by Russell Schiller and Jonathan Reynolds
In 1985 we produced a measurement of British Shopping Centres (both town centres and purpose-built shopping centres) and have now updated our exercise.
The aim is to produce a simple method of grading shopping centres which can be applied uniformly across the country: giving each centre a score according to its number of durable multiple branches is an easy way of achieving this end. National durable multiples are unusually strong in British retailing, taking more than two-thirds of clothing sales. The presence of key multiple branches, therefore, is a reasonable indicator of shopping potential.
Writing in Estates Gazette in March 1985 we pointed out that one of the advantages of the new method was that it would be relatively easy to update. The first exercise, using 1984 data, was a collaborative one with the University of Newcastle, supported by the RICS and the ESRC. The new, updated exercise has been completed in co-operation with the Oxford Institute of Retail Management and Chas E Goad Ltd (Goad Plans), using data for 1989, five years after the previous exercise.
Goad plans cover 1,079 centres, of which 826 were found to contain at least two of the key multiple fascias. Two multiples — typically Woolworths and Boots — were considered to be the minimum qualification for significant durable or comparison shopping.
Fascias
A list of 113 retail fascias was selected as suitable for scoring each shopping area. Multiples were selected from trades which tend to cluster in centres, such as fashion and jewellery. Care was taken to avoid regional bias by examining the regional distribution of chains before they were selected. National multiples tend to be stronger in the South East: it was necessary to allow for this.
It was not possible to use the same lists as in 1985 because of takeovers and new arrivals, such as Next. The list includes all significant department and variety stores plus a selection of fashion and other durable multiples providing a range from a minimum of 10 up to several hundred branches. The end result was to produce a matrix of 113 fascias against 1,079 Goad plan areas. The OXIRM computer did the rest.
Retail sales
Before looking at the results, a word needs to be said as to what the multiple branch score indicates. Shopping centres can be measured in several ways, such as sales, number of shops and quantity of floorspace, as well as multiple branch numbers. There is a good correlation between these four indicators, but they all measure different attributes. Declining centres might continue to maintain a high level of floorspace and shop numbers even though sales had fallen and multiples had moved out, to be replaced by independent traders requiring a lower level of sales.
Retail sales measure patronage by shoppers, but this can be affected by one or two large stores: the influence of Harrods on Knightsbridge is the most extreme example of this. Multiple shop branch numbers lie between sales and physical size as a measurement factor, relying on the trading experience of national retailers and their perception of the quality and potential of a centre. In this respect they are a sensitive indicator of change in a centre’s fortunes.
Hierarchy
The results range from the West End of London, with a score of 186, down to 133 small towns with a score of just two, such as Musselburgh near Edinburgh or Potters Bar in Hertfordshire. We have put the centres into rank order so that they can be compared with their rankings in 1984. We have also grouped them into a simple hierarchy.
London clearly stands out from any other centre. Below London there is a group of large provincial centres which we have classified as “metropolitan”. In the ranking there is a natural break between Nottingham with a score of 93, which is seventh, and Cardiff, which is eighth and scores only 74. We have therefore limited the metropolitan grade to only six centres, less than in 1984. The movement of the metropolitan centres since 1961 is shown in the chart. The most significant change is the recovery of Manchester to fourth place, the position which it occupied in 1961. Liverpool’s spectacular decline is shown to continue: it has dropped from third place in 1971 to 12th in 1989.
Below the metropolitan centres comes a group of 159 regional centres. These offer a substantial range of comparison shopping, and typically contain at least one department store and several variety stores. The 660 district centres offer a much reduced range of durable shopping, with perhaps one variety store. They cater for the regular weekly shopping trip which combines both food and non-food items.
It is interesting to note from the table that there has been a reduction in the number of centres in each of the three grades. In the 1984 analysis Boots and Woolworths each had around a thousand branches, more than twice the number of any other multiple on the list. Both have tended to close branches in smaller centres, and Woolworths in particular have reduced their branch numbers from some 900 in 1985 to 669 in 1989.
The reduction in the number of regional centres results from a more rigorous analysis of department and variety stores. The divide point between regional and district has been taken where the average number of stores falls below three per centre. Overall, the reduction in numbers might suggest a growing separation between a clearly defined group of modern durable centres served by national multiples and a remainder of local and smaller-scale shopping.
