Is collaborative consumption the new gamble to speed digital knowledge up in the accommodation sector?

Abstract

Purpose

The present article aims to examine the development of the theoretical framework surrounding collaborative consumption (CC) standards in recent years regarding European short-stay accommodation booking platforms. The sharing economy has significantly impacted the tourist accommodation market in recent years. Starting with the use of experimental data on CCs published on Eurostat in 2019, this article analyzes the correlation between choices of CCs for short-stay accommodation, employment and the economic crisis.

Design/methodology/approach

A vector autoregressive panel approach was applied to investigate the correlation between CC short-stay accommodation choices using panel organization data from 561 EU regions between 2018 and 2021.

Findings

Analyzing the connection between the main data panel variables, a positive correlation was found, followed by an increasing trend in CC use. A self-multiplying effect is generated; that is, the more people use CC, the more electronic captures occur. Consequently, the improvement of CC use and knowledge-intensive activities in short-stay accommodation is strongly linked with employment and GDP.

Originality/value

The originality of the investigation is to examine with a cross-sectional panel data overview the reasons that can push stakeholders to adopt CC and to clearly define a new perimeter of research in terms of the endpoint of CC in short-stay accommodation. Furthermore, the study seeks to assess the end-point congruence to utilize CC as a new gamble to accelerate digital knowledge in the hospitality sector.

Citation

Gravili, G., Avram, A. and Benvenuto, M. (2024), "Is collaborative consumption the new gamble to speed digital knowledge up in the accommodation sector?", Management Decision, Vol. ahead-of-print No. ahead-of-print. https://doi.org/10.1108/MD-05-2023-0873

1. Introduction

Over the past decade, collaborative consumption (CC) has disrupted and transformed traditional service providers into sectors, challenging existing business models, radically modifying managerial imperatives and incorporating profound changes in organizational structures, processes and cultures. CCs “have emerged as a new public channel for matching up knowledge providers with knowledge seekers” (Wang et al., 2020); thus, they represent new phenomena focused on cooperation and organizational innovation, tools that allow to dynamically change and review the knowledge creation processes. In this “sharing economy” (Park and Armstrong, 2017), the CCs become places where knowledge changes, allowing easy access to resources at a lower cost thanks to sharing (Geissinger et al., 2019) and changing thinking towards the concept of ownership (Botsman and Rogers, 2010). The CCs become innovative channels for generating new organizational learning processes. Due to their impetuous force, in recent years, the repercussions of these dynamics in organizations have been manifold; hence, CC represents not only an opportunity but also a threat to traditional supply chains (Murillo et al., 2017; Tescașiu et al., 2018).

In this still-open scenario, it is irrefutable that to unleash positive effects, CC-based organizations need to focus attention on the ability to attract individuals who can stimulate innovative behaviors or who know how sharing techniques and principles could be acquired. Consequently, they need to always identify and select individuals who adapt quickly to change (Oxer, 1998).

The CC phenomena-based companies' possibility to develop knowledge sharing processes depends on their ability to develop processes to support the acquisition, creation, sharing, use and storage of knowledge. The implicit assumption is that there is an organizational archetype that contributes to knowledge sharing. The challenge for these organizations is attracting or training employees and developing a culture of knowledge that involves the intentional actions of the organization and all its members (Gravili and Fait, 2016). Therefore, CC-based organizations are increasingly facing dynamic environments that require adaptation to high speed (Baskerville and Smithson, 1995). However, the time to change conventional structures is often short, so it is become easier for organizations to identify new structures in which interconnectivity and interdependence, driven by advanced information technologies, replace traditional organizational structures and linkages (Scott-Morton, 1991; Baskerville and Smithson, 1995).

For example, COVID-19 has increased computerized work activities, transforming a large number of workers into “knowledge workers” (Zuboff, 1984), who find themselves operating in a situation of continuous challenge in which the real objective is to understand how to improve the context of information and communication by facilitating the dissemination of explicit and tacit knowledge. In these contexts, CCs become effective places for the creation and dissemination of knowledge; they can retain dynamic knowledge and encourage evolutionary processes in real-time, thus becoming places in which to deposit static knowledge and where to develop formal and informal collaboration. The first is favored and encouraged by the organization, and the second is individuals' skills.

