Effects of COVID-19 on the Tourism Sector: Learning from the Azores Islands

Abstract

This research aims to examine the impact of COVID-19 on businesses operating in the tourism industry in the Azores. The objective of this survey, conducted in 2021 and 2022, was to investigate the preventive and management measures implemented by the tourist firms, in this outermost region of Portugal, in response to the pandemic, along with the effects on profitability. Additionally, the study aims to assess the contribution of financial support from the regional government toward these businesses’ sustainability and anticipate future expectations. The temporary closure of businesses was the most frequently adopted measure by the firms to adapt to the ongoing pandemic. Most companies reported a significant decline in profitability, with a reduction in customers being the primary contributing factor. Most firms availed themselves of the preceding financial support structures provided by the regional government. Looking forward, the firms expressed their intention to retain their current employees. At the same time, their biggest concern was the potential decrease in consumer demand, even in the context of well-established public health safety protocols.

1. Introduction

Tourism is a prominent and multifaceted sector in modern society, encompassing a range of industries. Its substantial employment opportunities make it a significant contributor to the economy [1,2,3,4]. Nevertheless, on 17 November 2019, the world witnessed the first recorded case of COVID-19, an illness caused by the SARS-CoV-2 coronavirus. The World Health Organization declared the outbreak a global pandemic on 11 March 2020. This unprecedented situation has resulted in widespread stagnation, affecting communities, businesses, and nations worldwide.

Among the various industries affected by the SARS-CoV-2 pandemic, tourism is among the most affected [5], due to the impediment of movement across borders and the fear of getting infected by the virus [6]. In the space of months, the framework of the global tourism system has changed from excessive tourism to nonexistence [7,8,9,10].

In this sense, before a vaccine appears, the most effective strategies to try to slow the disease’s contagion rate are mass confinements and social distancing measures. These measures have significantly impacted the sustainability of various sectors, including but not limited to the tourism industry. The temporary closure of businesses and restrictions on travel and gatherings have led to decreased revenue and increased financial strain on businesses, and tiny and medium-sized enterprises. In turn, this has reduced employment opportunities, causing significant economic and social impacts.

Additionally, these measures have reduced tourism activity and disrupted supply chains, further impacting the industry’s sustainability. While the long-term effects of these measures still need to be fully understood, it is clear they have had significant impacts on the sustainability of businesses and industries around the world. Besides that, the implementation of these measures is complex in economies intrinsically linked to the production and distribution of goods and tourism [11].

According to the study of Ponte et al. [12]: “In view of the above, if we consider outermost territories, as is the example of island regions, where, as a rule, tourism plays a fundamental role in their socio-economic sphere, the consequences of the COVID-19 pandemic crisis are even more evident (...)”.

Contextually, the European Union (EU) has nine outermost regions (OR), among which is the Azores. ORs present strategic potential and unique assets from which the EU can benefit, and the Azores is no exception. The peripheral and smaller regions tend to depend more significantly on tourism. In the Azores, this industry is important in developing the regional economy. The region has natural and cultural resources present in large quantities, which enabled the region to develop and consolidate competitive advantages in the nature tourism market.

However, a fundamental issue inherent to these regions is the strong dependence on an efficient transport system, which provides greater accessibility and, consequently, an increase in tourism. For a long time, the Azores depended exclusively on one airline, SATA Azores Airlines, and it was only in 2015 that a new air transport business model was implemented that allowed the entry of low-cost airlines. This change facilitated greater exposure to international markets through promotional campaigns promoted by these companies. There is a need to continuously create strategies to enhance local tourism resources, in which the creation of differentiating products, services, and experiences is often inherent. Another significant achievement for tourism in the region was the certification from the Global Sustainable Tourism Council (GSTC) in 2019 as a sustainable tourist destination, which opened doors to new competitive advantages. It has also helped raise awareness of sustainable tourism practices among tourists and industry stakeholders. It has encouraged other destinations to follow in the Azores’ footsteps in promoting sustainable tourism practices. Overall, the Azores is promoting tourism sustainability, such as developing a sustainable tourism action plan, promoting eco-tourism activities, and encouraging the use of renewable energy sources in tourism activities. These efforts aim to ensure sustainable tourism development, benefit local communities, and preserve natural and cultural resources [13].

