins, L.E. Frauson, C.N. Park, S.D. Barnard, R.A. Hummel, and CG.
Humiston. 197 8. Results of a two-year chronic toxicity and oncogenicity study of
2,3,7 , 8-tetrachlorodibenzo-p-dioxin in rats. Toxicology and Applied Pharmacology,
46:279-303.

Kuehl, D.W., P.M. Cook, A.R. Batterman, D.L. Lothenbach, and BC. Butterworth.
1987. Bioavailability of polychlorinated dibenzo-p-dioxins and dibenzofurans form
contaminated Wisconsin River sediment to carp. Chemosphere 16:667-679.

Long, C. 1973. The trout streams of Michigan no. 6: the Escanaba River.
Michigan Department of Natural Resources, Fisheries Division, Technical Report: 72-
18, Lansing, MI. '

Metcalf, R.L. and T. Liu. 1973. Environmental distribution and metabolic fate of
key industrial pollutants and pesticides in a model ecosystem. Research Report 69.
Water Resources Center, University of Illinois, Urbana, IL.

Michigan Department of Commerce. 1982. Pulp and paper opportunities in
Michigan. Chas. T. Main, Inc, Boston, MA.

158

Michigan Department of Natural Resources. 1984. Staff Report: Support document
for the proposed Rule 57 package. Environmental Protection Bureau, landing, MI.

Michigan Department of Natural Resources. 1985. Guidelines for Rule 57(2).
Environmental Protection Bureau, Lansing, MI.

Michigan Department of Natural Resources. 1987. Report of an industrial
wastewater survey conducted at Mead Paper Company - Publishing Paper Division.
Point Source Monitoring Section, Environmental Protection Bureau, Lansing, MI.

Michigan Department of Natural Resources. 1988a. Fish Contaminant Monitoring
Program: Menominee River. Lansing, MI.

Michigan Department of Natural Resources. 1988b. Fish Contaminant Monitoring
Program: Tittabawassee River. Lansing, MI.

Michigan Department of Natural Resources. 1989a. Cancer risk assessment
justiﬁcation 2,3,7,8-tetrachlorodibenzo-p—dioxin (TCDD) CAS #1746-01-6. Lansing,
MI.

Michigan Department of Natural Resources. . 1989b. Licensing Division. Personal
Communication.

Michigan Department of Natural Resources. 1989c. Michigan Water Resources
Commission authorization to discharge under the national pollutant discharge
elimination system: Mead Corporation, Publishing Paper Division, Escanaba,
Michigan. Lansing, MI.

Michigan Department of Public Health. 1986. Background Document Concerning
1986 Fish Consumption Advisory For Dioxin Contaminated Fish. Center for
Environmental Health, Lansing, MI. 13 pp.

Michigan Department of Public Health. 1988. Mitigating factors associated with the
US EPA risk assessment relating to ﬁsh contamination in the Midland area. Center
for Environmental Health Sciences, Lansing, MI.

Michigan Department of Public Health. 1989. Fish for 2,3,7,8-TCDD in parts per
trillion. ICHEQ-Investigations Division, Analytical Laboratory Section - Fish
Monitoring Program, Lansing, MI.

Moore, MR. 1984. Extending the economic theory of pollution control to
accumulating pollutants. The Environmental Professional 6:277-282.

159

Morris, William (ed). 1979. The american heritage dictionary of the english
language. Houghton Mifﬂin Company, Boston, MA.

Muir, D.C.G., A.L. Yarechewski, and G.R.B. Webster. 1985. Bioconcentration of
four chlorinated dioxins by rainbow trout and fathead minnows. In: Aquatic
Toxicology and Hazard Assessment: Eight Symposium. R.C. Bahner and DJ.
Harsen, Eds., ASTM STP 891 pp. 440-454.

Muldoon, P. and M. Valiante. 1989. Submission to the International Joint
Commission on behalf of the program for zero discharge. Canadian Institute for
Environmental Law and Policy. 26 pp.

Murray, F.J., Smith, P.A., Nitschke, K.D., Hameston, C.G., Kociba, R.J., and
Scwetz, BA. 1979. Three-generation reproduction study of rats given 2,3,7,8-
tetrachlorodibenzo-p—dioxin (TCDD) in the diet. Toxicology and Applied
Pharmocology, 50:241-252.

National Academy of Science. 1983. Risk assessment in the federal government:
managing the process, National Academy Press, Washington, DC.

Nicholson, W. 1985 . Microeconomic Theory: Basic Principles and Extensions. The
Dryden Press, Chigaco, IL. 768 pp.

