Benefits
7.15 The potential benefits of control relate to reducing any inefficiencies (allocative, productive and dynamic) and/or excess returns in a market. An analysis of performance in the counterfactual compared to an efficiently operating market could be used to measure these benefits. However, it cannot be assumed that all of the potential benefits would actually be realised in practice through the imposition of control. Clearly, different forms of control may be more or less effective. Further, control can create additional costs to those emerging from lighter forms of regulation, as discussed below.
7.16 A useful starting point for the analysis of the benefits of control remains the inefficiencies that may be present in the counterfactual. Chapter 4 presented the pricing principles and explained the three aspects of economic efficiency (allocative, productive and dynamic efficiency). The sources of potential benefit include:
- Allocative inefficiency being reduced or eliminated by control (with the resulting lower prices passed on to consumers). Inefficient levels of service quality for the price charged could also be addressed through control. There may also be indirect or spill-over benefits from any lower prices to related markets.
- Excess returns being reduced or eliminated by control, with a transfer of wealth from suppliers to consumers (being a net benefit to acquirers). The increase in consumers' wealth is matched by a reduction in suppliers' wealth (resulting in zero net public benefit).
- Productive inefficiency being reduced or eliminated by control (with resulting cost savings likely to be passed on to consumers in lower prices).
- Dynamic inefficiency being reduced or eliminated by control, because of better utilisation/allocation of resources. This would benefit New Zealand and potentially lower required revenue from landing charges (to cover costs) likely leading to lower prices for consumers.
7.17 The sources of potential benefits are now discussed below. The models used in the airport-specific chapters to quantify the potential benefits of control over time are also introduced.
Allocative Efficiency Gains
7.18 Allocative inefficiencies could exist in the past, present or future. The evaluation of allocative inefficiencies at the airports would require a calculation of the efficient price at each airport over time. The total revenue and cost for airfield activities could be used to do this. Cost would be measured by the sum of appropriate airfield expenses and a normal return on investment, the latter being calculated by multiplying the appropriate asset base by an appropriate WACC. Revenue would be measured by multiplying airfield charges by the level of output.
7.19 Where revenue exceeds cost, or equivalently, where the airport's actual returns on airfield activities (after allowing for expenses) are greater than normal returns, prices would be above the efficient level. From this, the potential benefits to acquirers can be estimated, if control were to have the effect of reducing price at each airport to the efficient level.
7.20 CIAL was critical of the building blocks approach taken by the Commission. CIAL's preferred approach was a DCF approach, which it argued could produce a smoother price trend.223 The implication was that "apparent" over-recovery at one point in time may be matched by "apparent" under-recovery at other times. The Commission considers that, properly applied, either a building blocks or a DCF model should produce similar results over the medium-term.
7.21 The Commission has undertaken current, forecast and historical analysis for each airport, based on both actual data and forecast information. The models used for these purposes are described in the sub-sections that follow.
The Models
Analysing Current and Future Performance
7.22 In the Draft Report, the Commission presented a model for measuring present and future allocative inefficiencies and excess returns. In light of submissions, the Commission has adjusted its model, as explained below.
7.23 Given the inelastic demand for airfield services, large price increases would likely have minimal adverse impact on demand by consumers. Deadweight losses (DWL) associated with inefficient pricing would emerge in the airfield services market, but these would be likely to be small relative to the size of the distribution effects (i.e., the wealth transfer from consumers to suppliers through the higher prices). These effects are explained in Figure 2, which shows a stylised demand and cost structure of a typical, single runway, airport.
7.24 The vertical and horizontal axes of the figure are scaled in terms of the average price per tonne, and the number of tonnes landed, respectively. The cost structure of the airfield activities is such that fixed costs make up a large proportion of total costs, while marginal costs are very low so long as excess capacity exists. The point at which the demand curve (D) meets the price axis is not shown on the chart, but is termed point A. The demand curve is assumed to be linear for simplicity.
