Silloth (Travel-to-work Area) (Hansard, 25 July 1985)

§ Mr. David Maclean (Penrith and The Border)

I am grateful to you, Mr. Speaker, for scheduling the Adjournment debates today to slot in an extra one, which has given me the opportunity to raise this matter. I am grateful for the help of your office, and I am grateful to my hon. Friend the Under-Secretary of State for Employment for cutting short an engagement to reply to the debate.

After a debate on space exploration, we must come down to the problems of Silloth on Solway, in my constituency. However, this is an important subject to me and to many of my constituents. In 1983 and 1984, the Department of Employment carried out a review of travel-to-work areas. The review used data about travel-to-work patterns from the 1981 census of population, and the results were published on 16 July 1984. The new travel-to-work areas and the wards in them make up the building blocks for the revised map of assisted areas published by the Department of Trade and Industry.

In making the calculations, the Department of Employment has made a fundamental error in assigning Silloth on Solway to the Carlisle travel-to-work area instead of the Workington travel-to-work area. To prove that point, I hope to show my hon. Friend that the new travel-to-work formula, based on a random 10 per cent. sample of the population, can go dangerously astray when dealing with such a small sample. In addition, I shall show from the evidence collected from main employers in the area that the majority of the work force in Silloth are Silloth-based and Workington-based rather than Carlisle-based.

It is necessary to understand how travel-to-work areas are defined and how individual electoral wards, which may be 50 or 100 miles apart, are allocated to them. The Department admits that there is no one theoretically correct algorithm for grouping areas in this way. Nevertheless, it has adopted a formula which is described simply as the commuting links formula. Algebraically, that formula is described as

Fa.b/Ra. . Fa.b/Wb + Fb.a/Rb . Fb.a/Wa

Fa.b is the number of journeys to work from area A to area B, Ra is the number of workers who live in area A, and Wa is the number of people who work in area A.

That is sufficient to deter most people from querying whether the Department of Employment has it right. Fortunately, the Employment Gazette described how the formula and algorithm were used in practice. The algorithm used to arrive at travel-to-work areas consisted of five stages. Since my hon. Friend knows all those stages, I will not go through them, except to say that I believe that it was at the fifth stage that the Department made the fundamental error over Silloth on Solway.

If Silloth on Solway was allocated to a TTWA on its merits, it would certainly be associated with Workington on the west coast. However, wards are considered not on their own, but in relation to the commuting links with the next door ward. To the east of Silloth ward lies Waver ward, a large rural area stretching well over to the east near to the Wigton wards. Wigton, naturally and rightly, has close commuting links with Carlisle TTWA, and is 1409 assigned to that. Because most of the rural population, particularly in the eastern part of Waver ward, has, after Carlisle, the closest commuting links with Wigton, Waver ward has been assigned to Carlisle TTWA.

The Department has got Silloth wrong because the industrial estate in Silloth on Solway is technically a few hundred yards within the Waver ward boundary. Most people know it as Silloth airfield industrial estate, and know that it is an integral part of Silloth town, so much so that the vast majority of workers in the factories there come from Silloth. Nevertheless, it completely distorts the figures because technically all the workers from Silloth commute daily to Waver ward a few hundred yards away. That is because Waver ward has been dragged inexorably towards Wigton, which has links with Carlisle, so Silloth is also shown to have the closest commuting links with Waver ward, and is technically also allocated to Carlisle. A study of a map would demonstrate this much more clearly than my explanation.

We thus have the anomaly of a large rural ward such as Waver covering 40 or 50 square miles, not densely populated, but with a majority of people in the eastern part looking to Wigton and Carlisle for their natural commuting links, while tucked away in a small corner on the western extremity of the ward is the industrial estate of Silloth, which then gives a totally wrong loading to the commuting links formula.

As one of the factors governing the choice of algorithm was its ability to produce contiguous areas, the geographic position of Silloth, bounded by the sea on one side and on the other three sides by Waver ward, must mean that it is extremely unlikely that Silloth will be allocated to a different TTWA from the one to which Waver is allocated. This view is confirmed by the travel-to-work patterns.

