M-A-G  MACC INFO  WIGGINS TDC MP's LINKS  NEW OOP's                          If you want to make comments or find out more about the Manston Airport Group, email on 
g r o u p @ m - a - g. f s n e t. c o. u k  or post to P.O. Box 311, Ramsgate CT11 9XL. ALL enquiries are confidential.

Emeritus Professor
J. Parry Lewis
Urban and Regional Economist
31 Chapel Place
Ramsgate, Kent CT11 9SB
Telephone: 01843 592668
e-mail: jparrylewis@riter.freeserve.co.uk 

_____________________________________________________________

A Comment on
An Environmental Statement for the Development of London Manston Airport

The Environmental Statement will be referred to by use of ADL, the initials of its authors, Arthur D Little. This comment also refers to

1. London Manston Airport, A Strategy for Success, 2001 (known as The Masterplan )
2. Oxford Economic Forecasting, The Contribution of the Aviation Industry to the UK Economy 1999, which was commissioned jointly by the Airport Operators Association, the British Air Transport Association and the DETR (referred to as OEF).
3. Department of the Environment, Transport and the Regions, The Future of Aviation, 2000 (referred to as DETR).

Tables quoted directly from these reports are numbered as in the original. Tables adapted or constructed for this comment carry the prefix L (for Lewis).

• INTRODUCTION AND SUMMARY
• 1. AIR TRAFFIC
• 2. EMPLOYMENT
• 3. INCOMES
• 4. OTHER ECONOMIC EFFECTS
• 5. THE FUNDAMENTAL ASSUMPTION
• APPENDIX 1

In this comment I am concerned mainly with ADL's economic analysis. This concentrates on employment and incomes and virtually ignores adverse economic impacts (paragraph 1.6 and Section 4) . The estimates of employment and income generation are wrong and seriously inflated (Section 2). I have therefore produced my own estimates based on evidence and assumptions cited in the text. The biggest assumption is that passenger and freight traffic will grow at the rates projected by Wiggins. Except for the last section, the whole of this comment is based on that assumption, which is challenged in the final paragraphs.

Three of the more important sources of error are that

1. they relate the rate of growth of employment simply to the rate of growth of passengers, and ignore the rate of growth of freight;
2. they make no allowance for changes in productivity (which OEF and DETR put at 3% per annum --- paragraphs 2.14-15), and
3. they ignore the fact that indirect jobs may be anywhere in the country.

The explanation of the revised estimates in Section 2 involves a lot of detail. Much of this can be skipped. The key sentences have been italicised.

Table L4 below is taken directly from the ADL report. It highlights Wiggins's own estimates of traffic. Comment appears in Section 1 below.

My revised estimates of employment of Thanet residents as given below, are based on detail in the main text, and on the assumption (made by Wiggins) that 10/17 of the workers will be Thanet residents.

It may be noticed that by 2010 the number of jobs created will be well in excess of the present level of unemployment. However, demographic factors will mean that there may still be a shortage of jobs, depending on other developments. When we look beyond 2015 this will be less true and we will be entering a period of labour shortage in Thanet which will add to the general overheating of the South East economy. We will also become increasingly dependent on one employer in a highly vulnerable sector, which does not bode well for a stable secure job structure.

The incomes received by these workers are estimated in Table L21 below.

____________________________________________________________

* These tables take no account of reductions in employment and incomes due to the adverse impacts of the airport, discussed in Section 4 of this report. They also depend on various assumptions that are detailed in the text. In reading Table L21 note carefully paragraph 3.10 below.

This comment ends with remarks on other economic impacts, mainly house prices, the impact on tourism, and a suggestion that quite apart from terrorism the trend in air travel is due for a correction due to a revolution in communications.

1.1 The impact of Manston airport on Thanet depends on the nature, volume and timing of air traffic. The broad levels of traffic indicated in the Masterplan are summarised in Table L1 below.

1.2 The Environmental Statement gives more detailed forecasts . Some of them are for 2002. There is no indication of how these have been made, and recent events have probably affected their reliability. We discuss this a little further in Section 5. There now seems to be little point in commenting on the 2002 forecasts. Accordingly I begin by assuming (for the time being) that the 2005 forecasts of flights and employment are correct, subject to modification as described below. ADL give details of traffic but no totals. The following totals are based on Table 4, which is attributed to Wiggins.

1.3 There is no indication of the basis of these forecasts. For the purpose of this Comment they have been accepted, but they are discussed further in Section 5. They have been used by Wiggins to produce estimates of air traffic movements (flights) and employment. In Table 5 they have been converted into forecasts of air traffic movements, on assumptions that involve growing aircraft sizes,. These assumptions are illustrated for selected types of trip in Table L3 below. It should be noted that Wiggins's assumption that the numbers of passengers to a plane will increase implies some increase in labour productivity. I return to this point.

