The Impact on Bed and Breakfast Prices: Evidence from Airbnb in China

Abstract

As a new type of accommodation and a new way of life, it is of great significance for the spatial optimization and price management of the tourist accommodation market to explore the spatial differentiation characteristics of and influencing factors on bed and breakfast (B&B) house prices. Taking Shandong B&B merchants on the Airbnb website as the research object, this paper discusses the spatial characteristics of and influencing factors on B&B prices in the Shandong province, combining spatial autocorrelation analysis and interpolation analysis to identify the B&B cluster region. Quantile regression was used to reveal the main influencing factors. The results show that: (1) the spatial agglomeration effect of B&B prices in the Shandong province is obvious, and the high value areas form a new pattern between the provincial economic circle and the Jiaodong Economic Circle; (2) the influence of different factors on B&B house prices is very uneven in space. From the regional point of view, there are “sub-regional effects” on the spatial distribution of the influences of various factors on B&B house prices. The results of the study provide references for reasonable pricing, scientific site selection, and spatial optimization of B&Bs.

1. Introduction

Common prosperity is one of the main goals of China and a common aspiration of the Chinese people. At present, China is firmly committed to achieving the goal of common prosperity. However, the problem of inadequate development imbalance in China is still outstanding, and there is a disparity between urban and rural areas, which hardly fulfills the promise of common prosperity. Implementing a rural revitalization strategy is the inevitable choice needed to realize the common prosperity of all people. Only by realizing rural revitalization can China fundamentally solve its poverty problem. Rural revitalization and industrial prosperity form the foundation as well as the focus. Tourism is a strategic pillar industry of China’s national economy and a focus of rural industrial revitalization.. As an advanced developmental form of rural tourism, B&Bs play a positive role in excavating and protecting human history and natural ecology, reshaping rural charm, increasing farmers’ incomes, and promoting the overall tourism industry chain. B&Bs empower rural revitalization, which in turn promotes common prosperity.

The B&B is a brand-new concept which has been increasingly used in the process of tourism development in mainland China in recent years. Mainly imported from Europe, the United States, Japan and Taiwan, B&Bs have gradually presented unique spatial entities and conceptual definitions in local practice. “Basic Requirements and Evaluation of Tourism B&B” (LB/T065-2019) was released by China’s Ministry of Culture and Tourism in July 2019, and “Tourism B&B” was formulated by the Department of Culture and Tourism of the Shandong Province in 2020. The meaning of the term “B&B” is clearly defined in both the Classification and Evaluation of B&Bs, which define a B&B as a small accommodation facility, rich in local culture and living characteristics, opened by residents through the use of relevant idle resources and with the help of family side businesses. As an emerging accommodation industry, B&Bs have received extensive attention from the government and academia for their authenticity and uniqueness.

The B&B is a product of the tourism market catering to the diversified needs of consumers. B&Bs, as a kind of non-standardized accommodation with temperature and mood, play an important role in helping rural revitalization, revitalizing idle resources, and increasing employment income. With the support of relevant policies, B&B companies have good development momentum. According to the data provided by the “B&B Blue Book: China Tourism B&B Development Report (2019)”, as of 30 September 2019, there were 169,477 B&Bs (inns) in mainland China, and the Shandong province ranked fifth in the comparative ranking of provinces, with 10,639 B&Bs [1]. In February 2020, with the approval of the provincial government, the Shandong Provincial Department of Culture and Tourism issued joint departmental guidance on the high-quality development of tourist B&Bs. This indicates that the tourist B&B industry in the Shandong province has entered a high-quality development stage. The high-quality development of the B&B industry not only invigorates tourism resources, but also played an important role in promoting industrial upgrading and revitalization. Although the consumer market is in a short-term downturn due to the COVID-19 epidemic, with remarkable achievements in the prevention and control of the epidemic in China and the policy bias of the central government to “accelerate the formation of a new development pattern with domestic big circulation as the main body and domestic and international double circulation promoting each other”, demand for tourism will soon be released. If businesses can apply the refined management and reasonable pricing strategy of the B&B, they will seize the opportunity to make a big turnaround after the epidemic.

Price is usually one of the most important factors for tourists to consider when choosing accommodation. This is also the key advantage for B&B businesses [2]. Behind B&B prices is a multiple-player game between B&B owners, tourists, and competitors. At present, existing research into influencing factors on B&B prices mainly focuses on location characteristics [3], architectural characteristics [4,5], neighborhood characteristics [6], etc. Research through characteristic price models or ordinary least squares with quantile regressions found that although B&B pricing weakens the influence of factors such as transportation facilities [6], B&B hardware features with its internal environment and B&B owner characteristics have a relatively large premium on its [4,7,8,9]. However, the hardware features of a B&B, its internal environment, and the features of its hosts lead to higher premiums. Meanwhile, Wu et al. incorporated trust and socialization into the index system of influencing factors to reveal the important role of non-economic factors in price [3]. Overall, the spatial differentiation of B&B prices is a mapping of the differentiated segregation of accommodation space and the imbalance of the relationship between urban socio-spatial subjects. As the level of urban economic development and people’s accommodation needs change, the influencing factors of B&B prices need to be constantly changing and improved in order to effectively reveal the characteristics of the spatial differentiation of B&B prices and their mechanisms of action.

