Shipping news: the implications of electronic commerce for logistics and freight transport

Resources, Conservation and Recycling 36 (2002) 211 /240 www.elsevier.com/locate/resconrec

Shipping news: the implications of electronic commerce for logistics and freight transport Markus Hesse * Department of Earth Sciences, Geographic Sciences and Urban Research, Free University of Berlin, Malteserstr. 74-100, 12249 Berlin, Germany Accepted 24 June 2002

Abstract The paper considers the significance of electronic commerce (e-commerce) for freight transport, logistics and physical distribution, regarding both business to business and business to consumer commerce. The possible implications of e-commerce are analysed in the broader context of structural change, going beyond narrow assessments that overstate the significance of e-commerce and its potential to make freight traffic more efficient. The main argument of the paper is threefold: first, most recent analyses of freight transport and logistics implications of e-commerce are overstating the current relevance of e-commerce applications on the one hand, and neglecting the influence of the underlying structural change in the entire logistics system on the other. Second, conventional analyses of certain efficiency benefits of ecommerce are probably too optimistic, whereas its negative effects are underestimated at the same time. E-commerce is likely to support longer transport distances and often higher delivery frequencies, increasing demand for land, due to the establishment of new transhipment points (distribution centres) and, to a certain extent, a shift towards truck and air freight transport modes. Third, e-commerce and IT are interrelated components of the structural change in distribution. They affect the environment in terms of vehicle miles, related emissions and energy consumption, by speeding up the time and increasing the geographic area of transport operations. Whether e-commerce contributes to a more efficient distribution system or not very much depends on particular regional circumstances, such as consumer habits, delivery modes and population density. Overall, there is some evidence that ecommerce is likely to reinforce longstanding trends of transport growth, rather than breaking them. Future research should investigate e-commerce more comprehensively, in relation to the entire distribution system and to its application in firms and households. This would also be

* Tel.:
/49-30-838-70209; fax:
/49-30-838-70749; http://www.geog.fu-berlin.de//teas/ E-mail address: [email protected] (M. Hesse). 0921-3449/02/$ - see front matter # 2002 Elsevier Science B.V. All rights reserved. PII: S 0 9 2 1 - 3 4 4 9 ( 0 2 ) 0 0 0 8 3 - 6

212

M. Hesse / Resources, Conservation and Recycling 36 (2002) 211 /240

more useful for dealing with a generic property of modern economy and society: increasing demand for flexibility, speed and mobility. # 2002 Elsevier Science B.V. All rights reserved. Keywords: E-commerce; Logistics; Physical distribution; Freight transport; Warehousing; On-line retail; Home delivery; Environment; Land use; Traffic generation; Structural change; Corporate re-organisation; User behaviour

1. Introduction This paper addresses the implications of electronic commerce (e-commerce) for logistics and freight transport operations. Whereas major transport-related analyses in the context of e-commerce are focussing on passenger transport, this paper investigates the physical distribution (PD) sector. PD comprises the entire system of goods handling and goods movement. The question is to what extent this sector is influenced by e-commerce. Similar to recent discussions on passenger transport, a popular perception of the potential effect of e-commerce on logistics and distribution is driven by conventional wisdom: It suggests that the electronic transfer of information through an optimised logistics system would lead to more efficient transport operations */by cutting out ‘unnecessary’ transactions, by avoiding redundant traffic flows and by eliminating under-utilised infrastructure (Meyer, 2001). Electronic exchange would make the logistics market transparent and allow for optimal organisation and allocation of transport services (Song and Regan, 2001). The situation is much more complicated than conventional wisdom would indicate. E-commerce is an outcome of the interplay of technological innovations and their adoption by certain users (firms, households), leading to particular spatial consequences (goods flow, transport operations, facilities, infrastructure). This needs careful consideration. With regard to the spatial implications of new information and communication technologies, (Graham and Marvin, 1996) distinguished several approaches to investigate the broader relationship between technology and society (Graham and Marvin, 1996, 79). One approach was characterised as ‘technologically deterministic’, highlighting simple impact- assessments that may result from new technologies. In this respect, common assessments of e-commerce within freight operations are based upon modelling exemplary logistics chains and measuring the concrete transport and environmental impact caused by average vehicle operations. In a prototypical case, the traditional way of book merchandising at a neighbourhood store was compared with on-line purchase and related delivery operations (Matthews et al., 2001, 2002). The comparison concluded that the Internet-driven supply mode appeared less transport-, energy- and emission-intensive than the traditional way had been. This kind of investigation leaves many questions unanswered: impact assessment as shown above appears to be too narrow in focus, because it ignores many conditions responsible for the current systems’ performance. The goods distribution system in most of the industrialised countries has been undergoing profound changes over the last two decades. The introduction of supply

M. Hesse / Resources, Conservation and Recycling 36 (2002) 211 /240

213

chain management and its integrated */instead of selective */approach to corporate re-organisation has fundamentally changed the entire system of goods processing (Bowersox et al., 2000). E-commerce appears as an extension of that trend, and has effects relevant for information exchange and, probably, for goods movement as well. But, at least recently, it represents nothing new for how goods are being handled and moved in economy and society. And the future prospects for some ecommerce applications are now regarded far less optimistically by most observers than they were 2 years ago, since a remarkable number of on-line retailers (‘e-tailers’) have gone out of business. With respect to both the complexity and volatility of the subject, this paper follows another approach named by Graham and Marvin for covering the relationship between technology and society. It is associated with the idea of technology as socially and politically ‘constructed’ (Graham and Marvin, 1996, 79). This is due to the very specific adoption of new technologies by individuals and within society. The possible effects of technology are regarded as multifaceted and basically open, to a large extent depending on cultural and social practices and political regulation. Consequently, this paper considers logistics and freight transport implications of ecommerce in a broader context, taking into account the structural change and corporate strategies of re-organisation and re-engineering, within the generic framework of logistics development. Beyond that, it sheds light not only on vehicle miles and emission assessments but also on land use requirements, social costs and the way they are subject to cumulative causation. The paper should help readers to better understand the distribution system and the way it is being operated by suppliers and customers, and certainly encourage them to evaluate the entire transport and environmental implications more carefully. In doing this, the paper reflects the observation that e-commerce is more than just a tool for optimising goods flows: It is a basic element of a newly emerging ‘sphere of circulation’, the system of information, finance and goods flows that is both a fundament and an outcome of the network economy (Castells, 1985, 1996)1. As a consequence, the question whether e-commerce is good, bad or neutral for our freight transport system and the environment cannot be answered without keeping this framework in mind.

2. The development of modern distribution systems By definition, the system of PD is the collective term for the range of activities involved in the movement of goods from point of production to final point of sale (McKinnon, 1988). It comprises all functions of movement and handling of goods, particularly transportation services (trucking, freight rail, air freight, inland water1

Whereas the term ‘circulation’ is grounded in Marxian theory, the current performance of information and commodity flows very much refers to the notion of the ‘space of flows’, which was first introduced by Manuel Castells in 1985 (14) and applied in his concept of the ‘network society’.

214

M. Hesse / Resources, Conservation and Recycling 36 (2002) 211 /240

ways, marine shipping, and pipelines), logistics and warehousing service (e.g. consignment, storage, inventory management, packaging), wholesale and, in principle, retail as well. ‘It is the job of PD to get the right assortment of raw materials and finished products to the right location in an efficient manner timely to marketing and manufacturing requirements (Bowersox et al., 1968). Corporate logistics used to be divided into three functions: supply, production and distribution. Many firms are now following a more integrated approach, which emerged as a result of the upcoming demand for flexibility. Earlier applications of this organisation principle were developed within the concept of lean management, primarily in manufacturing (Womack et al., 1990; Harrison, 1994). The flexibility and integration perspective is now directed towards the whole supply chain. Time has become critical in value creation. Recent developments of logistics and PD are an outcome of economic structural changes, and in turn, logistics is going to influence structural changes as well. These changes comprise as follows: Sectoral changes, i.e. the rise of service economies, the increasing share of goods with high value and low weight, the upcoming high tech and knowledge based sectors. Globalisation, which means (i) the spatial expansion of the economy, (ii) more complex global economic integration, (iii) the upcoming network of global flows and hubs. A power shift in market relations from a supplier-dominated to a buyer-oriented market, associated with completely new landscapes of interfirm competition. The politics of deregulation and liberalisation (effective for the US in the late 70s and early 80s, for Europe since the introduction of the Single European Market in 1992). The introduction of new information and communication technologies that allow for the integrated management and control of information, finance and goods flows. The flow-oriented mode of corporate management and organisation affects almost every single activity within the entire process of creation of value (Bovet and Martha, 2000). The backbone of this highly segmented structure is the supply chain, the timeand space-related arrangement of the whole goods flow between supply, manufacturing, distribution and consumption. Its main components also represent the major actors in this interplay, the supplier, the producer, the distributor (e.g. a wholesaler, a freight forwarder, a carrier), the retailer, and the end consumer. Prototypically, one of the primary changes associated with electronic commerce in terms of logistics is the process of disintermediation, leading to a reduced number of members in the distribution channel, compared with traditional supply chains (see Fig. 1). What used to be processed from the manufacturer through wholesale and retail to the customer (either a firm or the private consumer), is now going a different way, probably by eliminating one or more of the costly operations in the traditional supply chain organisation (Abbey et al., 2001). Delivery modes are thus changing as well.

