Traditional Small Retail Shops vs. Emerging Supermarkets and Shopping Malls in a Sri Lankan City

Street retail shops are an emblematic image of the Sri Lankan urban landscape. Retail shops vary in size, sell a diversity of household, mechanical and convenience items and appear as an integral part of urban form. While retails shops provide essential services to a variety of consumers generate employment and make urban streets vibrant, there have been few investigations into the nature of retail form in Sri Lanka. It appears that the global spread of supermarkets and shopping malls are transforming the retail space in Sri Lankan cities. Utilizing field observation data, this study investigates the extent to which traditional retail outlets dominate the retail form in the urban landscape, creates employment, and attracts customers. This study also examines to what extent city retail outlets are retained as shop houses. Results show that while Sri Lankan traditional small retail shops form a greater number of retail shops and continue to be an emblem of the urban landscape, the development of a few new shopping malls and supermarkets have impacted the daily shopping behaviour of people and are slowly transforming the retail form as well as urban landscape. ______________________________________________________________________________


Introduction
Retailing forms an important aspect of Sri Lanka's urban function and landscape. Retail shops are pervasive throughout urban areas, with higher concentrations in town and city centres and within linear developments along the major roads and neighbourhoods in Sri Lanka. Retail outlets are scattered everywhere because unlike North America cities, Sri Lankan urban and rural spaces are characterised by mixed uses. Similar to retail shops in Asia and elsewhere in the world, Sri Lankan retail stores sell a variety of items for different socio-economic groups. They sell vegetables, fruits, and electrical goods, pharmaceutical and other essential items. In terms of economic vitality, retail shops provide employment for people, particularly poor people, and improve the tax revenue of local authorities. Similar to other developing countries, Sri Lankan small traditional retail shops cater to middle and lower income groups, while supermarkets and modern shopping malls serve higher socio-economic status groups (Bromley, 1998). Sri Lankan retail outlets also provide economic and community vitality (Duany, Plater-Zyberk, & Speck, 2000;Jacobs, 1961). At the local level, retail shops, particularly in small traditional retail areas, make urban spaces vibrant. They provide spaces for people to interact, and thereby facilitate building and maintaining authentic communities (Gehl, 2010). Such small retail dominant areas not only make urban spaces live but also improve the perceived safety of pedestrians as retailing increases pedestrian movements in streets (Jacobs, 1961). Nevertheless, there are some planning implications associated with retail shops in most urban areas in Sri Lankan. For example, retailers tend to invade pavements/ footpaths though construction of illegal structures, ambulant traders often block pedestrian movements in street and footpaths and some markets are poorly integrated with other urban land uses.
Building on earlier work by Bromley (1998), this paper investigates the form of retailing in a Sri Lankan urban area and the extent to which traditional retail shops dominant the retail form in urban landscape. It also focuses the functioning of retailing in terms of _______________________________________________________________________ employment generation, customer attraction, and to what extent city retail shops are retained as shop houses. The findings reveal that small retail shops form a greater number of retail outlets and therefore they contribute to a larger part of urban development. The results also show changing aspects of retail form and the transformation of urban form in the case study area: Katubedda in the Moratuwa Municipal Council area in Sri Lanka. Since the research focuses on retail form in Sri Lankan urban areas, the nature of this study is exploratory rather than definitive. The study findings primarily interpret the form of retail shops and transformation of urban spaces due to retail stores in the study area as far as possible with the available data sources. This study is a part of a larger research project related to the form and the transformation of retailing in an urban area. Therefore, it needs to be noted that this paper does not examine more on how the retail sector affect planning practices in Sri Lanka.