However, existing B&B research is limited to investigating spatial differences. Studies have mainly explored the spatial distribution of B&Bs and the factors influencing their prices in regions with more mature B&B development systems. For example, previous studies have taken Shanghai, Beijing, Hangzhou, and other developed cities as research objects, and used methods like nuclear density [10], spatial nearest neighbor analysis [11], and geographical probe [12]. This paper studies the spatial differentiation characteristics of and price-influencing factors on B&B prices. However, the characteristics of the price mechanism formed in the mature stage of the B&B development system cannot be applied to the high-quality development stage of B&Bs, and the research on an individual city can no longer be used to fully explain the price mechanism changes caused by spatial differences. Therefore, it is necessary to explore the spatial differences in pricing decisions of B&Bs in the high-quality development stage, and further explore the price mechanism characteristics of the B&B market in the high-quality development stage.

Secondly, existing research mainly discusses price-influencing factors from the perspective of micro-factor differences within cities, while little of the literature analyzes the impact on the prices of B&Bs from the perspective of provinces. As tourists are the most frequent visitors through Airbnb, Airbnb’s independent label must be taken into account in price research [13]. Previous studies only considered variables such as distance, score, etc., but seldom considered the influence of the Airbnb label. Especially from the perspective of the province, the impact of the label will change a lot.

Therefore, in this study, so as to fill the above-mentioned gaps using big data to analyze the spatial heterogeneity of and influencing factors on provincial spatial B&B prices, we attempt to answer these research questions: (1) If spatial heterogeneity is considered at the provincial level, what is the distribution range of high and low price bands? (2) How do the spatial characteristics, architectural characteristics, service characteristics, and reputation characteristics of the region affect the provincial price? This study combines Arcgis 10.8, Stata 26.0, and other analysis software, and uses interpolation analysis and quantile-regression to identify clusters of and analyze influencing factors on B&B prices, focusing on exploring the influence of various factors on prices under the spatial imbalance of resources, and providing references for scientific site selection and reasonable pricing of B&Bs, as well as optimizing resource allocation. The structure of this research is shown below. We will review previous studies related to the spatial distribution of B&Bs and price-influencing factors and try to understand the spatial heterogeneity in pricing strategies. It will introduce the research fields and methods. Then, we present the spatial heterogeneity characteristics of prices in the Shandong province and discuss the influencing factors in terms of the whole area and each region, which will be followed by a discussion and conclusion.

2. Data and Methods

2.1. Study Area

The study area included the whole Shandong province, including 16 prefecture-level cities such as Jinan, Dezhou, Zibo, Liaocheng, Weifang, Binzhou, Tai’an, Heze, Linyi, Jining, Rizhao, Zaozhuang, Dongying, Qingdao, Yantai, and Weihai (Figure 1). The Shandong province is located on the east coast of China, at the lower reaches of the Yellow River, with a land area of 155,800 square kilometers and extremely rich tourism resources. In 2019, Shandong received 938 million tourists, and the total tourism revenue was CNY 110.8730 billion. In recent years, relying on the rich cultural resources and continued prosperity of the cultural tourism market, coupled with the encouragement and guidance of the party committees and government levels, the Shandong province’s B&B industry has taken advantage of this upward momentum.

2.2. Data Source

Airbnb is the originator of online short-term rent platforms for B&Bs, and its basic function is to provide private rooms for tourists [14,15]. Airbnb merchants are mostly personal hosts, which is more in line with the basic function of a B&B, and it is typical to use it as a data source (citation). The price point of the data in this study was January 2022. Based on web crawler software, 15,200 pieces of data were captured in the Shandong province. The data was cleaned on the basis of deleting inactive houses, houses with no reviews, or those with unrealistic prices, and finally 9712 pieces of data were retained. The data attributes obtained include the name of the B&B, number of reviews, rating, address, latitude and longitude, type of house, price per night, number of people that can be accommodated, description of the house (introduction to the B&B written by the owner), etc. There are various types of Airbnb rooms, and for the whole set of listings, this study takes the total price of all room types divided by the number of room types to find the price of the average room, so that the prices of B&Bs with different room types are comparable. In order to ensure the authenticity of the data, random telephone verification of the data was conducted according to the phone number provided by the website. The data is basically accurate.

There are various types of geospatial information which apply to a B&B. For the study, provincial and national road data were obtained through the website of the National Basic Geographic Information Center; city center and scenic area data were obtained through Gaode Map; and administrative boundary data were obtained through the website of the Institute of Geographical Sciences and Resources of the Chinese Academy of Sciences.

In this study, the longitudinal and latitudinal coordinates of 9712 B&B samples were projected onto a map using ArcGIS 10.8 software, as shown in Figure 2. It can be seen that the B&Bs in the Shandong province are mainly distributed around the coast, and clustering occurs in Jinan, Tai’an, Rizhao, Qingdao, Yantai, and Weihai.