Figure 2: Estimating Allocative Inefficiency in the Airfield Services Market
7.25 An airport must cover all of its costs, including fixed and overhead costs, so the competitive average price is assumed to be set above marginal cost (MC) at PC, with output at QC (PC includes an appropriate level of normal returns reflecting an appropriate asset base and WACC). In other words, if the airport's average cost curve were to be drawn on Figure 2, it would slope downward and intersect the demand curve at point F, the downward slope reflecting the spreading of overheads over the larger landed tonnage. At this position, gross surplus would be represented by the area OAFQC, from which variable costs of OCHQc would have to be deducted. The remaining net surplus (or net benefit from production) is split between consumers' surplus of PCAF, and suppliers' surplus - covering fixed costs and normal returns - of CPCFH.
Inefficient pricing would be reflected in the price being raised above the competitive level to, say, PM, with output in consequence shrinking to QM. This would result in:
- A loss of net surplus equal to the area BFHG. This loss is shared between acquirers' (consumer) surplus of BFE and the supplier's (producer) surplus of EFHG.
- Resources no longer required because of the reduction in output, represented by the area GHQCQM, which are assumed to be absorbed elsewhere in the economy, with no impact on welfare.
- Additional surplus gained by the supplier at the expense of acquirers, depicted by area PCPMBE, which is a wealth transfer from acquirers. In efficiency terms, this transfer is assumed to have no direct effect, since one party gains at the expense of the other.
7.27 Hence, the detriment arising from the loss of allocative efficiency in the airfield services market is represented by the area BFHG. The supplier earns excess returns equal to the value of area PCPMBE.224
7.28 An alternative possibility is that the actual price could be below the competitive price. To generate that outcome using Figure 2, the M subscript can now be treated as indicating the competitive position, and the C subscript the actual position. In this case, acquirers of airfield services benefit at the expense of the service provider, who earns less than normal returns. The total revenue produced by the service is represented by the area OPCFQC, and the total cost is equal to OPMJQC, leaving a loss to the airport of PCPMJF.225 The deadweight loss from the over-production by QMQC is shown by the triangular area BJF. In this scenario, as in the previous one, the deadweight loss is likely to be very small relative to the wealth transfer from, in this case, suppliers to consumers/acquirers.
7.29 The low price elasticity of demand for airfield services suggests that the output decrease between QM and QC could be quite small. This would suggest that the transfer of wealth from suppliers to acquirers, as represented by the area PCPMBE, associated with monopoly pricing would be likely to greatly exceed the loss of allocative efficiency, denoted by area BFHG. Applying the same logic, an attempt to return prices to the competitive level through the use of control would, if successful, reverse these changes. The wealth transfer of PcPmBE would revert back to acquirers, and allocative efficiency would improve by BFHG. From a narrow acquirers' perspective, they would benefit from the lower prices by the gain in consumers' surplus PCPMBF.
7.30 NERA argued that the marginal costs of airfield activities were not zero as the Commission has assumed in the Draft Report.226 They did not have their own assessment of marginal costs.227 If marginal costs are not zero then the potential producer surplus gains estimated by the Commission may be overstated.
7.31 To determine a marginal cost figure for the purpose of analysis, the Commission has looked at the repairs and maintenance costs on the runways and dividing this by tonnes landed. A figure so derived, however, is likely to be an average, rather than marginal, figure unless the two are the same because all costs are variable and increase in a linear fashion with tonnes landed. This being unlikely, the estimated marginal cost is likely to exceed the actual marginal cost. In addition, the Commission considered fluctuations over time caused by sporadic maintenance on the runways. Based on all the information available to it, the Commission considers a marginal cost of 50 cents per tonne landed is appropriate for all three airports for the purpose of analysis and should be treated as constant over the period of analysis.
7.32 CIAL objected to the model in the Draft Report on the grounds that the demand curve was a stylised portrayal of a derived demand. It contended that the link between output and final prices passengers faced was what determined the slope of the demand curve. It considered this would be invariant and no output change would occur.228 The Commission considers that the airlines are not likely to be completely invariant to price changes, and that output could adjust because of price changes that affected their overall costs. The calculation of the price elasticity of demand recognises that there is a marginal effect for both intermediate and final consumers. Even marginal effects can be significant, in principle, and change behaviour.
