Space tourism refers to members of the general public, such as civilians and space tourists, visiting space solely for recreational purposes. It covers spaceflights for various distances: suborbital, orbital, and beyond the Earth’s orbit. For example, Virgin Galactic provides suborbital trips that reaching around 100km altitude and passengers can enjoy several minutes of out-of-seat weightlessness and breathtaking views of Earth in space costing US$450,000. In 2022, it already attracted 800 customers signed up for the trips. SpaceX offers even more expensive orbital trips that reaching an altitude of almost 600km and can keep passengers in space for several days costing US$55 million. China is also catching up quickly (with 370 commercial space-related companies set up as of 2021) and it is expected to start suborbital travel in 2025, which costs around 2-3 million yuan (US$285,00-US$427,000).

With the continued advancement of space technology, space exploration has become a key mission for many nations including the U.S. and China. For example, the U.S. plans to return to the moon and establish a lunar colony over the next few years. China also built its Tiangong space station in 2022 and is planning to get to the Moon in 2030. The Chinese State Council pledged to “make the space industry an important part of the nation’s overall development strategy.” The rapid development of space technology will further lower the costs of space tourism and the space tourism market is expected to grow rapidly. The global market for space tourism is estimated to reach US$1.7 billion by 2027, and the U.S. and China are the two biggest potential markets for space tourism. North America is the leading market for space tourism, with a market share of more than 40 percent. Asia-Pacific trails closely behind, with a market share of about 30 percent. The Chinese space tourism industry is forecast to be worth US$401 million by 2027, making it the second largest in the world.

Having civilians travelling to space will be a breakthrough in human cultural development. As space tourism hits mainstream commercialisation, who would be more interested to spend money on enjoying the breath-taking views of Earth in space? Moreover, a luxury space hotel Voyager Station is scheduled open for business in 2027. In order to generate revenue, what activities and foods should space tourism companies provide to tourists once reach orbit? Since space tourism is still expected to remain relatively expensive, what about the masses who are still unable to afford space travel? Is there a way for them to experience space travel in a cheaper fashion? What lessons can companies seeking to develop space tourism learn?

These are some of the questions I sought to answer in this Research White Paper.

‘Stealing’ Digital Red Packets

The online shop studied usually give out higher value red packets to new customers and dedicated customers. However, it added a ‘stealing’ function so that even non-frequent customers are able to receive high value red packets from their friends who are loyal customers. Customers can request an unused red packet from friends in their social circles when they place orders. This customer can choose any friend in their social network to get the red packet from, and this friend cannot prevent this red packet from being given away but they can choose which ones they get to keep.

Under this feature, the size of the red packets not only depends on the customer’s status with the company but also the size of their online social network. To increase their chances of getting valuable red packets, users can actively recruit new customers by sending out referral links on the app to their friends who are not existing users. They can also send out special merchant offers to their friends who are existing users to encourage them to place more orders.

“By making digital red packets ‘social’, companies turn incentivising online purchases into a red packet game. It’s a win-win situation for both customers and the company. The former gets to participate in a fun and interesting experience, and the latter benefits from exposure to a new customer base as well as potentially increased business from existing clients,” Prof. Lin says.

The Effect of Sharing

After analysing the behaviours of hundreds of users on the food delivery platform from October 2016 to September 2017, the researchers found that when red packets are allowed to be shared, people who are infrequent customers saw their monetary rewards rise the most.

The study found that on average, when companies gave out red packets that were not allowed to be shared among customers, high value customers, such as new customers and loyal customers, received red packets with discounts of 12 percent and nine percent, respectively. On the other hand, existing customers who are not frequent buyers received discounts of just two percent. However, if the red packets were allowed to be circulated among users’ social circles, the average discounts in red packets for new and devoted customers came in at 11 percent and 10 percent, respectively, but the discounts received by existing but infrequent customers rose sharply to eight percent.