There are various factors influencing the prices of B&Bs. When data are available, four first-level indicators include location characteristics, architectural characteristics, service characteristics, and reputation characteristics, according to the spatial differentiation characteristics of B&B prices in the Shandong province, with reference to the literature on the influencing factors on B&B prices [3,16,17,18,19], hotel prices [20,21,22,23], and residential prices [24,25,26]. From the perspective of tourists’ demand, the choice of residential location is a trade-off between accommodation cost and tourism cost under the budget constraint of tourists, so the distance from provincial roads, national roads, scenic spots, and city centers will influence tourists’ accommodation choice [27]. At the same time, as people’s living standards improve, tourist’s requirements for the living environment become more refined, and the impact of architectural features, such as the size of the accommodation space and whether it has a garden terrace, on the price of the B&B becomes more prominent [6]. The B&B owner is the soul of the B&B, and their unique lifestyle and attitude towards life will have an impact on the price, as will the ability to cancel orders in a timely manner and the availability of additional services such as parking [28]. In this age of advanced information, the wide availability of the Internet has made online booking even more popular, so online ratings and post-stay reviews have become the main reference for tourists choosing B&Bs [29].

In summary, 12 indicators representing location characteristics, architectural characteristics, service characteristics, and reputation characteristics were used as explanatory variables, and B&B prices were used as explanatory variables to comprehensively measure the influencing factors on B&B prices in the Shandong province. The specific variation descriptions and quantification methods are shown in

Table 1. Simple description of model variables and descriptive statistics.

Feature Classification	Explanatory Variables	Variable Description	Variable Assignment	Average Value	Standard Deviation
Dependent variable	Price	Price per room per night on Airbnb (CNY)	According to the collected data	213.70	148.70
Location Features	Provincial roads	The distance of the B&B from the provincial road Distance (m)	According to ArcGIS calculation results	3811.00	2908.00
	National highway	The distance of the B&B from the provincial road Distance (m)	According to ArcGIS calculation results	7292.00	5953.00
	Scenic Area	Distance of B&B from the nearest scenic spot above 4A levels (m)	According to ArcGIS calculation results	3884.00	4262.00
	Downtown	Distance of B&B from city center Distance (m)	According to ArcGIS calculation results	16,855.00	22,889.00
Architectural Features	Accommodation space	Accommodation space per room Number (rooms)	According to the collected data	2.94	1.70
	Garden	With or without garden	Dummy variable, 1 or 0	0.05	0.22
	Terrace	Whether it has a terrace or not	Dummy variable, 1 or 0	0.02	0.12
Service Features	Awesome Landlord	Labeling according to certain rules of the platform	Dummy variable, 1 or 0	0.41	0.49
	Cancellation Policy	Flexible and timely cancellation policy	Dummy variable, 1 or 0	0.40	0.49
	Free parking	Free parking space	Dummy variable, 1 or 0	0.61	0.49
Reputation Features	Rating	Guest Ratings	According to the collected data	2.96	2.35
	Number of Comments	Number of Guest Reviews	According to the collected data	11.30	21.16

.

2.3. Methods

2.3.1. Global Spatial Autocorrelation Analysis

Global spatial autocorrelation analysis is mainly a description of the distribution of attribute values of spatial geographic elements across a whole region, and it is used to determine whether the attribute values have spatial agglomeration characteristics in space [30]. A global spatial autocorrelation analysis of B&B prices in the Shandong province can effectively determine the clustering characteristics of B&B prices in spatial distribution. In this study, the global Moran’s I index was used to measure the degree of global spatial autocorrelation of B&B prices in the Shandong province, which was calculated as follows [31]:

$$I=\frac{n}{S_0}\times \frac{{{\displaystyle \sum }}_{i=1}^n{{\displaystyle \sum }}_{j=1}^n{W}_{ij}\times \left(x_i-\overline{x}\right)\times \left(x_j-\overline{x}\right)}{{{\displaystyle \sum }}_{i=1}^n{\left(x_i-\overline{x}\right)}^2}$$

(1)

where:$n$ is the total number of spatial locations; $W_{ij}$ is the spatial weight matrix, with 1 for spatial adjacency and 0 for non-adjacency; $x_i, x_j$ denote the observed values on spatial geographical unit $i$ and $j$, respectively; $\overline{x }$ is the mean value of regional variables; and $S_0$ is the sum of all elements of the spatial weight matrix. The global Moran’s I takes values between −1 and 1; when Moran’s I is greater than 0, it indicates the existence of positive spatial autocorrelation; when Moran’s I is less than 0, it indicates the existence of negative spatial autocorrelation; when Moran’s I is close to or equal to 0, it indicates the absence of spatial autocorrelation [32].