7.33 WIAL argued it was given no credit in the model in the Draft Report for the efficiency of its pricing structure.229 The Commission accepts that, the above model (Figure 2) cannot deal with the structure of prices. In preparing the Draft Report, the Commission considered there might be scope for more efficient pricing structures, and the airports are not penalised for this.
7.34 At the Conference, Kerrin Vautier (an expert for CIAL) questioned whether the model was appropriate, because it appeared to be a model based on perfect competition and not one of workable and effective competition.230 As noted in Chapter 4, the Commission considers that wording in the Draft Report may have given this impression, but in its application, the approach taken by the Commission did apply a workable or effective competition standard. Using WACC to determine the level of normal returns on the asset base is consistent with a workable and effective standard rather than a perfect competition standard (where returns would be based on marginal costs).
7.35 Dr Lattimore (an expert for Christchurch City Council) suggested a bilateral monopoly model was appropriate for evaluating the behaviour of airports and airlines, which suggests prices are negotiated between two parties of equal power.231 The Commission considers that the situation for airfield services does not involve a single buyer and seller of airfield services, as there are a number of airlines buying airfield services (notwithstanding there are a few key buyers). The Commission also notes Scherer and Ross' argument that: "The theory of bilateral monopoly is indeterminate with a vengeance." It adds that, even if buyer and seller collaborate to establish a joint profit-maximisation output, the "price is indeterminate over a potentially wide range."232 In contrast to the bilateral monopoly model, which relies on prejudging the relative power of both parties, the outputs of the Commission's model reveal the relative power of both the airlines and airports. The Commission considers that, the bilateral monopoly model, regardless of whether it has merit, does not mean the approach the Commission has taken is inappropriate.
Analysing Historical Performance
7.36 The above model can best be applied in a current setting where the variables required to calculate prices and quantities are known. The year 2000 was the first year airports were required to disclose segmented financial accounts. Segmented forecast figures are also available. However, for prior years this is not always the case.
7.37 In preparing the Draft Report, the Commission considered that data availability limited its historical analysis to an examination of whether excess returns existed. Determining efficient prices for each of the years was not possible, only rates of return were computed historically.
7.38 As a general principle, rate-of-return figures must be used with care when assessing efficiency, as the returns reflect changes in both revenues (pricing) and costs. A firm with market power may earn high returns by raising prices rather than lowering costs. Excess returns might be present, but be absorbed in higher costs, so that allocative inefficiency is both obscured, and augmented by a further loss in the form of productive inefficiency. For these reasons, the efficiency with which resources are being used should ideally be assessed. However, the Commission is not able to do so in any detailed way.
7.39 In the Draft Report, the Commission had to extrapolate certain historical expense data. Both AIAL and CIAL criticised the Commission's estimation of historical airfield expenses.233 They suggested expenses were higher in earlier years than estimated by the Commission, because of the greater proportion of total revenues generated by airfield activities in those years (than present). This is certainly true. Airfield activities are a smaller part of AIAL's and CIAL's business now than they were at vesting (1988).
7.40 The Commission has reconsidered the data on which it conducted its analysis of historical performance in the Draft Report. This Commission has also reviewed the methodology and formulae used.234 A representation of the Commission's methodology follows, as well as discussion of revisions to any of the estimates of expenses that were made. It should be noted that the analysis is an economic one, not an accounting one. The results from the historical analysis, including data revisions, are presented in the airport-specific chapters.
7.41 In the Draft Report, the Commission computed percentage rate-of-return figures for historical years using the Accounting Rate of Profit (ARP) formula. The figures presented in this Report are not returns based on the ARP formula. Instead, the Commission has calculated dollar excess returns for each year according to the following formula:
Excess Returns ($) = Net Earnings - (Asset Base x WACC)
7.42 The first part of the formula, net earnings, represents an airport's actual earnings from airfield activities. Net earnings is computed as earnings before interest after tax, depreciation and operating expenses plus any revaluation gains or loses. In accordance with the principles on asset base determined by the Commission, the revaluations included are only those relating to revaluations of land to opportunity cost. The second element of the formula (Asset Base x WACC) represents the target returns. The asset base and WACC numbers used are those determined by the Commission. As with the Draft Report, interest is still excluded. The asset base used is "beginning year" (i.e., as at the start of the financial year).