In the specific regional case of the Azores archipelago (Figure 1), the industry recorded drops in the amount of 400 million EUR due to the pandemic. According to the Chamber of Commerce and Industry of Ponta Delgada (CCIPD, 2021), the loss was more than 2 million in overnight stays in just one year, with the most significant drop on the island of São Miguel, corresponding to 71.5%, followed by the islands of Terceira and Faial.

According to the Azores Tourism Observatory (OTA), the growth rate of overnight stays in the region between March and December and the volume of overnight stays was always negative. Regional tourist companies consider that the impact of the COVID-19 pandemic was negatively “very large”, with 38.8% of companies “remaining in full operation” and 34.9% “remaining in partial operation”. About 90.1% of companies say they “have not made any redundancies since the beginning of the pandemic”. Regarding turnover, 30.8% of companies had a reduction in turnover of more than 90% during the high season of 2020, with 71.5% resorting to layoffs in the same period, mainly due to a drop of 40% or more in turnover (60.2%); a total of 56.1% of tourist companies did not “(…) join the Regional Complement to the Normal Lay-off, which aims to encourage the training and qualification of workers (…)” and 62.3% did not “use the main lines of support for Tourism launched by the Regional Government” [15].

In general, tourist companies in the Azores agree with the health control measures applied in the region and consider the “Clean & Safe” seal to be an asset for companies in the industry [15]. One of the three main aspects on which companies consider that tourism in the Azores will have to invest urgently in the short and medium term is “improving air accessibility in the Azores, attracting new regular airlines operating to and from the region; Facilitate visits to all islands, improving the inter-island transport network; promote activities throughout the year, attracting specific segments of tourists to combat seasonality” [15]. The measures taken to prevent the spread of the SARS-CoV-2 virus implied restrictions on the mobility of citizens and a drastic reduction in economic activity, which could jeopardize the sustainability of many companies and give rise to a severe unemployment problem. Thus, intending to mitigate such risks, the regional government created several mechanisms to support companies and workers.

The impact of the COVID-19 pandemic has brought new dynamics and perspectives to the heart of companies operating in the tourism industry. Therefore, many are rethinking their tourist offer by introducing new promotion/communication channels and intercalating new products [14]. Moreover, according to Sousa et al. [16]: “(…) the Azores regional actors should follow the sustainable development strategy (including sustainable tourism) if this destination wants to prosper in the post-pandemic period as a potential rural tourism destination.” Therefore, tourism planning is fundamental for the competitiveness of tourist destinations, “(…) especially when it is required to take into account the new needs of tourists and the global trend towards meeting the principles of sustainable tourism.” [12]. In sum, it is believed that the current crisis highlights some crucial differences from the previous ones, not only in terms of the dimension of the sudden reduction in economic activity at a global level caused by social isolation but also by the uncertainty that has quickly destabilized the economic markets.

Santos and Cândido [17] emphasize that activities developed in a particular locality should reflect sustainability principles to maintain balance and equity between the environmental, social, and economic dimensions. Since tourism is an activity that directly impacts the locality where it is developed, it should be managed based on the balance and equity between the mentioned dimensions. In order to be aware of the principles of sustainability, Meller and Marfán [18] show that tourism, more than any other sector within the service segment, presents excellent sensitivity to the changes that occurred due to the COVID-19 pandemic.

2. Materials and Methods

The present study aims to identify the main measures adopted by companies that develop activities related to tourism during the COVID-19 pandemic and their impact in terms of profitability, as well as to evaluate the following formulated research hypotheses:

H1. 

“Company’s size” is correlated with the effects of COVID-19 on “Profitability fluctuation”.

H2. 

The level of concern of companies regarding “Reduction in customer demand, even in a context of control of the health situation” is correlated with the effects of COVID-19 on “Oscillation in profitability”.

H3. 

The effects of COVID-19 on “Profitability fluctuation” differ according to the “Sector of activity”.

H4. 

The effects of COVID-19 on “Profitability fluctuation” differ with the “company’s geographic location”.

H5. 

There is a relationship between the “Measures adopted by the company to adapt to the current pandemic situation” and the “Sector of activity”.

H6. 

There is a relationship between the “Measures adopted by the company to adapt to the current pandemic situation” and the “Number of years of activity of the company”.

H7. 

There is a relationship between the “Measures adopted by the company to adapt to the current pandemic situation” and the “Company’s size”.