1.4 The forecasts of air traffic movements have been converted by ADL into forecasts of daily and hourly flights, on the basis of 350 days of operation for scheduled flights, 210 days for chartered flights, and 365 days of operation for cargo flights. A 16 hour day is assumed in calculating the numbers of flights per hour. These forecasts, given in Table 6 and attributed to Wiggins, are shown below in Table L4.

1.5 Tables L1 - L3 can be used to get a first impression of the likely volume of activity, and so (with some modification) of the level of employment generated by the airport. Table L4 can be used to provide a first impression of the rates of growth of adverse impacts on the environment and the local economy due to noise and pollution. Between 2001 and 2010 we can expect these impacts to grow by a factor of over 30, and between 2001 and 2020 by a factor of over 100, subject only to the qualification that as planes are expected to become less noisy the noise impact will grow by somewhat less than this table implies.... but nobody has suggested that the noise from a plane will decrease by anything like the rate of growth of these numbers of flights. The figures come directly from Wiggins.

1.6 Unfortunately ADL have done virtually no analysis of the economic effects of noise and other pollution. An environmental study should have included, at the least, a count of numbers of dwellings within the various noise contours, a realistic estimate of the losses in property values, the numbers of these dwellings that take in all-year or holiday guests, an estimate of the impact on local traders of reduced spending due to reductions in holiday makers, foreign students at the language schools, and so on. It should also have looked at the costs of dealing with traffic, with pollution, of mitigation measures and of being subjected to pollution of various kinds. Instead, apart from a few tabulations that have negligible associated interpretation or other comment, it has concentrated on predicting the positive impacts of the airport on employment and incomes, for which it has made extravagant claims and has virtually ignored the negative impacts. There is a lot that it does not do. I argue below that much of what it does do is done wrongly.

2.1 The "Rule of Thumb"

Forecasts of employment and income for Manston are complicated by the fact that Wiggins and their advisors have presented a muddled picture. It begins with a statement of what has become called "the Rule of Thumb". In A Strategy for Success they refer to research that shows "a clear correlation between passenger throughput and on-site job creation, with typically 1000 new jobs being created on site for every additional one million passengers, and a similar number of new jobs for every 100,000 tonnes of cargo." I discuss the evidence for this "Rule of Thumb" in Appendix 1. Here I emphasise three points about it:

1. it is biased very heavily in favour of the two largest airports --- Heathrow and Gatwick.
2. it relates to total direct employment, which is defined below. Using it to estimate employment totals that are defined differently can lead to seriously misleading errors.
3.  as quoted, it seems to relate to 1998 and takes no account of recent or future changes in productivity.

2.2 I show in Appendix 1 that in any case the data on which this rule is based are flawed. Having stressed that, to avoid confusion and to concentrate on other objections to the ADL estimates I shall, for the moment, accept the Rule of Thumb as a rough guide under certain circumstances.

2.3 Acknowledging that technological change and economies of scale may reduce the ratios cited in the Rule of Thumb, Wiggins say that the passenger and cargo targets suggest that "on site employment at the airport could provide some 6000 job opportunities" by 2010. Note that applying the rule of thumb to the forecast activity for 2010 ,and assuming no change in productivity, produces 3,700 passenger associated jobs and 2,700 cargo associated jobs, totalling 6,400. But this rule of thumb relates to direct employment as defined by OEF (and quoted below) and Wiggins have used it differently. They have defined direct employment to exclude passenger handling services, shops and so on. If they adhere to that definition then the Rule of Thumb becomes inapplicable. In particular, the estimate of some 6000 jobs is unsupported

It is important to distinguish between three kinds of employment, to which the ADL report and the two main references (OEF and DETR) refer frequently :

2.5 Direct employment, which OEF define to be "employment that is wholly dependent on airport related activities, whether on-site at the airport or off-site" According to an OEF enquiry in 1998 about 60% of direct employees at UK airports work "for airlines and handling agents (e.g as flight crew, check-in staff, maintenance crews etc). Another 10% or so work directly for the airport operator (eg in airport management, maintenance, security, etc). Almost another 10% work in concessions (retail outlets, restaurants, etc), with 4% in freight/cargo businesses and 6% in control agencies (eg HM Customs and Excise, immigration). The remainder work in a variety of organisations, including on-site ancillary activities such as hotels." The Rule of Thumb relates to the totals of all these.

