Tourism and COVID-19: An economy-wide assessment

Abstract

The promotion of domestic tourism is widely identified as a promising recovery measure to revive tourism economies severely affected by the COVID-19 pandemic. The paper intends to estimate the economy-wide impacts of a hypothetical scenario of domestic tourism promotion in a small, emerging economy with heavy reliance on international tourism. The estimated impacts are compared with the economy-wide negative impacts of the pandemic via domestic and international tourism, to understand the potential level of compensation of domestic tourism promotion. The industry-level analysis indicates Dutch disease effects due to the decline in international tourism. A hypothetical boom in domestic tourism benefits non-tradable sectors to a significant level. Aggregate level analysis indicates a potential for economic gains through domestic tourism promotion, though such gains are inadequate to fully offset the foregone economic benefits due to loss in international tourism.

1. Introduction

Domestic tourism seemed to be a promising COVID-19 recovery strategy in developed economies with strong domestic tourism markets. However, for developing economies this remained an unanswered policy question at the time of the pandemic. The role of domestic tourism in crisis management in tourism is vital in enhancing adaptive capacities and resilience for future crises including climate change. The economic assessments of domestic tourism in developing countries have mainly focused on the immediate direct economic impacts only. However, a shock in tourism can create notable indirect and induced impacts due to tourism's significant linkages with the rest of the economy. Due to this reason, estimates that capture only the direct impacts of tourism crises are of limited usefulness in informing effective policy interventions.

The purpose of this paper is to assess and understand the system-wide effects of domestic tourism promotion as a crisis recovery strategy in a developing country context. The paper makes an important contribution to literature in several ways. Firstly, the paper investigates the potential of domestic tourism in easing the economic hardships due to reduced international tourism following the COVID-19 pandemic based on a developing country context. Due to the improved availability of data and models, developed economies such as Australia (Dwyer et al., 2006; Dwyer et al., 2003; Peter Forsyth, Dwyer, & Spurr, 2014; P. Forsyth, Dwyer, & Spurr, 2014; Meng & Pham, 2017), Spain (Liu & Wu, 2019), China (Li et al., 2010) and the UK (Blake and Sinclair; 2003) have been the focus of the research that aimed to understand the economy-wide effects of domestic tourism in crisis periods. These research studies conducted on larger and more developed economies have revealed that the economic impact of domestic tourism varies significantly based on the economic structure of the country. This paper seeks to contribute to this field of literature by providing empirical evidence from an emerging economy.

Secondly, the paper contributes to the resilience theory in tourism by exploring the potential of domestic tourism in improving the economic capacity to adapt to crises. As Cochrane (2010) identifies, resilience theory in tourism has evolved to “explain the cyclical and complex nature of systems, based on recovery from perturbation and the accumulation of various forms of capital which allow faster renewal and stronger structures”. While the analysis focuses on the COVID-19 pandemic, the lessons drawn from this analysis may be useful in making vulnerable tourism systems more resilient to future shocks. One strategy to achieve this goal is to promote industry diversification by emphasizing different tourism segments, such as domestic tourism. Thirdly, the paper presents a framework to analyze the Dutch disease effects for the changes in international and domestic tourism segments separately. More specifically the assessment captures the impacts on manufacturing and non-tradable industries and the potential implications on resilience.

Though the paper involves a single-country assessment, the study approach is highly generalizable to other economies with necessary modifications in models and shock values. The assessment is based on a case study of Sri Lanka, an emerging economy with a heavy reliance on international tourism. The impacts of a crisis in tourism in such countries are devastating due to the heavy dependence of such economies on tourism. In most cases, emerging economies lack economic diversification avenues when facing shocks like the COVID-19 pandemic. Also, there are significant constraints for recovery due to economic fragilities in such countries. The costs of reduced economic activity due to the COVID-19 pandemic are particularly significant in developing countries because of their low healthcare capacity, the presence of large informal sectors, their less developed financial market, less fiscal space, and poor governance (Loayza & Pennings, 2020). The study provides important policy implications which are useful in enhancing resilience in other developing tourism economies, including small island developing states.

On the empirical front, the paper makes an interesting contribution to the limited literature on the economy-wide impacts of promoting domestic tourism in responding to a crisis. To the best of our knowledge, limited research has focused on the system-wide impacts of crises on emerging small economies, which are substantially dependent on international tourism. The study approach can be replicated in other tourism destinations with similar economic structures. A relevant policy research question following the COVID-19 pandemic evolves around domestic tourism's role in facilitating recovery and resilience (Gössling et al., 2020). In this paper, we intend to focus on the following two research questions (RQ) concerning domestic tourism.