Prof. Lin adds that although infrequent buyers can improve their chances of getting red packets with steep discount vouchers by recruiting new users and encouraging their friends to make more purchases, it is more difficult for them to recruit new users than motivating existing users to make more repeat orders. This is because customers usually know what their friends would want to buy anyway. Because of this, the researchers conclude that “social” red packets are less effective at, and take a longer time in, acquiring new customers.

Customer Profiling

The researchers also found that some types of customers are more likely to use their online social circles to get the most out of these red packet discounts. First, customers who pay more attention to prices are more interested in getting red packets through their social network. Second, customers who are more socially active are more likely to take advantage of social red packets. Due to their strong ties with their friends on social networks and strong public profiles, they may have a higher chance of convincing their friends to place more orders and recruiting new users for the app.

Different age groups also behave differently. According to the study, older customers tend to share more merchant offers with their friends, while younger customers are keener on recruiting new users by sharing referral links. Thirdly, customers who like to buy other things on the food delivery platform the researchers studied, rather than just order meals, tend to be more interested in recruiting new customers but they were less likely to socialise with their existing friends using the platform.

In addition, users who have friends which are potential new customers or existing loyal customers are likely to socialise with these friends on their social networks to take advantage of the social red packet function. Furthermore, given they live in the same city, customers are more generally interested in sharing red packets with friends who are also existing users, than recruiting them as new customers.

It’s been called the holy grail of finance. Is it possible to harness the promise of artificial intelligence to make money trading stocks? Many have tried with varying degrees of success. For example, BlackRock, the world’s largest money manager, has said its AI algorithms have consistently beaten portfolios managed by human stock pickers. However, a recent research study by The Chinese University of Hong Kong (CUHK) reveals that the effectiveness of machine learning methods may require a second look.

The study, titled “Machine Learning versus Economic Restrictions: Evidence from Stock Return Predictability”, analysed a large sample of U.S. stocks between 1987 and 2017. Using three well-established deep-learning methods, researchers were able to generate a monthly value-weighted risk-adjusted return of as much as 0.75 percent to 1.88 percent, reflecting the success of machine learning in generating a superior payoff. However, the researchers found that this performance would attentuate if the machine learning algorithms were limited to working with stocks that were relatively easy and cheap to trade.

“We find that the return predictability of deep learning methods weakens considerably in the presence of standard economic restrictions in empirical finance, such as excluding microcaps or distressed firms,” says Si Cheng, Assistant Professor at CUHK Business School’s Department of Finance and one of the study’s authors.

Disappearing Returns

Prof. Cheng, along with her collaborators Prof. Doron Avramov at IDC Herzliya and Lior Metzker, a research student at Hebrew University of Jerusalem, found the portfolio payoff declined by 62 percent when excluding microcaps – stocks which can be difficult to trade because of their small market capitalisations, 68% lower when excluding non-rated firms – stocks which do not receive Standard & Poor’s long-term issuer credit rating, and 80 percent lower excluding distressed firms around credit rating downgrades.

According to the study, machine learning-based trading strategies are more profitable during periods when arbitrage becomes more difficult, such as when there is high investor sentiment, high market volatility, and low market liquidity.

“Machine learning methods require high turnover and taking extreme stock positions. An average investor would struggle to achieve alpha after taking transaction costs into account.”

Prof. Si Cheng

One caveat of the machine-learning based strategies highlighted by the study is high transaction costs. “Machine learning methods require high turnover and taking extreme stock positions. An average investor would struggle to achieve alpha after taking transaction costs into account,” she says, adding, however, that this finding did not imply that machine learning-based strategies are unprofitable for all traders.

“Instead, we show that machine learning methods studied here would struggle to achieve statistically and economically meaningful risk-adjusted performance in the presence of reasonable transaction costs. Investors thus should adjust their expectations of the potential net-of-fee performance,” says Prof. Cheng.

