Retail trade in Brussels: reconciling the city with a changing sector

In 2017, the Brussels-Capital Region 20 700 points of sale for goods and services, accounting for 87 % of the expenditure made by the inhabitants of Brussels. This retail trade provides about 9 % of regional employment. This synopsis reviews what is known about distribution in Brussels, an indispensable economic sector whose complexity makes it difficult to understand due to its diversity, and which is subject to profound changes in consumption, the urban environment and the internal organisation of the profession. After outlining the framework of the analysis by defining trade through its functions, forms and stakeholders, the authors draw up several observations associated with major dynamics in the sector (transformation of supply and demand and changes in public regulation in particular). The third part of the synopsis responds to these observations by identifying clear issues which, if taken into account by the sector and supported by public authorities, could reconcile trade with urban space.


Introduction 1
The medievalist Henri Pirenne  said that the city is the "daughter of commerce". This suggests an obvious and close link between an economic and social activity, stationary or itinerant retail trade, and a type of space, namely the urban space [Lemarchand, 2008b;Dugot, 2010;Renard-Grandmontagne, 2016]. 2 Today, the economic structure of cities -especially larger ones like Brussels -is much less dependent on retail trade than it was in medieval times. However, despite an expansion of commercial space, which has spread to the suburbs as well as cyberspace, the distribution of goods and the provision of services for the final consumer remain strong elements in the urban landscape and still provide many jobs. As a result, Brussels is both a place which supports retail trade (physically as well as in terms of professional and consumer practices) and a region which is produced in part by retail trade. This is an obvious fact which we sometimes tend to forget: cities concentrate the population and therefore also the economic activities which meet the needs of this population.

3
Like any economic activity, trade undergoes changes. This is due to changes in the organisation of companies (increase in productivity, economic integration in particular), and is also due to consumers. Trade is one of the few economic sectors which everyone is in direct contact with and also has a direct hold on. In big citiesprobably more than elsewhere due to the diversity of the offer -customers are not only "trolley pushers" [Navereau, 2011], but are also "enlightened consumers" [Lemarchand, 2011]. 4 The importance of exchange activity in the urban metabolism 1 and the evolution of the commercial offer and consumption patterns therefore fully justify the summarising and updating of information on the relationship between retail trade and the city [Renard-Grandmontagne, 2016], particularly in Brussels where the future of this relationship is shaped increasingly by regional competencies. 5 The purpose of this synopsis is to gather the extensive knowledge on retail trade in Brussels and its evolution and functioning. It is divided into three parts. The first sets out the definition of retail trade and the institutional framework it operates in. The second sets out a series of findings, supported by empirical data and converging academic literature. The third part deals with the retail trade issues in Brussels which emerge from the findings or which are currently being debated. This third part is presented in the form of recommendations, even if they do not necessarily provide turnkey solutions for consideration.
1. The framework: diversity of functions, multiplicity of stakeholders 6 Trade is the activity of selling goods or services. Retail trade resells them in quantities suitable for the final consumer, while wholesale trade supplies them to retail resellers or institutions. In the business world, this is referred to as "business to consumer" (B2C) and "business to business" (B2B) respectively. In practice, retail trade performs multiple functions: between the technical nature of the offer and the skill levels of consumers (IT, DIY and sports).
Finally, it should be emphasised that retail trade tends more and more to steer certain production chains (by defining products in particular) due to its knowledge of the market. It is no longer just an outlet for standardised industrial products. This new steering function tends to strengthen the power of large retail companies with respect to industrialists [Moati, 2011].  Vandercammen and Jospin-Pernet, 2010, pp.28-29 7 While for a long time these different functions have generally been fulfilled on the same site, i.e. the point of sale, one of the major changes in how retail trade operates is the progressive geographical (and temporal) decoupling of these different components of the trade. For example, nowadays it is much easier to place an order and pay online at home and pick up the goods in a shop, or order and pay in a shop and have the goods delivered to your home. terms (in the official classification of activities, such as in collective labour agreements), this covers 20 696 establishments in the Brussels-Capital Region in 2017: the sale of goods, which in Brussels represents 10 195 points of sale (49 % of active business establishments). This category includes, among other things, food businesses, personal and household equipment and the sale of goods related to recreation; cafés, restaurants and fast food businesses (HORECA) account for 5 100 establishments (25 %); commercial services (hair salon, bank branch, laundrette, etc.) represent 4 407 commercial establishments (21 %); vehicle sales and transport related services occupy 994 locations (5 %).  (1950-1969-1997-2017) Source: Université libre de Bruxelles, applied geography and geomarketing department and hub.brussels 9 As a result of varying trends according to sector of activity, retail trade in Brussels has experienced a progressive redistribution of the offer. Between 1950 and 2017, shops specialising in personal goods and, to a lesser extent, those offering everyday or leisure products, decreased proportionately in favour of an offer oriented towards services, restaurants and, to a lesser degree, businesses related to transport and household equipment.

The landscapes of urban retail trade: forms and places
10 With nearly 28 000 points of sale or provision of services of a commercial nature and 2,4 billion m 2 of net sales area for goods, the Brussels urban area has the largest concentration of commercial offer in Belgium by far.  Source: Université libre de Bruxelles, applied geography and geomarketing department, based on Locatus land inventories (2019), delimitation of the Urban Region according to Vanderstraeten and Van Hecke [2019] 11 In urban areas, over the centuries, sedentary trade has been organised spontaneously in more or less large clusters, often referred to as hubs. If the city is of sufficient size, these clusters are hierarchical. The main hubs are those constituting the commercial centre. A series of large commercial centres (100 to 300 shops) also radiate over part of the city and possibly the outskirts. Smaller and still quite diversified centres with 50 to 100 shops radiate over a district. In addition, there are smaller, looser clusters and scattered businesses [Grimmeau et al. 2004;Grimmeau et al. 2007].
Retail trade in Brussels: reconciling the city with a changing sector Brussels Studies , Notes de synthèse Source: hub.brussels [2017], Grimmeau and Wayens [2003] 12 Since the 1970s, groups of medium or large shops have emerged; these are generally specialised, but may include supermarkets or hypermarkets, built around car parks, outside towns or in the outskirts. The architecture of these commercial parks is very simple, to the point that they have been referred to as alignments of shoeboxes [Péron, 2004;Garcez and Mangin, 2014]. In some cases, particularly along roadways, these alignments may extend so far that shoppers find it useful to take their cars from one shop to another. 13 The grouping of shops can also be planned. Shopping centres appeared in Belgium at the end of the 1960s, in cities and on their outskirts, and include several shops under one roof in galleries which are usually covered. In an urban environment, shopping centres can be in continuity with the commercial fabric of a hub. In the centre of Brussels, City2 was built in continuity with Rue Neuve, and is an example of this.
14 Finally, dispersed retail trade is made up of points of sale which are located in towns and cities in a relatively isolated manner. Dispersed retail trade is not necessarily made up exclusively of small shops: a hypermarket or a large furniture shop can be isolated and cover several thousand square metres, which is equivalent to the cumulative surface area of many commercial centres. This example also shows that dispersed retail trade cannot be likened to local retail.  (2019) 15 In addition to these permanent or sedentary commercial spaces, there are temporary commercial spaces, i.e. markets. In 2017, there were 59 locations in the Brussels Region where at least one market was organised once a week, and a total of almost 100 weekly markets. The geographical distribution of the markets throughout the whole of Brussels is relatively homogeneous, with, however, a fairly logical concentration in the parts of the Region which are more densely populated or have a greater reach, with certain markets (the Midi and Abattoirs markets) having a supra-regional catchment area [Lambert et al., 2017].

Shops and merchants: statuses and profiles
16 A point of sale is not necessarily a company in itself, as the same company may have more than one establishment (also known as places of business). In the administrative sense, establishments differ in location. While a retail business may have several points of sale as well as an administrative headquarters and warehouses, each of these sites constitutes an establishment. The number of establishments is always greater than the number of companies.
17 Businesses are statistically identifiable because they are legal entities and are liable for VAT, which is collected by the tax authorities, and therefore means that there is a statistical record of their head offices, which are not necessarily places of business. Self-employed persons who often run these companies pay social security contributions to INASTI. If establishments have employees, they are identified in the Retail trade in Brussels: reconciling the city with a changing sector Brussels Studies , Notes de synthèse decentralised statistics of the ONSS, the body responsible for collecting social contributions and which is therefore a source of statistics on salaried employment. Establishments may be run on a day-to-day basis by a manager, who is an employee.
