Asia Pacific Journal of Marketing and Logistics

Emerald Article: Economic development and channel evolution in The People's Republic of China Ling Yi, Eugene D. Jaffe

Article information:
To cite this document: Ling Yi, Eugene D. Jaffe, (2007),"Economic development and channel evolution in The People's Republic of China", Asia Pacific Journal of Marketing and Logistics, Vol. 19 Iss: 1 pp. 22 - 39 Permanent link to this document: http://dx.doi.org/10.1108/13555850710720885 Downloaded on: 05-04-2012 References: This document contains references to 73 other documents To copy this document: permissions@emeraldinsight.com This document has been downloaded 2831 times.

Access to this document was granted through an Emerald subscription provided by LAHTI UNIVERSITY OF APPLIED SCIENCES For Authors: If you would like to write for this, or any other Emerald publication, then please use our Emerald for Authors service. Information about how to choose which publication to write for and submission guidelines are available for all. Additional help for authors is available for Emerald subscribers. Please visit www.emeraldinsight.com/authors for more information. About Emerald www.emeraldinsight.com With over forty years' experience, Emerald Group Publishing is a leading independent publisher of global research with impact in business, society, public policy and education. In total, Emerald publishes over 275 journals and more than 130 book series, as well as an extensive range of online products and services. Emerald is both COUNTER 3 and TRANSFER compliant. The organization is a partner of the Committee on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive preservation.
*Related content and download information correct at time of download.

The current issue and full text archive of this journal is available at www.emeraldinsight.com/1355-5855.htm

APJML 19,1

Economic development and channel evolution in The Peoples Republic of China

Ling Yi
ChongQing Commercial and Technology University, ChongQuin, China, and

22
Received December 2005 Revised April 2006 Accepted June 2006

Eugene D. Jaffe
Department of Marketing, Copenhagen Business School, Frederiksberg, Denmark
Abstract
Purpose The objective of this study is to develop a model for analyzing the relative importance of facilitating and constraining factors that determine changes in the structure of distribution channels as a country progresses along the path of economic development. Design/methodology/approach Data on channel structure were collected on national, regional and municipal levels. The hypotheses in this study were tested by using time series and crosssectional data. The ordinary least squares method was used for multiple regression and correlation analyses. Findings The most significant findings from the above research are that channel structure is determined by economic development and, to some extent, by government policy. Both of these factors work hand-in-hand to affect distribution reform. First, economic development provides the need for more efficient channels, first as more privately-owned intermediaries enter the distribution system, and later as the number of channel intermediaries contracts, partly stemming from vertical integration and from a shakeout of less efficient enterprises. Second, government policy can be formulated to encourage these developments. In order to promote distribution reform, the governments open policy, which allows for market access by foreign firms as well as local ones, should be continued. Research limitations/implications More cross-sectional data for China are needed in order to better analyze the impact of different economic sectors (at different levels of development) on channel evolution. As more data become available, these analyses will be possible. Originality/value Very little is known about the factors that shape a nations channel system. Improvements in distribution systems are crucial to upgrading marketing systems in developing countries. Understanding how improvements in distribution systems contribute to economic development can aid decision makers in developing countries to improve resource allocation. This study has contributed to the theory of channel evolution in developing countries. Keywords China, Channel flow, Economic development, Government policy, Marketing Paper type Research paper

Asia Pacific Journal of Marketing and Logistics Vol. 19 No. 1, 2007 pp. 22-39 # Emerald Group Publishing Limited 1355-5855 DOI 10.1108/13555850710720885

The objective of this paper is to examine the relationship between economic development and channel structure. A number of studies have raised the proposition that a countrys distribution channel structure is a function of economic development (Douglas, 1971; Mallen, 1975; Kaynak and Hudanah, 1987; Olsen and Granzin, 1990; Samiee, 1993; Mallen, 1996). Identifying and explaining the factors that influence channel structure is one of the most important tasks of channel research (Anderson, 1985; Bucklin et al., 1996; Dwyer and Welsh, 1985; John and Weitz, 1988; Klein et al., 1990; Stern and Reve, 1980). As the economy evolves from a planning orientation to a
The authors wish to acknowledge the helpful comments of Gorm Gabrielsen, Klaus Meyer, Tony Pecotich and Hans Stubbe Solgaard.

market orientation, distribution structure reform is a critical factor in this process (Luk and Li, 1997; Cundiff and Hilger, 1988; Olsen and Granzin, 1991). The focus of this paper is The Peoples Republic of China (PRC). The PRC is in transition from a planned to a market economy. Over the last 20 years the PRC economy has undergone significant transformations in the form of privatization, liberal trade policies and a movement away from a centrally planned distribution system (Garten, 1998; Jiang and Prater, 2002; Taylor, 2003). Distribution reform has been a key element in the governments objective of decentralizing the economy. The evolution of distribution structures of developing countries has been the subject of a number of studies, the objectives of which have varied significantly (Samiee, 1993). Some have focused on a single nation or particular channel member, while others have provided cross-cultural insights. However, an examination of the literature devoted to the evolutionary process of distribution structures shows little progress in theory development ( Jain, 1990; Olsen and Granzin, 1991; Mueller et al., 1993; Bucklin et al., 1996). Very little is known about the factors that shape a nations channel system (Sharma and Dominguez, 1992). Improvements in distribution systems are crucial to upgrading marketing systems in developing countries (Samiee, 1993). Understanding how improvements in distribution systems contribute to economic development can aid decision makers in developing countries to improve resource allocation. Therefore, the development of theory in channel evolution is a necessary condition for the implementation of marketing reform. Channel structure A study of channel structure may focus on one or more of three dimensions: the types of channels used, the number of channel levels and the type and number of institutions that comprise the channel (Bello and Williamson, 1985). Studies of channel structure have also investigated channel length (Dwyer and Welsh, 1985; Olsen and Granzin, 1990, Sharma and Dominguez, 1992; Couglan and Lal, 1992), bureaucratization ( John and Reve, 1982; Boyle et al., 1992) and ownership patterns. In this paper, channel structure is conceptualized as the number and size of marketing intuitions in the channel. Channel structure is important from a public policy perspective as many governments in developing countries have put in place vertical distribution systems, bypassing traditional channels. And distribution reform plays a very crucial role in a countrys economic development. The failure of Gorbachevs perestroika was primarily due to a breakdown in the economys distribution system (empty store shelves, widespread hoarding of goods, etc.). Such a breakdown in distribution was actually both a cause and effect of a general loss of public confidence in the former Soviet Unions economy ( Holtzman, 1991), which finally contributed to the fall of Gorbachev and the collapse of the Soviet Union. Some studies focus on the dimension of government economic policy and have demonstrated its influences on market entry decisions (Karakaya and Stahl, 1989; Root, 1994). In addition, several marketing scholars have studied the effects of the economic role of government and associated policy factors on distribution systems in different countries ( Fubara, 1991; Mahmoud and Rice, 1991; Mitchel, 1991). In addition to its importance at the macro level to researchers and public policy officials trying to better understand the forces that shape the organization of the exchange system, channel structure is strategically significant also at the micro level to individual firms and hierarchies attempting to enhance profits and lower costs.