2.6 Indirect employment is defined as "employment in firms outside aviation generated because they supply goods and services to the aviation industry. Examples include jobs in the energy sector generated because of airline purchases of aircraft fuel, or in the aerospace sector by airline purchases of aircraft equipment; the employment in the IT sector providing computer systems for airport and airline operators; construction workers building additional facilities at airports; and the workers required to manufacture the goods sold in airport retail outlets." These employees may live and work anywhere in the country or even abroad (in the case, for example, of imported equipment).

2.7 OEF estimate that in 1998 there were 180,000 direct jobs in UK airports, and that these generated an additional 200,000 indirect jobs in the UK. This gives a direct/indirect multiplier of 1.11, which is used by ADL in its calculations. If, as is possible with the growth of the European economy, more of the goods and services used by aviation are imported, the value of this multiplier will fall.

2.8 Induced employment is employment generated by both direct and indirect employees through the spending of their "income to purchase goods and services for their own consumption." OEF point out that the relevant amount of income to be used in calculating induced employment is not the total income of these employees but the difference between that and the total benefits they would receive if unemployed. We refer to this later.

Table 96 of the Environmental Report gives Wiggins's predictions of staff employed at LMA. The table gives no totals but these are easily derived and are given below, along with each figure expressed as a multiple of the preceding figure.

2.10 It may be noticed that the rates of growth of employment indicated by the multiples in the last row of this table are very similar to the rates of growth of predicted passenger numbers indicated in column 3 of table L2. This observation prompts closer examination of Table 96. We find that the grades of staff fall into two categories, whose names and predicted numbers for the years 2005, 2010 and 2015 are given below. The predictions are taken from Table 96.

It is clear that while numbers of the "core staff" in the first five rows of this table have been determined by other considerations and grow very little, numbers of "other staff" in 2010 and 2015 have been increased simply by multiplying by the growth of predicted passenger numbers. The same is true of predictions of "other staff" for 2020, which all have multiples of 3.14 or 3.15 compared with a passenger multiple of 3.15.

2.11 But LMA is also a freight port. Taking 100,000 tonnes of cargo as having roughly the same employment implications as 1,000,000 passengers (which is suggested by the Rule of Thumb), we can translate passengers plus freight into units of traffic, as indicated below where we also show the quinquennial changes

2.12 The multiples in the last column, indicating the growth of total traffic, have to be compared with those in table L2. Assuming that the predictions for traffic and employment in 2005 given in this table are correct, and that it is correct to equate the employment demands of 100,000 tonnes of cargo to those of 1,000,000 passengers, then  

• the rate of growth of employment between 2005 and 2010 should be 1.82 rather than 2.52. 
• Between 2010 and 2015 it should be just marginally different (1.27 compared with 1.25) 
• and between 2015 and 2020 it should be 2.28 rather than 3.15.

2.13 In other words, if we equate employment growth to all-traffic growth, and ignore changes in productivity, then totals employed become roughly as follows

These predictions are based on the assumption that the 2005 predictions are correct, and that later employment is determined by the growth of overall traffic, including passengers and freight., rather than simply by the growth of passenger traffic ( which is what LMA assumed).

2.14 However a further adjustment is necessary, because the above predictions (and LMA's original predictions) make no allowance for the effects of increased productivity over the period up to 2020. (The only allowance made for changes in productivity is in the estimation of air transport movements, when note is taken of bigger planes). An estimate made by OEF on the basis of official statistics, and adopted by the DETR in The Future of Aviation, shows that the number of passengers handled per employee has been rising by around 3% per annum. They expect direct employment to increase from 180,000 in 1998 to 210,000 in 2015 --- an increase of 16.7%. But in the same period the number of passengers is expected to more than double. This implies a growth in productivity of over 70% in this period of 17 years, giving an average annual compound rate of just over 3% per annum.

2.15 Instead of using this rate of 3%, I shall assume that there may be problems in sustaining this over the next eighteen years and that productivity grows by only 2.5% per annum. Using this as a compound rate we derive a reduction factor for predicted staff of 0.8811 per quinquennium compounded. Applying this to the estimates in Table L8, we obtain revised predictions of LMA staff as

2.16 At this point we must move from an attempt to forecast employment by Wiggins to the more difficult task of predicting direct employment, defined as by the OEF and the DETR to include many items that are excluded from the above calculations. The ADL report makes no attempt to do this. Instead it estimates indirect and induced employment by applying to the original Wiggins employment figures some multipliers that should be applied to estimates of total direct employment (of which Wiggins employment is but part). This means that if the Wiggins figures are correct (and we argue above that they are not ), and if they do indeed exclude a large part of direct employment, then indirect and induced employment have been underestimated. ADL note this, but make no attempt to correct for it..