18 In terms of labour law, a merchant can therefore be self-employed or receive a salary (employee or worker). 19 For the consumer and in the commercial landscape, this administrative distinction associated with employment status and social security systems is not obvious. Rather, a distinction is made between points of sale which are grouped together as a brand and operate in a coordinated manner, particularly in terms of logistics. This organised or integrated retail ("chains"), brings together points of sale operating as networks with various statuses (branches, franchises, voluntary chains, cooperatives) [Desse, 2008]. The points of sale in a network grouped under the same brand name may be managed as a branch (the manager is an employee) or as a franchise (managed by a selfemployed person in the context of a separate business, linked contractually to the franchiser). The two management methods can coexist within the same network of organised retail. All branches of the same company have the same VAT number, while each franchised establishment has its own number.
20 Unorganised retail therefore consists of points of sale operating alone, without being a branch and without a franchise agreement. The latter category is often regarded by the general public as being made up of small independent merchants. But legally, franchises, which are part of organised retail, are often also separate businesses run by self-employed people. 21 While the distinction between self-employed and salaried employees is obviously important in employment policy, in practice, when it comes to examining the economic and logistical rationale behind retail trade, the distinction between integrated or networked retail (branches and franchises) and separate independent retail is much more relevant. Distinguishing between small and large shops in terms of sales area is also relevant, provided that one is careful not to systematically equate large shops with organised retail, as this is not always the case.
1.4. Public management of retail trade 22 In the Belgian federal system, there are many levels of public authority which organise and regulate retail trade, forming a system of shared public responsibility which integrates certain forms of subsidiarity, for example between regions and municipalities for urban planning issues. 23 Moreover, due to the diversity and autonomy of the stakeholders within the same level of authority (the three regions and the municipalities in particular), the legal and regulatory framework as well as the day-to-day functioning of retail trade are greatly dependent on the location of the company or the establishment.  While the table may give the impression of a highly fragmented and uncoordinated public approach to retail trade, it should be borne in mind that integrated approaches at local and regional level do exist. As early as 1997, a first Town Centre Manager (TCM) was established in the centre of Brussels as part of the integrated retail initiative. But the public authorities soon took over and, in 1999, 12 units for the management of commercial centres were set up by the Brussels Region, for a selection of centres in difficulty in the old neighbourhoods. The approach was renewed and extended to 15 neighbourhoods in 2002. City centre management made up for the public authorities' lack of coordination in this area (safety, waste management, management of public spaces, search for new occupants for empty units) and professionalised the activities of shopkeepers' associations, with the risk of occupying too much of the manager's time. 25 In Brussels, the city centre management abandoned the activities, seen as not very productive, and became Atrium in 2005, covering 27 centres. In 2012, atrium.brussels became the regional agency for retail trade, and since then has taken care of all of the Region's commercial centres. In addition to data collection (which represents a significant need), atrium.brussels works to convince retailers and brands to occupy available units. This approach complements that of private brokers who only operate in the most profitable areas. Gradually, atrium.brussels has devoted a growing part of its activity to helping new merchants to fine-tune their project, find premises, carry out administrative procedures, and obtain an affordable rental and a loan. But atrium.brussels also continues to act locally to remove obsolete signs and advertising, improve shop fronts and harmonise them, provide visibility to shops and neighbourhoods, and even help municipalities to convert empty units into housing or Retail trade in Brussels: reconciling the city with a changing sector Brussels Studies , Notes de synthèse help shopkeepers move to more profitable locations [Grimmeau and Wayens, 2016], which are activities recognised as good practice in commercial revitalisation [Westvlaams Ekonomisch Studiebureau, 2015]. 26 Finally, in 2018, with a view to rationalising the business support sector, hub.brussels was formed by the merger of atrium.brussels, Impulse and Bruxelles Invest & Export, the latter two organisations managing the creation of companies as well as export aid and foreign investment respectively.
2. Findings: economic significance, reorganisation and weak regulation 2.1. Retail trade is a major source of employment 27 Employment in retail trade in the Brussels-Capital Region, including self-employment, restaurants and car sales (but not banks and insurance or non-stationary retail) was estimated at 61 000 full-time equivalents in 2018, of which more than 18 000 were in the restaurant sector and more than 3 000 were in the car and motorcycle sales and maintenance sector. This represents 9 % of regional employment. 28 In the rest of the Brussels urban area, the suburbs have more than 43 000 full-time equivalents, including almost 34 000 for the rest of the municipalities in the morphological urban area outside the Brussels Region. The 104 000 full-time equivalents in the wider Brussels urban region represent 20 % of national employment in retail trade. Most of the employment in retail trade is salaried employment: 72 % in Brussels and in the rest of the morphological urban area, and 49 % in the suburbs.
Retail trade in Brussels: reconciling the city with a changing sector Brussels Studies , Notes de synthèse The development of organised retail, as well as the growth in the size of points of sale observed both in organised retail and among self-employed individuals contributes to the spread of wage-earning in the retail trade. The competition between integrated and independent retail takes place through the mobilisation of household expenditure, as well as on the labour market. For example, some butcher shops at the end of their activity cannot find a buyer, as young butchers who have just finished their training prefer more comfortable working conditions in organised retail [Grimmeau et al., 2007]. 30 In the Brussels urban area, the diversity of shopping areas, their environment and their public has an impact on the density and nature of retail employment. This is, first of all, the result of a structural effect: not all activities offer the same employment density, have the same proportion of integrated retail or have the same wage rate. Beyond this structural effect, significant differences reflect local specificities [Wayens and Keutgen, 2015;Wayens and Keutgen, 2016]. 31 In the Brussels-Capital Region as well as in its outskirts, the employment density per unit of net sales area is well above the national average. In 2009, there were 3.1 jobs (FTEs) per 100 m² in Brussels, compared with only 2.1 on average for the country as a whole [Wayens and Keutgen, 2015;Wayens and Keutgen, 2016]. Some municipalities have lower values. They are all characterised by the presence of hypermarkets and large specialised shops. The typical activities of these large shops in the outskirts are those with lower employment densities per m². But even in these cases, the employment density is higher than for the same activities elsewhere in Belgium. This is Retail trade in Brussels: reconciling the city with a changing sector Brussels Studies , Notes de synthèse largely explained by the density of the market and the large clientele, as well as by the cost of property, which makes it necessary to optimise the use of space. In the central area of the city, population density and property prices are forcing the number of staff to increase more than the surface area, which contributes to the growth of the workforce, regardless of how the points of sale are operated [Wayens and Keutgen, 2015].
32 As regards the relative presence of integrated retail and the degree of salaried employment, there is clearly a specificity associated with the municipalities in the inner ring (from Saint-Josse to Forest via Molenbeek). The proportions of integrated retail and salaried employment are significantly lower, not only as a result of a structural effect, but also due to a local effect which reinforces the trend. This is in particular related to ethnic retail [Ma Mung, 1996;Lemarchand, 2008a], which satisfies less standardised needs which are therefore (for the time being) taken into account less by integrated retail. As retail is traditionally a sector of integration for immigrant populations [Zalc, 2010], entrepreneurial structures in these areas remain more family oriented. On the other hand, the municipalities in the Region or in the outskirts, where brands are most present and where salaried employment is the most developed, are home to the most modern forms of commercial infrastructure.
33 Retail trade is one of the sectors in which the rate of Brussels residents among workers is the highest [Ermans et al., 2019]. The Brussels Region is linked administratively to only part of the employment area and therefore does not necessarily benefit from the tax revenue generated by the jobs located there. The local anchoring of the commercial workforce is therefore very important and positive in the Brussels context, as the region is experiencing high unemployment, particularly among those who are least qualified. In addition to proactive training policies, maintaining a supply of low-skilled jobs in the economic fabric is a real social issue. Retail trade and the HORECA sector provide jobs which require low qualifications (at least in terms of diplomas), making it a particularly important sector for employment policy. In the Brussels Region, 65 % of retail trade employment is held by Brussels residents, and 70 % of these jobs are held by low-skilled workers. For HORECA, these proportions are 75 % and 77 % respectively [Observatoire bruxellois de l'Emploi, 2015].

Consumption is changing
34 A long-term perspective on household consumption in Belgium shows that per capita expenditure at constant prices in retail trade in the strict sense (sales of goods) has been stable at least since the end of the 1970s, while per capita income has continued to rise. The household income supplement was therefore not spent on the purchase of goods sold in the retail trade. In fact, it was devoted to commercial services, the costs of car use and other means of transportation and those related to housing (water, heating, electricity). The greater part of the increase in revenue has therefore not benefited retail trade [Grimmeau and Wayens, 2016]. 35 If we look at the different indicators at national level, we see that while household expenditure in retail trade has remained stagnant for a long time, retail turnover increased until 2000. Part of the turnover from the sale of goods can therefore not be attributed to households (B2C) and must come from professional purchases (B2B). These may include food products purchased by restaurant owners and confectioners, Retail trade in Brussels: reconciling the city with a changing sector Brussels Studies , Notes de synthèse building materials for construction companies, furniture, computers and office equipment for offices, work clothing, etc. The proportion of this "professional" consumption transiting through retail trade (excluding services) therefore increased from 30 % of turnover in 1978-1979to 47 % in 2012[Grimmeau and Wayens, 2016. The structure of consumption has therefore changed significantly in Belgium. While we generally keep an eye on household consumption statistics, we should not neglect the potential for retail trade with respect to expenditure in a professional context, particularly in Brussels, which is an important centre of activity.
