Air departure tax in Scotland: an economic assessment

An economic assessment on the impact of a 50% reduction in the overall burden of air departure tax and a plan for future monitoring and evaluation.

5. Outcomes – Demand Side & Supply Side

The proposed introduction of ADT, and associated reduction in taxation, has the potential to increase passenger demand for air travel and also improve air connectivity from Scotland. This section describes the approach taken to estimate the demand and supply side responses to reduced ADT, and associated estimates of ADT Payable Trips and total ADT receipts under each of the 9 scenarios described above.

Demand Side Approach

Theoretically, a reduction in tax should lead to a reduction in the cost of air travel, which in turn should generate a corresponding increase in demand. The demand side response to a change in air fares was estimated as below:

  • Existing APD charges were deducted from average air fares for each airport and passenger category to identify the base untaxed cost.
  • The ADT charged under the three proposed scenarios was then calculated together with how much of the saving would be retained by the airline/airports under all three sets of pass-through assumptions. For example, if a return flight from Aberdeen to London cost £150 in 2016, then the untaxed fare was £137. If Band A rates reduce by 50%, and 50% of the saving was passed through to passengers, then the fare under the new ADT scenario would be £146.75 (£137 + 0.5 * 0.5 * £13 = £146.75) .
  • The proportional change in price was calculated and then price elasticities of demand ( PED) applied. PEDs from the UK Aviation Forecasts were applied in the first instance (with a further two sets of alternative PEDs used in sensitivity analysis).
  • ‘Ramp-up’ factors (taken from the (rail) Passenger Demand Forecasting Handbook ( PDFH)) were then applied to reflect the fact that behavioural change, which results from a reduction in price may not be immediate.

Supply Side Approach

As noted previously, some airlines are likely to treat a reduction in air passenger taxation as a reduction in their cost base and therefore may:

  • fly new routes which are currently not viable;
  • reinstate routes which have been withdrawn due to viability issues; and / or
  • convert seasonal routes to year-round routes.

As discussed, the airline industry response is expected to depend on the specific sub-sector of the industry. EasyJet and Ryanair have already published statements that they plan to increase the number of flights they operate out of Scotland, if cuts to APD go ahead, while British Airways has said that while it is supportive of the cuts, it does not expect a substantive increase in supply.

The APEX modelling tool has also been used to test a sample of new routes across the different airports and market segments. These case studies have been used to develop a potential uplift in supply and, in turn, an estimate of passenger numbers which will emerge from this. The APEX modelling work is detailed in Appendix A. In summary, of the 24 routes analysed, the modelling suggested that tax reductions would have a material impact on the viability of 10 routes (42%) and a modest impact on the viability of a further four (17%), so the modelling does support the supposition that a significant supply side response is likely.

In the light of this analysis and the engagement with the industry, Table 5.1 sets out the assumed supply side responses used in the forecasts in the event of a zero pass-through of tax cut to fares i.e. full supply side impact (Scenarios 1c, 2c, 3c). These values were estimated on the basis of Scenario 3c (i.e. a 50% cut in in Band A and Band B, as shown in bold in the table below). In the scenarios where the tax in a band is reduced to zero (i.e. a 100% reduction), we have factored the 50% reduction uplift factors by 1.75. In the absence of suitable data, this figure is a study estimate intended to reflect the fact that doubling the tax cut may not necessarily double the supply side impact.

Similarly, where a partial pass-through was assumed (Scenarios 1b, 2b, 3b), we have halved these values (i.e. assumed half the supply side impact).

Where there is a full pass-through to fares (Scenarios 1a, 2a, 3a), for simplicity we have assumed no additional impact on the supply side other than that associated with accommodating the additional passengers on existing routes.

The biggest supply side impacts have been assumed to be for low cost long haul, but it should be noted that these uplifts are applied to a very low base figure.

