European outbound tourism expansion on the islands of Cape Verde

Abstract

The main objective of this article is to shed light on the determinants of international tourism demand expansion in Cape Verde. The research involves cointegration analysis to test the existence of a long-term equilibrium relationship between the variables included in tourism demand using the autoregressive distributed lag bounds testing approach. The results showed high income elasticity and price inelastic demand. Moreover, turmoil in traditional tourist destinations in non-European Mediterranean basin countries, coupled with the investments of tour operators and hotel groups who are key players in the European source markets, significantly and positively affected international tourism demand. Supply side changes associated with foreign investment were necessary to increase the number of tourists. Finally, lower level of income in European countries following the financial crisis of 2008 might have favored the shift of European tourists from more expensive destinations to Cape Verde. The concentration of tourism, economic growth, jobs creation, and population growth on two islands may have affected island communities, resources, and other aspects of inter- and intraisland relationships.

Keywords

ARDL cointegration Cape Verde foreign investment international tourism demand Small Island Developing State (SIDS)

Introduction

Many scholars have argued that the high growth of international tourism demand (Cobbinah et al., 2013; Gauci et al., 2002; Holden et al., 2011; WEF, 2017) could be a pillar to stimulate economic growth (Bouzahzah and El Menyari, 2013; Nunko et al., 2019), particularly in some island territories endowed with key characteristics for the development of the tourism sector (Seetanah, 2011). These benefits of tourism (Moreno et al., 2020) have led governments and international organizations to implement programs and policies with the objective of increasing economic growth and employment. However, obstacles such as insufficient investment in the tourism sector (El Menyari, 2021; Fauzel et al., 2017) or uncertainty dissuading people from visiting risky destinations (among others, Buigut and Amendah, 2016; Naudé and Saayman, 2005; Pizam and Fleischer, 2002) explain how low- or medium-income countries, particularly island territories endowed with natural, historical, and cultural resources, stay in a “vicious circle” and “low level of economic development” (Naftziger, 2006; Soubbotina, 2004).

The case of the Small Island Developing State (SIDS) of Cape Verde shows that foreign investment overcomes shortages of domestic capital and, coupled with other factors, raises the level of competitiveness in the international tourism market (UNCTAD, 2018). As a consequence, the number of tourists in hotels and similar establishments has more than quintupled over the last 20 years. Cape Verde has increased its market share in the international tourism market, and the archipelago is among the countries with the highest share of tourism in terms of value added, employment, and exports (WTTC, 2017). The WEF (2017) ranks Cape Verde 83rd out of the 136 countries included in the travel and tourism competitiveness index. It is ranked as the seventh African country with the highest level of competitiveness, behind South Africa (53rd), Mauritius (55th), Morocco (65th), Egypt (74th), Kenya (80th), and Namibia (82nd). Cape Verde improved considerably and rose six places in the ranking since 2011, the year when the country was included in the analysis for the first time (WEF, 2011). However, the fact that the archipelago is ranked behind other small island territories included in the index—Malta (32nd), Cyprus (52nd), Mauritius (55th), Barbados (58th), Jamaica (69th), and the Dominican Republic (76th)—shows space for further improvement.

Previous literature on the tourism–economic growth interaction has considered different factors to explain the evolution of tourism and their net economic benefits (Bouzahzah and El Manyari, 2013; Nunko et al., 2019). Among these factors, the replacement of a centralized model with a more open and marked based economy, inflows of foreign direct investment (FDI), and the sun, sand, and sea resources of the islands have played an important role in Cape Verde (UNCTAD, 2018). Although those supply side changes associated with FDI are necessary, the ultimate determinant of tourism arrivals is demand in the source markets, that is, tourists’ destination choices.

Several published studies on tourism demand have focused on destinations in Africa, among them Ibrahim (2013), Wahba and Elanani (2017), and Faragalla (2018) in Egypt; Holden et al. (2011) in Ghana; Buigut and Amendah (2016) in Kenya; Gooroochurn and Sinclair (2005) and Seetanah et al. (2015) in Mauritius; Awaritefe (2007), Bankole and Babatunde (2010), and Eja et al. (2012) in Nigeria; Burger et al. (2001); Saayman and Saayman (2015), Durbarry et al. (2009), Seetanah et al. (2010), Gil Alana (2011), and Moyo and Ziramba (2013) in South Africa; Ouerfelli (2008), Choyakh (2008), Ouerfelli (2010), Gasmi and Sassi (2015), Hathroubi (2011), and Sassi (2014) in Tunisia; Muchapondwa and Pimhidzai (2011), Chigora and Vutete (2015), and Mpofu (2009) in Zimbabwe; and Gauci et al. (2002), Naudé and Saayman (2005), Adeola et al. (2018), and Viljoen et al. (2019) included groups of countries. Despite the remarkable role that tourism has played in the economic performance of the SIDS of Cape Verde over the last few decades, to the best of our knowledge, no empirical study has been carried out. Thus, it would be convenient to examine the drivers behind the demand dynamics of visitors to Cape Verde. According to previous research on modeling international tourism demand,¹ the most salient determinants of the number of foreign visitors are consumers’ income in the source market and the relative prices of goods and services in destinations compared with prices in tourists’ countries of origin (Adeola et al., 2018; Saayman and Saayman, 2015; Saayman and De Klerk, 2019; Song and Witt, 2000; Viljoen et al., 2019, and more particularly Seetanah et al., 2015 for the case of islands’ tourism demand).

Apart from economic explanatory factors, other favorable or unfavorable circumstances might affect the evolution of tourism demand. Given that tourists are highly mobile, if they perceive a destination to be risky, they can omit it from their destination choices or change decisions they have already made (Buigut and Amendah, 2016; Hall and O’Sullivan, 1996; Naudé and Saayman, 2005; Pizam and Fleischer, 2002). The impact on tourism demand depends on tourists’ risk aversion (Lepp and Gibson, 2003) and the visiting country’s stage of development (Thompson, 2011). Following the Arab Spring (AS) in North African and Persian Gulf countries, riots and revolutionary movements have generated negative publicity since the beginning of 2010 and diverted tourism from these destinations. Two circumstances might have contributed to the diversion of international tourists from nations affected by the AS to Cape Verde. First, tour operators and hotel companies that are the main investors in the tourism industry in Cape Verde are also main players² in the European outbound tourism market. Second, European countries have been the main source markets of international tourism to countries affected by the AS. Most studies have focused on the loss of tourism receipts in different destinations where attacks were committed. However, less attention has been paid to gains in the number of international travelers to substitute destinations that have remained safe and peaceful in contrast to European outbound tourism destinations that have suffered from social and political instability, such as countries affected by the AS.

