Research Report 1/97 from The University of Karlskrona/Ronneby

How Planners and Small Business Entrepreneurs View Industrial Environments - the need and possibility for a planning dialogue
By
Anders Törnqvist* and Jukka Corander**
* Anders Törnqvist, Assoc. professor, Spatial planning, University of Karlskrona-Ronneby.

**Jukka Corander, MSc, Statistics, University of Stockholm.


Table of Contents:

Abstract

A picture sorting study with small groups of urban planners and small business entrepreneurs was carried out in order to explore the conceptualization of industrial environments. The participants were asked to sort and characterize 20 photographs of industrial buildings, selected to illustrate a broad range with regard to age, size, standard, exterior design and possible use. Multidimensional scaling analysis and other statistical techniques revealed significant differences between the two groups, where primarily the use of branch classifications discriminated between them. In contrast, consensus analysis of the sortings as such showed no significant differences, indicating the possibility of a dialogue. The results are interpreted and discussed in the context of communicative planning theory and potential conflicts in the industrial property market between owners and occupiers.














Introduction - the manager whose building had to be just right

A firm working with electrical and telecommunications installations made good money during the building boom in the late 1980s. The branch office in a middle-size town had offices, workshops and a warehouse at three different locations in town and wanted to invest in a new building to concentrate its activities. The branch office manager was concerned that the new building would look just right. It should not look too sumptious and extravagant because the customer would know he's the one to pay for it. But it should not look too plain either. The company was a solid, technologically advanced enterprise and the building should express this. The building would also be a long-term investment, providing space for future needs. The manager thought he got it right with a sober brick building in two yellow ochre tones and proper landscaping of the prominent site, visible from the main road. But when recession had set in in the beginning of the 1990s, soon after completion, the manager felt they had overdone it, after all. The building was now too big, and also, in the manager's view, looked too expensive and conspicuous.

This manager's concern that the design of a new building should express adequately the corporate identity is not unusual. What was surprising in this case was how narrow the interval was between the acceptable and the unacceptable. But interviews with other small business entrepreneurs, owning or renting far more plain and low-standard premises, confirmed the keen sense of appropriateness some entrepreneurs have concerning their buildings.

This surprised local planners, who thought that small business managers did not care much for the visual appearance of their premises. The observations have bearings on several planning issues such as renewal of old industrial districts, potential conflicts in the industrial property market between owners and occupiers, the increasing share of rented accommodation, the supply of accommodation for small enterprises. The need for studies developing the conceptualization of industrial property has been stressed (Ball & Pratt, 1994). This also seems relevant in view of the current interest in communicative planning theory (Forester, 1989, Healey, 1992, Sager, 1994). The emphasis has been switched from centralized planning based on extensive data and expert analysis to efforts to promote a dialogue between different interests and actors.

In order to explore the conceptualization of industrial environments among some of the actors involved, a picture sorting study was made, in which groups of planners and small business entrepreneurs sorted and characterized photographs of industrial buildings. The results of the study will be presented and discussed in the context of communicative planning theory and the regulation of the industrial property market.

Conflicts within the industrial property market

A number of writers have in various ways drawn attention to the conflict between owners and occupiers of industrial property. Fothergill et al (1987) investigated the quality of the British industrial building stock and argued that the majority is probably poorly suited for modern industrial use. The additions that can be made to the existing stock are necessarily small and the supply not always corresponding to demand. The proportion of owner-occupied buildings has declined in favour of rented premises, the design of which may not always be optimal for the individual firm. Institutional investors, however, focussing on return on investment rather than responding to the need of appropriate provision, prefer building standardized units in attractive locations (Ball & Pratt, 1994). Given the choice, for example, the private sector will develop high-tech/office space in preference to light industrial/warehouse space, according to Henneberry (1988)

Small firms have limited financial resources and are dependent on rented acommodation. This is not necessarily a disadvantage, since their needs change rapidly and they relocate frequently. But their information about the supply of suitable premises can be limited. Some landlords screen potential applicants, preferring financially stronger, long-term tenants. On the other hand small firms are seen as having low requirements of building quality and specific design and could make good use of the diversity of premises, potentially available in a large stock of older buildings (Fothergill et al, 1987, Green & Foley, 1986, Ball, 1994). In later years there have been many examples of successful conversions and rehabilitation of urban industrial buildings (Department of the Environment, 1987). But the entrepreneurial approach and the financing from several sources that characterize the low-cost examples still seem exceptions in urban areas, compared with institutional, large-scale rehabilitation projects.