36 In addition to the structure of consumption, i.e. what households spend their money on, it is also necessary to take into account the evolution in the means by which they spend their money. We are obviously referring to the rise in e-commerce. According to Comeos estimates, in 2018 this represented almost 12 % of retail turnover in Belgium, more than half of which was spent on platforms based outside Belgium. In 2019, nearly 70 % of Belgian consumers had used e-commerce in one way or another in the previous year [Postnord and Nepa, 2019]. Due to the socio-demographic structure of the population in the region [Beckers et al. 2018], and probably also to the density of the commercial offer in Brussels, the use of e-commerce is less frequent in Brussels compared to the Belgian average. That said, Eurostat estimated that in 2018, the proportion of online shoppers in Brussels (regular or occasional) was 58 %. In 2017, a survey of private consumers in Brussels, which took into account only the purchase of goods excluding services of a commercial nature, estimated the market share in value of e-commerce at 2,2 %, mainly for small relatively common purchases . 37 The development of electronic commerce obviously creates competition for traditional commerce. But the situation is actually not so simple. First of all, some of the stakeholders in e-commerce are merchants who also have physical points of sale. Secondly, in Belgium, 48 % of consumers surveyed said that they find information on the internet before making a purchase in a shop, and 26 % find information in a shop before making an online purchase [De Vuyst et al., 2019]. The issue of e-commerce is therefore not limited to online sales, but must be understood as the link between an online presence and a physical point of sale. The role of shops is evolving differently according to sectors of activity and location.

The commercial offer is evolving
38 Between 1950 and 2017, the number of active points of sale in the Brussels-Capital Region decreased from 42 712 to 20 696 shops. This is a decrease of more than 50 % in the commercial offer in 67 years, i.e. an average annual decrease of 0,8 %. However, this decrease is slowing down, at least in relative terms. There was an average annual decline rate of 1,3 % between 1950 and 1997, and only 0,4 % between 1997 and 2017. Measured in terms of the number of points of sale, the commercial offer in the Brussels-Capital Region is therefore trending downwards, but the decline is slowing down [Grimmeau et al., 2007;Vazquez Parras, 2017]. 39 The decrease in commercial density is more marked in the Pentagon and the inner ring. The commercial offer is shrinking on the major roads. In the rest of the Region, there is a progressive individualisation of the commercial centres and a widespread disappearance of retail in the interstitial spaces.
Retail trade in Brussels: reconciling the city with a changing sector Brussels Studies , Notes de synthèse . Some roads and avenues have seen an increase in the number of shops (Avenue Louise, Chaussée de Waterloo…). With continued urbanisation beyond the regional boundaries, much of the recent commercial development has taken place in the outskirts. However, some densification processes can also be observed in the central part of the city, either through former property developments (such as the construction of galleries, mainly in the upper part of the city during the post-war period) or more recent ones (creation of City2, renovation of the South Station).
41 If we focus on recent developments and extend the observation area to the Brussels suburbs, we can see that the reduction in the number of points of sale now also affects the rest of the metropolitan area, even if new local commercial facilities may reflect the continued development of the residential outskirts, which are generally far away.
42 As regards the sale of goods in the strict sense of the term, there were fewer points of sale in 2019 than in 2009, but on the other hand, the number of square metres of sales area had increased. It can therefore be clearly stated that the decrease in retail trade is a reality in terms of the number of establishments where goods are sold and not in terms of the commercial surface area used for this activity. With the growth in sales area taking place in a context of a reduction in the number of points of sale, the Retail trade in Brussels: reconciling the city with a changing sector Brussels Studies , Notes de synthèse average size of the points of sale is increasing in almost all business sectors. Although it can also be observed in dense urban areas, this growth in average shop size is more noticeable in the outskirts, where it is less affected by the inadequacy of traditional buildings or land parcels and competition for land from other functions. 43 The net sales area balance, which was positive between 2009 and 2019, shows very different trends depending on the type of point of sale and the way in which it is operated. Organised retail, which includes branches and franchises, is growing sharply, while independent retail in the strict sense is in sharp decline. This is the result of the reorganisation of retail trade: the degree of economic concentration in the sector is increasing. The growth in surface area is concentrated in organised retail, either because the net growth is almost exclusively driven by it in certain sectors or because it replaces independent retail in other sectors. The phenomenon of economic concentration can also be observed in the field of commercial services, but to a much more limited extent [Wayens et al., 2020].  regarding the disappearance of commercial activity, due to the fact that, as a largely consumer-led activity, it is less easily relocatable than others. Rather, the concern is about the extent of collateral damage which could be associated with a rapid reorganisation of the sector. 47 The first observation is an accelerated decline in the performance of physical retail. Turnover in retail trade in Belgium has stagnated since the beginning of the 2000s and is now only sustained by population growth, while turnover per capita has been falling sharply for the past 20 years. At the same time, the growth in surface area continues, although it has slowed down sharply over the past two decades. As the growth in net sales areas is faster than the growth in consumption via retail trade, turnover per square metre has been falling faster in recent years. As a result, the performance per square metre is now lower than in the past [Grimmeau and Wayens, 2016;Wayens et al. 2020].
48 In Brussels, demographic dynamics have probably compensated for the decrease in individual consumption. However, the sustained population growth observed over the last 20 years will not necessarily continue at the same pace. Moreover, the drop in performance per square metre may lead to several unfavourable developments in the Brussels Region. If rents do not adapt (soon enough) to the decrease in performance per square metre, property pressure may accelerate the weakening of retail, whose costs are becoming too high. This may lead to an increased tendency to reduce other operating costs in order to compensate, which is obviously not good news as regards employment, either in number or quality. One response to the decrease in performance per square metre is a sort of headlong rush: having larger shops in order to maintain net profits. In an urban environment, this is often more difficult, particularly as it requires expanding into the interior of a block or regrouping, and can compete with other functions which are more profitable for owners. There is therefore a risk that this will contribute to the attractiveness of peripheral areas, where property is cheaper and less "complicated" in terms of implementation.
49 A second observation is the low profit margins. In Belgium, the net profit margin for the retail sector in the strict sense is now only 1,5 % of turnover [Konings and Vanormelingen, 2013]. As a comparison, in the early 1950s it was estimated to be between 9 % and 16 % depending on the sector of activity [Evalenko and Michel, 1958]. The combination of declining performance and declining margins is probably related to the increase in bankruptcies in this sector. Even if the number of active businesses is not declining due to a large number of start-ups, the increase in the bankruptcy rate reflects greater financial fragility. 50 The third observation relates to the extreme versatility of retail trade: the points of sale which disappear due to a change in the type of business, a conversion into empty units, housing or offices, or the demolition of buildings, represent between 4 % and 9 % of initial businesses annually depending on the location [Grimmeau and Wayens, 2016]. This is obviously much more than the average annual rate of decrease in the number of businesses, which in Brussels was approximately -0,4 % between 1997 and 2017, as there have also been start-ups in addition to these bankruptcies, whether it is because one commercial activity replaced another or because a space which was not occupied by a business became a point of sale. There is thus some versatility within the commercial units, as well as between commercial and non-commercial space: between 1997 and 2017 in the Brussels-Capital Region, 7 227 land parcels lost their commercial function Retail trade in Brussels: reconciling the city with a changing sector Brussels Studies , Notes de synthèse and 4 501 were newly occupied by a business [Vazquez Parras et al., 2018]. The turbulence observed in the landscape can also be seen with respect to companies: in Brussels, 9 % of active companies have been removed from VAT registers, which is more or less compensated by the same rate of start-ups. With a rate of 18 % turbulence (start-ups and closures) at the level of VAT registers in 2018, there has also been a high degree of versatility among retail businesses with their head office in Brussels.
Whether at the level of the point of sale or at company level, versatility is increasing and the average lifespan of businesses is decreasing. 51 The versatility of retail helps to understand that a vacancy rate of 5 % to 10 % for commercial units is considered normal, as it is associated with rotation. Versatility and the associated frictional vacancy allow the commercial offer to adapt to the evolution of the company and to the local market and business environment. Today, however, empty units are not filled easily, and even if some of them have been converted and have left the commercial real estate sphere, the rate of empty units is increasing and is well over 10 % Wayens et al., 2020].

