The Determination of Occupational Health and
Safety Standards in Ontario, 1860-1982:
From the Market to Politics to…?
Eric Tucker*
The author reviews the historical develop-
ment of the decision-makingframeworks within
which courts and the Legislature have made
choices regarding the allocation of risks to
health and safety in the workplace. Arguing
that this development has been conditioned
by the necessity of satisfying in a capitalist
democracy conflicting demands to facilitate
capital accumulation and to justify to the
electorate the manner in which choices re-
garding the structure of the processes of pro-
duction have been made, the author contends
that recent pressure to adopt cost-benefit
analysis as the basis for health and safety
standard setting is mistaken. He considers the
failure of cost-benefit analysis to satisfy the
demands of legitimation and accumulation,
and challenges its adequacy as a normative
and a political principle. In setting his criti-
cism in the context of a broad view of the
political and historical aspects of legal rule-
making, the author can address the limits on
present allocations of risk imposed by the
structure of society, and discuss the possi-
bility of significant future reform.
L’auteur passe en revue l’6volution du pro-
cessus d6cisionnel suivi par la Iegislature et
les tribunaux ontariens dans le choix des po-
litiques de risque en mati~re de sant6 et de
s6curit6 au travail. I1 pr6tend que cette 6vo-
lution a dt6 temp6r6e par la n~cessit6, dans
une socit6 capitaliste et democratique, de
satisfaire aux demandes conflictuelles de fa-
ciliter l’accumulation du capital et de l6giti-
miser; aux yeux de l’61ectorat, Ia fagon de choisir
les proc6d6s de production. Consid6rant mal
fondees les pressions actuelles visant a faire
de l’analyse cofit-b6n6fice la m~thode de base
pour fixer les normes de sant6 et de s6cu-
rit6,l’auteur en d6montre l’chec, face aux de-
mandes de lgitimisation et de capitalisation.
I1 remet en question l’efficacit6 de cet outil
normatifet politique. L’auteur, en formulant
cette critique dans la large perspective des
aspects historiques et politiques du processus
16gislatif, peut ensuite apprcier A la fois les
limites impos~es aujourd’hui aux politiques
de risque et les possibilit6s d’une reforme ma-
jeure dans l’avenir.
*Assistant Professor of Law, Osgoode Hall Law School, York University.
1983]
OCCUPATIONAL STANDARDS IN ONTARIO
Synopsis
Introduction
I. The Contractarian Framework
A. Selection of the Framework
1. Historical and Economic Background
2. The Common Law Courts
B. The Failure of the Framework and Its Ensuing Transformation
H. State Standard Setting in Nineteenth-Century Ontario
A. First Steps
B. The Coming of Age
C. Note on Employer Liability and Standards
M. Interregnum: 1914-1960
IV. The Second Wave of State Regulation: 1961-1980
A. Process
B. Policy
1. The Approach of Cost-Benefit Analysis
a. The Adequacy of the Market
b.
c. Risk Premiums
Information
2. The Application of Cost-Benefit Analysis to Occupational Health
and Safety Standard Setting
a. Estimating the Benefits
b. Estimating the Costs
c. Conclusion
3. The Normative Assumptions of Cost-Benefit Analysis: A Brief
Critique
4. The Political Economy of Cost-Benefit Analysis
Conclusion
*
*
*
REVUE DE DROIT DE McGILL
[Vol. 29
Introduction
Ontario has recently set seven toxic substance exposure levels under
the Occupational Health and Safety Act.’ This is the first time in Ontario
that exposure to toxic substances in the workplace has been controlled by
legally binding standards. 2 Proponents of government regulation as a so-
lution to the problem of occupational health and safety may see this step
as a long-awaited breakthrough, or perhaps as a disappointment, depending
on their view of the adequacy of the regulation. Opponents of regulation
may bemoan the fact that we seem destined to repeat the mistakes of our
American neighbours. My purpose is neither to celebrate nor to mourn the
fact that we now have standards, or the particular standards that we have.
‘Designated substances: Lead-O. Reg. 536/81; Mercury-O. Reg. 141/82; Vinyl Chloride-
0. Reg. 516/82; Coke oven emissions-O. Reg. 517/82; Asbestos-O. Reg. 570/82; Isocyan-
ates-O. Reg. 455/83; Silica-O. Reg. 769/83. These Regulations were filed under the Occu-
pational Health and Safety Act, R.S.O. 1980, c. 321.
21n the absence of specific toxic substance exposure standards, the Ministry relies on general
regulations to the effect that “[a]ll measures necessary to prevent exposure to any toxic sub-
stance… shall be taken…” (0. Reg. 658/79, respecting industrial establishments). Similar reg-
ulations can be found with respect to construction projects and mines. This kind of general
language can be traced back to The Ontario Factories’Act, 1884, S.O. 1884, c. 39, s. 11(3):
Every factory shall be ventilated in such a manner as to render harmless, so far as
is reasonably practicable, all the gases, vapours, dust or other impurites generated
in the course of the manufacturing process or handicraft carried on therein that
may be injurious to health.
In enforcing this general regulation the Occupational Health and Safety Division (OHSD)
of the Ministry of Labour relied on data sheets to set exposure standards. Failure to comply
could, in theory, be considered a violation of the general regulation, but the vagueness of the
regulation probably meant that the data sheets were considered as guidelines rather than as
legally binding standards. See Franson & Lucas, “The Legal Control of Hazardous Products
in Canada” in Canadian Law and the Control of Exposure to Hazards (Science Council of
Canada, Background Study No. 39, 1977) 36. More recently the OHSD has issued a document
entitled “Exposure Criteria for Potentially Harmful Agents and Substances in Workplaces”
(Ontario, Ministry of Labour, May 15, 1981) which suggests criteria designed to assist people
involved in the administration of the Occupational Health and Safety Act in determining when
exposure to a wide range of substances might be considered harmful. Again, these are not
legally binding standards, although they may be adopted in the future as designated substance
regulations. However, the courts have not treated the failure to comply with these Exposure
Criteria as a violation of the Act. See R. v. Windsor Board of Education (1982) 1 CCH Can.
Employment Health and Safety Guide, para. 95-046 (Prov. Ct. Cr. D.).
1984]
OCCUPATIONAL STANDARDS IN ONTARIO
Rather, I am concerned with recent proposals for regulatory reform which
call for a requirement that standards be selected on the basis of cost-benefit
analysis. The movement to subject all major government regulations to such
analysis has made substantial advances in the United States during the
Reagan presidency 3 and seems to be gaining support here.4
This paper arises out of a concern that the adoption of cost-benefit
analysis as a basis for occupational health and safety standards in Ontario
would be a serious mistake which could leave thousands of workers legally
exposed to hazardous conditions. Because of this concern I have devoted
a significant portion of this paper to a review of the technical and ethical
objections that have been voiced with respect to economic analysis. I also
want to extend the familiar criticisms by placing cost-benefit analysis in an
historical and political-economic context, and to explore the links between
modern proposals that regulatory standards be subjected to cost-benefit anal-
ysis and the nineteenth-century common law of employers’ liability. While
the institutional framework for determining occupational health and safety
standards is clearly different than that which governed employer-employee
relations in the nineteenth century (public regulation as opposed to private
contract) there are strong ideological similarities between the two ap-
proaches to the assignment of risk. This can be seen in their common em-
phasis on the goal of allocative efficiency, to the exclusion of competing
considerations such as distributional equity. These two approaches are also
linked with respect to the view they take of the legitimate ends of state
activity. Both emphasize its function of facilitating the survival and expan-
sion of capitalism by maintaining conditions favourable to profitable capital
accumulation, through, for example, constituting the legal infrastructure for
a capitalist labour market.
Beyond this ideological similarity, I also intend to explore the dynamics
underlying the selection of decision frameworks within which occupational
health and safety standards are determined. For example, why was it that
courts in the early to mid-nineteenth century selected employer liability
3 1n the United States, President Reagan has required all major rules to be subject to a form
of cost-benefit analysis, as well as to review by the Office of Management and Budget and the
Task Force on Regulatory Relief. See Executive Order 12,291 (1981) 46 F.R. 13193.
However, in light of the United States Supreme Court ruling in American Textile Manufac-
turers Institute, Inc. v. Donovan 452 U.S. 490 (1981) that Congress had already defined the
relationship between costs and benefits by placing the benefit of worker health above all con-
siderations save those making the attainment of that benefit unachievable, it is unlikely that
the Executive Order will be applied to OHSA toxic substance standards (if any are proposed).
4See G. Doern, M. Prince & G. McNaughton, Living with Contradictions: health and safety
regulation and implementation in Ontario (Ontario Royal Commission on Matters of Health
and Safety Arising from the Use of Asbestos in Ontario, Study Series, No. 5, 1982) 5.6-5.7
(hereinafter Doern).
McGILL LAW JOURNAL
[Vol. 29
rules that were contractually-based and infused with presumptions derived
from the logic of a capitalist labour market to assign responsibility for injury
in the workplace when, by the last decades of the nineteenth century, leg-
islatures had intervened to impose minimum standards enforced by an
inspectorate? What do those events tell us about the changing role of the
state and the constraints operating upon it in the field of occupational health
and safety policy? What does this analysis have to reveal about the likelihood
and implications of the adoption of cost-benefit analysis?
In pursuing this line of inquiry I will be drawing upon recent attempts
to develop a Marxist theory of the state. 5 In general, these theories reject
the orthodox Marxist-Leninist view of the state in capitalist society as a
mere tool of the ruling class, reflexively expressing and enforcing its interests.
They also reject the liberal view of the state as a neutral arbiter between
conflicting claims of fragmented interest groups, including labour. Rather,
they seek to show how the state is relatively autonomous from the capitalist
class by identifying both the supports for the independence of state power
and the limits on its exercise. At the most abstract level we can examine
an economic system (capitalism, for example) and ask what limits it imposes
on the structure and policies of the state. To the extent that a range of such
structures and policies is compatible with capitalism, we can assert the
existence of the relative autonomy of the state, in the sense that no particular
structural or policy development is determined by the economic system,
although some developments are more likely than others. At the next level
of analysis we may ask what functions the state serves in a capitalist social
formation. If the state performed solely the function of facilitating the ac-
cumulation of capital, then one would expect a low level of autonomy. In
order to facilitate accumulation the state would take directions from the
capitalist class and would act autonomously only to the extent that the
capitalist class itself was fragmented or otherwise incapable of organizing
itself. However, if the state also has the function of obtaining and main-
taining the consent and loyalty of the public-at-large to existing socio-eco-
nomic arrangements (legitimation), then we may find that its roles conflict,
as the policy which best facilitates accumulation from an economic per-
spective may be politically unacceptable. Thus, at the level of functional
analysis, we may find further support for the relative autonomy of the state.
The precise nature of the relation between the state and the dominant eco-
nomic system cannot be predicted in the abstract, but only by an analysis
of the political and social forces operating in an historical conjunction.
5J. O’Connor, The Fiscal Crisis of the State (1973); R. Milliband, The State in Capitalist
Society (1973); J. Habermas, Legitimation Crisis (1975); L. Panitch, ed., The Canadian State:
Political Economy and Political Power (1977); E. Wright, Class, Crisis, and the State (1978).
For an excellent review of the literature see B. Jessop, The Capitalist State (1982), and Recent
Theories of the Capitalist State (1977) 1 Camb. J. of Eco. 353.
1984]
OCCUPATIONAL STANDARDS IN ONTARIO
In my analysis of the historical development of occupational health
and safety standard-setting from the period of its reliance on a market-based
contract model to the point at which it took on a political character, and
in my discussion of contemporary proposals to implement cost-benefit anal-
ysis, I have tried to achieve a balance between functional analysis at the
abstract level and concrete historical determinations. Nonetheless, the anal-
ysis has remained somewhat more schematic than I had desired, and it is
offered here in the hope that it will be built upon in the future through
further historical research.
I. The Contractarian Framework
A. Selection of the Framework
1. Historical and Economic Background
The selection by common law courts in the mid-nineteenth century of
a market-based contractual framework for determining conditions of em-
ployment relating to occupational health and safety can only be understood
in the context of the emergence of a capitalist labour market during this
period.6 This is not to suggest that prior to industrial capitalism workers
were not killed or injured in the course of their work. Obviously there are
hazards involved in any productive activity and some activities will be more
hazardous than others. What was new in the nineteenth century was the
introduction of technologies of production which exposed workers to dif-
ferent and probably greater hazards than were present in pre-industrial work
settings, 7 and the way in which decisions about the distribution of risks
61n England the principal cases are: Priestley v. Fowler (1837) 3 M. & W. 1, (1837) 150 E.R.
1030 (Ex. Ct.); Hutchinson v. York, Newcastle, and Berwick Ry Co. (1850) 5 Ex. 343, (1850)
155 E.R. 150. See generally Ingman, The Rise and Fall ofthe Doctrine of Common Employment
(1978) 23 The Juridical Review 106.
In the United States the principal case is Farwell v. Boston and Worcester Ry Co., 45 Mass.
(4 Met.) 49 (1842). See generally L. Levy, The Law of the Commonwealth and Chief Justice
Shaw (1957).
In Ontario, the principal case is Deverill v. Grank Trunk Ry Co. (1866) 25 U.C.Q.B. 517.
For a discussion see Risk, The Last Golden Age: Property and the Allocation of Losses in
Ontario in the Nineteenth Century (1977) 27 U.T.L.J. 199, 233-6.
7Records of industrial accidents and disease were not kept until the nineteenth century and
so it is impossible to compare statistics in order to prove that the new industrial technologies
were more hazardous to workers than the old. Certainly the number of workers exposed to
hazards in an industrial setting increased with the growth of the urban working class, and the
overwhelming majority of contemporary observers were convinced that the coming of industrial
capitalism brought the majority of working people “appalling hardships”. For a general review
of historical debates about the condition of the British working class in the first half of the
nineteenth century, see E. Hobsbawm, “History and ‘The Dark Satanic Mills”‘ in Labouring
Men: Studies in the History of Labour (1976) 123-40.
REVUE DE DROIT DE McGILL
[Vol. 29
amongst groups in society and tolerable levels of risk were made. In order
to place the selection of a contractarian framework in perspective it might
be useful to compare capitalist relations of production with those that pre-
ceded them. Feudal society was organized on the principles of contingent
property ownership (tenures) and unfree labour. The basis of productive
activity, land, was held in exchange for services and, at the lowest level,
labourers were given land to work for themselves in exchange for a pro-
portion of the product or the labourer’s time. The proportion of product or
time due to the lord was settled by custom. Hierarchical relations of dom-
ination and subordination were supported by a matrix of traditional and
religious values which imposed constraints on the lord as well as on his
subordinates. Of course physical coercion could also be brought to bear if
traditional sources of social order proved to be insufficient.
Even as the feudal- organization of production broke down and was
gradually replaced by wage labour, custom still played a major role in de-
termining the conditions of employment.8 Included in this matrix of custom
and values was an obligation of the employer to provide reasonably safe
working conditions. This obligation had been articulated by the common
law courts in the eighteenth century, persisted through the nineteenth cen-
tury,9 and can still be said to exist today.’ 0 However, the substantive content
of that obligation has undergone a number of changes.
Capitalism can be defined as a wage labour system of production for
profit. Owners of the means of production hire the direct producers, but
keep the product themselves for sale on the market at a profit or loss. The
profit or loss can be measured as the difference between the cost of inputs,
including labour, and the price that the outputs or commodities fetch. It is
crucial to the enterprise that labour costs be minimized. How then are labour
costs determined? For capitalism to function labour must be treated as an
ordinary commodity. Ownership of the commodity is initially assigned to
its possessor, the labourer, and as owne? he has the power to enter into
contracts for its purchase and sale. The price of the commodity will, as for
other commodities, be determined in the marketplace according to the prin-
ciples of supply and demand. For the capitalist, labouir power is a variable
input and not a capital asset; that is, it can be hired as needed, provided
that a sufficient pool of labour power of the right type is available upon
demand. In the absence of such a pool, capitalistic labour relations tend not
to develop. Thus, in capitalistic labour markets there tends to be a reserve
bouring Men, ibid., 405-35.