2.3.2. Empirical Bayesian Kriging Interpolation

The empirical Bayesian kriging interpolation method is based on spatial autocorrelation and unbiased optimal estimation of regionalized variables with a limited range through the structured nature of semi-variance functions, which can accurately predict data that are generally unstable in degree [33]. This method fully takes into account the structural characteristics of the distance relationship and the spatial distribution of observations and can transform each discrete price point data into a continuous surface to accurately analyze the spatial distribution of B&B prices in the Shandong province [5]. The specific function expressions are as follows [34]:

$$Z\left(x_0\right)={\displaystyle \sum }_{i=1}^n{\lambda }_{i}Z\left(x_i\right),$$

(2)

where: $Z\left(x_0\right)$ is the B&B price of the unknown sample point (CNY/night); $Z\left(x_i\right)$ is the B&B price of the known sample points around the unknown sample point (CNY/night); ${\lambda }_i$ is the weight of the B&B price of the $i$ known sample point (CNY/night) to the B&B price of the unknown sample point (CNY/night); and $n$ is the number of known sample points.

2.3.3. Quantile Regression

Quantile regression is a linear regression model of log unit B&B price (CNY/night) $y$ at different quantile points between 0 and 1, where B&B price influences $X$. Unlike the ordinary least squares (OLS), quantile regression uses a weighted average of the absolute values of the residuals as the minimization objective function, which can help to understand the different locations of the distribution of the disturbance terms, can explain the different marginal effects of the variable $X$ on the variable $y$, and is less susceptible to extreme values. The results are more robust [35]. This study uses a combination of both methods to study the factors influencing the price of B&B, which allows a more in-depth analysis of the trend situation of the influence of various factors on the price. The quantile regression model equation is [36]:

$$Q_{\theta }\left(y|X\right)=X^{\prime }\beta \left(\theta \right)$$

(3)

where: $y$ is the log unit B&B price (CNY/night); $X$ is the B&B price influence factor; $Q_{\theta }\left(y|X\right)$ is the value of B&B price (CNY/night) at the $X$ the quantile given the B&B price influence factor $\theta$; $\beta \left(\theta \right)$ is the regression coefficient of B&B price (CNY/night) at the $\theta$ the quantile.

3. Analysis and Discussion of Results

3.1. Analysis of Spatially Divergent Characteristics of B&B Prices in the Shandong Province

3.1.1. B&B Price Space Autocorrelation Analysis

Using GeoDa to calculate the global spatial autocorrelation of B&B prices in the Shandong province, the results show that the global Moran’s I index of B&B prices in the Shandong province is 0.2436, and Z = 45.6112, |Z| > 2.58, p = 0.001, p < 0.01 at a 0.01 significance level, which is statistically significant. This indicates to some extent that B&B prices in the Shandong province have a significant and positive spatial autocorrelation, i.e., the change of B&B prices in one city within the Shandong province is simultaneously affected by the joint effect of local and neighboring regions. Meanwhile, the points further away from the axis are mainly distributed in the first quadrant (high-high clustering), indicating that the global positive spatial correlation of B&B prices in the Shandong province is mainly caused by the spatially clustered distribution of areas with high B&B prices.

3.1.2. Analysis of the Spatial Differentiation Pattern of B&B Prices

Using Equation (2), the empirical Bayesian kriging method was used to spatially interpolate the B&B price data in the Shandong province, and the interpolation results were tested by applying the cross-validation method. The results showed that the standard mean value was −0.005646376, which is close to 0; the standard root mean square was 1.008707, which is close to 1; and the mean standard error was close to the root mean square. The test results indicated that the model prediction effect was more satisfactory and suitable for spatial interpolation prediction of B&B prices, and the spatial distribution map of B&B prices was obtained on this basis (Figure 3).

(1)

The Prices of B&Bs in the Shandong province Show Spatial Agglomeration

The prices of B&Bs in the Shandong province show an upward trend from west to east and from north to south in space. High-priced B&Bs are concentrated around provincial capitals and coastal cities, while medium- and low-priced B&Bs are concentrated in inland areas. The average price of a B&B is CNY 213.70/night, the price level is relatively low and needs to be further improved. From the perspective of each prefecture-level city, Jinan, Qingdao, Yantai, Weihai, Rizhao and most of Weifang and Tai’an have relatively high B&B prices, forming a cluster of high-value areas. Although B&B prices in the Shandong province have an upward trend in coastal areas, such areas account for a low proportion relative to the province and have little impact on overall prices, so from an overall perspective, B&B prices are still at a low to medium level.

(2)

B&B Price High-Value Group Distribution is a “Two Circles” New Pattern

The five cities of Qingdao, Yantai, Weihai, Rizhao, and Weifang form the heart of the Jiaodong agglomeration. From Figure 3, we can see that the high-value areas of B&Bs in the Shandong province tend to expand from the inside to the outside along the sea, forming a high-value cluster centered on the Jiaodong Economic Circle and the provincial capital economic circle, with obvious group location distribution.