The public regulation of commercial establishments is showing its limitations
52 In 2006, in order to promote free competition, the European "service directive", known as the "Bolkestein directive", called into question the criteria used until then in Belgium for the authorisation of a project: the impact on employment and existing retail trade or the interest of consumers can no longer justify a refusal to grant a site permit. Initially, the criteria were simply adapted, with the requirement that the decision had to be motivated by consumer protection, compliance with social and labour legislation, and protection of the urban environment. This last point already falls within the remit of the regional authorities. The transfer of jurisdiction to the Regions was implemented as part of the sixth state reform [Grimmeau and Wayens, 2016].
53 A May 2014 decree modified the Brussels Code of Urban Planning (CoBAT) by specifying that projects with a net commercial surface area of more than 400 m² represented an additional reason for obtaining planning permission. The question of commercial establishment was therefore integrated into general urban planning legislation in a very strict manner, as even a change in the activity carried out would require a permit. This became more flexible in 2017, with the removal of the requirement to obtain prior planning permission for a change in commercial activity in a building already assigned to this function, and with an increase in the surface thresholds requiring a report or impact study [Di Giaccomo, 2018]. 54 There are therefore three trends in the public regulation of commercial establishments: an easing of constraints for applicants, as seen in particular in application thresholds, evaluation criteria and reduced decision times, the shift from a logic of socio-economic regulation to one of urban and environmental regulation, the transfer of part of the regulation (via legislation, as well as its application) from the federal to the local level. In a federal context, this last point resonates in particular with the Brussels Region, whose morphological and functional reality goes beyond its narrow administrative boundaries. Many commercial projects are emerging not far from administrative borders, in increasingly different legal frameworks. In the absence of coordination, this leads to problems of interregional competition. 56 Moreover, while merchants are tending to increase their surface area in order to increase their market share, with stable demand, this also contributes to the decline in the overall performance per square metre. The lack of a possibility for direct intervention by the public authorities in order to limit this increase in surface areas therefore does not allow a form of common interest -even sectoral -to moderate individual wishes.
57 In addition, the regulation of retail trade by urban planning regulations requires the adaptation and updating of urban planning tools. In Brussels, the Regional Land Use Plan (PRAS) is the main urban planning framework, together with CoBAT. The PRAS included -in a rather innovative way at the time of its introduction in 1997 -specific requirements regulating commercial development. However, an analysis of the evolution of the commercial offer in Brussels over the last 20 years unfortunately does not show that the PRAS, which nevertheless introduced an innovative tool in terms of commercial urban planning -"commercial edges" 3 -has achieved its objectives in the area of retail. 58 The PRAS borders are undeniably areas where retail trade has been maintained. Although there has been an overall reduction in the commercial offer in Brussels, there has only been a slight decrease in the number of workers in areas within the borders of legal commercial centres (-0,1 % per year between 1997 and 2017). However, it must be noted that their role in terms of supervision and regulation has been limited, as the borders were modelled in 1997 on the areas which had already remained the densest at the time, and as there has been a significant increase in the footprint of shops outside the borders since then, with many points of sale of all types having been set up outside the borders since 1997. This can be understood by looking at the plot structure in the centres. The average size of the plots within the borders is smaller than in other areas of the region. In other words, it is difficult for big shops to find room there. And as there are many mechanisms to overcome the PRAS requirements, it is relatively easy for a business requiring large surfaces to consider setting up outside a border. The borders have therefore hardly contributed to maintaining an offer specific to the commercial centres [Vazquez Parras et al.2018]. This observation is confirmed by an analysis of building permits, with 87 % of the authorised new retail space located in places where there was no retail in 1997 [Grimmeau and Wayens, 2016].
Retail trade in Brussels: reconciling the city with a changing sector Brussels Studies , Notes de synthèse 3. The challenges: supporting transitions, reconciling the city with retail trade 3.1. Making the right diagnosis: not a decline, but a major reorganisation "Today, I sell a much wider range of products. Now there is competition. 30 years ago, people used to go to an organic shop to buy organic food. Now we have competition everywhere: in pharmacies, in supermarkets, in shops which are bigger than mine, in mini-markets, etc." (A shopkeeper from Brussels) multiplied, a range of paid services must be used increasingly for management and payments, and rents have increased [Grimmeau and Wayens, 2016]. Net profit has fallen and the turnover required to be profitable has increased, which means that the self-employed are under a high amount of pressure, given the greater difficulty they have in achieving economies of scale. 61 In order to understand the decrease in the number of shops and the increase in empty units, which is a corollary of this, the reasons for closure at individual level do not necessarily need to be questioned. Rather, it is necessary to ask why the shops which close are not taken over (or the [small] empty commercial units taken over by other commercial activities): productivity increases, economic concentration, unsuitability of smaller units for current operations and output, lack of new occupants due to competition on the labour market, etc. and, above all, a fall in demand for property in terms of the number of units (not in terms of square metres) due to the increase in the average size of points of sale. All of this without neglecting the effects of the end of growth in marketable expenditure per capita and therefore the fact that since the 2000s, turnover growth has been driven solely by population growth. Where demographics are not very dynamic, retail trade is threatened with decline [Grimmeau and Wayens, 2016]. In this respect, retail trade in Brussels can be thankful for the demographic boom in the Region over the past two decades.
62 Due to e-commerce, a growing proportion of purchases no longer transit through points of sale. This therefore contributes to the decrease in turnover from physical retail. But it also increases operating costs: managing a high-performance website and optimising logistics for online sales is easier for (new) stakeholders who are active exclusively online (so-called "pure players") and, to a certain extent, for organised retail. The digitisation of retail trade, which also affects supplies, is more difficult to implement for small independent shopkeepers. And it is not clear whether digitisation is cost-effective or appropriate for everyone [Gillis, 2014].
63 Retail trade is therefore not a decline which needs to be managed, but rather a major sectoral reorganisation which needs to be supported.

Brussels retail must take local demographics into account
"It's all well and good to say, 'You have to advertise, etc.,' but when people don't have money in their pockets, you can do whatever you want, they won't buy." (A shopkeeper in Brussels) 64 Retail trade is an induced activity, largely dependent on the market, whose determinants at regional and local level are, in particular, demographic structures and developments. 65 The link between retail trade and demographics is simply a matter of population density and its direct effect on retail density. The recent demographic growth of the Brussels Region is therefore favourable to commercial development, and over the last 20 years has compensated for the quantitative loss due to peri-urbanisation, which in turn has led to commercial facilities in the suburbs. 66 The size of the population must obviously be weighted by the level of household income. Wealthier people have a higher level of consumption. But income also influences the structure of expenditure, i.e. the proportion of different goods and services used by consumers. According to Engel's law [Houthakker, 1957], the share of Retail trade in Brussels: reconciling the city with a changing sector Brussels Studies , Notes de synthèse income allocated to food expenditure decreases when the income increases, even if food expenditure in absolute terms increases. The elasticity of food expenditure is therefore low, unlike other commercial goods. The level of income thus induces a variation in the proportion of different types of product (basic, comfort, luxury, etc.) in the local commercial offer.
67 Global and local changes in purchasing power are therefore an important factor in the health of retail trade. In terms of income, the situation in the centre of the metropolitan area is less favourable. In the long term, the reversal of population trends does not fully offset the decrease in the average income in the central part of the urban area. On the other hand, in the suburbs, especially the most remote ones, total purchasing power has increased. Differences in income levels and income trends are also significant within the Brussels Region. However, due to the greater reach of neighbourhoods in the centre of the Region and the presence of workers, the neighbourhoods have not experienced the effects of local demographic changes in the same way [Wayens and Grimmeau, 2009].
68 Age and household structure also generate specific consumption profiles. Areas characterised by a strong presence of young adults are particularly favourable to certain commercial dynamics: developed cultural and leisure practices (which is also favourable to HORECA), lower motorisation rate favouring the local economy, etc. Of course, this young population is also sensitive to new modes of consumption: online shopping, organic, short supply chains, ethical approach, use rather than purchase, etc. Finally, this young, flexible and diverse urban population can also contribute to commercial dynamics through the job market. This of course calls to mind the uberisation of retail (notably through deliveries), which, to a large extent, mobilises students and other young adults in transitional periods [Strale, 2016], as well as the specific dynamics of ethnic retail, a historical means of access to entrepreneurship for newcomers.