The question, therefore, is how adequate in quantity and quality this potential supply really is. Even in economically prosperous areas like South East England there are high proportions of vacant space, some of it apparently in old, large buildings that seem difficult to subdivide and convert to premises for small firms and other users. (Fothergill et al, 1987, King & Co, 1986). Ball (1994) highlights the large volume of potentially useful vacant industrial space that is not put on the market for several reasons. Lack of motives and financial resources for necessary refurbishment could be one reason. There are many practical difficulties, both in launching schemes linked to small, individually insignificant buildings and in promoting large-scale, high-cost redevelopment projects. In economically expansive urban areas the tendency for more profitable uses to crowd out industry is often present, as pointed out by D. Pratt (1994) and Henneberry (1994). The shift from Fordist production relations to a post-industrial economy often has meant the transformation of old industrial zones into consumption zones (retail, housing and offices), while a profit-driven private sector and a poorly funded public sector have been neglecting the needs of the new network economy of small, growing manufacturing and service firms, making use of the dense, diverse and well-connected urban structures in larger cities.

The situation in Sweden

Industrialisation came late to Sweden and the industrial building stock as a whole is comparatively young with a large proportion located in rural areas, close to the supply of raw materials, forests and iron ore (Lindgren, 1991). Urban industry developed rapidly in the late 1890s. The successful examples of engineering firms that started as small, innovative enterprises and rapidly grew to big multinational concerns, such as Alfa Laval, ABB, Ericsson and SKF, may have biased planners in bigger cities and induced less tolerance of aging industrial districts with a diversity of small firms that remain small.

A study of property tax data (Mattson, 1992), for example, shows that advance factories for rent in Stockholm make up a smaller proportion of the industrial building stock compared to the rest of the country. The building standard of the Stockholm units is also higher than the national average. The findings reflect deindustrialization in the Stockholm area but also underline the relative importance of the supply of rentable premises for smaller firms that remains.

The need for some incubator environments in the urban structure is usually acknowledged by planners. But in their opinion, the large semi-central industrial districts, many situated along the waterfronts of the big cities Stockholm and Göteborg, can be put to better use, for modern office development and housing. See for example the current structure plan for Stockholm. Preliminary planning proposals for these areas tend to emphasize large-scale renewal rather than recognize the value of diversity of use, something which could safe-guard still viable small-firm environments in the area. The latest property boom culminating around 1990 and the subsequent crisis left a large surplus of modern office space in many locations, and investors seem hesitant to get involved in new large-scale projects. The preliminary proposals have also met with protests from local business organisations, fearing relocation of established firms and a reduced supply of inexpensive accommodation for small businesses. On the other hand there is an accumulating need to maintain and improve the existing stock in some way or other, or the supply of suitable premises will be reduced by deterioration anyway.

The study presented here was carried out among planners and entrepreneurs in Göteborg, the second biggest city in Sweden, and Trollhättan, a smaller, heavily industrialized town. Both have a history of large-scale industrial development, leaving a heritage of old industrial buildings and districts, now used by smaller firms or in some cases transformed in comprehensive renewals.

The need for a dialogue

The need for some kind of dialogue and cooperation between the various interests involved in using and renewing industrial environments seems obvious. Market forces may not be sufficient to meet the demand for a diversity of reasonably priced and well situated premises for small firms. Public funding of property development is hardly an option in these times with reduced levels of public resources. Exclusionary zoning in favour of industry is a heavy-handed measure with possibly adverse side-effects (Hunter & Heikkila, 1985). The public sector nevertheless remains a dominant player with a responsibility to stabilize land and property values and to give confidence to developers and investors, according to Healey (1992a).