Retail development and physical planning in the case study area
This study was undertaken in Katubedda ward area in the Moratuwa Municipal Council area. In the case study area it seems that physical planning process in relation to retail development is reactive rather than proactive. Like many other urban areas in Sri Lanka, it appears that Moratuwa City Council promotes linear development along main roads without a proper zoning plan, planning guidelines and enforcement of zoning regulations (United Nations Centre for Human Settlements, 2002). It seems that planners may unintentionally contribute to the oversupply of small retail outlets particularly in existing residential areas. This undermines the residential character that is conducive to residential living leading to amenity impacts from proliferation of retail shops in urban areas. Planners and planning authorities sometimes fail to enforce laws to unauthorized retail establishments, unauthorized conversion of residential buildings for commercial activities and unauthorized filling of wetlands for such trading activities (Ratnayake & Butt, 2012). Similar to many other urban areas in Sri Lanka, retail activities have also affected mobility. Retail shops are located along the busy main roads that run through the town. In the case study area, these types of linear development activities are located along Galle Road as well as Campus Road (see Figure 1 and 2). Since many of these trading activities have direct access from the main road, it has naturally contributed to traffic congestion as well as a continuous inconvenience to the pedestrians (United Nations Centre for Human Settlements, 2002). The informal sector traders also fuel the situation by blocking the pavements or footpaths. The lack of parking facilities in retail areas also contributes to traffic congestion, as customers tend to park their vehicles on the roads.

Transforming of retail form and urban growth in Sri Lanka
Previous research has shown that the form of retailing for developed and less developed countries remain mostly different; however, there is growing evidence of global convergence. In cities of more developed countries, particularly cities in North American and Europe, the ubiquitous carbased shopping mall has become the standardized form of new retail development since the 1950s (Goodman & Coiacetto, 2012). The recent retail literature in Australian Traditional Small Retail Shops vs. Emerging Supermarkets and Shopping Malls in a Sri Lankan City ______________________________________________________________________________________ cities indicates that the global enclosed type shopping malls have been embraced to various degrees in different cities, and cities have proportionately a much smaller amount of traditional street-based independent retailing (Goodman & Coiacetto, 2012). The retail literature for the less developed countries highlights that market-place trade and streetbased independent shops continue to thrive within expanding cities and in linear developments alone the major roads (Bromley, 1998;Skinner, 1985). In less developed countries, studies have identified many retail forms: traditional street-based independent shops, market-place trade, supermarkets, ambulant or informal retail outlets and less common, shopping malls (Bromley, 1998;Romo & Reardson, 2009). Similar to other developing countries, retailing also comes many forms in the cities of Sri Lanka. For example, pola (farmer's markets), streetshops, informal/ambulant traders, supermarkets, traditional market centres and post-modern enclosed type shopping malls.
In Sri Lanka, it is not easy to define retail forms as boundaries between various forms are often blurred. However, they can be classified according to location, size, and type of goods on sale. Pola are usually open-air markets on a concrete platform or on streets. They often operate in fixed sites in towns or villages on periodic days. Street based retail shops usually operate in fixed sites and comprise permanent building structures. These shops are found on most of the main roads and street corners in cities and are also scattered in random fashion throughout neighbourhoods. Street based shops are generally small in size and operate throughout the week. Traditional market centres comprise many fixed shops that are occupied by traders operating on a daily basis or on market days. They are mostly located in the historic centre of the town or city and tend to be contained within one story buildings. Supermarkets, usually two storey or three storey buildings comprising a variety of shops and operate on market days. They are also mostly located in town centres. Supermarkets in Sri Lanka are therefore having some similarities with traditional markets. The shopping malls tend to be the largest markets, and they are found mostly in cities in Sri Lanka. Most of the modern supermarkets and shopping malls provide plenty of parking facilities. The informal or ambulant traders, often on streets or pavements, tend to be mobile and they may not have permanent structures. The ambulant traders remain stationary for long hours throughout the week. Similar to pola and fixed retail shops, they sell foodstuff, household items and a variety of manufactured items. Therefore, as indicated in past research, the boundaries between the forms of pola, streetbased retail shops, ambulant trading and supermarkets in terms of goods on sale are ambiguous (Bromley, 1998).