Based on this brief literature, the purpose of this study is to investigate the CC phenomenon to explore the real-world determinants of service transactions via CC knowledge-sharing economy platforms. The goal is to examine the correlation between CC choices for short-stay accommodation, employment and the COVID-19 crisis effect, starting from the recently used experimental CC data published in Eurostat in 2019 to clearly define a new perimeter of research in terms of the endpoint of CC in short-stay accommodation.

After an analysis of the literature on short-stay accommodation, it will be applied to investigate the correlation between CC short-stay accommodation choices using a balanced panel approach and data from 561 EU regions between 2018 and 2021 by consumers. The discussion and implications follow, based on the results obtained.

2. Theoretical background

The purpose is to present a brief overview of the existing literature on CC. Collaborative consumption is a phenomenon that has been analyzed since the 1970s, when it was first defined as an “event in which one or more persons consume economic goods and services in the process of engaging in joint activities with one or more others” (Felson and Spaeth, 1978). With the advancement of innovation, the concept has evolved, and today CC is considered a multidimensional construct (Lamberton and Rose, 2012; Bardhi and Eckhardt, 2012; Botsman and Rogers, 2010) that involves a multiplicity of aspects that reflect different approaches.

From a technological viewpoint, “collaborative consumption is a phenomenon strictly combined with the computer-mediated economy” (John, 2013; Harvey et al., 2017). From a collaborative perspective, CC is considered the most developed form of cooperation (Kaplan and Haenlein, 2010; Rodrigues and Druschel, 2010), which is an essential tool for knowledge creation and sharing. Knowledge sharing is, according to numerous studies, the key activity of the CC community (Di Gangi et al.,2012; Hartono and Sheng, 2016; Chua et al.,2019; Hsu et al.,2007; Hamari et al., 2016).

According to this perspective, Rodrigues and Druschel (2010) describe CC as systems in which knowledge generation is high, such as knowledge sharing between entrepreneurs employing strategies (Martin, 2016) and developing a lot of use of resources and products that already existed (Nwaorgu, 2018).

Consequently, CC phenomena are “pervasive contexts for diverse kinds of information sharing for a wide range of purposes by dispersed participants,” thus aiding in the sharing of knowledge between people inside, outside and across companies (Rice et al., 2019). Therefore, by allowing people to interact with outside actors directly and exchanging knowledge, CC phenomena affirm themselves as important environments for knowledge exchange (Malhotra and Majchrzak, 2019; Rice et al., 2019).

Thus, CCs have become an innovative phenomenon in which knowledge sharing does not represent a risk but an opportunity to develop companies' competitive behavior and contribute to a further boost to economic growth.

Collaborative consumption phenomena extend to a growing number of communities and cities around the world, which use technological networks to do more with less through activities such as renting, lending, swapping, bartering, gifting or sharing products on a scale that was previously unimaginable (Małecka et al., 2022). For these reasons, its economic impact is accompanied by a conceptual transformation of work that will make collaborative or participatory consumption a significant potential source of employment in the coming years (Dall-Orsoletta et al., 2022).

The performed analysis shows not only the fragmented nature of existing research on CC in short-day accommodation but also the progressive nature of this area of research (Huang and Kuo, 2020; Sembada and Kalantari, 2021). It becomes important to investigate this gap area and identify the directions for future research in this area.

This CC creates an innovative system in which sharing does not represent a risk but an opportunity to develop companies' competitive behavior and contribute to a further boost to economic growth.

The possibility to manage knowledge strongly depends on organizational capacity: to organize knowledge (legitimize sources, index knowledge and provide ways of reading knowledge); to create favorable conditions for knowledge exchange (physical proximity, virtual and cognitive proximity or, in other words, the generation of “community conditions”); to develop organizational context conditions favorable to the exchange of knowledge (work culture and standards, incentive and development mechanisms, etc.). In other words, if technology creates the premises for overcoming some barriers to the accumulation and diffusion of knowledge (Applegate et al., 1996), formal and social organizations contribute enormously to the fact that digital tools can produce their results.

For these reasons, it is believed that digital tools, as regards the conception of CC phenomena, need, however, to be oriented toward ensuring that this is accessible and easy to use, promoting a relationship of closeness and trust between users and the supplier of the service itself, providing sufficient and reassuring security measures and collecting user experiences to establish reputation thanks to the appropriate centralized Big Data databases (Kandampully et al., 2022; Stojkovski et al., 2022). The CC approach has transformed the “cylindricity” of companies.