NTP. 1982a. Carcinogenesis bioassay of 2,3,7,8-tetrachloro dibenzo-p-dioxin in
Osbome-Mendel rats and B6C3F1 mice (gavage study). National Toxicology
Program Technical Report Series No. 209, Research Triangle Park, NC.

NTP. 1982b. Carcinogenesis bioassay of 2,3,7,8-tetrachloro dibenzo-p-dioxin in
Swiss-Webster mice (dermal study). National Toxicology Program Technical Report
Series No. 201, Research Triangle Park, NC.

Olsen, CA. 1981. An Analysis of Wage Differential Received by Workers on
Dangerous Jobs. The Journal of Human Resources 16(2):167-185.

Paper mill dioxins pose threat EPA: Eating ﬁsh from downstream raises caner risk.
Port Huron Times Herald, 1 May 1990, p. A1.

Pearce, D.W. and R.K. Turner. Economic of natural resources and the
environment.The John Hopkins University Press. , Baltimor, MD. 378 pp.

Pollock, G.A, Y.A. Wieder, I.J. Uhaa, A.M. Fan, R.R. Cook. 1989. Risk
assessment of dioxin contamination of ﬁsh. Hazard Evaluation Section, Ofﬁce of
Environmental Health Hazard Assessment, California Epartment of Health SErvices,
Berkeley, CA.

160

Rainey, RH. 1967. Natural displacement of pollution from the Great Lakes.
Science 155 pp. 1242-1243.

Randall, A. 1987. Resource Economics: An Economic Approach to Natural
Resource And Environmental Policy. John Wiley & Son, New York, NY. 434 pp.

Rappe, C. 1988. Swedish view of the dioxin issue. Vatten 44: 137-144.

Sills, Robert. 1990. Toxicologist, Surface Water Quality Division, Michigan
Department of Natural Resources, Lansing, Michigan. Personal Communication.

Sly, P. 1967. Lake Erie and its basin. Journal of Fishery Resources Board of Canada
33 pp. 355-370.

Smith, V.K., and W.H. Desvousges. 1987. An empirical analysis of the economic
value of risk changes. Journal of Political Economy 95(1):89-111.

Smith, BF. and ED. Enger. 1988. A survey of attitudes and ﬁsh consumption of
anglers on the lower Tittabawasee River, Michigan. Center for Environmental Health
Sciences, Michigan Department of Public Health, Lansing, MI.

Smook, G. 1982. Handbook for pulp & paper technologists.Joint Executive
Committee of the Vocational Education Committees of the Pulp and Paper Industry,
Canada. 395 pp.

Stachiw, N.C., M.E. Zabik, A.M. Booren, and M.J. Zabik. 1988.
Tetrachlorodibenzo-p—dioxin residue reduction through cooking/processing or
restructured carp ﬁllets. Journal of Agricultural Food Chemistry 36:848-852.

Stalling, D.L., Smith, L.M., Petty, J.D., Hogan, J.W., Johnson, J.L., Rappe, C.,
and Buser, H.R. 1983. Residues of polychlorinated dibenzo-p—dioxins and
dibenzofurans in Laurentian Great Lakes Fish. In Human and Environmental Risks of
Chlorinated Dioxins and Related Compounds, R.E. Tucker, A.L. Young, and AP
Gray, eds. New York: Plenum Press, pp.221-240.

Staniec, C.L. 1989. U.S. EPA/Paper Industry Cooperative Dioxin Study Analytical
Results. Memo from Coordinator 104 Mill Study, U.S. Environmental Protection
Agency Region V, to 104 Mill State Contacts, March 16, Chicago, IL.

Taft, W. 1990. Interofﬁce Communication. Surface Water Quality Division,
Michigan Department of Natural Resources, Lansing, MI.

161

Talhelm, Daniel, R. 1990. Recommended values for computing fair restitution to
the citizens on Minnesota for ﬁsh and wildlife illegally killed, injured or possessed.
Resource Econometrics, East Lansing, MI

Tsushimoto, G., F. Matsumura and R. Sago. 1982. Fate of 2,3,7,8-
tetrachlordibenzo-p—dioxin (TCDD) in an outdoor pond and in model aquatic
ecosystems. Environmental Toxicology Chemistry 1:61-68.

Udd, E. 1985. Perceptions and behavioral responses of anglers to the presence of
toxic chemicals in Michigan rivers. Dissertation, Michigan State University. 131 pp.

University of California at Berkely. 31 August 1990 press release concerning article
appearing in Science, August 31, 1990, and three articles in press in the Proceedings
of the National Academy of Science by Bruce Ames and Lois Swirsky Gold.