It can argued that the type of environmental planning which will accomplish this is a communicative enterprise (Forester, 1989, Healey,1992b, Sager, 1994). Sager (1994:42) describes the paradigmatic core of communicative rationality as organizing "dialogue to promote democracy and personal growth and search for a solution agreed upon in undistorted communication". Planning should not be limited to rationalist decision-making, scientifically evaluating planning consequences and weighing the articulated goals and priorities of different interests, as in instrumental rationality. "Any action works not only as a means to an end but also as a promise, shaping expectations." (Forester, 1989: 141) Nor is it a mere power game. There is rather a need for efforts at "learning about the interests and perceptions of other participants and with that knowledge revising what each participant thinks about each other's and their own interests." (Healey,1992b:157) Planning, according to Healey, should not be restricted to a narrow aesthetic discourse about the relative merits of different urban forms, but this does not mean that the language of morality or aesthetics would be excluded from the dialogue. It may involve telling stories as well as doing analyses, because the differences involved concern systems of meaning as well as economical and social needs.

Conceptualizing industrial environments - a picture sorting study

Are premises for small firms mainly an economic issue, where the planners, sharing potential developers' concern for effective and profitable land use, conflict with entrepreneurs looking for acceptable accommodation at the lowest possible rent? Or is it an aesthetic issue, where planners condemn old and derelict industrial environments, which entrepreneurs, insensitive to aesthetic values, find quite adequate for their operations? Could there be different judgments of the functional suitability of small firm premises, entrepreneurs appreciating the greater variety offered by old industrial buildings, while planners believing the standardized contemporary sheds more flexible and adequate for modern business? What systems of meaning could be discerned among these groups concerning industrial property? In what ways do entrepreneurs and planners in fact see things, in this case, industrial environments, differently?

The method

The advantages of some type of sorting procedure for exploring concept formation and evaluation of environmental qualities have been outlined by Groat (1982) in a paper about meaning in post-modern architecture. In her study Groat (op.cit) showed 24 building photographs with examples of different architectural styles to two groups, 20 architects and 20 accountants. The results showed clear differences between the verbal constructs used by the two groups to describe the buildings. Compared with semantic differential techniques, the sorting procedure makes the use of prespecified concepts unnecessary, leaving the respondents' judgments uncontaminated by the investigator's preconceptions. Indeed, as Canter (1977) points out, studies of the former kind have for this reason generally failed to identify great differences between for example laymen and professionals. The sorting procedure is also less time-consuming compared with other techniques intended to identify concept formation processes, for example pair-wise similarity judgments or the repertory grid.

In the present study, with the purpose to explore differences between entrepreneurs and planners, the pictures were not chosen to illustrate particular types of industrial environments that the participants should identify. The focus was on identifying the verbal constructs used by the participants when sorting and describing a wide range of different industrial buildings and settings. To produce an assortment that could be described in many different ways, 20 photographs were ordered in four series with 4-6 pictures in each, showing respectively, differences of building age, building standard, type of enterprises likely to use the building and the design of the site outside the building. The environments shown covered a wide range from large 19th century textile mills and very simple storage sheds to modern, high-tech facilities with landscaped sites.

The participants were chosen among the entrepreneurs (N=26) in different industrial districts in Trollhättan and among municipal planners in Trollhättan (N=7) and Göteborg (N=11). Two consulting private planners were also asked. The entrepreneurs represented establishments with 1- 60 employees in manufacturing, repairs and private services (mostly consulting firms). No retailers were included. Five establishments were branch offices or affiliated with larger corporations. In those cases the local manager was the one interviewed and doing the sort. In the others the owner or one of the partners did the sort.

The participants were asked to sort the photographs in groups according to some kind of similarity and then characterize the environments grouped together. No restrictions on the number of groups were given, nor any clues as to the criteria that could guide the sort. Only one sort was made by each participant and extensive notes were taken of all the terms and expressions he/she used when describing the sort. In most cases the sorts were quickly made and many entrepreneurs were surprisingly quick in characterizing the different groups. When asked they could also quickly choose the group to which their current premises corresponded and also the group corresponding to accommodation they could considering moving to in the near future. In many cases the groups were the same, only the pictures differed.

