Chief Economist statement on Anase

Print Print page   Download PDF PDF image

Introduction

1. The ANASE Study – Attitudes to Noise from Aviation Sources in England - was launched in 2001 with the aim of updating the 1985 ANIS Study – the last Government research on aircraft noise annoyance.

2. The study had three specific objectives;

  • Re-assess attitudes to aircraft  noise in England
  • Re-assess their correlation with LAeq (recommended by the earlier ANIS study as the preferred measure of sound)
  • Attempt to put a monetary value on aircraft noise annoyance, to use in cost benefit analysis

3. The study has been carried out by a team of international experts lead by MVA Consulting and has benefited enormously from the guidance of two groups of distinguished peer reviewers.

  • On the valuation analysis – Professor Ian Bateman, Dr Brett Day, Professor Mark Wardman, and Professor Graham Loomes
  • On the noise annoyance analysis – Stephen Turner (Director of Acoustics at Bureau Veritas) and Peter Havelock, Head of Environmental Research and Consultancy Department at the Civil Aviation Authority.

4. The study contains innovative and ground breaking work - which make it a valuable contribution in this field. The analytical challenges have been significant and this means that the study has both taken longer, and been more resource intensive, than was originally anticipated. We are very grateful both to the study team and to the peer reviewers for their expert analysis in the delivery of the report.

5. Despite this good work - and perhaps inevitably, given the significant analytical challenges – there remain some   material differences of view between the study team and the peer reviewers on the robustness and interpretation of the study’s findings. In particular, the peer reviewers consider that ANASE’s results are not sufficiently robust to use quantitatively in certain aspects of policy making. For example, the peer reviewers say that "they recognise the very challenging task that faced the researchers" and that "the valuation exercises were designed and implemented to a high standard". But they "counsel against using the detailed results and conclusions from ANASE in the development of Government policy" and they argue that "the valuations reported coming from the study are not reliable".

6. The purpose of this note is to consider:

   (a) What can be concluded from the analysis, in particular given the peer reviewers comments.
   (b) What this means for the economics of aviation noise policy.
   (c) What further evidence needs remain.

Findings from the ANASE Study

7. The study’s main findings essentially consider four issues:

   I. Has concern about aircraft noise increased since the earlier ANIS study?
   II. Is there a threshold level of noise at which people start to be particularly concerned?
   III. Can we put a monetary value on people's concern about noise?
   IV. Are the metrics we use to measure sound a good indicator of people's concern?

   I. Has concern about aircraft noise increased since the earlier ANIS study?

8. The evidence in ANASE indicates, in my view, that it is highly probable that concern (or annoyance) with a particular level of aircraft noise is higher than found in the ANIS study in the early 1980s.This finding is in line with the emerging findings from the European Commission’s HYENA Study. It is also consistent with the more general finding that peoples' valuations of environmental impacts tend to increase over time as average incomes (or GDP) grows, a finding which is reflected in DfT's approach to the valuation of noise impacts from road and rail.

9. However, we cannot be certain about:

a) The precise quantitative size of the increase in annoyance since the earlier ANIS Study; people's experience of aircraft sound today is very different to the early 1980s and too many other changes have taken place over this period to enable a robust estimate to be made.

b) Nor can we be certain about the precise quantitative causes of the increase in annoyance; the study suggests that rising incomes and changing preferences are part of the story (as for other environmental impacts) but at least part of the explanation may be changes in the nature of aircraft noise (more quieter aircraft today, compared to fewer but noisier aircraft in the 1980s).  This may have implications for how we measure aircraft sound - see issue 4.

   II. Is there a threshold level of noise at which people start to be particularly concerned?

10. At present, one of the criteria set in the Air Transport White Paper for the expansion of Heathrow relates to the 57dB noise contour; the criterion is that this would not expand beyond its size in Summer 2002. The earlier ANIS study showed a mainly linear relationship between annoyance and Leq, but the researchers also reported some limited evidence of a ‘step’ at 57 dBA Leq (24 hours) although this was not entirely conclusive. The ANIS study stated that any decision on Leq values for aircraft noise policy must therefore be judgemental. Following consultation at the time, the Government subsequently adopted the onset of significant community annoyance as 57 dBA Leq (16 hours) in 1990.