The service capacities can be performed for the automaintenance competitive advantage, favoring the social value of the interaction without neglecting the variables and the differences between individuals. Thus, information becomes a strategic commodity that digital technology can produce, quantify and distribute at an accelerated pace. It is more difficult to imitate and is one of the organizational resources that increases in value through its use (Probst et al., 2000). Davenport and Prusak (1998) explain that “… knowledge grows with use: ideas develop innovative ideas, and the dissemination of knowledge remains in the one who gives it and enriches the one who receives it.” It is, therefore, necessary to understand how information can be increased and diffused. In this scenario, it becomes appropriate to reformulate the equation: credit + advertising + individual ownership = hyper-consumption, which reflects the spirit of the 20th century, with another, which aims to represent the current century: reputation + community + shared access = collaborative consumption (Dimitrova et al., 2022).

From the literature examined, it is evident the difficulty mentioned by many scholars in finding a global definition for CC: “the heterogeneity of practices encompassed within it” (Nguyen and Llosa, 2018). It appears that several practices are not new and already existed before the rise of the CC phenomenon: “more recently, the exchange of goods (with transfer of ownership) via an online platform (ex: eBay) started becoming popular towards the end of the 20th century.” This reflects the significant differences between authors positions with regards to CC characteristics, according to Nguyen and Llosa (2018).

Furthermore, it is important to estimate the effects of the CC phenomenon by investigating the impact of tourism on economic growth and other key macroeconomics. This perspective of investigation was brought into focus by Hubert (2021), Bennett (2022) and Škare et al. (2021).

The authors suggest that in a broad field of investigation, it is important to investigate the correlation between the choices of CCs for short-term accommodation and employment and the effect of the COVID-19 crisis to understand its effects.

From this apparent lack of consensus on CC theory, a new approach based on what is determinant to speed innovation in the phenomenon is necessary. It became crucial to identify what constitutes the core of CC to allow an appropriate conceptualization, theorization and modeling in short-stay accommodation case with a cross-sectional panel data overview.

Starting from the CC highlighted points of view literature, the following research hypotheses were formulated in the short-stay accommodation case:

H1.

Short-stay accommodation offered via CC will have a positive effect on GDP per capita.

The starting point is the positive impact of tourism on economic growth and thus on GDP per capita (Hubert, 2021). Moreover, taking into consideration the impact of increasing short-stay accommodation provided by the CC put into the framework by the online platforms, it is assumed, in ceteris paribus conditions, a positive effect of short-stay accommodation, provided by the CC, on economic growth and thus on GDP per capita.

H2.

Short-stay accommodation offered via CC will have a positive effect on employment in the accommodation industry.

The assumption is that by increasing the intensity and number of short-stay accommodations, there will be an increase in the demand for the workforce employed in the field of tourism and support services. Other authors (Barykin et al., 2021; Bennett, 2022) have different opinions expressed regarding the impact of CC and employment; however, in the paper, it is hypothesized that an increase in demand will always increase the need for the additional workforce.

H3.

The COVID-19 crisis had a negative effect on accommodation, employment and GDP.

The COVID-19 crisis had a negative impact on the tourism and accommodation industry and, of course, on employment and GDP (Škare et al., 2021). It is tested in H3 as a control CC hypothesis regarding its impact on other variables.

3. Methodology

The VAR methodology is one of the most versatile research tools used in the literature because of its broad applicability, regardless of the field and topic. A VAR model does not attempt to determine the relationship or dependency between variables but rather hypothesizes that the determinants of choices are influenced by a list of variables.

To demonstrate the interdependencies between the variables considered and their impact, an approach based on a panel vector autoregressive (VAR) model has been used. The adopted VAR model is an autoregressive unlimited panel vector (PVAR) since the variables are integrals of different order among themselves. The proposed model is original and follows the indications of Harvey (1990). According to Canova and Ciccarelli (2013), vector autoregressive (VAR) panel models are one of the best research tools for determining the explicit microstructure present in stochastic dynamic general equilibrium models.

The vector autoregressive panel approach was applied to demonstrate the divergence between the variables and provide useful answers to demonstrate the defined hypotheses.

Therefore, the recent Eurostat database on consumer use of CC platforms was used in 561 EU regions in the period 2018–2021, including the pandemic year 2020.

In this study, it emerged that consumers increasingly turned to CC to obtain information on short-term accommodations. Several variables are identified to assess the implications of this widespread phenomenon:

1. GN – Overnight stays in accommodation facilities for short stays.

2. E – Employment (thousand hours worked) by NUTS 2 regions in the industry of accommodation and food service activities.