M. Hesse / Resources, Conservation and Recycling 36 (2002) 211 /240

215

Fig. 1. Conventional and Internet driven supply chain.

The intermediate character of logistics is not new. It has already been stated more than 30 years ago: ‘If the physical distribution mission is aborted, little happens of a profitable nature. Thus, physical distribution is concerned with a vital area of corporate activity. The totality of the physical distribution process constitutes one of the main operational efforts of the firm’ (Bowersox et al., 1968). A major requirement of the implementation of PD was the invention of modern information and communication technologies. Features like electronic data interchange (EDI), the automatisation of product flow in dedicated warehouses and distribution centres (DCs), or the recent computerbased tracking-and-tracing systems (which offer online insight into the status of your shipment via the web) are primary sources of enormous productivity gains over the last two decades. The standardised container, which industrialised trade and transport, also improved productivity enormously. An important requirement for the successful implementation of these innovations was the expansion of traditional infrastructure such as highways, terminals and airports. Thus, the material foundation for the contemporary ‘network economy’ is based both on modern logistics and on traditional asphalt and concrete. In order to identify the really ‘new’ contribution of electronic commerce to business management it is useful to look back at previous applications of information technology (IT). Corporate strategies in PD were primarily associated with two technological innovations. They were developed in the late 1970s/early 1980s and almost revolutionised logistics: the bar-code and EDI. Both allowed for a completely new operation of inventory management, adjusting the entire material management according to customer’s demand and thus reducing inventory costs substantially (Abernathy et al., 2000). The main disadvantage of bar-code-based

216

M. Hesse / Resources, Conservation and Recycling 36 (2002) 211 /240

inventory management and particularly EDI-appliances was their high price and non-standardisation. Both reasons led to a limited degree of implementation, since only large firms could afford to invest in these systems. With the emergence of the Internet, the situation changed dramatically: prices went down and acceptance increased accordingly, while the http-protocol functioned as the main common standard for all web-suppliers and users. Now the web has become a major rationalisation tool, not only for individual firms and their subsidiaries, but also for managing the entire producer-customer relationship. The accelerated pace of technological innovation contributed to further changes, for instance by introducing Global Positioning Systems (GPS) that allow for identification and both efficient and flexible routing of vehicles in a way that was not known before.

3. E-commerce applications and performances in logistics The term ‘electronic commerce’ primarily refers to the exchange of information and to transactions between different commercial agents, such as firms (also called business-to-business electronic commerce or ‘B2B’), which is different from the relationship between firms and consumers (business-to-customer electronic commerce or ‘B2C’). Transactions between governments and other agents are omitted in this paper. Among the firms practising e-commerce are both general manufacturers, retailers or wholesalers (who are now developing additional distribution ‘channels’ and thus trying to optimise their market performance) and completely new, IT-based firms, who are concentrating on webdesign, IT applications, or pure on-line retail. Among bookstores for instance, ‘Barnes & Noble’ is an example of the former, and the largest Internet retailer Amazon.com represents the latter. Regarding the organisation of e-commerce transactions and PD, it is extremely important to distinguish between customer-related activities, such as order receiving, sales and marketing (the so-called ‘front end’ of e-commerce), and the processing and shipment of the ordered goods (the so-called ‘back end’). The latter also includes PD */the management and movement of goods to fulfil the customers’ needs. Most innovations by e-commerce refer to the ‘front end’: the way goods or services are being ordered by customers, and how they are successfully marketed in a highly competitive environment. The Internet functions as a main platform for information exchanges, transforming them into market transactions. Particularly in B2B relationships, this procedure offers certain advantages to both sides: The supplier receives orders with a substantially higher degree of reliability and often predictability. Business customers can accelerate the entire ordering process, receive delivery faster, and are able to follow the status of their order on-line. Changes can easily be processed within a fraction of the time required before. Large firms experience greater market transparency (offers, prices) and are able to increase their purchasing power, in order to achieve overall cost reduction. For the B2C relationship, the degree of change appears more limited: what once was ordered by phone or fax is now being transmitted through the computer and processed via the Internet.

M. Hesse / Resources, Conservation and Recycling 36 (2002) 211 /240

217

Totally different from these ‘front end’-operations is the way electronic commerce affects the ‘back end’ of the supply chain and of the PD process. It appears that this field of transaction has undergone only minor changes so far. Even the most prominent web-retailers often did not spend much attention on PD, which often raised costs and caused severe logistical problems. The ‘back end’ practices of delivery are in many cases not much different from the way goods were being shipped 5 years ago. Since B2B and B2C operations differ significantly in terms of structure, operations and potential outcome, some key features are discussed separately as follows. 3.1. B2B electronic commerce According to recent OECD research, about 70 /85% of all electronic commerce transactions refer to business-related exchanges (OECD, 2000). The ordering and supply system within interfirm relationships comprises the bulk of e-commerce. The advantage of B2B e-commerce for companies is that it offers almost perfect market information and the opportunity to re-allocate the vast purchasing power of firms. The cost reduction potential of this kind of e-commerce is estimated to be tremendous. A widely recognised study by Goldman Sachs established that B2B ecommerce is likely to contribute to cost reductions of up to 40% of corporate expenses in selected US industries (Brooks and Wahhaj, 2001). The Goldman Sachs study estimated that the freight transport industry would be able to mobilise cost reductions amounting to 15 /20% of current expenses (see Table 1). This rate is only exceeded by the savings available to the electronic components (29 /39%) and forest products (15 /25%) industries. Table 1 Potential cost savings from B2B e-commerce in the US, by sector Branches

Cost savings (%)

Electronic components Machining (metals) Forest products Freight Transport Life sciences Computing Media and advertising Aerospace and related parts Steel Chemicals Oil and gas Paper Health care Food ingredients Coal

29 /39 22 15 /25 15 /20 12 /19 11 /20 10 /15 11 11 10 5 /15 10 5 3 /5 2

Source: Martin Brookes and Zaki Wahhaj, ‘The Shocking Economic Effect of B2B’, Goldman Sachs Global Economics Paper: 37, 3 (February 2000).

218

M. Hesse / Resources, Conservation and Recycling 36 (2002) 211 /240

The immense economic potential of B2B e-commerce to lower production costs is */among other factors */primarily based upon the opportunity of a complete breakdown and cost efficient re-organisation of the supply chain (Reynolds, 2000). Most common e-commerce applications in distribution, either in the logistics industry or in manufacturing and retail, relate to: / / / / /

order processing and fulfilment, warehouse and inventory management, fleet management, vehicle routing, freight brokering (e-marketplaces), electronic payment.

In this respect, e-commerce is likely to reinforce the cost saving potential of integrated logistics concepts. This is exemplified by Dra¨ger Sicherheitstechnik in Lu¨beck/Germany, the winner of the German Logistics Award in 1999. The mediumsized company (about 2600 employees worldwide) specialises in manufacturing safety devices such as respiration protection technologies, mobile measurement technologies, gas detection instruments and sensors. By introducing the Safety Technology Integration Logistics (STIL) system, the company has managed to reduce its stocks by 24%, to raise its delivery punctuality to 90%, to streamline the number of sales items (from 25 000 to 8000) and to standardise many of its parts (the number of types and parts has been reduced from 22 000 to 10 000). The impact of the new logistics system on profits was remarkable: The firm’s earnings (before interest and tax) rose from DM 17.4 million in 1997 to DM 32.3 million in 1998, directly related to supply chain improvements (Dra¨ger, 2000). The potentials of B2B e-commerce for cost reduction are considered so attractive that even competitors make joint use of e-commerce. Some of the largest automobile manufacturers founded the ‘Covisint’ platform, in order to streamline their supply chain, to reduce complexity (redundancy) and to lower purchase cost. A basic promise of ‘Covisint’ is to reduce automotive development cycles from currently 4/6 years to 1/1.5 years (Cassinfo/ProLogis, 2000). In the distribution business, information exchange by ecommerce helps to improve inventory management and order fulfilment. Thus facilities and vehicles are used more efficiently and many processes are at least partially more predictable. At the same time, these technologies allow for a more demand-side, short-term oriented and thus unpredictable order behaviour. This can cause severe impacts on the reliability of business services as well, and may therefore hamper the firm’s performance in an increasingly competitive, time-sensitive environment. How far B2B e-commerce applications are in fact implemented in freight and distribution businesses is not yet precisely known. Information management and integrated processing of the entire supply chain are supposedly used to a large extent, particularly by medium sized and large firms. Inventory management is increasingly enforced with the assistance of handheld devices (as is routing or scheduling), even if that may not be regarded as a direct outcome of e-commerce (OECD, 2001a). Other innovations, such as ordering via on-line freight brokers, seem to be accepted less