Trends
One of the trends which emerged most strongly in 1984 was that many small attractive towns were ranked much higher than they had been in terms of sales in 1971 or 1961. Thus in 1984 Chester, Bath and Exeter ranked between 14th and 23rd, whereas in terms of 1971 sales they ranked between 34th and 54th. The results for 1989 show that this trend has intensified. Chester has risen from 14th to ninth and has, in fact, overtaken Liverpool, which has dropped from 10th to 12th. Bath, having been well behind Bristol Broadmead in 1985, has now climbed to the point where it is equal 17th with its larger neighbour. Similarly, in Scotland, Perth has now risen to match Dundee, a city roughly four times its population. Other examples are Canterbury (53rd) which has overtaken Sunderland (90th) and Truro, which ranks above Rochdale.
A similar pattern is found in the level of shop rents where, during the 1980s, small, attractive towns have in many cases overtaken northern cities with many times their populations. The evidence suggests that population is becoming a less reliable guide to the retail potential of a town. As the public becomes more mobile and has more leisure, retail spending seems to become closely linked to tourist patterns.
Development
When comparing the 1989 rankings against those for 1984 there are a number of cases where dramatic changes have occurred because of shopping centre openings. In 1984 Southend ranked only 55th, but by 1989 it was 19th, a rise which coincides with the opening of The Royals. Similarly, the opening of the Grafton Centre in Cambridge is the obvious explanation for a rise from 21st to 12th by that city. In both cases multiples have opened second branches. In the opposite direction, Dudley seems to have suffered badly from the development of Merryhill, dropping from 73rd to 139th.
The table shows 20 major regional centres that have experienced substantial rises in rank which are associated with new development. The second half of the 1980s has seen a major wave of town-centre development which has varied enormously in its impact on individual towns. In 1984 there was a group of towns which appeared to have fallen for no other reason than that they had lacked development. Three of these were identified in our article in 1985 — Carlisle, Leamington Spa and St Albans. All three now have major schemes and, as the table shows, they have risen greatly in rank.
As well as towns which have done well because of development, there is another group which has done badly while awaiting the results of development. Bromley, Aberdeen and Watford were not able to benefit from their major new schemes at the time of the multiple count in 1989 and show a fall in rank. From the evidence of other centres, this drop should be temporary.
Losers
In 1984 we identified a group of towns which had steadily lost rank owing to urban decline (inner-city problems, population and economic decline). These fell broadly into two groups, inner London suburban centres and northern industrial towns. For most of these the sad decline has continued in 1989. Towns to show consistent decline from 1961 to 1989 are:
Midlands and North:
Leicester, Hull, Bradford, Blackpool, Dundee, Stockton, Oldham, South Shields, Bishop Auckland, Keighley, and Ashton-under-Lyne.
Suburban centres:
Lewisham, Peckham, Sutton, Kilburn, Holloway, East Ham, Streatham and Tooting.
Included in this list are the two London centres where John Lewis have recently closed department stores (Holloway and Streatham). In addition to these groups, Luton and Gillingham also register persistent decline.
It is encouraging to report that not every member of the 1984 long-term decline list continued to fall in 1989. Derby, Newport, Halifax, Rotherham and St Helens managed small rises. In London, Woolwich and Brixton also managed to halt their decline.
Winners
Despite the overall reduction in the number of regional centres — from 188 in 1984 to 159 in 1989 — there are four new centres claiming regional rank. They are: Rhyl, Durham, Walthamstow and Metro Centre.
Since 1989 a number of out-of-town centres have been completed which would have an effect on the rankings if they were included. Three are of regional status: Merry Hill (40), Meadowhall (39) and Lakeside (36).
Among new district centres are Surrey Quays (18), which is included in the ranking, and three smaller district centres: Crystal Peaks, Sheffield (7), One Stop, Birmingham (5), and Foss Park, Leicester (4), which have opened since 1989.
Conclusion
The exercise described gives a consistent measurement of durable shopping centre size. A number of multiple retailers have found that it gives a good proxy for sales potential in individual towns. It also offers a simple and convenient method for comparing any centre with its neighbours. The preliminary analysis of the results given here shows that, as a measure, it reflects deep changes in the retail scene which are taking place in Britain today. We hope to be able to update the exercise with greater frequency in future to monitor the effects of shopping centre development more closely.(1)
(1) A wall map showing the location of the shopping area and a booklet listing details are available from Hillier Parker GOAD Plans and Oxford Institute of Retail Management. Price £50.