The target population comprises companies in the tourism industry operating in the Azores. Data collection was based on a questionnaire generated in Microsoft Forms, which was distributed online to companies that meet these requirements, explaining its objectives and guaranteeing the anonymity of the participants and the confidentiality of the data. The questionnaire consists of five parts (I—COVID-19 and measures adopted; II—Turnover and profitability; III—Governmental Support; IV—Future Perspectives; and V—Sociodemographic Data (sector; number of years in activity; location; company’s size)).

Among the several questions included in the questionnaire, particular emphasis will be given to the measures adopted by companies in the face of the pandemic and to a set of 14 items that assess the level of importance on a 5-point Likert scale, from 1 (Not at all important) to 5 (Extremely), that companies in the tourism sector attribute to each of these measures regarding the promotion of tourism in the Azores. The description of the items corresponding to these measures adopted in the region is as follows: A1—Improve the air accessibility of the Azores, attracting new regular airlines to and from the region; A2—Improve air accessibility in the Azores, attracting new low-cost airlines to and from the region; A3—Favor visits to all islands, improving the inter-island transport network; A4—Focus marketing actions on attracting tourists with high purchasing power, that is, attracting fewer tourists, but better tourists, those who spend more because they value our offer; A5—Promote activities throughout the year, attracting specific segments of tourists (for example, senior tourism, creative/cultural tourism) to combat seasonality; A6—Implement communication and promotion actions with emigrant communities in the USA and Canada; A7—Bet on other complementary products, in addition to nature tourism, diversifying the offer; A8—Encourage the continuous improvement of the quality of the visitor’s tourist experience through the training of workers; A9—Prioritize the attribution of support by the regional government aimed at improving the quality of the offer of existing tourist companies, especially small and medium-sized companies, including Local Accommodation; A10—Stimulating entrepreneurs to open new innovative and quality tourism businesses; A11—Support the use of new technologies to increase the productivity and competitiveness of tourist companies and the Azores destination; A12—Invest in the formation of assets aimed at the new needs resulting from the pandemic; A13—Preserving local communities through the application of sustainability principles; and A14—Support strategic plans to combat the current crisis and define tourist strategies for the medium and long term, reinforcing the specificities of each municipality.

The sample is composed of 122 companies of the tourist sector (71 (58.2%) are in the eastern group (The Azores Archipelago is divided into three groups: (1) Eastern—with the Islands of Santa Maria and São Miguel; (2) Central—with the Islands of Terceira, Graciosa, Faial, Pico, and São Jorge; and (3) Western—composed by the islands of Corvo, and Flores), 49 (40.2%) in the central group, and 2 (1.6%) in the western group), with tourist entertainment being the most represented sector of activity, with about 51 (41.8%) of the responses. Approximately 52 (42.6%) of these companies have been operating for less than six years, and 63 (51.6%) for at least seven years. The majority (82%) have between 1 and 2 employees. In addition, it should be noted that companies located on the largest island of the Azores (São Miguel) correspond to more than half (55.7%) of the participating companies.

Statistical data analysis was carried out using summary statistics, Spearman’s correlation coefficient, and the respective hypothesis test; some nonparametric tests (Kruskal–Wallis’s test, Dunn’s test, chi-square independence test; categorical principal component analysis (CatPCA, also known as NLPCA); and some hierarchical algorithms in the scope of cluster analysis.

Descriptive statistics are essential and very useful in today’s era of big data to summarize the characteristics and distribution of a set of data. Concerning the summary statistics, we used frequencies, percentages, mode (the most frequent value(s) of a variable in the data), quartiles (values that divide the set of ordered data into four parts: Q1 (first quartile), Q2 (second quartile = median), Q3 (third quartile). It should be noted that in terms of percentiles (values that divide the set of ordered data points into 100 parts), Q1 is the 25th percentile (P25) and indicates that at least 25% of the data are less than or equal to this value; Q2 indicates that at least 50% of the data points are less than or equal to this value (coincides with the median and the 50th percentile (P50)); and Q3 is the 75th percentile (P75) and indicates that at least 75% of the data points are less than or equal to this value.

Spearman’s rank correlation coefficient, ρ, is a nonparametric measure of the strength and direction (direct or inverse) of the relationship (linear or nonlinear) between a pair of variables measured on at least an ordinal scale when the relationship under analysis is monotonic. This correlation coefficient varies between −1 (perfect negative relationship) and 1 (perfect positive relationship). The further the value of this coefficient is from zero, the stronger the relationship between the pair of variables. A positive sign means that as one variable increases, the other tends to increase, while a negative sign means that as one variable increases, the other variable tends to decrease. Moreover, Spearman’s correlation coefficient should be applied in the case of quantitative data when the assumption of normality, required for the calculation of the Pearson’s product-moment correlation, fails, or the relationship between the two variables is nonlinear, as well as in the case of one or both variables being measured on an ordinal scale. In the present study, we applied this coefficient due to the ordinal nature of the data concerning the first (H1) and second (H2) research hypotheses.