3. GDP – Gross domestic product (GDP) at current market prices by NUTS 2 regions.

4. Dummy – Dummy variable – for the year 2020, it has used an indicator of 1 to show the apparition of the crisis, and for all the other years, the value is 0, as no crisis has manifested.

An econometric model has been defined using a system of four equations suitable for an unrestricted panel vector autoregressive (PVAR), compatible with the Cobb–Douglas production function, which is suitable for macroeconomic models.

(1)GN1,t=α2+∆∑j=1tβ1,j GN1,t‐j+∆∑j=1tδ1,j E1,t‐j+∆∑j=1tε1,j GDP1,t‐j+ϵ1,j,Dummy1,t‐j+u1t

(2)E1,t=α2+∆∑j=1tβ1,j E1,t‐j+∆∑j=1tδ1,j GDP1,t‐j+∆∑j=1tε1,j GN1,t‐j+∆∑k=1tϵ1,j,Dummy1,t‐j+u1t

(3)GDP1,t=α2+∆∑j=1tβ1,j GDP1,t‐j+∆∑j=1tδ1,j GN1,t‐j+∆∑j=1tε1,j E1,t‐j+∆∑k=1tϵ1,j,Dummy1,t‐j+u1t

(4)Dummy=α2+∆∑j=1tβ1,j Dummy1,t‐j+∆∑j=1tδ1,j E1,t‐j+∆∑j=1tε1,j GN1,t‐j+∆∑k=1tϵ1,j,GDP1,t‐j+u1t

where:

1. GN – Guest nights spent at short-stay accommodations offered via collaborative economy platforms.

2. E – Employment (thousand hours worked) by NUTS 2 regions in the industry of accommodation and food service activities.

3. GDP – Gross domestic product (GDP) at current market prices by NUTS 2 regions.

4. Dummy – Dummy variable – for 2020.

The equations presented above have been validated once the stability conditions are met, according to the standard test for the VAR methodology.

4. Results

The stability condition of the methodology is satisfied by the VAR, as shown in Table 1. Thus, the essential requirement for VAR stability is achieved. From this point onward, it is possible to proceed with the study. Table 2 shows that the number of lags was set to five. Therefore, a modification in the independent variable's unit will have an impact on the dependent variable's value within the next period, causing the effect to be almost instant. This lag is reasonable, as the consumer will conduct their own set of searches.

Table 3 shows that the VAR satisfies the autocorrelation condition. This condition is respected by using the Portmanteau autocorrelation test, which is necessary for validating the VAR technique.

In addition, Table 4 shows that the VAR model respects the heteroskedasticity conditions, allowing it to proceed with impulse response functions.

After the verification of VAR stability conditions and data accuracy tests, presented in Tables 1–3 and Table 5, impulse functions were performed to observe the impact of the variables chosen and demonstrate the hypothesis, as shown in the results section.

The number of lags chosen, as it is possible to observe in Table 2, is one. Thus, the effects of the dependent variable will be observable during the next period; therefore, the effects are rapid and consistent, as is observed in the impulse response functions presented below. The first hypothesis that short-stay accommodation offered via online CC will have a positive effect on GDP per capita, as shown in Figure 1, was confirmed.

The second hypothesis, i.e. short-stay accommodation offered through online CC will have a positive effect on employment in the hospitality sector, remains confirmed as it is clearly observable (Figure 2) that the occupation variable understood as a key factor is strictly related to the accommodation sector, therefore it will positively influence the consumption of short-stay accommodation offered through sharing economy platforms. Both variables influence each other. As can be seen, the impact is strong and lasting for the period analyzed.

The third hypothesis, i.e. understanding whether the COVID-19 crisis has had a negative effect on housing, employment and GDP, is demonstrated as defined in figures 3–5. The reasons for this confirmation are dictated by the fact that, despite the increased access to and use of CC platforms, this has led to distorting effects on employment. Therefore, the negative relationship between the investigated variables is in line with consumer habits in crisis conditions. The latter scenario included a dataset of monthly employment rates that contained a sudden and sharp increase in the unemployment rate caused by a short and severe crisis (COVID-19). For these data, the regression used to model the influence on employment, the use of CC and GDP via the dummy variable to estimate the expected impact of the recession on the variables finds confirmation of recessionary effects in all cases.