Economic development and channel evolution 23

APJML 19,1

24

Most studies of channel evolution have attempted to determine a relationship between economic development and channel length. However, as Olsen and Granzin (1991, p. 61) assert, the few empirical investigations have been severely limited in scope. Only about half of the studies reviewed in Sharma and Dominguez (1992) were tested empirically, resulting in conflicting findings. Wadinambiaratchi (1965) and Olsen and Granzin (1991) found that channels lengthen as economic development occurs. On the other hand, Livesay and Porter (1969) found that channels shorten. Forman and Riegelhaupt (1970) and Douglas (1971) found that the function is U-shaped, i.e. channels first lengthen and then shorten as was also the case in Olsen and Granzins (1991) empirical study of agricultural channels. Other difficulties in interpreting the findings stem from the diverse methodologies used. Some researchers focused on developed countries; others on developing countries. Most relied on crosssectional data, while only Laytons (1989) and Livesay and Porter (1969) analyze timeseries data. Such differences in methodology may account for the apparent differences in empirical findings (Boddewyn, 1981; El-Ansary and Liebrenz, 1982). Determinants of channel structure Sharma and Dominguez (1992) proposed that channel evolution is a function of economic development and consequent changes in shopping behavior, management style, government intervention and the degree of urbanization. Samiee (1993) proposed similar variables for studying distribution structures in developing economies. Neither Sharma and Dominguez nor Samiee empirically tested the propositions raised in their articles, but gave some suggestions as to how the constructs might be measured. Based on the literature of channel structure and distribution reform, we posit that channel structure is a function of economic development, mobility and environmental variables, which include government policy and urbanization. A test of these relationships is reported below. Research model Our research model posits that channel length is a function economic and environmental variables as conceptualized in Figure 1. Economic development It is believed that channel structure evolves with economic development (Kaynak, 1986; Olsen and Granzin, 1991). Growth in retail and wholesale institutions are attributed to a countrys level of development. Higgins (1959), Samli and Mentzer (1981), Slater (1970) and Wadinambiaratchi (1965, p. 77) associate more levels of distribution and retail outlets with higher economic development. In economies where the producer is also the consumer, or merchandise is exchanged with others in the vicinity of its production, as in agrarian economies, channel functions are transferred to specialists who can perform them more efficiently ( Mallen, 1973, 1996; Olsen and Granzin, 1991; Coughlan and Lal, 1992). At low levels of development there is little specialization between distribution and non-distribution functions ( Mallen, 1975, 1996). As market size grows, distribution becomes more specialized, competitive intensity increases and as a result, distribution firms become larger while less efficient firms exit the distribution system (Olsen and Granzin, 1990; Sharma and Dominguez, 1992; Kim and Frazier, 1996). Thus, it is expected that channels become more productive as measured by average sales per establishment (wholesale and retail) as economic development progresses.

Economic Development
GDP/ per capita

Economic development and channel evolution 25

Structure

Channel

Environment
Urbanization Government Mobility

Figure 1.
Research model

An increase in the standard of living of a developing country is often equated with the consumption of material goods by the private sector. However, while governments expend much effort to increase the production of consumer goods, the distribution sector of the economy is frequently ignored or superficially treated ( Holton, 1953). The principle of convergence ( Bartels, 1981) suggests that the entire marketing system is significantly influenced by the countrys level of development (Samli, 1964; Samli and Mentzer, 1981; Slater, 1970). In addition, the increased sophistication of consumer demand stimulates the growth of the service sector (Taylor, 2003). Developing countries go through a number of stages of economic growth (see, for example, Rostow, 1960). While the PRC as a whole may be close to the takeoff stage according to Rostows terminology, vast parts of the country are still at the traditional and preconditions for takeoff stages. Thus, markets in the PRC are diverse, and becoming more so (Veeck and Veeck, 2000), so that any study of marketing reform in the PRC must use cross-sectional data. That is the approach used in this paper as explained below. The relationship between economic development and channel structure suggested by H1 is similar to an inverted U. As a countrys economy develops there is a concomitant increase in the demand for channel services resulting in increasing numbers of intermediaries, given an absence of administrative entry barriers. The point where the direction of change in channel structure is reversed occurs when a society has begun to achieve a high level of income and wealth accumulation (Sharma and Dominguez, 1992). In the case of the PRC, the inverted U shape may take the form as shown in Figure 2.