USEPA. 1984. Ambient water quality criteria for 2,3,7,8-Tetrachlorodibenzo-p-
dioxin. EPA 440/5-84-007, Washington, D.C.

USEPA. 1985. Health assessment document for polychlorinated dibenzo-p-dioxins.
Ofﬁce of Health and Environmental Assessment. EPA/600/8-84/014F, Washington,
D.C.

USEPA. 1988a. U.S. EPA/paper industry cooperative dioxin screening study.
Ofﬁce of Water Regulations, EPA-440/1-88-025 , Washington D.C.

USEPA. 1988b. Interim strategy for the regulation of pulp and paper mill dioxin
discharges to the waters of the United States. Ofﬁce of Water, Washington, D.C.'

USEPA. 1989a. Great Lakes Water Quality Initiative. Washington, D.C. 5 pp.

USEPA. 1989b. Approach to regulation of dioxin discharges from pulp and paper
mills that bleach with chlorine. Regional 5, Chicago, IL.

USEPA. 1990a. Summary of technologies for the control and reduction of
chlorinated organics from the bleaching chemical pulping subcategories of the pulp
and paper industry. Ofﬁce of Water Regulations and Standards, Ofﬁce of Water
Enforcement and Permits, Washington, D.C. 85 pp.

USEPA. 1990b. EPA to regulate dioxin in paper industry. Draft Environmental
News, Public Affairs, Washington, D.C.

162

USEPA. 1990c. Risk Assessment for 2378-TCDD and 2378-TCDF Contaminated
Receiving Waters From U.S. Chlorine-Bleaching Pulp and Paper Mills. Ofﬁce of
Water Regulations and Standards, Assessment and Watershed Protection Division,
Washington, D.C. 36 pp.

Van Strum, C. and P. Merrell. 1987. No margin of safety: a preliminary report on
dioxin pollution and the need for emergency action in the pulp and paper industry.
Greenpeace USA, Inc., Toronto, Ontario.

Verway, David, I. 1987. Michigan statistical abstract, twentieth edition 1986-87.
Bureau of Business Research, School of Business Administration, Wayne State
University, Detroit, MI.

Wentz, CA. 1989. Hazardous waste management. McGraw-Hill Book Company,
New York, NY. p. 53.

Wheelan, Gary. 1990. Fisheries Biologist, Fisheries Research Laboratory, Michigan
Department of Natural Resources, Ann Arbor, MI. Personal Communication.

Wolfe, W.H., J .E. Michalek, and LC. Miner. 1989. An epidemiologic investigation
of health effects in Air Force personnel following exposure to herbicide - mortality
update. Poster presented at the Ninth International Symposium on Chlorinated
Dioxins and Related Compounds, Toronto.

Young, A.L., L.G. Cockerham, and CE. Thalken. 1987. A long-term study of
ecosystem contamination with 2,3,7 ,8-tetrachlorodibenzo-p-dioxin. Chemosphere
16(8/9):1791-1815.

Zugger, RD. 1990. Michigan Water Resources Commission Authorization to
Discharge Under the National Pollutant Discharge Elimination System, Mead
Corporation, Publishing Paper Division, County Road 426, Escanaba, MI. Michigan
Department of Natural Resources, Lansing, Michigan.

APPENDIX A

Calculation of a Regional Bioaccumulation Factor for the
Escanaba River for Dioxin (Anderson et al., 1990)

WT TCDD WT TCDD
Eiah_ﬁneaiea llhal 129i) Eiﬁh_522£iﬁﬁ Libs) 122:1
Northern Pike 6.1 3.31 Smallmouth Base 2.0 5.51
Northern Pike 6.3 23.40 White Sucker 3.7 8.44
Northern Pike 5.7 10.20 White Sucker 3.0 8.29
Northern Pike 5.7 12.00 White Sucker 4.4 17.60
Northern Pike 5.0 13.60 White Sucker 3.7 9.06
Northern Pike 5.2 24.40 White Sucker 3.0 13.40
Northern Pike 4.0 7.15 White Sucker 2.9 15-30
Northern Pike 3.7 4.90 White Sucker 3.1 12.40
Northern Pike 4.6 9.81 White Sucker 3.9 8.80
Northern Pike 3.3 2.86 White Sucker 3.5 8.55
Northern Pike 2.5 6.62 White Sucker 4.1 9-51

Mean TCDD concentration I 10.69 ppt

‘ Effluent Effluent Weighted Average Effluent
Plow TCDD Cone. TCDD Concentration - 26.20 ppq
Dix: _nnsnl__ . ..12221_—

. 12/87 35 17.2 River TCDD Concentration
03/88 33 39.9 .26120_nns.x_36_nsd
06/88 39 18.6 602.63 mgd + 36 mgd
09/88 37 47.9 s 1.48 ppq
12/88 34 20.6
03/89 43 40.0
06/89 39 11.9

13.7

09/89 39
BAP calculation

lQa§2_RPS

1.4a ppq . 7238

163

.AJHHENﬂlﬂxlB i

Rule 57 (Mich. Admin. Code. r.323.1 05 7)

R 323.1057. Toxic substances.