The hypotheses

The original assumption was that there is a significant difference in the way entrepreneurs and planners, respectively, perceive, describe and evaluate industrial environments. A small pilot study with 5-6 entrepreneurs and planners seemed to give some support to this and indicated furthermore that the difference could be related to the tendency of each group, respectively, to use concepts either describing the physical qualities of the built environments or the businesses likely to use them.

1st hypothesis: There is a significant difference in the way entrepreneurs and planners use verbal constructs in sorting and characterizing pictures of industrial environment.

2nd hypothesis: Planners tend to use constructs describing the properties of the built environment as such and entrepreneurs tend to use constructs referring to the businesses likely to use the buildings shown.

The analysis

The first task when analysing the material was identifying the constructs behind the various terms used by the participants. This was in most cases no difficult task. "Old buildings", "Newer accommodation", were easily categorized as a construct referring to Building age. The term "Modern", used in some cases of buildings and in others of enterprises, was in these cases interpreted as different constructs, referring to Building standard and to Modern in contrast to Traditional businesses, respectively. It must be mentioned that the constructs cannot be expected to form a system of logically distinct categories. They are intended to summarize within a limited range all the categories that could possibly differentiate between the ways different individuals perceive and describe industrial environments.

The total list of constructs formed a range of constructs clearly referring to either the physical/technical/ economical qualities of the built environment or the businesses/users. In the middle of the range are some constructs that are ambiguous or of a transitional nature, such as Mixed area, Building Type, Versatile use and Attractiveness. Among these the constructs Building type required some careful consideration in view of the second hypothesis. Many entrepreneurs used terms such as "Office", "Workshop", "Warehouse", and it was usually not clear if they referred to the building or to the use. Planners were usually more articulate, making it clear they referred to the building. When describing environments as "Attractive", "Suitable", "Appropriate", it was likewise usually clear that both planners and entrepreneurs referred to a certain use or type of business. When classifying the constructs in two distinct groups, Building type was therefore put among the Building constructs. In that way the second hypothesis was made easier to test and falsify. Versatile use and Attractiveness were put among the User constructs.

Originally 35 constructs were established. Table 1. Entrepreneurs introduced 16 of the 17 User constructs. Ten of these were also used by planners who only contributed one additional construct. Planners introduced 17 of the 18 Environment constructs. Thirteen of these were also used by entrepreneurs, who only contributed one additional construct, that of the Odd building they could not place. This seems to give some support to the two hypotheses, but since some constructs were used by very few participants and the variety in the total number of constructs used by the participants was great (2-13), a more sophisticated analysis was needed. Constructs that were not used by at least three participants in either group were eliminated before the actual statistical analysiswas made, which left 12 Environment constructs and 9 User constructs. The purpose was to minimize the effect of individual variations in the relatively small groups.

Several statistical techniques were used to analyse the distribution of constructs used by the two groups of participants - cluster analysis, lattice analysis, multidimensional scaling analysis and consensus analysis. Lattice analysis and consensus analysis are described in technical detail by Corander (1996). For cluster analysis and multidimensional scaling the reader is referred to Arabie et.al. (1996) and Green et.al. (1989), respectively. All techniques yielded results that tended to support the two hypotheses concerning differences in the use of constructs by entrepreneurs and planners respectively. Here the main findings are summarized with particular attention to the results of the multidimensional scaling analysis and consensus analysis.

Table 1. Constructs used by entrepreneurs and planners, ordered according to frequency