The Future of Machine Learning

“However, our findings should not be taken as evidence against applying machine learning techniques in quantitative investing,” Prof. Cheng explains. “On the contrary, machine learning-based trading strategies hold considerable promise for asset management.” For instance, they have the capability to process and combine multiple weak stock trading signals into meaningful information that could form the basis for a coherent trading strategy.

Machine learning-based strategies display less downside risk and continue to generate positive payoff during crisis periods. The study found that during several major market downturns, such as the 1987 market crash, the Russian default, the burst of the tech bubble, and the recent financial crisis, the best machine-learning investment method generated a monthly value-weighted return of 3.56 percent, excluding microcaps, while the market return came in at a negative 6.91 percent during the same period.

Prof. Cheng says that the profitability of trading strategies based on identifying individual stock market anomalies – stocks whose behaviour run counter to conventional capital market pricing theory predictions – is primarily driven by short positions and is disappearing in recent years. However, machine-learning based strategies are more profitable in long positions and remain viable in the post-2001 period.

“This could be particularly valuable for real-time trading, risk management, and long-only institutions. In addition, machine learning methods are more likely to specialise in stock picking than industry rotation,” Prof. Cheng adds, referring to strategies which seek to capitalise on the next stage of economic cycles by moving funds from one industry to the next.

The study is the first to provide large-scale evidence on the economic importance of machine learning methods, she adds.

“The collective evidence shows that most machine learning techniques face the usual challenge of cross-sectional return predictability, and the anomalous return patterns are concentrated in difficult-to-arbitrage stocks and during episodes of high limits to arbitrage,” Prof. Cheng says. “Therefore, even though machine learning offers unprecedented opportunities to shape our understanding of asset pricing formulations, it is important to consider the common economic restrictions in assessing the success of newly developed methods, and confirm the external validity of machine learning models before applying them to different settings.”

We live in the age of the social influencer. Nowadays, with the development of the internet and the growth of social media, it is increasingly easy for consumers to post a comment or a picture of a product and to notice whether somebody else uses or praises a product or a brand, making word of mouth and opinion leader recommendations more important than ever.

Online, a recommendation may come in a form of a selfie on Instagram, a short Tweet, a “like” on Facebook, or the fact that an opinion leader uses certain products. An entire cottage industry has seemingly risen overnight where internet celebrities – people who acquired or developed their fame online – make their entire livings from corporate endorsement, using their relative notability to peddle products or experiences to their followers.

However, using social influencers to market a company’s products or services can be tricky. Consumers nowadays are savvy enough when they encounter a product endorsement to consider not only that it is likely sponsored, but also whether the recommendation is purely based on high product quality. It is highly likely they would also think about whether it was because the social influencer selected a specific product (from a variety of offerings) that suited their particular taste, says Chenxi Liao, Assistant Professor at The Chinese University of Hong Kong Business School’s Department of Marketing.

Under such a scenario, is there an ideal number of varieties of a product a company could offer to maximise profits? This is what drove a group of researchers to look at the impact of product variety when consumers rely on the product evaluations of opinion leaders or experts to make purchase decisions. Entitled Opinion Leaders and Product Variety, the study was conducted by Prof. Liao in collaboration with Prof. Dmitri Kuksov of the University of Texas at Dallas.

“Absent costs of variety, to increase profit, an intuitive solution is to increase product variety. With more alternatives available, it is more likely that an expert can find a variant that fits their personal preference, and post a positive product opinion,” says Prof. Liao, adding that since consumers expect quality to be higher when the expert opinion is positive than when it is negative, they are willing to pay a higher price.

When More Variety is Bad

From this perspective, increased assortment can benefit the firm. However, when many product variants are available, consumers may expect an expert to find a better fit, and this consideration would then reduce consumer expectations of the product quality, leading to lower profits.

There is also the issue of market noise. Consider that opinion leaders are likely to be more knowledgeable and have a greater desire to educate themselves about a certain product, since this is typically how they gain popularity and the trust of their followers in the first place. These experts are likely to be able to better understand and choose which variant fits them best.