• RQ1What are the economy-wide impacts of tourism loss due to COVID-19 pandemic?
• RQ2What is the magnitude of potential benefits of domestic tourism promotion following COVID-19 pandemic?

The following section reviews the domestic tourism and crises literature. Methodological details are discussed in section 3. Section 4 presents the analysis results and section 5 includes conclusions and proposed policy directions.

2. Domestic tourism and tourism crises

Comprehensive system-wide assessments of the potential role of domestic tourism over crisis periods are rare in literature. Previous assessments indicate the contribution of domestic tourism in facilitating resilience in tourism economies based on the direct economic benefits. For instance, Canh and Thanh (2020), based on an analysis of 79 countries, find that domestic tourism spending (as a percentage of gross domestic product (GDP)) facilitated economic resilience, while international tourism may increase the vulnerability of economies. This impact is consistent in low- and lower-middle-income countries, while it is non-significant in upper-middle- and high-income countries. Regarding the global financial crisis, Sheldon and Dwyer (2010) highlight the potential role of domestic tourism to provide a basis for recovery. The promotion of domestic tourism following the global financial crisis helped Malaysia to stabilize its economy (de Sausmarez, 2007).

Based on a systematic literature review, Wickramasinghe and Naranpanawa (2021) suggest the use of general equilibrium models to simulate policy interventions aiming at enhancing resilience in tourism. These models can capture the economy-wide impacts which include both direct and indirect economic consequences. When assessing the economy-wide impacts of crises on tourism, studies have focused on the differential impacts of crises on international and domestic tourism. Blake and Sinclair (2003) found that after the September 11th terrorist attack, the overall economic cost of this reduction in domestic tourism was greater than the reduction in demand by international tourists. Several other studies differentiated Australian domestic and inbound tourism impacts under various policy scenarios (Dwyer et al., 2006; Dwyer et al., 2003; Peter Forsyth, Dwyer, & Spurr, 2014; P. Forsyth, Dwyer, & Spurr, 2014; Meng & Pham, 2017). Similar studies were based on other countries such as Spain (Liu & Wu, 2019), China (Li et al., 2010), and Indonesia (Mahadevan et al., 2017). These studies have not considered the potential impacts of domestic tourism promotion on improving resilience in respective tourism destinations. Further, the studies that focus on the previous crises have limited usefulness in informing COVID-19-related policy due to its global impacts in comparison with previous health crises.

Existing system-wide assessments with regard to the COVID-19 pandemic revolve around its impacts (Pham et al., 2021), and the impacts of prevention and control strategies of the pandemic (Li et al., 2022; Wang et al., 2021). Notably, literature is yet to pay enough attention to tourism-related recovery strategies, including domestic tourism promotion that can enhance the adaptive capacity of tourism economies. Even during normal periods, well-developed domestic tourism markets are important in enhancing economic benefits in tourism economies. Particularly, domestic tourism plays an important role in reducing leakages from the economy and generating income and employment opportunities (Sheldon & Dwyer, 2010). Domestic tourism can help in addressing the seasonal effects of international tourism (Nyaupane et al., 2020).

The role of domestic tourism in promoting recovery and long-term resilience presents an important research gap in post-pandemic recovery (Gössling et al., 2020). However, a switch from inbound tourism to domestic tourism may be based on country-based issues, which need further investigation (Yang & Chen, 2009). Based on the evidence from the Caribbean and the Maldives, Mekharat and Traore (2020) highlight that the impact of the COVID-19 pandemic on tourism will be influenced by the degree of dependence the destination has on international tourism.

The pandemic has led to increased competition in domestic tourism markets, as the demand for international tourism demand is dramatically reduced following the pandemic (Duro et al., 2022). There is likely to be an uneven recovery in domestic tourism markets in response to promotional campaigns to encourage domestic tourism following the pandemic (Hall et al., 2020). The potential role of domestic tourism as a crisis recovery strategy is yet to be well established in terms of its economy-wide implications.

3. Method

To understand the potential role of domestic tourism in compensating for the economic losses due to the COVID-19 pandemic, we undertake two simulation experiments using a tourism CGE model. Simulation experiments helps in assessing and understanding “what if” questions regarding two scenarios. The first simulation experiment shows how the economy will look if it undergoes a reduction in tourism demand in international and domestic segments. The second experiment indicates how a policy intervention to promote domestic tourism would make changes in the economy. The simulation experiment allows us to measure how the tourism shocks influence the economy through their system-wide impacts. These experiments do not consider the other changes that take place in the economy due a shock event.