The Kruskal–Wallis test (nonparametric equivalent of the one-way ANOVA) can be used for testing whether samples originate from the same distribution (e.g., [19,20]). Therefore, it allows us to verify if there are statistically significant differences between two or more independent groups regarding the values of a quantitative or ordinal dependent variable. If the null hypothesis is rejected using the Kruskal–Wallis test (that is, there are statistically significant differences between at least two of the k (k ≥ 3) independent groups), it is useful to conduct the Dunn’s Test (nonparametric pairwise multiple comparisons procedure) to determine which pairs of groups are statistically significantly different considering a specific significance level. This test was conducted to evaluate the hypotheses H3 and H5.

Finally, the chi-square test of independence is used to determine if there is a significant association/relationship between two (categorical) variables. Here, we used this nonparametric test in the case of nominal variables concerning the remaining hypotheses (H5 to H7).

According to Linting et al. [21,22,23], contrary to PCA, “NLPCA can handle the analysis of possibly nonlinearly related variables with different types of measurement level. The method is particularly suited to analyze nominal (qualitative) and ordinal (e.g., Likert-type) data, possibly combined with numeric data”. In the present study, the CatPCA, with varimax rotation with Kaiser normalization, was based on the submatrix (122 × 14) containing the 14 items (A1 to A14) mentioned above. The application of the ascendant hierarchical cluster analysis (ACHA), or hierarchical clustering, on the same submatrix was carried out based on the Spearman’s correlation coefficient as a measure of comparison between the items and on three aggregation criteria, namely, Single Linkage, Complete Linkage, and Average Linkage.

3. Results

3.1. Measures Implemented during the Pandemic Period

Regarding the measures adopted by companies to adapt to the current pandemic situation, it is noteworthy that the most applied measure was temporary closure, mentioned by approximately 62.3% of respondents, and the least-used measure, mentioned by only 10.7% of business owners, was salary reduction, as shown in

Table 1. Measures adopted by companies—Multiple response analysis (absolute frequency and %; number of valid cases (n) = 122).

Measures	Yes
	Frequency	% of Cases
Teleworking	31	25.4
Reorganization of work teams	36	29.5
Salary reductions	13	10.7
Temporary closure	76	62.3
Permanent closure	0	0
Remained fully operational	49	40.2
Remained partially operational	60	49.2
Introduction of new communication/contact channels with customers	35	28.7
Offer diversification	56	45.9

.

The main measure implemented by the majority (62.3%) of companies during the pandemic crisis was temporary closure, while the least adopted measure, mentioned only by 10.7% of respondents, was salary reduction. Approximately 32.8% of companies underwent workforce restructuring, with the majority (64.8%) resorting to partial layoff measures, 24.1% to total layoff, 5.6% to nonrenewal of contracts, 1.9% to dismissals, and 3.7% to other measures. Moreover, around 21.3% of the participants mentioned that their company uses alternative channels for contacting customers. Of these channels, online sales were the most widely used, mentioned by 37.9% of participants, followed by remote service provision (17.2%), which can be attributed to the higher response rate from tourism animation companies.

In order to reduce contagion within companies, it became imperative to implement permanent restructuring in the way of working. According to the respondent’s perception, the most adopted measure for this purpose was decreased business travel (38.5%), while the least applied measure was the more intensive use of telework (17.2%).

Among the companies that implemented telework, the majority (61.3%) did not provide technological equipment or training for employees. Regarding companies that continued to work in person, the vast majority adopted health control measures to maintain their regular operation, with the costs related to the availability of safety materials having some impact on company expenses. Concerning companies’ investment in training aimed at obtaining seals in the context of COVID-19, the majority (65.6%) competed for the “Clean and Safe” seal.

