Benefits and Costs
10.102 Chapter 7 outlined the Commission's approach to deriving estimates of the potential benefits and costs of controlling airfield activities. The models developed in Chapter 7 are now applied to the airfield services supplied by CIAL to aircraft operators.
10.103 The Commission's analysis of the potential benefits of control involves a number of distinct parts: calculation of returns from vesting to date (1989-2001); forecasting of returns into the future (through to the end of the period for which landing charges are presently set, i.e., 2003); assessing any allocative inefficiencies associated with current landing charges; assessing any productive inefficiency; and assessing any dynamic inefficiency. Each are now discussed.
10.104 All the results presented in this Chapter are based on the Commission's assessed airfield asset base and the WACC range for CIAL estimated by the Commission. Appendix 17 contains full workings of the analysis, as well as results and sensitivity analysis based on alternative asset base assumptions.
Historic Analysis of Returns
Introduction
10.105 From an economic perspective, CIAL should be able, on average over time, to earn a normal return on the optimised assets used in providing the services of airfield activities. WACC is used to determine the normal or target return CIAL's assets used for airfield activities, on the grounds that a return equal to the WACC for an entity is a return commensurate with the opportunity cost of capital for that entity. A return in excess of that would suggest CIAL was earning an excessive return, unless those returns reflected efficiency gains and superior performance.
The Calculations
10.106 The Commission has conducted an analysis of the historical returns of the airfield activities of CIAL over the period since vesting, comparing actual returns with target returns (based on the Commission's views on asset base and WACC). The returns have been calculated based on the formula provided in Chapter 7:
Excess Returns ($) = Net Earnings - (Asset Base x WACC)
10.107 The first part of the equation, net earnings, represents CIAL's actual earnings from airfield activities. As noted in Chapter 7, net earnings is computed as earnings before interest, but after tax, depreciation and operating expenses, plus revaluations. In accordance with the principles on asset base determined by the Commission, the revaluations included are those relating only to any revaluations of land to opportunity cost. The second element of the equation (asset base x WACC) represents the target returns.
10.108 The returns are computed annually for each financial year from vesting (1989-2001) separately for the lower bound, upper bound and point estimates of WACC (relevant to that financial year, based on the last price reset). In order to look at trends over time, and not create an outlier in the returns derived in years where there are substantial revaluations, revaluations are spread back to vesting, or the last revaluation.434
10.109 The framework for the analysis is largely the same as that in the Draft Report. However, the spreadsheets used for the analysis have been revised and simplified, and the analysis has been updated to include the 2001 financial year results for CIAL (unavailable when the Draft Report was released). Inputs and assumptions have also been modified where appropriate (as discussed in this Report).
Assumptions and Inputs
10.110 As noted above, for full details of the data used and of the analysis of CIAL, readers are referred to Appendix 17. These include an analysis of the sensitivity of the results to different assumptions or scenarios regarding the appropriate asset base reported below.435 The key assumptions and inputs in the Commission's analysis of historical returns are detailed below.
10.111 Revenue and expenses figures for CIAL's airfield activities are sourced from information provided by CIAL in its submission on the Draft Report and supporting computations since provided. The only figures that have been estimated are those of "other revenue" for 1989-1997, where an estimate of $200,000 is included each year.
10.112 Expense figures for 1999-2001 are sourced from CIAL's financial system and information disclosures. For 1989-1998, expenses were not recorded in terms of airfield activities, so estimates are required. In the analysis included in the Draft Report, the Commission derived estimates of the airfield expenses for these years by extrapolating back from 2000. In response to the Draft Report, CIAL produced its own estimates of airfield expenses for 1989 to 1998. The figures for 1989 are based on the 1989 Travers Morgan pricing model developed for CIAL. Where possible, expense figures for 1990-1998 are directly allocated. Elsewhere, CIAL has interpolated between the proportions in the 1988 model and the actual allocations for 1999 and 2000. This approach applies to expenses other than depreciation. CIAL has derived airfield depreciation figures by constructing an asset register for airfield assets back to 1989. CIAL's approach to estimating expenses takes account of the change in the focus of CIAL's business over time.