In this study we consider domestic tourists as residents who travel within the country. Foreign tourists imply foreign nationals who are visiting from abroad. The following sections present the details of model development, database, and simulation experiments.

3.1. Model development

We developed a single country tourism CGE model for Sri Lanka, to estimate the economic cost of COVID-19 and the role of domestic tourism to potentially ease these costs. The CGE model is based on the ORANI-G model, a comparative static model developed by Horridge (2000). The theoretical structure of the ORANI-G model comprises equations that describe demand and supply equations for producers, demand equations regarding capital formation, households, exports, and government. Further the equations also describe the relationship between basic values to production costs and to purchasers' prices. Another set of equations describe the market-clearing conditions for commodities and primary factors (Horridge, 2000). The model is based on neoclassical assumptions. These include.

• ⁃A perfectly competitive economy with constant returns to scale,
• ⁃That industries minimize costs,
• ⁃That households maximize utility,
• ⁃Labour is perfectly mobile across industries while capital is treated as industry-specific,
• ⁃There is continuous market clearing, and
• ⁃Zero profits are generated for all industries (Meng & Siriwardana, 2016).

The standard ORANI-G model does not include either domestic or international tourism as intermediate industries. The structure of the standard CGE model includes five major demand sectors namely production, investment, household, government, and exports. Based on the input-output data from Sri Lanka, we performed two modifications to the standard ORANI-G model to incorporate the domestic and international tourism sectors as intermediate industries. The modifications in the standard ORANI-G model to include tourism components were based on two characteristics of tourism. Firstly, tourism is considered a multifaceted sector that offers a composite product. Secondly, tourism sells composite products to consumers by purchasing the components of the product from an array of other sectors (Ihalanayake, 2012).

The first modification involves the extraction of international tourism component from exports and including it as a separate industry in the intermediate matrix. The export column then refers to exports excluding tourism. In the second modification, household consumption of domestic tourism is segregated and included in the intermediate matrix as a domestic tourism component. Therefore, the new household consumption column excludes the tourism component. Both domestic and international tourism components in the model purchase commodities from other industries and sells the composite product to the final demand sectors. There was no purchasing of primary inputs in this case (Ihalanayake, 2012). Thus, the tourism CGE model developed for this study includes two separate industries to represent international and domestic tourism segments. This allows us to simulate different shock magnitudes for the two tourism segments and understand the potential impacts of domestic tourism in the absence in international tourism.

3.2. Database

We use the official input-output table (IOT) developed by the Sri Lankan Department of Census and Statistics (DCS) as the foundation database for simulations. We consider 2018 as the baseline year, as it was the most recent year with no significant internal or external shocks affecting both domestic and international tourism. The IOT is symmetric in nature with 127 industries and 127 commodities. In the conventional IOTs, domestic and international tourism expenditures were embedded in the household final consumption vector and export vector (Dwyer et al., 2016; Pham et al., 2015).

Our first step involved deriving a cost structure for domestic tourism. Domestic tourism expenditure was estimated based on data from the Sri Lanka Tourism Expenditure Survey 2017/2018. The survey covers nearly 25,750 households representing all districts in Sri Lanka. The dataset is used to calculate the proportions of different tourism expenditure categories. In the absence of post-COVID estimates, it is assumed that these proportions remain the same in the short run so that it will not impact the results substantially. The cost structure for international tourism was developed using the data from the Survey of Departing Foreign Tourists from Sri Lanka 2018. The cost structures used in deriving new industries to represent international and domestic tourism components are given in Table 1 .

Table 1.

Percentage expenditure by cost category.

Cost item	Domestic tourism	International tourism
Accommodation, food, and beverages	24	35
Transport	33	20
Wholesale and retail trade	43	45

3.3. Experiment design

We undertook two simulation experiments to estimate the economy-wide impacts of the pandemic via tourism and to understand the potential impacts of a hypothetical scenario of domestic tourism promotion. The shocks for the simulation experiments were calculated as annual percentage changes from the base year. The experiments use two shocks to represent international and domestic tourism and the two shocks are now described in detail.