2.17 According to OEF, total employment by UK airport operators accounts on average for only 11% of total direct employment in aviation. On the basis of current trends in productivity it predicts that in the year 2015 total direct employment will be about 210,000, implying a total employment by airport operators of about 11% of this, which is 23,100.

2.18 My revised prediction of employment by LMA in the year 2015 is 2206 which amounts to 10.5% of the UK total. This seems to be rather high. It looks as if the LMA staff numbers may include some categories of workers who are excluded from the definition of airport operator employment used by OEF.

2.19 Some confirmation of this thought is provided by traffic forecasts for total UK airports. These suggest that in 2015 there will be about 325 million passengers and 6.8 million tonnes of freight. The traffic index combining these (with freight weighted by a factor of 10) comes to 393, compared with an index value of 8.12 for LMA (see table L7). This means that in 2015 predicted traffic at LMA will be 2.07 % of the national total.

2.20 We can use these comparisons to make a necessarily crude estimate of direct employment, as defined by OEF and to which the multipliers can be properly applied. We could assume that in 2015 the percentage of national direct employment to be found at LMA is also 2.07%. But that would ignore the point that the national average is very much influenced by the larger airports for whom the ratio of employees to passengers is higher than it is for small airports. I have therefore assumed that for Manston the percentage of national direct employment in 2015 is lower than the percentage of national traffic. More precisely, I have assumed that it is 1.70%, which gives Manston direct employment in 2015 as 1.70% of 210,000 which is 3570. This may be compared with our final estimate of LMA employment of 2206. Thus in 2015 Direct Employment is 1.62 times the estimated LMA employment. (This value of 1.62 obviously depends on the assumptions I have made. I can say only that I have made them after careful thought and that estimating direct employment by multiplying LMA employment by 1.62 seems preferable to doing as ADL have done and treating LMA employment as though it is the whole of Direct Employment. Taken as a whole, the effect of making this [or a similar] assumption is to increase the importance of LMA as a creator of jobs.).

2.21 It would be wrong to assume that this ratio applies to all airports in all years. The national figures are very much weighted by experience at Heathrow and Gatwick. Much of the direct employment, such as employment by airlines and concessions, grows in steps. For example, only when an airport reaches a certain critical size will national chains open cafetarias; and only when a substantially higher critical size is reached will there be waiter service restaurants. Accordingly I assume that the ratio of 1.62 is reached in stages, being 1.2 in 2005 and 1.41 in 2010, 1.62 in 2015 and 1.83 in 2020. This produces our estimates of direct employment as

2.22 Thus using these assumptions and the original Wiggins LMA staff figures instead of estimates of direct employment results in an underestimation for early years but an overestimation for later years. Part of the difference arises from Wiggins failure to take account of productivity changes.

2.25 These proportions have been used by ADL, and I shall do the same. If we apply them to Table L12 we have the following estimates of Thanet residents in direct employment.

2.26 Estimates of Indirect and Induced Employment

Table 100 of the ADL report contains estimates of indirect and induced employment. They have been derived by applying multipliers to the LMA staff employment figures. These multipliers have been calculated by OEF, using the definitions given above, and should therefore be applied to figures of direct employment, rather than of LMA staff..

2.27 It is clear from the definition that indirect employment may be located anywhere. For example, aircraft equipment and goods sold in shops at the airport are unlikely to have been made in Thanet. There will be some local jobs arising out of the supply of aircraft fuel, but many of the construction workers will either live in the one of the other districts or, as is the custom for construction workers, have homes well away from their place of work and do most of their spending out of Thanet. That is why Table 100 describes indirect jobs as "location non specific". It is also why it is quite wrong to include total "indirect jobs" in an evaluation of the impact on Thanet. Yet on page 134 we read that the airport "will lead to greater economic growth in Thanet and it has been estimated could lead to a further 6,500 indirect and induced jobs in 2010..... in addition these benefits will lead to income injections both within Thanet and the local area." The fact that almost two-thirds of the calculated 6,500 further jobs are indirect, or induced from indirect, and could be anywhere, is ignored. We are presented with a highly misleading statement, which is, in any case, based on incorrect calculations. See paragraph 2.34 below.

2.28 Table 100 has seven rows. The first, labelled "Indirect jobs -- location non specific" has been obtained by multiplying the LMA staff employed figures of Table 96 by the OEF indirect jobs multiplier of 1.11. I have recalculated this row, using our estimated Direct Employment (Table L10) instead of the LMA figure (Table L5). The result appears below.

2.29 After considering the definition of indirect jobs, and the locations of suppliers of goods and services to the airport and its users, I am assuming that one fifth of the indirect jobs are taken up by residents of Thanet. Some will be hundreds of miles away.