On the other hand, the everyday consumer may not be as familiar with the product or are not willing to spend the time and effort. This means they may not be able to choose the best-fitting alternative and may benefit less from having many alternatives.

“The idea is that the promotion of a smaller number of products makes communication easier and clearer.”

Prof. Chenxi Liao

Taken all of this into consideration, the researchers formulated a complex mathematical model and found that there does exist a point beyond which increasing product variety netted no significant benefits to a company.

Prof. Liao explains that two opposing forces come into play when a company increases the variety. First of all, when product variety increases, it raises the probability of getting a positive expert opinion and therefore the product being recognized by consumers as of high quality. On the other hand, it also decreases certainty by the consumer that the product received a good review because of its inherent high quality.

Although the first effect is positive, the second is negative. It turns out that for a small number of variants, the first effect dominates the second, and for a large number of variants, the second effect dominates the first.

“We further find that the optimal number of product variants increases if the importance of fit for the expert or the expert’s unwillingness to provide a positive recommendation is higher,” Prof. Liao adds.

Another consideration is whether the company has a good grasp of how consumers would rate the quality of its products. The researchers cite as examples two similar products in recent years – the Apple iPod and the Microsoft Zune music players. Whereas the second flopped, the first became a phenomenal success, but because they were both conceptually new products, it was difficult to tell how consumers would rate their quality.

The research also found that, unsurprisingly, when the firm has a good understanding of how consumers will perceive the quality of its products, the firm with the lower quality product may seek to hide its inferiority by limiting the information transmitted through expert opinions and mimic the variety provided by a firm with a similar product but of a higher quality.

Real World Implications

Relating the findings to real world business practices, Prof. Liao notes that it is not uncommon for a company to introduce a product line with a small number of products, or even just one, promoted to opinion leaders.

“The idea is that the promotion of a smaller number of products makes communication easier and clearer,” she says, citing a successful promotional campaign by the U.S. luxury department store chain Lord & Taylor. In introducing its Design Laboratory collection, the company chose to promote a single dress to many influencers on Instagram.

The fact that a number of social influencers all agreed to create a post on Instagram with the same dress signaled to their followers that the choice was based on the quality of the dress itself, rather than because it complimented a specific fit or complexion.

Another prediction of the model is that when opinion leaders are more likely to be happy with the product regardless of the exact quality they see, a firm should prefer to offer a smaller product selection, says Prof. Liao. For example, Apple does not provide as many customization opportunities in iOS for iPhones as Android systems usually do. However, since many opinion leaders like Apple products more, they are likely to give a positive review.

Prof. Liao adds that one important assumption was that consumers are unable to tell exactly how much of a positive product review is because the product is suited to the reviewer’s tastes. This is changing in that many websites, such as the American NGO Consumer Reports, are providing more detailed information regarding the reviewer’s context. In these circumstances, the negative effect of having a large number of product variants will be diminished.

Looking at opportunities for future research, Prof. Liao notes that another assumption is that the expert always posts a product evaluation. “In practice, opinion leaders could be silent on many products. An absence of a recommendation may be interpreted as a negative opinion, as in the expert did not find it worthwhile to choose the product, or could be reducing the informativeness of the recommendations,” she says.

One interesting avenue is to consider the uncertainties facing the expert before and after the purchase, and how it affects the likelihood of a positive review, she adds.

The COVID-19 pandemic has accelerated the adoption of technology in the service industries. As hotels and restaurants increasingly turn to artificial intelligence and install robotic concierge and waiters, as well as service kiosks, what is the best way for them to communicate with customers?

For example, if a customer wants to know whether a tourist attraction is worth visiting, should a robot respond with the plain and informative “The view is excellent” or the more conversational “The views there will blow your mind away”?

This is basis of a new research study, titled “How May I Help You?” Says a Robot: Examining Language Styles in the Service Encounter. Looking at the use of language in service encounters, the researchers found that, more often than not, providing information in a clear and straightforward manner is better than the more colourful alternative.