APJML 19,1

Channel Structure

26
c b a

Figure 2. Hypothesized channel structure dynamics as a function of economic development or time

Economic Development or Time

Figure 2 shows a hypothesized change in channel structure for the PRC as a function of economic development or time. We expect that the number of intermediaries increases with economic development, and then at some point, the increase slows while average size (as measured by sales) increases. This inverted U-shape function may be explained by looking at several points on the curve in Figure 2. The least developed areas in China should be at area a on the curve, developing areas at areas b and c and the most developed at area d, when average firm size is greatest. Using cross sectional data, time (t) is constant. In time series data of a single region, the economy develops over time. In mixed cross-sectional and time series data together, at each t, different regions are at different points on the curve and move along it over time. To summarize, the inverted U relationship stems from the following:
.

In a developing economy, there is a need for more intermediaries to accumulate and re-sort goods from different suppliers. Demand for a greater variety of products will lead to an expansion in the merchandise-mix ( Luk and Li, 1997). As such, more intermediaries enter the distribution system. The evolution toward larger intermediaries may be explained by two phenomena: (a) As the economy further develops and becomes more service oriented, there is less of a need for indirect channels leading to the development of vertically integrated, larger intermediaries. (b) The growth of urban centers encourages vertical integration in channels in order to improve efficiency.

The preceding discussion proposes that the level of economic development significantly influences the structure of distribution channels. Observations and case studies of markets in the PRC also support this notion ( Xu, 1990; Veeck and Veeck, 2000). Using GDP per capita (GDP/n), a U-shaped function is expected between GDP/n and channel length. Therefore, Hypothesis 1: H1. As GDP/n increases, channel structure takes on a U-shaped function.

Environmental variables Mobility. Automobile ownership enables consumers to shop further from home and to conveniently reach larger stores located in shopping centers. They also enable buying in larger quantities. The car extends both the distance accessed and the weight to be carried, thus reducing the number of shopping trips. Chinas private automobile ownership increased significantly mainly in urban areas. The number of private automobiles was 284,900 in 1985 and increased to 5,338,800 in 1999. Hall et al. (1961), Takeuchi and Bucklin (1977), Ingene and Luschs (1981) measure mobility by the number of automobiles owned by the shoppers. Actually, their measure of mobility is imperfect, pointed out by Ingene and Lusch (1981) because public transport is not considered (Chow, 1995). In this study, consumer mobility is measured by the number of per-capita civil and private vehicle seats, which takes into account both privately-owned vehicles and public transport. Since much of Chinas individual enterprises are located in free market areas that may not be close to residential areas, transportation is necessary for these shoppers. If mobility is increased, shoppers can have a better comparison of the products that they would like to buy and their potential choice sets should move closer to their ideal ones. Consequently, shoppers are going to increase their purchases. Increased mobility may also contribute to greater consumer outreach. The concept of consumer outreach was developed by Goldman (1974). It is defined as the ability and willingness of consumers to venture out of traditional territories and activities (Goldman, 1974, p. 9). The lure of modern retailing institutions, such as supermarkets and discount stores, may not be sufficient to motivate consumers to change their shopping habits. On the one hand, consumers may lack the mobility to reach more distant, but larger stores, because of low incomes and/or the lack of convenient transportation. On the other hand, the habit of shopping in different stores for various consumption items several times a week may be so ingrained that it is difficult to change. If either or both cases prevail, then consumer outreach will be constrained. Therefore: H2. As mobility increases, the average size of channel intermediaries increases. Urbanization. The rapid pace of urbanization is one of the most significant aspects of contemporary economic geography, particularly so in the developing areas. Urbanization, and more specifically, concentration of population in large cities, has significant transactional cost implications. Large urban centers pose distribution challenges and opportunities that small ones do not. Small urban centers can be provisioned directly from surrounding areas, whereas large cities require a vast array of producers from farther distances (e.g. Forman and Riegelhaupt, 1970). Large urban centers afford producers and intermediaries the opportunity to achieve sufficient economies of scale. High urban concentration accentuates the advantages of internal governance to large producers and intermediaries. Large intermediaries are no longer content to rely on small suppliers; they prefer to work directly with large producers in order to be assured of supply continuity (Forman and Riegelhaupt, 1970). In addition to economic development and higher incomes, one of the factors facilitating the growth of supermarkets in China has been the rapid growth of urbanization ( Lo et al., 2001). In summary, it is expected that the presence of large urban markets will spur the growth of large-scale intermediaries, where there is already a sufficient level of development and competition. This leads to hypothesis 3: H3. As urbanization increases, the average size of channel intermediaries increases.

Economic development and channel evolution 27

APJML 19,1

28

The role of government. Several marketing scholars have studied the effects of the economic role of government and the associated policy factors on the distribution system in different countries (Fubara, 1991; Mahmoud and Rice, 1991; Mitchell, 1991; Naor, 1990). Their findings suggest that the role of government in marketing could be regulatory, facilitative, or entrepreneurial (Mitchell, 1991). A government that assumes a regulatory role attempts to establish and maintain an orderly framework for economic activities, whereas the facilitative role requires more direct involvement in infrastructure projects and supporting marketing programs by the government. The entrepreneurial role requires the government to actively invest in and operate business projects. Therefore, the knowledge of the aims of a government and its economic policies is important as this could have profound impact on negotiations and transaction behavior ( Wells, 1977). The PRC implemented central planning more than 30 years ago. Before economic reform, China like other communist countries had a shortage economy. The production and distribution of commodities were centrally planned. Nearly all the existing marketing channels at that time were state-owned. The distribution system was established not according to economic functions, but according to political and administrative functions (Mun, 1998; Qiang and Harris, 1990). At least five problems resulted from this system: (1) The marketing channel was long and steeped in red tape. The final retail price that consumers paid was several times the initial price as the product had to pass through many government middlemen. Moreover, it was difficult for products manufactured in cities to reach consumers in rural areas; similarly, agricultural products also took weeks to reach consumers in nearby cities. (2) The system was highly rigid and impervious to changes. (3) All products were bought and sold by the state distribution system according to the price set by the planning bureau. There was no correlation between the profits or losses of the state distributors and their performance. (4) Products offered to consumers were of limited variety and were of inferior quality. Opportunities for enterprises to vary the quality or quantity of their products were highly restricted. Managers were also unwilling to allow for such opportunities because of the absence of incentives. (5) With insufficient and backward storage facilities, many of the goods were simply lost in storage or shipping processes. The problems that existed at that time were similar to what Gajewski (1992) and Iwinska-knop (1992) discussed in the case of Poland and the case of the Soviet Union mentioned by Holtzman (1991) under central planning. The whole system is basically driven by a central planner rather than by consumer sovereignty. Compounded by the monopolization of distribution by state enterprises, marketing simply has no role to play in such a shortage economy (Ennew et al., 1993; Hooley, 1993). The PRCs economic reform model emphasizes competition over privatization. The Chinese government at first did not attempt to privatize state-owned enterprises directly. Instead, it allowed or encouraged the non-state enterprises sector to compete with and gradually outgrow the state sector. The majority of these non-state enterprises are collective enterprises that are basically a communal organization or a vaguely defined cooperative ( Weitzman, 1993). The collective enterprises are exactly opposite the type of private organizations that are at the center of the East European