5. General discussion

The results are in line with the impact of the collaborative economy on the tourist accommodation market in the last decade. On the one hand, CC phenomena facilitate the dissemination of information by users, but on the other hand, they require greater attention to the information updating that consumers expect for this market segment.

The summary results identify interesting correlations concerning the hypotheses outlined. From the point of view of employment trajectories and GDP growth in different scenarios, it also examines consumption data at the national, regional and European level of cities on overnight stays in 2019. The most recent data, which also covers the impact of the COVID-19 crisis on the short-stay accommodation sector, demonstrate how important the CC will be in the future.

This statement is in line with the literature investigated; in fact, the collaborative economy has both a positive and negative impact on tourism. Its proponents believe it provides easy access to a wide range of services that are often higher quality and more affordable than those provided by their traditional commercial counterparts.

The starting point is the positive impact of tourism on economic growth and thus on GDP per capita (Hubert, 2021). Moreover, taking into consideration the impact of increasing short-stay accommodation provided by the CC put into a framework by the online platforms, the results assume, in ceteris paribus conditions, a positive effect of short-stay accommodation provided by the CC, on economic growth and thus on GDP per capita.

The analysis highlighted how the short-stay accommodation offered via CC will have a positive effect on GDP, confirming the first hypothesis. The second hypothesis also remains confirmed, since short-stay accommodation offered through online CC will have a positive effect on employment in the hospitality sector and the occupation variable understood as a key factor is strictly related to the accommodation sector, therefore it will positively influence the consumption of short-stay accommodation offered through sharing economy platforms. In the third hypothesis, it is interesting to observe how decisive the crisis in the short-day accommodation sector was in the COVID-19 period. It had a negative effect, confirming the hypothesis. The result, although apparently obvious, highlights how access to tourist consumption through CC can improve the effects of the restart of consumption concerning the three variables observed. This also allows us to confirm the results of the third hypothesis, i.e. that the COVID-19 crisis had a negative impact on the tourism and accommodation industries and, of course, on employment and GDP.

It follows that the CC phenomenon can be defined as “platform tourism” according to the definition between the European Commission and four large online collaborative economy platforms (Airbnb, Booking, Expedia Group and Tripadvisor), signed in March 2020.

The COVID-19 pandemic has had major impacts on all tourism sectors between 2020 and 2022, but platform tourism has recovered and is surpassing pre-pandemic levels. This article, focusing on European-level data related to short-day accommodation, helps us understand how important collaborative digital culture is to accelerating digital knowledge in the hospitality sector. This allows us to state, in general, that the economic restart of the sector examined in the study has been faster than the CC phenomenon. These discussions address possible policy actions and good practices for consideration by public authorities to define CC as the new gamble to speed up digital knowledge in the accommodation sector.

6. Implications and conclusion

In the CC phenomenon, consumers take individual values and behavioral models with them that guide knowledge sharing, becoming a learning mechanism able to influence decision-making processes. The CC scenarios give life to spontaneous processes that are continuously turned on to generate new knowledge.

A second implication is the development of CC dynamics. They facilitate widespread learning processes, allowing tourism to benefit from data resources and holistic individual skills simultaneously. According to this perspective, the development of CC dynamics facilitates widespread learning processes, allowing tourism to take advantage of resources and, at the same time, skills.

CC phenomena are machine learning systems capable of accelerating the organizational learning process in an innovative space in which knowledge and information “travel” from one subject to another without the use of hierarchical channels, modifying traditional cooperative methods. CC systems are intelligence systems capable of accelerating the organizational learning process in an innovative space in which knowledge and information “travel” from one subject to another without the use of hierarchical channels, modifying traditional cooperative methods.

Thus, knowledge becomes an anomalous good whose theoretical value grows according to its reuse and the possibility of recombination. The interrelationship continues between the formalized-generalized knowledge of the company and informal knowledge, which resides in individual consumers and becomes the basis of the creative process of competitive advantage. Through learning, the knowledge generated is supported by the experiences of individuals to be integrated and used coherently; this process finds its drive and motivation even in the CC platforms.

Fostering knowledge sharing presupposes that organizational coordination mechanisms are adopted and partially implemented through the technological infrastructure of the CC platforms, which become an effective tool for creating knowledge as, on the one hand, they stimulate discussion, communication and mutual confrontation and, on the other hand, favor the birth of mutual trust among collaborative consumers.