7.43 The returns are computed annually for each airport for each financial year from vesting to 2001, separately for the lower bound, upper bound and point estimates of WACC (relevant to that financial year, based on the last price reset).
7.44 As in the Draft Report, revaluations are spread back over time to the last revaluation or vesting. However, the basis by which revaluations are spread has changed. Revaluations are now spread entirely based on the Housing Group of the Consumer Price Index (CPI) for each airport's region. In the Draft Report, the Commission used the New Zealand-wide all groups CPI, with a wash-up based on revenue. The change in approach has been made in light of submissions from AIAL which questioned the relevance of the calculations in the Draft Report.235 Use of regional Housing Group CPI figures is likely to be more reflective of changing land values than the All Groups CPI used in the Draft Report.
7.45 The operating revenue, expense and asset-base figures for each airport are based on a combination of data sourced from the published financial statements of the airports, recent disclosure financial statements, and additional information supplied by the airports to the Commission during consultation. Where gaps still existed in data, the Commission extrapolated from the data available to derive estimates for the missing figures. Submissions on the Draft Report have helped refine these estimates. Explanations of the calculations and figures used for each airport are detailed in the airport-specific chapters.
7.46 In the Draft Report the Commission assumed that profits on airfield activities were taxed at 33%, even though an airport company as a whole could have paid a different effective tax rate. AIAL submitted that its effective tax rate on airfield activities was between 42-36% (between 1989 to 1996), due largely to the non-deductibility (or lower deductibility rate) of depreciation on certain airfield assets.236 At the Conference, AIAL explained that the higher effective tax rate claimed by AIAL in its submission could largely be explained by different treatment by AIAL and the Inland Revenue Department (IRD) of depreciation on sealed surfaces. AIAL believed depreciation of the runway should have been over 16 years, roughly matching the useful life of the runway. However, AIAL claimed the IRD allowed depreciation claims on the runway based on a 40-year period. There was, as a result, a significant difference in tax expense under the two approaches, with AIAL's approach leading to a much higher actual tax expense.237
7.47 In cross submissions, the airlines argued that "proper consideration of the useful lives of assets" could see a reduction in difference between AIAL's and IRD's depreciation expenses. They conclude that "given the uncertainties surrounding tax rates it is considered that the use of the 33% rate provides the most logical approach to the issue."238
7.48 The Commission considers that, for consistency in the treatment, if it accepts an airport's expenses, it should also use its effective tax rate.239 Where the Commission does not accept the airport's expenses (e.g., if it does not accept the airport's depreciation expense, compared to that of the IRD), or if effective tax rates are not available, then the 33% statutory rate should be used (although, the latter is not necessary in the context of this Inquiry).
7.49 In terms of taxation, the Commission now uses an effective tax rate in its analysis of returns. 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.
Spill-Over Effects
7.50 The presence of excess returns and allocative inefficiency losses in the airfield services market is likely to lead to some further inefficiencies in the form of spill-over effects to other markets. These could arise in two different ways:
- The additional profits accruing to the airport could be spent on relatively inefficient new investment spending or inflated operating costs.
- Outcomes in downstream markets related to the aircraft movement market will be distorted. Such markets could include those servicing domestic passenger travel, international passenger travel, domestic freight, and international freight. Where these are competitive, even small rises in costs may have significant output effects, creating dead weight losses in both those markets, and in other markets associated with them. The effects may be smaller in the associated markets, and more dispersed, but could potentially create a significant cumulative effect.
7.51 Although these effects are difficult to measure, they should ideally be incorporated into the assessment of the effects of monopoly pricing. The Commission did not try to quantify these effects in the Draft Report. Airlines submitted that these effects should be qualitatively considered in the Commission's report to the Minister.240
Service Quality
7.52 The primary focus on price implicitly assumes that service quality is maintained at the level consumers desire and are prepared to pay for. The airports currently provide information on interruptions to their services, pursuant to the disclosure requirements in the Airport Authorities (Airport Companies Information Disclosure) Regulations 1999. The number and duration of interruptions disclosed cover those relating to runway services, stand position services, airbridge services, and baggage handling systems.