model. Although the performance of the state sector also improves under competitive pressure, it may not be enough to save them in the long run. In contrast to the East European model, the Chinese model places more stress on the design of a competitive market structure which, in turn, gradually exerts pressure on inefficient state enterprises that must improve their productivity in order to survive in the market (Chow, 1994). McMillan and Naughton (1992) report that the Chinese reform strategy followed three stages: (1) massive entry of non-state enterprises; (2) a dramatic increase in competition; and (3) improvements in the performance of state-owned enterprises resulting from state-imposed market-like incentives. The PRC has demonstrated how it has successfully used the fundamental market forces of entry and competition in moving its centrally planned economy toward a market economy (Chow, 1994). The evolution of PRC government reformist policy toward its distribution system may be divided into three stages: (1) Introduction of market forces 1978 to 1985. The distribution system was opened to individual businessmen, thereby encouraging private entrepreneurs to compete. Government distribution policy enabled the growth of small businessmen getihu in the distribution system. Greater autonomy was granted to enterprises, especially those run by the state, in managing their business activities ( Luk, 1998). (2) Introduction of market-driven incentives end of the 1980s. A dual distribution system was formed to establish horizontal connections between enterprises across different industries and/or provinces (Luk, 1998). The pricing of products was gradually left to supply and demand. Manufacturers were permitted to sell directly to retailers, bypassing state distributors. (3) Consolidation of wholesale markets mid 1990s. The government promoted supermarket chains in urban areas. In 1992, the retail sector was opened to foreign enterprises, in order to promote the state-owned enterprises reform. Most state-owned distributors were transformed to public companies listed on the stock exchange. Greater autonomy was granted to commercial enterprises and large manufacturers to import and export finished goods and components. By the end of 1996, the role of non-state enterprises had accounted for 73 per cent of consumer goods sales ( Xu, 2000). The above discussion leads to the following hypothesis: H4. As government policy becomes market driven, the average size of channel intermediaries increases. Variable measures Table I lists the criterion and predictor variables, hypotheses and variable measurement. The productivity of a nations aggregate channel system is the composite of all channels. As a result, direct measurement is infeasible. Instead, proxy measures may be used (Sharma and Dominguez, 1992). In this study, we use a measure based on the ratio of the number of wholesale and retail establishments per Yuan

Economic development and channel evolution 29

APJML 19,1

Variables Channel structure

Hypotheses

Measures CTI. Equal to the number of establishments per Yuan 100 million sales GDP/per-capita Mobility measured by total civil and private vehicle seats divided by regional population Population density in urban areas Dummy variable, before 1985 0; from 1985 to 1992 1; from 1992 to 1997 2; after 1997 3

30

Economic development Mobility (M)

As GDP/n increases, channel structure takes on a U-shaped function As mobility increases, the average size of channel intermediaries increases

Environment Urbanization (Ur) Government (G) Table I. Measurement criteria

As urbanization increases, the average size of channel intermediaries increases As government policy becomes market driven, the average size of channel intermediaries increases

100 million sales as an indicator of distribution intensity, which we term channel throughput index (CTI). An increase in the CTI indicates that the average size of distribution intermediaries decreases, and vice-versa, a decrease in the index indicates that average size increases. Model specification and data used in the study The hypotheses in this study were tested by using time series and cross-sectional data. In a country the size of the PRC, regional economic differences are significant ( Jiang and Prater, 2003). This has resulted in uneven economic development between regions and variability in reform policy (Luk, 1998). The ordinary least squares method was used for multiple regression and correlation analyses. The linear regression function is given as: Clt 0 1 GDP=nt 2 Urt 4 Gt 6 Pat t t 1980, 1981, 1982, . . . , 1999 year where: (1) Clt value of the dependent variable in the tth trial, or observation. Clt is channel structure (CTI). (2) GDP/n, Ur, G, Pa the specified value of the independent variable in the tth trial or observation. (3) GDP/n is GDP per-capita; Ur is the urbanization ratio; Pa stands for mobility (seats per-capita) and G is government.  0 intercept of the regression equation.  16 parameters of the regression equation, which indicates the contribution ratio of each factor. 1

Data set Two types of data are the source for hypothesis testing in this study. First, time series data on a national basis for China are available from 1980 to 2000. Partial time series data are also available for Shanghai from 1980 to 2001, for Guangzhou from 1980 to 1998 and for Beijing from 1995 to 2000. Cross-sectional data for 31 administrative regions and cities are available for 1995, 1996, 1998, 1999 and 2002. Thus, total sampling units comprise 155 (number of regions and cities times years) and 755 total data units (number of sampling units times variables). The large sample will reduce the influence of multi or co-linearity among independent variables. All data used in this study were culled from the China Statistical Yearbooks, 1980 to 1999. Findings Regression analysis results We first test for a relationship between channel structure and economic development using both time series and cross-sectional data. As shown in Table II, we find that from 1980 to 1984 (introduction of market forces, see above), average establishment size decreased and in 1985 onwards, average establishment size increased. Overall, there was a steady increase in the number of wholesaling and retailing establishments from 1.7 million in 1980 to 17.7 million in 1999, but as pointed out above, average establishment size increased from 1985. The relationship between GDP/n and the CTI was tested in a regression analysis that resulted in a low R2, but a significant F 0.01[1]. Regression analyses by region using cross-sectional data showed (see Table III) similar results. GDP/n was a significant predictor of channel structure in every year except 2002. U-shaped quadratic functions resulted from both the time series and