The main contribution of the present study is to shed light on international tourism expansion in Cape Verde. First, we define and estimate explanatory factors of tourism demand as fundamental conditions for tourism expansion within a tourism demand model and using cointegration analysis. Second, we specifically take into account the gains in international tourism in Cape Verde, a destination that remained safe and peaceful in contrast to European outbound tourism destinations in the Mediterranean that suffered impacts of social unrest and terrorism. Better knowledge of the effects of determinants can help policy makers and managers in their decision-making processes not only in Cape Verde but also in other SIDSs medium-haul destinations endowed with attractive natural resources (e.g. sun, sea, and sand) for improving tourism competitiveness.

The structure of the article will be as follows. First, the article will describe the evolution of tourism on the Islands of Cape Verde and their source markets in the second section. Second, the third section will present estimates of the explanatory factors of international mass tourism within a tourism-demand framework. Finally, the fourth section will conclude.

Overview of the evolution of tourism in the islands of Cape Verde

Cape Verde (Fernandes, 2005) is one of the archipelagos that are part of Macaronesia, which also includes Azores in the southwest of Europe and the archipelagos of Madeira and the Canary Islands in the northwest of Africa (Figure 1). Population on the nine islands increased from about 200 thousand people in 1960 to over half a million in 2016 (INE, 2016, 2017). There are international airports on the two most populated islands, Santiago (where the country’s capital Praia is located) and São Vicente, and on the two main touristic islands, Sal and Boa Vista. There are also airports with domestic flight connections on the islands of Fogo, São Nicolau, and Maio.

Figure 1.

The Cape Verde archipelago.

Given the absence of sufficient domestic funds for investment, the development of tourism in Cape Verde has relied on foreign investment (UNCTAD, 2018). Figure 2 displays both the evolution of total FDI in the tourism sector and the number of tourist arrivals. FDI inflows to Cape Verde underwent a great increase from around a yearly average of 10 million euro in the 2001–2005 period to nearly 80 million in 2006–2010 and over 50 million in 2011–2019. The number of international tourists grew from approximately 152 thousand per annum in 2001 to 280 thousand in 2006, 475 thousand in 2011 and over 800 thousand in 2019. The country received these inflows after transformation from a centralized planned economy into a more open and liberalized market economy (UNCTAD, 2018). The archipelagic state was able to provide foreign investors, mainly from Europe, with sound economic policies and specific island resources to develop “all-inclusive” tourism projects. The main investors have been well-known tourism operators and hotel groups (different companies of the TUI group: TUI, Thomson, RIU; the disappeared Thomas Cook group, DER touristic group, Resort Group; Iberostar, Meliá; Barceló). Thus, a key supply side expansion of the tourism industry in Cape Verde, driven by tour operators and characterized by low priced packages and charter flights, took place.

Figure 2.

Evolution of FDI and international tourism demand in Cape Verde (2000–2019).

Source markets development

Figure 3 displays the development of international tourism demand, measured in number of nights stayed in hotels or similar accommodations in Cape Verde and separately the two main destination islands of the archipelago. From 2000 to 2019, the number of nights spent by international tourists in Cape Verde increased by around 4.4 million—more than 3.6 million of them on the islands of Sal and Boa Vista and only 0.8 million on the rest of the islands, including the more populated Santiago, São Vicente, and Santo Ãntao (see Figure 1). The islands of Sal and Boa Vista increased their share in number of nights spent by international tourists from nearly 70% in 2000 to 90% in 2019.

Figure 3.

Evolution of number of nights spent by international tourists on main tourist islands (2000–2019).

Figure 3 shows the development of nights spent by international tourists from the 12 major countries of origin included separately in the tourism statistics of Cape Verde. The 10 largest European³ source markets for tourists (Austria, Belgium, France, Germany, Italy, the Netherlands, Portugal, Spain, Switzerland, and the United Kingdom) account for over 80% of international tourism to Cape Verde.⁴ The share of source markets has changed over the period of study. Italy and Portugal were the main source markets in 2000, with 20% of international tourists each (Figure 4). Although all the main source markets increased in absolute terms, the United Kingdom showed the highest growth and became the main country of origin. The share of UK residents increased⁵ from fewer than 1% of the visitors arriving in Cape Verde in 2000 to over 25% in 2019. Belgium and the Netherlands showed the second-highest growth, increasing their participation from 2% to nearly 10%. Germany maintained its relative value (12%) but quadrupled its number of visitors. Finally, Portugal and Italy reduced their participation in relative terms to less than half of their values in 2000 but Portugal tripled the numbers of visitors in the same period.

Figure 4.

International tourists by main source markets (2000–2019).

Figure 5 shows that the average length of stay increased from 4.2 nights in 2000 to 6 in 2019. Visitors from France and Portugal showed the lowest length of stay, while the United Kingdom was the highest, followed by Belgium, the Netherlands, and Germany.

Figure 5.

Development of average stay by country from main source markets. Source: Author’s own calculations based on data obtained from the Statistical Office of Cape Verde.

Population and gross domestic product in main tourist spots

The concentration of foreign investment and tourism on the islands of Sal and Boa Vista has affected the distribution of populations, employment, and income. First, Figure 6 shows the annual average growth rate (AAGR) of the population for the islands. Whereas Sal and Boa Vista achieved growth rates close to 9% and 17%, respectively, the archipelago of Cape Verde had an AAGR close to 1%. Some islands, such as São Nicolau, Brava, Santo Antão, and Fogo, even showed negative population growth rates. As can be seen in Figure 6, the period between 1970 and 1990, prior to the development of tourism, showed low differences in population growth among the islands. Regarding employment, Figure 7 displays the differences in employment across the islands in terms of AAGR. The islands of Sal and Boa Vista show an AAGR that is three and five times higher than Cape Verde’s average, reflecting the employment created from coastal tourism. This concentration of tourism activities, population, and employment may have caused overdevelopment migration as well as changes in transportation and mobility, environmental quality, heritage, and community values (UNCTAD, 2018).

Figure 6.

Population AAGR on the island before (1970–1990) and during tourism boom (2000–2016). Source: Author’s own calculations based on data obtained from the Statistical Office of Cape Verde. AAGR: annual average growth rate.

Figure 7.

Employment AAGR for population over the age of 15 (2011–2016). Source: Author’s own calculations based on data obtained from the Statistical Office of Cape Verde. AAGR: annual average growth rate.