8See Hobsbawm, “Custom, Wages and Workload in Nineteenth Century Industry” in La-
9See Ingman, supra, note 6, 109 and cases cited therein.
1oJ. Fleming, The Law of Torts, 6th ed. (1983) 478-9; W. Prosser, Handbook of the Law of
Torts, 4th ed. (1971) 526.
19841
OCCUPATIONAL STANDARDS IN ONTARIO
pool of labour, or to put it simply, supply exceeds demand. If workers are
unable to gain some control over the supply, labour costs are likely to
decline.
At a minimum, a legal framework consonant with capitalism must allow
for and protect private property and recognize and enforce exchange through
contract. In particular, labour power must be recognized as a commodity
which can be sold by its possessor, the potential direct producer. However,
capitalist production can occur even if the wage relationship does not exist
in its perfect form. Some of the conditions governing the sale of labour
power may still be determined by non-market criteria. For example, in
industries which still depend on craft skills the measure of what constitutes
“a fair day’s work” may depend on the custom of the trade. It is with this
background in mind that I would now like to examine the selection of a
contractarian framework for determining occupational health and safety
standards in Ontario.
At the beginning of the nineteenth century capitalism was hardly the
dominant mode of economic organization in the territory that now com-
prises the province of Ontario. Even by 1812 the territory was sparsely
settled, with the population probably not exceeding 100,000. Land was read-
ily available and the majority of settlers were engaged in primary commodity
production. Wheat and lumber were exported to England under favourable
trading arrangements. 1I To the extent that employment relations were es-
tablished they often did not take the form of wage labour, but rather of
“personal” labour relations, or forms of unfree labour such as indentured
servitude. The shortage of labour was a chronic complaint and under such
conditions it was unlikely that wage labour would emerge as the dominant
relation of production.12
The population continued to grow slowly until 1825. In the next two
decades there was explosive growth from immigration, and by 1842 the
population had tripled to 450,000, doubling again by 1851. The large pool
of labour was absorbed by the construction of a transportation infrastructure
to handle the movement of primary products. Canal construction in the
1830’s and 1840’s and the boom in railway construction in the 1850’s were
the sectors in which wage labour expanded. Factory production expanded
“This overview of economic developments is derived from secondary sources, the ‘most
important being W. Easterbrook & H. Aitken, Canadian Economic History (1956) and R. Harris
& J. Warkentin, Canada Before Confederation: a study in historical geography (1974).
2See H. Pentland, Labour and Capital in Canada, 1650-1860 (1981), particularly ch. 2. From
1
the employer’s perspective wage labour will not be preferable to pre-capitalist arrangements
unless there is an excess of supply over demand. In its absence, supply of the right quality of
labour at the right time cannot be guaranteed and the cost of labour is likely to rise.
McGILL LAW JOURNAL
[Vol. 29
as increased population, urbanization and requirements for railway con-
struction combined to generate a steady level of internal demand capable
of attracting capital investment. Industrial processes were introduced as
manufacture became increasingly concentrated in larger firms and cities.
Thus, by 1860, industrial capitalism had firm roots in Southern Ontario
and wage labour had become a major, if not the dominant, relation of
production. 13
It is appropriate to remark that workers did not necessarily flock to
become wage labourers. 14 As I have already noted, the ready availability of
land either in Canada or the United States provided an attractive alternative
to wage labour for large numbers of immigrants, and contributed to a chronic
labour shortage in the early part of the nineteenth century. Craft producers
also resisted the transformation of the labour process. Perhaps the major
group to seek wage labour were the Irish immigrants who were fleeing from
intolerable conditions in their homeland and had already been stripped of
their resources prior to their arrival in Ontario.’ 5 The emergence of wage
labour can be explained by a combination of factors including superior
organization of production, economies of scale achieved both by more ex-
tensive and more intensive exploitation of labour, concerted political action
by the capitalist class to support their economic power, and the relative
inability of those who were being drawn into the capitalist labour market
to mount effective resistance to the destruction of their way of life.’ 6
In the absence of statistics,’ 7 it seems reasonable to assume that the
introduction of industrial processes and the more extensive and intensive
13For recent discussions of the development of industrial capitalism in Ontario see G. Kealey,
Toronto Workers Respond to Industrial Capitalism, 1867-1892 (1980), ch. 2 and B. Palmer, A
Culture in Conflict: skilled workers and industrial capitalism in Hamilton, Ontario, 1860-1914
(1979). For a critical review of the emphasis of these historians on the role and place of the
industrial proletariat in the formation of the Canadian working class see D. Drache, “The
Formation and Fragmentation of the Canadian Working Class: A Dissenting View” (forthcoming).
14For example, the destruction of the economic foundation of the English peasantry pushed
them into wage labour or onto poor relief. For a discussion of this process see B. Moore, Social
Origins of Dictatorship and Democracy: Lord and Peasant in the making of the modern world
(1967) 20-9.
15See H. Pentland, supra, note 12, ch. 2.
16For an excellent review of the process of proletarianization in the United States see D.
Gordon, R. Edwards & M. Reich, Segmented work, divided workers [:] The historical trans-
formation of labour in the United States (1982) ch. 3 [hereinafter Gordon].
17The first requirement for employers to systematically report injuries to their employees
from particular causes (i.e. fire, machines) was not introduced until The Ontario Factories’Act,
1884, S.O. 1884, c. 39, ss 18, 19. Earlier, concern about injuries to railway employees led the
Legislative Assembly to strike a committee which conducted a survey of deaths and injuries
to railway employees. See Report of the Select Committee on Railway Accidents (1880) Vol.
XIII, Journals of the Legislative Assembly (Ontario), Appendix I.
1984]
OCCUPATIONAL STANDARDS IN ONTARIO
employment of labour combined to increase the incidence of employment-
related injuries and deaths.18 Widows and workers sought relief.19 One forum
to which they turned was the common law courts. This raised the problem
of the selection of legal liability rules, which in turn required that the courts
select a set of rules for determining the standards against which employment
conditions relating to occupational health and safety were to be measured.
2. The Common Law Courts
Of course, Ontario courts did not select legal rules in a vacuum. England
and the United States had already chosen a contractarian legal framework.20
Ontario’s courts remained colonial institutions even after the Constitution
Act, 1867 and thus, in a formal sense, it could be said that Ontario courts
did not select legal rules at all, but merely applied those rules which had
been selected in England.21 A reading of the reported employer liability cases
between 1861 and 1886 (when the Ontario Legislative Assembly adopted a
new set of decision rules), reveals that the rules were applied strictly and
18Some evidence for this assertion may be provided by the reported employer liability cases
in which the overwhelming majority of claims arose from injuries or deaths caused to workers
involved with mechanical processes. See E. Tucker, “Employer Liability Law in Nineteenth-
Century Ontario”, unpub. ms. However, factors influencing the selection of litigated cases (see
note 19 below) require one to be cautious in inferring that a high proportion of litigated claims
reflects a higher-than-average injury rate.
191t would be interesting to speculate as to the factors which influenced the decision to seek
redress in the courts. The availability or adequacy of ex gratia payments by the employer or
self-insurance might be one factor. As well, a worker who did not suffer serious impairment
and was able to resume work with his employer would be unlikely to seek damages as it would
probably result in immediate dismissal. Thus, it is most likely that widows and the severely
disabled would seek legal redress. Further, unless there were contingent fees or crusading ad-
vocates, it is likely that the litigants would come from the better paid segments of the working
class.
20Supra, note 6.
21See generally Roberts-Wray, Commonwealth and Colonial Law (1966) 565-7. For a dis-
cussion of the attitude of Ontario judges towards English precedents, see Risk, The Law and
the Economy in Mid-Nineteenth Century Ontario: A Perspective (1977) 27 U.T.L.J. 403, 422
where he states:
For the courts, faithfulness to the precedents of English courts was a constitutional
obligation, which was supported by loyalty, habit and convenience.
Evidence of this approach in the area of employers’ liability law comes from a case in which
the court was presented with American cases which had modified the fellow servant rule. In
declining to follow that path Hagarty, C.J.O. tersely stated:
There is a marked difference between English and American laws on the “fellow
servant” question. Of course we have to follow the former.
See Matthews v. Hamilton Powder Co. (1887) 14 O.A.R. 261, 265 (Ont. C.A.).
REVUE DE DROIT DE McGILL
[Vol. 29
with little hesitation. 22 The tenor of the judgments suggests that even if our
judges had been more independent it was unlikely that they would have
chosen differently.
The proposition that an employer was obliged to provide reasonably
safe conditions of work was not rejected by the courts. Rather, they devel-
oped a number of doctrines which provided strong defences to any employer
who was sued on the basis that he had failed to fulfil this obligation. The
first defence, that of contributory negligence, went to the issue of causation.
If it could be shown that the employee’s own negligence had contributed
to the occurrence of the accident, then the employee was absolutely barred
from recovery. This was not a doctrine that was specific to employer liability
law, and, although it did provide a strong defence to employers, it did not
directly address the problem of the standard against which the conditions
of employment relating to safety could be measured for the purpose of
determining whether the employer was liable.
The second defence was based on voluntary assumption of risk, and
the selection of this doctrine constituted a fundamental decision about the
way in which the problem of employer liability would be resolved. Essen-
tially, the doctrine provided that if employees were aware or ought to have
been aware of the hazards present in the workplace, then they were deemed
to have assumed those risks as conditions of their employment and could
not sue their employer for damages in the event that an injury was subse-
quently caused by those hazards. The theory articulated by the court in
support of this position was that the employee who was aware of hazards
would demand a higher wage in order to incur the risk of injury. The contract
of employment would provide the mechanism for compensation, not ju-
dicial enforcement of customary standards. 23
22See Tucker, supra, note 18, and Plant v. Grand Trunk Ry Co. (1867) 27 U.C.Q.B. 78, in
which an employee was killed because of brake failure. Liability was denied on the basis that
fellow employees had failed to discover and remedy the defect in the brakes and because Mr
Plant was contributorily negligent in running down the track to escape injury, rather than
stepping to the side. To the widow and children of Mr Plant, Draper, C.J. (at p. 86) could only
offer sympathy:
The loss and misfortune to the plaintiff and her children is doubtless very serious
and sad, but we must not be drawn out of our path of duty, even by our feelings
for the widow and the orphans.
That Ontario judges were not drawn out of their “path of duty” is established by the fact
that employees lost seventeen of the twenty reported employer liability cases decided prior to
statutory reform of the common law of employers’ liability.
23Mr Justice Shaw in Farwell v. Boston and Worcester Ry Co., supra, note 6, and Lord
Bramwell in numerous judgments and public statements most clearly expressed this rationale
for the selection of the rule.
1984]
OCCUPATIONAL STANDARDS IN ONTARIO
On this view, the aggregate of market transactions would determine the
level of health and safety present in the workplace. If the workers demanded
higher wage premiums in order to incur the risks present in the workplace,
then employers would choose either to eliminate the risk or pay the pre-
mium, whichever was cheaper. Private choices made by autonomous in-
dividuals would produce the socially desired level of safety. State imposition
of minimum safety standards was an unwarranted intrusion on private
property and exchange relationships. Formally voluntary exchange rela-
tionships became the measure of justice, or, to put it in more contemporary
language, efficiency itself was defined as justice, displacing other conceptions
grounded in tradition or moral theory. The market, not the court, deter-
mined what was reasonable. 24
The fellow-servant doctrine, which held that an employee could not
recover damages from his employer where the accident had been, at least
in part, caused by the action of a fellow worker, can be viewed as an extension
of the voluntary assumption of risk doctrine. One of the risks present in an
industrial setting was that you might be injured by the activity of a co-
worker. This was presumed to have been taken into account in the wage
bargain. Thus, the market would also provide a mechanism for dealing with
careless workers who caused injuries to others within the enterprise. If work-
ers demanded a higher premium to work alongside a careless co-worker,
employers would take greater care to hire competent workmen, provided
the cost of doing so did not exceed the aggregate of the incremental pre-
miums being demanded. The level of health and safety conditions in the
workplace would be determined by discrete “voluntary” transactions be-
tween workers and their employers, and not by the imposition of external
standards generated by courts or legislatures. Although the structure of cap-
italist relations of production did not require that health and safety con-
ditions be determined by the contract of employment, the selection of the
contractarian framework can be said to have facilitated the development of
conditions favourable to capitalism. By matching the economic power of
the capitalist class to employ labour in hazardous conditions with a legal
right to do so, the court generated a legal infrastructure which supported
the perfection of wage labour. Further, on the plausible assumption that
24Horwitz, commenting on nineteenth-century employer liability cases, concludes:
Thus, the contractarian ideology above all expressed a market conception of legal
relations. Wages were the carefully calibrated instrument by which supposedly equal
parties would bargain to arrive at the proper “mix” of risk and wages. In such a
world the old ideal of legal relations shaped by a normative standard of substantive
justice could scarcely coexist. Since the only measure ofjustice was the parties’ own
agreement, all pre-existing legal duties were inevitably subordinated to the contract
relation.
M. Horwitz, The Transformation of American Law, 1780-1860 (1977) 209.
McGILL LAW JOURNAL
[Vol. 29
externally imposed standards would have been higher than those that work-
ers were deemed to have voluntarily assumed, the decision not to apply
external standards would have also held down the cost of production.
The fact that a legal framework can be said to facilitate the development
of capitalist relations of production does not guarantee its selection. That
requires the intervention of judges where the matter is initially presented
as an issue for the common law. Although the selection of the law for Ontario
can be formally explained by its selection in England, the ideological prem-
ises underlying the choice seem to have gained widespread acceptance amongst
members of the judiciary on both sides of the Atlantic. 25
25The ideological framework that I refer to is that of classic nineteenth-century liberalism,
which placed the autonomous individual at the center of the universe and sought to maximize
the liberty of the individual to pursue his own ends. Thus, Risk notes that the “most powerful
and pervasive value [of the common law of Ontario in the mid-nineteenth century] was the
facilitation and encouragement of private initiative.” See Risk, supra, note 21, 417.
Notwithstanding the pervasiveness of these values in terms of the substance of the law, the
formalist stance of the Ontario judges inhibited the expression of an ideological world view
in their judgments. I have also been unable to find any extra-judicial statements.
In England, Lord Bramwell was not afflicted with such inhibitions. He articulated his ide-
ological premises in judgments, books and testimony before various Parliamentary committees.
My favourite quotation comes from Lord Bramwell’s testimony before the Select Committee
on Employers Liability for Injuries to Their Servants, on May 4, 1877:
1116 …. Everybody must know (it is a very respectable feeling in one point of view, though
I think an erroneous one), that servants will never complain of the conduct of each other to
the master. Though I think, upon a more strict view of their duty, they should do so, they do
not; they let things go on wrong time after time, and mischief be done and danger be incurred,
and they make no complaint at all. I think it is very desirable to guard against that. Then there
is another thing which, really, I can speak to of my own knowledge, from the cases which I
have tried; the recklessness of workpeople is wonderful, and it is astonishing the risks they
run. They get familiarised with them; they forget the old proverb that the pitcher goes often
to the well and is broken at last. I think all those considerations go to show that it is really
much more the master who requires to be guarded against the negligence of his servants where
he is not to blame himself, than it is the servant who requires to be protected against the
negligence of a fellow servant ….