The form and location of retail shops have been affected by the introduction of open economic policies in Sri Lanka since the early 1980s and the impact of these policies seem to support trade as retail sector has continued to flourish in the last two decades. Since the 1980s, a variety of state-owned supermarkets such as Sathosa supermarket and LakSathosa chain of supermarkets have been established in most of the urban areas. The state also encouraged private sector investors to establish modern supermarkets such as Cargill's, Laughs and Keels supermarkets, particularly during the last two decades. These supermarkets have been located in main urban centres and they attract mainly upper class and middle income customers. There has also been a growth of planned shopping malls in inner city areas during the last three decades (Liberty Plaza, Magestic city and K-Zone). Nevertheless, the growth of closed type shopping malls has been relatively slow compared with modern supermarkets and street based traditional retails. These shopping malls mostly attract high status customers.
Apart from the previous retail types, new special economic trade centres have been established under the Sri Lankan Ministry of Trade and Commerce. These centres are located in high-density urban areas of Dabulla, Ratmalana, Norochcchola, Talawa, Narahenpita, Anuradhapura and Meegoda (Wedaarachchci, 2009). Similar to the Victoria Market in Melbourne (Goodman & Coote, 2007), these trade centres are large in size compared to traditional open air or semi-open air 'pola' (farmers' market). They comprise a variety of retail outlets such as fish, vegetables, eateries, household items and other essentials. It appears that the new special trading centres attract particularly middle and low-income customers as these traders sell their items at wholesale prices.
In developing countries, the changes to the retail form and proliferation of traditional street based retail shops have been related to urban growth, increasing household participation in marketing process and increasing efficiency of transport (Skinner, 1985). It was observed that during the last three decades, the population densities in major cities in Sri Lanka were increasing at an average rate of four percent per annum (Department of Census and Statistics Sri Lanka, 2010). The commuter population has also increased in major cities due to increasing activities in urban areas such as service sector and industrial sector employment (Urban Development Authority, 1999). In recent years, It has also been observed that the frequency of public and private transport facilities, particularly public buses, have increased in urban areas alone with the urban growth in Sri Lanka. Therefore, in Sri Lanka, similar to many other developing countries, the growth of population densities, the frequency of public and private transport facilities and household participation in market process tended to increase demand density and foster emergent alternative global retail forms such as supermarkets, hyper or modern supermarkets and shopping malls (Skinner, 1985;Bromley, 1999). The demand density continues to support street-based independent retail shops, traditional market centres and ambulant traders or pavement hawkers.
In developing countries, the growth of streetbased retail outlets and pavement hawkers or 'informal sector' is associated with not only demand density but also poverty in these countries. In less developed countries, due to poor economic conditions of the state, the poor do not receive sufficient welfare payments and in some instances, there are no safety nets for youth. However, the cost of living is increasing on a daily basis (Dholakia, Dholakia, & Chattopadhyay, 2012;Lawson, 1995). This situation has tended to result in increasing numbers of people engaging in petty retail shops such as street-based small shops and ambulant trades or pavement hawkers (United Nations Centre for Human Settlements, 1996). Although many studies explored the evolution and the transformation of retail trade in developed and less developed countries, the focus of the form of retail trade, its transformation and operation in urban areas in Sri Lanka has been neglected. Therefore, this study attempts to shed light on the form, changing aspects of retailing and transformation of urban form due to unplanned retail expansion in a Sri Lankan city area.