Finally, tourism has also effected changes in gross domestic product (GDP). The two main destination islands showed an AAGR for the period between 2007 and 2016 of up to four times the Cape Verde average (Figure 8). Moreover, people living on the tourism islands have a higher income on average—approximately US$7000 in Sal and US$5000 in Boa Vista, compared to US$3000 in Cape Verde as a whole (Figure 9). Those differences in levels of income imply that Boa Vista and Sal are in a higher income group than the rest of the Cape Verde islands, according to the World Bank’s (2018) country income classifications. Sal and Boa Vista share the income group of island states such as Grenada, Jamaica, the Maldives, the Marshall Islands, Mauritius, Saint Lucia, and Saint Vincent and the Grenadines, whereas the rest of the Cape Verde islands belong to the same income group as Sao Tome and Principe, Sri Lanka, and the Philippines.

Figure 8.

Real GDP growth* (2007–2016). Source: Author’s own calculations based on data obtained from the Statistical Office of Cape Verde. *The GDP at current prices was deflated by the GDP deflator for Cape Verde. GDP: gross domestic product.

Figure 9.

GDP per capita (US$) in 2016. Source: Author’s own calculations based on data obtained from the Statistical Office of Cape Verde. GDP: gross domestic product.

COVID-19

Tourists are afraid of catching COVID-19 while on a plane (OAG, 2020) or needing health care in foreign countries, which makes them hesitant to travel. This combined with quarantines and travel bans imposed by governments in the tourists’ countries of origin has had devastating effects on the tourism industry in Cape Verde. Many holiday resorts have remained closed in Cape Verde since March 2020 and international tourism receipts shrank by 11% and 91% during the first and second quarters of 2020, respectively, compared with the same periods in 2019 (INE, 2020; UNWTO, 2020a). Businesses in the tourism industry, including local small and medium enterprises that were viable before the pandemic, might now be vulnerable (UNWTO, 2020b). To support business survival, the government in Cape Verde has addressed liquidity shortages with assistance from international organizations.⁶ Moreover, resort owners in Cape Verde (companies of the TUI group,^7,8 among others) have received state loans in the countries where their headquarters are located. However, the financial resources might not be enough to protect jobs and income to mitigate the effects of the crisis on workers in this labor-intensive sector (UNWTO, 2020b). The existence of effective vaccines against COVID-19 is an important step in the fight against the pandemic, but it is too early to assess the effects of the pandemic on the tourism sector in the country. It will depend not only on the effects of the lockdown but also on when a full return to tourism activities will be possible. Stakeholders and authorities in Cape Verde and other destinations seek to restore tourists’ confidence and restart tourism, for example through the creation of safety corridors and other specific measures to stimulate international tourism. Balanced measures that avoid both delays in reopening and moving too quickly are needed. On the one side, delays in reopening might cause job losses and business closures. On the other side, moving too quickly might cause a backslide in visitors’ recovery of confidence. Cape Verde is heavily reliant on intercontinental tourism, and, according to the results of the study conducted by Kourentzes et al. (2021), long-distance travelling (e.g. from France, the United Kingdom, Germany, or the United States to South Africa and/or Mauritius) is predicted to experience longer recovery time to register similar tourist arrivals compared to before the pandemic.

Main determinants of international tourism demand

Methodology and data

Our analysis of the explanatory factors of the evolution of Cape Verde’s international tourism demand (Song and Witt, 2000; Saayman and Saayman, 2015; Seetanah et al. 2015) can be defined as follows:

ln T_{t} = β_{0} + β_{1} \cdot ln P_{t} + β_{2} \cdot ln {GDP}_{t} + β_{3} \cdot AS + β_{4} \cdot CRISIS + u_{t} \forall t = 1, ..... T

where T_t is the international tourism demand variable at time t; P_t is the price of tourism in Cape Verde at time t; GDP _t is the income level in the source markets at time t; AS is a dummy variable that collects the influence of the political and social instability that followed the AS in many non-European Mediterranean basin countries, CRISIS is a dummy variable that captures the effect of the international financial crisis and sovereign debt crisis in Europe, and u_t represents the disturbance term.

The demand function in equation (1) can be expressed as the corresponding error correction model (ECM) using a linear transformation (Banerjee et al., 1993). This modeling representation, which is called autoregressive distributed lag (ARDL-ECM), assumes a unique model that integrates the short-run dynamics with the long-run equilibrium (Cetin et al., 2015) and has the merit of simplifying the computation of the variances of the coefficients (Balg and Metcalf, 2010). The ARDL-ECM approach entails estimating the following expression:

Δ ln T_{t} = γ_{1} {CRISIS}_{t - 1} + γ_{2} {AS}_{t - 1} + \sum_{j = 1}^{p - 1} α_{j} \cdot Δ ln T_{t - j} + \sum_{j = 0}^{p - 1} δ_{1 j} Δ ln {Price}_{t - j} + ​ \sum_{j = 0}^{p - 1} δ_{2 j} Δ ln {GDP}_{t - j} + ϕ ln ​ T_{t - 1} + ϑ_{1} ln {Price}_{t - 1} + ϑ_{2} ln {GDP}_{t - 1} + ε_{t}

where Δ is the first-difference operator, $α_{j}$ and $δ_{i j}$ reflect the short-term dynamics of the analysis, $ϕ$ and $ϑ_{i}$ are the parameters that represent the long-term relationship and γ_i describes the effect of the dummy variables. p is the lag length and $ε$ _t is the white-noise term. T_t is the dependent variable that measures international tourism demand, approximated by the number of guests from the 12 major source markets registered in hotels and similar establishments. The data were obtained from the Cape Verdean Statistical Office.

The justification of the chosen variables is as follows. A potential determinant of demand is income in the tourist source markets. A higher level of income in the EU countries is expected to raise the number of visitors in Cape Verde. The most common variable taken as an approximation for consumers’ income is GDP defined as

{GDP}_{t}^{} = \sum_{j = 1}^{5} {GDP}_{j t} w_{j t}

The weights w_jt are the shares of source markets considered separately in the tourism statistics of Cape Verde at time t (the United Kingdom, South Africa, Switzerland, the United States, and eight Euro area member states⁹: Austria, Belgium, France, Germany, Italy, the Netherlands, Portugal, and Spain). GDP time series for South Africa and the United States come from the International Monetary Fund (IMF) database. The GDP data for the United Kingdom, Switzerland, and the Euro area were obtained from the Statistical Office of the EU (Eurostat). These time series were adjusted to control the seasonality of the data.