M. Hesse / Resources, Conservation and Recycling 36 (2002) 211 /240

219

extensively. There is no evidence that all shippers and carriers would join auctions and other electronic marketplaces to lower production cost and to boost delivery efficiency, as many expected. Surveys show that a large number of firms in the US (nearly 70%) and in Europe (about 86%) still operate on the basis of contracts, mostly signed for annual or longer periods (Morgan et al., 2000). The ‘spot market’ for short-term demand for transport services may offer cost advantages. But this is normally balanced against other factors, such as reliability and predictability. Many of the electronic marketplaces (freight platforms) that were being established in the late 90s no longer existed 2 years later. It is unlikely that this mode of operation will gain significant importance, except in market niches. From this perspective, the specific potential of e-commerce to re-organise and improve freight operations appears to be limited. Examples such as Covisint also demonstrate the limits of e-commerce purchase platforms, since it is reported that one major advantage of the platform */cost transparency */can also be regarded as a disadvantage. Platform participants, notably the subcontractors and suppliers, are reported to have distinct reservations against publishing cost structures, due to the high pressure on subcontractors exerted by customers, and because they are concerned about revealing proprietary information (Bretzke, 2001). Retailers may have good reasons for carefully watching co-operating participants in electronic supply chains, since they are potentially challenged by the attempt of manufacturers to introduce direct distribution through disintermediation. By ‘pirating the value chain’, manufacturers could keep a larger share of the profit for themselves, particularly once they have good access to customers (e.g. via the web). This risk seems to be an important barrier to broader implementation of B2B e-commerce (OECD, 2001b). Another limitation for the extension of e-commerce is logistics */regarding the basic fact that overcoming time and space constraints still remains a costly but necessary activity for almost any business, regardless of the question whether the front end is working properly or not. In general, it is more likely that those e-commerce applications that explicitly fit into the predominant mode of re-organisation and improving competitiveness will become widely implemented. Others, particularly those with an unclear cost-benefit ratio, may not gain that importance. Thus the use of B2B-technologies will be limited. Where it is mostly practised (in the field of information transfer), it does not necessarily shape PD and material goods flow. As a consequence, the direct implications of B2B e-commerce on logistics and freight transport might still be hard to assess. However, in the long term, the indirect outcome of predominant logistics practices may increase, although they are currently difficult to predict. 3.2. B2C electronic commerce B2C-related electronic commerce has gained more public attention than have interfirm exchanges, although B2C is significantly less important in terms of number and value of operations. Many people became familiar with the idea of consumer goods home delivery from the very beginning. Among e-commerce products, those that were predominant were either well suited for mail-order and delivery (such as

220

M. Hesse / Resources, Conservation and Recycling 36 (2002) 211 /240

books and CDs, computers, airline tickets or hotel reservations), or were ordered both frequently and in bulk (such as groceries). It is no coincidence that among the most important B2C on-line retailers were, for example, the book-and-more-store Amazon.com, the personal computer and devices manufacturer and retailer Dell, and */up to July 2001 */the on-line grocer Webvan.com. The extent to which B2C e-commerce contributes to major changes in logistics, wholesale or retail distribution processes (and in doing so, how it may affect transport and the environment) is controversial. Both questions should hence be subject to careful consideration. Certainly, it first of all depends on the rate of ecommerce adoption by users and market participants, which still seems to be limited. On-line retailing was expected to experience dramatic increases according to early estimates. More recently, expectations have been more modest. For the fourth quarter of 1999, the market share of on-line retail in the US was estimated to total 0.64%. It rose to a peak of almost 1.1% exactly 1 year later, falling to 0.9% in the second quarter of 2001 (US Department of Commerce, 2000). Recent observers estimate the market potential of on-line grocery retail to a share of no more than 2% in 2005 (OECD, 2001c). It is notable that Webvan.com, the second largest on-line retailer, went out of business in July 2001, due to rising costs and limited number of customers. The firm reportedly lost a total US$ 830 m, without ever having achieved the desired market share. Amazon.com, still the largest ‘e-tailer’, has received the first net earnings from its operations no earlier than in the fourth quarter of 2001, after 6 years in business (Amazon.com, 2002). Second, estimates how far information exchange by e-commerce leads to corresponding changes in the patterns of PD */and hence how it influences transport and the environment */are hampered by the lack of precise data, particularly those that can be generalised. Freight implications of e-commerce are certainly dependent on the characteristics of the finished product. On-line retail is mostly practised for specific commodities (see Table 2). In some rare cases physical transport is eliminated, for example, once documents or music files can be downloaded from webpages, instead of being packaged and mailed to a shop or directly to the Table 2 Categories of goods purchased by on-line shoppers Product categories

Bought by percent of on-line shoppers (%)

Books CDs, recorded music Computers, related products Air travel reservations Videos, filmed entertainment Flowers Event tickets (sport, enterainment) Food, drink Men’s clothing Women’s clothing

66 58 38 26 19 18 17 13 12 12

Source: Ernst and Young, 2000. Global on-line retailing, January (cited in Browne 2001, p. 5).

M. Hesse / Resources, Conservation and Recycling 36 (2002) 211 /240

221

customer. Electronic ticketing by airlines may spare the delivery of the tickets to the travel agent or customer. However, as far as it is possible to make a judgement on this issue, those transactions constitute a minority of on-line retail. If they are just replacing the transfer of information or documents, the net effect may only be minor savings. In most cases, the innovation by on-line retail is limited to a new ‘front end’ for sales and marketing purposes. In this respect, on-line retail is comparable with the traditional catalogue business; what used to be the mail-order catalogue is now becoming the Internet. The degree of change seems to be limited, though it is still an open question how far catalogue retail will be affected by the Internet at all. The back end of B2C is probably far less sophisticated and more ubiquitous. PD is still required for delivering most products or services. In many cases of e-commerce, this task is fulfilled by logistics service providers (so-called third party logistics or 3PL), and to a certain extent by couriers, express and parcel services, due to the large quantity of small parcels and packages in e-commerce related merchandise. In taking this approach, on-line retail makes use of highly efficient distribution systems, though it is associated with certain environmental effects. In other cases, on-line retailers such as Webvan.com were operating their own DCs and delivery fleet, which was clearly too expensive, keeping in mind the small profit margins in grocery retail. Recent attempts to implement home-delivery highlight severe deficiencies both from the supply and demand sides of this distribution channel (see below). In most cases, this approach is too expensive and is not yet accepted by a majority of consumers. A major impact on the back-end of on-line supply chains is the elimination of one of the members of the distribution channel through disintermediation, compared with the usual marketing process operated by manufacturer, wholesaler, freight forwarder, retailer and customer (see Fig. 1). There is indeed empirical evidence that this can happen. The computer manufacturer and marketer Dell has successfully established a system of direct merchandising, combined with the integration of customer demands into the configuration of the final product (‘built to order’). The opposite can also be true: Middlemen such as wholesalers usually fulfil an important cost-minimising function in the entire delivery chain (Robeson and Kollat, 1985), and in practice they have not lost real influence and significance (see Fig. 2). ‘The Internet story highlights a little understood trend: Wholesale distribution intermediaries are growing, not shrinking. Sales through distribution of non-durable goods, which include everything from books to industrial supplies, have been expanding at about 1.4 times GDP growth. Many pundits have predicted disintermediation, but the data show the complete opposite trend. In other words, reports of the death of the middleman have been greatly exaggerated (The LehmanBrothers, 1999) One of the recent outcomes of the cost-crisis in pure on-line retail is the rising competitiveness of the so-called click-and-mortar companies. They combine on-line sales and ordering systems with distribution out of their stores. One well known example is ‘Fnac’, the largest book and music retailer in France who bought the ‘Societe´ Francaise du Livre’, an established book wholesaler and distributor specialised in logistics and warehousing, capable of delivering 60 000 titles in 48 h,

222

M. Hesse / Resources, Conservation and Recycling 36 (2002) 211 /240

Fig. 2. The intermediate role of wholesale in distribution channels.

in order to expand the Fnac on-line branch. The new distribution channel does not replace the traditional logistics system, responsible for 99% of the firm’s orders, but adds to it (OECD, 2001d). Similarly, Safeway, one of the largest US grocery retailers, in conjunction with Tesco plc. (UK), is trying to develop home-delivery services in Portland/OR and Vancouver/WA out of regular neighbourhood stores, not out of specialised but costly DCs. Overall, the entire logistics chain often remains complex and lacking in transparency. Even if on-line retail means that a product is being delivered directly either from the manufacturer (although that is rare) or, more likely, distributed out of a dedicated DC to the final consumer, it is unclear how many agents are financially involved in the transactions, nor how many operate within the physical transfer (Reynolds, 2000, 430). This questions needs more empirical evidence to be answered. Until such data are available, we must be careful with general statements and judgements.