2.30 Induced employment generated by spending by direct employees is calculated by multiplying the numbers of direct employees (instead of the LMA employment figures) by a multiplier of 0.25. I assume that spending by direct and indirect employees is mainly in their places of residence, which means that 10/17 of this induced employment is in Thanet. (This is also implied by the ADL estimates). Accordingly we get the estimates of induced employment in Thanet shown below.

2.31 This replaces the second row of Table 100. I do not think that we should be concerned about the impact on employment outside Thanet, and so I have not calculated new versions of the next three rows., but they could easily be calculated using Table L11.

2.32 Row 6 in Table 100 is also misleading. We have seen that indirect employment may be anywhere in the country. The location of employment induced by the spending of these employees is equally vague. Nevertheless we need to make some allowance for it. I have assumed that the indirect jobs located in Thanet generate some induced jobs given by operating on Table L13 above with the OEF multiplier of 0.25. This produces

2.33 Row 7 in Table 100, labelled "Total induced jobs non specific", is in fact, the total of rows 2 - 6. It thus adds total induced jobs anywhere, due to direct or indirect employment anywhere. It has no place in a serious analysis of the impact on Thanet. We replace it with another row in the summary below.

2.34 This shows that in paragraph 5.8.7. (reproduced in paragraph 2.27 above) the figure of 6,500 indirect and induced jobs should be replaced by 1186.

2.35 Adding Tables L15 and L19 gives the following estimate of total direct, indirect and induced employment of Thanet residents as a result of LMA. It has to be corrected by the subtraction of jobs lost because of the adverse economic impact of noise and other pollution.

We have to remember one important thing about the Thanet residents who will be in these jobs. Some of them will have moved into Thanet because of their new jobs. There will also be people migrating from afar into the surrounding areas. Some of these migrant workers will have partners who will be seeking work and may well take jobs at the airport or elsewhere in Thanet. We do not know how many of these there will be, but every migrant who takes one of these jobs will reduce the number of new jobs available to the unemployed of Thanet; and if the migrant has a successful job-seeking partner there will be a further reduction.

Table 101 and the comment on it in paragraph 5.8.7 reveal a surprising confusion about economic theory and social accounting. This necessitates a brief digression.

3.2 If the residents of an island have no trade or other financial links with outside, and spend their money buying and selling goods and services to each other, so that one resident's income is another resident's expenditure, the total of payments made (and received) in a year is known as the gross domestic product. The passage of money between these people, around and around, is described as the circular flow of income.

3.3 If somebody (or a government) not living on the island pays money to some unemployed islanders, perhaps for some good or service that is exported, or as a social benefit, the islanders are likely to spend some of this money buying from their fellow islanders. They in turn will spend some of their increased income buying from other islanders; and they from others, and so on. All of these increases in income will add up to more than the original payment, and so the gross domestic product will increase by more than the original payment. The ratio of this increase to the size of the original payment is known as the income multiplier. If there is enough unemployment on the island this extra spending will generate further (induced) jobs and eventually employment will rise by more than the number of people given jobs by the original payment. This eventual increase in employment divided by the original increase is known as the employment multiplier.

3.4 The original payment by somebody outside the island (and so outside its circular flow of income) to people living on the island increases the quantity of money that can circulate. The increase will be less than the original payment because there will be associated reductions in unemployment and other state benefits. The net gain in income represented by the original payment minus benefit reductions is known as the net income injection. It is responsible for all the other increases in income which arise through the multiplier process described above. It is completely wrong to include in an estimation of an income injection any induced income effects (which arise from it).

3.5 Yet this is what ADL have done in Table 101. The payments made by Wiggins to workers at LMA are given in Table 97. This also presents information about payments of tax and insurance, but only the top row, described as "total salaries" represents money paid into the pockets of the LMA workers. It is this, plus payments made to the comparatively few residents of Thanet who will be indirectly employed, that at first sight constitutes the income injection. But the additional employment created directly and indirectly by Wiggins means that less money is being received in unemployment and other state benefits. The income injection is thus reduced by the amount of these benefits. (One way of taking account of the reductions in benefit is to incorporate them in the calculation of the multiplier, as OEF has done. This artificially increases the size of the injection but compensates by submitting it to a lower multiplier.)

3.6 But ADL has produced a table headed "Income injection from spend by local and commuting staff", which completely ignores all this. For example, the first (top left) entry in it is labelled "Income injection by local staff" for 2002. This has been obtained by starting with 10/17 of the total LMA salaries for that year (amounting to £6,630,110 x 10/17 = £3,900,064). This is part of the original payment and so is part of the injection. But ADL has multiplied this by 1.25 (to obtain the first entry of £4,875,081) and called THIS an injection. It is not. It includes the multiplier income generated by the original payment.