Second, if the price of visiting Cape Verde in the currencies of foreign visitors increases, the number of tourists will decrease and vice versa. We approximated the price at time t (Price t) by the cost of living using levels of inflation (consumer price index (CPI)¹⁰), adjusted by exchange rate:

{Price}_{t} = \sum_{j = 1}^{5} \frac{{CPI}_{t}^{CV}}{{CPI}_{j t}} w_{j t} ​ \frac{1}{{ER}_{j t}}

where ER _jt are the nominal exchange rates between the currencies of the tourist destination and the source markets at time t, in other words, the amount of Cape Verdean escudos (CVE) needed to buy one euro (€), one pound (£), one Swiss franc (CHF), one South African rand (ZAR), or one US dollar ($). The weights w_j are the share of source markets. CPI _t ^CV and CPI _jt are the CPI for Cape Verde and the major source markets at time t, respectively. The CPI for Cape Verde CPI^CV was downloaded from the web page of the National Statistical Office of Cape Verde. The CPI for the Euro area and United Kingdom are available on the web page of the Statistical Office of the European Union (Eurostat). CPIs for South Africa, Switzerland, and the United States come from the IMF database. The exchange rates were obtained from the Central Bank of Cape Verde (Banco Central da Republica de Cabo Verde).

Researchers agree that tourists’ decision-making, particularly the destination, can be severely affected by external shocks such as terrorist attacks, instability, or violence (Buigut and Amendah, 2016; Naudé and Saayman, 2005; Pizam and Fleischer, 2002). When considering European outbound tourism, social and political instability associated with the AS in non-European Mediterranean basin countries may have diverted European outbound tourism flows to alternative destinations, including Cape Verde. Two circumstances may have contributed to the diversion of international tourists from nations affected by the AS to Cap Verde. On the one hand, the main investors in tourism projects in Cape Verde are also key players in the flows of European outbound tourism (TUI group, DER Touristik group, Resort group, Meliá, Iberostar, Barceló). On the other hand, European countries have been traditionally the main source markets of visitors to African and Asian Mediterranean basin countries. AS is a dummy variable that aims to consider the effects of the rebellion waves that began in Tunisia and spread with speed and magnitude to other countries of the Mediterranean basin. The dummy takes the value 1 for the period between April 2010 and December 2019 and 0 for the period between January 2001 and March 2010. The global financial crisis that started 2007/2008 hit Europe hard. According to the European Commission (2017), the GDP fell by more than 5% from the last quarter of 2007 to the last quarter of 2008. The recovery from the sovereign debt crisis that followed the financial crisis started in 2013. Different studies have analyzed tourists’ behavior and economizing strategies under crisis times. Papatheodorou et al. (2010) and Campos-Soria et al. (2015) state that under crisis it is more likely that tourists stay closer to home and scale back their expenditure per night or their total holiday expenditure. The dummy variable CRISIS is included to account for the effects of this deep recession in the source markets on outbound tourism (ETC, 2009; ONS, 2017) to Cape Verde. The variable takes a value of 1 for the period 2008: Q1-2013: Q1 and 0 the rest.

The variables have been transformed using a logarithm to reduce the variability of the data and to interpret the estimated coefficients of the explanatory variables as elasticities. Table 1 shows main statistics descriptive of the variables. The study is based on 76 quarterly observations and covers the period from 2001: Q1 to 2019: Q4.

Table 1.

Descriptive statistics.

Variable	Max	Min	Mean	SD
$Ln$ T	12.36	10.46	11.44	0.56
$Ln GDP$	4.7	4.20	4.65	0.14
Ln price	4.62	4.49	4.57	0.03
$Arab Spring$	1	0	0.46
Crisis	1	0	0.27

Results

We considered different cointegration techniques to study the existence of a long-run equilibrium relationship between the variables included in the tourism demand. We made use of the ARDL approach to cointegration developed by Pesaran et al. (2001) because this methodology offers advantages in comparison with the traditional cointegration techniques (El Engle and Granger, 1987; Johansen and Juselius, 1990). First, the ARDL approach can be used irrespectively of whether the underlying explanatory variables are I(0), I(1), or a mixture of both (De Vita and Abbott, 2002). Second, this technique has been used in time series analyses in a number of fields (e.g. tourism: Saayman and Saayman, 2015; finance: Belke and Polleit, 2006); agriculture: Muchapondwa, 2009; energy: Narayan and Smyth, 2005). Third, the ARDL model takes a sufficient number of lags to appropriately represent the data-generating process and different variables can be assigned different lag-lengths. Another reason that justifies the use of this approach is that even with small or finite sample data sizes, it is consistent and more efficient than the traditional cointegration techniques (Pesaran and Shin, 1999). However, despite this great advantage, it is necessary to check that the variables are not I(2). If this were the case, the approach would produce spurious results.

Therefore, it is necessary to know the order of integration of the variables included in the estimation (Pesaran et al., 2001). Table 2 shows the results of the ADF and KPSS unit root tests. In our observations, the tests provide clear evidence that none of the variables are I(2) and, consequently, the ARDL bounds testing approach to cointegration is applicable for studying the existence of a cointegration relationship between the variables defined in equation (2).

Table 2.

ADF and KSPSS unit root tests results.

	ADF test $H_{0} : Unit Root$	KPSS Test $H_{0} : Stationary$
T	−0.446 (11)	0.114 (6)***
GDP	−0.872 (11)	1.013 (6)***
Price	−1.31 (0)	1.07 (6)***
ΔT	−5.010 (11)***	0.084 (13)
ΔGDP	−3.631 (10)***	0.134 (14)
ΔPrice	−8.1 (0)***	0.078 (9)

Note: GDP: gross domestic product. The symbol *** means rejection of the null hypothesis at the 1%, level. For the KPSS tests, the number of bandwidth are shown in brackets according to the Newey–West Criterion using Bartlett Kernel. For the ADF test, the number of lags are shown in brackets according to the Schwarz information Criterion.

However, we must first determine an appropriate lag length p for the ARDL model represented in equation (2). Pesaran and Shin (1999) recommended choosing a maximum of eight lags when the periodicity of the data is quarterly. In our study, the optimal number of lags is selected according to the Schwarz Bayesian information criterion, which is found to be (3, 1, 0).

The cointegration procedure is based on two tests to check the null hypothesis that the variables are not cointegrated (Pesaran et al., 2001). First, an F-statistic to test the joint nullity of all long-run regression coefficients $H_{0} : ϕ = ϑ_{1} = ϑ_{2} = 0$ . The second statistic is a t-test that checks the nullity of the parameter associated with the lagged dependent variable $H_{0} : ϕ = 0$ . Pesaran et al. (2001) proposed upper and lower critical value bounds that need to be compared to the calculated F and t statistics. There are three possible results: (i) inconclusive inference if the values of the tests are between the upper and lower critical bounds; (ii) rejection of the null hypothesis on cointegration and evidence of a relationship between the variables if the statistics of the tests are above the upper critical value of the bound; and (iii) nonrejection of the null hypothesis on no cointegration and nonexistence of a relationship if the statistics of the tests are below the lower critical value bounds.