Economic development and channel evolution 31

Year 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999

CTI 963.8009 1,186.6973 1,381.4612 2,220.2635 2,644.4906 2,480.5941 2,237.0757 2,118.1075 1,746.2035 1,464.1817 1,484.6428 1,396.1674 1,325.2632 1,115.5183 1,202.6517 1,084.0767 986.6748 953.6123 972.0606 861.1769

GDP/n (unit: Yuan) 460.0000 477.5391 504.0922 549.5180 638.3409 721.7809 761.3587 818.6683 848.6997 803.9346 850.9309 950.9446 1,098.131 1,246.638 1,367.313 1,473.696 1,595.568 1,718.314 1,838.213 1,963.271

Private automobiles (10,000 units) 15.91000 18.68000 21.47000 24.06000 27.88000 28.49000 34.70790 42.29000 60.42000 73.12000 81.62000 96.04000 118.2000 155.7700 205.4200 249.9600 289.6700 358.3600 423.6500 533.8800

Urban population weight % 19.39000 20.16000 21.13000 21.62000 23.01000 23.71000 24.52000 25.32000 25.81000 26.21000 26.41000 26.37000 27.63000 28.14000 28.62000 29.04000 29.37000 29.92000 30.40000 30.89000

Government policy 0.00 0.00 0.00 0.00 0.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00

Table II.
Time series data of China from 1980 to 1999

APJML 19,1
Region Beijing Tianjin Hebei Shanxi Inner Mongolia Liaoning Jilin Heilongjiang Shanghai Jiangsu Zhejiang Anhui Fujian Jiangxi Shandong Henan Hubei Hunan Guangdong Guangxi Hainan Sichuan Guizhou Yunnan Tibet Shaanxi Gansu Qinghai Ningxia Xinjiang

CTI Number of establishments per 100 million Yuan sales 91.74304 163.0772 661.8942 618.6833 579.7862 286.7955 732.6048 503.3868 47.10559 165.7756 275.4633 553.9071 373.0633 717.0123 549.7500 441.3953 551.3011 761.1778 237.7740 767.3872 799.9030 721.0648 726.2984 279.1375 1,911.613 533.4139 728.2960 404.1277 443.5847 283.7227

GDP per capita unit: Yuan 10,265.00 8,164.000 3,376.000 2,819.000 3,013.000 6,103.000 3,703.000 4,427.000 15,204.00 5,785.000 6,149.000 2,521.000 5,386.000 2,376.000 4,473.000 2,475.000 3,341.000 2,701.000 6,380.000 2,772.000 4,820.000 2,516.000 1,553.000 2,490.000 1,984.000 2,344.000 1,925.000 2,910.000 2,685.000 3,953.000

Total civil and private vehicle seats (unit: 10,000 seats) 428.2000 78.95000 329.8500 128.2900 109.6800 455.5400 81.44000 273.3600 174.8400 260.4900 143.4800 120.7000 88.63000 90.80000 267.6100 293.7600 163.2100 168.9300 384.5600 111.1600 48.30000 344.0900 73.77000 119.8500 11.59000 96.95000 185.7300 24.61000 22.73000 132.1800

Urbanization population weight (over total population) % 64.81424 56.77917 17.12181 24.32389 32.52767 44.50946 42.25228 44.73674 70.82538 24.87438 18.36513 17.39451 18.65556 20.88274 24.94431 16.22204 26.23383 17.86090 29.98162 16.55363 23.49954 16.94193 13.51695 13.63844 13.79369 20.54655 17.82777 28.15971 26.92412 34.56147

32

Table III. CTI and economic/ environmental variables by region (1999)

cross-sectional analyses as illustrated by the shape of the curves shown in Figure 3. These findings support H1. Environmental variables as modifiers of channel structure Now we determine to what extent variables other than economic development contribute to changes in channel structure. Both time series and cross-sectional data were used for this purpose. Using again the data contained in Table II, linear regression was used to determine the significance of the predictor variables. The significant F 0.011 is smaller than 0.05 (normal significant F level), so we can conclude that the entire CTI function is significant. According to the results of the test for each independent variables coefficients, urbanization is not significant. However, government policy, GDP/n and mobility are significant. Next, we use cross-sectional data for testing across regions. To illustrate, Table III shows the CTI, GDP/n, mobility and urbanization for 31 regions and cities in the PRC during 1999. Similar data were available for the years 1995, 1996, 1998, 1999 and 2002. In this case, a regression analysis shows that GDP/n, mobility and urbanization are