4. Freight transport and environmental implications Electronic commerce is certainly not only a matter of virtual information flows but also has a material goods movement dimension. With more anecdotal reference, this fact became widely known in public by a special event in the year 2000. At the end of July 2000, the fourth book in the Harry Potter series */Harry Potter and the Goblet of Fire*/was released in the US, rapidly becoming one of the most popular children‘s books in the country. In order to guarantee same-day delivery, 2 50 000 copies had been reserved for Amazon.com for express shipping. This was a recordbreaking number in the on-line retail of a single e-commerce company. In

M. Hesse / Resources, Conservation and Recycling 36 (2002) 211 /240

223

cooperation with the freight integrator Federal Express (FedEx), free delivery on publication day was offered to all preordering customers of Amazon.com. The delivery schedule was extremely time-sensitive and associated with particular transport consequences, since it required the separate operation of 100 air freight planes and 9000 trucks (Matthews et al., 2000). At first sight, the freight transport related impacts of e-commerce in this particular case (both in terms of traffic volume and logistics quality) seem to be quite clear. Competition drives service level and time requirements, and in order to fulfil these needs, a certain transport and environmental expense is generated. However, the whole issue is more complicated and far from being easily answered. Both in B2B and B2C e-commerce, information transfer is associated with distinct patterns of freight transport. The possible net effects in this relationship can occur at least in three ways: substituting for physical transport (‘e-limination’ of goods movement, taking into account certain interdependencies with passenger transport), being more or less neutral for freight transport, or generating additional demand for both information and goods transfer (‘e-combustion’)2. Recent papers addressing the freight transport-related impact of e-commerce end up with contradictory findings, based on conceptual and empirical findings in France, Germany, the Netherlands, the UK, the US or Sweden. This is due to the different subjects and the varying practices that have been investigated. Some make general statements on substitution or reduction of freight transport due to electronic commerce and home-delivery, even if the calculated examples represent very specific cases (Matthews et al., 2001). Others are more cautious, tending to optimistic scenarios, yet without neglecting that there is indeed the option for increasing freight transport as well (Orremo et al., 1999). A third group is making more critical statements on the freight transport generation impact of e-commerce, e.g. in conjunction with the generic structural change (Browne, 2001; Colin, 2001; Bundesministerium, 2001; Fla¨mig, 2001). Following Dutch estimations, freight transport is expected to increase until the year 2005 by 38% of all vehicles trips (Transport en Logistiek Nederland, 2000). Seventeen percent of this growth belongs to ‘new economy’-related freight movements, 21% to ‘old economy’ sectors (see Table 3). This estimation indicates that it makes sense not to separate the discussion of e-commerce from other, more general developments in the field of structural change and logistics. Some studies highlight the generic property of e-commerce to compress delivery times and to increase the geographic areas of supply and distribution (Bahrami, 2001; Hultkrantz et al., 2000; Janz, 2001). Since this is just a small selection of a probably large number of different cases to be studied, a detailed investigation would exceed the scope of this paper. In order to provide some basic knowledge, some of 2 See for the basic relationship between communication and passenger transport and relevant expriences the overview by Mokhtarian, P., 2000. Telecommunications and travel. Millennium paper of the Committee on Telecommunications and Travel Behavior of the Transportation Research Board: 1. Washington DC.

224

M. Hesse / Resources, Conservation and Recycling 36 (2002) 211 /240

Table 3 Freight transport growth by 2005 in the Netherlands Transport segments

Changes by 2005 (number of vehicle trips) (%)

Freight transport ‘old economy’ Freight transport B2C Freight transport B2B Total freight traffic Passenger traffic


/21
/8
/9
/38
/10

Source: Transport en Logistiek Nederland, 2000. Assumptions: Online orders in B2C increase by 10 / 15% until 2005; the market share of online merchandise achieves about 15% in non-food, about 10% in the food sector.

the most important and typical features of e-commerce-related freight transport consequences are summarised below. Environmental impacts such as energy consumption, air pollution and noise disturbances are caused by all freight vehicle operations. The only exceptions are bike couriers who can achieve a certain market share of the shipment of documents and small parcels. This is primarily happening in core, dense urban areas, where many new economy and e-commerce firms are located. Regarding all fossil-fuel based transport vehicles, it is widely recognised that the greatest environmental burden stems from truck and air freight operations, with railway and waterway freight being less intensive in energy consumptions and emissions (OECD, 1999; INFRAS/IWW, 2000). The environmental impact particularly of road freight transport relative to passenger transport has become much more important over the past two decades due to higher growth rates and to still limited efforts in introducing clean technologies. Several recent studies have investigated energy consumption patterns and emission impacts of the different freight transport modes (Schipper et al., 1997a; Forkenbrock, 2001). 4.1. Truck transport efficiency and frequency Some have claimed that, due to integrated data transfer and information management, flexible distribution and e-commerce would allow the processing and shipping only of those materials that have been ordered, nothing more. In doing this, the system would work more efficiently than it did previously, when transport and storage were operated for an unknown demand. Once distribution could be customised more properly, returns could be avoided to a far greater extent than before. All this is supposed to happen as efficiently as possible, since firms are committed to reduce cost and not to waste time, workforce or transport capacity (Komor, 1995; Pflaum et al., 2000). The breakdown of loads can lead to different consequences. On the one hand, it may generate no direct additional demand for transport services, if on-line ordered products are shipped via full truckload or parcel services. On the other, the share of bundled freight flows (and thus of more efficient portions of freight transport) can

M. Hesse / Resources, Conservation and Recycling 36 (2002) 211 /240

225

decrease as well, following more demand-side oriented logistics. This was also the case once new concepts in lean management were introduced, contributing to lower weight at the expense of higher trip frequencies (Ten Hompel and Siebel, 2001). Also influenced by rising carrier competition, increasing order flexibility supports on-time delivery, often with less than a full truck load, at higher frequencies and with smaller vehicles than before. This is particularly true for urban distribution patterns that are less efficient than long-haul operations (McKinnon, 1998). Increasing distances and vehicle miles are due to market integration at a larger geographical scale. Particularly, the hub-and-spoke systems that most logistics companies and parcel services operate today contribute to the extension of transport distances. In some instances, the option of a more efficient freight transport is likely, once shipments are processed through large logistics networks, thus mobilising the advantage of economics of scale. In many others, it is likely that vehicle miles may increase simply due to rising distances between the transhipment hubs (Bahrami, 2001).

4.2. Bulk freight carriers The breakdown of smaller loads according to the splintering of the order- and delivery-cycles supports a certain ‘atomisation’ of freight flows, which once were more bulky in volume and more buffered (less time-critical) in logistics than today. Rail freight and inland waterway systems, to many extents the losers of structural changes in logistics, may again be driven out of business, since they cannot cope with the demand for flexibility as well as with the decreasing size of the truckloads. ‘A significant and possibly alarming outcome of the increased market penetration of IT and EC [e-commerce] is the likely acceleration in the shift in domestic freight mode share away from rail and waterways, towards all-highway movements.’ [. . .] ‘By all indicators, the widening gap between truck and non-highway modes in domestic freight service will continue to increase with the growth in EC. The April 2000 shipper survey conducted by Morgan Stanley Dean Witter indicated that 70% of the respondents expected their freight transportation needs to change as a result of EC. Of these, 65% said they would use more parcel/express envelope service, 57% expected to use more regional or national less-than-truckload (LTL) carriers, and 53% expected to use more local trucking or courier service. Twenty one percent of the respondents expected to use less rail as EC becomes more prevalent (O’Kelly, 1998). It is not clear whether the bulk oriented freight carriers can become competitive in those goods sections that are likely to be processed via e-commerce (Bahrami, 2001). Rail and waterway freight carriers are probably still concentrated on the old industrial, less splintered and less time-critical freight markets */exactly the opposite of the nature of e-commerce generated transactions. Container trains and inland ships were supposed to cover a certain portion of domestic and particularly international freight transport growth (e.g. by specialised parcel-express trains). Until now, they have not developed as hoped.

226

M. Hesse / Resources, Conservation and Recycling 36 (2002) 211 /240

4.3. Air freight Judging from recent surveys and insights, PD in general and e-commerce in particular are extremely time sensitive. This characteristic implies a particular role of air freight, which is both the fastest growing and least environmentally acceptable mode of air freight. Since almost half of the air freight is shipped via dedicated freight planes (different from the so-called belly freight in passenger planes), ecommerce is likely to generate a certain amount of direct air freight, by lean inventory strategies and shortened delivery times to the customer (Broens et al., 2000). It is no coincidence that two major US parcel services, UPS and FedEx, both have established their own front and back end systems for the e-commerce market, with associated air freight operations and newly constructed dedicated facilities for parcel consolidation and distribution. The environmental impact of air freight transport is characterised by significant levels of energy use, air pollution and noise disturbances (Clancy and Hoppin, 2000). Beyond this quantitative share of certain emission sectors, air freight contributes to an explicit burden in a more qualitative manner. First, aircraft emissions are the only manmade damage of the ecosystem in certain parts of the atmosphere. The relationship between climate change and high-level aircraft emissions seems to be evident (Penner et al., 1999). Second, aircraft-related noise disturbances happen very often at night. Many transport services in industrialised countries are operating overnight, and air carriers, postal services and couriers often rely on night-time starting and landing privileges (whereas regular air services are often prohibited between 22:00 and 6:00 h). Then, air traffic noise emission can cause major disturbances and may be perceived negatively by the public. As discussed earlier, e-commerce is not exclusively responsible for these damages and disturbances. However, due to the undoubted preference of many e-commerce shipments for express and parcel services, it is likely that this particular component of air freight further contributes to generic air transport growth (Jindel, 2001; Burnson, 2001; Sobie, 2000). Thus, air transport is critical for the sustainability of the entire transport system. In the longer term, e-commerce as a major customer of air freight is also likely to increase (even indirectly) the pressure on airports for further growth and further expansion. As long as major airports worldwide continue working at limits and plan to expand infrastructure and facilities (even if we have to consider recent slowdowns as a result of Sep 11, 2001), e-commerce by air freight will be associated with future air traffic growth and related problems. 4.4. Customer (shopping) travel As a consequence of home-delivery, consumers presumably do not need to walk or, more important, drive to the shop to pick up their orders, so on-line-purchase may spare passenger trips */at least in theory (Matthews et al., 2001). Most comparisons on the logistics impact of e-commerce highlight this ‘last mile’ topic, hoping for reduced passenger miles as an outcome of delivery convenience. This suggestion is being made without reflecting both the complex design of local freight