1157. If the fellow workman in a mine sees that workmen are employed by the employers,
and he has no control whatever, neither directly nor indirectly on their employment, would
you regard it that a workman should be liable to all the injuries and accidents that persons so
employed would cause?-I am very glad you have asked me that question, because it reminds
me to say something I have wished to say. Certainly I would give him no remedy. Why does
he not leave the employment if he knows it is dangerous? To my mind, it is a sad thing to
hear men come into court, as I have heard them, and excuse themselves for not having done
that, on the ground that their bread depended upon it, or something of that sort. I should like
to see a more independent feeling on the part of workmen, so that they would say, “I will have
nothing to do with a man who employs dangerous things or dangerous persons”. And, besides,
you must remember that when people are engaged in a trade dangerous, either from the nature
of the trade itself, or from the means or the persons who are employed, they charge all the
more for it, they are compensated, they receive a sort of insurance premium; and it would be
19841
OCCUPATIONAL STANDARDS IN ONTARIO
B. The Failure of the Framework and Its Ensuing Transformation
If we are analyzing the success of a legal framework for social decision-
making in terms of its functional compatibility with capitalism, we must
not only consider the extent to which it facilitates particular kinds of eco-
nomic activity, but also whether it generates outcomes which are perceived
as legitimate on process or substantive grounds. Capitalist accumulation of
wealth is not merely an economic phenomenon; it takes place within a set
of social conditions, including legal frameworks, which can impede or fa-
cilitate the process. Conditions that seem to encourage accumulation be-
cause they facilitate a particular kind of economic activity may prove, in
the long run, to be unsuccessful for the capitalist because they fail to legi-
timize the process of accumulation in the eyes of potentially powerful forces.
Those forces may then be mobilized and threaten to upset the stability of
the conditions of accumulation, either through economic or political action.
I will argue that the contractarian framework selected by the courts failed
on both substantive and process grounds to legitimate the practice of ex-
posing workers to hazardous conditions without a concomitant obligation
to pay compensation in the event of an injury, and as a result was altered
within a relatively short period of time.
The expansion of industrial capitalism created in its wake a significant
class of proletarianized workers, or wage labourers. These wage labourers
tended to be concentrated in location. First, as individual firms grew in size
the number of employees literally found under one roof also increased.26
Second, industrial development tended to be concentrated in urban envi-
ronments, so that working class neighbourhoods began to develop.2 7 As well,
very unreasonable that they should receive damages as well. Suppose the master said to a
servant I will subscribe a shilling a week to an accident fund for you, but I must not be liable
for any damage you may receive in my service, surely that would be a reasonable arrangement
that ought to bind. Suppose he subsequently said I will give you the shilling, the case would
be the same, and that is now the case in effect in dangerous trades.
Reprinted in Irish University Press Series ofBritish ParliamentaryPapers [:] IndustrialRelations
19, 196, 198 and 201.
26Examining 1871 census figures for Toronto, Kealey, supra, note 13, concludes at p. 29:
The major sectors of the Toronto economy in 1871 then were highly industrialized
with large concentrations of workers, extensive mechanization, and an elaborate
division of labour.
27Concern with the living conditions of urban industrial workers arises simultaneously with
concern about factory conditions themselves. In England there were numerous studies and
reports exposing the unsanitary conditions of English cities. Indeed, Engels in his first work,
The Condition of the Working Class in England (1884), focuses not on factory conditions but
on the phenomenon of urbanization. For an interesting discussion see S. Marcus, Engels,
Manchester, dnd the Working Class (1974).
REVUE DE DROIT DE McGILL
[Vol. 29
workers sharing common ethnic or religious backgrounds were often con-
centrated within particular firms, industries and regions.2 8 Shared economic
conditions, physical locations and cultural backgrounds provided fertile ground
for the emergence of working class organizations which sought to promote
working class interests through political, economic, cultural, educational and
social activities.
The emergence of an organized working class, however, does not nec-
essarily lead to direct attacks on capitalism itself. Hobsbawm has written
that there are two great watersheds in the history of the employment of
industrial labour in the nineteenth century.29 The first is when employers
and employees learn the rules of the game, the game being a self-regulating
market economy. In this context, the common law courts both selected the
rules and then proceeded to educate employees as to their effect, making it
clear that workers would have to express their risk tastes in the market
because the state would not protect them by imposing its own view of what
was reasonably safe. The second watershed is the complete learning of the
rules so that employees demand what the market will bear, while employers
develop better techniques to exploit labour more intensively.
Implied in the proposition that having learned the rules of the game
one responds by developing strategies to play it is the notion that either one
accepts the rules as legitimate or recognizes that it is beyond one’s powers
to alter them. Sources of legitimation might include shared ideologies or a
perception that one’s position is improved by the selection of the rules. For
whatever reasons, 30 abolition of wage labour was not the major objective
of most Ontario labour organizations. 31
28The influence of ethnic concentrations in the emergence of working class organizations has
been discussed by Pentland, supra, note 12 and Kealey, supra, note 13. Commenting on the
Orange Order, a prominent Protestant Irish working class organization in nineteenth-century
Toronto, Kealey notes at p. 121:
The Orange Order played an ambiguous role in the life of the Toronto working
class. Although clearly dividing the working-class community in two, it nevertheless
provided some of its Protestant members with a number of strengths that were
usefully carried into the realm of unionism.
29Hobsbawm, supra, note 8, 406-7.
30The reasons for the relatively successful integration of workers into the capitalist order, in
the sense that revolutionary rejection of capitalism was the exception rather than the rule, are
far from clear and require more detailed study. For an excellent discussion of the problem that
draws on the history of the German working class, see B. Moore Jr., Injustice: The Social Bases
of Obedience and Revolt (1978). With respect to Canada, Pentland has argued that an important
factor affecting industrial relations was a cultural heritage that emphasized and supported a
hierarchical conception of employer authority and employee subordination. See H. Pentland,
A Study of the Changing Social, Economic, and Political Background of the Canadian System
of Industrial Relations (unpub. draft study for the Task Force on Labour Relations, Ottawa,
1968) 19-37. This study is cited with the kind permission of the estate of Professor Pentland.
31The Knights of Labour were perhaps the most radical of the labour organizations to gain
19841
OCCUPATIONAL STANDARDS IN ONTARIO
Nevertheless, it would be inaccurate to suggest that all aspects of the
legal foundation of the accumulation process were accepted by working class
organizations or, indeed, by other segments of the population. The con-
tractarian framework for determining workplace conditions was perhaps
considered one of the least legitimate of the newly selected game rules. Some
evidence of the failure of these rules to gain widespread acceptance is the
judicially noted fact that, if employer liability cases were allowed to go to
juries, juries almost invariably found the employer liable.32
Furthermore, the creation of an income franchise for provincial elec-
tions in 1824 in addition to the property franchise, and its gradual extension
until universal male suffrage was achieved in 1888, 33 provided both a focus
and a foundation for working class political activity. Workers developed
effective political organizations and by the 1880’s they had close ties with
the ruling Liberal Party headed by Oliver Mowat.34 Both the Liberals and
the Tories continued to compete for the working class vote by advancing
issues such as Factory Act legislation and reform of employer liability law.
Even though it is unclear what priority those issues had on organized la-
bour’s legislative agenda, 35 the Provincial Legislative Assembly enacted leg-
islation in the 1880’s which drastically changed the legal framework for
determining health and safety conditions. The essence of the new approach
substantial support in nineteenth-century Ontario. However, Kealey notes that the “Knights
conceived social transformation grandly but shunned conflict”. Their popularity lasted for a
short period in the 1880’s. See Kealey, supra, note 13, 215 and generally ch. 10.
32Where there was insufficient evidence, cases were withdrawn from the jury precisely because
of the problem of jury sympathy for the plaintiff. For example, in Deverill v. Grand Trunk Ry
Co., supra, note 6, 526, Hagarty, J. noted:
To leave such a case to the jury is, as has been remarked by English judges, simply
to direct a verdict for the plaintiff, where a railway company are defendants [foot-
notes omitted].
33For a general history of the legislation extending the franchise in Ontario see J. Boyer,
Political Rights: the Legal Framework of Elections in Canada (1981) 134-6. For a brief dis-
cussion of the causes of this extension and of the percentage of males eligible under these
statutes see Kealey, supra, note 13, 330, 367-8.
34Kealey, supra, note 13, 218, remarks that the “[Toronto Trades and Labour Council], the
Knights of Labour…, and their allies throughout the province enjoyed an intimacy with the
Mowat government unknown before the 1880’s and seldom seen after in Canadian working
class history.”
35Kealey, supra, note 13, makes several references to the presence of factory and employer
liability legislation as a labour demand, but does not discuss any concerted efforts that might
have been made to have such legislation enacted. Further, employer liability legislation was
first introduced by the opposition in 1885. Mowat was caught off guard and, in defence of the
failure of his government to introduce similar measures, claimed that workmen’s societies had
exerted pressure for other legislation, but that there had been no demand for employer liability
reform. See Legis. Debates, The Globe, February 27, 1885. In view of his close connections
with labour organizations his claim merits serious consideration.
McGILL LAW JOURNAL
[Vol. 29
was quite simple: the state now guaranteed that no worker could be legally
exposed to hazards that it judged to be unreasonable.
The replacement of the market by a political process of social decision-
making about health and safety conditions in the workplace can be viewed
as part of a general transformation of the role of the state in capitalist society.
The contractarian rule system selected by the courts was typical of a state
whose role was largely restricted to one of facilitating capitalist economic
activity through, inter alia, the provision of a legal infrastructure which
protected the core institutions of private property and market exchange.
However, to the extent that the market economy produced outcomes which
were unpopular, pressure was placed on the state to expand its functions,
both in order to direct the economy so as to avoid economic crises and to
secure the loyalty of the citizenry by, for example, ensuring that minimum
standards of living were maintained. Through various regulatory mecha-
nisms, which included direct policing, the state increasingly extended its
control over areas that were formerly considered private. As one of its
earliest acts in this new role the legislature imposed minimum standards of
health and safety in the workplace.
Why had the contract model failed to produce politically acceptable
results? Perhaps this was a case of classic market failure involving imperfect
information and high transaction costs? 36 The case for imperfect informa-
tion is not very convincing. Most problems of this era related to unsafe as
opposed to unhealthy workplace conditions, and safety experts have stated
that experts know little more about unsafe conditions than do workers. 37
Further, in many industries the labour processes remained artisanal; that
is, although workers no longer owned the means of production, they still
possessed the knowledge of the techniques of production and thus would
be as aware of the risks as the employer.38 What about transaction costs? If
361n microeconomic theory, exchange produces allocative efficiency only when there is perfect
competition and perfect markets. In order for perfect markets to operate, each individual must
have complete knowledge of all relevant factors and be able to engage in costless transactions.
In the context of health and safety, imperfect knowledge might exist where workers are ignorant
of the actual risks in the workplace. High transaction costs would arise simply by virtue of the
fact that it would be time-consuming to bargain with each individual employee. Where infor-
mation is imperfect or there are high transaction costs the market may fail to achieve allocative
efficiency. For a general discussion of the problem see J. Hirshleifer, Price Theory and Appli.
cations, 2d ed. (1980) 230-49.
37Mendeloff, Regulating Safety (1979) 14.
38Reliance on craft production techniques in many major industries persisted well into the
twentieth century. See Gordon, supra, note 16, 79-81.
It would not be surprising if we were to find that injury rates were lowest in firms and
industries that were heavily dependent on craft production techniques. Workers would have
the greatest control over production and organize it so as to minimize risks. Indeed, railways
and spinning mills, two industries which did not have to overcome craft traditions, were well
represented in the reported liability cases. If this hypothesis holds, it might also explain why
1984]
OCCUPATIONAL STANDARDS IN ONTARIO
277
indeed workers and employees were bargaining individually over general
terms and conditions of employment, there is little reason to believe that
additional bargaining over health and safety would have substantially in-
creased what would have already been high transaction costs. Alternatively,
if in reality practically no individual bargaining was taking place, then the
problem of marginally higher transaction costs that would be caused by the
addition of bargaining over health and safety is purely theoretical, and there-
fore irrelevant. 39
Rather than rely on classic market failure for an historical explanation
of the rejection of the contractual framework, I would suggest that the radical
inequality in bargaining power between employers and employees produced
results that were so unpopular that they threw the framework into disrepute.
Attempts to rectify the imbalance in bargaining power through workers’
collective action were fiercely resisted.40 Even the courts came to acknowl-
edge that the individual employment contract could not be realistically
treated as a voluntary exchange relationship. 41 The state responded to the
demands of workers, to substantive conceptions of fairness which were in
conflict with the market definition of justice, and perhaps to the long term
interests of capitalists as a class in not diminishing the available labour
power and in preserving social order.42 The conflicts generated in the realms
of production and exchange were transferred to the realm of the state.43
a labour movement dominated by the crafts would not have employer liability and Factory
Act legislation at the top of its political agenda.
39To the extent that the transaction costs of individual bargaining were prohibitive, collective
bargaining might have been seen as a mechanism for reducing these costs. Needless to say,
employers were not enthusiastic about this solution.
40For a good overview of trade union struggles see C. Lipton, The Trade Union Movement
of Canada: 1827-1959, 3d ed. (1973).
41For example, Wilson, C.J., of the Ontario Court of Queen’s Bench stated in a rare moment
of candour:
It is hard to apply that maxim [volenti non fit injuria] in many cases against a
workman when the choice before him is to run the risk or give up his work, or
starve.
Dean v. Ontario Cotton Mills Co. (1887), 14 O.R. 119, 128.
42The suggestion that nineteenth-century Factory Act legislation was passed in the interest
of the capitalist class as a whole originates with Marx’s discussion of English legislation limiting
the length of the working day. Marx was concerned with the question of why the Parliament
of the ruling class would pass such legislation. His explanation involved several elements,
including working class organization, but he also made the point that capitalists were physically
debilitating the available pool of industrial labour.
Apart from the working-class movement that daily grew more threatening, the lim-
iting of factory labour was dictated by the same necessity which spread guano over
the English fields. The same blind eagerness for plunder that in one case exhausted
the soil, had in the other, torn up by the roots the living force of the nation. Periodical
epidemics speak on this point as clearly as the diminishing military standard in
Germany and France.
K. Marx, Capital, vol. 1 (New York: International Publishers, 1967) 239. For similar comments
also see 269-70.
REVUE DE DROITDE McGILL
[Vol. 29
H. State Standard Setting in Nineteenth-Century Ontario
A. First Steps
The approach of the early legislation to standard-setting was quite sim-
ple: the hazard was identified and the corrective action to be taken was
stipulated. The first legislation directed at the regulation of workplace con-
ditions was enacted in 1874 and entitled An Act to require the owners of
Thrashing and other Machines to guard against accidents.44 The Act required
that owners and operators of machines connected to a “horse-power” by
certain means take specified measures to protect persons working on or near
those machines. The next workplace setting to receive legislative attention
was the railway. After some delay and study,45 the Railway Accidents Act,
1881 was approved by the Legislative Assembly. The preamble expressed
an attitude toward standards that stood in sharp contrast to the thinking
that still prevailed in the common law courts:
WHEREAS frequent accidents to railway servants and others are occasioned
by the neglect of railway companies to provide a fair and reasonable measure
of protection against their occurrence; and whereas a proper construction of
railway bridges and certain precautions in the construction and maintenance
of railway frogs, wing-rails, guard-rails, and freight cars would greatly lessen,
if not entirely prevent, the happening of such accidents.. .[emphasis added].46
The Act then went on to specify a series of measures which were to be
taken by railroads in order to satisfy the Legislative Assembly’s requirement
of fair and reasonable protection. These included a provision that there be
a clearance of seven feet between the bottoms of overhead bridges and the
tops of railway cars and that existing bridges were to be brought into com-
pliance at the owner’s expense within a year.
43Although my concern in this paper is with occupational health and safety, it is important
to keep in mind that the resolution of other aspects of employment relations were also trans-
ferred from the market to the State. Thus The Ontario Factories’ Act, S.O. 1884, c. 39, ss 4-
6, also contained provisions governing child labour, the employment of women and legal limits
on hours of work. Minimum wage laws were first introduced in Ontario in 1920 but only
applied to women (The Minimum WageAct, S.O. 1920, c. 87). For a somewhat sketchy account
of the development of employment standards legislation see P. Malles, Canadian Labour Stan-
dards in Law, Agreement, and Practice (Economic Council of Canada, 1976) 3-18.