Table 3 shows the values of the F and t_I statistics and the corresponding critical value bounds for two explanatory variables represented in Pesaran et al. (2001). As we can see, the F-statistic is F = 3.95 and t-statistic is |t| = 9.04, which in both cases are above their respective critical bounds at the 5% and 1% significance level, respectively. Therefore, the null hypotheses on no cointegration is rejected and there is conclusive evidence of a long-run relationship between the number of visitors to Cape Verde and the explanatory factors (e.g. GDP, Price, AS, and CRISIS).

Table 3.

Critical values and bound test for cointegration.

Test value	Lower bound critical value	Upper bound critical value	Long-run relationship
F_I = 3.95	2.35**	3.63**	YES
$\|t_{I}\| =$ 9.04	3.42**	4.84***	YES

Note: F_I is the value of the F-statistic associated with case I (Pesaran at al., 2001. The critical values are for a level of significance of α = 0.05**. t_I is the t-statistic associated with case I (Pesaran at al., 2001). The critical values are for a level of significance of α = 0.01***.

Table 4 provides the ARDL-ECM that captures this relationship. The signs of the long-run coefficients are consistent with the economic theory and are statistically significant. They are estimated to be −0.19 and 1.13 for price (Price) and income (GDP), respectively. Moreover, the coefficients of the dummy variables AS and CRISIS are also statistically significant and show the expected signs. First, the less stable political and social background in these competing non-European Mediterranean tourism destinations has lent a sizable advantage for European outbound tourism to Cape Verde and its competitive advantage relative to them has improved. Terrorist attacks and political and social unrest have affected some traditional European outbound tourism destinations, mainly Tunisia and Egypt, whereas there has been a lack of security concerns in Cape Verde. Second, the variable CRISIS has a statistically significant positive impact on European inbound tourism to Cape Verde, indicating that more European tourists chose this tour operator driven market destination characterized by low priced packages and charter flights during the crisis. This suggests that Cape Verde fitted the needs of European tourists, who opted to cut back on holiday expenditures during the crisis period. These results are in line with previous literature that found evidence that tourists economize in holiday expenditures rather than take fewer holidays. (Campos-Soria et al., 2015; Eugenio-Martin and Campos-Soria, 2014; Papatheodorou et al., 2010).

Table 4.

Long-run and short-run coefficients and diagnosis tests.

	Estimated value	p-Value
Estimated long-run coefficients
ΔT (lagged dependent variable)	−0.42	0.00
GDP	1.13	0.00
Price	−0.19	0.05
Arab Spring	0.104	0.06
Crisis	0.075	0.01
Estimated short-run coefficients
ΔPrice _t	0.28	0.61
ΔGDP _t	0.926	0.16
ΔT_t₋₁	−0.25	0.04
ΔGDP _t _-1	1.97	0.00
ΔT_t₋₂	−0.46	0.00
ΔT_t₋₃	−0.38	0.00

Finally, the estimated coefficient $\hat{ϕ}$ of the lagged dependent variable $_{T_{t - 1}}$ is equal to −0.42. This estimated coefficient gives us valuable information on the speed of adjustment to the long-run equilibrium in case of a short-term shock on demand. Specifically, any deviation from the long-run equilibrium caused by a shock is corrected by nearly 42% over the following quarter. Moreover, the negative sign and the high significance of the coefficient corroborate the earlier finding of the existence of a long-run relationship (Granger et al., 2000).

Table 5 also shows the diagnostic tests that verify the econometric validity and strength of the estimated coefficients. Based on tests commonly used in the literature, the null hypothesis on homoscedasticity (Breusch–Pagan–Godfrey test), no serial correlation (LM test and Ljung–Box Q tests), normality (Jarque–Bera test), and no misspecification (Ramsey test) were rejected. Furthermore, the goodness of fit (adjusted R² = 0.74) was relatively high.

Table 5.

Diagnosis tests.

			Value	p-value
Adjusted-R²			0.74	—
Autocorrelation	Lagrange multiplier (LM) test	LM(1)	0.63	0.43
Autocorrelation	Lagrange multiplier (LM) test	LM(2)	1.31	0.0.28
Heteroscedasticity	Breusch–Pagan–Godfrey test		1.47	0.16
Normality	Jarque–Bera test		3.65	0.16
Misspecification	Ramsey test		0.59	0.56

To quantify the effects, we calculate the point estimates of the elasticities using the expression (equation (5)) proposed by Bardsen (1989):

{\hat{ϕ}}_{i} = - \frac{{\hat{ϑ}}_{i}}{\hat{φ}} \forall i = 1, ... K

where ${\hat{ϑ}}_{i}$ and $\hat{ϕ}$ are the estimated long-run coefficients in equation (2). The estimated values of the elasticities allow us to assert that a 1% rise in the GDP of the countries of origin would increase tourism to Cape Verde by nearly 2.7%, and international tourism to Cape Verde can be regarded as a luxury good. This finding is in line with previous elasticity estimates for outbound tourism in European countries (Alvarez-Diaz et al., 2015; Daniel and Ramos, 2002; Rey et al., 2011) and destinations in Africa (Saayman and Saayman, 2015; Viljoen et al., 2019). It implies that even moderate GDP growth rates in the EU increase international tourism to Cape Verde.

The price elasticity tells us that a 1% price increase reduces international tourism to the archipelago by 0.45%, indicating a price inelastic international tourism demand for Cape Verde. Khandaroo and Seetanah (2008) and Saayman and Saayman (2015) among others have also found a relatively price inelastic demand for tourism destinations in Africa.

The calculated price elasticity also indicates that the fall in revenue caused by a price decline in holiday packages would be higher than the revenue increase due to the greater number of tourists (Song et al., 2003), unless the price decline raised the international tourism revenue by extending the length of guests’ stay as Song et al. (2010) reported.

The coefficients of the price variable in the short run are not statistically significant. This implies that short-run price does not play a major role in determining the demand for Cape Verde international tourism and may suggest that tourists cannot fully adjust their behaviors in the short run when the price changes (Li et al., 2004).

Regarding the dummy variables, turmoil in the Mediterranean basin countries of North Africa and Asia has led to a 10.4% increase in international tourism to Cape Verde and the number of arrivals increased by 7.5% during the 2008–2013 crisis period.