M. Hesse / Resources, Conservation and Recycling 36 (2002) 211 /240

227

distribution patterns and the multifaceted nature of shopping as a cultural and social activity. First, speaking logistically, an environmentally acceptable and efficient ecommerce delivery primarily depends on the specific local framework of distribution. The literature on this topic is still scarce, though calculations support the suggestion that cost-efficient home-delivery systems depend on strong requirements (Orremo et al., 1999; Hultkrantz and Lumsden, 2000; Cairns, 1996). Major factors responsible for distribution performance are the distances between DC and market areas, the density of population (customers), and finally optimised load factors and routing of delivery vehicles3. Swedish calculations (Orremo et al., 1999) are based on certain assumptions for conventional food retail: average distance to the next food store is 7.81 km; modal share of shopping trips is 60% car, 4% public transit, 36% walking/ biking (see Table 4). The assumptions for on-line merchandise are as follows: a market share of online food retail of 10, 25 or 50%, the required population density is 20 people per m2 (/10% market share), 8 people/sqm (/25%) or 4 people per m2 (/50%); the value of an average online purchase is supposed to be four times higher than in conventional retail, the distribution area comprises a radius of 30 km, and each delivery van holds 25 orders per route. The findings are summarised in Table 4. As a consequence, a cost-efficient home-delivery service that is environmentally favourable can be logistically organised in different ways. On the basis of the above assumptions, Fig. 3 presents four distinct distribution models. Models 1 and 4 are reported to be favourable for higher density urban areas, model 3 is likely to support efficient distribution in more dispersed areas. The most common idea of delivery out

Table 4 Change in energy consumption by home shopping (B2C) in the food sector of Sweden Market share of home shopping (%)

Change in energy consumption (50 km Change in energy consumption (90 km delivery route) (%) delivery route) (%)

10 25 50

/7 /18 /36

/5 /13 /26

Source: Orremo et al., 1999. Assumptions: conventional food retail: The average distance to the next food store is 781 km. The modal share of shopping trips is 60% car, 4% public transit, 36% walking/biking. Online-merchandise: The market share of online food retail achieves 10, 25 or 50%. The corresponding population density is 20 people/sqm ( /10% market share), eight people/sqm ( /25%), four people/sqm ( /50%). The value of an average online purchase is four times higher than in conventional retail. The distribution area comprises a radius of 30 km. The delivery van holds 25 orders per route. Energy consumption is calculated with current data for passenger car, delivery van and truck (1999).

3

For comparisons between different physical environments of urban retail see also: Aoyama, Y., 2001. Structural foundations for e-commerce adoption: a comparative organization of retail trade between Japan and the United States, Urban Geography 2001;22,2:130 /153.

228

M. Hesse / Resources, Conservation and Recycling 36 (2002) 211 /240

Fig. 3. Local distribution models.

of a DC to customers’ homes, as featured in model 1 of Fig. 3, requires high population density and consumer participation, full van or truck loads (average number of shipments per load: 25) and home runs about less than 180 km (with average ten shipments and 75 km each tour). Population density and optimised logistics, both an expression of high degree of consumer participation and high load factors, seem to be essential. Regarding the organisational and spatial concentration of transhipment facilities, particularly of modern DCs or warehouses (see the section below), and the deconcentration of population as a result of sub- und exurbanisation processes, providing for efficient home-delivery services is at least difficult, if not unlikely. Against this background, and taking into account the extremely low profit margins in retail (i.e. grocery), it is no surprise that even such advanced delivery models like those offered by Webvan.com, kozmo.com or peapod.com failed in the past. As already suggested above, it is more likely that the combination of Internet platforms on the ‘front end’ (customer order) with neighbourhood stores for optimised distribution on the ‘back end’ will be more effective. Even in more e-commerce-favourable product categories such as book or CD distribution, and in the case of technologically highly advanced

M. Hesse / Resources, Conservation and Recycling 36 (2002) 211 /240

229

firms such as Amazon.com, distribution efficiency as a function of density has been described as critical4. Second, it is questionable under what circumstances a higher degree of efficiency in goods delivery may really correspond with changing customer behaviour regarding shopping trips. If additional home-delivery trips are generated, a positive net effect may only be achieved once a certain number of passenger car shopping trips are eliminated. Consequently, people would have to stay at home and receive their delivery (or do something else), instead of driving. Recent experience shows that there is not as much substitution as expected, due to customer habits and routines, and also because shopping is often seen as a social event, increasingly being mixed with leisure or other activities (Casas et al., 1999; Gould et al., 1998). This example underpins that new technologies are deeply embedded in culture and in social practices, in daily routines and habits that are far from being rapidly changed by new order-and-delivery services. Finally: as a consequence of Internet shopping experiences, customers are becoming accustomed to short-term order processing */even without any real necessity, just because consumers have the choice to do so. Changing order habits may support the desire for the fastest possible receipt of delivery. In many ways, this relates to the truck or the airplane: ‘Web shoppers, by placing a premium on nextday delivery of the goods and on direct control and tracking of the delivery process, are relying increasingly on the expedited highway and air modes (Bahrami, 2001).

4.5. The social costs of freight transport Data collections on social costs have been established in order to investigate comprehensive environmental and social impacts associated with a given technology. Whereas certain impact assessments of e-commerce related to transport and the environment primarily focused on energy consumption and air pollution, it is evident that freight transport is also the source of other impacts that are not paid for by users. These so-called externalities are worth citing here, since they also include the cost of congestion, the cost of infrastructure construction and maintenance. Road and air freight also contributes to severe traffic noise disturbances, which are known to increase dramatically, at least in urban areas. As far as e-commerce supports general structural changes associated with more truck and air freight transport, it may contribute to certain externalities as well.

4 According to a press interview in the British Guardian on Feb 9, 2002, Amazon.com-CEO Jeff Bezos explained the more profitable nature of its businesses in Europe with higher customer densities, related lower distribution cost and higher real estate prices, compared with the US. The cost of overcoming time and space constraints on the back-end of the supply chain made Amazon.com also strengthen their warehousing operations, particularly by closing down one out of then seven US fulfillment centres in Georgia, January 2001.

230

M. Hesse / Resources, Conservation and Recycling 36 (2002) 211 /240

The different transport modes are distinct in terms of their fatalities, infrastructure overuse, energy use and environmental degradation. The reported externalities turn out to be different, according to the specific transport mode’s characteristics. Recently, the ‘external costs’ of all transport modes in Western Europe were estimated at 658 bn Euros (a) in 1995, one third coming from freight transportation. This sum is supposed to have reached 700 bn a in 2000 (in 1995 prices). Air freight transport contributes to this with the highest external costs per unit (205 a/1000 t km, mostly related to climate change and noise disturbances), compared with 88 a/1000 t km of road, 19 a/1000 t km of rail and 17 a/1000 t km of waterways5. Regarding the specific contribution of truck and air freight traffic to noise emissions, infrastructure overuse (truck) and global warming (air), it is likely that a widespread practice of ecommerce might support a further increase in these particular social costs of transport.

5. Freight and land use (warehouses and distribution centres) One of the most obvious issues in the freight context of e-commerce is the increasing need for transhipment points, such as warehouses and DCs (and, eventually, pick-up points for customers as well). DCs are operated in order to control the more complex, mass-customised freight flows without increased logistics costs. The rising demand for DCs and warehouses is currently a major feature of the ongoing structural change in logistics and freight distribution. It is by no means exclusively related to e-commerce, but may become accelerated by its contribution to the re-organisation of distribution. In terms of facility development, it is important to know that the new patterns of demand and supply are shaping the old warehouses into new DCs, or ‘High Throughput Centres’ (Abbey et al., 2001). The facility is no longer needed for storage (at least not primarily) but for the efficient consolidation and processing of materials flow. ‘A DC is the antithesis of a warehouse. Warehouses serve as the physical expression of the need to store large inventories of goods, the main artifact of traditional retailing. DCs, in contrast, form the nexus between retailers and their suppliers. [. . .] Rather than being a place for storage, a DC consists of bays for in-bound and outgoing trucks, an automated, fast-moving conveyor network connecting them, and a sophisticated information system to control movement from receiving to shipping docks as well as to process the transactions relating to the shipments’ (Abernathy et al., 2000, 10).

5 All external cost data is based on the research studies ‘Internalisierung der externen Kosten: Instrumente’, INFRAS Consulting Group, Zurich, February 1999; ‘Externe Kosten des Verkehrs-, Unfall, Umwelt- und Staukosten in Westeuropa’, by INFRAS, Consulting Group, Zurich, IWW, University of Karlsruhe, March 2000, both in summary published by: UIC-Union Internationale de Chemins de fer/ International Union of Railways (ed), 2000. Der Weg zur Nachhaltigen Mobilita¨t. Die externen Kosten des Verkehrs reduzieren. Paris.