3.7 There are also other errors in this table, which we attempt to remedy below. In particular we completely reject, as meaningless in any analysis of an impact on Thanet, the row labelled "Induced investment from spend". Spending of direct and indirect incomes will eventually generate investment in the plant and machinery needed to make the goods that are bought, but this investment may be anywhere. (How much of the produce bought in Thanet shops is made locally?)

3.8 The total salaries presented by Wiggins in Table 97 are net payments to the workers mentioned in Table 96. In 2002 the structure of employment is necessarily top-heavy and the average net payment per worker is £19,500. By 2005 workforce expansion, especially at the lower end, has reduced this to £16,980. In successive fifth years it falls to £16360, £16350 and £16,109. These averages have to be taken in real terms. Quite properly, they make no allowance for inflation or pay rises, however these may arise.

3.9 To provide acceptable versions of Tables 100 and 101 we need to calculate the money paid into the pockets of all direct employees at the airport. Our estimates of the numbers of employees, after allowing for all air traffic growth and for productivity changes, are given in Table L10 above. If we assume that the average of payments made to all direct employees is the same as that made to Wiggins employees, we obtain the following estimates for the totals, irrespective of location, and for Thanet

3.10 We also have to take account of those indirect employees who live in Thanet, and of induced employees, as indicated in Table L21. If we assume that they, too, have the same average wages as employees of LMA we have the following estimates of total payments for all workers

The payments totalled in this table constitute the Income Injection, which triggers induced employment as described above. Operating against it is the cessation of unemployment benefits (and other payments from outside) for all these newly employed persons. This item can be estimated by multiplying the total employment figures shown in Table L21 by the average received in benefits annually by the registered unemployed of Thanet, which is a figure that Thanet Council should be able to acquire. It should include all benefits that will be lost when an unemployed person takes a job, other than those paid for local taxes without reimbursement from central government or other sources outside Thanet.

3.11 Other items that reduce the economic benefits are discussed in Section 4 below.

The ADL report says very little about adverse economic impacts. Some of the more important of these are discussed briefly below. A convincing quantitative analysis would require greater data-gathering resources than a private individual such as myself is likely to command, which makes it all the more regrettable that the ADL report does not provide more information. There are several pages of readily accessible information about tourism, house prices, and other matters but there is little attempt to provide anything new, or to use the information it presents in an economic analysis of the impact of noise.

4.2 It is, for example, surprising that the ADL report does not contain statistics of the numbers of properties within different noise contours, and other information that would facilitate a more thorough analysis. The contour maps that I have been able to download from the Wiggins website do not enable me to do this for myself. The report is also defective in not giving detail about how the noise contours have been calculated, and the precise information on which they are based. Since the assessment of environmental impact and such matters as double glazing requirements depends critically on them it should be possible for somebody like myself to check the correctness of the contours. This is made especially important by the use of noise contours to assess entitlements to compensation or double glazing.

There is abundant evidence that house prices in the vicinity of airports are lower than they would have been but for the noise of the aircraft. The reduction in price obviously depends not only on the the volume and timing of the noise, and the numbers of flights, but also on many other factors; and it is difficult to make reliable estimates for one airport on the basis of enquiries conducted at other airports.

4.4 ADL cite work by Tinch and Schipper which suggests that a rise of 1dBA in noise is likely to reduce house prices by 0.5 - 1%. Remarking that "the noise contours published for this assessment identify that noise levels between 2001 and 2005 will improve and not get worse" they conclude that they anticipate no "negative impact on house prices" during that period. "More detailed assessment of changes in noise levels beyond 2005 will provide an indication of whether houseprices beyond 2005 will be affected by the airport."

4.5 It is true that the published noise contour maps show an improvement between 2001 and 2005. This is difficult to believe. According to tables in the ADL report, Wiggins plan to increase the number of flights per day from 3 in 2001 to 50 in 2005. It is asserted (see, for example, paragraph 5.6.19 of the ADL report) that during this period planes will become quieter, but it is hard to see how 50 "quieter" planes will create less noise nuisance than 3 present-day planes. If a model suggests that they will then either there is something wrong with the model or the measure of noise nuisance is faulty. That is one reason for saying that we need to know exactly how, and on what exact basis, these contours have been calculated. It is interesting that no contours have been published for 2010, to say nothing of later years. Apparently ADL was advised that it "would not be worthwhile" to do this because of anticipated technological change.

