An employer owes duties to employees under Common Law and statute. The common law employers’ duties have been developed by the courts as they decide cases on accidents at work.
The employer’s Common Law duties are:
a) To provide a safe place of work
b) To provide proper tools and equipment
c) To provide a safe system of working
d) To provide competent staff

In addition an employer owes duties under statute to safeguard employees in the workplace.
Under the Safety, Health and Welfare at Work Act 2005, (”SHWWA”) (Section 8)) the employer’s duty is to ensure the safety of employees and in particular;

a) to provide by management and action, safety at work;
b) to provide by management and action that improper conduct does not occur;
c) to provide a safe place of work;
d) to provide safe plant or equipment;
e) to provide safe a safe system of work;
f) to provide safety information, instruction, training and supervision to employees;
g) to make a risk assessment and to implement measures to protect workers from those risks;
h) to provide protective clothing or equipment where hazards can not be completely eliminated;
i) to make emergency plans; and revise them as required;
j) a guard against hazards of particular articles or substances;
k) to provide welfare facilities and maintain them;
l) to provide a competent person to ensure safety and health at work of his employees.

Assessment of risk is the subject of academic study. It has the appearance of common sense but is far removed from that. For the purposes of SHWWA 2005, it involves four aspects.

• The risk must be identified in the first place.

• Secondly, it must be than assessed or quantified as a statistical possibility. In other words the likelihood of the perceived possibility must be measured.

• Thirdly, the measure of damage potentially arising must be identified.

• Fourthly, the practicability of avoiding the risk must be assessed.

This last matter is a question firstly, of physical possibility and secondly, cost. There is no certainty that each of these aspects can be perfectly ascertained. This is the norm In risk assessment
The standard of knowledge demanded of an employer in discovering the hazards in the workplace is that of the best available information. If the information becomes available, the employer is expected to promptly become aware of it. Such information is likely to first become available in countries other than Ireland. Employers must monitor the field of occupational safety and health on a global basis. Most occupational hazards in Ireland will, already, be well known as such.
Having recognized the hazard or risk, the employer must calculate the probability of its occurrence. This assessment is normally based upon inadequate evidence from a purely scientific point of view. This is not a real handicap or valid criticism. “Scientific” standards are not the norm in practical affairs and risk assessment is a practical matter. In addition, if scientific research is available to assist, it is very valuable, but it need not be conclusive. Usually, the assessment is based upon information arising from previous accidents, insight and skill in engineering or other technology, and/or information arising from experiments on animals with subsequent extrapolation to humans.
This mix is unlikely to be sufficient to produce a perfectly scientific statistical assessment of risk. So what? In practice decisions cannot be postponed. Only the available information can be used and, therefore, it must be used.
It may become more obvious in some circumstances than in others that risk assessment involves some ordering of values or priorities. Even continuing the status quo is a decision based on a value judgement. Cost effectiveness is a frequent determinant on such occasions. The value which society places on human health or life is another, even if no reference is made to it overtly.
If the full extent of the potential for damage amounts to something no worse, to speak in terms of farce, than the scuffing of the patent shoes of the sales manager, then the risk to be guarded against is negligible. This judgement remains valid even if the statistical probability shows that there is a risk of the damage occurring every six months. If, however, the full extent of the potential damage amounts to the loss of a limb of a worker or the death of a worker the risk to be guarded against is very large. This will apply even if the measured probability of the occurrence is as remote as once every 50 years.
What if the probability is very remote and the foreseeable damage is catastrophic? The final arbiter will be cost. This is not a reference to the capacity of the employer to pay for the avoidance of the hazard. That issue is irrelevant. It is a social judgement on the value of the money spent on avoiding the risk. If the amount of money required to be spent is large compared to the remoteness of the risk and the potential for damage, the risk need not be guarded against. Following the Alpha Piper oil-rig disaster in the North Sea, the U.K. Government set up a committee of inquiry into the event. Following the delivery of the report of Lord Cullen, the UK Department of Energy stated that mathematical standards of acceptable and non-acceptable risks would be introduced throughout the industry for the first time. The report itself went on to criticise Occidental Oil, the owners of the oil-rig, and the Department of Energy. It recommended that all offshore installations adopt a formal safety assessment system “involving the identification and assessment of hazards over the whole life cycle of a project……. The techniques used include hazard and operability studies; quantitative risk assessment; fault free analysis; human factors analysis; and safety audits”.
The sum considered proper to be spent to guard against a catastrophe is substantial, probably irrespective of the remoteness of the risk.
What does this mean in every day terms? It might require the closing down of a factory. This might happen because the risk cannot be guarded against, or the employer cannot, or will not, spend the money required to install the protection. There are mechanisms in the SHWWA 2005 for imposing this judgement on an employer. However, the decision may remain with the employer. There are hard reasons why the employer will not close down where the circumstances call for it.
There are in place standards for the assessment of value of loss or damage. Under the Civil Liability Act, the basis for compensating dependents and close relations of a deceased person have been in use for many years. The principle of compensation is approached from two directions. The Act sets out the sums of money to be paid for the pain and grief on the loss of the deceased. It varies with the relationship. Beyond that, the compensation is calculated on the basis of the financial loss suffered by the dependent on the death of the deceased.
There are serious drawbacks with this approach. It means that all lives are not of equal value. In addition, it hardly reflects the views of the deceased person. Put another way, the prospect of death from a known and avoidable risk is not readily faced by most people in circumstances where it is not guarded against for reasons of cost benefit.
Cost benefit has been shown to produce strange results. Dr. John Adams of the Department of Geography at University College London has suggested, on the basis of the Roskill Inquiry report (on the third London airport), that the net benefit to the economy, given the cost benefit value of a female human life, lay in having an aircraft crash on a woman.
Cost benefit analysis is of greater familiarity to managers than risk assessment. The opinion of where a cost benefit lies varies with the industry or business. It also varies with the society. It is a fact that the levels of compensation for personal injury in Britain are low. In the United States they may be too high. Given such variation, cost benefit analysis cannot be said to be straightforward and generally acceptable. For a manager making a decision to spend money, or not to spend money, such matters may not appear relevant. He will feel, having ascertained the statistical probability of an event, that a particular risk is acceptable. He will know nothing of the survey methods of the study that produced the probability figure. It may have included people who are not and never were at risk. They were not in the “danger zone”. Their presence in the survey makes a risk appear more remote than it really is, for the people who really matter, the people who are actually exposed.
There is another view to be taken: there is a social cost to a death or injury as much as an industry cost. The Environmental Risk Assessment Unit of the University of East Anglia identified a United Kingdom social cost of a life at about £266,000. This included £135,000 for lost output; £128,000 for pain and suffering; £1,300 police and medical costs; £1,750 damage to property.
This is higher than the amount the Nuclear Industry in Britain is prepared to spend on saving a life, if the person whose life is in issue is an employee. For an employee it will spend £500,000: for a member of the public it will not spend more than £200,000. The difference is based upon the perceived differences of exposure to risk.