7.53 Of the above disclosures, runway interruptions are a relevant indicator of the service quality of airfield activities. Such interruptions appear to have been infrequent, and do not appear to suggest inferior service quality. BARNZ submitted that inferior service quality for airfield activities is not an issue at any of the airports at present.241 The Commission considers no allocative efficiency benefits could be attained from control in this regard.
Excess Returns (Wealth Transfers)
7.54 As discussed in Chapter 4 on pricing principles, airports should be able, over the medium-term, to earn a normal return on the assets used in providing the services of airfield activities. An appropriate WACC and asset base can be used to determine the normal return on airfield activities at each airport.
7.55 An actual return in excess of a normal return over the medium-term would suggest that the entity was earning excess returns, unless those returns reflected superior performance (assuming costs are minimised). These excess returns represent a transfer of wealth from consumers to producers, and imply a loss of allocative efficiency.
7.56 From an efficiency perspective, wealth transfers between different groups within the economy are ignored. The efficiency standard is concerned with increasing welfare overall, regardless of who benefits directly. Therefore, any reduction in excess returns would not be considered a benefit from an efficiency perspective, although the presence of excess returns is indicative that allocative inefficiencies may exist and may also have other spill-over efficiency effects, as described above.
Summary
7.57 The preceding section has highlighted the potential benefits of control emerging from allocative inefficiencies and excess returns. The Commission uses two models for determining these benefits, one is a current/forecast analysis, while the other is a historical analysis.
Productive Efficiency Gains
7.58 A productively efficient operation is one that meets demand at the lowest possible cost. The impact of productive inefficiencies in the airfield activities of an airport can be modelled by further developing Figure 2, as shown in Figure 3. The further assumptions built into the model are:
- The competitive price and output is assumed to be found, as before, at the point where the existing average cost (AC) curve intersects with the demand (D) curve.
- All productive inefficiency is assumed to be felt in fixed costs, so that average fixed costs are inflated, and the AC curve is "too high". This assumption is made to simplify the graphical illustration of the effects of productive inefficiency.242 The level of the average cost curve when costs are minimised is at AC'.
Figure 3: Productive Inefficiency in the Airfield Services Market
7.59 The inefficiently high level of costs results in welfare effects that can be analysed at two levels. First, as a productive inefficiency, the wastage of resources is an outright loss, as their transfer to other productive employments would lead to no loss of output in airfield activities. In terms of Figure 3, this loss is measured at a given output by the vertical distance between AC and AC', multiplied by that output.
7.60 Second, in an efficient setting the inflated costs would not be present, so that the competitive average cost curve would be AC', not AC as assumed so far. This, in turn, would mean that the efficient price and output would be PC' and QC' respectively, not PC and QC as assumed in Figure 2. As a consequence, the allocative inefficiency loss and wealth transfer flowing from price at PM being above the efficient level is larger than previously estimated. The allocative efficiency loss increases from BEF to BJK, and the transfer increases from PCPMBE to PC'PMBJ.
7.61 In summary, the model used in Figure 3 shows that, if productive inefficiency in the counterfactual were found in the costs, and if those inefficiencies were to be eliminated under control through the pressure of lower prices forcing greater efficiency, this would allow a further reduction in prices beyond that described in Figure 2.
7.62 However, without a precise measure of the slope of the AC' curve, it is not possible to calculate the additional allocative efficiency effect (or those proportions that reflect consumer, and producer, surplus gains respectively). Accordingly, a conservative approach is taken, with only wasted resource measured as a potential benefit of control. This is the same approach as in the Draft Report.
7.63 A difficulty lies in estimating the extent of productive inefficiency (if any) in practice. In the Draft Report, the Commission wanted to get a feel as to how significant those inefficiencies might be by assuming they amounted to 1% of relevant operating costs. This approach was intended (and so stated) to elicit comments from interested parties.
7.64 The airports were critical of the Commission's approach of applying a blanket 1% across all airports. They argued that the evaluation should be done separately. The airlines equally argued that different amounts of inefficiency existed at each airport.