Table 5.1: Supply side uplift by scenario and airline type (2019-2021), Zero Pass-through

Scenario Airline Sector Band A Uplift Band B Uplift
1c) 100% cut in Band A only Network / National / Regional 0.0% n/a
Chartered 2.8% per annum / 9% overall n/a
Low Cost Carrier 8.1% per annum / 26% overall n/a
2c) 100% cut in Band B only Network / National / Regional n/a 8.1% per annum / 26% overall
Chartered n/a 5.5% per annum / 17% overall
Low Cost Carrier n/a 17.3% per annum / 61% overall
3c) 50% cut in Band A and 50% cut in Band B Network / National / Regional 0.0% 4.8% per annum / 15% overall
Chartered 1.6% per annum / 5% overall 3.2% per annum / 10% overall
Low Cost Carrier 4.8% per annum / 15% overall 10.5% per annum / 35% overall

A supply side reaction is unlikely to be immediate (depending on other commitments / timetable plans etc.), and consultations with airlines suggest that the lag time will vary by market segment. Our modelling has assumed that any supply side reaction would be staggered over three years, beginning in 2019. It should be noted that the supply side responses here are conservative compared to the public pronouncements made by some airlines in relation to their planned response to cuts in APD / ADT. [28]

Note that we have assumed that the demand associated with the new routes accrues at the same rate as existing routes – i.e. the demand response to supply uplifts is perfectly elastic, the logic being that airlines would presumably anticipate load factors on the new routes in line with existing routes. The analysis also assumes a uniform response regardless of the level of supply side uplift – again on the assumption that airlines would only operate routes if commercially viable.

Published plans from airlines reflect growth on existing conditions and so the corresponding uplift in passengers has been calculated by applying these values to the number of 2016 APD Payable Trips.

Market Composition

In order to provide further context for the analysis, the figures below show firstly the estimated market breakdown by key market segment in total, and secondly the same sectors split by short haul ( ADT Band A) and long haul ( ADT Band B).

These figures have been derived by PBA from analysis of the CAA passenger survey data for Aberdeen, Edinburgh, Glasgow and Prestwick airports and are split by travel by UK residents and Foreign Residents.

Figure 5.1: Scotland’s Air Market by Segment ( CAA Passenger Survey Data)

Figure 5.1: Scotland’s Air Market by Segment (CAA Passenger Survey Data)

Figure 5.2: Scotland’s Air Market by Segment and Short / Long Haul ( CAA Passenger Survey Data)

Figure 5.2: Scotland’s Air Market by Segment and Short / Long Haul (CAA Passenger Survey Data)

The key points which emerge from this in the context of the ADT Scenarios are:

  • The very large majority of ADT paying passengers pay the Band A (short haul) rate. Any change in Band A tax rates, and therefore fares and / or supply will therefore have a bigger absolute impact than the same changes to Band B.
  • Band B comprises a higher proportion of Foreign Leisure at 29% than Band A (13%) – on a proportional basis, cuts to Band B would therefore be expected to have a bigger impact on inbound tourism numbers than cuts to Band A.
  • Around 30% of the market is for business travel – the evidence from e.g. DfT’s UK Aviation Forecasts 2013 suggests that these trips are far less sensitive to changes in fares but would still be responsive to changes in supply.

Headline Results

This section discusses the forecast changes to ADT payable movements and ADT receipts under each of the three scenarios set-out by the Scottish Government and associated variant pass-through assumptions. In this section, a 5-year period from the introduction of ADT in 2018 is reported (i.e. 2018 to 2022). For reference the nine scenarios are listed in the table below.

Table 5.2: ADT Scenarios Modelled

Scenario Degree of Pass-through of Tax Cut to Fares Reductions Supply Side Response [29]
1) 100% cut in Band A, no change in Band B 1a: Full pass-through (100%) None
1b: Partial pass-through (50%) Partial
1c: None (0%) Full
2) 100% cut in Band B, no change in Band A 2a: Full pass-through (100%) None
2b: Partial pass-through (50%) Partial
2c: None (0%) Full
3) 50% cut in Band A and 50% cut in Band B 3a: Full pass-through (100%) None
3b: Partial pass-through (50%) Partial
3c: None (0%) Full

The figures below provide a high-level comparison of the nine combined ADT and pass-through sub-scenarios in terms of ADT payable movements and ADT receipts. These figures reflect our core assumptions of:

  • Underlying Growth: DfT Central Unconstrained Growth (inc. LCLH factor); and
  • Price Elasticities of Demand: DfT UK Aviation Forecasts 2013.