Concluding remarks

The aim of this empirical study was to shed light on the determinants of European tourism arrivals in Cape Verde as one of the most recent SIDSs to experience expansion of international tourism demand. Consequently, tourism has become its main source of tax revenue, job creation, GDP-per-capita growth, and foreign exchange supplies. The main results and practical implications derived from our study can be summarized in the following points:

As expected, our findings reveal that income in visitors’ country of origin of has a long-run effect on the number of arrivals. It is a main explanatory variable of European tourism arrivals to Cape Verde. This result could be helpful for the design of more efficient operational strategies; for example, the estimated income elasticity can be useful to assess how economic cycles in the source markets affect future international tourism to Cape Verde.

The relatively price inelastic demand is also in line with previous studies for destinations in Africa. Despite this price inelastic demand, authorities should be aware of inflation differentials with respect to the source markets adjusted by the exchange rates that despite the price inelastic demand would cause declines in European tourism arrivals to Cape Verde. The calculated price elasticity provides information for professionals in the tourism industry and policy makers to maximize tourism revenue.

It is important to note the stable macroeconomic background with low inflation rates during the period of analysis. Consequently, macroeconomic stability has successfully helped to attract foreign investment to the tourism sector. Moreover, given that inflation rate in Cape Verde shows a similar development as in the tourist’s origin countries, there was no need for a depreciation of the domestic currency (Cape Verdean Escudo) against the currencies of the tourists’ source markets, mainly euro and pound, to offset the lost tourism price competitiveness of Cape Verde relative to other destinations. The depreciation of the own currency against currencies of tourist’s origin countries could have harmed the revenue of inbound tourism, as Chao et al. (2013) reported. In addition, according to previous studies, exchange rate volatility can affect visitor’s expenditure and arrivals (Saayman and Saayman, 2013).

Qualitative factors were found to be relevant in explaining tourism arrivals to this medium-haul destination. First, we found a positive effect of the years of deep economic crisis in the source markets of tourists. This suggests that European tourists, who economized in holiday expenditures, chose the destination of Cape Verde characterized by low priced packages and charter flights.

Second, results indicate that the instability that followed the AS in many non-European Mediterranean basin countries accelerated the expansion of international tourism in Cape Verde. The political and social instability in those countries appears to be the most relevant determinant of international tourism to Cape Verde. Three circumstances might have played a key role in taking advantage of the diversion of international tourists from non-European Mediterranean basin countries to Cape Verde. The first circumstance is the country’s relative geographical closeness to Europe (Viljoen at al., 2019). Second, the main investors in Cape Verde tourism projects are key players (UNCTAD, 2018) in the flows of European outbound tourism (DER Touristik group, Meliá, Iberostar, Resort Group, TUI group). Third, European countries are also the main source markets of international tourism arrivals to Mediterranean basin countries of North Africa and Asia. Investments have already been undertaken, and multinational firms have now been operating in Cape Verde for many years. Therefore, it is difficult to foresee how the possible restoration of tourism destinations affected by the AS will influence international demand for tourism to Cape Verde. Nevertheless, hotels and governments should undertake measures to create a safe environment for tourists and avoid crime or social unrest that could hurt the tourism industry.

The absence of time series data for the period of analysis constitutes the main limitation of this study, for example, CPI for package holidays, tourism demand time series by country or region of origin, and qualitative factors that were found to be relevant in explaining tourism arrivals to African destinations such as infrastructure, level of development, or health risk (Adeola et al., 2018; Naudé and Saayman, 2005; Saayman and De Klerk, 2019; Viljoen et al., 2019). It might be worthwhile for further research to overcome these limitations when data availability makes it possible.

COVID-19 brought the tourism expansion in Cape Verde to a halt. The effects of the pandemic are devastating on this medium-haul developing destination. The existence of effective vaccines is an important step to recover tourists’ confidence. However, vaccination will take time and people might be afraid of flying or needing health care in foreign countries. This makes medium-haul destinations less likely to be selected for traveling. Under these circumstances, stakeholders have no other option but to reduce this uncertainty brought by COVID-19 by reassuring travelers that both flying to and staying in Cape Verde on holidays is safe. The findings reported in our study provide information that helps to characterize the demand and can be of great assistance for different actors in the tourism sector to encourage tourists back to visit the islands.

Footnotes

Acknowledgments

The author thanks both the University of Cape for the invitation to teach in the doctoral program that was the starting point for this study and the support from the Statistical National Office of Cape Verde for the facility to access the data. He also acknowledges the sabbatical year granted by his employee, which gave him the opportunity to devote full time to research.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The study was funded by University of Capetown.

ORCID iD

Manuel González-Gómez

Notes

References

Adeola

Boso

Evans

(2018) Drivers of international tourism demand in Africa. Business Economics 53(1): 25–36.

Alvarez-Diaz

Gonzalez-Gomez

Otero Giraldez

(2015) Estimating price and income demand elasticities for Spain separately by the major source market. Tourism Economics 21(5): 1103–1110.

Awaritefe

(2007) Evaluating tourism resource areas in Nigeria for development. Tourism Review 62(2): 37–45.

Balg

Metcalf

(2010) Modeling exchange rate volatility. Review of International Economics 18(1): 109–120.

Banerjee

Dolado

Galbraith

, et al. (1993) Co-Integration, Error Correction, and the Econometric Analysis of Non-Stationary Data. Oxford: Oxford University Press.

Bankole

Babatunde

(2010) Elasticities of demand for tourism in Nigeria: an ARDL approach. Tourism Analysis 15(4): 489–496.

Bardsen

(1989) Estimation of long run coefficients in error correction models. Oxford Bulletin of Economics and Statistics 51(3): 345–350.

Belke

Polleit

(2006) Monetary policy and dividend growth in Germany: long run structural modelling versus bounds testing approach. Applied Economics 38: 1409–1423.

Bouzahzah

El Menyari

(2013) International tourism and economic growth: the case of Morocco and Tunisia. The Journal of North African Studies 18(4): 592–607.

10.

Buigut

Amendah

(2016) Effect of terrorism on demand for tourism in Kenya. Tourism Economics 22(5): 928–938.

11.

Burger

CJSC

Dohnal

Kathrada

(2001) A practitioners guide to time-series methods for tourism demand forecasting—a case study of Durban, South Africa. Tourism Management 22(4): 403–409.

12.

Campos-Soria

Inchausti-Sintes

Eugenio-Martin

(2015) Understanding tourists’ economizing strategies during the global economic crisis. Tourism Management 48: 164–173.

13.

Çetin

Ecevit

Seker

, et al. (2015) Financial development and energy consumption in Turkey: empirical evidence from cointegration and causality tests. In: Zeynep

(ed) Handbook of Research on Behavioral Finance and Investment Strategies. Hershey, PA: IGI Global.

14.