M. Hesse / Resources, Conservation and Recycling 36 (2002) 211 /240

231

Warehousing in terms of keeping certain commodities ready for quick shipment is becoming increasingly important, as the economy is increasingly demand-side oriented. The ability to deliver is becoming a major issue of competition. The more diversified and fluctuating the demand is becoming, the more difficult is the task to provide for appropriate, fast and cost-efficient response. ‘The act of warehousing exists because companies are unable to predict demand and prefer to provide a buffer for themselves that accommodates spikes and lulls in the sales process. E-commerce tools, which enable the instant sharing of data among trading partners, dramatically improve the ability to predict demand. Aggregate demand for traditional warehousing space should decline over time, as the enabling technology is widely adopted and implemented. Today’s state-of-the-art warehouses feature high-cube space with clear heights of at least 30 feet. However, as the new technology enables continual movements of products in the supply chain, the need to stack inventory begins to diminish. Traditional storage space must start housing activities that involve more horizontal movement rather than vertical stacking (. . .).’ (Kirschbraun and Bomba, 2000). Owing to the competition between the primary distribution locations, all major freight hubs (large ports, freight airports, inland hubs) are currently committed to plans for expanding related infrastructure. This is, among other factors, a consequence of the high pressure on supply chains, caused by accelerated information transfer and changing consumer habits. ‘As the delivery of products moves toward greater degrees of customisation, real estate requirements will include additional power sources, docks (supporting greater throughput) and people to manage the process. Although existing high-cube facilities are not necessarily obsolete, most likely they will require more land for employee parking and queue trucks. Effectively, then, the world of industrial real estate will encompass more distribution and less warehousing space. The value of speed is so high in the New Economy that modernised production and distribution facilities increasingly will cluster around major airfreight hubs. Depending on the cost structure of various classes of product, additional clustering around truck and rail logistics hubs will occur. Logistics models will drive distribution facilities to strategic locations, often closer to customers’ (Kirschbraun and Bomba, 2000). The new DCs, designed for rapid material flow, are increasingly being established at strategic, time- and space-sensitive locations. These have traditionally been the large gateways of goods flow, particularly large ports and major airports, the centres of manufacturing, and, more recently, highway interchanges with access to a large number of customers. Regarding the large-scale national and international distribution, favourite locations for the future establishment of DCs are those gateways and transportation corridors with access both to traditional logistics interfaces and to large agglomerations and thus consumer markets. Beside some competitive container ports and air freight hubs, two features are interesting in this context. First, major logistics and integrator companies represent such a large freight volume that they can afford to own DCs: FedEx established its major air hub in Memphis/Tennessee, UPS did similarly in Louisville/Kentucky. Second: so-called Inland Hubs are becoming more and more important, where primarily road freight is consolidated,

232

M. Hesse / Resources, Conservation and Recycling 36 (2002) 211 /240

e.g. in the New York/New Jersey suburbs, in Columbus/Ohio or in Indianapolis/ Indiana. Those locations are ideally suited to serve major markets both at the East Coast and in the Midwest (US Congress, 1995). Within urban regions, a second trend toward decentralisation is prevalent. New warehouses, DCs and logistics facilities are moving at the edge of the agglomeration. Given the respective locational behaviour of early logistics companies, this ‘Outward Drift of Freight-Handling’ has already been described in the New York region 40 years ago: ‘Wholesale establishments, warehouses, and terminals for both water and land transport */all of which are characterised by continual in-and-out freight movements*/have been increasing their employment fastest in places outside the congested heart of the Region. And in this respect they closely resemble manufacturing itself’ (Chinitz, 1960). More recently, this might become a significant trend for a majority of the urbanised industrial countries (Hesse, 2002a). Due to decentralisation, the demand for developed land for distribution purposes is likely to increase (Hesse, 2002b). In many cases, new DCs are being established on former greenfields */at places out of the congested urban cores with excellent highway access. In some cases, distribution companies went into existing buildings and brought in their new high-technology equipment, like Webvan.com did in Oakland, CA. The establishment of a new warehouse on a brownfield site, subject to redevelopment in New Jersey, seems to be a rare example (Strauss-Wieder, 2001). However, most existing buildings do not fit into the advanced technological requirements of automated warehouse operations. This means that the apparently positive effect of concentrated distribution in e-commerce (/reduced number of DCs) is challenged by the fact that new e-commerce buildings, as well as most modern DCs, contribute to further land consumption. In the case that new pick-up points have to be established, where customers can receive their e-commerce parcels (instead of staying present at home all day), the respective infrastructure and demand for space has to be taken into account as well. In order to define the particular contribution of e-commerce to this issue, we would need to know how far the current construction activity for warehouses and DCs is related to on-line order and fulfilment. This question is hard to answer, because much of the e-commerce-related delivery is operated by third party logistics providers (3PL). In most cases, 3PL do not exclusively serve e-commerce order fulfilment but offer distribution and warehousing services for a number of customers. Currently, the total number of DCs and large warehouses in the US is estimated to be about 2 50 000 (Dixon, 1999). Compared with that total, the number of facilities specifically devoted to e-commerce retailers or to their service providers is small. E-fulfilment centres were established by UPS and FedEx, being used in addition to the firm’s existing hubs and terminals. Large on-line retailers such as Amazon.com operate six DCs in the US (and three in Europe), and Barnes and Noble.com is distributing out of two new e-commerce DCs (Annual Reports, 2000). These examples show that the physical impact of e-commerce is still limited and its implications are hardly predictable. They also show that e-commerce is far from being virtual or weightless.

M. Hesse / Resources, Conservation and Recycling 36 (2002) 211 /240

233



6. Straight assessments in a complex world /a commentary The purpose of this paper was to explore the implications of e-commerce for freight transport and to comment on how best to assess them. A crucial point is, at least momentarily, that a general statement on the net effect of e-commerce on freight transport and the environment is almost impossible to make. This is due to three factors: the broad variety of distribution practices (supply side), the unpredictable behaviour of the users (demand side), and the weak empirical data both in terms of selected case studies and of the broader context of interrelationships and spatial outcome. We also know almost nothing about cumulative causation: how new technologies, after becoming embedded into significant practices, are related to their social and economic environment and what their long term, so-called second order effects are. Some early judgements about the environmental benefits of electronic commerce in logistics and freight transport are oversimplifications. This is true for the assumption that the delivery of on-line-ordered products to the household eliminates customers’ shopping trips. In most cases, a certain environmental benefit calculated in this incremental approach primarily relies on the elimination of passenger transport, it does not refer to a more efficient freight delivery pattern. First it has to be acknowledged that the logistics system is complex and*/regarding functionality and impacts */generates interdependencies. Second, consumers’ habits usually tend to inertia and unpredicted behaviour rather than calculable action and effects. The discourse on telecommunication and travel has already shown that large substitutional effects as an outcome of new technologies have not materialised yet to the desired extent, and that the opposite is also possible (Mokhtarian, 2000). In turn, this means that the general assertion that ordering books via the Internet benefits the environment still has a weak basis. It is suggested that under real life conditions */not in the quasi-experimental comparison */people behave differently from theory. Predicting individual passenger’s behaviour, which was already called the ‘wild card’ in the context of transportation issues, is known to be difficult (Skinner, 2000). This is also true for corporate strategies. Companies do not necessarily act always on the basis of perfect market information and rational decision making. Competition often drives firms to a certain market presence that may ‘rationally’ not be justified. The notion of the firm as a cold, rational machinery that consists of nothing but the optimal combination of production factors is widely criticised in economic and social science research (Schoenberger, 1997; Hollingsworth et al., 1997). The expectation that ecommerce would, first, be overwhelmingly used by firms and, second, would dramatically increase the efficiency of PD, and third, would thus reduce the environmental burden of freight transport, is constructed and exaggerated. It refers more to conventional wisdom than to real life experience. There are also further doubts about too positivistic judgements. Is it really good for the environment if traffic emissions are likely to grow more slowly than GDP */ as long as they are still growing, and as long as there are no signs of a different trend? By the way, data already show a decoupling of freight transport energy use and GDP

234

M. Hesse / Resources, Conservation and Recycling 36 (2002) 211 /240

growth in most OECD countries since 1973 (Schipper et al., 1997b). Is it a real environmental benefit if Amazon.com constructs a 2 million square feet warehouse that allows for a certain increase in average space efficiency per book */if the DC’s building base is simply added to the existing grey- and brownfields? Most of the infrastructure for the new economy does, at least for now, not replace any of the existing infrastructures and land uses, just as once the telephone did not supersede material travel. There is much evidence for a predominant trend in freight transport to continue: toward smaller delivery units, faster services, increasing distance, higher market shares of truck, van and air services. This is an outcome of interrelated forces that are heading in the same direction. It also means that the overall trends toward higher energy consumption, increased emissions, further land use and additional social cost of transport are likely to continue. Of course, this is an individual judgement, not a statistical finding. It is hard to prove this hypothesis, even if it is plausible. As long as precise data are scant (and models are models, often different from real life), we should at least be careful with general statements.