10.113 In terms of taxation - as noted in Chapter 7 - the Commission now uses an effective tax rate in its analysis. The effective tax rate is unlevered to fit with the way returns are computed (i.e., before interest). In recent years, the unlevered effective tax rate and the statutory corporate tax rate are the same. The statutory tax rate continues to be applied in the forecast return analysis beyond 2001.
10.114 The Commission's assessment of the appropriate airfield asset base for CIAL as at 30 June 2001 was detailed earlier in this Chapter. In analysing historical returns, the Commission also needed to determine the asset base in each year from vesting through to 2001. The Commission has been able to source asset base figures for earlier years from a combination of the 1999 valuation reports, information disclosures, CIAL's pricing model from the recent consultation round, and the airfield asset register constructed by CIAL to determine depreciation expenses.
10.115 Adjustments to the asset base (and revaluations included as earnings) are made over the period 1988-2000. One adjustment for optimisation over this period relates to the land held for development of the airfield, for which the historic cost values (as advised by CIAL) are removed. Other adjustments to the asset base due to the Commission's chosen method of valuing assets over this period are due to reduction or removal of spread revaluations.
The Results
10.116 The Commission's assessment of the returns earned historically on airfield activities by CIAL are summarised in Table 67. Table 67 provides three different representations of the results: average returns from vesting to date (1989-2001), average returns over the last five years (1997-2001), and the present value of returns from vesting to date as at the end of CIAL's 2001 financial year (30 June 2001). For results for individual years, refer to the detailed results provided in Appendix 17.
Table 67: Returns on Airfield Activities Supplied by CIAL Since Vesting ($000s)
| | Over WACC Range | At Point Estimate of WACC |
|---|
| Average 1989-2001 | -843 to 76 | -348 |
| Average 1997-2001 | -1,525 to -479 | -962 |
| Present Value 1989-2001 | -17,116 to 1,509 | -7,087 |
10.117 The figures in Table 67 suggest results of varying magnitude, but all with largely the same overall result - negative returns. Small positive returns have been earned since vesting at the lower end of the Commission's WACC range.
10.118 As with WIAL and AIAL, the Commission tested the influence of a depreciating asset base on returns. In CIAL's case, the few instances of positive returns appear to be due to the effect of a depreciating asset base. Without needing to examine the influence of productive efficiency gains, the Commission concludes that CIAL has not earned excess returns historically.
10.119 These findings suggest the conclusion that CIAL has not used its market power in airfield activities historically. This appears reasonable, given that CIAL kept its landing charges constant from 1991 to 2001.
Current and Forecast Analysis
10.120 As discussed in Chapter 7, the counterfactual in CIAL's case will be the status quo.
10.121 While analysis of historical returns is useful for evaluating behaviour of CIAL in the past, an analysis of the forecast returns helps to determine whether such results are an indication of the future. The future analysis also presents an evaluation of the efficiency effects of CIAL behaviour, assuming that behaviour in the past continues.
10.122 The Commission uses the year 2001 as a base year for introducing the forward-looking models, as this is the most recent year from which projections can be made.436 The analysis for CIAL projects future returns and inefficiencies out to 2003. The approach is designed to be consistent with the historical analysis, in particular, it takes into account any unrealised capital gains or losses.
Allocative Inefficiency and Excessive Returns
Determining PC, PM and QC
10.123 The Commission has calculated an average price per tonne (Pm) for CIAL's 2001 year based on the landing charge revenues and tonnes landed (Qm) in the 2001 financial year for CIAL. Pm is used to compute Pc. CIAL's price elasticity of demand of [...] (calculated in Chapter 3) is used in calculating Qc, per the model in Chapter 7 (Figure 2). The results of these calculations are presented in Table 68.