44S.O. 1874, c. 12.
45An Act to protect brakemen was first introduced in 1875, but failed to get past second
reading. In 1877 a return of all railway accidents from 1874 to 1876 was called for and was
published in 1878. See Ontario SessionalPapers (1878), vol. 10, part 3, no. 14. In 1880 a Select
Standing Committee on railways was appointed and its report was published as Appendix no.
1 of vol. 13 of the Ontario Journals of the Legislative Assembly in 1880.
46S.O. 1881, c. 22.
1984]
OCCUPATIONAL STANDARDS IN ONTARIO
These early Acts show signs that the move toward a new legal frame-
work was still tentative and that politicians were uncertain as to just what
role they should play. Was the Legislature itself going to study each problem
in every industry and devise the appropriate technical solution? Indeed,
were only the hazards specifically identified by the Legislature going to be
dealt with by regulation, while the remainder would be left to be resolved
through contract? Most importantly, what mode of enforcement would be
selected to implement the decisions being made about standards?
The establishment of a full-time state inspectorate would have involved
a significant departure from the facilitative role that laissez-faire capitalism
envisioned for the state, in that it entails direct policing of private economic
activity. The first two Acts avoided taking that step, and instead relied on
enforcement mechanisms which might be of interest to contemporary reg-
ulatory reformers. The thrashing machines legislation relied on private pros-
ecutions which were encouraged by awarding the prosecutor half of the fine
that was imposed and on a curious provision which barred actions to recover
monies owed for services rendered by a machine which did not comply
with the statutory requirements.4 7 The only enforcement mechanism pro-
vided in the Railway Accidents Act, 1881 was that, subject to a number of
conditions, employees injured as a consequence of the employer’s negligent
non-compliance could sue for compensation as if they were not employees.48
Thus, a rational railway would only comply with the law where their ex-
pected losses under the new liability rules would exceed the cost of compliance.49
B. The Coming of Age
It was not until The Ontario Factories’ Act, 1884 that the Ontario Leg-
islature committed itself to a regulatory scheme that included both wide-
ranging standards and a full-time state inspectorate. 50 With regard to stan-
dards, the Act provided special protection for women and children, including
a limit on the length of the working day in Ontario, and restrictions on
tasks that women and children could be assigned. For factory employees of
47An Act to require the owners of Thrashing and other Machines to guard against accidents,
48S.O. 1881, c. 22, ss 7 and 8. Note that a ceiling was put on the amount of compensation
S.O. 1874, c. 12, ss 2-4.
that could be obtained under this Act.
49The reported case law provides no information on the extent to which employers complied
with these standards nor on the extent to which injured employees sought compensation under
The Railway Accidents Act, S.O. 1881, c. 22.
5oS.o. 1884, c. 39.
McGILL LAW JOURNAL
[Vol. 29
either sex and any age, section 14 of the Act contained a strongly-worded
general prohibition of unsafe and unhealthy work conditions:
It shall not be lawful to keep a factory so that the safety of any person employed
therein is endangered, or so that the health of any person therein is likely to
be permanently injured….51
This general penalty provision was supported by a number of more
specific provisions dealing with sanitary conditions, machine guarding and
fire prevention. 52 Breach of such a provision was deemed to be a breach of
section 14. It must be noted that many of these standards differed substan-
tially from the standards set in the earlier legislation. Rather than identifying
a specific hazard and selecting a particular solution, the Act tended to set
general standards, sometimes qualified by language like “reasonably prac-
ticable”, and left it to the inspector and/or the Lieutenant Governor in
Council (Cabinet) to determine the operative meaning of those standards
and how they could be complied with.53 Indeed, with regard to fire pre-
vention, where the Act provided more specific standards, inspectors were
given authority to dispense with any of those requirements altogether. If
section 14 were read to have meaning independent of the surrounding sec-
tions, one would have to conclude that the Act was internally inconsistent 54
With respect to enforcement, the Act established a permanent inspec-
torate to police factory conditions, and this was a major step in the process
of transforming the state from facilitator to regulator. However, the line
between standard-setting and enforcement is quite blurred under the Act.
The powers of the inspectors were so varied and so highly discretionary
that, in a practical sense, the two functions became merged in a single actor.
This phenomenon might be explained on the basis of administrative needs
and political convenience. From an administrative standpoint, the Legis-
lature understood that it was no longer competent to determine exactly what
SIThe Ontario Factories’Act, 1884, S.O. 1884, c. 39, s. 14.
52The Ontario Factories’Act, 1884, S.O. 1884, c. 39, ss 11-13, 15 and 16 respectively.
53For example, see The Ontario Factories’Act, 1884, S.O. 1884, c. 39, s. 12(l):
In every factory where, contrary to the provisions of this Act, there is any omission,
act, neglect or default in relation to any overcrowding, ventilation, drain, privy,
earth-closet, water-closet, ash-pit, water supply, nuisance or other matter whereby
the health of persons employed in the factory may be affected, the employer shall
within a reasonable time take such action thereon as the Inspector, acting under
the regulations made in respect to such subjects, notifies the employer to be proper
and necessary….
54 For example, with respect to ventilation s. 11(3) of The Ontario Factories’Act, 1884, S.O.
1884, c. 39 requires that “Every factory shall be ventilated in such a manner as to render
harmless, so far as is reasonably practicable, all the gases, vapours, dust or other impurities
generated … therein that may be injurious to health”.
If it had not been reasonably practicable to eliminate those injurious conditions would they
be tolerated under s. 14?
1984]
OCCUPATIONAL STANDARDS IN ONTARIO
should be done in each and every circumstance that might be encountered
in a factory setting, and chose to take advantage of the availability of a full-
time inspectorate by delegating to it the authority to make those determi-
nations. From a political perspective, the Liberals could tell their working
class supporters that they had taken action to require safe and healthy work-
places and that responsibility was now in the hands of experts, while at the
same time they could tell their industrialist friends that the Act was am-
biguously worded and would not be likely to impose unreasonable costs.
Once the state begins to assume direct responsibility for determining
conditions of employment, relations of production lose their apolitical form,
in the sense that they can no longer be said to be governed by a self-regulating
market economy. Political struggles in regard to state policy now have a
direct impact on the content of those relations. The structure of capitalist
relations of production may impose limits on the selection of state policies,
but within those limits there is a broad range of possible outcomes. Some
outcomes may be more or less favourable to accumulation than others, but
the organizational strength of different classes will have more to do with
the selection of particular policies than will the economic structure itself.
However, the economic structure of a society will have an influence on the
strength of different classes in the political sphere. First, a small number of
people who control a disproportionate share of the wealth of a society are
much easier to organize than a large number of persons who individually
do not command much economic power. Second, concentrated economic
power has a privileged position within the welfare state. As the state takes
over responsibility for steering the economy it is faced with the fact that
unless it is prepared to challenge private property, private investment de-
cisions will to a large extent determine economic performance. Thus, the
steering function of government often becomes one of providing incentives
to capitalists to invest. The public impact of private investment decisions
often leads toward an identification of the general interests of business with
those of society at large. On the other hand, the claims of the disenchanted
and displaced appear as special interests making claims against society as
a whole. According to this analysis, the state is not simply a pluralist in-
stitution reflecting the competing claims of different interest groups. Rather,
there is a structural tendency for the state to favour the particular interests
of large private investors over competing claims. 55
In this context the enactment of the Factories’ Act should be viewed
as an advance for labour, in that it signalled their ability to get the state to
intervene in the labour market on behalf of workers, and to expand its social
55In addition to the Marxist writings cited supra, note 5, an excellent non-Marxist explanation
of this phenomenon can be found in C. Lindblom, Politics and Markets (1977) 161-233.
REVUE DE DROIT DE McGILL
[Vol. 29
welfare functions. This was in part a reflection of their own organizational
capacity as well as of the fact that political parties were competing for the
newly enfranchised workers. The partial politicization of relations of pro-
duction created new legitimation problems which could only be resolved if
the state met enough of the demands of working class voters to retain their
loyalty. However, the advance was limited. The Factories’ Act was itself
inconsistent, the standards it set uncertain and its enforcement was highly
discretionary. 56 These weaknesses in the Act were not determined in any
simple way by the economic structure of late nineteenty-century Ontario,
but that structure affected the organizational capacity of different classes
and the weight their interests were assigned by the state in its selection of
specific policies.
Over the next thirty years, legislation of a similar character was enacted
to regulate workplace conditions in shops, 57 mines, 58 construction 59 and
offices. 60 This legislation remained in force with little or no change until
the 1960’s.
C. Note on Employer Liability and Standards
Regardless of the strength or weakness of the standards or the degree
of administrative enforcement, the state had adopted a public regulatory
scheme in which substantive decisions about what constituted reasonably
safe work conditions were being made. What then of employer’s liability?
Was the contractarian legal framework still to govern the determination of
whether or not an injured workman was to be compensated as a result of
a workplace injury? What if the injury was caused by a failure to comply
with the statutory standards? What if the employee knew of that failure and
still continued to work notwithstanding the violation? Was liability for in-
juries still regarded as a mechanism in addition to the Factories’ Act for
56Marx, supra, note 42, 494, in discussing English Factory Act legislation, observed a similar
phenomenon:
What strikes us, then, in the English legislation of 1867, is, on the one hand, the
necessity imposed on the Parliament of the ruling classes, of adopting in principle
measures so extraordinary, and on so great a scale, against the excesses of capitalist
exploitation; and on the other hand, the hesitation; the repugnance, and the bad
faith, with which it lent itself to the task of carrying those measures into practice.
57The Ontario Shops’ Regulation Act, 1888, S.O. 1888, c. 33; The Bake Shops’ Act, 1896,
58The Mining Operations Act, 1890, S.O. 1890, c. 10; The Mines Act, 1892, S.O. 1892, c. 9
59The Buildings Trades Protection Act, S.O. 1911, c. 71.
60The Factory, Shop and Office Building Act, S.O. 1913, c. 60 (s. 85 of this Act is the repeal
(s. 76 of this Act repeals The Mining Operations Act, 1890).
S.O. 1896, c. 64.
section).
1984]
OCCUPATIONAL STANDARDS IN ONTARIO
establishing conditions of employment? A brief examination of the reso-
lution of these problems follows.
The contractual principles of the common law were largely replaced by
legislation imposing negligence rules in 1886.61 The fellow-servant doctrine
was severely circumscribed and the presumption of voluntary assumption
of risk was weakened. Thus, employees could recover for injuries caused
by the negligence of most fellow-servants and for defects in the production
process in respect of which the employer was negligent. The courts, and
subsequently the Ontario Legislative Assembly, specified that prior knowl-
edge of a defect by the worker did not give rise to a presumption that she
had voluntarily assumed the risk. 62 Failure to comply with the statutory
standards of the Factories’ Act was held not to give rise to strict liability,
but rather provided evidence of negligence to go to the jury. In practice,
however, it seemed as if the liability of the employer for injuries caused by
violations of the Act was strict, although that verbal formulation was avoided. 63
In 1899 the Legislature placed the onus on the defendant who was alleged
not to have complied with the Factories’ Act to disprove the allegation if
he was to avoid liability for an injury caused by the alleged defect. 6 4
How then did employers’ liability relate to the establishment of health
and safety conditions in the workplace? The dominant view, and the policy
expressed in the legislation, was that the issues were separate. The Factories’
Act required that all workplaces be kept in a safe condition65 and provided
an inspectorate to enforce those standards. Compensation was payable where
the employer negligently failed to provide safe working conditions. The view
that health and safety standards were to be determined primarily by the
legislature was clearly expressed by Professor James Mayor in his study of
the desirability of introducing no-fault workers’ compensation in 1900:
If accidents occur which are preventable by means of legislation, humanity
would determine that legislation should be passed without delay. But this is
not a case of that kind. It is rather a question of the distribution of the costs
of accidents which ex hypothesi are not preventable. 66
61The Workmen’s Compensation for Injuries Act, 1886, S.O. 1886, c. 28.
62Dean v. Ontario Cotton Mills Co. (1887) 14 O.R. 119, 125-6 (Q.B.) per Armour, J.; The
Workmen’s Compensation for Injuries Amendment Act, 1889, S.O. 1889, c. 23, s. 7.
631n Finlay v. Miscampbell (1890) 20 O.R. 29, 38 (Ch.D.) the Court explicitly rejected the
proposition that a breach of The Ontario Factories’Act, 1884, S.O. 1884, c. 39 would by itself
give rise to liability. Nevertheless, the finding of a breach was considered evidence of negligence:
McCloherty v. Gale Manufacturing Co. (1892) 19 O.A.R. 117, 120.
64The Workmen’s Compensation For Injuries Act, 1899, S.O. 1899, c. 18, s. 3.
65The Ontario Factories’ Act, 1884, S.O. 1884, c. 39, s. 14.
66J. Mavor, “Report on Workmen’s Compensation for Injuries” in Ontario Sessional Papers
(1900) no. 48, 46-7.
McGILL LAW JOURNAL
[Vol. 29
Actual practice may have been quite different. The standards estab-
lished by the Factories’ Act were uncertain and their enforcement highly
discretionary. As such, the actual level of safety in the workplace may have
reflected economic rather than political forces. If compliance with the Fac-
tories’Act was not voluntary or enforced, then the actual level of safety may
still have been determined by the minimization of the sum of the employer’s
accident costs and accident prevention costs, where accident costs include
not only lost production time, but also ex ante compensation for injury
according to the new employer liability rules. The enactment of no-fault
compensation in 191467 would have affected the actual quality of health
and safety conditions to the extent that it caused a net increase in accident
costs, thereby creating an incentive to invest in additional accident prevention.
In sum, the legislative reforms of the nineteenth century only partially
politicized the wage relationship, still leaving economic forces a direct role
in the determination of health and safety conditions.
M. Interregnum: 1914-1960
This pattern of relatively weak health and safety legislation in con-
junction with no-fault compensation stood for nearly fifty years. Why? The
question is intriguing, and has been asked more generally with respect to
the overall inability of Canadian workers to make significant gains in in-
dustrial relations, either through legislative action or collective bargaining,
from 1908 to 1943.68 There are several factors which would need to be
considered in any attempt to provide a comprehensive answer to the ques-
tion. Many of the early labour groups were organized on craft lines. This
was appropriate in a society in which the labour process still relied on skilled
workers, even though they were now brought together under one roof. How-
ever, by the end of the nineteenth century, efforts were being made to trans-
form the labour process. This transformation involved increasing mechanisation
and supervision, so that knowledge of production techniques became vested
in the employer as opposed to the skilled labourer. The benefits to an em-
ployer were clear. With control over the production process employers were
able to increase productivity per worker, eliminate skilled workers (who
tended to be unionized) and replace them by “homogenized” labour which
could be drawn from the general pool of wage labourers at lower cost.69
Craft unions faced with a challenge to their existence fought rear-guard
67The Workmen’s Compensation Act, S.O. 1914, c. 25.
68Pentland, supra, note 30, 160.
69For accounts of the transformation of the labour process in the late nineteenth and early
twentieth centuries, see Gordon, supra, note 16, 112-35 and H. Baverman, Labour and Mo-
nopoly Capital [:] The Degradation of Work in the Twentieth Century (1975) 85-138.
1984]
OCCUPATIONAL STANDARDS IN ONTARIO
actions to protect their diminishing membership rather than support in-
dustrial unionism. 70
The growth of a general pool of industrial labourers did give rise to
industrial unions, but their organizing efforts were strenuously resisted by
Canadian employers. Their relative weakness was reinforced by divisions
within the labour movement, and this left the labour movement as a whole
unable to respond to management’s offensive to gain stronger control over
labour in the production process. 7 1 The onset of the depression hardly cre-
ated conditions that would be conducive to worker initiatives for safer and
healthier work environments.72 Rather, struggles for survival dominated
labour’s agenda until between the two World Wars.