4.6 In the case of LMA some house sellers living under the flight path report that prospective purchasers have been more concerned about the prospect of future noise than about current noise levels, and with this mind have offered as much as £12,000 less than they would otherwise have been willing to pay. In assessing the plausibility of this report, we may note that the Tinch and Schipper result quoted above, shows that an increase in noise level of 10dBA could reduce the price of a house by between 5% and 10% --- so a house priced at £120,000 might suffer a fall of between £6000 and £12,000.)

4.7 If a house owner who is contemplating sale finds that his realistic asking price is now about £12,000 less than it would otherwise have been then it means that he has that amount of money less to buy a house elsewhere. It is little different from having his savings reduced by that amount. Even if he decides not to move, because of this, he is still £12,000 poorer. All existing owners of houses under the flight path will be affected to some extent in this way, even if they do not sell for several years.

4.8 The fall in price will depend upon the level of noise. If we had information about the numbers of houses within the different (and credible) contours then it would be possible to make some estimates of the total amount of money lost to house-owners. My guess is that it would be a few million pounds ---- less than the pay received by LMA workers but still substantial both in total and as an ensemble of losses imposed on individuals.

4.9 Language Schools and Tourism

Many of the houses under the flight path take in holiday visitors or foreign students of English. The impact of noise on these visitors will depend on the timing of the noise. There seems to be something of a smoke screen over this, which two sentences may help to dispel:

1. The Section 106 agreement about night noise limits not night flights but regular night flights.
2. A statement that there are no plans to have night flights is not saying that there are plans to have no night flights; it says nothing about what may be planned tomorrow.

4.10 Paragraph 4.8.6 and Table 54 of the ADL report indicate that between January 2001 and July 2001 LMA received six complaints about night noise and 83 about day-time noise, with all the night complaints and 72 of the day time complaints coming from residents of Windermere Avenue, which is one of several streets on the Nethercourt estate that must have almost identical noise experiences. One of the procedures for complaining is to telephone and leave a message on the answering machine asking for a complaint form. My own experience is that the machine is not always working and the form does not always arrive.

4.11 Even if there are no night flights, residents and visitors will be disturbed by noise. In terms of flight regulations, there is a strong tendency for "night" to begin rather late and to end rather early.

4.12 Visitors have less time than residents to become to some extent adjusted to noise. I have argued in my Comment on the Thanet Local Plan that noise and especially nocturnal noise will have a severe impact on visitors to Thanet, especially Ramsgate. There is no need to repeat here the detail of the argument. Briefly, not only will visitors be less likely to come again, but they will spread the word and discourage their friends from holidaying in Ramsgate or using language schools there. Noise over the beaches will also deter some visitors. All in all, it will mean reduced incomes for landladies, declines of sales by shops and places of entertainment, and reduced viability of some of the language schools. Tourism and language schools collectively play an important role in the local economy, but once these sectors are seen to be declining they will become increasingly less popular. One consequence of the decline in staying visitors or students will be a fall in the demand for, and prices of, both registered boarding houses and the many large houses used unofficially to accommodate visitors.

The ADL report says very little about the costs of preventing, reducing or coping with pollution. We also need to know more about the impact on transport costs, road programmes and local authority finance.

5. THE FUNDAMENTAL ASSUMPTION

5.1 While estimates of employment, incomes and noise levels, and the various effects of these, depend on a variety of assumptions, such as those spelt out in this comment, the fundamental assumption on which the whole analysis rests is that passenger numbers and freight are going to grow in the way that has been "predicted" by Wiggins and summarised in Table L2 at the front of this comment. Although the totals have been broken down by type of flight there is no indication of how these components have been calculated. The figures seem to be targets rather than reasoned predictions, and there seems to be no way of assessing the likelihood of their being achieved.

5.2 This likelihood may, however, be affected by a social trend that has not been given much attention. Even before September 11th, airlines were beginning to experience unexpected declines in their fortunes. While this is commonly blamed on the slowing down of world economies, there is also a more fundamental reason which sets a trend that is more enduring than fluctuations in economic growth.

5.3 Briefly, it is that the revolution in communications brought about by the internet, satellite communications and text messaging is reducing the need to travel. The greatest effect will be on business travel. Exchanging information and views is becoming increasingly easier and sophisticated. Travelling by plane, and being out of the office for several hours, is becoming an increasingly inefficient option. Improved telephone and other communications already mean that more and more office and other workers now do their work from home. But the technological revolution is gathering pace. As business executives become more aware of the reality of terrorist threats, and of the airport delays arising out of security checks, they will reassess their options and in some cases discover that flying is not the best solution. The rising trend will falter.