Table 68: Average Prices Relative to Competitive Benchmark Prices for CIAL for its 2001 Financial Year
| Over WACC Range | At Point Estimate of WACC |
|---|
| Actual Price (PM) | $6.71 | |
| Efficient Price (PC) | $7.53 to $8.07 | $7.78 |
| Difference, PM-PC | -1.36 -0.82 | -1.07 |
| Actual Output (QM) | 1,779,665 | |
| Efficient Output (QC) | [...] | [...] |
| Difference, QC-QM | [...] | [...] |
10.124 CIAL's actual price for 2001 is below the Commission's efficient price. Similarly, CIAL's actual output for 2001 is above the Commission's efficient output. Allocative inefficiencies are losses of producer surplus, and negative excess returns are also apparent (representing less than normal returns).437
10.125 The Commission has computed the same figures for each of the forecast years (2002-2003). These figures are not detailed here, but can be found in Appendix 17. The figures for 2002 and 2003 include the impact of a full 12 months of new landing charges (2001 data included only six months of increased prices). This has the effect of producing different results for 2002 and 2003. In these years, CIAL's actual price exceeds the efficient price, and CIAL's actual output is less than the efficient output. Slight allocative inefficiencies are forecast for 2002 and 2003, and are losses of consumer surplus. The impact of this is reflected in the averages for 2001 to 2003 in subsequent tables.
Estimates of Allocative Inefficiency and Excess Returns
10.126 The figures above have provided the information to calibrate the model in Figure 2 of Chapter 7. The model can now be used to estimate the potential allocative inefficiencies and excess returns associated with CIAL's 2001 price levels. Because the precise values of marginal cost are not available, although they are known to be very low, it is assumed that marginal cost is 50 cents. It is also assumed, for the purpose of analysing allocative inefficiencies, that there are no productive inefficiencies or cost misallocations, and that levels of service quality demanded by the airlines are being provided (to be discussed below). On this basis, the Commission has estimated that the likely size of the allocative inefficiencies associated with pricing of airfield activities in the 2001 financial year of CIAL.
10.127 Forecast returns are computed using the same formula as that used in the historical analysis. CIAL's actual results for 2001 are used as the base, modified as appropriate based on CIAL's forecasts and growth estimates. In computing forecast returns under scenario 1 for 2002-2003, the Commission has adopted CIAL's forecasts of MCTOW (tonnes landed), expenses and changes in the asset base (i.e., capital expenditure and depreciation). Those figures are then adjusted as to be consistent with the Commission's view on how to calculate the appropriate asset base (e.g., adjustments to depreciation in respect of the move from ODRC to DHC).
10.128 Allocative inefficiency consists of consumer surplus (area BFE in Figure 2, Chapter 7) and producer surplus (area EFHG in Figure 2, Chapter 7). In addition, the excess returns stemming from prices being above the efficient levels cause a redistribution of wealth from acquirers to suppliers (as measured by area PCPMBE in Figure 2). Estimates of these distribution effects are given in Table 69. These transfers are proportionally much larger than the associated allocative inefficiencies, as would be expected, given the highly inelastic demand for airport services. It should be noted, however, that transfers are distributional in nature, not losses to economic efficiency. The figures shown in Table 69 are an average of the three years 2001-2003. Results for individual years are shown in Appendix 17.
Table 69: Estimated Allocative Inefficiencies and Excess Returns for CIAL($000s)
| | Over WACC Range | At Point Estimate of WACC |
|---|
| Consumer Surplus | -4 to 0.3 | -2 |
| Producer Surplus | -43 to 10 | -13 |
| Excess Returns | -758 to 246 | -217 |
10.129 As with allocative inefficiencies, the Commission's findings on excess returns change in 2002. For 2001, the Commission finds negative returns (consistent with the finding of no excess returns historically), but when the new charges take full effect in 2002 and 2003, excess returns of up to, on average, $1.5m per annum are forecast. The levels of allocative inefficiency are immaterial relative to the excess returns forecast.
10.130 It should be noted that for the purpose of the forecast analysis, no expected revaluation gains are estimated for and, although the Commission considers that there may be revaluation gains over the forecast period. If this were the case, this would be likely to suggest that the figures above understate forecast allocative inefficiencies and excess returns.
Cross-Subsidisation
10.131 In Chapter 4, the Commission presented the way it would assess whether there was any cross subsidisation associated with airfield activities by considering:
- Prices charged by aircraft type per landing, with prices per landing dependent on weight bands.
- Cost allocation between airfield activities and other airport activities.
10.132 CIAL determines the price charged by aircraft type per landing by first using a cost allocation model and then establishing weight bands and prices into which different aircraft fall. For airlines, the landing charge they pay for a given aircraft landing is calculated by multiplying a dollar charge per MCTOW by the MCTOW of that aircraft. The key drivers of CIAL's cost allocation model are summarised in Table 70.