Rule 57. (1) Toxic substances shall not be present in the waters of
the state at levels which are or may become injurious to the public
health, safety, or welfare; plant and animal life; or the designated uses
of those waters. Allowable levels of toxic substances shall be deter-
mined by the commission using appropriate scientific data.

(2) All of the following provisions apply for purposes of developing
allowable levels of toxic substances in the surface waters of the state
applicable to point source discharge permits issued pursuant to. Act ‘
No. 245 of the Public Acts of 1929, as amended. being 5323.1 et seq. of
the Michigan Compiled Laws:

(a) Water quality-based effluent limits developed pursuant to this
subrule shall be used only when they are more restrictive than technology-
based limitations required pursuant to R 323.2137 and R 323.2140.

(b) The toxic substances to which this subrule shall apply are those
on the 1984 Michigan critical materials register established pursuant to
Act No. 245 of the Public Acts of 1929. as amended. being 5323.1 et seq.
of the Michigan Compiled Laws; the priority pollutants and hazardous
chemicals in 40 C.P.R. 5122.21. appendix D (1983); and any other toxic
substances as the commission may determine are of concern at a specific
site.

(c) Allowabli'levels of toxic substances in the surface water after a
discharge is mixed with the receiving stream volume specified in R 323.1082
shall be determined by applying an adequate margin of safety to the MATC,
NOAEL. or other appropriate effect end points. based on knowledge of the
behavior of the toxic substance. characteristics of the receiving water.
and the organisms to be protected.

(d) In addition to restrictions pursuant to subdivision (c) of this
subrule. a discharge of carcinogens. not determined to cause cancer by a
threshold mechanism. shall not create a level of risk to the public
health greater than 1 in 100,000 in the surface water after mixing with
the allowable receiving stream volume specified in R 323.1082. The
commission may require a greater degree of protection pursuant to R 323.1098
where achievable through utilization of control measures already in place
or where otherwise determined necessary.

(e) Guidelines shall be adopted pursuant to Act No. 306 of the Public
Acts of 1969. as amended. being 524.201 et seq. of the Michigan Compiled
Laws. setting forth procedures to be used by staff in the development of
recommendations to the commission on allowable levels of toxic substances
and the minimum data necessary to derive such recommendations. The
commission may require the applicant to provide the minimum data when
otherwise not available for derivation of allowable levels of toxic
substances.

(f) For existing discharges. the commission may issue a scheduled
abatement permit pursuant to R 323.2145 upon a determination by the
commission that the applicant has demonstrated that each of the following
conditions is met:

(i) Immediate attainment of the allowable level of a toxic substance
is not economically or technically feasible.

(ii) No prudent alternative exists.

(iii) During the period of scheduled abatement. the permitted discharge
will be consistent with the protection of the public health, safety, and
welfare.

(iv) Reasonable progress will be made toward compliance with this rule
over the term of the permit, as provided for in a schedule in the permit.

164

APPENDIX C

Michigan Department of Natural Resources Rule 57 (2) and
Water Quality Based Efﬂuent Limit for Mead, Escanaba Calculations

 

Rule 57 (2)1
D x Wk
C = WC + (F x BCF)
where: C = Concentration of carcinogen (mg/1)
D = Dose (6.62 x 10" ng/kg/day)
Wh = Weight (70 kg)
WC = 0.01 l/day for surface water
F = Fish consumption (0.0065 kg/day)

BCF = Bioconcentration factor (51,600)

6.62 x 10'5 x 70 kg
C = 0.01 l + (0.0065 kg x 51,600) = 1.4 x 10'5 ng/l

Water Quality Based Effluent Limit for Mead, Escanaba2

V((1/4 QR) + QB) - CRQR

WQBEL = QE
where: V = Rule 57 (2) value of 1.4 x 10" ng/l
QE = Mead outfall design maximum ﬂow (50 mgd)
QR = 95% exceedence flow for the Escanaba River (122.8 mgd)
CR = Ambient river dioxin concentration (zero assumed)

1.4 x IQ” ng/l((.2§ x 122.8) + SQ mgd) - (Q x 122.8)
50 mgd

= WQBEL = 2.2596 x 10'5 ng/l = 0.022 pg/l = 0.022 ppq

 