The high-priced agglomeration area, with the Jiaodong Economic Circle as its core, mainly includes the five cities of Qingdao, Yantai, Weihai, Rizhao, and Weifang. A total of 7312 B&Bs were counted in the region, and their B&B prices fluctuated from CNY 48 to CNY 1768 per night, with an average B&B price of CNY 223.90 per night, higher than the average B&B price in the Shandong province. In the Jiaodong Economic Circle, Qingdao, Yantai, and Weihai all have small price peaks, which are related to the quality of urban economic development and consumption levels. In terms of GDP per capita in 2021, Qingdao, Yantai, and Weihai are ranked among the top five in the Shandong province, with CNY 140,400/person, CNY 122,700/person, and CNY 119,200/person, respectively. The increase in GDP per capita has led to an increase in consumption levels, and the Qingdao, Yantai, and Weihai regions, with their superior geographical and economic advantages, have made a reasonable allocation of resources to B&Bs and reflected them in their prices, with the help of the city’s own outward expansion and internal development.

The high-priced agglomeration area, with the provincial capital economic circle as its core, mainly includes the cities of Tai’an and Jinan. A total of 1621 B&Bs’ data were obtained in this region, with B&B prices fluctuating from CNY 45 to CNY 1591/night, and the average B&B price was CNY 179.92/night. The price level of B&Bs in the provincial capital economic circle is significantly lower than of the Jiaodong Economic Circle, and lower than the average price of B&Bs in the Shandong province. This is mainly because there are many low-end B&Bs in the region, which lowers the overall price level, so the average price is low. Although there is a spatial gradient in the price of B&Bs in the provincial capital economic circle, the development guidance and support policies are relatively sound. For example, the city of Jinan has issued “Implementation Opinions on Accelerating the Development of the B&B Industry” and “Jinan City B&B Management Measures”. Under the guidance of these policies, high-end B&Bs have emerged, mainly in the vicinity of Jiu Ru Shan Waterfall Group Scenic Area, Red Leaf Valley Ecological and Cultural Tourism Area, Zhujiayu Tourism Area, Taishan Forest Hot Spring Tourism Area, and Taishan Fantasy Fun World in the Taishan District. The prices of B&Bs relying on these scenic spots are higher than those in the surrounding areas, thus pushing up the prices in the surrounding areas, forming a small price peak.

(3)

The Spatial Gradient Difference in the Low Value of B&B Prices is Obvious with the Clustering in the Northwestern and Southern Regions of Lu

The northwestern and southern Lu regions mainly include nine cities, namely Zibo, Liaocheng, Dezhou, Binzhou, Dongying, Heze, Jining, Zaozhuang, and Linyi. A total of 779 B&Bs were found in this region, with B&B prices fluctuating from CNY 40 to 790 per night, and the average price of B&Bs was CNY 158.45 per night, which is lower than the price level of B&Bs in the Shandong province. The number of B&Bs in the northwest and south of the Shandong province is small, and the overall price level is not high. The uneven regional economic development makes obvious differences to the spatial distribution of B&Bs, and the differences in tourism resources and economic development levels make B&B prices lower in this spatial context.

3.2. Analysis of Factors Influencing the Prices of B&Bs in the Shandong Province

3.2.1. Analysis of the Results of the Factors Influencing the Prices of B&Bs from a Region-Wide Perspective

The comparison table of OLS and quantile regression (

Table 2. OLS and quantile regression comparison.

Variables	OLS	0.1	0.25	0.5	0.75	0.9
Constants	266.756 ***	122.255 ***	158.677 ***	204.660 ***	295.689 ***	458.564 ***
Provincial Roads	−3.437 ***	−0.001 ***	−0.002 ***	−0.003 ***	−0.005 ***	−0.006 ***
National Highway	2.146 ***	0.000 **	0.001 ***	0.002 ***	0.003 ***	0.005 ***
Scenic Area	−2.874 ***	−0.001 ***	−0.001 ***	−0.001 ***	−0.003 ***	−0.005 ***
Downtown	−0.000 ***	−0.000	−0.000	−0.000 ***	−0.000 ***	−0.001 ***
Accommo-dation Space	9.928 ***	9.887 ***	12.926 ***	12.731 ***	8.671 ***	7.222 *
Garden	−16.965 ***	−17.170 ***	−19.788 ***	−12.731 ***	−15.641 ***	−21.419 ***
Terrace	56.671 ***	27.592 ***	35.107 ***	60.187 ***	99.079 ***	131.387 ***
Superb Landlord	−13.262 ***	−1.182	−4.160	−5.815 ***	−14.247 **	−26.312 **
Cancellati On Policy	−11.767 ***	−4.253 ***	0.221	−3.970 ***	−11.639 ***	−35.777 ***
Free Parking	−28.058 ***	−17.144 ***	−21.826 ***	−24.584 ***	−32.710 ***	−42.588 ***
Rating	−4.278 ***	−0.111	−0.502 **	−1.135 ***	−4.963 ***	−12.534 ***
Number of Comments	−0.223 ***	−0.024	−0.045	−0.164 ***	−0.283 ***	−0.181
Observatio-ns	9268.000	9268.000	9268.000	9268.000	9268.000	9268.000

