​​​​​​Application for a new nickel refinery facility by a publically listed resources company

Introduction

This case study looks at the challenging aspects arising from the provision of limited assurance over forward-looking subject matter, the quantity of production reasonably likely to be produced by a new nickel refinery for the year ending 30 June 2019. The case study is based upon the application of the assurance standard ASAE3450 and is set out as a narrative that outlines the types of evidence that the applicant may use and include in its basis of preparation.

It is expected that the operators of the new facility should have a basis or preparation and to have prepared a detailed evidence pack to support their application. It is noted that it is a requirement of conducting assurance in accordance with ASAE3450 that the assurance practitioner shall, prior to agreeing the terms of the engagement, obtain agreement from the responsible party that they:

1. understand and accept the type of assurance to be expressed on each of the three different elements of the prospective financial information (paragraph A85-A86 of ASAE 3450), and
2. acknowledge and understand their responsibility for the preparation of the prospective financial information:
   1. in accordance with the stated basis of preparation, and
   2. ​based on assumptions selected by the responsible party that:
      • provide reasonable grounds for the preparation of the prospective financial information, if it is to be included in a public document
      • are clearly realistic for the preparation of the prospective financial information, if it is to be included in a non-public document, or
      • are not misleading or deceptive, having regard to applicable law and regulation.

Background

A large resources company listed on the Australian stock exchange is building a new nickel refinery in Australia and are in the process of applying for a calculated baseline under the new facility criteria. They have engaged a registered greenhouse and energy auditor to provide assurance over the application, including reasonable level assurance over the criteria and limited assurance over the quantity of the production that is reasonably likely to be produced and the expected emission intensity.

Technical details

The refinery will operate on the well-established Sherritt Gordon pressure leaching process and has been designed by a combination of the applicant’s own engineering team and a global consultancy that specialises in chemical process design. Conceptually the process consists of six individual unit operations, five of which will be of standard design with the same unit in use in other nickel refineries in operation somewhere in the world. However one of the units, the final nickel reduction stage, has a design aspect that improves its energy efficiency. This modification will impact the overall covered emission intensity of the production. The application has stated that facility operators expect the highest annual production will occur in financial year 2018-2019 and will be 80 000 nickel tonnes (Ni t) in the form of 99.99 per cent nickel briquettes (London Metal Exchange investment grade) and powder.

Company details

The company that will build and operate the new refinery are a relatively large, Australia-based business and are publically listed on the ASX. The company has a strong track record for hitting production targets, a good cash flow and is regarded by market analysts as well run. The company makes regular investor briefings and plans to release a market update on its expected performance of the new facility as soon as it has been granted a calculated baseline.

Testing the claim of reasonably likely production

The critical aspect that this case study deals with is how the registered greenhouse and energy auditor will be able to gain sufficient and appropriate evidence to reach a limited assurance opinion over a forecast production quantity. The key aspects that frame the assurance work are:

1. The application of ASAE 3450 – when the assurance practitioner is requested to provide assurance on prospective (financial) information prepared by the responsible party based on their assumptions (excluding material hypothetical assumptions).
2. The need for the information provided in the application to comply with the NGER requirement for accuracy. Accuracy refers to having regard to the availability of reasonable resources by a registered corporation or responsible entity and the requirements of the audit determination handbook, uncertainties in emission estimates must be minimised and any estimates must neither be over nor under estimates of the true values at a 95 per cent confidence level.

Sufficient and appropriate evidence

The new facility is a major investment for the company and consequently requires an extensive range of budgeting and planning to be carried out in order to support capital raising activities, inform the internal governance processes and to facilitate effective communication with investors and stock markets. There should be extensive design and planning information available to support the claims and clear evidence of this being subject to management control. These controls are expected to include those focused on ensuring the veracity of any external communications to investors by senior management and those related to the oversight of investment strategy by the Board via its governance frameworks.

The auditors should be able to seek sufficient and appropriate evidence to support the claim along a number of lines of enquiry. These are expected to include the:

• documented financial and engineering plans for the new facility
• internal reporting and the operation of the controls and sign offs relating to the investment approval and communication with investors, and
• application of standard forecasting and analytical techniques by the applicant in their calculation of the reasonably likely level of production.

Where practical the applicants will have included this type of evidence in its basis of preparation.

Financial and engineering

The basis of preparation could include the following supporting evidence:

• the technical and engineering design parameters, plans, mass balance or material flow calculations
• the supply chain information for the input materials such as the nickel matte and nickel concentrate (e.g. contracted inputs or contracted outputs), and
• business case studies and the internal project assessment and design approval documentation (often quite extensive for a major capital project).

Governance and management oversight

Where relevant to the application, the basis of preparation should also include details of how the company management or Board exercise oversight over the release of market-sensitive information such as production forecasts. For example the Board Charter of the company includes an investment oversight committee as part of the overall governance framework (see example below).

The inclusion of this type of information in the basis of preparation will allow the registered greenhouse and energy auditors identify the appropriate controls and seek representations from the company staff charged with governance and oversight.

Production planning data – forward estimates

A particular requirement for the registered greenhouse and energy auditors is to assess the reasonableness of the forecasts of production for a new facility for which there is no historical data. The applicant has developed a forecast of future production of the new facility based on its extensive experience in running a similar site. It has included in the basis of preparation detailed analysis that shows the correlation between the monthly planned production and actual production at a similar facility. The approach that the registered greenhouse and energy auditors may choose to take in the examination of this is discussed below. It should be noted that this approach would need to be complemented by a wider understanding of the company’s systems and controls with particular regard to the applicability of the historical data from one site to the forecast data of another.