#Category ConstructEntrepreneurs N=26 Planners
N=20
27User Enterprise size
62 %
60 %
18Building Building type
58 %
40 %
23User Manufacturing
58 %
30 %
1Building Building age
46 %
60 %
13Building Location
46 %
30 %
26User Public services
38 %
10 %
29User Modernity
38 %
10 %
5Building Condition
35 %
60 %
11Building Odd building
27 %
0 %
20User Attractiveness
23 %
25 %
22User Private services
23 %
10 %
25User Repairs
23 %
10 %
4Building Standard
19 %
30 %
2Building Building size
19 %
20 %
28User Enterprise age
19 %
10 %
12Building Tidiness
15 %
10 %
3Building Rent level
12 %
30 %
6Building Building material
12 %
20 %
21User Trade
12 %
20 %
7Building Exterior design
8 %
25 %
31User Technology
8 %
20 %
19User Versatile use
8 %
15 %
14 Building Parking
8 %
10 %
24User Sub-contractor
8 %
0 %
35User Symbolic name
8 %
0 %
33User Concern for environment
4 %
30 %
17Building Mixed area
4 %
20 %
16Building Unified design
4 %
15 %
10Building Character
4 %
15 %
32User Profitability
4 %
10 %
34User Export industry
0 %
5 %
15Building Bus service
0 %
5 %
30User Business form
0 %
2 %
8Building No of storeys
0 %
1 %
9Building Modular bldg
0 %
1 %

The lattice analysis shows nodes of participants using the same small cluster of constructs with links to nodes with other constructs used by the same participants. This analysis indicated that planners use building constructs more frequently than entrepreneurs. They also used a greater range of building constructs.

The method of multidimensional scaling (MDS) was used in order to investigate in what ways entrepreneurs and planners differ with respect to their use of constructs. The dissimilarity measure percent disagreement between the participants was used in MDS. This and other measures of disimilarity can be found in Gower (1985).) With this method the participants are represented as points in a multi-dimensional space, where a greater number of dimensions produces a more faithful reproduction of the original (dissimilarity) matrix, describing how participants use or do not use the different constructs. Reducing the number of dimensions produces a more intelligible but less accurate view of the data structure, and through iterative calculations the total error (measured by Alienation and Stress-values) with regard to the original matrix is minimized. In a two- or three-dimensional representation the clustering of the participants into more or less distinct groups can be assessed.

A scaling in more dimensions will be more accurate, however, and will support a more detailed interpretation of what the different dimension may represent. This interpretation is to some extent subjective and does not follow from the statistical method as such. For a five dimensional scaling the Stress value was = .10 and the Alienation value = .11.

Interpretation of the five dimensions:

1. This dimension reflects the extent to which participants use branch classification constructs (no 21-26) when characterizing their groups of sorted pictures. Participants with high values in this dimension frequently use branch classification constructs but also other constructs. Participants with low values tend not to use branch classification constructs at all. In this dimension there is a statistically significant difference between entrepreneurs and planners. A non-parametric test (Mann-Whitney U-test) shows that this difference is significant at the 5 % -level (p=.044).

2. This dimension reflects the extent to which participants use building constructs (no 1-17) when characterizing their groups of sorted pictures. Participants with very low values in this dimension use more building constructs than user constructs and vice-versa. In this dimension there is no statistically significant difference between entrepreneurs and planners.

3. This dimension reflects the use of the constructs Building age (1), Building size (2), Modernity (29) and Technology (31). Participants with very low values in this dimension frequently use the two former constructs and not the latter ones, and vice versa. There is no statistically significant difference between entrepreneurs and planners in this dimension.

4. This dimension reflects the use of the construct Building size (2). Participants with high values in this dimension use Building size rather than Enterprise size when characterizing their groups of sorted pictures. These participants are often entrepreneurs, while mostly planners have low values in this dimension, signifying that they do not use the construct Building size. In this dimension there is a statistically significant difference between entrepreneurs and planners. A non-parametric test (Mann-Whitney U-test) shows that this difference is significant at the 5 % -level (p=.003).

5. This dimension reflects the use of the constructs Odd Building (11), Location (17), Private Services (Consultancy firms) (22) and Manufacturing (23). Participants with high values in this dimension frequently use these constructs while participants with low values tend to prefer other constructs. There is a small difference between the entrepreneurs and planners here, but it is not statistically significant.

In an attempt to summarize and interpret the results of the multi-dimensional scaling and of the other analyses of the data it would seem that the participants to a great extent share basic constructs referring to age and size when characterizing industrial environments. Both groups refer to Building age and Enterprise size, while some entrepreneurs stand out in preferring Building size to Enterprise size.