Table 70: Basis of Cost Allocation
| | CIAL |
|---|
| Return on the capital cost of land | Landings and m² runway area used except for 1984 runway extension which is based on seats landed438 (number of seats x number of landings) |
| Return on the capital cost of runways and taxiways | Equivalent landings of design aircraft439 and m² runway area used except for 1984 runway extension which is based on seats landed |
| Return on the capital cost of aprons | Seats landed |
| Runway damage (operating costs of sealed surfaces) | Equivalent landings and m² runway area used except for 1984 runway extension which is based on equivalent seats landed (number of seats x equivalent number of landings) |
| Rescue fire service costs | Landings |
10.133 The cost allocation model attempts to identify the causes of costs, and to allocate costs accordingly. The cost of runway damage aims to take account of the wear-and-tear on the runway, and associated taxiway and aprons, caused by aircraft movements, with heavier aircraft causing greater damage. As with WIAL, the cost of rescue fire is allocated based on the number of landings (not on seat capacity). The returns on the various capital costs are related to the size of the aircraft and, therefore, the number of passengers, reflecting demand conditions. This seems to be a reasonable attempt to recover the costs involved.
10.134 CIAL is a multi-product business, and serves a variety of customers. This suggests there is potential for cross-subsidisation to occur across its different activities. Because of the throughput of passengers generated by airfield activities, CIAL can undertake other integrated aeronautical activities (such as the provision of both airfield and terminal facilities) together with significant complementary commercial activities (such as the provision of retail and commercial premises). There are incremental and common costs associated with these activities.
10.135 BARNZ argue that airlines have not received sufficient information from CIAL on the apportionment of common costs to commercial activities and airfield activities to judge whether cross-subsidisation is occurring. It considers disclosures could be enhanced to assist in assessments of such allocations.440
10.136 The Commission considers there is no economically appropriate way to allocate costs, except indirectly via Ramsey Pricing. MCTOW based pricing approximates Ramsey Pricing. The Commission also notes that an analysis of the adequacy of information disclosure regulations is outside the scope of this Inquiry.
10.137 Given the information available, the Commission considers the scope for cross-subsidisation is minimised by CIAL's use of a multiple till approach, which where possible does try to associate costs with their cause and, to some degree, the demands of the various user groups.
Productive Inefficiency
Introduction
10.138 Airports are predominantly fixed costs businesses characterised by economies of scale. As traffic builds up, the runway facilities are better utilised and the fixed costs are spread over a larger number of landings or passengers. In general, therefore, unit costs would go down with increased use, unless an airport invests too much or too soon in new facilities. However, despite the importance of fixed costs for efficiency, the operating costs at airports are also significant.
10.139 The pricing principles in Chapter 4 suggest a productively efficient operation is one that, over the medium-term, meets demand at the lowest possible cost, commensurate with the level of service quality demanded. The Commission has considered the submissions on CIAL's productive efficiency from interested parties. It presents below some of the evidence that has informed its decision on this matter.
10.140 In its submission, CIAL argued that in principle the efficient cost curve depicted as AC' in Figure 3 (Chapter 7) "makes no reference to the likelihood that AC is already lower than it would otherwise be, because of economies of scope".441 It notes that the Commission recognises that economies of scope exist and were mentioned in the competition analysis. CIAL argues it should be given a credit for this, as it believes AC in Figure 3 may be already be lower than the Commission expects. The Commission considers that economies of scope are a relevant consideration for determining the efficient level of average costs (AC'). Whether these economies of scope are captured in AC or AC' is irrelevant to determining what the efficient level of average costs. The Commission does not double count this effect, so it does not agree that credits should be given for cost savings that should be achieved were CIAL behaving efficiently.
Measures of Operating Costs and their Trends
10.141 The major operating expenses of CIAL are depreciation, employee remuneration, repairs and maintenance, and rescue fire services. Of the operating expenses, all but depreciation would appear to be potentially susceptible to productive inefficiency over the medium-term. These might arise, for example, because of overly lavish maintenance expenditure, over-staffing, or excessive levels of staff remuneration.