7. Conclusions This paper considered the significance of electronic commerce for freight transport and PD. Hence, it was trying to go beyond narrow analytical approaches, by reflecting the generic structural change. The main argument of the paper was threefold: first, most recent analyses of freight transport and logistics implications of e-commerce are overstating the current relevance of e-commerce applications on the one hand, and neglecting the influence of the underlying structural change in the entire logistics system on the other. Second, early assessments of the possible contribution of e-commerce to making the distribution system more efficient are probably too optimistic, and underestimate its negative effects. Third, electronic commerce and IT is interrelated components of structural change in distribution. They generically affect the environment by speeding up the time and increasing the geographical area of freight transport operations. E-commerce as a part of the entire distribution system can contribute to certain interrelated impacts: to longer transport distances and often to higher delivery frequencies, to increasing demand for land, due to new transhipment points, and to more frequent air transport. It also has the potential for the opposite, though this is being questioned under current circumstances. This is particularly due to unrealistic assumptions about both distribution efficiency and consumer acceptance that are given in respective e-commerce models. In general, the business related promises of e-commerce stand in remarkable contrast to its plausible negative effects on transport and logistics (see Tables 5 and 6). Possible contributions of e-commerce to a more efficient and sustainable distribution system very much depend on particular regional circumstances, such as consumer habits, certain delivery modes and the important question of population density. In order to be effective, local innovations may also rely on political support

Table 5 E-commerce and logistics /an ambiguous relationship: the promise of e-commerce: business effects Supply management

Manufacturing/assembly

Improve product development

Reduce sourcing costs Reduce inventory costs

Lower transaction costs Strengthen customer relationship Lower work in progress Research new user segcosts ments

Enable collaborative development across firms

Marketing

Distribution

Customer Service

Reduce sales and distribution costs Promote new products and services

Improve customer service Lower customer interaction costs

Source: McKinsey and Company, 2001 (cited in OECD, 2001a Science and Technology Industry/STI, 2001. Fifth Business and Industry Policy Forum: The Internet and Business Performance, p. 39. Paris).

Table 6 The dark side of e-commerce: logistics and transportation Supply chain management

Long distance transport

Warehousing

Local distribution

Introduce and strengthen flexible sourcing and distribution strategies (e.g. ‘Just-in-time’ concepts) Introduce and strengthen large-scale (e.g. global) sourcing and distribution strategies

Increase the market share of trucking and air freight transportation Lower the market share of mass freight transportation modes

Create demand for new high- Decrease local distribution eftechnology, ‘big box’ ware- ficiency through ‘atomisation’ houses (DCs) of consignments Abandon existing storage facilities, due to special ICTrequirements

The last mile: home delivery to customers Increase the volume of local distribution (trips or mileage) Inspire short-term order behaviour by customers

M. Hesse / Resources, Conservation and Recycling 36 (2002) 211 /240

Product/service development

Source: own.

235

236

M. Hesse / Resources, Conservation and Recycling 36 (2002) 211 /240

at state, federal and international levels6. Until now, the economic success and logistics efficiency of e-commerce are challenged by the costly frictions of physical space, way beyond the splendour of sophisticated webpages. Distance, density and topography have by no means vanished. Since the material and the virtual world are inextricably intertwined, goods movement is unlikely to ‘de-materialise’ through the establishment of electronic distribution channels */as telecommuting did not free a majority of people from the daily drive to work, shopping and leisure. In contrast to the notion of the ‘End of Geography’, recent research supports the argument that the apparently ‘weightless’ new economy performs distinct ‘spatialities’, by linking virtual and real spaces in specific ways, and even by following common patterns of regional and urban development (Graham and Marvin, 1996; Pratt, 2000; Zook, 2000). In this context, a final evaluation of the net-effect of transport generating and substituting forces is not possible yet, due to the lack of precise data, case studies and knowledge that allow for general statements on the relationship between this particular technology, society and the environment. However, there is some evidence that e-commerce is likely to reinforce longstanding trends of transport growth, rather than breaking them. Speaking in more general terms, passenger and freight transport have a fundamental societal dimension that has to be considered: They play a major role in the process of ‘mobilising’ modern societies, as far as they allow for extended individual activities, accelerated spatial division of labour, and global trade and exchange. E-commerce is a small constituent in more complex arrangements. Distribution, logistics and transport are major requirements of structural change, and they are shaping the physical and social environment related to those changes. Deterministic assessments do not reflect these broader foundations of technological change. Some of the related developments have already been investigated, primarily with reference to mobility and passenger transport (Gillespie et al., 1998; Rondinelli, 2001). However, we still do not know very much about the role of logistics and freight transport in this context (Hu, 2001). As a consequence, future research should be directed towards the various implications of logistics technologies, organisation and infrastructure (in terms of supply and demand, customer behaviour, environmental outcome, spatial dynamics). This research should also cover the full range of feedbacks and cumulative causations between technology, users and their environment.

6

Sustainable Distribution, 1999. Political regulation, incentives and action plans seem to be essential for incorporating any local and regional activity into a broader policy context. Good examples are given e.g. by the UK strategy ‘Sustainable Distribution’, within the 10 Year Integrated Transport Plan of the British Government of 1999, or the ‘Freight Sustainability Development Program’, recently submitted by ‘Transport Canada’.

M. Hesse / Resources, Conservation and Recycling 36 (2002) 211 /240

237

Acknowledgements ‘‘I am indebted to Heike Fla¨mig, Technical University of Hamburg-Harburg, Germany, for critical discussion and many insights into the world of e-commerce and its implications for transport and the environment.’’

References Abbey DD, Twist DC, Koomen LJ. The Need for Speed: Impact on Supply Chain Real Estate. AMB Investment Management, Inc. White Paper, January 2001, 1. Abernathy FH, Dunlop JT, Hammond JH, Weil D. Retailing and supply chains in the information age. Technology in Society 2000a;22:5 /31. Amazon.com Inc., Form 10-K For the Fiscal Year Ended December 31, 2001: 21. Retrieved from the Internet, www.amazon.com, Feb 26, 2002. Annual Reports, 2000. cf. the Annual Reports 2000 of Amazon (www.amazon.com) or Barnes and Noble (www.bn.com). Bahrami, B., 2001. E-Commerce: Implications for Supply Chain Productivity, Carrier Competitiveness, and Efficient Allocation of the External Costs of Transportation. Paper submitted to The Transportation Research Board, 80th Annual Meeting, January 7 /11, 2001, Washington, DC. Bovet D, Martha J. Value Nets. Breaking the Supply Chain to Unlock Hidden Profits. New York: Wiley, 2000. Bowersox DJ, Smykay EW, LaLonde BJ. Physical Distribution Management. Logistics Problems of the Firm. New York/London: MacMillan, 1968:4. Bowersox DJ, Closs DJ, Stank TP. Ten mega-trends that will revolutionise supply chain logistics. Journal of Business Logistics 2000;21(2):1 /16. Bretzke W-R./KPMG Consulting AG, 2001. Die Zukunft elektronischer Transportmarktplaetze: Nur wenige kommen durch. www.mylogistics.net, 2001. Broens DF, Vanroye K, Demkes R.E-Commerce, Supply Chain Management and Intermodality. Background paper for the Third EU-USA Forum on Freight Intermodalism. Delft/NL, 2000. Brooks M, Wahhaj Z. The economic effects of business to business internet activity. National Institute Economic Review 2001;175:100. Browne M. E-Commerce and urban transport. Paper presented at the joint OECD/ECMT-Seminar ‘The Impact of E-commerce on Transport’, Session 4: Transport and Local Distribution. Paris, 2001. Bundesministerium, 2001. fu¨r Verkehr, Bau- und Wohnungswesen, 2001. Auswirkungen neuer Informations- und Kommunikationstechniken auf Verkehrsaufkommen und innovative Arbeitspla¨tze im Verkehrsbereich. (Federal Ministry of Transport, Building and Housing of Germany, 2001. The impact of new information and communication technologies on transport and innovative jobs in the transport sector). Berlin, 2001. Burnson P. Seeking Solution for E-commerce. With the Internet business clicking off, air freight companies are looking for the legacy in Web systems. Air Cargo World Online, 2001, Feature 4. Cairns S. Delivering Alternatives. Transport Policy 1996;3(4):155 /76. Casas J, Zmud J, Bricka S. Impact of Shopping Via Internet on Travel for Shopping Purposes. Paper submitted to 80th Annual Meeting of the Transportation Research Board, Washington DC, 2001. Cassinfo/ProLogis (eds.), 2000. State of the Logistic’s 12th annual Report. www.cassinfo.com/bob.html. Castells, M., 1985. High Technology, Economic Restructuring and the Urban-Regional Process in the United States. In: Castells, M., editor. High Technology, Space and Society. Urban Affairs Annual Reviews. vol. 28, Sage, Beverley Hills, pp. 11 /40. Castells M. The rise of the network society. In: The Information Age: Economy, Society, and Culture, vol. 1. Oxford, UK: Blackwell, 1996. Chinitz B. Freight and the metropolis. In: The Impact of America’s Transport Revolutions on the New York Region. Cambridge, MA: Harvard University Press, 1960:153.