What primarily explains the clustering of planners and entrepreneurs in two fairly distinct groups is the entrepreneurs' much more frequent use of constructs referring to different branch classifications. This interpretation is also supported by observations made while defining the original constructs. A branch classification construct such as Manufacturing summarized a much greater number of different terms used by the entrepreneurs (9 terms) than by the planners (3 terms). In addition, entrepreneurs differ from planners in preferring the general construct Building type (Factory, Office, Warehouse) to other, more developed building constructs.

A cluster analysis was also made of the groups of pictures sorted together by the participants in order to identify possible similarities. In contrast to Groat's study, it was not one of the purposes of this study to explore the participants' ability to identify certain characteristics as such in the pictures presented. It is natural to expect, however, that obviously similar pictures, of for example, old buildings, would be put in the same group by several participants. Possible differences in the sortings would also be of interest, particularly if the differences could be related to the type of participant - entrepreneur or planner. On the other hand, the absence of significant differences in sorting would highlight differences in the use of constructs between the groups of participants.

On the basis of clusters of pictures sorted, three measures of association between entrepreneurs and planners were calculated. One measure is the proportion of clustered pairs of pictures similarly classified by participants belonging to the two different groups. This value, which can vary between 0 and 1, was equal to 0.83, which means that a high proportion of the pictures were similarly classified by entrepreneurs and planners.

Finally, consensus analysis was used. The general model, presented by Batchelder & Romney (1988) was modified along the lines found in Batchelder, Kumbasar & Boyd (1995). Table 2 shows that the differences between entrepreneurs and planners is small for both the estimation methods used. (The degrees of consensus are significant at the 0.001 level in each case.)

Table 2. The degree of consensus for participants in two groups.


Estimation method
Entrepreneurs (N=26) Planners (N=20)Consensus between groups
Unweighted
0.736
0.659
0.784
Bayesian
0.736
0.679
0.826

Conclusion

In conclusion, the first hypothesis concerning significant differences between entrepreneurs and planners has received support by the study. The differences are most prominent between entrepreneurs and the Trollhättan planners, whereas the Gothenburg planners show greater individual variations.

The second hypothesis also has received support, with the modification that it is mainly in the frequent use of branch classifications that entrepreneurs differ from planners in general. Entrepreneurs and planners share some basic concepts for describing buildings, with the difference that entrepreneurs tend to use more general constructs for describing buildings, such as type and size, while planners use more developed environment concepts, such as condition and age.

These characteristic differences between the ways entrepreneurs and planners describe and interpret pictures of industrial environments are further enhanced by the fact that both groups tend to perceive the similarities between the pictures as such and sort them accordingly.

The meaning of buildings for small business entrepreneurs and planners

The results of the study agree well with Groat's study and many similar studies in environmental psychology. According to a summary by Rapoport (1982) users and non-professional people do tend to perceive and appreciate associational values in the built environment. They associate building forms, ornamental details, planning patterns etc with socially attractive or religiously commended life-styles and behavior. Professionals, such as architects, designers, critics, on the other hand emphasize perceptional, formal values.

Canter (1977) finds a logical explanation for these differences in the different roles in relation to the environment these two categories of people take. Users look for benefits, both practical and symbolical which they can gain from an environment. This interpretation is supported by the realistic choice the entrepreneurs made of pictures illustrating possible new premises in the near future. The small entrepreneur in an old low-standard building did not choose the fancy, science park high-tech complex as the next step, which the well established engineering firm in a modern building, would do, however.

The professional attitude toward the environment is a result both of the training and of the actual tasks and responsibilities professional practice entails. In this case on could point out that it is the planners' responsibility to take many different interests into account, not giving undue preference to any one group of users. They are also usually given a specific responsibility for the aesthetic, holistic qualities of the urban environment.

In this perspective, the conceptualizations of the two groups of entrepreneurs and planners seem to reflect quite clearly their respective life-worlds. Economic aspects do not discriminate between the groups, according to the multidimensional scaling analysis, although planners tend to use the construct of rent levels somewhat more frequently than entrepreneurs. (Table 1). A natural hypothesis, which should be tested in another study, is that if investors and developers are included these aspects would receive a more prominent place.