*** p < 0.01, ** p < 0.05, * p < 0.1.

) shows that the effects estimated using OLS regression are not constant in quantile, and most of the confidence intervals of quantile regression fall outside the confidence intervals of OLS. An in-depth analysis of the factors influencing B&B prices based on the OLS and quantile regression comparison table (Table 2) shows the following results:

(1) From the perspective of location characteristics, the impact of provincial roads on B&B house prices is significantly negative, indicating that the closer the B&B is to the provincial road, the higher the house price will be. According to the estimation of quantile regression, the coefficient decreases with an increase in one quantile, which indicates that the influence of provincial roads is more significant for low-priced properties. There is a positive correlation between the national highway and the house prices; i.e., the closer to the national highway, the lower the house price. The reason for this situation may be that, on the one hand, with the change of traffic tools, the impact of national roads on tourists gradually decreases, while the role of provincial roads as arteries connecting cities and towns in the province gradually strengthens, linking various areas together, and facilitating the flow of tourists between different scenic spots [4]. On the other hand, it is also possible that B&Bs are close to the national highways in order to form economies of scale, attract more tourists, and use the convenience of transportation to increase their competitiveness. The coefficient of a scenic spot is 0.01. The attraction coefficient is negative at the significant level of 0.01, indicating that the closer you are to the attraction, the higher the price, which is in line with the normal expected result. Downtown location is significant under the OLS regression and the variable passes the significance test and has a negative chef - fi - cent at all quartiles except for the 0.1 and 0.25 quartiles under the quantile regression. Downtown locations offer more convenience to visitors in terms of transportation and entertainment, and the increased convenience of living will increase the propensity to live and therefore the price of accommodation will be higher.

(2) As far as building features, living space and terraces are concerned, their coefficients are significantly positive in OLS and quantile regression results. The larger the accommodation space, the higher the price of the room, which is consistent with a study by Gibbs et al. [13]. Generally speaking, an increase in accommodation space increases the actual area of the property, so the house price of the B&B is higher. It brings a high-quality experience to tourists, so the price of the B&B is also relatively high. The coefficients from the terrace quantile regressions show an increasing trend, indicating that terraces have a more pronounced positive impact on higher-priced properties than on lower-priced properties. In the OLS and quantile regression results, the garden coefficient is negative. It is found that B&Bs with gardens are mostly rented rooms by residents of the community, and the gardens are public gardens within the community rather than exclusive gardens that the B&B offers as included. The landlords used the garden as an attraction and also took advantage of the low price of the house in the hope of gaining more visitors.

(3) As far as service characteristics are concerned, under OLS regression, the coefficient of the most senior landlord is significantly negative. Under the quantile regression, the variable passed the significance test at all quartiles except for the 0.1 and 0.25 quartiles, indicating that B&Bs with the Superlative Host label were instead less expensive. For a B&B owner to earn the Super Awesome Host label, he or she must have completed at least 10 trips or three orders in a year, have received a total of 100 nights of accommodation, have a response rate of no less than 90%, a cancellation rate of no more than 1%, and maintain an overall rating of 4.8 or above. To meet these requirements set by the platform, it is necessary to attract customers with lower prices and increase the occupancy rate of the house. After receiving the awesome landlord label, the exposure of the B&B will increase and the revenue potential will naturally increase as a result. The cancellation policy has a significant negative coefficient in all but the 0.25 quantile, which is consistent with the findings of Benítez-Aurioles [29]. Generally speaking, free parking is negatively correlated with room price; that is, B&Bs which provide free parking have lower prices. Due to the non-steerable nature of the accommodation industry, B&B owners may combine low prices with additional services such as liberal cancellation policies and free parking to attract visitors. Meanwhile, Wang et al. [8] argue that the reason B&B owners set up cancellation policies closely relates to their emotional factors; they do not care about the amount of revenue, but simply offer a reasonable price for tourists who really want to stay.

(4) As far as reputation characteristics are concerned, the rating coefficient under OLS regression is significantly negative, indicating that the higher the rating, the lower the price. In addition, in the quantile regressions, all quantiles except 0.1 quantile passed the significance test, and their coefficients gradually decreased with an increase in quantiles, which may be because the ratings are more significant for the list of low-priced ones. According to the inference of Zhang et al. [37], this may be because tourists have lower expectations of Airbnb’s low-priced listings, and thus are easily satisfied and likely to give higher scores. The coefficient on the number of comments is significantly negative at the level of 0.01, indicating that the more comments, the lower the price. This is consistent with studies by Zhang et al. [37] and Gibbs et al. [13], for the reasons that can be found in their conclusion: Airbnb B&B owners are more likely to attract more guests by lowering their prices, and more reviews represent less information asymmetry and are therefore less likely to overestimate prices.