3.3. Brussels must harness customer flows or retain them 70 Given that today, the (local) growth in the number of customers and not just the increase in average spending per customer is contributing to the growth of the retail market, the question of capturing non-resident customer flows has become crucial. From this point of view, tourism is one of the ways of broadening the economic base of retail trade. 71 The tourism sector plays a crucial role in economic growth and the creation of jobs in the city which are unlikely to be outsourced. While it offers new opportunities for the Retail trade in Brussels: reconciling the city with a changing sector Brussels Studies , Notes de synthèse creative industries and the enhancement of cultural heritage, its local impact is concentrated above all in two sectors of activity: accommodation and retail (restaurants and cafés, as well as the sale of goods and services). The recreational turn [Stock, 2007] is now affecting large multifunctional cities where tourism has long been one of several economic activities. Tourism is becoming a bigger determining factor in urbanism and urban attractiveness. In this context, the spaces in large cities such as Brussels are organised more and more by and for tourists. Retail trade plays a major role in this reorganisation, either passively (by evolving towards an adapted offer) or actively (by making retail one of the elements of tourist attraction, in the same way as events and major conference, cultural and leisure facilities). 72 The proportion of commercial activity which is generated directly by tourism varies according to sector (it is not the same for a souvenir shop as for a DIY shop) as well as according to geographical location, which is a very important factor in a big city like Brussels (the proportion is higher in some neighbourhoods) Wayens and Grimmeau, 2006]. 73 In Brussels, the materialisation of the recreational turn in the area of retail trade is seen in particular in two major urban projects, which have given rise to concern and resistance among inhabitants, in particular due to nuisance caused by festivities in the city. In the city centre, the redevelopment of the boulevards is in keeping with the objective to redirect the commercial offer in order to reach tourists in particular, which has led to legitimate concerns regarding the offer aimed at meeting the demands of the inhabitants. On the Heysel plateau, another major centre for tourism and leisure activities, the NEO project will have an important commercial component. It should reap the benefits of the conference centre and leisure complex traffic, as well as ensure the stability of property development by financing public facilities. Retail trade should also contribute to the overall attractiveness of the site. Again, these choices are a source of concern and contention. Some people are beginning to oppose the reorganisation of space in favour of tourists, who, in their view, threaten the viability, proper functioning and sustainability of the city for its inhabitants and other users [Hubert et al., 2017;Fenton et al., 2020].
74 It has to be said that today, there is too much of a separation between the management of tourism policy and the management of retail trade policy. This separation is conceivable if we consider retail trade as an induced activity, but it is precisely time to stop viewing retail trade as a mere beneficiary of tourism, and instead consider it as a contributor to the urban tourism ecosystem.
75 Tourism can be strongly impacted by geopolitical events. The terrorist attacks (2015, 2016) had a relatively short yet significant impact on the turnover of shops which depend on tourism. This has contributed to the increased volatility of retail businesses, as not all shopkeepers have the cash flow to withstand this type of crisis for long periods of time [Vanneste et al., 2016]. Similarly, the Covid-19 health crisis in 2020 and its immediate repercussions in terms of tourist travel will have a pronounced impact in the long term on businesses which depend more widely on supra-local demand.
76 Having these reserve funds therefore implies harnessing flows which are potentially more stable than those generated by tourism. This involves in particular limiting shopping outside the region. It is of course unrealistic to try to prevent Brussels residents from shopping outside the regional boundaries. Fortunately, despite 50 years of retail trade development in the outskirts, only about 13 % of consumption by the Retail trade in Brussels: reconciling the city with a changing sector Brussels Studies , Notes de synthèse inhabitants of the Brussels-Capital Region takes place in points of sale outside the Region. In fact, half of this consumption takes place in the province of Flemish Brabant, more precisely in the immediate outskirts [Vazquez Parras et al., 2019b]. Conversely, little is known about the consumption habits of commuters in Brussels, but this client segment probably contributes significantly to the vitality of certain segments of the Brussels retail sector. Perhaps more attention should be paid to this segment, as it represents a significant potential [Ermans et al., 2019] which is relatively predictable and regular, with its presence in Brussels for work.
3.4. E-commerce and the city: not just a question of logistics 77 In addition to purchases made in "physical" shops, the shift in purchasing power takes place through expenditure in other distribution channels. More and more, online shopping is becoming an alternative. The majority of online spending is on relatively common, relatively small (facilitating transport) and non-perishable (facilitating inventory management) products [Vazquez Parras et al., 2019b]. It is difficult to identify the location of these flows on an inter-regional scale in Belgium, but, according to Comeos, more than half of online spending by Belgian consumers benefits operators based abroad.
78 A part of retail trade operates increasingly on a multi-channel basis, whereby consumers and merchants use both online and offline channels to organise purchases. The result is an increase in the number of parcels being sent to consumers' homes, putting a strain on urban freight systems in terms of efficiency and sustainability. Nearly 45 000 e-commerce deliveries were made daily in Brussels in 2017, with operators expecting growth in the order of 20-30 % per year for the following years [Buldeo Rai et al., 2019]. In order to limit congestion, the environmental impact and the economic cost which accompany this increasingly individualised flow of goods [Cárdenas et al., 2017], the phenomenon must be organised or regulated. However, as the Brussels market is primarily a Belgian market in terms of the framework for commercial practices, and as operators are generally active in all three regions, action must ideally be taken in a coordinated manner.
79 More broadly, the issue of deliveries should not become a purely logistical and local issue, at the risk of developing an approach which would leave existing retail trade out of the picture (or worse, of developing unfavourable solutions). Physical retail could benefit from improvements in terms of delivery and (re-) establish a place and role in the new organisation of deliveries [Strale and Wayens, 2013;Strale et al. 2015]. In particular, this could involve a qualitative improvement and an extension of the services offered by the pick-up points in order to limit delivery failures to customers' homes, or a better link between storage and a local supply of goods in high demand, supplemented by an off-site supply with relatively rapid delivery [Delaporte et al., 2016]. Inevitably, this would also have an impact on the use of commercial property and on working time, which would have to be taken into account in regulations [Van Ossel and Devoldere, 2017]. Furthermore, training in e-commerce or, more precisely, in managing the presence of online merchants is a major issue, which is particularly difficult to take into account given the diversity of merchants and the other stakeholders involved.
Retail trade in Brussels: reconciling the city with a changing sector Brussels Studies , Notes de synthèse 3.5. The environmental transition will not happen without retail trade "Today's problem is the boxes: they're supposed to come through this morning, it poured last night, the boxes are soaking wet so we can't get them in because they're so wet, and no lorry has come yet. This street is disgusting." (A shopkeeper in Brussels) 80 With their high density in terms of housing, employment and leisure, cities are first and foremost consumption areas. At a time when Brussels is adopting an ambitious economic and climate transition policy, this is far from insignificant. Obviously, for the public authorities, it is easier to work with major stakeholders (the energy or waste sectors, for example) or with emerging and dynamic stakeholders, whose numbers are still small (the urban agriculture sector), than it is to work with the commercial sector, characterised by a multitude of stakeholders. However, with more than 20 000 establishments in the region (and 1,3 million square metres just for the sale of goods in the strict sense), retail trade is an important, if not essential, lever for any change in consumption in Brussels, and therefore for its economy and environmental impact.
81 In addition to the logistical function of retail trade -which has an inexorable impact on the environmental performance of the Region, whether through transport flows or the energy performance of points of sale and warehouses -the various functions of retail trade (see point 1.1) are levers to be put to (better) use. The assortment function determines the place given to different products in the basket of household goods. The evolution of the structure of consumption, often seen as a form of consumer freedom, is also induced by the information and advertising provided by producers, as well as by merchants, who have a true power of prescription, or at least, of direction or popularisation. For example, the development of the organic food offer obviously corresponds to the expectations of some consumers. However, it has also become much more accessible, thus attracting more consumers. While retailers play a role in changing consumer behaviour, the public authorities can also encourage or accelerate certain processes, or even help to make them more widespread, as we have seen with the gradual ban on non-reusable plastic bags in the retail sector. 4 82 In the context of developing an economic and environmental policy more marked by circularity (or at least by an extension of the period of use of goods), we should not fall into the trap of encouraging new emerging stakeholders or innovative solutions while forgetting reuse or recovery practices which already exist in the retail sector. These include commercial crafts (including the repair and adaptation of goods) and the second-hand sector. Although these sectors are promoted, they are nevertheless fragile. The development of e-commerce and the collaborative economy also affects second-hand goods, sometimes significantly reducing or degrading the pool of goods available to the historical stakeholders, without even taking into account the intrinsic decline in quality of new products. 5 Moreover, there is sometimes a lack of new craftsmen, due to the lack of a proactive training policy, but also because of working conditions considered too difficult by today's standards [Mahieu and Godin, 2014]. 83 With a view to a more coherent economy in terms of material flows, the question of waste prevention and management in the 20 000 shops in Brussels should be the subject of further reflection. Although they are often associated spatially with the residential function, shops have specific needs. And the solutions proposed for (big) companies in Retail trade in Brussels: reconciling the city with a changing sector Brussels Studies , Notes de synthèse general are not necessarily adapted to urban commercial centres. In environmental terms, as well as in terms of the image of the shopping district, there is room for substantial progress in waste prevention, reduction and collection practices [Godart and Strale, 2019].