Concerning companies that continued to work in person, the majority adopted sanitary control measures to maintain their regular operation, particularly the use of masks (96.7%) and hand sanitizers or disinfectants (96.7%), which were the most applied measures. Surface or product disinfection (93.4%) ranked second, followed by cleaning and disinfection of spaces (92.6%). Most companies provide the necessary materials to maintain in-person work. However, most did not offer face shields (73.8%) or thermometers (53.3%). It is also worth noting that based on the collected data, the majority (61.3%) of companies implementing telework did not provide technological equipment or specialized training for employees (78.7%).

Approximately 51.6% of the participants stated that the COVID-19 pandemic had a substantial effect on the profitability of their company (mode = 5 (Very large effect), P25 = 4 (Large impact), P50 = 5, P75 = 5), as shown in

Table 2. Effect of the COVID-19 pandemic on profitability (absolute frequency and %), according to the respondents’ perception.

		Frequency	%
Valid	Non-effect	1	0.8
	Little effect	2	1.6
	Moderate effect	20	16.4
	Large effect	36	29.5
	Very large effect	63	51.6
	Total	122	100.0

. Furthermore, the majority (57.4%) fully agree that the fluctuation in the company’s profitability is directly related to the lower current level of orders/customers.

Table 3. Degree of importance attributed to the package of measures presented by the government for the liquidity situation of companies due to COVID-19 (absolute frequency and %).

		Frequency	%
Valid	Not important	5	4.1
	Slightly important	11	9.0
	Important	32	26.2
	Very important	17	13.9
	Extremely important	57	46.7
	Total	122	100.0

summarizes the respondents’ answers regarding the importance attributed by their respective companies to the package of measures presented by the regional government for liquidity support during the COVID-19 pandemic. It is worth noting that the majority (approximately 87%) consider these measures to be globally necessary, very important, or extremely important (mode = 5 (Extremely important), P25 = 3 (Important), P50 = 4 (Very important), P75 = 5). Furthermore, it is noteworthy that 46.7% consider these measures to be essential.

Table 4. Degree of importance attributed to support measures for maintaining their company’s activity (absolute frequency and %).

		Frequency	%
Valid	Not important	4	3.3
	Slightly important	7	5.7
	Important	20	16.4
	Very important	25	20.5
	Extremely important	52	42.6
	The company did not benefit from any support measures since the beginning of the pandemic.	14	11.5
	Total	122	100.0

summarizes the responses given by the respondents regarding the degree of importance attributed to support measures for maintaining their company’s activity. It is worth mentioning that 42.6% of company employees stated that these measures are critical. Furthermore, it is noteworthy that 11.5% of the companies in the sample did not benefit from any support measures since the beginning of the pandemic (mode = 5 (Extremely important), P25 = 3 (Important), P50 = 5, P75 = 5).

The fourth part of the questionnaire focuses on the future perspectives of the companies. In this context, initially, five possible scenarios were presented to understand the perceptions of the respondents regarding the level of concern (1—Not concerning; 2—Slightly concerning; 3—Concerning; 4—Very concerning; 5—Extremely concerning) that these scenarios may evoke. In this regard, it is noteworthy that the collected data indicate that the reduction in customer demand, even in the context of control of the current health situation, is considered an extremely concerning scenario by a significant portion (44.3%) of the companies, as shown in

Table 5. Degree of concern of companies regarding “Reduction in customer demand” (absolute frequency and %).

	Not Concerning		Slightly Concerning		Concerning		Very Concerning		Extremely Concerning
	Freq.	%	Freq.	%	Freq.	%	Freq.	%	Freq.	%
Reduction in customer demand, even in a context of control of the health situation.	2	1.6	7	5.7	16	13.1	43	35.2	54	44.3

.

3.2. Perceptions Regarding the Level of Importance Concerning Some Measures to Promote Tourism in the Azores

3.2.1. Univariate Data Analysis

Table 6. Distribution of respondents according to the perceptions regarding the level of importance concerning some measures (A1 to A2) to promote tourism in the Azores.