11. However, the ANASE study shows no evidence of a threshold, or step point, at which people become very much more annoyed. The results show that some people are concerned - and value - aircraft noise at relatively low levels and that peoples' annoyance - and valuation - increases as noise levels increase. There is no particular threshold at which the increase in annoyance (and valuation) accelerates significantly. This finding is in line with the research evidence on noise from road transport, and consistent with DfT's approach to the valuation of noise impacts from road and rail.

   III. Can we put a monetary value on people's concern about noise?

12. The aim of this part of the study was to provide evidence which would enable aircraft noise to be valued in cost-benefit analysis in the same manner as presently happens for road and rail projects. However, this proved to be a challenging analytical task and the study's innovative analysis has not produced robust overall valuations of aircraft noise at this stage; it is possible that further work may enable reliable valuations to be estimated from the ANASE data, but this is not certain. In the interim, the valuations which DfT uses for road and rail noise impacts probably provide the best guide (with appropriate sensitivity analysis).

13. The study results do, however, provide good evidence that, as might be expected, night time noise is valued substantially more highly than daytime noise.

   IV. Are the metrics we use to measure sound a good indicator of people's concern?

14. The ANASE analysis shows that the present LAeq metric provides a reasonable indicator to measure people's annoyance, or valuation, of aircraft noise at a particular point in time.

15. But the study's results suggest two potential areas for improvement which are valuable areas for further investigation:

a) First, the study shows robustly that people are far more concerned about aircraft noise at night than they are during the day - the value of night time noise is almost double that of day time noise in the ANASE results (with early morning and late evening noise also of greater concern, although not to the same extent)

b) Second, the study provides indicative evidence to suggest that people may be more concerned about the numbers of aircraft (and slightly less concerned about the sound level of an individual aircraft) than the present LAeq indicator assumes;

Possible implications for the economics of aviation noise policy

16. At present, DfT includes valuations of any increase (or reduction) in noise impacts in the cost-benefit analyses which are carried out to assess the value for money of road and rail projects, However, allowance for noise impacts is not usually included in cost-benefit analyses of aviation infrastructure. The ANASE finding that aircraft noise is becoming of more concern to people, suggests that it would be valuable to do this. And this approach would take into account the impact of increases in noise upon everyone (not just these around the 57dB contour) in line with the second ANASE finding on the absence of a threshold.

17. This would not have been possible at the time of the ANIS study – or for many years thereafter – because reliable evidence on the valuation of noise simply did not exist. This remains the case specifically for aircraft noise. As noted above, the findings from the ANASE study are insufficiently robust for this purpose. Nevertheless the valuations which DfT uses for road and rail noise (introduced, following extensive research, in 2006) could provide a reasonable interim measure (with appropriate sensitivity analysis). These valuations make provision for the rise in value which is expected in the future as GDP and people's incomes rise.

18. Second, the finding that there is no threshold level of sound at which annoyance (and valuation) accelerates significantly may have some implications for the economic validity of the current system of controls on aircraft noise.

19. In terms of economic efficiency, a benchmark or standard (like the 57dB contour at Heathrow) tends to work best at controlling the disbenefits from a pollutant in circumstances where there is a threshold level at which the disbenefit start to rapidly accelerate. It is important in these circumstances to keep below the threshold level and setting a benchmark or standard will usually be the most cost effective way to do this. In contrast, in circumstances where the disbenefits increase incrementally then setting a benchmark or standard will often be inefficient. A better approach is usually to “price” the disbenefits of the pollutant into decision making, either explicitly or, more usually, through cost benefit analysis, enabling the costs of pollution to be balanced against the costs of additional mitigation.

20. The findings from ANASE are consistent with the second of these cases. They show that people’s annoyance, and valuation, increases incrementally as noise levels increase. Although this means that a benchmark will give less good value for money in controlling overall levels of aircraft noise than a cost-benefit approach, nevertheless there remains a role for benchmarks in protecting the interests of those most severely affected by aircraft noise.

Further evidence needs

21. The findings from ANASE suggest that further work would be useful in a number of areas:

  • Further work on numbers and noise
  • Additional stated preference work
  • Further work as recommended by peer reviewers
  • Seek the views of other experts