7.65 AIAL argued in submissions that the 1% efficiency figure on operating costs used by the Commission was in effect a 4% efficiency figure on variable costs. To determine this 4% figure, AIAL assumed 75% of operating expenses (including employee costs, which is the greatest expense) were fixed.243 However, the Commission applied the 1% figure only to operating costs. Depreciation expenses were considered fixed and were deducted from operating costs before the 1% figure was applied. It is arguable, however, that some costs classified by AIAL as being fixed are also variable to some extent, particularly if a longer-term perspective were taken. Nevertheless, the Commission conservatively assumed that these were all fixed.
7.66 The measuring of historical productive inefficiencies is complicated by data availability. Nonetheless, the airports claim to have made improvements in productive efficiency in the past. However, the airlines suggest that CIAL and AIAL should have been able to achieve cost savings commensurate with WIAL and that CIAL and AIAL should achieve 3% productivity improvements into the future. BARNZ also referred to an ACCC decision in 2001 that cost savings of 4% were possible at Sydney Airport.244 It believed 3% for New Zealand airports was not unreasonable.
7.67 The potential productive inefficiencies argued in submissions, therefore, ranged from 0% to 3% of operating costs.
7.68 The Commission considers it desirable to evaluate each airport separately in terms of productive efficiency; however, this is difficult in practice. For example, declining costs over time may simply reflect increasing output in the presence of economies of scale, rather than any improvement in productive efficiency per se. Nonetheless, an assessment for each airport is presented in the airport-specific chapters.
Dynamic Efficiency Gains
7.69 In the Draft Report, dynamic inefficiencies were evaluated by considering whether surplus assets earned a return commensurate with what could be earned in that next best alternative use. This was evaluated in the context of the relatively high valuations given to such assets by the airports.
7.70 AIAL objected to the Commission's assessment of dynamic inefficiency on the grounds that optimising the land from the asset base for charging purposes, and evaluating the efficiency of the investment itself, constituted double counting.245 CIAL argued along similar lines. As they had removed the land for charging purposes, CIAL argued that it would be inappropriate for the Commission to evaluate the efficiency or otherwise of its current use of the land.246
7.71 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. Given submitters comments, however, the Commission has re-evaluated whether land held for future airfield activities should be seen as forming part of the airfield services markets over the medium-term.
7.72 It seems clear that both CIAL and AIAL see a certain amount of the land they hold for future development as forming part of the airfield activities (even if the land is excluded for charging purposes in CIAL's case). Accordingly, it may indeed be relevant for the Commission to evaluate the efficiency of holding such land which may not be used over the medium-term.
7.73 The Commission considers that, where it is reasonable to hold land for future airfield activities (e.g., because it is prudent and efficient to do so over the medium-term), then no dynamic inefficiencies should be included as potential benefits of control. However, such land could still be excluded for pricing purposes, as discussed in Chapter 5 on Asset Base.
7.74 Where it is not reasonable to hold such land (or a proportion thereof) for future airfield activities over the medium-term, then the dynamic inefficiencies associated with holding such land could be a potential benefit of control. The approach taken by the Commission in the Draft Report for measuring these dynamic inefficiencies, although not perfect, does provide a relevant approximation. No alternative approach was raised in submissions. The mathematical approach to measuring dynamic inefficiencies involves:
- Determining the annual returns on the optimised land that is not presently used or useful.
- Treating the annual return on this land as a perpetuity (since land does not depreciate), allowing a valuation for the land in its current use to be derived.
- Comparing the value of the land in its current use to the opportunity cost value of that land. If the opportunity cost value exceeds the value derived from its current use, then dynamic inefficiencies exist, otherwise they do not. To determine an annual figure for any dynamic inefficiencies, the difference between the two values is converted into an annual figure by calculating the perpetuity which, when discounted at the risk free rate of return, equals the difference.
7.75 The above approach is largely the same as in the Draft Report, although rather than using the airports' value of the optimised land (which is not used or useful), the Commission uses the opportunity cost valuation of such land, as the Commission considers this best reflects what could be earned from the land in its next best alternative use.
Back to Top