The observed differences between entrepreneurs and planners does not necessarily mean that small business entrepreneurs are indifferent to the formal/aesthetic/expressive values of their buildings, even if they do not often use concepts describing them. According to Rapoport (op.cit.) associational aspects presuppose perceptual ones. One must notice differences in form in order to associate to uses and well-known images. Sensory, aesthetic qualities, could possibly serve as clues to entrepreneurs and their peers when they move on to other and often better accommodation. In spite of a less developed range of concepts to describe the built environments as such, entrepreneurs do not differ noticeably from planners in their actual perception of differences and similarities. They are only quick to interpret them in terms of potential use.

A way to regulate the industrial property market?

The question whether the property market operating within current planning regulations is efficiently serving the needs of expansive industrial enterprises is complex. The contribution of the study described above can only be a small one. But it seems that it does direct attention to the sometimes neglected notion that both the quality and the quantity of the supply of property are in fact relative concepts. What is an adequate standard differs not only between different types of enterprises but also at different points in time for a specific enterprise. Since quite a volume of industrial space is lying vacant or is insufficiently used, this means that the supply of industrial space in the short term is therefore also relative and depends on whether the perceptions of the actors involved coincide. Judgments, based on parameters such as age, building configuration and technical standard, concerning the suitability of industrial property for present use seem less relevant for the very small businesses that are in focus here. The issue is not so much whether the quality of the industrial building stock is high or low in some absolute sense. It is rather if there exists a range of premises sufficiently diverse in size, location, building design and technical standard.(Fothergill et al 1987:141) The relativity of the key concepts involved supports the idea of a local dialogue involving several of the actors concerned, planners, developers, and small businesses.

The possibility for a planning dialogue

The study presented has only explored the conceptualization of two types of actors - small business entrepreneurs and local planners. The next step would be to extend this type of studies to other groups: investors, developers and managers of industrial property. Still, there seems to be some preliminary lessons that can be drawn already, concerning the need and possibility for a planning dialogue. First, there are obviously some difficulties in having people from these different groups understand each other concerning the demand and supply of industrial property. Not only are there differences of values and opinions. There are also different systems of conceptualization. It appears that one does not even talk about the same things. One group talks about the age and conditions, location, probable rent levels of the properties involved. The other talks about enterprises, potential users, functional qualities.

On the other hand, they actually see the same things. They see differences and similarities of actual physical environments in a largely congruent way. This offers another, complementary perspective on Healey's view that: "We see things differently because words, phrases, expressions, objects are interpreted differently according to our frame of reference." (Healey, 1992b:152). In this particular case, one essential part of the frame of reference is the same. It is the built environment, or its representation, that both groups see before them. One can refer to Schutz's thesis on mutual perspectives, expressed in the saying: "We both see the 'same' flying bird in spite of the differences of our spatial position, sex, age, and the fact that you want to shoot it and I just to enjoy it." (Schutz, 1962:316).

This means there should be a possibility for a dialogue. In Forester's (1989) Habermas-inspired terminology one could say that at least two of the conditions for undistorted communication seems to exist when planners and small business entrepreneurs describe industrial environments (op.cit: 145-147). Their common frame of reference, their congruent perceptions of the actual buildings, facilitates comprehensibility. Their different ways of conceptualizing them also seem legitimate, considering their professional and personal interests. Still, in actual planning situations, the truthfulness may be in doubt, factual claims may be unwarranted, and the sincerity of proposals and responses may be lacking.

This is where planning research can contribute. Critically examining the facts put forward and the use made of them, exposing inaccuracies and inconsistencies of argument, comparing words and actions, are obvious tasks of planning research. Another role is to help fill the need for translation skills, further supporting comprehensibility (Healey, 1992b:154). The different discourses brought into play by planning issues - social fears and aspirations, ideological convictions, technical and economical analyses, are not always clearly articulated and immediately comprehensible to other discursive communities. Research can contribute in articulating the values and life-worlds of different social groups, helping parties involved see the other perspectives and perhaps explain the reasons for them. There is never a straight road from analysis to implementation. Attention-shaping is one part of the journey and an important task of the planner, involved in communicative action, according to Forester (op.cit). It could also be a task of planning research, exploring the perceptions and values that tend to escape attention - like the qualities of old industrial environments for small businesses.