10.142 In the Draft Report the Commission suggested that productive inefficiencies may be 1% of operating costs (excluding depreciation) for CIAL. In its submission, CIAL questioned the empirical evidence supporting that view442, stating that "[i]n the absence of any empirical insight, it is not possible to judge whether the Commission is "correct" in its approach". CIAL suggested various comparators that could be used, including an assessment of operating costs and inter-airport comparisons.
10.143 CIAL's operating expenses (excluding depreciation) have gone up by about 60% from vesting to 2000. However, aggregate operating costs on their own do not provide sufficient information for evaluating productive efficiency. Relative (per unit) measures of operating costs are needed. Since vesting, Christchurch International Airport has experienced growth in passenger, aircraft movements, cargo and tonnes landed. Passenger and landing data for Christchurch provide a complete record since vesting. It is in relation to landings and passengers, therefore, that CIAL's productive efficiency can be evaluated.443
10.144 On a per passenger and per landing basis, the operating costs (excluding depreciation) of CIAL have risen on average by 1.3% pa and 0.6% pa respectively, when comparing the operating costs in 1989 to those in 2000. CIAL has forecast operating costs (excluding depreciation) over the next three years (2001-2003) to fall and then rise in nominal terms. Despite the reduction in flights from Air New Zealand and the fallout of the 11 September terrorist attacks in the US, forecast passenger numbers and landings are expected to stay the same over the same period. Together, these figures indicate that operating costs (excluding depreciation) per passenger and landing will increased on average by 4% and 7% pa respectively, comparing 2000 figures to those forecast for 2003.
10.145 BARNZ argued in submissions that CIAL should have been able to achieve the same cost reductions as WIAL, which BARNZ claim achieved 7.2% per annum reduction in airfield expenses over the period 1998 to 2000, "while operating under a Deed which specified prices".444 BARNZ argued the 1% figure suggested by the Commission was too low for CIAL. It suggested a 3% figure. BARNZ also noted that airports should benefit from economies of scale and that CIAL should be able "to achieve efficiencies above the level of total factor productivity improvement experienced by the economy as a whole."445
10.146 The Commission notes that, in setting current prices, CIAL has assumed an ongoing annual improvement in productive efficiency of 1% in line with the wider economy.
Benchmarking
10.147 The Commission considers that benchmarking of CIAL's productive efficiency has merit. However, it is also difficult to do. There were no benchmarking exercises presented regarding CIAL's productive efficiency, although CIAL was used as a comparator in the NECG response on LEK's study of AIAL's productive efficiency.446 In this study, CIAL's productive efficiency compares favourably with that of AIAL.
10.148 Despite the trend in rising operating costs (excluding depreciation) on a per passenger and landing basis, CIAL still had lower operating costs (excluding depreciation) per passenger and per landing than AIAL for 2000 (10% and 56% respectively), although it had higher operating costs (excluding depreciation) per passenger and per landing than WIAL for 2000 (18% and 11% respectively).
Summary
10.149 The Commission has adopted a pragmatic approach (to an issue that involves significant uncertainty) by presenting productive inefficiencies as a percentage range of operating costs (excluding depreciation).
10.150 Given the considerations above, the Commission considers there is likely to be some room for improvement in the productive efficiency of the airfield activities at Christchurch. Operating costs (excluding depreciation) have risen (and are expected as a general trend to continue to rise) on a per passenger and landing basis. CIAL has lower operating costs (excluding depreciation) on a per landing and passenger basis compared to AIAL, but higher costs compared to WIAL. The Commission considers there is scope for these figures to be improved upon over the forecast period. The Commission considers there to be productive inefficiency of the order of 1% to 2% of operating costs (excluding depreciation) at Christchurch.
10.151 Table 71 presents the levels of potential productive efficiency benefits based on this range. The figures shown in Table 71 include an average of the three years 2001-2003.
Table 71: Potential Productive Inefficiency at Christchurch International Airport ($000s)
| | 1-2% Range |
|---|
| 2001 | 83 to 166 |
| 2002 | 74 to 149 |
| 2003 | 80 to 121 |
| Average 2001-2003 | 79 to 159 |
Dynamic Inefficiency
10.152 Dynamic efficiency relates to minimising costs over time through investment, and to the quantity and quality of assets used by an entity. Inefficiencies can arise where investments that would be optimal are not made (or made at the wrong time), or investment has led to too many assets being acquired - meaning that some assets are not "used or useful" in meeting demand - or because some assets are "gold plated". Given the nature of airfield activities, the acquisition of too many assets (most likely land) is more likely to be a potential source of dynamic inefficiency than "gold plating". The issue then becomes one of whether the optimal amount of assets is being used to provide the service.