' MDNR, 1989
2 Anderson et al., 1990

165

Average Fishing Population in Delta County42

APPENDIX D

 

Type of

License 1986 1987 1988

Resident 7, 129 7,241 7,005
Senior 1,174 1,146 1,205
Average

Non

Resident 1,645 2,070 1 ,95 2
Daily

Permit 3,183 3,545 3,291

 

 

‘2 MDNR, 1989b

166

1989

6,909

1,160

2,175

3,410

Average

7,071

1,171

AAPTHHNEHDIIE

Mead Escanaba Pulp Mill Draft Dioxin Minimization Plan for
Wastewater Treatment and Kraft Pulp Mill Modifications (Abbott, 1990)

ﬁg! Astign Plans ,7

D

 

Continue the evaluation of alternative brownstock
washing defoamers to minimize 080 and UHF precursor
contamination while providing satisfactory washing
performance.

$urvey the process for other sources of 080 and 08F
contamination in the brownstock pulp before bleaching
(e.g., evaporator condensates used on the brownstock
washer showers).

Initiate the use of a new brownstock washing defoamers

and other appropriate process changes which can lower

080 and DBF contamination in the brownstock pulp to the
levels present as a result of background levels of 080

and DBF contamination in the environment (e.g., mill
fresh water).

Study the Operation of the wastewater treatment
facility to identify operating conditions which
improve the removal of TCDD and TCDF.

Convert the existing SVP process Clozigenerator to

the R-8 process increasing the production capacity
for C102.

Improve the efficiency of application of Cl; and
£102 in the first bleaching stage by replacing the
existing static mixer with two new static mixers in
series.

Convert the first bleaching stage from a simultaneous
to a D + c, sequential bleaching chemical
application.

Increase the £10, substitution for Cl2 beyond the
10% level. ,

Optimize the D + c, first bleaching stage using
the following process variables:

- C10; split between the D and Cu application
- Temperature
- Water recycle practices

- Etc. '

167

Timgtgblg
Nov.,1989-Dec.,1990

Nov..1989-Dec.,1990

Nov.,l989-DEC.,199O

May, 1990-Jan., 1991

May. 1990

May, 1990

May, 1990

June, 1990

May, 1990-Jan.. 1991

WM ' 11mm

 

. Study process control improvements to minimize the Feb.,1990-Jan.,1991
molecular chlorine usage in the first bleaching
stage.

. Continue pilot plant process research studies to Nov.,1989-Jan.,1991

minimize TCDD and TCDF production during the
bleaching of kraft pulps. Key process variables
to be investigated are:

Level of Clo. substitution
Cloz split between the D and C. application

pH of 0 + C stage

Eo versus second bleaching stage with
resulting mzlecular Cl2 reductions

E. or E... bleaching stage before D + C. stage
E. (D + C.) E. D
or
E... (D + C.) E... D
versus
(D + C.) E. D E D

- Assuming promising results are obtained from a pilot Nov., 1990
plant evaluation of adding hydrogen peroxide to an
oxidative extraction stage (i.e., E ) to reduce
molecular chlorine usage. conduct full scale mill
trials evaluating the benefits of E0. versus E.
bleaching.

. If significant process modifications are required
beyond the actions being taken in 1990 to meet the
goals of the DMP, the following additional process
improvements are projected based on today’s knowledge:

- Convert E stage to full tineE operation Oct., 1991
by adding° new H.D. storage, hanEdling. and
process control equipment.

- Convert the (D + c.) E .DED bleaching line to Oct., 1992
a E. (D + C.) E D sequence through extensive
modification and additions (e. g. ., long
delivery time high shear mixers).

168

APPENDIX F

Base Year Value Calculations for Willingness to Pay Values

GNP deflator

U.S. Department of Labor
Consumer Price Index Monthly Press Release
CPI-U All Urban Consumers

 

 

 

Annual Average Percent
1967 Base Change

1986 328.4
1987 340.4 .035
1988 354.3 .039
1989 371.3 .046
1990
(up to 7/90) 386.2 .039
Ave. .040

Z = X(1/r)"
Where Z = Corrected S

X = 3 amount

r = average inﬂation rate from 1986-90

n = equals number of years in period

1986 1990
Value Equivalent‘

Study ($) ($)
Olsen (1981) 8,000,000 6,801,622
Gegax et a1 (1985) 1,600,000 1,360,323
Blomquist (1979) 610,000 518,624

 

1. 1986 dollars have been adjusted to reﬂect 1990 purchasing power.

169

"I7'11iiiiii'il‘l'i‘iiii