3.3.1. Estimating the economy-wide impact of COVID-19 via tourism

The number of positive cases of COVID-19 with a Sri Lankan origin was reported in March 2020 and a subsequent increase in the number of positive cases was observed in the following months (Fig. 1 ). The Government of Sri Lanka imposed travel bans for international tourists on the March 18, 2020 and a country-wide curfew and area-based curfew to contain the spread of the pandemic. International tourist arrivals have been non-existent since then. We considered tourist arrival data from 2018 to 2020 to calculate the shock value for international tourism. Since Sri Lanka went through a significant decline in tourist arrivals in 2019 due to the Easter Sunday attack, the calculation did not involve data from 2019. International tourist arrivals decreased from 2,333,796 in 2018 to 507,704 in 2020 (SLTDA, 2018), recording a 78 percent decline.

Fig. 1.

International tourist arrivals by month.

Given a lack of data on the number of domestic tourists, we used local tourist guest night data to represent the decline in domestic tourism due to the pandemic (Fig. 2 ). Domestic tourism declined drastically during the pandemic-related curfew periods in April and May of 2020. From June to October 2020 the restrictions on domestic tourism were eased, and there was a slight increase in domestic tourism. The number of local guest nights has reduced from 1,984,891 in 2018 to 876,426 in 2020 (SLTDA, 2018). This represents a 56 percent decline in domestic tourism due to the pandemic. We incorporate percentage changes in local guest night as the shock value for domestic tourism in our simulation experiments. Therefore, this does not negatively impact the accuracy of the results.

Fig. 2.

Local guest nights by month.

Over the pandemic period, the nature of travel restrictions within the country were same for domestic and international tourists. However, international tourists had specific restrictions when entering the country while the borders were open. When the borders remained closed for international tourists, there was some level of domestic tourism within the country.

3.3.2. Estimating the economy-wide impact of domestic tourism promotion

The second experiment simulates a hypothetical scenario of domestic tourism promotion, while international tourism remains at the pandemic level. In this scenario, we assume that the outbound tourism expenditure of Sri Lankans is translated into domestic travel expenditure, due to restrictions on international travel. Further, we assume that the government intervention intends to provide incentives for domestic tourism and to support tourism businesses to improve their productivity. Further, existing evidence shows that many destination marketing companies in Sri Lanka have been turning their attention toward the domestic traveller segment through the promotion of excursions over the pandemic period (Arachchi & Gnanapala, 2020). Further, Sivesan (2020) suggests there is a need in Sri Lanka to promote domestic tourism, in addition to international tourism, to facilitate long-term recovery following the pandemic.

A recent survey conducted by the Department of Census and Statistics of Sri Lanka that outbound tourism consists of 51 percent of total tourism expenditure (total tourism expenditure includes both outbound and domestic tourism expenditure) (Department of Census and Statistics & Ministry of Finance, 2020). When this portion of tourism expenditure is translated to domestic tourism because of changes in travel behaviour and government support, we can potentially see a 100 percent increase in domestic tourism expenditure. The CGE models provide linear estimates which are proportionate to shock values. In this analysis, our results are corresponding to a 100 percent increase in domestic tourism. The impacts can easily be computed for smaller shock values based on these results. Table 2 summarises the shock levels for each simulation experiment. The shock level indicates the positive and negative signs to show the direction of the shocks.

Table 2.

Shocks under each simulation experiment.^a.

Shock scenarios	Experiment 1b Impact of the pandemic	Experiment 2c Impact of domestic tourism promotion
International tourism	−78%	−78%
Domestic tourism	−56%	+100%

^aThe values presented in the table denotes percentage changes compared to the base year 2018.

^bSimulation experiment 1 aims to assess the total economic cost of COVID-19 via tourism.

^cSimulation experiment 2 assessed the impact of the policy intervention to actively promote domestic tourism.

The simulation experiments involve a short-run macroeconomic environment (or model closure) as indicated in Fig. 3 . The short-run macroeconomic environment here implies that there is at least one input fixed over the period. It may not be accurate to assume a long-run scenario as the period under consideration is too short for the quantities of all inputs in the economy to vary. In the short-run closure, all sectoral capital was exogenized and a slack labour market was assumed, leaving total employment as endogenized. Sectoral rates of return were also considered endogenous. The major shocks to the economy are taking place due to the decrease in international tourism expenditure and domestic tourism expenditure. From the demand side of GDP, we keep non-tourism private expenditure, government expenditure, and non-tourism net exports endogenous. The following discussion presents the results of the simulation experiments.

Fig. 3.

Macroeconomic environment (closure) in the short run.

