- 2.1 Appropriate level of protection
- 2.2 Risk assessment methodology
- 2.3 Estimation of consequences
- 2.4 Risk management
2. Risk analysis methodology
2.1 Appropriate level of protection
In the draft IRA document Australia's appropriate level of protection (ALOP) has been illustrated. This ALOP is expressed as an "iso-risk" curve of probability vs. the consequences of entry, establishment and spread of the pest of concern. New Zealand welcomes this statement as it improves the overall transparency of the risk analysis.
2.2 Risk assessment methodology
AFFA has elected to assess the risk of entry, establishment and spread of pests associated with apple fruit in qualitative terms (as is AFFA's right). This has been done by breaking down the importation pathway into a series of steps and assessing the risk at each of those steps. The likelihoods are then combined into a summary of risk using a matrix.
This methodology has not, however, been consistently applied in the risk assessment. The biological pathway has been broken down into importation (4 steps), distribution (4 steps) and spread (one step). The risks are considered at each step, however an assessment of risk (as per Table 6 of the draft IRA) has not been made at every step. For example, in the assessment of fire blight the conclusion on importation step one was that "some fruit" may be infested/infected.
New Zealand believes that the qualitative assessment of risk should be applied in a consistent manner to all stages of the risk assessment, as phytosanitary measures are based on these assessed risks. There are also instances where a risk has been considered to be "possible" or "conceivable". New Zealand submits that at all steps of the assessment AFFA should provide some evaluation or estimation of likelihood or probability (qualitative or quantitative), as opposed to the mere possibility of a risk.
AFFA has stated that Australia's ALOP is highly conservative, and has also been conservative in the consideration of data. There are instances where a logical conclusion based on science has been reached but because of "uncertainty" (or perhaps consequences) further conservatism is introduced. New Zealand submits that consideration of Australia's ALOP should only be introduced at the risk management stage of the draft IRA, and not during the risk assessment stage. ALOP is not relevant when identifying pests of quarantine concern and assessing the risk associated with each. It is only relevant when determining what risk management measures are required in order to achieve the ALOP. The ALOP should not be allowed to influence the objective process of estimation of risk from scientific evidence.
It is stated (page 36) that marketing factors such as import volumes, end use patterns and product distribution patterns are taken into account in assessing risk. However these factors do not appear to have been addressed in the assessment of risk.
New Zealand is concerned that the rationale behind the method of combining risks is not transparent. AFFA has provided a table explaining the result of the combination of two probabilities, but not the logic behind the decision (e.g., it is not clear why low x low = low, yet moderate x low also equals low).
The matrix is also highly simplistic, in that the lowest likelihood prevails in all cases except moderate x moderate. This is inconsistent with the evaluation of "sequential conditional steps" (p84, l18), which implies a multiplicative effect on probability. When the likelihoods from four steps are combined using the matrix the multiplicative effect of combining probabilities is lost. All of the steps in the scenarios identified in the draft IRA are dependent sequential events. Each likelihood is a probability of less than one and will reduce the assessed at the previous step risk (AFFA associates the term probability with likelihood in Table 6). Logically, a series of four dependent steps each with a low risk will produce an overall risk that is considerably less than "low". The matrix contained in the draft IRA therefore overestimates risk unless it is used for combining only two probabilities. New Zealand requests that the rationale behind the method of combining risks is made more transparent.
2.3 Estimation of consequences
New Zealand asserts that AFFA has introduced into pest risk assessment concepts that go beyond the internationally accepted guidelines2 for assessing the economic impact of a pest. These concepts include "recognition", "concern", "values", and "wellbeing" (p.46 of the draft IRA).
For a pest to be considered to be a quarantine pest it must be of potential economic importance to the area endangered by the pest (FAO, 1999). The concept of, and requirement for, economic impact assessment is further elaborated upon in "Guidelines for Pest Risk Analysis" (FAO, 1996). This international standard discusses the various ways of assessing potential economic importance. In all cases, the standard refers to "economic importance". All impacts of a pest must be expressed in economic terms. AFFA has departed from this international standard by stating that the consequences assessed by AFFA "may be interpreted in dollar terms, in terms of particular societal values or social wellbeing, or as a combination of both". New Zealand submits that the international standard is quite precise in the requirement that these concepts be presented in terms of "economic importance". New Zealand asks that AFFA reviews the basis of their economic impact assessment to ensure that the impacts are expressed in economic terms.
New Zealand is also concerned that many of the terms used by AFFA in the estimation of economic impact are undefined. New Zealand asks that AFFA provides a clear explanation of what is meant by the following terms: "recognised within an affected geographic region"; "recognised at the national level"; "national concern". None of these terms are suggested as acceptable measures of economic impact in the relevant international standard.3
AFFA has indicated that economic impact can also be expressed as the impact on "societal values" or "social wellbeing", however these terms have not been defined. At no point has AFFA indicated that any survey has been undertaken to define the "societal values" of the potentially affected areas nor what the Australian populace considers to be a state of "wellbeing". Clearly such research is necessary before the impact of a quarantine pest4 on these values can be assessed. We suggest that AFFA has not conducted such research and that any assessment of impact on these criteria is completely subjective and in no way complies with the international guideline for assessment of economic impact. New Zealand submits that the risk assessment methodology relating to consequence assessment should better take into account risk assessment techniques developed by relevant international organisations.
2.4 Risk management
In determining whether phytosanitary measures are necessary in order to manage risk, AFFA has not considered other pathways for the introduction of quarantine pests. For example Australia currently imports commodities that are host to many of the quarantine pests identified in the draft IRA. The measures prescribed by AFFA for the management of these pests in the importation of apples are more restrictive than the measures required to manage the risk of the same pests being introduced through alternative pathways. Also, AFFA has not considered the smuggling pathway. In the previous IRA (AQIS, 1998b), the likelihood of introducing fire blight through smuggling was estimated to be higher than one introduction in 663 years.
Article 2.2 of the SPS agreement requires WTO members to ensure that any sanitary or phytosanitary measure is applied only to the extent necessary to protect human, animal or plant life or health, is based on scientific principles and is not maintained without sufficient scientific evidence. This provision finds fuller elaboration in Article 5. New Zealand believes that there are a number of measures proposed by AFFA which are not based on scientific principles. Furthermore, the rationale for the application of these measures is often unclear.
| New Zealand requests that AFFA explain the way in which each measure reduces risk (i.e., the scientific basis of the measure), and also the extent to which the measure is believed to reduce risk. |
3 i.e., FAO (1996)
4 as defined in FAO (1999)
Contact for Enquiries
MAF Information Services
Pastoral House
25 The Terrace
PO Box 2526
Wellington, NEW ZEALAND
Fax: +64 4 894 0721
Contact this person