10.153 As noted earlier, there appears to be some over-investment by CIAL in airfield activities in the form of land held for future development. This land has been optimised out of CIAL's asset base (including by CIAL itself). However, there may also be dynamic efficiency implications regarding this optimised land particularly if the land is not put to its best alternative use until it is used for airfield activities.
10.154 CIAL argued in its submission to the Draft Report that optimising out the land from the asset base for charging purposes, and evaluating the efficiency of the investment itself, constituted double counting.447
10.155 In the Draft Report, the Commission did not consider the calculation of allocative inefficiencies and dynamic inefficiencies as double counting, but merely taking account of all efficiency effects. having considered CIAL's submission, the Commission considers that, if the optimised land will form part of the airfield activities markets in the medium-term, and was a prudent investment, then consideration of any inefficiencies in the interim may constitute double counting. Double counting would be inconsistent with the Commission's principles in Chapter 4, in particular it would compromise prices being dynamically efficient over the medium-term.
10.156 The Commission has attempted to quantify the extent of dynamic inefficiencies at CIAL by first determining that proportion of land that may have resulted in dynamic inefficiency (either through the purchase or subsequent use). Determining whether the optimised land, even if it is not presently used or useful, is prudently held over the medium-term provided basis on which the decision could be made. That land which is not prudently held over the medium-term is subject to dynamic inefficiencies and should be used elsewhere (or not acquired in the first place) as it will not earn its opportunity cost in the medium-term in the present case. Land that is prudently held is expected to enter the asset base on which charges are based over the medium-term, and will, therefore, earn its opportunity cost in the future.
10.157 The portion of land potentially subject to dynamic inefficiency at Christchurch is that proportion of land which will not be needed for airfield activities over the medium-term. In 2001, the proportion of land was 32% of the total airfield land. This proportion of land was optimised by CIAL in determining its charges. The presence of this excess land holding could nonetheless imply sub-optimal use of this land, and could imply dynamic inefficiencies still existed.
10.158 However, as the optimal use for this land (while the airport continues to operate) is deemed to be rural land uses, and because it is currently used for these purposes, there are unlikely to be dynamic inefficiencies associated with the use of this land. No dynamic inefficiencies have therefore been attached to these surplus assets at CIAL.
Costs of Control
10.159 Costs of control are forward looking by the very nature of this Inquiry. The Commission considers that the direct costs of control (including both the regulators' and market participants' costs) for a single airport might be $1-$2 million in a review year, and $0.5-$1 million in other years. Over a five-year period, with one review, this suggested an annual average of between $0.6-$1.2 million per year at each airport.
10.160 The total costs of control are not easy to estimate. In Chapter 7, the Commission considered that, in the absence of any superior alternatives, the indirect costs of control could be measured by considering what extent of the potential benefits of control could be realised by control. The Commission determined that the indirect costs of control as a proportion of potential benefits would be 25% of any excess returns and producer surplus, 43.75% of any consumer surplus, and from 50% to 100% of any dynamic inefficiencies. The productive efficiency costs of control are estimated at 0 to 2% of operating costs (less depreciation) and are offset against the range of possible benefits of control regarding productive inefficiencies.
10.161 Table 72 summarises the likely indirect costs, per annum, of controlling the airfield services supplied by CIAL to aircraft operators. The figures shown in Table 72 are an average of the three years 2001-2003. Results for individual years are shown in Appendix 17.
Table 72: Likely Indirect Costs of Controlling CIAL ($000s)
| | Over WACC Range | At Point Estimate |
|---|
| Consumer Surplus | -2 to 0.1 | -0.9 |
| Producer Surplus | -10 to 2 | -3 |
| Excess Returns | 48 to 182 | 103 |
| Productive Inefficiency | 0 to 159 | 79 |
| Dynamic Inefficiency | 0 | 0 |
Back to Top