References

Arabie, P., Hubert, L. and DeSoete, G., 1996, (eds.) Clustering and classification, Singapore:World Scientific.

Ball, R, & Pratt, A C, (eds) 1994, Industrial Property, Policy and economic development, London and New York: Routledge.

Ball, R, 1994, "Charting the uncharted, Vacant industrial premises and the local industrial property area", in R Ball & A C Pratt (eds), 1994.

Batchelder & Romney, 1988, "Test theory without an answer key", Psychometrica, 53, 193-224.

Batchelder, Kumbasar & Boyd, 1995, Consensus analysis of three-way social network data, Technical Report: Institute for Mathematical Behavioral Sciences, University of California.

Canter, D, 1977, The Psychology of Place, London: Architectural Press.

Corander, J, 1996, Statistical Analysis of Subjective Similaritiy Data, Research Report, Department of Statistics, Stockholm University.

Fothergill, S, Monk, S and Perry, M, 1987, Property and Industrial Development, London: Hutchinson.

Forester, J, 1989, Planning in the Face of Power, Berkely: University of California Press.

Gower, J., 1985, "Measures of similarity, dissimilarity and distance", In S Kotz, N Johnson and C Read (eds.) Encyclopedia of Statistical Sciences (Vol. 5), 145-146. New York, Wiley

Green, P, Carmone, F, & Smith, S, 1989, Multi-dimensional scaling, Needham Heights (Mass): Allyn and Bacon.

Green, H & Foley, P, 1986, Redundant space: a Productive Asset, London: Harper & Row.

Groat, L, 1982, "Meaning in Post-Modern Architecture: An examination using the multiple sorting task", Journal of Environmental Psychology, 2, pp 3-22.

Healey, P, 1992a, Urban Policy and Property Development - globalisation and localisation in real estate development in an old industrial region, Paper presented at the AESOP Congress, June 3-6 1992, Stockholm.

Healey, P, 1992b, "Planning through debate, The communicative turn in planning theory", Town Planning Review, 63 (2) 143-162.

Henneberry, J M, 1988, "Conflict in the industrial property market", Town Planning Review, 59 (3), 241-62.

Henneberry, J M, (1994), "High technology firms and the property market", in R Ball & A C Pratt (eds), 1994.

Heikkila, E & Hunter T A, 1985, "Toward an Evaluative Framework for Land Use Policy in Industrial Districts of the Urban Core: A Qualitative Analysis of the Exclusionary Zoning Approach", Urban Studies, 22, 47-60.

King & Co, 1986, Industrial Floorspace Survey, London: King & Co.

Lindgren, H., 1991, "Sveriges industribyggnader-en kartläggning", Tidskrift för Arkitekturforskning, Vol 4, No 3, pp 9-29.

Mattson, H, 1992, "The Industrial Property Market in Sweden", in B Wood & R H Williams (eds) 1992, Industrial Property Markets in Europe, 193-226, London: Spon.

Pratt, D, 1994, "New land uses: the recommodification of land for new uses on the city fringe", in in R Ball & A C Pratt (eds), 1994.

Rapoport, A, 1982, The Meaning of the Built Environment, A Nonverbal Communication Approach, Beverly Hills/London/New Delhi: SAGE Publications.

Schutz, A, 1962, Collected papers, The Hague: Nijhoff.

Schön, D, 1983, The Reflective Practitioner, How Professionals Think in Action, New York: Basic Books.

Törnqvist, A, 1995, "Development and Growth of Small Firms: Studies of Transfers between Urban Industrial Districts", Paper presented at The 35th European Conference of the Regional Science Association, Odense, Denmark, 22-25 August, 1995.

Wood, B & Williams R H, (eds) 1992, Industrial Property Markets in Europe, London: Spon.