3.6. Commercial property must adapt to the needs of retail, not the other way round 84 In a context of stagnating demand and retail difficulties, it is legitimate to question the forces behind the growth in surface areas. Admittedly, part of this is due to the demand from merchants who are trying to compensate for the drop in performance per square metre by increasing the number of square metres or by optimising their logistics operations.
85 But the fact that new retail space is now a product of property development should not be underestimated. Developers active in commercial property, who are increasingly disconnected from the world of retail in the strict sense, are building shopping centres and business parks with equity and bank loans. They have seen their activity boosted by low interest rates and the mass of capital which has to be invested defensively, with little risk; an activity which is all the more abundant, as property offers a better return than other low-risk investments. For several decades, developers have benefited from a considerable increase in the value of commercial property. Investment funds have enjoyed both the high returns from retail rents and the fact that walls act as a guarantee of solvency [Madry, 2011]. 86 Obviously, this whole system is based on the growth or maintenance of the mass of commercial rents which can be collected. The acceleration of the decline in performance per square metre and the end of the growth in expenditure in physical retail are calling into question the foundations of this system. In other words, not all new developments necessarily find a buyer, and if they do, not necessarily with the expected level of rent and therefore property return [Madry, 2011;Ancion and Assouad, 2018].
87 This development may undermine some public/private partnerships. Some commercial property developments are in keeping with the interests of local authorities, as they can add value to a brownfield site, compete with a neighbouring commercial development or provide financing within a more global project for an unprofitable infrastructure wanted by local authorities lacking in resources (e.g. a stadium or conference centre). But this financing scheme, which was common in the first decade of the 2000s, is probably much less realistic today [Madry, 2012]. 88 It is clear that the disconnection between property logic and commercial and consumption logic accentuates the gap between the growth of retail space and needs. It also reflects a growing inability of all stakeholders (shopkeepers, investors and public authorities) to (self-) regulate. In this context, public authorities must (once again) become fully involved in commercial urban planning, encouraging an intensive (rather than extensive) mode of growth of retail trade. They must provide a framework for commercial property development based on objective needs: the property offer must be justified by consumption, and not the other way round. Finally, it is necessary to avoid contributing to capital gains from property by selling off public land and making property values a simple adjustment variable in an opportunistic approach [Madry, 2011]. 89 As socioeconomic regulation has been abandoned in Brussels due to a need for coherence with European legislation (see point 2.5), urban planning legislation is now largely responsible for regulating retail trade. However, it is not clear that the production of commercial space can, as it stands, be regulated effectively through urban planning regulations [Mérenne-Schoumaker, 2010;Vazquez Parras et al., 2018], even though property production is the result of economic logic which is increasingly disconnected from the regions, whether through the growing economic integration of distribution stakeholders or through the development of new consumer interfaces which are beyond its reach (e-commerce, mobile pop-up shop applications, short supply chains, etc.) [Madry, 2012;Desse et al., 2016].

Acting on mobility is acting on retail trade
"A pedestrianised area is made with the merchants who are already there -it can't be made only with the big international chains." (A shopkeeper from Brussels) 90 The loss of accessibility by car is a recurring fear on the part of merchants. Retail representatives often associate the increase in the attractiveness of retail with the maintenance of the number of parking spaces and a reduction in the cost of parking.
91 This concern is partly well-founded, although it is also clear that merchants often overestimate the importance of motorists in their customer base. For the most distant customers, i.e. those in the immediate outskirts who are highly dependent on cars [Strale, 2019], it is more a question of access time and traffic than of parking. In the city centre (the Pentagon or the upper part of the city), the public car parks, for which there is a fee, are rarely full. These car parks are often run by private operators, and are too often seen as a source of income (for public stakeholders as well) and not as a way to manage mobility (in particular by relieving the pressure of parking in public spaces). In the rest of the Region, with the widespread standardisation of paid parking in the streets, the initial role of parking meters, i.e. to ensure the rotation of vehicles, has been forgotten, particularly in retail areas. Consequently, it is probably legitimate that merchants feel that they benefit less and less from parking policies. 92 The use of cars for shopping will not disappear in the short term. This is not necessarily desirable or possible everywhere. The question of the accessibility of shops therefore deserves some constructive thought. It should be based on a systematic objectification of the mobility mix necessary for the proper commercial functioning of each area. This reflection should not be limited to parking and should integrate (the perception of) the accessibility of the centre by all modes, taking into account the catchment area and reach of the commercial centre. Depending on the centres concerned, this could lead to a panel of recommendations and stakeholders, ranging from very local measures (improving pedestrian accessibility, parking rotation, creation of delivery areas, etc.) to metropolitan issues (role of SNCB and park and ride facilities in commercial traffic, etc.). This would contribute to developing a mobility policy which is less focused on home-to-work travel [Hubert et al., 2013]. Reflection on the mobility mix must also include the issue of deliveries, including those related to e-commerce. Many points of sale are in the process of becoming pick-up points, which has an impact on the number or frequency of customer visits and deliveries [Libeskind, 2015]. As with waste, e-Retail trade in Brussels: reconciling the city with a changing sector Brussels Studies , Notes de synthèse commerce is likely to increase outgoing flows from shops, whereas today, ingoing flows are mainly taken into account. 93 The issue of the accessibility of shops has been brought to the fore by the expansion of the pedestrian area in the city centre in 2015. However, the pedestrianisation of shopping areas began in the 1970s, and the impact of these transformations on the functioning of retail trade has been the subject of numerous studies [Mérenne-Schoumaker, 2008], which have revealed the common features of these developments: an increase in changes of unit operators; an increase in the proportion of integrated retail and a decrease in the proportion of selfemployed shopkeepers; a decrease in the supply of commonplace products and a reorientation towards relatively common or special products, as well as towards leisure products; the strengthening of a range focused on mass consumption, high profit margin and a low presence of luxury retail; a virtual disappearance of activities dependent on direct car access; a development of HORECA, which benefits from the possibility of extending terraces. 94 The extension of the pedestrian area in the city centre as well as the drastic limitation of the presence of cars in Chaussée d'Ixelles have recently increased the areas where pedestrians have priority, to an extent not seen since the advent of the car. Although it is not possible to examine the evolution of commercial structures in retrospect, we can see changes in the offer in these areas which are consistent with the observations made concerning the previous wave of pedestrianisation, with the exception of a few details. The relative weight of the restaurant sector is likely to be much greater, as it is both a growing sector in the most central urban areas and one of the major beneficiaries of the recreational turn or, more broadly, of changing lifestyles. The extended pedestrian area will probably also be less exclusionary for the food sector, although this will probably be to the benefit of the organised retail sector, which is developing new, adapted formats. On the other hand, it is not certain that this will contribute to the strengthening of the clothing retail (at least in terms of the number of points of sale), which is subject to the dual pressure of e-commerce and developments in the outskirts. It will probably be reluctant to increase the number of urban points of sale.
95 This last observation underlines the extent to which the dynamics of a pedestrian area cannot be isolated from its metropolitan context, as the small amount of recent literature shows that the success of such areas can be explained in part by the presence of a critical mass of residents (or tourists) in the city centre, the existence of an efficient public transport network and a low level of dependence on cars in the catchment area, as well as by the implementation of policies discouraging the development of shopping areas in the outskirts [Boussauw, 2016]. As all of these criteria are not necessarily met, this strongly limits the transferability of the pedestrianisation approach to other shopping streets in Brussels. 96 The works in Chaussée d'Ixelles, the widening of the pedestrian area and the future north metro also remind us how much the implementation of major infrastructure works -which disrupt the use of public space -is a potential source of inconvenience for retail trade [Grimmeau et al., 2004]. The diversity of the impacts according to the type of activity, as well as the fact that they can persist well after the end of the works, means that the logic of compensation -which, in Brussels, is subject to many conditions and requires an active administrative approach by the shopkeeper 6 -is Retail trade in Brussels: reconciling the city with a changing sector Brussels Studies , Notes de synthèse hardly a solution, however praiseworthy it may be in terms of the symbolic recognition of the nuisance, especially given the short lifespan of shops. The (low) compensation often comes too late.
97 It is much more important to work upstream with better planning of the worksites, and during the works, with stricter control of the process and the nuisance it causes.