	Not at All Important		Slightly Important		Important		Very Important		Extremely Important
	Freq.	%	Freq.	%	Freq.	%	Freq.	%	Freq.	%
A1	6	4.9	6	4.9	20	16.4	28	23.0	62	50.8
A2	17	13.9	18	14.8	22	18.0	20	16.4	45	36.9
A3	5	4.1	7	5.7	17	13.9	31	25.4	62	50.8
A4	2	1.6	5	4.1	20	16.4	24	19.7	71	58.2
A5	2	1.6	4	3.3	20	16.4	29	23.8	67	54.9
A6	6	4.9	12	9.8	33	27.0	32	26.2	39	32.0
A7	3	2.5	3	7.4	28	23.0	38	31.1	44	36.1
A8	1	0.8	5	4.1	21	17.2	37	30.3	58	47.5
A9	2	1.6	8	6.6	20	16.4	37	30.3	55	45.1
A10	9	7.4	13	10.7	23	18.9	35	28.7	42	34.4
A11	2	1.6	9	7.4	37	30.3	38	31.1	36	29.5
A12	3	2.5	10	8.2	33	27.0	36	29.5	40	32.8
A13	1	0.8	4	3.3	15	12.3	32	26.2	70	57.4
A14	1	0.8	1	0.8	25	20.5	42	34.4	53	43.4

and Figure 2 contain some measures of descriptive statistics (frequencies, percentages, mode, and percentiles P25 (first quartile), P50 (median), and P75 (third quartile)) regarding the level of importance (1—“Not at all important” to 5—“Extremely important”) that companies attribute to the measures described in Table 1, to promote tourism in the Azores. Regarding the notation used in Table 1, “Freq.” is an abbreviation for frequency (the number of times the value occurs in the data).

The measures that were most appreciated by participants to promote tourism in the region were those corresponding to the items A1, A3, A4, A5, and A13 (see the corresponding values of the mode and percentiles P25, P50, and P75 shown in Figure 2). The values of these percentiles are represented, respectively, by the blue, orange, and green lines in this figure.

3.2.2. Multivariate Data Analysis: Categorical Principal Component Analysis

The application of CatPCA on the submatrix that contains the 14 items (A1–A14) allowed the extraction of five principal components that explain about 83.16% of the data variance, having verified that all eigenvalues are greater than one and that the values of Cronbach’s alpha coefficient (measure of internal consistency of the items) point to an acceptable internal consistency of the items (see

Table 7. CatPCA—Matrix of Component Loadings, after rotation (14 items: A1 to A14) and Model Summary.

		PC1	PC2	PC3	PC4	PC5
Measures	A1			0.909
	A2			0.871
	A3					0.873
	A4	0.788
	A5	0.661
	A6				0.884
	A7		0.516
	A8	0.976
	A9		0.665
	A10		0.820
	A11		0.871
	A12		0.754
	A13	0.977
	A14	0.975
	Cronbach’s Alpha	00.839	00.809	00.707	00.695	0.608
	Total (eigenvalue)	40.048	30.005	10.931	10.453	10.205
	% of variance	280.917	210.464	130.794	100.379	80.606

Note: The blank spaces correspond to loadings whose absolute values are less than 0.5.

). Note that, according to Gliem [20], “(…) the closer Cronbach’s alpha coefficient is to 1.0 the greater the internal consistency of the items, and values less than 0.5 are unacceptable”.

The loadings corresponding to the essential items for each component (dimensions) are in bold in Table 3. Thus, the five extracted dimensions, PC1 to PC5, were named, respectively, “Tourism promoting strategies”; “Innovation and competitiveness of the tourist offer”; “Improvement of aerial accessibility”; “Promotion actions with emigrant communities in the USA and Canada”; and “Improvement of the inter-island transport network”. These dimensions constitute five strategic aspects that should be considered when planning tourism development in the region.

3.2.3. Multivariate Data Analysis: Hierarchical Cluster Analysis (Searching for a Typology of the Items)

Concerning the results provided by the AHCA, the best partition was the following partition into six classes (clusters), supplied by the average linkage method, at level 8 of the aggregation process; Cluster 1: {A11, A12, A10, A9}; Cluster 2: {A5, A14, A7, A8, A13}; Cluster 3: {A4}; Cluster 4 {A6}; Cluster 5: {A1, A2}; Cluster 6: {A3}. Contextually, interpreting the dendrogram from top to bottom, it should be emphasized that the items included in the first cluster reflect the second principal component (PC2); the items included in the second cluster reflect the PC1; the third cluster contains only the item A4 (“Concentrate marketing actions on attracting tourists with high purchasing power and high spending; and the fourth, fifth, and sixth clusters correspond, respectively, to PC4, PC3, and PC5. Also, it can be analyzed in the following Figure 3.

Comparing the results obtained with CatPCA and AHCA, there is a remarkable agreement between the results obtained by the two techniques of multivariate data analysis.