World War II provided a golden opportunity for the labour movement
to regain the strength it had developed in the late nineteenth century. The
economy was stimulated, manufacture expanded and the domestic civilian
labour supply tightened. Union membership doubled between 1939 and
1944 and the number of workers involved in labour disputes dramatically
increased between 1937 and 1943. In response to these pressures the wartime
Government issued PC. 1003, which laid the legislative foundations for the
industrial relations system that emerged from the War.73 Essentially, the
provincial labour relations statutes which followed provided for union cert-
ification, an obligation on employers to bargain in good faith with certified
unions, and a grievance system for resolution of contract disputes as well
as for enforcement of collective agreements. 74 Under this regime labour was
prohibited from taking collective action in the form of a strike during the
life of the collective agreement. There was also a tacit acceptance of man-
agerial rights to control the production process, which included choice of
technology and direction of the work force. Unions focused their demands
on improving wages, benefits and security provisions, leaving management
to manage. This basic trade-off, combined with a recession in the 1950’s,
provide some explanation of the fact that occupational health and safety
did not emerge as a substantial demand of labour following the War. Indeed,
it was not until the Hogg’s Hollow disaster of 1960 in which five workers
were killed that the inadequacy of the existing regulatory system was brought
to light, both with respect to state enforcement and employer responsibility.
The coroner’s jury reached the following conclusions:
The death of Pasquale Allegrezza and his fellow workers is the inevitable result
of the failure to implement and enforce regulations made under the Department
70See Gordon, supra, note 16, 159-60 and Pentland, supra, note 30, 116-8.
71See Gordon, supra, note 16, 143-4, 157-9 and Pentland, supra, note 30, 120-4.
72See Pentland, supra, note 30, 170-91.
73See Pentland, supra, note 30, 192-224.
74See, e.g., The Labour Relations Act, 1950, S.O. 1950, c. 34.
REVUE DE DROIT DE McGILL
[Vol. 29
of Labour Act governing the protection of persons working in compressed
air….
According to the evidence presented, almost all the safety regulations governing
this tunnel project were violated at one time or another, and many of the
regulations were violated continuously. The attitude of management towards
the safety of the individual worker can be described as no less than callous.75
IV. The Second Wave of State Regulation: 1961-1980
Increasing state intervention in the workplace is not merely a local
event, but has been common in many western industrialized countries,
including the United States. 76 Nor have demands for increased state inter-
vention been restricted to the workplace. The environmental movement is
often pointed to as one of the factors that spurred demand for workplace
safety, and helped create a receptive political environment. 77 Since the Hogg’s
Hollow inquest we have seen state activity in occupational health and safety
regulation increase dramatically. There have been numerous Royal
Commissions 78 and much legislative reform, 79 culminating in the Occu-
pational Health and Safety Act, 1978.80
The Act contains a number of innovations as compared with its nine-
teenth-century predecessors. These innovations reflect the orientation of the
Ham Report 81 and its emphasis on the importance of establishing an internal
responsibility system to complement the external policing mechanism which
was available through government inspection and prosecution. The internal
responsibility system consisted of direct accountability of management for
workplace safety, and was supplemented by mechanisms that provided for
worker contribution and participation in the development of safe work con-
ditions. This was possible through joint health and safety committees, work-
er’s inspections, increased worker access to information and the right to
refuse unsafe work. It was thought that health and safety would be best
75The verdict of the coroner’s jury in the Hogg’s Hollow inquiry was reproduced in P. McAndrew
(Chairman), Report of the Royal Commission on Industrial Safety (Ontario, 1961) Appendix
-E.76The major legislation in the United States is the Occupational Health and Safety Act of
1970, 29 U.S.C. 651 et seq. (1970).
77N. Ashford, Crisis in the Workplace: Occupational Disease and Injury (1976) 46.
78J. Ham (Chairman), Report of the Royal Commission on the Health and Safety of Workers
in Mines (Ontario, 1976) and K. Burkett (Chairman), Towards Safe Production, (Report of the
Joint Federal-Provincial Inquiry Commission into Safety in Mines and Mining Plants in Ontario,
1981).
79Construction Safety Act, 1961-62, S.O. 1961-62, c. 18; Industrial Safety Act, 1964, S.O. 1964,
c. 45. These Acts were amended and consolidated and then replaced by The Employees’Health
and Safety Act, 1976, S.O. 1976 (2d Sess.), c.79.
80S.O. 1978, c. 83 as consolidated by R.S.O. 1980, c. 321.
81See supra, note 78.
1984]
OCCUPATIONAL STANDARDS IN ONTARIO
achieved if the problem of regulation were approached cooperatively by
labour, management and the state rather than on a confrontational basis. 82
Although it is clear that the actual levels of health and safety achieved
in the workplace may reflect discrete bargains that are struck at the level of
enforcement, rather than legislative or regulatory standards, more empirical
work on the effect of these innovations is required before the utility of the
Act can be assessed. Furthermore, the internal responsibility system, as its
name implies, shifts at least some of the responsibility that the state had
previously assumed back to the workplace, thus making the issue of safety
standards less public. While this development is significant and will be
adverted to later on, it is necessary to remember that the state has still
retained ultimate responsibility for defining the minimally safe health con-
ditions which all employers must provide. This is particularly true with
respect to exposure to toxic substances, and it is in determining maximum
exposure levels that conflicts over the role of the state have been made most
visible. In the following sections I will examine the process of standard-
setting and debates over appropriate policy in this area, with a particular
emphasis on the proposal to adopt cost-benefit analysis as a measure of
regulatory reform. The terms of this debate will help elucidate the tensions
generated in the modern state by the demand that it simultaneously facilitate
capital’s accumulation of wealth and maintain its legitimacy by ensuring
“fair” outcomes of exchange transactions.
A. Process
The distinction between policy and process is particularly sharp in the
Occupational Health and Safety Act, although perhaps no more so than in
other regulatory legislation in Ontario. To put it simply, the Act contains
no substantive policy content whatsoever with respect to toxic substance
exposure standards. A comparison of the Ontario legislation with its United
States counterpart should make this point quite clear.
The United States Congress delared that its purpose and policy in pass-
ing the Occupational Health and Safety Act of 1970 was “to assure so far
as possible every working man and woman in the Nation safe and healthful
working conditions and to preserve our human resources”. 83 With respect
to toxic material standards, the American legislation directs the Secretary
of Labour to “set the standard which most adequately assures, to the extent
82For a recent evaluation of these aspects oftheAct see Doern, supra, note 4, 3.37-3.48, 3.60-
3.76, and generally Swinton, “Enforcement of Occupational Health and Safety Legislation: The
Role of the Internal Responsibility System” in Swan & Swinton, eds, Studies in Labour Law
(1983) 143-75.
8329 U.S.C. 651 (1970).
McGILL LAW JOURNAL
[Vol. 29
feasible, on the basis of the best available evidence, that no employee will
suffer material impairment of health or functional capacity even if such
employee has regular exposure to the hazard dealt with by such standard
for the period of his working life”, s4 The Supreme Court of the United
States has recently interpreted that section in American Textile Manufac-
turers Institute, Inc. v. Donovan:
In effect then, as the Court of Appeals held, Congress itself defined the basic
relationship between costs and benefits, by placing the “benefit” of worker
health above all other considerations save those making attainment of this
“benefit” unachievable. 85
The Ontario legislation provides no such direction. Employers and su-
pervisors are directed to “take every precaution reasonable in the circum-
stances for the protection of the worker”. 86 With respect to general regulation-
making authority the Act provides:
The Lieutenant Governor in Council may make such regulations as are ad-
visable for the health or safety of persons in or about a workplace.87
Authority to make regulations with respect to toxic substances is then
specifically granted.88 Nowhere in the Act does the Legislature express a
view on the crucial policy issues that must be addressed in regulating ex-
posure to toxic substances in the workplace. What are the relevant consid-
erations in setting standards? What weight should be given to costs? Should
Cabinet be risk averse? “Reasonableness” or “advisability” are devoid of
meaning in the absence of some evaluative criteria. In sum, the Legislature
has delegated an unconfined discretion to Cabinet to formulate its own
policy, subject only to the procedural constraints of the regulatory process.
There is no general statutory requirement in Ontario for public partic-
ipation in the regulatory process similar to the notice and comment pro-
cedures in the Administrative Procedure Act in the United States.8 9 Further,
English and Canadian courts have yet to attach natural justice or fairness
84 Occupational Health and Safety Act of 1970, 29 U.S.C. 655(b)(5) (1970).
85Supra, note 3, 509.
86 Occupational Health and Safety Act, R.S.O. 1980, c. 321, ss 14(2)(c) and 14(2)(g).
87Occupational Health and Safety Act, R.S.O. 1980, c. 321, s. 41.
88Occupational Health and Safety Act, R.S.O. 1980, c. 321, s. 41(2)(13)-(14).
895 U.S.C. 553 (1976).
1984]
OCCUPATIONAL STANDARDS IN ONTARIO
requirements to legislative functions.90 Yet, in the Occupational Health and
Safety Act, the Legislature took steps to open up the regulatory process.
Section 22 requires that, prior to designating a toxic substance, the Minister
of Labour publish a notice that the substance may be so designated and call
for briefs or submissions. Further, the proposed regulation is to be published
at least sixty days before it is filed. The Act also establishes a standing
Advisory Council on Occupational Health and Occupational Safety (ACO-
HOS) to advise and make recommendations to the Minister. This body
represents labour, management, technical and professional persons and the
public, and provides another potential source of public input into the reg-
ulatory process.91
The only other procedural constraint on the process is the Regulations
Act which requires that a designated member of the Executive Council of
Cabinet review all proposed regulations. This legislation also establishes a
Standing Committee on Regulations in the Legislature with authority to
ensure that regulations are not ultra vires. The Committee is specifically
precluded from conducting a policy review.92
In practice, the Ministry of Labour has developed an even more open
process than is required by the Act. In particular, a public meeting is held
in which the Ministry explains how and why it has or has not responded
to submissions by making revisions to the proposed regulation. ACOHOS
is then consulted with respect to any changes made after that public meeting.93
Clearly, the setting of designated substance standards in Ontario is a political
process, in which the state participates, subject to the conflicting demands
to facilitate accumulation and to legitimate the social relations of produc-
tion. This is not to say that technical expertise does not play a significant
role in laying the foundations and establishing the boundary conditions
within which choices will be made. However, when it comes to making
choices between alternatives, the process seems to encourage bargaining in
9The decision of the Supreme Court of Canada in A.G. Canada v. Inuit Tapirisat of Canada
[1980] 2 S.C.R. 735, (1981) 115 D.L.R. (3d) 1, confirms that the Court is disinclined to impose
procedural requirements on the exercise of a delegated power to make general decisions based
on matters of public policy and convenience. See also Momex Realty and Development Co. v.
Village of Wyoming [1980] 2 S.C.R. 1011, (1981) 116 D.L.R. (3d) 1. More recently, the Supreme
Court considered the fact that Cabinet had given persons affected a reasonable right to be heard
as a ground for not strildng down an Order in Council where it was alleged to have been made
in bad faith. See Thorne’s Hardware Ltd. v. The Queen (1983) 143 D.L.R. (3d) 577, (1983) 46
N.R. 91 (S.C.C.). While this decision indicates judicial approval of such a practice, it does not
suggest that the Court will be prepared to impose an implied right to be heard. In general, see
J. Evans, de Smith’s Judicial Review ofAdministrative Action, 4th ed. (1980) 181-2 and R. Reid
& H. David, Administrative Law and Practice (1978) 97-8.
91Occupational Health and Safety Act, R.S.O. 1980, c. 321, s. 10.
92R.S.O. 1980, c. 446, ss 6, 12.
93See Doern, supra, note 4, 3.25-3.27.
REVUE DE DROIT DE McGILL
[Vol. 29
which both labour and management are granted concessions. A recent study
of the process of setting an asbestos standard described the strategy of the
Ministry as one of “splitting the difference”. 94
The politicization of the process of standard-setting is, at its core, a
continuation of the developments we saw in the late nineteenth century.
However, there are a number of significant changes. First, the Legislature
sets fewer specific standards. Indeed, in the present Act, the Legislature did
not even set a general standard similar to section 14 of the 1884 Factories’
Act. Instead it set up a process through which general policies and specific
standards would be determined. In part this reflects an institutional con-
straint. The Legislature has neither the time nor the expertise to resolve
each particular problem. But political considerations may play a role as well.
Even with an open regulatory process, the controversy generated by such
proceedings is less likely to receive the same level of publicity as proceedings
conducted in the Legislature. For example, the vinyl chloride exposure level
was set according to the principle of acceptable risk, which in this instance
was defined as 50 excess deaths per 100,000 exposed workers. 95 One can
imagine the publicity and outrage that would have been generated in an
open legislative debate over the quantity of excess worker deaths that was
“acceptable”. Further, in dealing with substances individually without leg-
islatively mandated decision criteria, the Ministry has chosen different cri-
teria for selecting exposure levels for different substances. 96 This kind of ad
hoc decision-making may also provide a useful way of avoiding serious
political confrontations over fundamental principles.
Secondly, the regulatory process reflects the fact that organized labour
has more or less been accepted as a permanent feature of the industrial
relations system. Although the state still mediates class conflict, the means
by which it does so has shifted: collective bargaining now provides an al-
ternative to direct state intervention in dispute resolution. This is possible
because the collective bargaining process, in part at least, rectifies the severe
imbalance that was present in the negotiation of an individual contract of
employment. Thus, conflicts over workplace health and safety can be fed
back into an extension of the bargaining system (Joint Committees), with
the state monitoring and mediating through the mechanism of the inspector.
94See Doern, supra, note 4, 3.32.
950. Reg. 516/82. For a discussion of the analytic process used in setting this standard see
Advisory Council on Occupational Health and Occupational Safety (ACOHOS), Fifth Annual
Report (Ontario, 1983), vol. 1, memo 82-I.
96The shifting basis on which maximum levels of exposure to toxic substances are set has
been noted by ACOHOS in its fourth and fifth annual reports. See generally ACOHOS, Fourth
Annual Report (Ontario, 1981) and Fifth Annual Report (Ontario, 1982-83).
1984]
OCCUPATIONAL STANDARDS IN ONTARIO
As well, at the level of formal standard-setting, because a trade union pres-
ence has been institutionalized, the state can act as a mediator in a form
of industry-wide or multi-industry-wide bargaining between labour and
management, rather than appearing as an intervenor after a bargaining pro-
cess has already failed to reach substantively acceptable results.
Finally, there are many difficult technical and scientific issues involved
in toxic substance regulation. If major problems can be said to require
scientific determinations rather than political choices, conflict can be min-
imized. An emphasis on expertise tends to transform effective public par-
ticipation in the decision-making process into public “education”. 97
B. Policy
So far I have focused on the procedural aspects of standard setting,
without addressing the question of substantive policy other than to point
out that the legislation itself failed to articulate any particular policy goal(s).
But because substantive policy and process are clearly inter-connected, this
focus may be distorting. If your policy preference is that each standard be
determined through political negotiation mediated by the state, then you
would defeat your purpose by stating a preference for one result rather than
another in advance of the outcome of those negotiations. Similarly, if you
thought that substantive policy should be formulated solely on the basis of
technical criteria you might design a process which minimized the oppor-
tunities to bring political pressure to bear by, for example, assigning the task
to a professional body. Thus, by selecting a generally open regulatory pro-
cess, the Legislature may have indicated a policy preference: that standards
be politically determined from time to time. However, this will not satisfy
those who insist good administration requires that future decision-making
be guided by some basic policy choices, rather than determined by ad hoc
political considerations. A policy decision to promulgate the most protective
standard feasible, for example, could be applied to particular future cases,
without the risk that worker safety would be compromised by concessions
to political pressures.
In an attempt to rationalize its own decision-making the Ontario Cab-
inet has issued guidelines governing the submission of policy proposals and
regulations for Cabinet approval. 98 One of those guidelines calls for the
preparation of an economic impact statement, to include an assessment of
97Bryden, Public Input into policy-making and administration: the present situation and some
requirements for the future (1982) 25 Can. Pub. Admin. 81.