Respecting the deadlines and timing of the different phases is an important element. Some shopkeepers may modify some aspects of their operations in order to adapt to the situation (moving their annual holidays, rescheduling deliveries, adapting inventory, etc.). This obviously requires clear information, permanent dialogue and guarantees with respect to phasing. It should be borne in mind that some useful support initiatives such as providing alternative car parks can become completely ineffective due to inadequate customer information [Durand and Gillet, 2017]. Moreover, the multiple renovations in Rue Neuve have shown the importance of a judicious choice of finishing materials, in order to avoid the inconvenience of a multitude of small worksites for carrying out repairs.
98 From a distance and from the perspective of the media, worksites and the difficulties they create may appear to be much more unfavourable than they are in reality. Communication focused on the practical aspects, specifying the hours and days when works cannot be carried out (Saturdays, for example), as well as on the future qualitative aspects of developments, can be useful in maintaining a welcoming image of the district. Finally, dramatisations of the most spectacular phases of the worksites are welcome. The huge hole in Avenue de la Toison d'Or in 2013, preceded by an installation by a well-known artist, attracted a lot of people. Clearly, all of these measures require the active involvement of often multiple owners and contractors. However, reflection on the stealth of construction sites is gradually gaining ground internationally [Kaufmann et al., 2017]. The gradual integration of this concept into the specifications and practices of site planning and monitoring in Brussels could be an example of a smart city particularly appreciated by shopkeepers and beneficial to the local economy.

Retail, an area of tension
"I have customers who say: 'Listen, I'm not buying anything today, but I need to tell you something!'" (A shopkeeper in Brussels) sometimes the only other function present -are neighbourhood quarrels. Inevitably, retail is a source of nuisance: deliveries, cooling and ventilation units, quantity of waste, customer flows, etc. Solutions must be found which suit the various stakeholders, but which maintain a balance between nuisance and necessity. We must obviously avoid an intensification of the NIMBY phenomenon (Not In My Back Yard) in the area of retail, at the risk of reinforcing peri-urbanisation. More positively, when consumers are also residents and not users from "outside" the neighbourhood, shops offer a service to them (which legitimises certain types of nuisance) and it is in the shopkeepers' interest to maintain good relations with their local environment, which is also their market. This is a subtle balance which is usually established over the long term, and can be upset when the nature and operation of activities change. In the context of an evolving offer in the neighbourhoods, it is thus necessary to be particularly attentive to the growing proportion of HORECA establishments and leisure activities as well as to the question of opening hours. In the event of a major change in the offer and temporalities of a neighbourhood, as has been observed at Saint-Géry, 7 Cimetière d'Ixelles, Place Flagey, Parvis de Saint-Gilles and Place Jourdan in Etterbeek, it is necessary to provide support and regulate. Initiatives such as the development of charters and the creation of night councils 8 can encourage self-regulation, but can only work if there is true awareness and careful monitoring by the authorities. And the question of the legitimacy of the nuisance should not be underestimated when commercial activities are clearly disconnected from the inhabitants, in particular because of their reach.
104 On the other hand, many shopkeepers deplore the increasing presence of homeless people and begging in shopping areas. The increase in homelessness is widespread in Brussels [Quittelier and Horvat, 2019] and is reinforced by the flows of migrants associated with political instability and many conflicts at international level. As long as care for these people is insufficient and limited to the night, shopping areas will remain, for many reasons (social controls, specific urban morphology, availability of boxes, pedestrian flows, etc.) havens for marginalised populations in the urban space [Malherbe and Rosa, 2017]. This is detrimental to retail operations in terms of image, but also sometimes in very concrete terms (difficult early morning deliveries, extra cleaning, for example). While some of them actively support solidarity initiatives (notably through effective management of unsold goods, now encouraged by legislation), retailers feel somewhat abandoned in this respect. A security-based response is obviously inappropriate given the scale of this social crisis, but it is high time to invest massively, including with shopkeepers, in a pragmatic and humane management of the issue of homelessness and marginality in the public space, Retail trade in Brussels: reconciling the city with a changing sector Brussels Studies , Notes de synthèse especially when the latter is in close connection with a semi-public space (actually accessible to the public) such as a commercial area.
105 Supporting shopkeepers in the management of negative externalities generated by social and migration crises is all the more justified as this sector itself plays a major role in integration, offering many jobs with a wide range of qualifications, and has long been a vector of economic integration for populations of immigrant origin. While shopkeepers are sometimes guilty of discrimination [Bourabain and Verhaeghe, 2019], it is reasonable to assume that in Brussels, they are just as much victims as perpetrators of this type of behaviour. Consequently, their demands regarding the management of public order should not automatically be considered as exclusionary behaviour, but should also be examined in terms of the constraints which these crises place on commercial activities. 106 In the conflictuality related to retail trade, within businesses and commercial spaces, the risk of increasing social conflicts associated with the restructuring and reorganisation of the sector should not be underestimated. The small size of businesses and the dispersion of points of sale still strongly limit the structuring and rise of social conflicts in retail trade, but convergence with other urban struggles or new forms of protest is emerging. In this respect, it should be noted that the first strikes by uberised bicycle delivery workers emerged in major European cities under the pressure of rapidly deteriorating pay conditions. The blocking of mass distribution logistics centres -structures which are much more concentrated and linked to the life of a company than the wholesale trade in the past -is also happening more and more. 107 We must also not forget the tensions encountered by self-employed merchants at individual level. Systemic constraints beyond their control and a status based on the notion of independence lead to an ambivalent work experience. The absence of subordination defines their legal and social status as well as their identity and gives them a true ability to take ownership of their work, unlike most employees. But this autonomous business operation is coupled with direct confrontation -not mediated by a third party employer -with the constraints of the market and its regulations. Their well-being is dependent on their ability to work within these constraints. Surveys show that the self-employed in the commercial sector deal with this tension in two ways. As customer relations are at the heart of their identity and status, professional training, tools and routines help to reduce uncertainty in most cases. This includes maintaining the right distance with respect to the customer, which allows them to have more control over their working and employment conditions. Moreover, true professional capacities and skills allow the self-employed person to have rewarding prospects. However, their ability to cope with their working conditions depends on the state of the labour market, in particular due to the fact that their identity is the result of the deliberate choice to be self-employed, which calls for extreme caution in the context of employment (or job placement) policies for fragile groups which place too much emphasis on an approach centred on entrepreneurship. In addition, the biographical accounts highlight the role of family or ethnic networks which the merchants belong to, and which are networks of solidarity and constraint [Mahieu and Godin, 2014;Mahieu and Godin, 2015;Debroux, 2018;.
108 In summary, the surveys revealed an ambivalent work experience: merchants value the independence of their activity and the identity associated with the ability to appropriate their work, but also recognise that their activity is highly dependent on an Retail trade in Brussels: reconciling the city with a changing sector Brussels Studies , Notes de synthèse environment of territorialised interdependencies, regulations and sociabilities over which they have little direct control. Due to the small size of the shops and a strong personalisation of the relationship with employees, many merchants exhibit a sense of isolation, which obviously does not help in the management of tensions of any kind. Nor with getting help. It is therefore absolutely necessary to maintain and develop training and support systems which take into account the specificities of shopkeepers.
3.9. There is a need to clarify the framework for public action in the area of retail trade "Too much help ruins the help! There is a lot, but it is far too diverse. It takes many hours to read about what you're entitled to and what you're not entitled to. That was the hardest part for me." (A Brussels shopkeeper) 109 The development of public policy is based on the representations which public stakeholders and social partners have of a given economic and socio-political context and the options available to them to respond to it. Traditionally, the discussions and pursuit of coherence are more concerned with the objectives, instruments and resources used in implementing policies than with the standards or goals which underlie them [Belley, 2014;Lascoumes and Le Galès, 2018].
110 As far as retail trade in Brussels is concerned, it should be noted that, at least since the founding of the Region, discussions have mainly been about instruments and have remained limited in terms of the standards and purposes assigned to retail trade. In the considerations of the Plan Régional de Développement Durable of 2018 or the 2025 strategy for Brussels, the induced and job-providing character of retail trade is predominant, and is often associated with a local role. On the other hand, the Brussels public discourse on retail trade does not include much rhetoric in the economic field. The cluster concept and its different variants such as company clusters, competitiveness clusters or local production systems are used regularly for other sectors, including in relatively vague terms (creative industries, productive activities, circular economy), but hardly ever for retail trade. Quite paradoxically, retail trade is not even present as a sector of activity highlighted in policies to strengthen the local economy (unlike productive activities or even the social economy sector). The concept of the presence economy [Davezies, 2009], based on the actual population present, is largely absent from the Brussels political framework documents, even though it is a theoretical reference system which is very relevant to retail trade.
productivity and associated training needs). In addition to the societal missions assigned to retail (enabling a neighbourhood city, ensuring employment) which tend to liken it to a facility, the objectives assigned to Brussels public policies in the area of retail trade should therefore also be based on a coordinated sectoral economic policy, with the aim to make it a profitable activity in the financial sense for retail trade (and therefore also in the fiscal sense for the public authorities).