Chao

C-C

L-J

Lai

C-C

, et al. (2013) Devaluation, pass-through and foreign reserves dynamics in a tourism economy. Economic Modelling 30(1): 456–461.

15.

Chigora

Vutete

(2015) Elasticity of demand a critical form tourism market position: understanding Zimbabwe tourism destination. Case Studies in Business and Management 2(2): 18–26.

16.

Choyakh

(2008) A model of tourism demand for Tunisia: inclusion of the tourism investment variable. Tourism Economics 14(4): 819–838.

17.

Cobbinah

Black

Thwaites

(2013) Tourism planning in developing countries. Review of concepts and sustainability issues. International Journal of Social, Human Science and Engineering 7(4): 313–320.

18.

Daniel

ACM

Ramos

FFR

(2002) Modelling inbound international tourism demand to Portugal. International Journal of Tourism Research 4: 193–209.

19.

De Vita

Abbott

(2002) Are saving and investment cointegrated? An ARDL bounds testing approach. Economics Letters 77: 293–299.

20.

Durbarry

Nicolas

Seetanah

(2009) The determinants of tourism demand in South Africa using a dynamic panel data approach. Tourism Analysis 14(3): 375–385.

21.

Eja

Joseph

Ojong

(2012) Success factors determining Nigeria as a tourist destination. Journal of Emerging Trends in Educational Research and Policy Studies 3(4): 426–432.

22.

El Engle

Granger

CWJ

(1987) Cointegration and error correction: representatión, estimation and testing. Econometrica 55: 251–276.

23.

El Menyari

(2021) Effect of tourism FDI and international tourism to the economic growth in Morocco: Evidence from ARDL bound testing approach. Journal of Policy Research in Tourism, Leisure and Events 13(2): 222–242.

24.

ETC. European travel Commission (2009) European Tourism in 2009: Trends and Prospects 2009. Quarterly Report (Q1/2009) Available at: https://etc-corporate.org/uploads/reports/ETC-Insights-Q12009-prn.pdf (accessed 20 May 2020).

25.

Eugenio-Martin

Campos-Soria

(2014) Economic crisis and tourism expenditure cutback decision. Annals of Tourism Research 44: 53–73.

26.

European Commission (2017) Reflection paper on the deepening of the Economic and Monetary Union. COM(2017) 291 of 31 May 2017.

27.

Faragalla

WSS

(2018) Econometric analysis for tourism demand function in Egypt: a dynamic panel data approach. Facta Universitatis Series: Economics and Organization 14: 321–332.

28.

Fauzel

Seetanah

Sannassee

(2017) Analysing the impact of tourism foreign direct investment on economic growth: evidence from a small island developing state. Tourism Economics 23(5): 1042–1055.

29.

Fernandes

(2005) Em Busca da Nação: notas para uma reinterpretação do Cabo Verde Crioulo. Editora da UFSC, IBNL.

30.

Gasmi

Sassi

(2015) International tourism demand in Tunisia: evidence from dynamic panel data model. Economics Bulletin 35(1): 507–518.

31.

Gauci

Gerosa

Mwalwanda

(2002). Tourism in Africa and the Multilateral Trading System: Challenges and Opportunities, Background Paper for the Economic Commission for Africa.

32.

Gil Alana

(2011) Tourism in South Africa-time series persistence and the nature of the shocks: Are they transitory or permanent? African Journal of Business Management 5(12): 4686–4698.

33.

Gooroochurn

Sinclair

(2005) Economics of tourism taxation: evidence from Mauritius. Annals of Tourism Research 32(2): 478–498.

34.

Granger

CWJ

Huang

Yang

(2000) A bivariate causality between stock prices and exchange rates: evidence from recent Asian Flu. The Quarterly Review of Economics and Finance 40: 337–354.

35.

Hall

O’Sullivan

(1996) Tourism, political instability and violence. In: Pizam

Mansfeld

(eds) Tourism, Crime and International Security Issues, New York, NY: John Wiley, pp. 105–121.

36.

Hathroubi

(2011) Modelling and forecasting international tourism demand for Tunisia: A time varying parameter approach. Available from261-279 Interdisciplinary Journal of contemporary research in business 3(7): 261–279.

37.

Holden

Sonne

Novelli

(2011) Tourism and poverty reduction: An interpretation by the poor of Elmina, Ghana. Tourism Planning and Development 8(3): 317–334.

38.

Ibrahim

MAMA

(2013) The determinants of international tourism demand for Egypt: panel data evidence. European Journal of Economics, Finance and Administrative Sciences 30: 50–58.

39.

INE (2016) Evolução da população residente em Cabo Verde por Ilha Concelho (1900 -2010). Available at: http://ine.cv/quadros/evolucao-da-populacao-residente-em-cabo-verde-1900-2010/. (accessed October 2018).

40.

INE (2017) Projecções demográficas CV por sexo e faixa etária, 2010 a 2030. Available at: http://ine.cv/quadros/projeccoes-demograficas-cv-sexo-faixa-etarias-simples-2010-2030/ (accessed October 2018).

41.

INE (2020) Estatísticas do Turismo-Movimentação de Hóspedes. 1º Trimestre 2020. Available at: http://ine.cv/publicacoes/estatisticas-do-turismo-movimentacao-hospedes-1o-trimestre-2020/ (accessed January 2020).

42.

Johansen

Juselius

(1990) Maximun likelihood estimation and inference on cointegration with applications to the demand for money. Oxford Bulletin of Economics and Statistics 52: 169–210.

43.

Khandaroo

Seetanah

(2008) The role of transport infrastructure in international tourism development: a gravity model approach. Tourism Management 29(5): 831–840.

44.

Kourentzes

Saayman

Jean-Pierre

, et al. (2021) Visitors arrivlas forecasts amid COVID-19: a perspective from the Africa team. Annals of Tourism Research 88: 103197.

45.

Lepp

Gibson

(2003) Tourist roles, perceived risk and international tourism. Annals of Tourism Research 30: 606–624.

46.

Song

Witt

(2004) Modelling tourism demand: a dynamic linear AIDS approach. Journal of Travel Research 43(2): 141–150.

47.

Moreno

Seetanah

Sannassee

, et al. (2020) An analysis of the sources of growth in Mauritius. In: Seetanah

Sannassee

Nunkoo

(eds) Mauritius: A Successful Small Island Developing State. Abingdon: Routledge, pp. 14–39.

48.

Moyo

Ziramba

(2013) The impact of crime on inbound tourism to South Africa: an application of the bounds test. African Security Review 22(1): 4–18.

49.

Mpofu

TPZ

(2009) An analysis of factors that caused a downturn in tourism demand trends and patterns in Zimbabwe from 2000 to 2005. Journal of African Development Studies 2(1): 35–49.