3.2.2. Analysis of the Results of the Factors Influencing the Prices of B&Bs from a Regional Perspective

In order to study the similarities and differences of the influencing factors of Airbnb listing prices in different areas of the Shandong province, according to the spatial differentiation characteristics of B&B prices in the Shandong province, 16 cities in the Shandong province are divided into three categories: the Jiaodong Economic Circle, the provincial economic circle, and northwestern and southern Shandong. OLS and quantile regression were used to study whether there was any change in the influencing factors of room prices in these three types of regions. The specific influencing factors are shown in

Table 3. Results of OLS and quantile regression correspondence in the Shandong province.

Main Influencing Factors	Jiaodong Economic Circle	Provincial Capital Economic Circle	Northwest Lu and South Lu Regions
	Rizhao, Weifang, Qingdao, Yantai, Weihai	Jinan, Tai’an	Liaocheng, Dezhou, Binzhou. Dongying, Heze, Jining. Zaozhuang, Linyi, Zibo
Provincial Roads	(-)		(-)
National Highway	(+)	(+)
Scenic Area			(+)
Downtown	(-)	(+)
Accommodation space	(+)	(+)	(+)
Garden	(-)
Terrace	(+)	(+)
Superb landlord	(-)	(-)	(-)
Cancellation Policy	(-)		(-)
Free Parking	(-)	(-)	(-)
Rating	(-)	(-)	(-)
Number of Comments	(-)	(+)	(+)

Note: The positive and negative signs in parentheses represent the different directions of influence, with “-” indicating that the factor has a significant negative impact on prices and “+” indicating that the factor has a significant positive impact on prices.

.

(1) Among many factors, accommodation space in architectural features, superb landlord and free parking in service features, and rating in reputation features have significant impact on accommodation price in various regions of this province. Therefore, for provincial B&Bs, expanding accommodation space can effectively help B&B owners raise the price of their B&Bs. Therefore, an initial investment in B&B space is conducive to the subsequent long-term business income. Extra services, such as awesome landlords and free parking, are also common marketing tactics of B&B owners. At the same time, the actual situation of B&Bs in the Shandong province is not in line with the expectations of tourists, and there is a trend of high prices and low scores. Therefore, it is important to consider the reasonableness of the price and focus on the score from the client in the subsequent development of the B&B.

(2) Among the location characteristics, provincial roads, national roads, city centers, and scenic spots have different degrees of influence on B&B prices in all regions of the province. They have a significant negative impact in the Jiaodong Economic Circle, and northwestern and southern Shandong. National highways and city centers have a significant impact in the Jiaodong Economic Circle and the provincial economic circle. National highways have the same impact direction, but the city centers have the opposite impact direction. Scenic spots only have a significant positive impact in the northwestern and southern areas of Shandong. In the Jiaodong Economic Circle, and northwest and south Lu, most of the B&Bs are concentrated in the countryside, and accessibility becomes the key factor for tourists choosing accommodation. The Shandong Provincial Department of Transportation issued the “Adjustment Plan for Ordinary Provincial Roads in Shandong Province (2015–2030)”, which clearly stated that the coverage rate of provincial roads and towns was over 90%. The popularity and optimization of provincial road resources have promoted the development of B&Bs in both areas, and tourists directly enjoy the convenience of transportation brought by trunk roads. The prices of B&Bs near provincial roads are higher, so there is a significant negative effect in terms of B&B prices. In the two economic circles, economic development has led to the development of high-speed rail and airplane transportation, so that tourists have more choices to travel, and the influence of national highways on them gradually lessens and becomes a breakthrough for low-priced marketing, so national highways have a positive influence on the price of B&Bs in the economic circle. The positive impact on the city center in the provincial capital and economic circle may be due to the difference in the type of properties, where the high-priced properties are mostly single-family houses and located in the suburbs, so the price will increase. In northwest and south Lu, B&Bs started relatively late and depend on the distribution of scenic spots, so prices are high.

(3) Among the architectural features, gardens have a significant negative effect in the Jiaodong Economic Circle, and terraces have a significant positive effect in the two economic circles. For B&B owners in the Jiaodong Economic Circle, gardens have gradually been integrated into the marketing strategy, so B&B owners can raise the premium price of B&Bs by providing rich room configurations. Patios are a plus for B&B prices, enriching both the spatial configuration of B&Bs and enhancing communication among people, which enhances the ability of B&Bs to generate sustainable income.

(4) Among the characteristics of service and word-of-mouth, a good cancellation policy has a significant impact only in the Jiaodong Economic Circle, while the number of comments has a different influence direction in different region. The quality level of B&B development in the Jiaodong Economic Circle is high, and the competition among B&B owners is becoming increasingly fierce, making it more difficult to make profits, so the cancellation policy can be used to attract potential users’ booking demand and increase profitability. The direction of the influence of the number of reviews on B&Bs prices within the Jiaodong Economic Circle is consistent with the overall situation in the Shandong province, but the provincial capital economic circle and the north-western and southern regions of Lu show a local positive correlation in terms of the influence of the number of reviews on B&B prices. The higher the number of reviews, the higher the price, which is consistent with the findings of Lu. The higher the number of comments, the greater the customer need. Tourists in these two areas may be price-sensitive, and they are more inclined to express their feelings because they value the increase in marginal utility caused by price increase, so price is positively correlated with the number of comments.