238

M. Hesse / Resources, Conservation and Recycling 36 (2002) 211 /240

Clancy B, Hoppin D. Converging on Air Freight. The Merge Global 2001 World Air Freight Forecast. In: Air Cargo World online, 2001. Special feature. Colin J. The impact of e-commerce on logistics. Paper presented at the joint OECD/ECMT-Seminar ‘The Impact of E-commerce on Transport’, Session 2: E-commerce and logistics, Paris, 2001. Dixon R. How to Select an Optimal Distribution Site, Site Selection magazine, 1999:3. Dra¨ger, 2000. Review 85 (2000): 28. Lu¨beck, Germany. Ernst and Young, 2000. Global on-line, January (cited in Browne 2001, p. 5). Fla¨mig, H., 2001. E-commerce, Ernaehrung und Umwelt. Gutachten fuer den Ausschuss fuer Bildung, Forschung und Technikfolgenabschaetzung des Deutschen Bundestags. Unveroeff. Forschungsbericht der TU Hamburg-Harburg. Hamburg. (E-commerce, food and the environment. Unpublished expertise for the Commission on Education, Research and Technology Assessment of the German Bundestag. Unpublished Report of the Technical University of Hamburg-Harburg. Hamburg). Forkenbrock D. Comparison of external costs of rail and truck freight transportation. Transportation Research A 2001;35:321 /37. Gillespie A, Healy P, Robins K. Movement and mobility in the Post-Fordist city. In: Banister D, editor. Transport Policy and the Environment. London: E FN Spon, 1998:243 /66. Gould J, Golob TF, Barwise P.Why do People Drive to Shop? Future Travel and Telecommunications Tradeoffs. Working Papers of the Center for Activity Systems Analysis (CASA), UC Davis, WP-98-1. 1998. Graham S, Marvin S. Telecommunications and the City, Electronic Spaces, Urban Places. London, New York: Routledge, 1996:79. Harrison B. Lean and Mean. In: The Changing Landscape of Corporate Power in the Age of Flexibility. New York: Basic Books, 1994. Hesse M. Physical Distribution and The City: Goods movement as an indicator of urban and regional development. In: Davies W, Townshend I. editors. Monitoring Cities of Tomorrow. Proceedings of the 2001 Meeting of the Commission on Urban Development of the International Geographical Union. August 2001. Calgary, Canada (forthcoming) 2002a. Hesse M. Transport and Logistics in City Regions: Driving Forces for Counter-Urbanization? Investigations on the Postmodern Urban Landscape. In: Black WR, Nijkamp P. editors. Social Change and Sustainable Transport. Indiana University Press, Bloomington, (forthcoming) 2002b. Hollingsworth JR, Boyer R, editors. Contemporary Capitalism. The Embeddedness of Institutions. Cambridge, UK: Cambridge University Press, 1997. Hu P. Developing an ‘E-Sensible’ Transportation System. What Are the Research and Data Needs? TR News 2001;216:4 /8. Transportation Research Board, National Research Council. Hultkrantz, Lumsden editors. 2000. Hultkrantz O, Lumsden K, editors. E-commerce and Logistical Consequences. Gothenburg: Chalmers University of Technology, 2000. INFRAS/IWW, 2000. External Costs of Transport: Accident, Environmental and Congestion Costs in Western Europe. Report for the International Union of Railways, Paris, France. Zurich/Switzerland, Karlsruhe/Germany. Janz O. Mehr Verkehr durch E-Commerce? /Eine analyse der Auswirkungen des E-CommerceWachstums (B2C) auf den Verkehr. Zeitschrift fu¨r Verkehrswissenschaften , 1 2001;72(1):48 /69. Jindel S. Delivering E-Commerce. The retail world is in upheaval over E-Commerce */can transport carriers be far behind? Air Cargo World online, 2001, Feature 1. Kirschbraun T, Bomba T. The New Economy. Effects on corporate real estate strategies. In: Jones Lang LaSalle: Property Futures. Occupier Strategies in the New Economy: Chicago, London 16, 2000. Komor P. Reducing energy use in US freight transport. Transport Policy 1995;2(2):119 /28. Matthews H, Hendrickson C, Lave, L. Harry Potter and the Hole in the Ozone Layer. While on-line shopping may appear to cause less impact on the environment than driving to the mall, the net effect of e-commerce remains unclear. In: IEEE Spectrum, 2000. Matthews HS, Hendrickson CT, Soh DL. Environmental and economic effects of e-commerce: a case study of book publishing and retail logistics. Transportation Research Record 2001;1763:6 /12. Matthews HS, Williams E, Tagami T, Hendrickson CT. Energy Implications of Online Book Retailing in the United States and Japan. Environmental Impact Assessment Review, 2002. in press.

M. Hesse / Resources, Conservation and Recycling 36 (2002) 211 /240

239

McKinnon A. Physical distribution. In: Marshall JN, editor. Services and Uneven Development. Oxford, UK: Oxford University Press, 1988:133. McKinnon A. Logistical restructuring, freight traffic growth and the environment. In: Banister D, editor. Transport Policy and the Environment. London: E FN Spon, 1998:97 /109. McKinsey and Company, 2001. Cited in OECD, Science and Technology Industry/STI, 2001. 5th Business and Industry Policy Forum: The Internet and Business Performance, p. 39, Paris. Meyer MD. Delivering The Future: E-Freight. Provided by the Foundation for Intermodal Research and Education, Greenbelt, MD 20770, USA (on the web: http://intermodal.org/FIRE/meyerpaper.html). Morgan, Stanley, Dean, Witter. Internet Strategies: Surveying the Freight Carriers. Freight Transportation and Logistics Industry Report. New York, August 2000. O’Kelly M. A geographer’s analysis of hub-and-spoke systems. Journal of Transport Geography 1998;6(3):171 /86. OECD, 1999. Environmental Data Compendium. Paris, France. OECD, 2000. Organization for Economic Development and Cooperation, 2000. The Economic and Social Impacts of Electronic Commerce: Preliminary Findings and Research Agenda, Paris. OECD, 2001a. Science and Technology Industry (STI), 2001a. Fifth Business and Industry Policy Forum: The Internet and Business Performance, 17 /18, Paris. OECD, 2001b. Working Party on the Information Economy, 2001b. Business-to-Business electronic commerce in publishing, retail distribution and pharmaceuticals distribution in France. Paris: 29 ff. (III. Impact on Market Structures and Competition). OECD, 2001c. Business to consumer electronic commerce. An update on the statistics. Paris. OECD, 2001d. Working Party on the Information Economy, 2001d. Business-to-Business electronic commerce in publishing, retail distribution and pharmaceuticals distribution in France, Paris, 11. Orremo F, Wallin C, Joensson G, Ringsberg K. IT, mat och miljo¨. (IT, food and the environment). English summary of a master’s thesis at Lund Technical University, Sweden, 1999 (unpubl. manuscript). Penner J, Lister D, Griggs D, Dokken D, McFarland M. Aviation and the Global Atmosphere. Special. Pflaum A, Kille C, Wilhelm M, Prockl G. Consumer Direct */The Last Mile. Fraunhofer-Anwendungszentrum fu¨r Verkehrslogistik und Kommunikationstechnik. 2000, Nu¨rnberg. Pratt AC. New media, the new economy and new spaces. Geoforum 2000;31:425 /36. Reynolds J. E-commerce-a critical review. International Journal of Retail and Distribution Management 2000a;28(10):417 /44. Robeson JF, Kollat DT. Channels of distribution: structure and change. In: Robeson JF, House RG, editors. The Distribution Handbook. New York, London: FreePress, 1985:225 /34. Rondinelli D. Making metropolitan areas competitive and sustainable in the new economy. Journal of Urban Technology 2001;8(1):1 /21. Schipper L, Scholl L, Price L. Energy use and carbon emissions from freight in 10 industrialised countries: an analysis of trends from 1973 to 1992. Transportation Research D 1997a;2(1):57 /76. Schipper L, Scholl L, Price L. Energy use and carbon emissions from freight in 10 industrialised countries: an analysis of trends from 1973 to 1992. Transportation Research D 1997b;2(1):57 /76. Schoenberger E. The Cultural Crisis of the Firm. Oxford, UK: Blackwell, 1997. Skinner RE. Transportation in the 21st Century. www.tfhrc.gov/pubrds/septoct00/skinner.htm, 2000. Sobie B. Hubs of Commerce. Airports across the central United States want a piece of electronic commerce, but only the biggest of the big know how to feel fulfilled, Air CargoWorld online, 2000. Feature 1. Song J, Regan A. Transition or transformation? Emerging freight transportation intermediaries. Transportation Research Record 2001;1763:1 /5. Strauss-Wieder A. Warehousing and Distribution Center Context. NJPTA Brownfield Economic Redevelopment Project. Prepared for the New Jersey Institute of Technology and the North Jersey Transportation Planning Authority, 2001. Ten Hompel M, Siebel L. Logistik und E-commerce. Konzepte fu¨r Ballungszentren. Eine Studie des Fraunhofer-IML im Auftrag des Landes Nordrhein-Westfalen. Dortmund. (Logistics and Ecommerce. Concepts for urban regions. A study by the Fraunhofer-IML on behalf of the State of North Rhine-Westfalia. Dortmund), 2001.

240

M. Hesse / Resources, Conservation and Recycling 36 (2002) 211 /240

The Lehman-Brothers, Macro Shock: How Wholesale Industries Are Being Revolutionised. Transcript of Lehman-Brothers sponsored Conference Call featuring Adam Fein, Ph.D., June 22, 1999: 5. www.nawpubs.org. Transport en Logistiek Nederland. Nieuwe wijn in oude zakken. Zoetermeer (Transport and Logistics Association of the Netherlands, 2000. Old wine in new bottles. Zoetermeer). US Congress, Office of Technology Assessment, The Technological Reshaping of Metropolitan America, OTA-ETI-643. Washington, DC: US Government Printing Office, 1995. US Department of Commerce,Digital Economy 2000, Washington DC.:vi, 2000. Womack J, Jones D, Roos D. The Machine that Changed the World. New York: Rawson, 1990. Zook MA. The web of production: the economic geography of commercial internet content production in the United States. Environment and Planning A 2000;32:411 /26.