The application includes a forward production plans for the new facility for financial year 2018–2019 in which the highest point of production is indicated to occur in late 2018.

Table 1. Planned nickel production at a new facility: Site A.

The estimates of the production quantity have been developed from engineering, production scheduling and material supply data all of which is available to the registered greenhouse and energy auditors and has been included in the basis of preparation and the evidence pack.

An example of analytical approach to gain evidence of reasonableness

A good initial test of the ‘reasonably likely’ claims over the forecast production quantity will be the examination of how proficient the facility operator has been at meeting their production planning targets at any similar facilities. It is common practice for the large resource operators to be involved in ongoing performance benchmarking programs – often run by the relevant trade association or by specialist global consultancies who acts as an honest broker. Comparable information may be available to the registered greenhouse and energy auditors, if required.

In this case study the applicant has provided data that they used to form the production forecast for the new facility. They used data (monthly planned vs. actual) for a comparable facility (called Site A) that it operates in Queensland for the calendar year 2015.

Details from the comparable facility, Site A

The facility (Site A) has been operating for 15 years and has a design production capacity of 50 000 Ni t per year. The operator has provided monthly production data, both planned and actual, in Ni t.

The total planned production for 2015 was 46 730 Ni t. The actual production was 46 544 Ni t only 186 Ni t or 0.4 per cent below the planned value. Figure 1 shows the time series plot of the planned vs. the actual monthly production for Site A. The point of highest actual production, 4 327 Ni t is in July, the same month as predicted but has actually exceeded the maximum forecast amount of 4 160 Ni t.

At first appearance this would seem to provide evidence that the applicant has a good understanding of the nickel refining process and is able achieve their planned production in a reliable manner. However, although the overall total and the monthly maximum for the planned vs. actual are close the registered greenhouse and energy auditors observes that the monthly data shows a degree of variability. The counterfactual to the claim is that the good agreement observed in the overall total may just be random.

Figure 1: Planned and actual production 2015: Site A

The information as presented is unlikely to provide the registered greenhouse and energy auditors with sufficient and appropriate audit evidence to support the applicant’s claim that the monthly production for their new facility will be around 6 670 Ni t per month. The evidence needs to be tested further, firstly to establish that there is in fact a reasonable correlation between planning and production in the historical data and secondly, to establish how the variability in the historical planned vs. actual translates into variability in the foreword estimates.

The test of the correlation between the planned and actual production for Site A is presented in Figure 2. Despite a degree of variance, the overall relationship between the two quantities is strong (an R2 value >0.5). The auditor would be able to conclude that there was a reasonable basis for the production planned by the operator to be reasonably likely to occur.

Figure 2: Correlation of planned and actual production 2015: Site A.

However, the auditor would note that when the operators of Site A forecast a monthly production of 4 000 Ni t the actual outcome was a value of mean of 3 969 Ni t and that judging from the data points in Figure 2 there appears to be a degree of variability involved. There is still the need for the registered greenhouse and energy auditors to obtain evidence to establish that the quantity stated by the applicant complies with the NGER requirement that ‘any estimates must neither be over nor under estimates of the true values at a 95 per cent confidence level’. Consequently, the registered greenhouse and energy auditors should require the applicant to establish and express the upper and lower bounds of the mean value.

The applicant performs that calculation. The result is that the planned production of 4 000 Ni t gave a mean of 3 969 ± plus or minus120 Ni t (exact values are + 130 and – 109, with the slight asymmetry caused by the distribution in the original data not having an exact normal distribution) at a 95 per cent level of confidence. See Figure 2. However, the registered greenhouse and energy auditors would note that has all been calculated from historical data. What are needed next are the values for the 95% interval for the forecast. These values are important as they allow the registered greenhouse and energy auditors to see the level of uncertainty in the forecast and to apply the assurance materiality.

It should be noted that there are many different techniques to forecasting and that registered greenhouse and energy auditors would need to be comfortable that the simple comparative approach used by the applicant would be appropriate.

Confidence intervals and prediction intervals

Within the discipline of statistics, the expression of 95 per cent confidence when applied to predictions (as is the case here) is called the prediction interval (PI). This is calculated using a modified version of the confidence interval formulae. The spread in the values for the 95 per cent PI will be wider than those for the 95 per cent CI. It is important to note that the application of the interval calculation is agnostic as to the choice of forecasting technique. Whatever forecast technique is selected the registered greenhouse and energy auditors should expect the NGER requirement for 95 per cent bounds to be applied.

The applicant performs the PI calculation. The result for Site A is that the planned production of 4 000 Ni t as a valid forward projection would be expected to give an outcome of a mean of 3,969 ± 360 Ni t (exact values are + 371 and – 349) at a 95 per cent level of confidence.

The relationship between the planned vs. actual, the CI and the PI 95 per cent bounds for the data from Site A are shown in Figure 3.

Figure 3: Confidence and prediction intervals: Site A.

Conclusion

The analysis performed above informs the registered greenhouse and energy auditors that the forecast values for the new facility based upon a similar site have an uncertainty of ± 9%. Consequently the applicant’s forecast of monthly production for the ne​w facility of 6 667 Ni t per month when expressed at the 95 per cent interval required by NGER is 6,667 ± 600 Ni t per month. The registered greenhouse and energy auditors would need to include this information in the application of their planned assurance materiality. The applicant will need address any concerns that the registered greenhouse and energy auditors may have on the appropriateness and comparability of Site A to the new facility, which will form part of the audit risk assessment and planning.