In this paper, we use Python crawler to automatically collect price information of online and offline electrical appliances on e-commerce websites, collect price information of offline electrical appliances market through field research, and compare and analyze commodity price and dispersion of online and offline electrical appliances market using market price dispersion model. The results show that although there is some price dispersion in the online market, the online household appliances market is still lower than the offline market, and the market efficiency is higher. Moreover, the discrete degree of online and offline prices has obvious “festival effect” in China’s “Double Eleventh and Double Twelfth” online shopping festivals.
In the 1990s, some scholars put forward the theoretical hypothesis that e-commerce market will lead to “frictionless market”―The development of e-commerce will lead to the gradual disappearance of price dispersion among different retailers in the market: e-commerce market can effectively reduce search costs, improve information transmission efficiency and transparency, so as to enhance market efficiency [
Stigler [
According to the China E-Commerce Report [
Past literature shows that in recent years, the main research objects of this kind of empirical research are [
Based on previous studies, the author selected the most representative household electronic products as the research object. Considering the author’s energy and convenience, the scope of data collection and research is set in Guangzhou, China. As a first-tier city in China, Guangzhou has a large number of e-commerce businesses, sufficient market inventory, fast price updates, and rich data resources. According to China E-Commerce Report [
After a comprehensive survey of the strength and data capacity of major e-commerce platforms. The author chooses four online e-commerce platforms, namely Jingdong, Suning, Tianmao(the official flagship store of Taobao) and Gome, to price the selected electrical appliances. In order to ensure data quality, the author tries to select official flagship stores and self-owned commodity data. Off-line data collection points are Gome Electrical Appliances in Victoria Square, Suning Electrical Appliances in Tianhe Road and Suning Electrical Appliances in Zhengjia Square (Suning and Guomei are the two largest offline shopping malls for household appliances in China), totally 3 stores in Tianhe District of Guangzhou City.
The author selected nine categories of electrical appliances (washing machines, refrigerators, air conditioners, mobile phones, TV, computers, kitchen appliances, toilet appliances, living appliances), a total of 90 kinds of household appliances. Data acquisition once a week. The observation time is from August 26, 2018 to December 16, 2018. We continuously observed 20 periods of online and offline data, collected 6988 sample values of online commodity prices and 6732 sample values of offline commodity prices, totaling 13,720 sample values.
The core part of this subject is data acquisition and collation. This paper needs a large number of accurate and scientific data with long time span to support the hypothesis, which makes the conclusions drawn in this paper more scientific and illustrative. In the past, the data acquisition methods of this kind of research are manual acquisition, with low efficiency and accuracy, and there are many household appliances, long data acquisition period and huge collection time cost. The author compiles Python program to meet the needs of data acquisition, and realizes the automatic collection of online data.
The target data is the online price of the same type of goods (a total of 90 goods) on four shopping websites. The author obtains the price data through the analysis of the commodity details page. The price loading mode of the target website is asynchronous loading, and the price data can’t be obtained by the conventional data crawler. The author uses selenium (python-controlled tools for testing web applications) and chrome to implement price data crawler, and uses xlwt package to write the final result to excel file.
The efficiency of e-commerce market can be examined from the perspectives of price level, price dispersion coefficient and price adjustment frequency. Following is a statistical analysis of the sample data collected by the above data collection methods to meet the requirements of comparability and integrity.
There are several indices to measure the dispersion: range, variance, standard deviation, coefficient of variation, fractional difference, coefficient of dispersion, etc. The coefficient of variation (the standard deviation of the same commodity price divided by the average price of the commodity) is selected as the index to measure the discrete situation. Specific price discretization is shown in
1) Stigler [
2) Overall, there is a certain gap between the online and offline price dispersion of all kinds of electrical appliances, and the offline price dispersion is greater than the online price dispersion. The overall difference was 7.94%.
3) The difference of online and offline dispersion of different types of electrical commodities is different. Among them, the disparity between online and offline prices of computers is 11.45%, while the disparity between online and offline prices of mobile phone (4.44%), air conditioner (5.00%) and TV (7.60%) is relatively small.
In order to test whether the difference in sales price between the two types of retailers is significant statistically, the author further tests the sample data with non-parametric tests that do not require too high sample size and distribution. The original hypothesis H0 is: there is no significant difference in the sales price between the two types of retailers. The alternative hypothesis H1 is that there is a significant difference in the selling price between the two types of retailers. In this paper, the test results of 20 sample data are shown in
Category | Washing machine | TV | Mobile phone | Computer | Refrigerator |
---|---|---|---|---|---|
Online average price | 4260 | 7265 | 3841 | 5552 | 5193 |
Offline average price | 5832 | 9104 | 4207 | 8858 | 6994 |
Online price dispersion coefficient (%) | 11.18 | 10.82 | 4.56 | 8.46 | 12.93 |
Offline price dispersion coefficient (%) | 20.32 | 18.42 | 9.00 | 19.91 | 22.03 |
Category | Air conditioner | Kitchen appliances | Toilet appliances | Living appliances | Population |
Online average price | 5765 | 2925 | 782 | 1825 | 3517 |
Offline average price | 6546 | 4176 | 1020 | 2927 | 4704 |
Online price dispersion coefficient (%) | 8.74 | 14.04 | 16.93 | 12.08 | 10.47 |
Offline price dispersion coefficient (%) | 13.74 | 22.40 | 25.11 | 21.33 | 18.42 |
Phase | 1st | 2nd | 3rd | 4th | 5th | 6th | 7th | 8th | 9th | 10th |
---|---|---|---|---|---|---|---|---|---|---|
Z value | −2.135 | −1.067 | −0.077 | 0.845 | 1.137 | −0.314 | −1.766 | −0.309 | 2.118 | 0.972 |
P value | 0.454 | 0.286 | 0.939 | 0.524 | 0.276 | 0.753 | 0.077** | 0.379 | 0.056** | 0.191 |
phase | 11th | 12th | 13th | 14th | 15th | 16th | 17th | 18th | 19th | 20th |
Z value | −1.503 | 1.177 | 2.010 | −1.004 | 1.521 | 0.003 | −0.235 | 1.033 | −2.482 | −5.05 |
P value | 0.133 | 0.246 | 0.044* | 0.334 | 0.136 | 0.998 | 0.814 | 0.321 | 0.013* | 0.613 |
*: There is a significant difference in the price between the two types of retailers at the 5% significant level. **: There is a significant difference between the price of the two types of retailers at 10% significant level.
As can be seen from
To sum up, this paper collects the prices of online and offline electrical appliances by means of programming, field research and other methods. And statistical analysis of these price data to verify whether e-commerce can effectively reduce market friction. The results show that the price of online and offline electrical appliances market has a large degree of discreteness, but the market efficiency of online market is higher, and its price discreteness is less than offline discreteness, the final empirical conclusion supports the “frictionless hypothesis” (e-commerce reduces market dispersion). Moreover, there is obvious “festival effect” during the online shopping festival.
The author declares no conflicts of interest regarding the publication of this paper.
Fu, R.R. (2019) An Empirical Study of Online and Offline Price Deviation in B2C Market. Open Journal of Business and Management, 7, 519-524. https://doi.org/10.4236/ojbm.2019.72035