98Guidelines for Preparing Cabinet Submissions, cited in G. Doern, Rationalizing the Reg-
ulatory Decision-Making Process: The Prospects for Reform (Economic Council of Canada,
Regulation Conference, Working Paper No. 2, 1979) 94-6.
McGILL LAW JOURNAL
[Vol. 29
the consequences that the regulation is expected to have on job creation
and job loss, capital investment, consumer prices, work incentives and cost
of compliance to private industry. In addition, the submission requires a
statement of the problem, background and options, and must address po-
litical considerations such as the source of the demand for regulation and
the likely response of those who are to be governed by it. These guidelines
seem to indicate a pragmatic approach to Cabinet decision-making which,
as we shall see, differs considerably from the approach of advocates of cost-
benefit analysis. It should be noted that the Cabinet guideline regarding
preparation of an economic impact statement was not followed by the Min-
istry in preparing its asbestos regulation. 99
If the Legislature or Cabinet decide to adopt a substantive policy with
respect to permissible exposure to risk several options are open. One pos-
sibility is the “no-risk” principle, which would ban substances from the
workplace unless they could be demonstrated to have no harmful effect.
This policy underlies the Delaney Amendment to the Federal Food, Drug,
and Cosmetic Act’ 00 in the United States which prohibits food additives
found to be carcinogenic in either animals or humans.’ 0′ A second alter-
native is the “safest standard feasible” test adopted by the United States
Congress.102 A third alternative is that acceptable levels of risk should be
determined by cost-benefit analysis. This approach has received much sup-
port in the United States, 103 and has its Canadian advocates as well, although
enthusiasm for exclusive reliance on this technique is tempered. 0 4
It is unlikely that any of the above approaches will be adopted formally
as government policy in Ontario in the near future. Nonetheless, cost-benefit
analysis is of particular interest because it forms part of a strong version of
an argument in favour of shifting the direction of state intervention towards
greater concern with the creation of conditions favourable to capital ac-
cumulation at the expense of workers and other vulnerable groups that some
governments are finding persuasive. Further, arguments in favour of such
a shift are often presented as if what was at stake was more a question of
administrative rationality than of basic political choices over the direction
of public policy. My decision to subject cost-berfefit analysis to close scrutiny
is not only politically motivated; I do wish to provide arguments against
99Doern, supra, note 4, 3.33.
1021 U.S.C. 301 et seq. (1976).
’01FoodAdditives Amendment of 1958, Pub. L. No. 85-929, codified at 21 U.S.C. 347 (1976).
102See supra, notes 84-5 and accompanying text.
103See supra, note 3.
‘4See Doern, supra, note 4; also see Tuohy, Regulation and Scientific Complexity: Decision
Rules and Processes in the Occupational Health Arena (1982) 20 Osgoode Hall L.J. 562, 609.
While Tuohy is sensitive to the limitations of cost-benefit analysis, he sees a role for the
technique as an adjunct to deliberative and participatory processes.
1984]
OCCUPATIONAL STANDARDS IN ONTARIO
its adoption, but I also believe that the discussion will provide an oppor-
tunity to classify and elaborate the issues involved in regulating exposure
to unsafe and particularly to unhealthy conditions in the workplace. In the
following sections I argue that not only is cost-benefit analysis technically
inadequate on its own terms, but that it calls for a radical shift away from
the welfare state model that has dominated modem Canadian political practice.
1. The Approach of Cost-Benefit Analysis
Cost-benefit analysis is a technique for evaluating legal rules so that it
may be seen if the net benefits flowing from a rule can be said to exceed
its costs. The best choice will be the rule which generates the highest net
surplus of benefits over costs. In the field of accident liability law, Calabresi
has given us the classic formulation of the goal of this technique. He tells
us that leaving questions of justice aside, it is axiomatic that a rational goal
of accident policy is to minimize the sum of accident costs and accident
prevention costs. 0 5 If we apply this approach to occupational health and
safety we would say that higher levels of protection should be mandated
only to the extent that it costs less to prevent damage to workers than to
allow the damage to occur. At that point resources are most efficiently al-
located: additional investment in protective measures would produce as
much of a net social loss as would insufficient investment in prevention. 0 6
While this formula applies with regard to both health and safety, my focus
will be on health hazards in the workplace because it is in the context of
toxic substance exposure standards that most of the debate has taken place.
a. The Adequacy of the Market
It is a touchstone of economic theory that under certain conditions the
market will produce an efficient allocation of resources given an intial dis-
tribution of property. 0 7 Under what conditions will the market produce an
efficient allocation of resources to health and safety?
’05G. Calabresi, The Costs of Accidents (1970) 26.
106Mendeloff, supra, note 37, 7.
’07Adam Smith, in An Inquiry into the Nature & Causes of the Wealth of Nations (London:
Strahan Cadell, 1793), vol. 2, 181, gave a classic formulation:
But it is only for the sake of profit that any man employs a capital in the support
of industry; and he will always, therefore, endeavour to employ it in the support
of that industry of which the produce is likely to be of the greatest value, or to
exchange it for the greatest quantity of either money or of other goods …. [H]e is
in this, as in many other cases led by an invisible hand to promote an end which
was no part of his intention. Nor is it always the worse for the society that it was
no part of it. By pursuing his own interest he frequently promotes that of the society
more effectually than when he really intends to promote it.
REVUE DE DROIT DE McGILL
[Vol. 29
Mendeloff says that the adequacy of the market depends on two as-
sumptions: first, that firms know enough about health to choose the most
cost-effective technique of prevention; second, that all production costs are
imposed on the firm. 10 8 The first assumption might be broken down into
an assertion about the need for adequate, low-cost information (if not com-
plete, costless information) and a requirement that employers act rationally
as profit maximizers. The second assumption requires a mechanism that
will impose the full costs of production on employers. That mechanism is
the labour market. Other things being equal, workers will demand premiums
in order to engage in more hazardous work. The quantum of those premiums
will reflect the risk tastes of the employees. If the costs of reducing the level
of risk in the workplace are less than the amount that would have to be
paid out as risk premiums, then the employer will improve conditions of
work (again if the first assumption is correct). Instead of risk premiums,
one might also agree to provide ex post compensation. The particular ar-
rangement can be negotiated between employer and employee. To what
extent are these conditions satisfied by actual markets?
b.
Information
Nichols and Zeckhauser identified three requirements which must be
satisfied with respect to information if the market is to function efficiently:
“the information must be available, it must be transmitted to the affected
parties, and individual workers and firms must be able to understand it.” 109
In the field of occupational health, problems exist at every level. First, with
regard to the availability of information, it is clear that both employers and
employees do not have complete information about the health risks present
in the workplace. The cost of such information is frequently quite high and
the potential for severe free rider problems would serve as a further dis-
incentive to its production. 110 Indeed, employers have incentives not to
produce information about workplace hazards in order to avoid insurance
or risk premiums associated with exposure to such hazards. Employees
(1977) 49 The Public Interest 39, 43.
08Supra, note 37, 7.
’09Nichols & Zeckhauser, Government Cdmes to the Workplace: An Assessment of OSHA
10Free rider problems arise where investments produce public goods, that is, goods that can
be enjoyed by non-investors as well as investors. A dilemma is created. A firm that is deciding
whether or not to invest will consider the possibility that someone else might bear the cost
and that they will enjoy the benefit for free, or that if they bear the cost themselves, others
will get the benefit for free and gain a competitive advantage. If these problems cannot be
resolved by internal organization the result is likely to be an under-investment in the production
of public goods. For a general discussion see Hirshleifer, supra, note 36, 561-5. For a classic
treatment of the implications of this problem see M. Olson, The Logic of Collective Action
(1971).
1984]
OCCUPATIONAL STANDARDS IN ONTARIO
generally do not gain knowledge about health hazards from on-the-job ex-
perience in the same way as they gain knowledge about safety hazards. Often
there are lengthy latency periods before the effect of exposure manifests
itself and then it may be difficult to link the disease to the occupational
exposure if there is a fairly high background level of the disease in the
population at large. Finally, even if greater efforts were made to identify
hazards, it is not clear that we have the scientific knowledge necessary to
provide definitive answers about the health effects of exposure to all sub-
stances and processes.
Once information is produced, it must be transmitted if market forces
are going to respond to it. There are numerous barriers to the transmission
of information. The protection of trade secrets may prevent both employers
and employees from discovering the chemical composition of the substances
being used in the workplace. Beyond trade secrets, producers of toxic sub-
stances have other incentives for suppressing information about the health
effects of their products. Consumers might switch to less hazardous substi-
tutes and third parties might sue to recover damages. Finally, employers
have incentives to suppress knowledge of hazards from their employees,
again to avoid having to bear the costs of the consequences of exposure to
those hazards.” I
Even if the knowledge is produced and transmitted, the recipients of
that information must be capable of comprehending it. Most of the infor-
mation about health hazards comes in the form of probabilities. Employees
may have difficulty processing the fact that their risk of developing a form
of cancer is increased from two in a thousand to four in a thousand as a
result of exposure to a toxic substance in the workplace. The sum of these
information problems points to the strong possibility of market failure.
” ‘One instance of such a cover up was thoroughly documented by P. Brodeur in Expendable
Americans (1974). It involved asbestos workers in Tyler, Texas who were never informed by
local doctors to whom they were referred by the company that their chest ailments might be
related to asbestos.
An even more shocking case of industry suppression of information of the effects of asbestos
and the concealment from workers of the fact that they had contracted asbestosis was revealed
in a series of documents dubbed the Asbestos Pentagon Papers, which came to light in the
course of pre-trial discovery proceedings in a products liability action. One of those incidents
involved a medical survey conducted by Johns-Manville of workers in one of their Canadian
plants. Seven workers were positively identified as having asbestosis but the medical director
deemed it inadvisable to warn the men of their peril. See S. Epstein, The Politics of Cancer
(San Francisco: Sierra Club, 1978) 89-94.
296
McGILL LAW JOURNAL
[Vol. 29
c. Risk Premiums
Beyond flawed information are there other defects in the labour market
contributing to market failure? One such defect might lie in imperfect mo-
bility. The structure of the labour market might be so fragmented as to
impede the movement of workers across trade or sectoral boundaries.”12
Workers may not be able to choose the characteristics of their jobs inde-
pendently of each other, so that risk tastes are not truly reflected in job
choices, given the market structure of employment possibilites.”13
Given all of these factors, it is not surprising that empirical evidence
on the existence and size of risk premiums is far from conclusive.” 14 Men-
deloff reviewed the literature and found that it offers some weak evidence
of risk premiums as a component of wages. 15 Gunderson and Swinton, in
a more recent survey, found that studies on the whole tend to confirm the
existence of risk premiums but that this was not a firm conclusion of all
the studies. 16 Even if risk premiums exist, most authors are hesitant to
draw the conclusion that they are an accurate reflection of workers’ risk
tastes, especially with respect to health hazards. 17 To the extent that the
wage bargain does not adequately reflect these concerns, the firm is not
being required to bear the true costs of impaired worker health and therefore
will be under-investing in protection.”i8 If the worker is unable to bear these
Gordon, supra, note 16, 170-215.
” 2The segmentation of labour markets in the period after World War II is emphasized in
113M. Gunderson & K. Swinton, Collective Bargaining and Asbestos Dangers at the Work-
place, (Study Paper prepared for the Royal Commission on Matters of Health and Safety Arising
from the Use of Asbestos in Ontario, 1980) 4.7. They also discuss the problem of labour
immobility in a stagnant or declining economy at 4.8-4.9.
‘ 4Although the technical sophistication of the studies far surpasses the kinds of casual
observations that were relied upon in the nineteenth century, I am still not sure that the debate
has moved much beyond the following exchange with Lord Bramwell that occurred in the
course of his testimony cited supra, note 25, 201. Lord Bramwell had asserted that risk pre-
miums exist and was challenged on this point:
1164. Will you give me an instance?-Ceteris paribus surely the more disagreeable
a trade the higher the pay will be.
1165. Give me an instance at the present moment existing of a dangerous employ-
ment where the men have better pay?-I give every instance. An agricultural la-
bourer gets less than a miner.
t15Supra, note 37, 11.
116Supra, note 113, Appendix A.
117See e.g., Mendeloff, supra, note 37 and Nichols & Zeckhauser, supra, note 109, 43.
1 SIt is also possible that structural rigidities in wage rates cause employers to pay too much.
An employer who introduces a new technology that reduces the risk of a particular job may
find that he is unable to easily alter the existing structure of wage rates within the firm or
industry and thus is unable to reduce costs. This could result in a future decision not to introduce
safer technology if the only advantage contemplated is a reduction in risk premiums.
19841
OCCUPATIONAL STANDARDS IN ONTARIO
costs himself and they are shifted onto society at large then there is an
externality with which the efficiency-promoting state will be concerned. 19
From the perspective of economic analysis, market failure may justify state
regulation provided that such regulation will not simply magnify the problem. 20
2. The Application of Cost-Benefit Analysis to Occupational Health and
Safety Standard Setting
State intervention in the face of market failure can take on a variety
of forms. Shifting liability rules (i.e. no-fault workers’ compensation), im-
posing injury taxes (i.e. penalty assessments) and providing free information
to the affected parties are some of the strategies that have been adopted or
proposed. In addition to these steps, direct regulation of maximum expo-
sures to unhealthy conditions, and particularly to toxic substances, has emerged
as a common instrument of state intervention.’ 2′ However, once the state
undertakes to establish standards, the problem of selecting the appropriate
exposure level arises. Cost-benefit analysis suggests that the appropriate
standard is the one that would have been produced by a well-functioning
market and that where the market is imperfect the state should employ its
centralized decision-making and planning apparatus as a substitute for the
market mechanism. In other words, the state must administratively deter-
mine the economic benefits and costs of different standards and select the
standard whose application maximizes social wealth. In the following pages
I will outline the techniques employed in performing cost-benefit analysis
and criticize their technical adequacy.
a. Estimating the Benefits
The first problem that arises in estimating the benefits that might flow
from a regulation reducing current levels of exposure is that we frequently
do not know what the effects of current exposure levels are. This is partic-
ularly true with respect to carcinogens. Indeed it is not easily established
whether a substance is carcinogenic at all, and, if it is, whether there is any
safe level of exposure. The current state of scientific knowledge simply does
“9An harmful externality arises where one firm can shift the costs of its activities onto some
third party without legal sanction. As a result the price does not accurately reflect the true costs
of the activity, resulting in an over-investment or a non-optimal allocation of resources. To
the extent that the state is concerned with promoting efficiency, some form of intervention is
justified. It is even more likely to occur when the state itself is forced to bear the costs, unless
it chooses to subsidize the activity in question. For a general discussion of the concept of
externalities and possible responses see Hirshleifer, supra, note 36, 532-9.
1’20For an elaboration of a theory of the failure of government regulation, see Wolf, A Theory
of Nonmnarket Failure: Framework for Implementation Analysis (1979) 22 J.L. & Econ. 107.
1′ For a survey of Canadian occupational health and safety legislation, see Brown, Canadian
Occupational Health and Safety Legislation (1982) 20 Osgoode Hall L. J. 90.
–
REVUE DE DROIT DE McGILL
[Vol. 29
not provide us with the information needed to develop accurate damage
functions for different exposure levels.122 However, this is a problem that
is faced in any regulatory approach and is not unique to cost-benefit analysis.