50.

Muchapondwa

(2009) Supply response of Zimbabwean agriculture: 1970–1999. African Journal of Agricultural and Resource Economics (AfJARE) 3(1): 28–48.

51.

Muchapondwa

Pimhidzai

(2011) Modelling international tourism demand for Zimbabwe. International Journal of Business and Social Science 2(2): 71–81.

52.

Naftziger

(2006) Economic Development, 4th ed. New York, NY: Cambridge University Press.

53.

Narayan

Smyth

(2005) The residential demand for electricity in Australia: an application of bounds test approach to cointegration. Journal of Energy Policy 33: 267–474.

54.

Naudé

Saayman

(2005) Determinants of tourists arrivals in Africa: a panel data regression analysis. Tourism Economics 11(3): 365–391.

55.

Nunkoo

Seetanah

Jaffur

ZRK

, et al. (2019) Tourism and economic growth: a meta-regression analysis. Journal of Travel Research 59(3): 404–423.

56.

OAG (2020) Coronavirus. How and when will aviation recovers from COVID-19? Available at: https://www.oag.com/coronavirus-airline-sheducles-data (accessed 10 January 2021).

57.

ONS-Office for National Statistics (2017) Holidays in the 1990s and now. Available at: https://www.ons.gov.uk/peoplepopulationandcommunity/leisureandtourism/articles/holidaysinthe1990sandnow/2017-08-07. (accessed 20 April 2020).

58.

Ouerfelli

(2008) Co-integration analysis of quarterly European tourism demand in Tunisia. Tourism Management 29: 127–137.

59.

Ouerfelli

(2010) Analysis of European tourism demand for Tunisia: a new approach. International Journal of Tourism Policy 3(3): 223–236.

60.

Papatheodorou

Rosselló

Xiao

(2010) Global economic crisis and tourism: consequences and perspectives. Journal of Travel Research 49(1): 39–45.

61.

Pesaran

Shin

Smith

(2001) Bounds testing approaches to the analysis of level relationships. Journal of Applied Econometrics 16: 289–326.

62.

Pesaran

Shin

(1999) An autoregressive distributed lag modelling approach to cointegration analysis. In: Strom

Diamond

(eds) Econometrics and Economic Theory in the 20th Century: The Ragnar Frisch Centennial Symposium, Cambridge: Cambridge University Press.

63.

Pizam

Fleischer

(2002) Severity versus frequency of acts of terrorism: Which has a larger impact on tourism demand? Journal of Travel Research 40(3): 337–339.

64.

Rey

Myro

Galera

(2011) Effect of low-cost airlines on tourism in Spain. A dynamic panel data model. Journal of Air Transport Management 17(3): 163–167.

65.

Saayman

De Klerk

(2019) Forecasting tourists arrivals using multivariate singular spectrum analysis. Tourism Economics 25(3): 330–354.

66.

Sassi

(2014) Combination forecasts of international demand for tourism in Tunisia. Journal of quantitative economics: journal of the Indian Econometric Society 12(2): 94–109.

67.

Saayman

(2013) Exchange rate volatility and tourism—revisting the nature of the relationship. European Journal of Tourism Research 6(2): 104–121.

68.

Saayman

(2015) An ARDL bounds testing approach to modelling tourist expenditure in South Africa. Tourism Economics 21(1): 49–66.

69.

Seetanah

(2011) Assessing the dynamic economic impact of tourism for island economies. Annals of Tourism Research 38(1): 291–308.

70.

Seetanah

Durbarry

Ragodoo

JFN

(2010) Using the panel cointegration approach to analyse the determinants of tourism demand in South Africa. Tourism Economics 16(3): 715–729.

71.

Seetanah

Sannassee

Sawkut

(2015) The impact of relative prices on tourism demand for Mauritius: an empirical analysis. Development Southern Africa 32(3): 363–376.

72.

Song

Witt

(2000) Tourism Demand Modelling and Forecasting: Modern Econometric Approaches. Oxford: Pergamon Oxford.

73.

Song

Witt

, et al. (2010) Tourism demand modelling and forecasting: How should demand be measured? Tourism Economics 16(1): 63–81.

74.

Song

Wong

KKF

Chon

KKS

(2003) Modelling and forecasting the demand for Hong Kong tourism. International Journal of Hopitality Management 22: 435–451.

75.

Soubbotina

(2004) Beyond Economic Growth: An Introduction to Sustainable Development (English). WBI learning resources series. Washington DC: World Bank. Available at: http://documents.worldbank.org/curated/en/454041468780615049/Beyond-economic-growth-an-introduction-to-sustainable-development. (accessed 20 Febrauary 2019).

76.

Thompson

(2011) Terrorism and tourism in developed versus developing countries. Tourism Economics 17(3): 693–700.

77.

UNCTAD United Nations Conference on Trade and Development (2018) Investment Policy Review. Cabo Verde Available at: https://unctad.org/en/PublicationsLibrary/diaepcb2018d2_en.pdf (2018). (accessed 08 January 2021).

78.

UNWTO (2020a) International tourism and COVID-19. International tourist arrivals and International tourism receipts by country. Available at: https://www.unwto.org/international-tourism-and-covid-19. (accessed 08 January 2021).

79.

UNWTO (2020b) How are countries supporting tourism recovery? UNWTO Briefing note-Tourism and COVID-19. Issue 2. June 2020 .

80.

Viljoen

Saayman

(2019) Determinants influencing inbound arrivlas to Africa. Tourism Economics 25(6): 856–883.

81.

Wahba

Elanani

(2017) The exchange rate fluctuations and its effect on international tourism demand case study: Egypt the period (2000–2016). Journal of Tourism Research 17(1): 128–139.

82.

WEF (2011) The Travel & Tourism Competitiveness Report 2011. Cologny: World Economic Forum. Available at: http://www3.weforum.org/docs/TTCR11/ (accessed 08 March 2019).

83.

WEF (2017) The Travel & Tourism Competitiveness Report 2017. Cologny: World Economic Forum. Available at: https://www.weforum.org/reports/the-travel-tourism-competitiveness-report-2017 (accessed 08 March 2019).

84.

World Bank (2018) Available at: https://datahelpdesk.worldbank.org/knowledgebase/articles/906519-world-bank-country-and-lending-groups (accessed 08 March 2019).

85.

WTTC—World Trade Tourism Council (2017) World Trade Tourism Council. Travel Tourism Economic Impact 2017 Cape Verde. Available at: https://www.wttc.org/-/media/files/reports/economic-impact-research/countries-2017/capeverde2017.pdf (accessed 08 March 2019).