On the basis of the figures given in answer to my parliamentary question of 15 April 1985, the approximate commuting link factor between Silloth and Waver, based on those in employment only, is 0.07, which is much stronger than any of the factors given in answer to my parliamentary question of 30 April when I asked: what was the value of the calculations of the measure commuting links formula for (a) Silloth ward and the wards 2, 6, 7, 8, 9, 10 and 13 of Carlisle … area".—[Official Report, 30 April 1985; Vol. 78, c. 106–7.] The answer was 0.00702. The factor between Silloth and Aspatria and Ellen wards was 0.00009, and the factor between Silloth and Castle, Dalton, Netherhall, Northside, St. Michael's, Stainburn and Westfield wards was 0.00010. I also asked about the factor between Waver ward and Wampool and Wigton wards, to which the answer was 0.00014. Between Waver ward and Aspatria and Ellen wards, it was 0.00006, and between Castle, Dalton, Netherhall, St. Michael's, Stainburn and Westfield wards, it was 0.0. If one studies this carefully, it can thus be seen that there is a very strong commuting link factor between Silloth and Waver, which everybody knew without the necessity for the formula which I have described. Thus any argument I make to my hon. Friend that Silloth has been wrongly allocated must also include consideration of Waver ward. The evidence from the census for allocating Silloth and Waver to Carlisle was very slight.

The crucial point is that the commuting factor between Silloth and Waver was 0.07. However, the commuting factor between Waver and Aspatria and Ellen was 0.00006. The commuting link between Waver and 1410 Wampool and Wigton was 0.00014. On those grounds alone, Waver was allocated to Wampool and Wigton, which thus had a stronger commuting link by a figure as splendid and as large as 0.00008. That is an incredibly small difference. I estimate that it could represent fewer than 10 journeys in each direction. On that basis alone, Waver was allocated to Wampool and Wigton. The travel-to-work patterns are based on a random 10 per cent. sample of census forms. Thus there could be a number of people commuting from Waver to Aspatria without that being reflected in the sample.

Let us suppose that the true number of journeys from Waver to Aspatria and Ellen was 10. There is roughly a 35 per cent. chance that none of the forms mentioning those journeys would appear in the sample. Similarly, if the actual number of journeys was 20 or 30, there would be probabilities of 12 per cent. and 4 per cent. respectively, but the sample would still suggest that there were no journeys.

If the number of journeys from Waver to Aspatria and Ellen had been 10 rather than zero, the commuting link factor would have been 0.00016—a stronger link than between Waver and Wampool and Wigton. Therefore, the allocation of Waver to the Wampool and Wigton Proto travel-to-work area may have depended on the random selection or otherwise of a single census form.

Because I believe that a random 10 per cent. sample is far too dangerously low a formula to use in a sparsely populated rural area, I conducted my own survey of the five major firms that account for most employment in the Silloth area. My sample shows a very different pattern of commuting from that shown in the 1981 census. However, in some respects our figures agree, which shows that my sample is based on valid criteria. The 1981 census identified 670 people working in Silloth; my sample of five firms four years later identified 700. The 1981 census identified 440 of those workers resident in Silloth. My sample identified 380. The remaining 60 will be made up of employees of smaller firms which I did not survey. Our figures are so close that no one can argue that my survey is inaccurate.

However, the figures show a different result for those commuting into Silloth from other areas. The random 10 per cent. sample of the 1981 census suggested that 110 of the Silloth workers commuted from the Carlisle TTWA and only 40 from Workington. My comprehensive survey, carried out in 1985, showed that only 84 commuted from Carlisle and 236 from Workington. Those are not minor statistical aberrations — they are glaring differences between the figures.

Let me read to the House extracts from letters I have received from the managing directors of the various firms concerned. One firm employing 55 workers states: Our staff locations done in a recent survey are as follows — Silloth 21, Maryport 11. Aspatria 11, Wigton 6, Abbeytown 6. 90 per cent. of our staff, therefore, come from the Maryport catchment area. That managing director goes on to make the point I have endeavoured to make. He says: Firstly, all the factories on this airfield of which we are one are not in the Silloth Parish area and are included in the Waver District. The Waver area is allied to Wigton which is allied to Carlisle. With our intimate knowledge of Silloth we can only assume that all people travelling, to Silloth Aerodrome and to British Sidac in Wigton are counted as travelling to Carlisle area. In our experience of employing people, our catchment area covers Maryport District in nearly every case. Applications for 1411 jobs come overwhelmingly from the Maryport area. We have had hundreds of applications and have never, ever had an application from Carlisle. All the other four firms make the same point. One states: Total number employed 149. From Silloth 93. From Carlisle 12. From Workington 44. I do not need an algebraic formula to determine where the strongest commuting links are in that case.