3.3. Results of the Research Hypotheses

Although hypothesis tests should only be carried out in the case of probabilistic samples to minimize potential biases, some nonparametric tests were applied in this study. It should be noted, however, that the results obtained are valid only for the sample under investigation and cannot be generalized to the population due to the use of a nonprobabilistic sample.

Regarding the first research hypothesis (H1), no significant correlation was found between the variables “Profitability fluctuation” and “Company’s size” using Spearman’s correlation coefficient. Therefore, the collected data do not support H1 (r_s = 0.032, p = 0.723).

About the second hypothesis (H2), there is a significant correlation between the effects of COVID-19 on “Profitability fluctuation” (variable “Profitability fluctuation”) and the level of concern regarding the reduction in customer demand, even when there is control of the COVID-19 pandemic. This finding reinforces the second hypothesis (H2) (r_s = 0.405, p < 0.001).

Using the Kruskal–Wallis test, no significant differences were found between companies in different sectors of activity regarding the effects of COVID-19 on “Profitability fluctuation” (dependent variable). Thus, the third research hypothesis (H3) was not validated. This finding may be attributed to the support provided by the regional government.

In this regard, significant differences were found between at least two of the islands of the Azores. Based on Dunn’s test, significant differences were observed regarding the oscillation of profitability between the following pairs of islands: Flores and Santa Maria, Graciosa and São Miguel, Graciosa and Faial, Graciosa and Santa Maria, Pico and São Miguel, Pico and Faial, and Pico and Santa Maria. These results support hypothesis H4.

The fifth research hypothesis (H5) needed to be validated due to the necessary assumptions for applying the chi-square test of independence not being fulfilled.

Regarding the “measures adopted by the company to adapt to the current pandemic situation,” the chi-square independence test revealed a statistically significant association between “Telework” and “Number of years of activity in the company” (χ² = 6.217, p = 0.013). Companies with more years of activity (seven or more years) are the ones that implement telework the most. Therefore, hypothesis H6 is partially confirmed. In addition, a statistically significant association was found between the following pairs of variables: “Telework” and “Company’s size” (χ² = 12.021, p = 0.001); and “Reorganization of work teams” and “Company’s size” (χ² = 8.089, p = 0.004). The conditions for applying the chi-square independence test were not satisfied for the remaining measures. Thus, hypothesis H7 is partially validated.

4. Discussion

The COVID-19 pandemic has profoundly impacted various industries worldwide, with tourism being one of the most severe impacts. The Azores Islands, known for their picturesque landscapes and vibrant tourism industry, offer a unique case study to examine the effects of the pandemic on tourism and the lessons that can be learned for future resilience. In fact, we can divide those into themes (

Table 8. Most relevant sectors and shifts observed in the Azores Archipelago due to the COVID-19 pandemic.

1. Decline in Tourist Arrivals (in a first phase)

2. Economic Implications

3. Adaptation and Innovation

4. Sustainability and Overtourism

5. Government Interventions

6. Changing Traveler Behavior

7. Collaborative Networks

8. Resilience and Future Preparedness

).

Contextually, if we analyze Table 8 in a more detailed way and corroborate those findings with previous and similar research, we have the following:

• Drastic Decline in Tourist Arrivals: The Azores Islands, like many other tourist destinations, experienced a significant decline in tourist arrivals due to travel restrictions, lockdowns, and health concerns [21,22,23]. In fact, such a decline led to an abrupt halt in the local economy, affecting businesses directly related to tourism and peripheral sectors such as transportation, hospitality, and retail [24].
• Economic Implications: The pandemic-induced downturn in the tourism sector has revealed the vulnerability of economies heavily reliant on tourism. The Azores Islands, which depended heavily on tourism, faced severe economic challenges, including job losses and reduced local revenue [25]. Therefore, it serves as a lesson for diversifying economic activities to avoid over-dependence on a single sector.
• Adaptation and Innovation: In response to the crisis, many businesses in the Azores Islands had to adapt and innovate to survive [12,26]. This period saw the rise of virtual tours, online experiences, and other digital initiatives to engage potential travelers—such adaptations underscore the importance of embracing technology to enhance the tourism experience, even in traditional and remote destinations [26].
• Sustainability and Overtourism: The sudden decline in tourism provided a temporary respite to popular tourist destinations grappling with overtourism. Thus, this experience highlights the significance of sustainable tourism practices, which focus on preserving the environment, supporting local communities, and maintaining a balance between economic growth and ecological conservation [27].
• Government Interventions: Government intervention played a crucial role in mitigating the impact of the pandemic on the tourism sector. Measures such as financial support for businesses, training programs for workers, and health protocols for reopening were instrumental in ensuring a gradual recovery [28]. However, there is a lesson in balancing short-term relief and long-term resilience.
• Changing Traveler Behavior: The pandemic reshaped traveler behavior, with safety and health concerns becoming top priorities. The Azores Islands, in response, had to implement stringent health and safety measures to regain the trust of travelers. This shift emphasizes the need for destinations to prioritize health infrastructure and communication to reassure potential visitors; in fact, it occurred a bit for all the Portuguese territories (see: [29]).
• Collaborative Networks: The crisis highlighted the importance of collaboration between stakeholders in the tourism sector, including governments, local communities, businesses, and tourists. The Azores Islands’ experience underscores the significance of building resilient networks that can coordinate responses to crises and work collectively toward recovery [30].
• Resilience and Future Preparedness: The Azores Islands’ journey toward recovery is a testament to resilience and adaptability. This experience serves as a reminder that disruptions, though challenging, can also offer opportunities for growth and transformation. Learning from the Azores Islands’ response can inform global strategies for future preparedness as the world grapples with the pandemic’s effects [31].