In conclusion, in this paper, the usual concepts linked to collaborative consumption are embedded. It brings to light that in short-day accommodation, the traditional interaction with the territory is joined by a virtual one that bridges the physical distances with the greater or lesser slowness of the transmission lines. The empirical observations of the paper lead to reflection on the importance of CC platforms characteristics: users' anonymity and new forms of consumption in which systems of the hierarchy are abandoned, commanded and controlled in favor of a more fluid. Fast and truthful diffusion of knowledge is facilitated by systems such as relationships, sociability and spatiality.

Therefore, this is the beginning of a new phenomenon that will lead organizations that invest in CC platforms to win the gamble to speed up digital knowledge in the accommodation sector.

Figures

Figure 1

Impact of CC on GDP

Figure 2

Impact of CC on employment

Figure 3

Impact of COVID-19 crisis on GDP

Figure 4

Impact of COVID-19 crisis on employment

Figure 5

Impact of COVID-19 crisis on CC

Table 1

Roots of characteristic polynomial

Endogenous variables: GN – E − GDP – dummy
Exogenous variables: consumers
Lag specification: 11
Root	Modulus
0.995004	0.995004
0.645060	0.645060
0.473979	0.473979
0.270301	0.270301

Note(s): No root lies outside the unit circle

VAR satisfies the stability condition

Table 2

VAR lag order selection criteria

Endogenous variables: GN – E − GDP – dummy
Exogenous variables: consumers
Included observations: 712
Lag	LogL	LR	FPE	AIC	SC	HQ
0	−2873.684	NA	0.038078	8.083381	8.109045	8.093294
1	−1059.530	3602.827	0.000244	3.032388	3.160705*	3.081951
2	−1011.731	94.39052*	0.000223*	2.943064*	3.174034	3.032277*

Note(s): * indicates lag order selected by the criterion

LR: Sequential modified LR test statistic (each test at 5% level)

FPE: Final prediction error

AIC: Akaike information criterion

SC: Schwarz information criterion

HQ: Hannan–Quinn information criterion

Table 3

VAR residual serial correlation LM tests

Null hypothesis: no serial correlation at lag order h
Included observations: 1,071
Lags	LM-Stat	Prob
1	54.46286	0.0000
2	31.35285	0.0121

Note(s): Probs from chi-square with 16 df

Table 4

VAR residual heteroskedasticity tests: no cross terms (only levels and squares)

Sample: 2018–2021
Included observations: 1,071
Joint test
Chi-sq	Df	Prob
422.8634	70	0.0000

Individual components
Dependent	R-squared	F(71,063)	Prob	Chi-sq(7)	Prob
res1*res1	0.117816	20.28061	0.0000	126.1811	0.0000
res2*res2	0.086888	14.45011	0.0000	93.05707	0.0000
res3*res3	0.051539	8.251790	0.0000	55.19784	0.0000
res4*res4	0.009689	1.485742	0.1684	10.37694	0.1682
res2*res1	0.077840	12.81831	0.0000	83.36647	0.0000
res3*res1	0.035769	5.633243	0.0000	38.30839	0.0000
res3*res2	0.046729	7.443960	0.0000	50.04662	0.0000
res4*res1	0.002952	0.449630	0.8708	3.161733	0.8696
res4*res2	0.002499	0.380381	0.9141	2.676000	0.9133
res4*res3	0.004776	0.728716	0.6477	5.114858	0.6459

Table 5

VAR residual normality tests

Orthogonalization: Cholesky (Lutkepohl)
Null hypothesis: residuals are multivariate normal
Sample: 2018–2021
Included observations: 1,071
Component	Skewness	Chi-sq	df	Prob
1	−0.092948	1.542134	1	0.2143
2	0.063931	0.729552	1	0.3930
3	0.227863	9.268030	1	0.0023
4	0.090722	1.469148	1	0.2255
Joint		13.00886	4	0.0112

Component	Kurtosis	Chi-sq	df	Prob
1	4.123189	56.29680	1	0.0000
2	4.006250	45.18456	1	0.0000
3	5.867968	367.0514	1	0.0000
4	532.4405	12,508,709	1	0.0000
Joint		12,509,177	4	0.0000

Component	Jarque–Bera	df	Prob
1	57.83893	2	0.0000
2	45.91411	2	0.0000
3	376.3194	2	0.0000
4	12,508,710	2	0.0000
Joint	12,509,190	8	0.0000