Channel Throughput Index 3000

Economic development and channel evolution 33

2000

1000

Observed 0 400 600 800 1000 1200 1400 1600 1800 2000 Quadratic

Figure 3.
Chinas channel throughput index change as a function of GDP/n

GDP per capita

significant predictors of the CTI. A summary of the data analyses for the above five years in Table IV indicates that apart from GDP/n, urbanization and mobility, in that order, were predictors of channel structure. In essence, average size of wholesale and retail firms increased as economic development progressed. By region, the transformation to larger-sized intermediaries was greatest where economic development increased the most. This finding also buttresses H1. Because of its nature (dummy variable over time), government policy could not be included in the cross-sectional regression equations. Nevertheless, it was a predictor of channel structure in the time series analysis above. Thus, H4 is also accepted. However, contrary to expectations, wholesaling and retailing institutions did not increase in size as urbanization increased. Also, the mobility indicator, while having a positive influence on institution size, was only significant for 1999. These findings lead to a rejection of H2 and H3. Discussion The above analysis confirms that economic development and government policy influence channel structure. However, the question remains as to why urbanization and mobility were not significant predictors and why urbanization worked in the opposite direction as hypothesized. A possible reason for this is that China with more than 800 million people employed in agriculture has a very low urbanization rate. In 1980, only 19.39 per cent of its people lived in cities or towns. In 1999 Chinas urbanization ratio increased to 30 per cent, but compared with the average urbanization ratio of more than 70 per cent in Western developed countries, its lower urbanization rate constrains the development of the market system. However, the situation in large cities like Shanghai is totally different. For example, Shanghais urbanization ratio increased from 61 per cent in 1980 to 75 per cent in 2001. As the biggest commercial city in China, the urbanization ratio of Shanghai was the highest in China from 1980 to 2001.

APJML 19,1

Year 1995

Formula CTI 8276.92 102 GDP/n (T 8.785), (T 3.773) F 14.234, Significant F 0.001 R 0.581, R2 0.337 Adjusted R2 0.313, DW 2.468 CTI 187.9038.6 102 GDP/n (T 0.743), (T 4.633), (T 2.783) R 0.666, R2 0.443 Adjusted R2 0.403 F 11.144, Significant F 0.000 CTI 1.6761.01 104 GDP/n + 1.767 102 Ur (T 0.613), (T 3.587), (T 1.794) R 0.613, R2 0.375 Adjusted R2 0.331 F 8.414, Significant F 0.001 CTI 1.6537.69 105 GDP/n 6.13 104 M 1.614 102 Ur (T 9.324), (T 3.274), (T 1.853), (T 1.823) R 0.656, R2 0.431, Adjusted R2 0.368 F 6.813, Significant F 0.001 CTI 1.7491.24 102 Ur (T 5.725), (T 1.729) R 0.306, R2 0.093, Adjusted R2 0.062 F 2.988, Significant F 0.095

Significant independent variable GDP/n

Co-linearity No

34
1996

GDP/n

No

1998

GDP/n Ur

No

1999 Table IV. Summary statistics of CTI as a function of economic and environmental variables cross-sectional data

GDP/n M Ur Ur

No

2002

No

Urbanization became the most significant factor affecting channel structure, and in one of the first open cities in China, Shanghais distribution structure developed faster. Many foreign retail companies entered this area in 1985, after local distribution was opened to the private sector. By the end of the 1980s, state-owned distribution companies began to vertically integrate and be converted to stock companies. So, after the 1990s many state-owned companies in the distribution system were converted to stock companies and subsequently listed on the stock exchange. The first supermarket chain was set up in the early 1990s in Shanghai. After that, many commercial corporate groups started to establish chain operations and large distribution groups developed. Supermarkets and chain operations were more integrated and needed fewer channel levels thus increasing channel productivity. For China as a whole, urbanization is still at a low level and the governments main strategy to foster urbanization is to develop small and medium towns and cities. When small and medium cities are developed, more retail and distribution companies will enter the distribution sector, but most of them are small in size. Thus, the average size of distribution establishments decreases when urbanization grows in China over beginning stages of economic development. As the growth of urbanization continues, it is expected that the average size of distribution intermediaries will increase. Summary and conclusions The most significant findings from the above research are that channel structure is determined by economic development and to some extent, by government policy. Both of these factors work hand-in-hand to affect distribution reform. First, economic

development provides the need for more efficient channels, first as more privatelyowned intermediaries enter the distribution system, and later as the number of channel intermediaries contracts partly stemming from vertical integration and from a shakeout of less efficient enterprises. Second, government policy can be formulated to encourage these developments. In order to promote distribution reform, the governments open policy, which allows for market access by foreign firms as well as local ones, should be continued. Moreover, the government has permitted, since 1986, the sale of goods by manufacturers directly to retailers, thereby making some state distributors redundant. In addition, chain stores sell a wide range of goods from necessities to durables, with food products accounting for 63 per cent and clothing 4 per cent of sales (Taylor, 2003). An increase in the proportion of chain stores in distribution channels will increase their leverage over manufacturers, perhaps leading to backward vertical integration (Sun, 2000; Xu, 2000). Moreover, the larger size of distributors and retailers will afford them buying power over manufacturers, resulting in both lower prices to consumers and better quality goods. While the above examples of market access were a condition for Chinas entry into the World Trade Organization, it is expected that most of the restrictions affecting the distribution sector will soon be phased out. Removing such restrictions will encourage more foreign investment, which in turn, will contribute to Chinas economic development. As Wan (1992, p. 307) had pointed put, no matter whether the state firms should be privatized or revitalized, a thriving, self-confident private sector holds the key for reform. While these findings indicate that channel evolution for the country as a whole is a function of economic and environmental variables, some of the results of data analyses show that because a developing or emerging economy has sectors or regions that are in different stages of development, the effect of predictor variables may be different by region. As more cross-sectional data for China becomes available, this issue can be investigated further. Forty years ago, Bucklin (1965, p. 31) stated that the study of the structure of distribution channels was one of the most neglected areas of marketing scholars. Expanding our knowledge of the relationship between economic development and channel structure not only will contribute to the development of marketing theory, but also has practical implications. Understanding these relationships may aid in the more successful transfer of marketing techniques between countries at different stages of economic development. It may also reduce the reliance on trial and error in the implementation of marketing reform in emerging economies. Another question that can be investigated is the extent of efficiency in channels. As channels become shorter, how more efficient do they become? In order to answer these questions, more research of the sort presented in this paper should be carried out in additional countries, especially those in transition toward a market economy. In the case of China, follow-up research can be done as additional data becomes available. Such research would be a welcome addition to theories of channel evolution and structure.
Note 1. The regression analyses may be obtained by contacting the authors. References Anderson, E. (1985), The salesperson as outside agent or employee: a transaction cost analysis, Marketing Science, Vol. 4, pp. 234-54. Bartels, R. (1981), Global Development and Marketing, Grid Publishing Company, Columbus, Ohio.