5.4 The consequent reduction in demand for business flights, and the expansion of capacity at Heathrow, will result in less pressure on leisure flights from existing London airports, and LMA will find it more difficult than it has predicted to attract new business.

THE RULE OF THUMB.

A.1 In The Masterplan it is stated that "research has shown there to be a clear correlation between passenger throughput and on site job creation, with typically 1,000 new jobs for every million passengers and a similar number of jobs for every 100,000 tonnes of cargo." This relationship is often called the Rule of Thumb.

A.2 I have not been able to trace the origin of this Rule of Thumb, but the OEF report sheds light on the correlation. The OEF report was not written in any attempt to establish relationships such as the Rule of Thumb. The following criticisms of the its use as a possible basis for the rule of thumb should not be taken as necessarily criticising its usefulness as a study of the contribution of the whole industry to the national economy, which was its purpose.

A.3 The OEF report says that in recent years there have been studies at "nearly all of the major airports ... to assess how much employment is associated with their activities." Detailed questionnaires have been sent to "companies based at airports and in the local vicinity asking how many of their staff are employed on tasks related to the airport".

A.4 The results of these studies have been used to compile a table showing the numbers of terminal passengers in 1998 and an estimate for the same year of direct employment (as defined in paragraph 2.5 of this comment) for fourteen airports. OEF explains that these studies were conducted at different airports at different times, and so in some cases the findings had to be updated to provide estimates for 1998. This was done by assuming that productivity growth at any individual airport had been "in line with the national average for the industry from the date of the original survey to 1998". This means that for an unstated number of airports the estimated direct employment in 1998 has been obtained by using information about other airports. Unfortunately this means that the employment estimates are not independent of each other, and so the usual calculations of regression and correlation coefficients are not very meaningful (in that standard errors cannot be properly calculated).

A.5 It also means that the employment estimates for those airports for which the surveys were undertaken before 1998 have been squeezed towards the "national average" which is dominated by a few large airports.

A.6 Another point should be noted. OEF reports that "different studies have adopted different conventions for adjusting the number of part-time employees to a full-time equivalent basis". To put the studies on a common basis they assumed that two part-time posts were equivalent to one-full-time post. Taken along with the italicised wording in paragraph A.3 above, this raises questions. In particular 

1. if a full-time employee works only part-time on "tasks related to the airport" does he count as full-time or part-time? 
2. if a part-time employee spends only part (perhaps a very small part) of his time working on such tasks does he still count as half a full- time worker? 

On the face of it, employment seems likely to be over-estimated.

A.6 The OEF estimates of direct employment and the numbers of passengers in 1998 have been tabulated in their report. The numbers of employees per million passengers range from 527 (for Edinburgh) to 1888 (for East Midlands). A rule of thumb with that degree of discrepancy is not very reliable. Not surprisingly, if the data are represented on a scatter diagram they show a very loose relationship, dominated by Heathrow and Gatwick. If we use the device mentioned in paragraph 2.11 of this comment and calculate values of a traffic index (defined as millions of passengers plus 10 x thousands of tonnes of freight), the scatter diagram of employment against traffic improves considerably, but the problems caused by the interdependence of the data, and their artificial squeezing towards an "average" dominated by a few large values, remain.

A. 7 The productivity ratios (millions of passengers to thousands of employees) for Heathrow and Gatwick in 1998 were 1.17 and 1.15 respectively, while the national average (based on the procedure outlined above) was 1.13. According to the OEF report Stanstead had 6.83 million passengers and 7,990 employees in 1998, giving a ratio of 1.17. However an Issues Brief published by the BAA between October 2000 and May 2001 says that there were "currently ... around 12 million passengers" and 9500 employees (giving a ratio of about 0.79), and that if the airport were to expand to 25 million passengers the employment would increase to around 16,000 (giving a ratio of about 0.64).

A.8 There are several possible explanations of the apparent differences between the OEF and BAA figures. However the important point for now is the BAA ratio falls 0.79 to 0.64. This emphasises a point made throughout this comment that productivity is changing rapidly. It follows that even if the Rule of Thumb of 1000 employees per million passengers was correct in 1998 it would be provide increasingly large over-estimates. Even if we assume that productivity at LMA grows in future years at only 2.5% per annum (compared with the rate of 3.0% used by OEF and DETR) the employment estimate per million passengers would fall from 1000 in 1998 to 927 in 2001, and to 838 in 2005. By 2010 it would be down to 738, and by 2020 to 573.

A.9 Thus the Rule of Thumb 

1. is based on not very reliable, and biased, data; 
2. is subject to a wide range of error; and 
3. takes no account of changes in productivity, 

so that the more we look into the future the less useful it becomes.