What is unique about cost-benefit analysis is that once we know the
health effects of different levels of exposure, we are required to quantify the
value of lives saved and health improved. Economic analysts typically meas-
ure value in terms of “aggregate consumer willingness to pay”. 123 Where
there is a market for a commodity its value is revealed directly by consumer
behaviour. However, in the absence of markets for health and lives, indirect
measures of willingness to pay must be found. One could, for example,
conduct a survey. But it is easy to see that there would be a strong incentive
for workers at risk to misrepresent their preferences so as to inflate the value
that policy makers would assign to safer conditions in the workplace. An
alternate approach favoured by economists is to find a surrogate market in
which preferences will be revealed by actual behaviour. In this area econ-
omists have looked to the presence of risk premiums. By seeing how much
workers actually demand to be exposed to different levels of risk, one can
infer the value that they place on their lives and health. We have already
noted the inconclusive character of such studies. 124 The inadequacy of val-
uations based on market transactions can be traced to the same problems
that created market imperfections to begin with. Risk premiums based on
incomplete information are not an accurate indication of workers’ prefer-
ences. Limited job mobility and other structural problems of the labour
market may also distort the analysis of surrogate markets. Unskilled workers
may not enjoy a wide choice of jobs and may take hazardous work at
relatively low pay. It is not clear that the value assigned to life and health
should depend on the socio-economic class of the individual at risk. 125
1
22For a review of the current state of medical knowledge about carcinogenesis and the
identification of carcinogens, see “Report of the Special Advisory Committee on Carcinogens”
in ACOHOS, Fourth Annual Report, supra, note 96, 143-78. With respect to low dosages, see
Ashford, supra, note 77, 115-24 and Maugh, Chemical Carcinogens: How dangerous are Low
Doses (1978) 202 Science 37. For a good discussion of the problems involved in epidemiological
research and its implications for standard setting, see Cranor, Epidemiology and Procedural
Protections for Workplace Health in the Aftermath of the ‘Benzene’ Case (1983) 5 Ind. Rel. L.J.
372.
123See R. Posner, Economic Analysis of Law, 2d ed. (1977) 10.
124See supra, notes 115-8 and accompanying text.
125Can we assume that poor people value their lives less than wealthy ones and that therefore
wealthier people should be better protected? While such results might be good news to em-
ployers in low-paying industries who want to avoid expensive regulations, for others it might
constitute an indictment of the distribution of wealth in our society and the appropriateness
of ignoring that distribution in formulating policy. For a critique of these econometric studies,
see M. Green & N. Waitzman, Business War on the Law: An Analysis of the Benefits of Federal
Health/Safety Enforcement, 2d ed. (Corporate Accountability Research Group, 1981) 47-60.
1984]
OCCUPATIONAL STANDARDS IN ONTARIO
Further, preferences are not formed in a void. Ashford pointed to the prob-
lem of socialization patterns that emphasized “machismo” and neurotic
risk taking. 126 While this pattern might translate into a low willingness to
pay for improved health, it is also not clear that such behaviour should be
relied upon as a basis for assigning value to a worker’s well-being.
So far the value of improved health and safety has been measured only
in terms of the willingness to pay of workers directly at risk. We also have
to consider benefits that may accrue to persons outside the employment
relationship. Mendeloff noted that the concerns of a worker’s family and
friends may not be reflected in his or her own willingness to pay. 127 Yet, if
we wish to assess aggregate consumer willingness to pay, these concerns
must be taken into account. Although their preferences are rarely expressed
in the marketplace, there are members of the public who would be prepared
to pay for the satisfaction of knowing that workers were provided with a
healthier environment, and the conclusion that no such demand exists is
not justified; rather, it may be that organizational costs and the structure of
the market do not readily enable small, widely dispersed demands to be
recognized, and there is no shadow market which would allow the state to
measure willingness to pay from actual behaviour. If these more remote
preferences are to be considered, then the calculation of benefits loses any
claim to being a scientific endeavour. 128 If they are excluded, then market
failure is simply being reproduced and the exercise is pointless. This suggests
that there is no scientific basis upon which damage functions for particular
exposures can be accurately calculated. There is no technical solution to the
problem of valuing lives and health. A quantification of the benefits of a
regulatory standard will reflect policy choices, not technical criteria.
b. Estimating the Costs
Bacow notes that regulation imposes costs on employers in three ways:
(1) fines for non-compliance; (2) costs of purchasing and maintaining the
technology required for compliance and (3) in some cases, impaired worker
productivity. 129 A number of surveys on the cost of compliance with OSHA
regulations have been conducted. A 1973 McGraw-Hill survey found a 26%
’26Supra, note 77, 357-9.
127Supra, note 37, 67.
128Once you accept a broad definition of externalities that goes beyond physical harm, it
becomes possible to justify almost any rule on efficiency grounds by giving high valuations to
the public’s willingness to pay in order to avoid moral outrage, diminution of aesthetic values,
etc. For an excellent critique of the liberal use of cost-benefit analysis, see Kennedy, Cost-
Benefit Analysis of Entitlement Problems: A Critique (1981) 33 Stan. L. Rev. 387, and partic-
ularly 398-400.
129L. Bacow, Bargaining for Job Safety and Health (1980) 29.
McGILL LAW JOURNAL
[Vol. 29
increase in planned expenditures on health and safety between 1972 and
1973. The National Association of Manufacturers (NAM) surveyed its mem-
bers in 1974 and their estimates of the costs of OSHA compliance ranged
from an average of $33,000 for firms with under 100 employees to $7,146,000
for firms of more than 5,000 employees, with a per worker cost which
decreases in larger firms. Dunn’s Review, relying on business sources, esti-
mated OSHA would raise costs in many industries by five to ten percent. 130
Of course, such studies must be viewed warily. The McGraw-Hill study
does not tell us what proportion of an increase is attributable to OSHA
regulation. Estimates about expected costs are likely to be inflated for stra-
tegic purposes. Mendeloff reports that when the vinyl chloride standard was
proposed, industry predicted dire consequences if the regulation was pro-
mulgated. A 1976 survey, however, indicated that only two plants had closed
and that there had been a six percent rise in PVC prices. 13 1 Further, costs
of compliance with any regulatory scheme are likely to be highest in the
period immediately following the introduction of the legislation because of
a “catch-up” effect, both with respect to the number of regulations and the
historic under-investment in health technology. Thus, the learning curve
should reduce compliance costs over time.’ 32 Green and Waitzman have
reviewed the question of cost extensively and argue that conventional stud-
ies fail to properly identify compliance costs as those which are above ex-
penditures that otherwise would have been incurred. As well, they claim
that many costs usually associated with regulation, such as declines in pro-
ductivity and employment, do not stand up under close scrutiny. More
stringent health standards, for example, may act as a stimulus to techno-
logical innovation which in fact increases productivity. Improved worker
health and a more pleasant work environment could lead to less sick leave
and absenteeism, which again will improve productivity. 133
Although the problems of estimating costs are considerably fewer than
those involved in calculating benefits, it is still not a simple matter of plug-
ging figures into a formula. Neutral information about the potential costs
of compliance with any standard is notoriously hard to come by, if only
because the most knowledgeable people are generally those who will be
subject to regulation. Even if we agreed on what costs should be included
and on the impact of dynamic effects, reliable information may still be
difficult to obtain. 34
130Cited in Ashford, supra, note 77, 317-9.
131Mendeloff, supra, note 37, 55-6.
132This effect was noted by Ashford. See supra, note 77, 338-9.
133Green & Waitzman, supra, note 125, 27-40.
1341f employers were faced with the real threat that fines for non-compliance with a regulation
would be raised to a level that would make compliance the less expensive alternative, the
incentive to inflate the estimated costs of compliance might be diminished.
1984]
OCCUPATIONAL STANDARDS IN ONTARIO
c. Conclusion
The performance of cost-benefit analysis has its own costs, and before
we decide to subject standards to that analysis, we should decide whether
the method can justify its own application.135 Of course, there are a number
of advantages that make cost-benefit analysis attractive to politicians, but
which are not likely to be openly offered as reasons for its adoption. The
most important of these is that cost-benefit analysis tends to mask what are
essentially political choices as scientific or purely technical calculations. The
selection of assumptions regarding the long-term effects of exposures to low
levels of toxic substances in the absence of scientific proof is a political
choice. The decision to consider third party externalities and the determi-
nation of the weight they should be given are also political choices. Yet,
where cost-benefit analysis is applied, these political decisions tend to appear
to be merely technical and beyond the comprehension of the non-specialist.
If the claims of objectivity can be unmasked, then perhaps the popularity
of this kind of analysis will decline. 36 In the following sections I will con-
sider the normative questions and the issues in political economy that are
at stake in the decision to adopt a policy of setting standards through cost-
benefit analysis.
3. The Normative Assumptions of Cost-Benefit Analysis: A Brief Critique
Early in this discussion of cost-benefit analysis I cited Calabresi’s well-
known formulation which began, “Apart from the requirements ofjustice, I
take it as axiomatic.. .”(emphasis added). 37 The statement requires some
clarification. What compelled Calabresi to begin with that qualification? Can
any claims be made in favour of wealth maximization as a normative ideal?
Is wealth maximization a goal that can be justified independently of the
claims of justice? If so, should justice and wealth maximization be traded
or “mixed” in pursuit of some conception of the social good? These are
fundamental issues which have been the subject of recent debates in the
135Not surprisingly, economists who stand to benefit both in terms of employment and status
opportunities within government often seem to be the strongest advocates of this kind of
analysis.
136For an extensive challenge to the objectivity of cost-benefit analysis on its own terms, see
Kennedy, supra, note 128.
137Supra, note 105.
REVUE DE DROIT DE McGILL
[Vol. 29
law journal literature.138 It is clearly arguable that the normative claims for
wealth maximization as an appropriate government policy are highly prob-
lematic. From a Kantian perspective wealth maximization is objectionable
because it fails to respect individual autonomy. It would allow some persons
to be made worse off if a net increase in social wealth results. From a
utilitarian perspective wealth maximization fails as a principle because there
is no necesary correlation between increases in total wealth and increases
in total happiness or utility. Indeed, once wealth is separated from happiness
it becomes extremely unclear what value wealth represents, unless it is an
end in itself, and one becomes what Dworkin has aptly identified as “a
fetishist of little green paper”. 139
But let us assume that we are fetishists, or that wealth maximization
is tied to some undefined bundle of values we want to increase. There may
be normative constraints on our fetishism, and some actions taken to max-
imize wealth can have the impact of decreasing the total bundle of values
we wish to maximize. With this in mind Calabresi chose to qualify his
formulation of the goals of accident law policy and to state recently that
“an appropriate blend of efficiency and distribution is highly instrumental
towards, and closely correlated with, achieving what many would view as
a just society”. 140 Thus, according to Calabresi, distributive criteria are also
a component of justice that might impose constraints on policies designed
to maximize net social wealth.
Against this background, let us now turn to the most ethically troubling
aspect of cost-benefit analysis of occupational health standards: reliance on
the market to determine the value of lives saved and health improved.
Calabresi and Bobbitt pointed to two “ethical” difficulties involved in leav-
ing allocative decisions about tragically scarce resources to the market. First,
they noted a problem with respect to the “costs of costing”:
The present problem reflects, instead, the external costs – moralisms and the
affront to values, for example – of market determinations that say or imply
that the value of a life or some precious activity integral to life is reducible to
a money figure.’ 4′
‘ 38The principal articles are: Posner, Utilitarianism, Economics, and Legal Theory (1979) 8
J. Legal Stud. 103; Dworkin, Is Wealth a Value? (1980) 9 J. Legal Stud. 191; Kronman, Wealth
Maximization as a Normative Principle (1980) 9 J. Legal Stud. 227; Posner, The Value of
Wealth: A Comment on Dworkin and Kronman 9 J. Legal Stud. 243; Posner, The Ethical and
Political Basis ofthe Efficiency Norm in Common Law Adjudication (1980) 8 Hofstra L. Rev.
487; Calabresi, About Law and Economics: A Letter to Ronald Dworkin (1980) 8 Hofstra L.
Rev. 553; Dworkin, Why Efficiency? A Response to Professors Calabresi and Posner (1980) 8
Hofstra L. Rev. 563.
139Dworkin, Is Wealth a Value?, supra, note 138, 201.
’14Calabresi, supra, note 138, 558.
141G. Calabresi & P. Bobbitt, Tragic Choices (1978) 32.
1984]
OCCUPATIONAL STANDARDS IN ONTARIO
Second, they point to the dependence of the market on the existing
distribution of wealth, and argue that decisions made under such conditions
generate social costs such as outrage. Thus, we do not allow indentured
labour notwithstanding that, given certain conditions, individuals might
choose it:
The willingness of a poor man, confronting a tragic situation, to choose money
rather than the tragically scarce resource always represents an unquiet indict-
ment of society’s distribution of wealth. 42
Economists might label this concern mere paternalism, but that re-
sponse is not an adequate answer when the question is whether concern
about the well-being of others should be counted in the formulation of public
policy. It is not inappropriate for individuals to care about others and for
policy makers to take those concerns into account.
Kelman has recently published a critique of cost-benefit analysis that
develops the first line of ethical objections noted by Calabresi and Bobbitt. 143
He argues that utilitarianism is an insufficient moral position in that it fails
to accommodate the view that certain values and rights are prima facie
morally valid, independent of their costs and benefits. Thus, the “right to
a safe work place” may be analogous to the right of free speech, and the
moral weight attached to such a right must be considered independently of
costs and benefits, without ignoring these important factors.
We have already explored some of the technical problems that beset
economists’ attempts to measure willingness to pay for non-market things
such as lives and health in order to monetize all values. Kelman highlights
some of the ethical difficulties in this valuing process. First, there may be
a distinction between how people value things in private transactions and
how they would wish those things to be valued publicly. Private transactions
may not capture the true range of individual preferences and ethical judg-
ments. Other-regarding behaviour embodying “higher” values is more likely
to be expressed in the political process than in private market transactions.
Cost-benefit analysis, by relying on market valuations, implicitly denies the
legitimacy of these expressions of “higher” values.
Further, the very act of monetizing non-market things may, by itself,
reduce the value of those things. When we declare life and health to be
priceless, we are not necessarily saying that we would be willing to trade an
infinite quantity of other goods to preserve a single life another day. Rather,
it is a statement about non-economic values upon which we do not want
142lbid., 39.
143Kelman, Cost-Benefit Analysis-An Ethical Critique (1981) 5 Regulation 33-40 (January/
February).
McGILL LAW JOURNAL
[Vol. 29
a market price to be placed, precisely because it would undermine those
values to do so. While we will continue to make decisions that have the
effect of valuing lives, the refusal to do so in advance by pricing human life
as an input into cost-benefit analysis preserves the value we place on life
and forces a confrontation between conflicting social values.
The second ethical problem noted by Calabresi and Bobbit relates to
the question of wealth distribution. If the existing distribution of wealth
cannot be judged to be fair, then Pareto-superiority may lose much of its
force as a moral justification.144 True, one party may be better off while the
other is no worse off, but the position of the weaker party may be such that
the conditions for voluntary exchange between autonomous individuals do
not exist. Thus, we might say that individuals do not normally sell their
lives and health unless they have no other resources. A person in that po-
sition is motivated more by compulsion than by free choice, 145 and, after
the transaction is completed, there is little left of the autonomous individual.
A moral system that claims to respect human autonomy cannot tolerate
social conditions which induce individuals to enter into agreements that
substantially reduce their autonomy.
Cost-benefit analysis, in relying on ability and willingness to pay, takes
the existing distribution of wealth as a given. It therefore favours those who
are already better off, since by definition their ability to pay is greater than
those worse off than themselves. Wealth effects may also result in an in-
creased willingness to pay for certain goods. 46 Thus, a corporate executive
of an asbestos mining corporation may be willng and able to spend consid-
erably more to have asbestos insulation removed from his office so as to
reduce his risk of contracting cancer or asbestosis from 0.05% to 0.01% than
a miner would to have ventilation improved so as to reduce his risk from
0.5% to 0.1%. The willingness-to-pay criteria suggests that it might be ap-
propriate to set higher standards of safety in the workplace for corporate
executives than for miners. I think many might share the judgment ofKronman:
Even if there is no justification for making those who are already wealthy share
what they have, there is something offensive in the suggestion that their wealth
is a reason for giving them even more.147
144A Pareto-superior transaction is one in which at least one party believes himself to be
better off and no one believes himself to be worse off. For a simple and clear explanation, see
B. Ackerman, Economic Foundations of Property Law (1975) xi-xiv.
’45As noted earlier, even the common law courts of the nineteenth century came to recognize
that fact. See Dean v. Ontario Cotton Mills Co., supra, note 41.