Case study area
The research project was carried out in the Katubedda area. The case study area is known for its skilled traditional carpentry works and furniture industry (Moratuwa Municipal Council, 2002). Kadubedda and its nearby areas also have rubber product, transformers, and batter industries. Katubedda is also home to Moratuwa University, the most well-known technological University in Sri Lanka. This suburb includes the famous Arthur C Clark centre and popular high schools such as Prince of Wales and St. Sebastian's College. Katubeddaarea also comprises many banks and financial institutions such as the Bank of Ceylon, People's Bank, Seylan Bank, Commercial Bank and Seylinco Insurance companies. Furthermore, an international cricket stadium, Tyrone Fernando Stadium, is situated in Katubedda. In recent years, it has been observed that the population growth and the above-mentioned urban activities have created a demand density that helps the retail sector to flourish in the case study area(Urban Development Authority of Sri Lanka, 2005).

Method of study
This study is focused on field observations and interviews undertaken during 2012 along the major roads with linear retail development in Three stages were involved in the research. The first stage comprised field observations that were utilized to identify and map retail shop locations, measure the size, form of retailing (mall, street-based retail shop, ambulant retail outlet or a shop house,) and type of activity (type of goods on sale). In Sri Lanka, it is common for municipal councils to monitor and maintain a registry of fixed retail outlets in their council areas. However, local authorities often do not have reliable data sources in terms of ambulant traders, type of activity, classification or spatial maps of retail outlets.
The Second stage involved the classification of retail ships and form of retail shops in the project area. Since the local authority or any other local administration agencies do not have a formal retail classification system in Sri Lanka, the authors classified retail shops according to the type of consumer goods on sale. In the retail literature, retail classification has been conducted in a similar way; therefore the retail activity groups classified in this study are consistent with past studies (Ghani, 2005;Juhasz, Seres, & Stauder, 2005).
When it comes to retail form, it is not easy to distinguish one form from the other because retail forms or types are blurred. However, in the retail literature (Romo & Reardson, 2009), there have been attempts to define retail forms in many ways e.g.: size, ownership, scale of operation and environment. Using the retail literature as a guide this study identified and classified four forms of retails in the case study area. The retail forms were classified based on the following characteristics: 1. Traditional street-based fixed retail shops: typically small/ medium scale, single-owner, often run by a husband and wife or 'mom and pop' or a family member or employee, not selfservice, and typically carry dry goods/foods, packaged foods, and non-food such as detergents, or wet goods (produce, poultry, or meat). The classification of retail shops and forms of retail outlets in the study area are shown in Figure 2 and Tables 1 and 2 below.
The third staged involved the observation of customers' behaviour in each fixed retail shops. At this stage, the number of customers that visited each fixed-retail shop in the study area was observed. Interviews were held with shop owners to collect information regarding the number of employees and age of the retail shops. The findings of all three stages were then analyzed to identify the dominant form of retails, growth of mall type retail outlets, and transformation of urban landscape due to retail shops.

Study Findings
Results from the field observations and interviews about retail locations, retail forms, and the transformation of urban landscape due to retail shops are discussed in this section. The study area reflects the vibrant character of the retail sector and a variety of retail options in typical urban areas in Sri Lanka. The findings indicate that a large number of retail shops tend to be the small street based retail shops and their growth has been surprising during the last decade. The study area consists of a considerable number of shop houses. As the data reveal, although there has been a slow growth of modern supermarkets and shopping centres in the study area, these retail stores proportionately attract a large number of customers. The findings also indicate that retail shops, particularly street-based small shops have contributed to convert residential lands uses into commercial land users.

Retail Forms in the study area
The field observation data identified three forms of retail provision in the study area: traditional street-based retails; informal sector traders; and shopping malls. The findings reveal that the majority of the retails (87%) were traditional street based retails. The study area also comprised 11% (n = 23) ambulant traders, a modern supermarket and a shopping mall. The study did not find traditional markets and Pola (farmers' market) in the study area although these retail forms exist in most of the urban areas in Sri Lanka. This is may be due to the small scale of the case study area and this area had not been a traditional urban centre. As previously discussed, traditional market centres and weekly Pola are mostly found in traditional city centres or towns in Sri Lanka. The retails forms in the study area are shown below. As shown in the table above, the study identified a total of 201 retail outlets in the project area. Of 201 retail shops, 178 (87%) retail shops were identified as fixed retail shops. The remaining 23 (%) shops were informal traders who did not occupy fixed places to sell their goods (See Figure 2). They usually occupied footpaths or pavements to sell their items. Although informal trading has been a pronounced feature and it offers a wide variety of cheap food and manufactured items for all socio-economic groups in the study area and more general in all urban areas in Sri Lanka, this study biased towards the identification of nature, form and changing aspects of fixed retails. Therefore, this study did not take into account informal sector traders when analyzing retail groups or categories, number of customers visited for retail shops and age of the retail stores.