4. Conclusions and Implications

4.1. Conclusions

In this study, spatial autocorrelation analysis and interpolation analysis were used to analyze the spatially divergent characteristics of the prices of B&Bs in the Shandong province, and the influencing factors were also analyzed with the help of the least squares method and quantile regression. The main conclusions are as follows.

First, the spatial differentiation of B&B prices in the Shandong province has obvious clustering characteristics, forming a pattern of high price distribution in “two circles”. Although the average price of a B&B in the Shandong province is low, the areas with high B&B prices show a certain level of clustering, forming a high-priced agglomeration area, with the Jiaodong Economic Circle and the provincial capital economic circle as its cores. With good geographical and location advantages, these areas concentrate various high-quality resources such as commercial transportation, and the resulting space dividends has a significant premium on B&B house prices. The space provided by B&Bs is a refraction of social space at the residential level, which has dual attributes of physical space and social space. In a certain space, the similarity of B&B characteristics and the external effect of public facilities tend to make B&B prices gather in space, and further evolve into a spatial stereotype of high-priced B&Bs. At the same time, due to the imbalance of regional economic development level, the low-priced areas are mainly distributed in the northwest and south of Shandong.

Secondly, the degree of influence of different factors on B&B prices is spatially heterogeneous. From a region-wide perspective, accessibility and tourist experience are important indicators to consider when pricing B&Bs, and different marketing strategies can lead to deviations in B&B pricing. Specifically, location characteristics such as distance from provincial roads, downtown locations, and scenic spots have significant negative effects on B&B prices; the closer you are, the higher the price. Among the architectural features, the living space and terraces had significant positive impact on B&B house prices. In terms of service features, superb hosting, good cancellation policy, and free parking had significant negative impact on B&B house prices. In terms of reputation characteristics, B&Bs with higher ratings tended to have lower prices, while the higher the number of comments, the lower the price. From a regional perspective, the influencing factors that B&B owners need to pay attention to vary from region to region. B&Bs in the Jiaodong Economic Circle and the provincial capital economic circle can seek to improve their house prices through improving accessibility and convenience, and by improving the hardware and facilities of their houses. B&Bs in northwest and south Lu can adopt a policy of selling more at lower prices, but they should pay attention to the number of tourist reviews and meet their needs in a timely manner.

4.2. Implications

The B&B is a new lodging form derived from the tide of the sharing economy. This study examines and portrays the spatially divergent characteristics of B&B prices in the Shandong province and provides insights into the factors affecting B&B prices. We propose the following developmental suggestions.

(1) Strengthening the location advantages and enhancing the spatial relevance of B&Bs. The spatial differentiation of B&Bs in the Shandong province shows that high-priced B&Bs have significant spatial agglomeration characteristics, and the analysis of influencing factors also shows that location advantages and traffic accessibility have significant influences on prices. Therefore, for the provincial economic circle and the Jiaodong Economic Circle, it is suggested that government departments can further strengthen the construction of transportation infrastructure and give financial support to cities with low transportation accessibility, so as to enhance the correlation between B&Bs and transportation accessibility environment and attract large market groups with high purchasing power. B&B owners should strengthen their location advantage, improve the hardware facilities of their B&B, and increase the product premium with convenient transportation and perfect infrastructure.

(2) Enriching additional services and reducing travel costs for tourists. Additional services such as “cooking, free parking, flexible cancellation, luggage storage” have increased the high frequency of communication between hosts and tourists and made the tourists’ accommodation experience gain higher value than expected. Tourists appear to gain more through such additional services than the value generated by the original facilities and necessary services. For B&Bs in northwest and south Lu, the provision of additional services enhances room availability, improves tourist satisfaction, and reduces tourist travel costs, making it easier to achieve a marketing strategy that results in more sales at lower margins.

(3) Enhancing tourists’ sense of experience and reasonable pricing for room availability. Compared with the traditional hotel accommodation industry, the B&B as a new form of accommodation industry, due to the influence of non-standardized forms and information asymmetry, poses certain challenges to tourists’ sense of experience in terms of product services and accommodation quality. Therefore, B&Bs should pay attention to spatial variability in the design process, so as to promote the communication between tourists. In the process of B&B pricing, the relevant departments should guide B&B owners to employ rational pricing, taking into account the impact of room features and supporting facilities on prices, narrowing the gap between tourists’ interests and accommodation prices, and enhancing tourists’ consumer experience.

This paper systematically analyzes and discusses the spatial distribution characteristics of and influencing factors on B&B prices in the Shandong province, but it has some shortcomings. First of all, this study only collected the house price and POI data of a B&B for a certain period and did not reflect the spatial and temporal distribution characteristics of B&B house prices. Secondly, the lack of research on the landlord’s attributes, government policies, housing vacancy rate, and other factors may lead to the incomplete construction of the theory of influencing factors.