Figure 5.3: Number of ADT payable trips by scenario

Figure 5.3: Number of ADT payable trips by scenario

It can therefore be seen that even in the ‘Do Nothing’ case (i.e. current APD rates), ADT paying passengers are projected to increase from 12.4m to 13.2m per annum by 2022. By 2022, under Scenarios 1c and 3c, the figures rise to 14.6m and 14.2m respectively.

To provide greater clarity on the impacts of each scenario, the change from the Do Nothing case for each is shown in the figure below.

Figure 5.4: Additional ADT payable trips – change from Do Nothing

Figure 5.4: Additional ADT payable trips – change from Do Nothing

As has been noted there are two main transmission mechanisms through which lower rates of ADT will translate through to an impact on passenger numbers: (i) through reduced fares; and (ii) through increased supply resulting from new routes – this additional capacity is assumed to be utilised at the prevailing rate (otherwise these routes would not be operated).

These have differential impacts on demand as noted below:

  • In terms of total volumes, Scenario 1 (short haul abolition) and Scenario 3 (short and long haul cuts) typically produce greater increases in ADT payable passengers than Scenario 2 (long haul abolition). The smallest passenger number increases are seen with Scenario 2, where the PEDs and supply side responses are being applied to a lower base volume.
  • For any scenario, as the degree of assumed tax reduction pass-through reduces (i.e. as you progress from Variant A to Variant C), the level of assumed supply side response (in the shape of new routes operated) increases. As the assumed supply side response is greater than the fares impact, under all scenarios, Variant C sees the greatest increase in passenger numbers.
  • By some margin, the biggest increase in passenger numbers is associated with Scenario 1c, where there is a significant demand side response, which affects some of the main market segments.
  • The incremental demand side expansion can be seen in 2018, 2019 and 2020 and the incremental supply side expansion can be seen in 2019, 2020 and 2021. After this the underlying growth rate (as per the Do Nothing) resumes meaning that the scenarios with no reductions in fares (all Variant C scenarios) flat line. The other scenarios continue to see growth relative to the do nothing due to the lower fares.

The figure below now shows the associated ADT revenue by scenario by year.

Figure 5.5: ADT Tax Revenue by Scenario

Figure 5.5: ADT Tax Revenue by Scenario

As shown here, under the Do Nothing scenario (and the Core underlying growth assumption), the level of ADT tax would be expected to grow to around £270m by 2022. If the DfT Low and High growth scenarios were applied however, this figure could range from £259m to £290m, representing a significant uncertainty. Again, for clarity, the change in ADT revenue from the Do Nothing for each scenario is shown below.

Figure 5.6: Reduction in ADT Tax Revenue from Do Nothing by Scenario

Figure 5.6: Reduction in ADT Tax Revenue from Do Nothing by Scenario

In Scenarios 1 and 2, the tax is abolished for Band A and Band B respectively. This means that the tax raised relates to only Band B or only Band A, and these will therefore be unchanged from the Do Nothing. Given the balance of tax take by band, this means that the tax take would reduce by around 56% and 44% under Scenarios 1 and 2 respectively. Under Scenario 3, once fully established, the tax take would be reduced by between 46% and 50%. These figures account for the additional tax attributable to the additional passengers resulting from lower fares and increased supply. These figures also suggest a degree of honing of the tax rates would be required to align the future tax take with the manifesto commitment of a 50% reduction in tax take. This should become a possibility when ADT outturn data begins to emerge early in 2018-19.

Scenario Based Sensitivity Tests

The impact that alternative underlying growth and PED assumptions would have on these ‘core’ values has been analysed. The purpose of this was to demonstrate how the forecasts are impacted by changes in the underlying assumptions.

In addition to the core test ( DfT Central (unconstrained) Growth / UK Aviation Forecasts Elasticities), the eight sensitivities are: [30]

  • DfT Low Growth / Core ( UK Aviation Forecasts) Elasticities
  • DfT Low Growth / Intervistas Elasticities (higher than DfT)
  • DfT Low Growth / Gillen et al Elasticities (higher than DfT and Intervistas)
  • DfT Core (Central (unconstrained)) Growth / Intervistas Elasticities
  • DfT Core (Central (unconstrained)) Growth / Gillen et al Elasticities
  • DfT High Growth / Core ( UK Aviation Forecasts) Elasticities
  • DfT High Growth / Intervistas Elasticities
  • DfT High Growth / Gillen et al Elasticities

To provide an indication of the degree of sensitivity associated with the differing underlying growth scenarios and elasticities, the table below shows, for each scenario:

  • the Core ADT paying passengers and associated revenue for 2022; and
  • the lowest and highest figures determined from applying the various elasticities and underlying growth scenarios.