116 Finally, the shift from an urban planning approach to a more entrepreneurial approach by the main regional public stakeholder in charge of retail trade will not automatically lead to a better consideration of retail trade as an economic stakeholder. Atrium.brussels was slowly but surely becoming an indispensable Brussels platform for sectoral consultation, both because of the dispersion of stakeholders in the retail sector (UCM, UNIZO, Comeos, various HORECA federations, etc.) and the multiplicity of public stakeholders involved (region and municipalities, urbanism, mobility, economy, employment and environment). Hub.brussels will probably need to reassert its role as a coordinator in the sectoral landscape, as well as its visibility. This was clearly the intention of the draft commercial development plan released in 2018. This plan Retail trade in Brussels: reconciling the city with a changing sector Brussels Studies , Notes de synthèse resolutely abandons the urban planning approach to propose a way to deal with 17 identified issues so that retail trade in Brussels is better taken into account in the various regional policies, so that retail trade is better integrated into the urban environment, so that its regulatory framework is more coherent and so that the satisfaction of users and workers in the retail sector is maximised. 117 The visibility of this coordination may be achieved through the implementation of a regional platform specifically dedicated to retail trade (or at least to the local economy), equipped with real means of coordination and capable of supporting stakeholders through a programming of actions, as the Plan Régional en Économie Circulaire (PREC) has done at Brussels level.
118 The evolution of regional policy in the area of support for retail trade has gradually moved away from more local and collective approaches. The 106 associations of shopkeepers active in the Brussels-Capital Region probably felt a certain emptiness following the "regionalisation" of atrium.brussels activities in 2012 and the recent change in the positioning of hub.brussels.
119 With one association for 226 shopkeepers, the Region seems to be fertile ground for these associations. But this also suggests that they are positioned on a very local scale, sometimes being active in only part of a neighbourhood [Dadic et al., 2017]. By including municipal stakeholders, we can see both the potential of field work which takes into account local realities, as well as the risk of an excessive fragmentation of local action and associated resources. There is therefore probably a need to reorganise the network of local stakeholders, as well as to strengthen the role of municipalities as a public stakeholder supporting retail trade, in cooperation and not in competition with the regional authorities. This will probably require a greater professionalisation of the civil servants assigned to this task, as well as a professionalisation of the work of associations which are too dependent on a limited number of volunteers acting as relays vis-à-vis the authorities. Two trends can be encouraged in this respect: A clarification of the missions of the shopkeepers' association (or co-operative): local activities and representation, as well as coordination and co-management of certain services useful to shopkeepers and the public. This clarification must be combined with an explanation of the methods of operation and financing of co-operatives which can benefit from public support. It should be noted that in some areas there has also been a shift in these associations towards public-private partnership formats at the local level 9 ; A diversification of the stakeholders involved, which, in a local development perspective, should probably also better integrate consumers, inhabitants and local civil society stakeholders.
120 In doing so, a local level of coordination could work on the daily management of commercial spaces to ensure that they function properly on a day-to-day basis, not only for the benefit of shopkeepers, but also with the aim to improve the quality of life of the inhabitants, for example by better prevention of neighbourhood conflicts, or by facilitating the intervention of the public authorities, which would then have a clearly identified interlocutor. Obviously, nothing prevents the public authorities from setting up this type of management themselves, but they must then move from operating in sectoral silos (economy, waste management, security, etc.) throughout the region, to a cross-cutting operation for a specific part of the region. In addition, care should be taken to establish a daily dialogue [Hubert et al., 2017]. This synopsis makes the observation that retail trade is an important economic sector in Brussels, which is confronted with profound changes in its organisation and environment. Instead of disappearing, retail trade is changing and reorganising itself, demonstrating spectacular adaptability and resilience. In doing so, it continues to fulfil its role as an interface in the urban ecosystem, even if it has meant a forced-march adaptation, which generates turbulence and raises many economic, social, environmental and urban planning issues.
122 Often considered simply as an induced activity even when part of the offer contributes to the city's reach, retail trade has also long been treated as a facility and not as an economic sector in its own right. At a time when regulatory frameworks for public action are evolving rapidly, its hybrid nature should be acknowledged: it is an economic activity which must be profitable, but whose profitability allows it to provide many services, including at the margin of the commercial sphere. While the environment of the point of sale is the essential basis for its viability, retail trade is becoming increasingly integrated into a more global economic ecosystem (through relations with suppliers, the economic concentration of stakeholders, the development of e-commerce, etc.).
123 Retail trade therefore has a special status. It is an essential facility for the smooth running of the city, made up of a very large number of operators who must ensure their profitability without direct public intervention. The Covid-19 crisis has highlighted the fragility of a sector which generally has few financial reserves and specific public aid, as well as the importance of local and online retail, whether independent or organised, for the confined inhabitants. Merchants are totally independent and financially responsible for their future, as well as being extremely dependent on the positive and negative externalities of their environment, over which they have little direct control.
124 For public stakeholders, recognising this fact means abandoning the utopian wish to plan retail trade directly (indicating which businesses or forms of retail should be present) in favour of planning which considers the needs of trade (ensuring a hospitable environment for the different types and modes of retail, arbitrating and avoiding conflicts of use, etc.). This approach is better adapted to the economic reality of the sector, the rapid evolution of its organisation and consumption patterns and the needs of the city, and does not at all advocate disinvestment by the public authorities in this area. Rather, it argues for more cross-cutting involvement, with better coordination between levels of power. It also has the virtue of raising the question of responsibilities: are the various public authorities responsible for the commercial offer, or for the setting (economic, social, urban planning, environmental) in which retail trade evolves?
9. This is already the case in the city centre with Shopera, a private initiative of Comeos, BECI and UCM.

ABSTRACTS
In 2017, the Brussels-Capital Region 20 700 points of sale for goods and services, accounting for 87 % of the expenditure made by the inhabitants of Brussels. This retail trade provides about 9 % of regional employment. This synopsis reviews what is known about distribution in Brussels, an indispensable economic sector whose complexity makes it difficult to understand due to its diversity, and which is subject to profound changes in consumption, the urban environment and the internal organisation of the profession. After outlining the framework of the analysis by defining trade through its functions, forms and stakeholders, the authors draw up several observations associated with major dynamics in the sector (transformation of supply and demand and changes in public regulation in particular). The third part of the synopsis responds to these observations by identifying clear issues which, if taken into account by the sector and supported by public authorities, could reconcile trade with urban space.
In 2017 telde het Brussels Hoofdstedelijk Gewest 20 700 verkooppunten van goederen en diensten, via dewelke de Brusselaars 87 % van hun uitgaven besteden. De detailhandel vertegenwoordigt ongeveer 9 % van de werkgelegenheid in het gewest. Deze synthesenota geeft een overzicht van de kennis over de handel in Brussel, een essentiële economische sector die echter moeilijk te vatten is wegens de diversiteit. Bovendien wordt deze sector geconfronteerd met ingrijpende veranderingen in de consumptie, de stadsomgeving en de interne organisatie van het beroep. Na een toelichting van het analysekader, waarbij de handel wordt gedefinieerd aan de hand van de functies, vormen en spelers, komen de auteurs tot een aantal bevindingen die verband houden met belangrijke dynamieken in de sector (verandering van aanbod en vraag, ontwikkeling van de overheidsregulering...). Het derde deel van de nota biedt een antwoord hierop door te bepalen met welke grote uitdagingen de sector, met de steun van de overheid, rekening zal moeten houden om handel en stad weer met elkaar te kunnen verzoenen.   His research focuses on the location of activities and metropolitan mobility issues. As regards retail trade in particular, he analyses supply chains and the interaction between urban development and transport policies and access to retail space for goods and customers.

EMMANUEL D'IETEREN
Emmanuel d'Ieteren is an agricultural engineer whose doctoral thesis addressed the issue of the environment, and has participated in the development of several evaluation systems (environmental scoreboards, environmental reports, prospective monitoring systems). In 2009, he joined the Groupe One and Ecores teams, which are active in sustainable economy. He developed multidisciplinary expertise in the development of green economic sectors, the support for entrepreneurs in the implementation of sustainable business models and the creation of innovative mechanisms for the regions. Currently, he is both a project manager working in the field and a consultant for companies and regions seeking greater resilience and sustainability.
Retail trade in Brussels: reconciling the city with a changing sector Brussels Studies , Notes de synthèse