Economic development and channel evolution 35

APJML 19,1

36

Bello, D. and Williamson, N. (1985), Contractual arrangements and marketing practices in the direct export channel, Journal of International Business Studies, Vol. 16, pp. 65-82. Boddewyn, J. (1981), Comparative marketing: the first twenty-five years, Journal of International Business Studies, Vol. 17, pp. 61-79. Boyle, B., Dwyer, R., Robicheaux, R. and Simpson, J. (1992), Influence strategies in marketing channels: measures and use in different relationship structures, Journal of Marketing Research, Vol. 29 No. 4, pp. 462-74. Bucklin, L. (1965), Postponement, speculation and the structure of distribution channels, Journal of Marketing, Vol. 2, pp. 26-31. Bucklin, L., Ramaswamy, V. and Majumdar, N. (1996), Analyzing channel structures of business markets via the structure-output paradigm, International Journal of Research in Marketing, Vol. 13, pp. 73-87. Chow, K.W. (1994), Entry and exit process of small businesses in P.R. Chinas retail sector, International Small Business Journal, Vol. 15 No. 1, pp. 41-58. Chow, K.W. (1995), Distribution reform and retail structure in China: an empirical analysis of entries and exits of enterprises, Asia Pacific Journal of Marketing and Logistics, Vol. 7 No. 3, pp. 3-25. Coughlan, A. and Lal, R. (1992), Retail pricing: does channel length matter, Managerial and Decision Economics, Vol. 13 No. 3, pp. 201-14. Cundiff, E. and Hilger, M. (1988), Marketing in the International Environment, 2 ed., Prentice-Hall, Englewood Cliffs, NJ. Douglas, S.P. (1971), Patterns and parallels of marketing structures in several countries, MSU Business Topics, Vol. 19, Spring, pp. 38-48. Dwyer, R. and Welsh, A. (1985), Environmental relationships of the internal political economy of marketing channels, Journal of Marketing Research, Vol. 22, pp. 257-62. El-Ansary, A. and Liebrenz, M.L. (1982), A multistage approach to comparative marketing analysis, Journal of Macromarketing, Vol. 2, pp. 59-65. Ennew, C.T., Filatochev, I., Wright, M. and Buck, T.W. (1993), Constraints on the adoption of the marketing concept: the case of the former Soviet Union, European Journal of Marketing, Vol. 27, pp. 21-34. Forman, S. and Riegelhaupt, J.F. (1970), Market place and marketing system: toward a theory of peasant economic integration, Comparative Studies in Society and History, Vol. 12, January, pp. 188-212. Fubara, B. (1991), Socializing grain distribution in kongi statistics for the international market, in Kaynak, E. (Ed.), Sociopolitical Aspects of International Marketing, The Haworth Press, New York, NY, pp. 243-56. Gajewski, S. (1992), Consumer behavior in economies of shortage, Journal of Business Research, Vol. 24, pp. 5-10. Garten, J. (1998), The Big Ten: The Big Emerging Markets and How They Will Change Our Lives, Basic Books, New York, NY. Goldman, A. (1974), Outreach of consumers and the modernization of urban food retailing in developing countries, Journal of Marketing, Vol. 38 No. 4, October, pp. 8-16. Hall, M., Knapp, J. and Winstin, C. (1961), Distribution in Great Britain and North America, Oxford University Press, London. Higgins, B. (1959), Economic Development, Norton and Company, Inc., New York, NY. Holton, R.H. (1953), Marketing structure and economic development, Quarterly Journal of Economics, Vol. 67, August, pp. 344-61.

Holtzman, F.D. (1991), Moving toward ruble convertibility, Comparative Economic Studies, Vol. 33, Fall, pp. 3-66. Hooley, G.J. (1993), Raising the iron curtain: marketing in a period of transition, European Journal of Marketing, Vol. 27, pp. 6-20. Ingene, C.A. and Lusch, R.F. (1981), A model of retail structure, Research in Marketing, Vol. 5, pp. 101-64. Iwinska-knop, K. (1992), Distribution as a barrier to application of marketing in the centrally planned economy, Journal of Business Research, Vol. 24, pp. 19-26. Jain, S. (1990), International Marketing Management, Kent-PWS Publishing, Boston. Jiang, B. and Prater, E. (2002), Distribution and logistics development in China, International Journal of Physical Distribution and Logistics Management, Vol. 32, No. 9/10, pp. 783-98. John, G. and Reve, T. (1982), The reliability and validity of key informant data from dyadic relationships in marketing, Journal of Marketing Research, Vol. 19, No. 4, pp. 517-25. John, G. and Weitz, B. (1988), Forward integration into distribution: empirical test of transactional cost analysis, Journal of Law, Economics and Organization, Vol. 4, pp. 121-39. Karakaya, F. and Stahl, M. (1989), Barriers to entry and market entry decisions in consumer and industrial goods markets, Journal of Marketing, Vol. 53, pp. 80-91. Kaynak, E. (1986), Marketing and Economic Development, Praeger, New York, NY. Kaynak, E. and Hudanah, A. (1987), Operationalizing the relationship between marketing and economic development: some insights from less developed countries, European Journal of Marketing, Vol. 17, pp. 196-208. Kim, K. and Frazier, G. (1996), A typology of distribution channel systems: a contextual approach, International Marketing Review, Vol. 13 No. 1, pp. 19-31. Klein, S., Frazier, G. and Roth, V. (1990), A transactional cost model of channel integration in international markets, Journal of Marketing Research, Vol. 17, pp. 196-208. Layton, R.A. (1989), Measures of structural change in macromarketing systems, Journal of Macromarketing, Vol. 9, Spring, pp. 5-15. Livesay, H.C. and Porter, P.G. (1969), Vertical integration in American manufacturing, 18991948, Journal of Economic History, Vol. 29, September, pp. 494-500. Lo, T., Lau, H. and Lin, G. (2001), Problems and prospects of supermarket development in China, International Journal of Retail and Distribution Management, Vol. 29 No. 2, pp. 66-75. Luk, S.T.K. (1998), Structural changes in Chinas distribution system, International Journal of Physical Distribution and Logistics Management, Vol. 28 No. 1, pp. 44-67. Luk, S.T.K. and Li, H. (1997), Distribution reform in China: a macro perspective and implications for channel choices, Journal of Marketing Channels, Vol. 6 No. 1, pp. 77-104. Mahmoud, E. and Rice, G. (1991), Sociopolitical analysis for international marketing: an examination using the case of Egypts infitah, in Kaynak, E. (Ed.), Sociopolitical Aspects of International Marketing, The Haworth Press, New York, NY, pp. 146-64. Mallen, B. (1973), Functional spin-off: a key to anticipating change in distribution structure, Journal of Marketing, Vol. 37, July, pp. 18-25. Mallen, B. (1975), Marketing channels and economic development: a literature overview, International Journal of Physical Distribution and Materials Management, Vol. 5 No. 5, pp. 230-7. Mallen, B. (1996), Marketing channels and economic development: a literature review, International Journal of Physical Distribution, Vol. 26 No. 5, pp. 42-57.