Table 5.3: Summary of ADT Scenarios Modelled

ADT Paying Passengers (millions, 2022) ADT Revenue (2022)
DfT passenger growth scenario Core Low High Core Low High
1a 13.8 13.2 15.6 £117m £113m £127m
1b 14.2 13.6 15.5 £117m £113m £127m
1c 14.6 14.0 15.5 £117m £113m £127m
2a 13.2 12.7 14.3 £153m £146m £164m
2b 13.4 12.8 14.4 £153m £146m £164m
2c 13.5 13.0 14.5 £153m £146m £164m
3a 13.5 12.9 14.9 £138m £133m £154m
3b 13.8 13.3 15.0 £143m £138m £156m
3c 14.2 13.6 15.1 £149m £143m £159m

It can therefore be seen that these values typically fall within -5% to +15% of the core values by 2022. These results are reported in more detail in Appendix B where each chart provides a ‘fan’ of results which compares each of the underlying growth and PED combinations with the Core forecast.

Other Issues to Consider

The preceding analysis set out potential outcomes associated with the different scenarios related to the introduction of ADT. The analysis undertaken assumes that, out-with the reduction in duty, all other things remain equal. Whilst this is a pragmatic research assumption, there are a number of wider issues and events that could have a material impact on the forecasts. These are set out below:

  • Opportunities
    • Improvements in aircraft technology are creating a climate for the emergence of low cost long-haul. This could be a major ‘market disruptor’ and presents an important opportunity for Scottish airports. The main prospect in the Scottish context currently is Norwegian, which is offering low cost flights to a number of cities in the United States, predominantly on the East Coast. However, Level, an IAG (International Airlines Group) subsidiary, has commenced flights from Barcelona to the United States, the Dominican Republic and Argentina. If this segment grows in a similar manner to how the short-haul low cost sector did in the 1990s and early 2000s, it could represent a fundamental evolution in the aviation industry. We have included an estimate of the scale of potential impact of this sector in the above analysis, but in reality the overall impact remains uncertain.
  • Risks
    • The industry consultations highlighted the significant risk posed by the United Kingdom’s planned departure from the European Union. On both the demand and supply side, aviation is an industry which relies as far as possible on the free movement of people and assets. The industry consultation highlighted the significant risks to the aviation market posed by a so-called ‘hard Brexit’ and explained that, if these risks materialise, the impacts will outweigh the positive impacts of introducing ADT at lower levels than the current APD regime.
    • The introduction of ADT will stimulate demand for additional air travel, either directly through lower prices or indirectly via additional routes / connections / frequency. In order to keep pace with demand, there may need to be a series of supply side responses to address constraints associated with e.g. terminal capacity, stand capacity and ultimately runway capacity over time. Investment may therefore be required to deliver the long-term demand forecast (this is captured in airport masterplans to some degree), although the forecasts undertaken here do not suggest that this point will be reached in the short term.
    • It is conceivable that reducing the level of ADT would have very little or no impact on either the fares paid by passengers or the level of connectivity provided out of Scotland. In this case, the tax cut would be subsumed within the aviation sector with no discernible supply side impact and therefore no wider economic impact. However, we consider the likelihood of this outcome to be low given the level of competition in the air market in Scotland, the evidence from cuts in similar taxes elsewhere, and the public pronouncements on this matter made by some airlines to date.

It is also worth noting that:

  • Passenger traffic through Aberdeen Airport is disproportionately affected by the performance of the oil and gas market. In the context of business travel from Aberdeen, the trend in the oil price is likely to have a much more significant impact on passenger numbers than the introduction of ADT at lower levels than the current APD regime.
  • APD is not currently paid for trips originating from the Highlands & Islands. It is assumed in this study that this position will continue when ADT is introduced.


Back to top