423-9.
’46For a discussion of wealth effects in cost-benefit analysis, see Kennedy, supra, note 128,
47Kronman, supra, note 138, 240.
1984]
OCCUPATIONAL STANDARDS IN ONTARIO
4. The Political Economy of Cost-Benefit Analysis
The doubtful status of wealth maximization as a normative ideal might
be relied on in framing a broader challenge to the ethical foundations of
capitalism. Indeed, the logic of capitalist accumulation seems to lead to the
fetishism of little green papers and to support of the principle that the holder
of accumulated wealth is entitled to more than one without accumulated
wealth. 48 However, I do not intend to pursue this line of critical analysis.
Rather, in this section I wish to focus on the constraints that the state is
under when selecting particular policies and on why the logic of accumu-
lation, as expressed in cost-benefit analysis, has been unable to gain a mo-
nopoly as the criteria for selecting state policy. Further, I will argue that the
adoption of cost-benefit analysis would portend a fundamental re-orienta-
tion of the role of the state.
I might begin by briefly summarizing the argument I have been making.
The role of the state began to be transformed in the nineteenth century,
from one of constituting the conditions for, and facilitating the development
of, capitalist production and market exchange to one of substituting itself
for the market, both with respect to the co-ordination of economic activity
and the maintenance of social integration. This process of transformation
has continued in the twentieth century, reaching unprecedented levels. How-
ever, these roles are potentially in conflict and the conflict becomes increas-
ingly manifest in a stagnating economy.
The Canadian economy was never laissez-faire: from the outset the
state played an active role, both in planning and providing a technical
infrastructure when private capital was unavailable. 149 Regardless of our
starting point, the role of the state in organizing the economy has expanded
substantially. The planning of fiscal and energy policy, the provision of
investment incentives, the supervision of marketing boards, as well as direct
involvement in government purchasing are only the beginning of a list of
economic functions assumed by the state. Yet government planning still
takes place largely within the limits set by private ownership and freedom
of investment. The government of the day finds itself in the unenviable
position of being held politically responsible for the management of an
essentially private economy. It is therefore limited in its selection of eco-
nomic policies by the continual threat that business confidence will be un-
dermined, leading to a decline in investment and a downturn in the economy.
1
4 8 0f course, some defenders of capitalism are not only prepared to concede that this state-
ment is accurate, but in fact would argue that fetishism and inequality should be viewed as
virtues. See Gilder, Wealth and Poverty (1981).
149Panitch, “The role and nature of the Canadian state” in L. Panitch, ed., The Canadian
State: Political Economy and Political Power (1977) 3, 14.
REVUE DE DROIT DE McGILL
[Vol. 29
Because the government has assumed responsibility for steering the econ-
omy, such results leave it politically vulnerable to charges of economic
mismanagement and weaken its electoral support amongst those who have
suffered as a result, and this includes workers.
In addition, the market has proven itself unable to fulfill a social in-
tegration or self-legitimation function. While capitalism undermines tra-
ditional bases of the social order, the ideology and practice of exchange of
equivalents or wealth maximization does not seem to be an adequate re-
placement upon which a social order can be built.150 Some market outcomes
seem to produce harsh consequences which engender politically effective
responses. The state increasingly takes on the responsibility of compensating
for the failure of the market to produce outcomes that are in accord with
widely-held beliefs about social justice that are not rooted in market con-
ceptions ofjustice. Thus, another constraint is placed upon the state policy
selection process. Policies must be judged to be legitimate on the basis of
some non-market criteria of fairness or equality. Regulation of occupational
health and safety conditions was one of the earlier state interventions mo-
tivated largely by legitimation concerns. It has been observed that while the
Canadian state played an exceptionally active role in steering the economy,
it has been a laggard in the provision of welfare benefits.’ 51 Nevertheless,
the welfare role of the state has expanded and the demands being made on
the state seem to be increasing. Aside from occupational health and safety,
there are demands for state regulation of environmental quality, consumer
transactions, housing quality and price as well as straightforward economic
demands for higher and more comprehensive benefits under a variety of
income security programs.
Failure to respond to these demands, at least to some extent, leaves
the government vulnerable at elections and also creates the possibility that
unsatisfied demands will destabilize basic social and economic arrange-
ments. Yet, if the state is to satisfy these demands it runs the risk of impeding
accumulation. The imposition of regulations may impose costs on industry
that cannot be passed on to the consumer because of competition. Faced
with a potential decline in profits, investment may decline. Further, the
party in power (assuming it is not a socialist or labour party) is likely to
lose the support of many of its financial backers.
Given these constraints, cost-benefit analysis, if it is to be successful,
must produce regulations which satisfy the requirements of accumulation
1’5 For a recent analysis of the effects of capitalism on culture and the social order, see C.
Lasch, The Culture of Narcissism (1978).
‘5’Panitch, “The role and nature of the Canadian state”, supra, note 149, 22.
1984]
OCCUPATIONAL STANDARDS IN ONTARIO
and legitimation. From the perspective of accumulation, cost-benefit anal-
ysis may provide the most attractive basis on which to have standards set,
short of de-regulation.152 As we have already noted, the goal of cost-benefit
analysis is to replicate the allocation of resources that would have resulted
in a well-functioning market given the existing distribution of wealth. In a
society in which there is great inequality in the distribution of wealth, the
willingness and ability to pay of the less well-off for health and safety is
likely to be fairly low, which will lead to a low quantification of the value
of improved worker health in cost-benefit analysis. A lower benefit function
would result in the setting of higher permissible exposures which would be
less costly for employers to comply with, and lower production costs increase
profitability.
If cost-benefit analysis will provide policies that are consistent with
accumulation, can it also provide policies that satisfy demands for legiti-
mation? Legitimation becomes increasingly important once the decision
framework is a public one in which policies must be consciously fashioned.
Legitimation deficits produced in the realm of exchange not only stimulate
the transformation of the state but, once that transformation occurs, con-
tinue to restrict the selection of specific state interventions. It is necessary
to consider therefore, the potential bases for the legitimation of cost-benefit
analysis as the cornerstone of standard-setting.
One source of legitimacy for cost-benefit analysis might be found in
the market itself. If we take the position that government regulation should
only occur in response to market failure based on high information and
transaction costs, then a technique which tells us how to mimic the outcome
of a well-functioning market would surely be desirable. However, the legit-
imacy of the technique can be no greater than the legitimacy of the market
itself. If, as I have suggested, it is not classic market failure but the underlying
distribution of wealth which has generated pressure for government inter-
vention, the market can offer almost no legitimation for government policy
because it itself produces the deficit which requires correction.
A second potential source of legitimacy for cost-benefit analysis would
rest on the claim that it is a neutral tool of rational administration. On the
surface, such a claim seems plausible. It is difficult to argue with the prop-
osition that the benefits of a particular policy should outweigh its costs.
However, as I have tried to show, the selection of assumptions that must
be made in order to conduct cost-benefit analysis of health standards in-
volves inherently political (as opposed to merely technical) decisions. More-
over, a decision to set standards on the basis of cost-benefit analysis involves
’52De-regulation would be attractive to industrialists because the burden of market failure
is carried by the workers themselves or by the state, but rarely by the employer.
McGILL LAW JOURNAL
[Vol. 29
a decision to ignore wealth distribution in that process. However else one
might describe that decision, it does not appear as a theorem of rational
administration.
A variation on this theme would be a version of Posner’s consent ar-
gument. 53 If cost-benefit analysis aims at wealth maximization, then every-
one should consent to such a rule provided the winners pay off the losers.
But this is just another attempt to rationalize the market outcome where
all workers are implicitly compensated ex ante for exposing themselves to
risk, where there has been a “voluntary assumption of risk”. If the problem
with the market stems from the distribution of wealth, then notwithstanding
the fact that employees receive ex ante compensation, they will not consent
to the market solution either under a contractarian framework or under a
state regulatory framework, because neither addresses the source of their
dissatisfaction. This raises again the problem of the doubtful normative
status of wealth maximization principles.
In practice, the major source of legitimacy for cost-benefit analysis rests
with its ability to disguise politics and ideology as economic science per-
formed by sophisticated technicians. However, this source of legitimacy has
proved rather weak. We have remarked the large body of academic literature
debunking the claims of cost-benefit analysis.’ 54 Organized labour in the
United States, with the support of the Occupational Safety and Health
Administration (at least prior to the Reagan administration) has persistently
opposed the imposition of a cost-benefit requirement for toxic substance
exposure standards, and the Supreme Court has interpreted the Occupa-
tional Health and Safety Act of 1970 as imposing a feasibility standard. 155
Presumably the feasibility standard will in most cases be more protective
than the cost-benefit standard. 56 If Ontario were to adopt a policy that
produced consistently less protective standards than those in the United
States, one doubts whether the supposedly scientific character of the process
supporting the policy would prove to be a strong source of legitimation to
those seeking the benefit of regulation. If the outcome of a decision-making
cation, supra, note 138.
’53Posner, The Ethical and Political Basis of the Efficiency Norm in Common Law Adjudi-
154Aside from the critical articles I have already cited see Baker, TheIdeology ofthe Economic
Analysis of Law (1975) 5 Phil. and Pub. Aff. 3 and P. Self, Econocrats and the Policy Process:
The Politics and Philosophy of Cost-Benefit Analysis (1975).
155American Textile Manufactures Institute, Inc. v. Donovan, supra, note 3.
561n theory it is entirely possible that cost-benefit analysis would lead to a standard that
1
was not economically feasible, in the sense that it would require an entire industry to shut
down. See Morrall “Cotton Dust: An Economist’s View” in R. Crandall & L. Lave, eds, The
Scientific Basis of Health and Safety Regulation (Brookings Institute, 1981) 93, 98. Of course
the economist finds that in this case the feasibility standard was more stringent than the cost-
benefit standard.
1984]
OCCUPATIONAL STANDARDS IN ONTARIO
process can be evaluated against criteria that are external to the process
itself, the legitimacy that adherence to the process can confer on the outcome
may be limited.157
The same considerations that led to the rejection of the contractarian
framework for setting the level of occupational health and safety in the
workplace should lead to the rejection of cost-benefit analysis as a guide for
government standard-setting. Both models fail to satisfy the criteria of le-
gitimacy, the former by producing undesirable consequences in the form of
unacceptably low standards, the latter by mimicking them. Perhaps the
better alternative to standard-setting would be a direct and massive re-
distribution of wealth. This would address the underlying problem directly.
In the meantime this is not on the agenda (at least if we are talking about
a more egalitarian re-distribution); regulation of the workplace is. Loss of
efficiency has its costs, but so too does loss of legitimacy. This is the cost-
benefit analysis that has guided the selection of state policy.
Conclusion
The theme of this paper has been that the development of occupational
health and safety standard-setting can be best understood by analyzing the
contradictory demands on the state in capitalist society to facilitate capitalist
accumulation while at the same time legitimating the social relations of
production. The state must ensure that citizens are provided with minimally
acceptable levels of economic well-being, which includes minimally safe
conditions of work. Policies that fail to satisfy this requirement of legiti-
mation, such as market-based contract rules or public regulations that mimic
the outcome of a health market discovered through cost-benefit analysis,
are unlikely to prove successful. It is for this reason that the Government
of Ontario has avoided adopting a substantive policy with respect to stan-
dard-setting and instead has chosen to rely on an ad hoc bargaining pro-
cedure. The enforcement of standards also takes place to a large extent in
an ad hoc manner, which permits employers to strike discrete bargains with
employees through the internal responsibility system. External policing is
only brought into play in the event of a breakdown in the bargaining process.
1571 do not deny that process values can serve as a source of legitimation. As Mashaw notes
in Administrative Due Process: The Quest for a Dignitary Theory (1981) 61 B.U.L. Rev. 885
at 888, “[w]e do distinguish between losing and being treated unfairly.” My point is that, if
under a particular process one group always seems to lose, they are quite likely to question
the fairness of that process. If the substantive justification for the process is that it is scientific
and not that it affords participation, the foundations for that claim would have to be beyond
dispute. Further, if the result was indeed scientific, it would have to be shown that the scientific
perspective was the only valid one. I do not think qost-benefit analysis of occupational health
standards comes close to satisfying these conditions.
REVUE DE DROIT DE McGILL
[Vol. 29
The balance between legitimation and accumulation is not determined
in the abstract, nor is it a direct reflection of the structural requirements of
capitalism. Rather the balance is the product of a specific historical context,
in which the organizational capacities and powers of different classes are
brought to bear. Thus the appearance of Factory Act legislation and the
reform of employers’ liability law in the 1880’s can be explained to a large
extent by the emergence of politically active workers’ organizations at that
time, and by competition for workers’ votes. Continuing this developmenft,
the NDP has been instrumental in the most recent wave of legislative reform,
serving as an instigator of organized labour’s concern with occupational
health and safety issues, and as a vehicle for the expression of this concern.
The ability of the state to be successful in juggling competing demands
for accumulation and legitimation may have depended in part on rapid
economic growth. Large increases in total economic wealth allow for overall
improvement in the conditions of employment, without significantly erod-
ing conditions favourable to capital accumulation. However, where the econ-
omy is stagnant or shrinking, the pressure on the state to develop and
implement policies that will improve the environment for private invest-
ment tends to increase. As well, the labour movement is likely to find its
strength decreasing as it is forced to direct resources toward the protection
of past economic and political gains, while at the same time its membership
is declining due to unemployment in organized sectors of the economy.
Under such conditions the balance of forces is likely to shift so that state
policy increasingly favours demands to facilitate accumulation.
Thus, even though the standard-setting process that is in place is es-
sentially a political one, it operates under the influence of a legally sanc-
tioned, but largely uncontrolled, exercise of concentrated economic power.
The threat that this power will be exercised by shifting investments to other
jurisdictions or countries in response to the imposition of standards that
are costly to comply with is real. The organizational capacity of the working
class is relatively weak: less than forty percent of workers are organized by
trade unions, and unorganized labourers lack effective representation in the
political process. It is unlikely that substantial progress will be made through
existing regulatory processes, unless effective strategies are developed for
enhancing the political power of workers and their allies.
A more radical approach to the problem of the distribution and allo-
cation of risk in production would require that those who are involved in
the production process decide democratically what levels of risk they are
willing to incur. This goes far beyond the limited freedom of employees to
choose between alternative jobs that the labour market currently offers, and
would involve worker ownership and control of the productive enterprise
1984]
OCCUPATIONAL STANDARDS IN ONTARIO
itself. Under such a regime the question of allocation of resources does not
disappear. Trade-offs between productivity and safety will still be necessary.
The difference is that decisions will be made by the individuals who stand
both to benefit directly by increased productivity and to suffer exposure to
the risks of production.15 8
Aside from the sporadic establishment of workers’ cooperatives, there
are few signs on the horizon that a decentralized socialist economy will
emerge in Canada in the near future. Reforms within the existing framework
of capitalist relations of production that will give workers more control over
production (such as a requirement that workers approve, or at least are
consulted about, the introduction of new technologies and substances in the
workplace) would be a step toward the kind of structural reform necessary
to implement more equitable approaches to the determination of occupa-
tional health and safety standards. 159 However, such reforms will only take
place where the political capacity to impose them is present, and will ne-
cessitate a transformation of Canadian society that is at present not easy to
anticipate.
L. Rev. 695.
158This approach has been developed by Abel in A Socialist Approach to Risk (1982) 41 Md
159For a stimulating and seminal contribution to the problem of developing strategies for
social change in Western capitalist societies, see A. Gorz, Strategy for Labour [:] A Radical
Proposal (1967). See also Wright, supra, note 5.