Furthermore, studying the effects of COVID-19 on the tourism sector in the Azores Islands offers valuable insights for other island territories facing similar challenges. Lessons learned from the Azores include diversifying tourism offerings, promoting domestic and regional tourism, prioritizing sustainable development, embracing digital tools, preparing for emergencies, fostering collaboration, investing in health and safety measures, building resilience, involving local communities, and having a long-term vision. Thereby, these lessons can guide other island territories in developing a more resilient and sustainable tourism sector in the face of COVID-19 and related challenges.

5. Conclusions

Therefore, the COVID-19 pandemic’s effects on the tourism sector, as observed through the lens of the Azores Islands, highlight the importance of adaptability, sustainability, collaboration, and innovation. As the world goes toward recovery, the lessons learned from this unique case study can serve as a valuable guide for rebuilding a more resilient and inclusive tourism industry.

In addition, the most adopted measure by companies to adapt to the current pandemic was temporary closure. At the same time, those that implemented a reorganization of work teams mostly resorted to partial layoff measures. In the case of teleworking, most companies did not provide technological equipment or training. On the other hand, for companies with in-person work, the necessary materials were made available for employee protection, and the costs of these materials had some impact on company expenses. Most companies invested in training to obtain the “Clean and Safe” security seal. Among the companies that used alternative customer contact channels, online sales were the most commonly used method. In fact, such outcomes could be corroborated by previous studies (see [32,33,34,35,36,37,38,39,40]).

Companies believe that COVID-19 significantly impacted their profitability, primarily due to a decrease in current orders and customers. Most companies availed themselves of the main support programs for tourism introduced by the regional government and consider these measures essential for sustaining normal business operations.

Regarding prospects, companies in the region intend to retain their workforce. Still, their main concern is the potential decrease in customer demand, even with control over the current health situation.

Regarding the study’s limitations, it should be noted that the hypothesis tests were performed for illustrative purposes only, as the sample used was nonprobabilistic. Therefore, the results are only valid for the specific sample under investigation and may not be generalized to the entire population. Additionally, the sample size is another limitation of the study.

Study Limitations and Future Research

The primary constraints of this study pertain to the utilization of the pandemic years for evaluating the impact, a timeframe that effectively extends into 2021.

Consequently, shortly, there will be an opportunity to appraise the analysis over the two years most significantly influenced by the COVID-19 health crisis. This assessment can then be compared against the pessimistic projections of managers within the tourism sector for 2021. In the context of this study, it is essential to underscore its originality, stemming from its application in gauging the effects of the recent health crisis on the Azores’ economy. This insular region notably saw the prominence of the tourism sector in 2019, drawing from a substantial pool of tourism industry entrepreneurs and managers within the autonomous region of the Azores. This research’s potential for replication in other geographic locales contributes to a comprehensive comparative examination of COVID-19’s impact on the tourism sector across islands and archipelagos. These regions exhibit heightened dependence on this sector, which underscores their vulnerabilities. Undeniably, this sector plays a pivotal role in the economies of most islands.

One additional constraint lies in the sampling method, where the study relies on a convenience sample, thereby hindering the generalizability of results to the entire population. Nonetheless, this limitation is mitigated by the considerable sample size and the focus on the most robust findings.