Retail categories
All fixed retail shops were classified into 12 retail categories based on the types of goods for sale. The findings show that the area has diverse and relatively well-balanced retail sector. It appears that the prominent retail shops in the study area tended to be food and beverages (19%), stationeries and communication (14%), groceries (13%), health and personal care stores (pharmacies and herbal products) (11%), home furnishing and appliance (9%), spare parts shops (10%), and to a lesser extent, hardware, clothing, perishables, timber, aluminum, gift and bric-abracs shops. Retail categories are shown in Figure 2 and Table 2. Table 3 shows the floor area of traditional street-based retail shops, modern supermarket and modern shopping mall. In the absence of reliable data sources, the retail space data were calculated using Geographical Information Systems (GIS). Several field visits have been conducted to verify the reliability of data. The results show modern shopping centres consume relatively large amount of urban space and they seem to threaten the rich retail mix in the study area. From Table 3 it can be seen that most of the street-based retail stores (87%) appear to be relatively small with the floor space less than 200m² compared with shopping mall and super market. Furthermore, the findings show that the average floor area of street-based shop was 121 m². The minimum and maximum of floor area of these shops were 13m2 and 1870m² respectively. It seems that the size of the street-based retail shop is smaller than retail locations in cities in developed countries. For example, one study shows that the average floor space of traditional street-based retail outlet was 250m² in Melbourne and 207m² in Brisbane in 1991(Goodman & Coiacetto, 2012.

Persons employed in retail shops and customers visited per retail location
In the study area, it seems the retail sector generates a number of employments. The field observation data show that a total of 563 were employed in the retail shops. As shown in Table 4, the average person employed per traditional street-based shop was relatively low (2.7) compared to super market (28) and shopping mall (50). This may be due to the small size of street-based shops and the low numbers of customer visits per small retail outlets. This may be also due to a large number of employees distributed throughout many shops in particularly the shopping mall. In order to get a more accurate picture of customer flows to retail shops in the study are, this study also observed customer visits to all forms of retails in the morning (8am-9 am), the afternoon (1pm-2pm) and the evening (5pm-6pm) on a week day. The results show that more than 12,500 customers visited all types of shops in the study area within these three hours. It seems that the majority of customers (more than 10,000) visited traditional street-based stores. However, the results indicate that a larger number of customers visited the supermarket and the shopping mall in the study area (See Table 5).

Longevity of the retail outlets
The field data suggest that the study area has experienced a greater rate of retail change during the last the last 10 years. What is most remarkable when looking at this increase is that the number of traditional street-based shops has increased from 9 to a total of 160 while modern super market and shopping mall have increased by 2 during the same time period, as shown in Table 6. This may be due to the growth of population in the area, and increased pedestrian movements due to growth of service sector activities such as banks and education institutions. It could be also attributed to the initiative of people who are unemployed and live in poverty developing small-scale trades to earn money. These data further suggests that the small street-based retail outlets are the dominant feature in the study area.

Spatial distribution of retail shops
The growth of retailing particularly the streetbased traditional shops over the last three decades and how they transform the urban space are better understood through retail location maps.
For this purpose, all forms of retail sites have been mapped out by using field observation data. The maps (Figure 3) show location of all fixed trading sites on main roads over the last three decades. The evidence of fixed trading sites clearly indicates the growth in the number of retail sites associated with the increased number of traditional street-based small retail outlets. It also shows how residential lands on main streets, particularly alone the Campus Road, have converted into retail outlets and the dramatic expansion of small retail shops on adjacent sites (Figure 4).
It was calculated that about 10000 square metres or 1 hectare of new retail/commercial land has been added to the study area. As the Figure 4 shows below, the sites which have become retail land use after 2002 have been mainly residential land uses before 2002. Furthermore, findings show the gradual increase of relatively large shopping centre (shopping mall and supermarket) and how these two large sites of shopping centres have transformed the urban space.