Economic development and channel evolution 37

APJML 19,1

38

McMillan, J. and Naughton, B. (1992), How to reform a planned economy: lessons from China, Oxford Review of Economic Policy, Vol. 8, pp. 130-43. Mitchell, L. (1991), The political element and marketing in Grenada, in Kaynak, E. (Ed.), Sociopolitical Aspects of International Marketing, The Haworth Press, New York, NY, pp. 207-21. Mueller, R., Wenthe, J. and Baron, P. (1993), Case note: the evolution of distribution systems, International Marketing Review, Vol. 10 No. 4, pp. 36-52. Mun, K. (1998), Chinas retailing in a changing environment, in Kaynak, E. (Ed.), Sociopolitical Aspects of International Marketing, Walter de Gruyter and Co., Berlin, pp. 211-26. Naor, J. (1990), Normative and political aspects of socialist marketing policy in transition, European Journal of Marketing, Vol. 24 No. 1, pp. 44-61. Olsen, J. and Grazin, K. (1990), Economic development and channel structure: a multinational study, Journal of Marketing, Vol. 10, Fall, pp. 61-77. Olsen, J. and Granzin, K. (1991), Economic development and channel structure, Journal of Macromarketing, Vol. 10, Fall, pp. 61-77. Qiang, Z. and Harris, P. (1990), Retailing reform and trends in China, International Journal of Retail and Distribution Management, Vol. 18 No. 5, pp. 31-9. Root, F. (1994), Entry Strategies for International Markets, Lexington Books, New York, NY. Rostow, W.W. (1960), The Stages of Economic Growth, Cambridge University Press, New York, NY. Samiee, S. (1993), Retail and channel considerations in developing countries: a review and research propositions, Journal of Business Research, Vol. 27, pp. 103-30. Samli, A. (1964), Wholesaling in an economy of scarcity: Turkey, Journal of Marketing, Vol. 28 No. 3, pp. 55-59. Samli, C. and Mentzer, J. (1981), A model for marketing in economic development, Columbia Journal of World Business, Vol. 16, pp. 91-101. Sharma, A. and Dominguez, L. (1992), Channel evolution: a framework for analysis, Journal of the Academy of Marketing Science, Vol. 20 No. 1, pp. 1-15. Slater, C. (1970), Market channel coordination and economic development, in Bucklin, L.P. (Ed.), Vertical Marketing Systems, Scott, Foresman, Glenview, IL, pp. 135-47. Stern, L. and Reve, T. (1980), Distribution channels as political economies, Journal of Marketing, Vol. 44, pp. 52-64. Sun, P. (2000), New characteristics of Chinas consumer markets, Jingji Daobao, 8 May. Takeuchi, H. and Bucklin, L. (1977), Productivity in retailing: retail structure and public policy, Journal of Retailing, Vol. 53, Spring, pp. 35-46. Taylor, R. (2003), Chinas consumer revolution: distribution reform, foreign investment and the impact of the WTO, Asian Business and Management, Vol. 2, pp. 187-204. Veeck, A. and Veeck, G. (2000), Consumer segmentation and changing food purchase patterns in Nanjing, PRC, World Development, Vol. 28 No. 3, pp. 457-71. Wadinambiaratchi, G.H. (1965), Channel of distribution in developing economies, The Business Quarterly, Vol. 30, Winter, pp. 74-82. Wan, H.J., Jr. (1992), On markets, planning & marketing reform, Academia Economic Papers, Vol. 20 No. 2, pp. 291-312. Weitzman, M.L. (1993), Economic transition: can theory help?, European Economic Review, Vol. 37, pp. 549-55. Wells, L.T. (1977), Negotiating with third world governments, Harvard Business Review, January-February, pp. 72-83. Xu, B. (1990), Marketing in China: One Billion New Customers, NTC Business Books, Lincolnwood.

Xu, X.M. (2000), The development of Chinas retail sector in the twenty-first century, Shangye Jingji, Vol. 7, pp. 21-4. Further reading Bucklin, L. (1970), National income accounting and distributive trade cost, Journal of Marketing, Vol. 34, pp. 14-22. Rosenbloom, B. (1987), Marketing Channels, 3rd ed., Dryden, Chicago. Corresponding author Eugene D. Jaffe can be contacted at: ej.marktg@cbs.dk

Economic development and channel evolution 39

To purchase reprints of this article please e-mail: reprints@emeraldinsight.com Or visit our web site for further details: www.emeraldinsight.com/reprints