The fifth major employer in my survey gave me this information: Total number of staff employed 184. Number from Silloth 46. Number travelling from Carlisle TTWA 22 and number travelling from Workington 116. The managing director went on to state: As you can see from these figures, there is no doubt in my mind that in the case of our company we are providing a valuable service to the Workington TTWA, which has a high level of unemployment, especially in the non-skilled working area which we are mainly employing. We run transport daily, at our cost, from Workington via Maryport to the works and it is clear from the figures that we draw mainly on this area for our labour force. The 1981 census and my survey show that 60 per cent.of those working in Silloth reside in Silloth. A random 10 per cent. sample from the 1981 census shows that 16 per cent. of those working in Silloth travel from the Carlisle area, and on that basis—a basis of 16 per cent—the Department of Employment has allocated Silloth to the Carlisle TTWA.

That would be a poor basis on which to make an allocation even if the figure was correct, but I submit that the figure is grossly inaccurate. My comprehensive survey shows that only 12 per cent. of workers in Silloth have commuting links with Carlisle. For Workington, the random 10 per cent. sample shows that only 5 per cent. of the Silloth workers have commuting links with Workington.

However, the correct figure, as evident from my survey, is 34 per cent. There is no doubt that a serious error has been made, and I know that when my hon. Friend has had a chance to read my speech in Hansard tomorrow, and study the figures for himself, he will inevitably come to the same conclusion.

I do not expect the Minister to give me an instant answer to the problem I have presented him with tonight. He will have seen from my parliamentary questions that it has taken me eight months of painstaking research to discover the fault in the algorithm and the formula. When confronted with Fa.b/Ra. . Fa.b/Wb + Fb.a/Rb . Fb.a/Wa it is difficult for an ordinary hon. Member, without the facilities at the disposal of my hon. Friend, to know what is the correct answer. In this case, I did not even know the correct questions.

I have not mentioned the effect of this error. I have not dwelt on it because it has been my sole intention to prove that the Department has made a mistake in the formula calculations, rather than to recite the catalogue of disaster which will befall the Silloth area if this eror is not corrected. The Minister will know that I am referring to the loss of assisted area status, and I urge him to look at the map and see the isolated and difficult position in which Silloth finds itself. The managing director of one firm said to me: The consequent loss of regional development status is quite serious to us, given the fact that a lot of the investment in our 1412 company has been made possible by regional grants, and should these be lost, the long-term effects due to the high transportation costs of our product will be quite serious. Another firm said: We brought our business from London because of the regional assistance afforded in the area and the high quality of staff available. If the incentives are taken away, then why come? Another firm stated: Currently we have proposals pending, with our parent company overseas, for large capital expenditure. The receipt of approval for this expenditure. The receipt of approval from this expenditure and the securing of jobs that this will affect could be heavily dependent on the availability of grant facilities. Currenly, our overseas facilities, with their advantageous tax and grant situations, have received approval for their expenditure, which are already well advanced in action. The loss of our regional grant status could incur unfavourable comparisons and weaken the case for our 1985 capital programme, as well as in subsequent years. I know that that is not the province of my hon. Friend, but he knows that the Department of Trade and Industry map for regional assistance is based on the figures produced by his Department for TTWAs. In my constituency, the largest in England, with 1,250 square miles. I can find many areas that I would dearly like to see quality for assisted area status, but I am restricting my request only to the town of Silloth on Solway, and for one reason only, and that is because it has been erroneously removed from the Workington TTWA and allocated to Carlisle.

I know that my hon. Friend, with his natural sense of justice and fairness, will want to correct that mistake as soon as possible. I will gladly send him the file and all the papers, and I look forward to a favourable response, if not in his reply tonight, at least when he has had a chance to study the case personally.