Discussion
The aim of this research to was to investigate different forms of retail provision and to understand to extent to which retail development transform the urban landscape in a rapidly changing urban area in Sri Lanka.
The study also carried out a field observation survey to document advantages of retailing sector as a generator of employment and generator of vibrant urban spaces, a vital part of the everyday urban experience of people in Sri Lanka.
In cities in developing countries, the traditional street-based retail sector, flourishes and contributes a substantial amount to economic growth (Rajkumar & Jacob, 2010). In India, the organised retail sector in 2011 is estimated to be growing at a rate of 15-20 per cent per annum (Global Retail Development Index, 2012). In Sri Lanka, it appears that post-war stability, political stability, unprecedented levels of tourism growth (46 percent in 2010 and 31 percent in 2011) and infrastructure development facilitate the expansion in the retail footprint (Global Retail Development Index, 2012). Although the organized retail sector makes up only about 3 percent in the Gross Domestic Product (GDP), it is expected to grow steadily. Modern supermarkets continue to grow in city centres. For example, Arpico increased its number of outlets in different city centres in Sri Lanka from 10 to 31 in 2011 (Global Retai Development Index, 2012).

_______________________________________________________________________
The findings of the present research highlight that small street-based retail sector has been growing at an increasing rate during the last two decades and also contributes in generating employment. It appears that the growth of modern supermarkets and shopping malls have been slow; however, it is expected that they will continue to grow and become popular in the future. In that respect, the present study findings support past studies that show the dominance of small retail shops and the steady growth of shopping malls and hyper markets in cities in developing countries (Bromley, 1998;Duany et al., 2000). Although the present findings reflect a realistic picture of small retail development in urban areas in Sri Lanka, it needs extension to the whole Moratuwa city council area and few more urban areas in order to get a more complete picture of the growth of supermarkets and shopping malls in urban areas in Sri Lanka. Nevertheless, the study area findings show that the area has a rich retail mix which provides a variety of retail options with respect to goods on sale for customers.
The study counted the number of customers who visited each retail shop in the area. From this data, it appears that retailing sector, more particularly; traditional street-based retail sector helps to make the urban street busy and vibrant. Since retail shops operate until late night on weekdays and weekends, this will improve the spatial confidence of pedestrians when using urban spaces at night. Further, this study analyse the floor area of all retail shops in the area and this data reflect that modern markets consume relatively a large floor area in the urban landscape compared to traditional street based retails. This has changed the landscape in the Katubedda area.
The development of large-scale market may threaten the existence of small retail outlets in the future in this area. Such development trends will have an effect on the urban experience of people, and thereby the vitality of the area. This study findings support to compare average floor area size of street based retail shops in cites in developed and developing countries. In contrast to that of the findings of average floor area size of street based traditional retail shops in Melbourne in Australia, the present study results show that the average size of street based retail shops are relatively small in Katubedda area. Furthermore, past research highlights that large shopping malls and modern market stores tend to locate on outskirts in cities in   (Bromley, 1998;Ghani, 2005). However, the present study findings show that large new markets are located along the main roads within close proximity to (within 500m) the urban centre. It appears that these large markets work well with the aid of efficient public transportation system in this area.
The research findings show retail development has negative influence on the other land uses. Haphazard development of retail outlets along the roads and unplanned transformation of residential areas into commercial land uses will cause liveability issues in the existing residential areas as well as causing traffic problems more generally in urban areas. The present study did not study about liveability issues associated with retail sector and it needs to be focused in the future studies. Nevertheless, planning authorities do not seem to be aware of this development. They seem to be unintentionally promoting retail sector expansion along existing residential areas. Since it is the primary responsibility of planning authorities to control and guide the urban development, local planning authorities need to have a consistent, holistic retail development policy to sustain an appropriate mix of retail development as well as other land uses in urban areas.
There has been a dearth of academic publications of systematic survey-based research of transformation of urban spaces and the impacts of retail sector expansion in urban areas in Sri Lanka and more generally in developing countries. Given the already increasing importance of transforming urban spaces due to different formats of retail shops and their associated impacts on other land uses, the retail sector's development in urban areas in Sri Lanka should be studied more systematically with field surveys in order to further analyse residents' perception of retail sector expansion in urban area, consumer satisfaction on existing retail sector, retailers' views ongoing retail sector development in cities, health issues associated with retail shops and infrastructure problems related to the retail sector. These appear to be the important gaps in retail sector studies in cities in developing countries that urgently need addressing in urban policy research.