“Verbal communications plays a key role in customer service encounter evaluations, but we know very little about how language styles affect customer satisfaction,” says Prof. Sungwoo Choi, Research Assistant Professor in the School of Hotel and Tourism Management at The Chinese University of Hong Kong Business School and one of the co-authors of the study.

“Today’s service encounters are increasingly infused with innovative technologies, and most notably, smart service robots have become ubiquitous. However, we know very little about what is the best ways these robotic service providers should address customers.”

Prof. Choi and his co-authors, who are Prof. Stephanie Liu at Ohio State University and Prof. Anna Mattila at the Pennsylvania State University, specifically looked at whether literal or figurative language was superior in service encounters, and whether this was different depending on whether the information was being provided by a human being, a robot, or a service kiosk.

Literal vs Figurative Language

When people use literal language, they are being straightforward and saying exactly what they mean. On the other hand, figurative language refers to the use of metaphors, similes, hyperbole or personification to describe something, often by comparing it with something else. Figurative language are sometimes used to evoke an emotionally intense response.

The researchers recruited 173 adult consumers in the U.S. for the study. The participants were asked to imagine themselves in a scenario where they stayed at a fictitious hotel and had to get ideas for dining options from the hotel concierge, which was either human, a robot or a service kiosk. The concierge either used literal language such as “the restaurant has a nice interior design” or figurative language such as “the restaurant looks more stunning than a palace”. The participants then completed a series of questions to evaluate their service encounters.

“When customers deal with human service personnel, the use of literal language led to marginally higher service encounter evaluation. This was also the case when customers were dealing with service robots.” says Prof. Choi. “The results also revealed that, when customers are dealing with human service representatives, using literal language also led to higher credibility. This effect again extended to service robots.”

“When customers deal with human service personnel, the use of literal language led to marginally higher service encounter evaluation.”

Prof. Sungwoo Choi

According to the study, service kiosks were unaffected by language style because they resembled objects rather than people. Language didn’t matter in this case because service kiosks didn’t look human and thus the expectations that govern human-to-human communication didn’t apply.

“The main difference between service robots and kiosks is their external shape. A service robot is designed to have an appearance or character resembling a human being, whereas a service kiosk looks more like an object. Past research have suggested that people tend to apply their beliefs and knowledge about humans to non-human objects when they have humanlike features,” Prof. Choi explains.

The researchers suggest hospitality managers to recognize the importance of language styles used by frontline employees and develop their training protocols accordingly.
“We show that literal language is more appropriate in face-to-face service interactions. Guests often inquire about places to visit. They would ask for recommendations for shopping, restaurants, or tourist attractions or about hotel facilities. Similarly, servers often give customers recommendations on menu items. In such service interactions, frontline employees should avoid using figurative expressions. Rather, providing straightforward and clear information tends to improve service encounter evaluation,” Prof. Choi comments.

Future Research Directions

Prof. Choi said it would be interesting to explore language styles in failed service encounters as the current study focused on the effect of language styles in a successful service encounter.

“Would people still expect a service provider to use literal language when offering explanations or an apology since service recovery perceptions are driven by competency and reliability? Or, would people expect the service provider to use emotionally intense language as a means to recover a damaged relationship?” Prof. Choi asks.

In addition, Prof. Choi said it would be worth extending the investigation to loyal customers who have an existing relationship with the service provider and to test the theory in the field and collect data from real interactions between guests and service robots. In addition, he said it would also be interesting to examine how language styles influence consumers’ brand perceptions, social media engagement, and loyalty.

“There are many other language styles that are worth exploring in the context of service encounters, such as assertive vs. nonassertive language, informal vs. formal language, and abstract vs. concrete language. It would be interesting to gain insight into consumer reactions to service robots employing such language features. Testing the effects could deepen our understanding of customer evaluation with technology-infused service encounters,” says Prof. Choi.