4. Simulation experiment results

Tourism shocks create direct, indirect, and induced impacts across the economy through tourism's backward linkages with other industries. The economy-wide impacts, as revealed by our simulation experiments capture the total effects. The impacts can be explained at the macroeconomic and industrial levels.

4.1. Macroeconomic impacts

The economy-wide assessment indicates that the pandemic-related tourism shocks create a 1.3 percent reduction in the country's real gross domestic product (GDP). With the promotion of domestic tourism, the decline in real GDP is reduced to 1.2 percent (Table 3 ). Though domestic tourism promotion does not fully compensate for the forgone benefits due to the reduction in international tourism, it has helped in reducing the negative impacts. The results support the notion presented by Mekharat and Traore (2020), that domestic tourism may help initiate the recovery, though it cannot fill the entire gap caused due to reduction in international tourism.

Table 3.

Macroeconomic impacts.

Macroeconomic indicator	Experiment 1: Impact of COVID-19	Experiment 2: Impact of domestic tourism promotion
Aggregate employment	−4.406	−4.177
Real GDP	−1.316	−1.247
Real GNE	−3.114	−3.048
Imports	−11.131	−11.169
Real household consumption	−4.375	−4.282
Exports	−4.348	−4.420
Real government demands	−4.375	−4.282

Note: Values indicate percentage changes from base scenario.

Our simulation experiment confirmed that the pandemic had a substantial impact on the overall unemployment rate of the country, as the aggregate employment loss is estimated to be 4.4 percent compared to the base case. Further, the fall in aggregate employment was much higher than the fall in real GDP in the two experiments. This was because the affected industries shed labour to reduce the cost of labour during the shock period. As a result, aggregate private consumption showed a 4.4 percent loss compared to the base case. The decline in aggregate employment is 4.2 percent compared to the base case in Experiment 2, due to the promotion of domestic tourism. It shows that domestic tourism has helped in improving the employment level to a certain extent.

To compare the results of this general equilibrium analysis, we also undertake a partial equilibrium analysis using time-series data. Here the assessment is based on a regression analysis between quarterly GDP as the dependent variable and quarterly local guest nights as the independent variable. The regression analysis indicates a statistically significant coefficient of 0.592 (at 1% level) for local guest nights with an adjusted R-squared value of 0.29. These results also show the same direction of impacts indicated by the CGE analysis. Other models which assess the long-run association between variables such as cointegration analysis will not fit for the purpose, as our CGE model is based on a short-run scenario.

The magnitude of the negative shock via other key macroeconomic variables was greater than that of real GDP and real GNE. The greatest impact of COVID-19-related shocks was reflected in imports (−11.13%). Imports tended to further decrease under domestic tourism promotion. Recovery in local economic activities and the substitution of imports for locally produced goods and the changes in foreign currency rates could explain this change. The reduction in exports (−4.3%) can mainly be attributed to the loss in international tourism revenue. Since international tourism is not recovering within Experiment 2, there was no notable recovery in exports there.

Overall, we noted that the promotion of domestic tourism in the absence of international tourism is capable of partially offsetting the foregone economic impacts of the pandemic within the Sri Lankan economy. The findings seem to be reasonable, given the relatively low value of tourism expenditure by domestic tourists compared to that of international tourists in Sri Lanka. As expected, it is therefore clear from our simulation experiments that, in the short run, the promotion of domestic tourism provides some economic stimulus for recovery.

Due to methodological differences, the results of Arbulú et al. (2021) which finds the capacity of domestic tourism to relieve reduction in international tourism demand, are not directly comparable with our results. Our results reflect the general equilibrium effects which incorporates both direct and indirect economic effects. The assessment undertaken by Arbulú et al. (2021) is based on a partial equilibrium approach. However, the study also empirically supports the possibility of considering domestic tourism as a helpful strategy to survive the tourism sector during the COVID-19 crisis. The study finds that the reorientation of outbound tourism to the domestic travel market can generate 10 to 70 percent of pre-crisis overnight stays, depending on specific regions in Spain.

4.2. Industrial impacts

The industry-level impacts of the pandemic vary significantly in terms of the magnitude and direction across 127 industries in our simulation experiments. The decline in international tourism and domestic tourism in Experiment 1 has benefited manufacturing industries (Table 4 ). This indicates the of the Dutch disease impact on tourism. Manufacturing industries and industries related to agriculture (such as tobacco, prepared, and preserved vegetables) benefit when tourism activity declines in the economy. The benefits primarily take place in the form of labour movements to these industries from negatively impacted industries. As the wage rates in our model closure are held exogenous, we do not see changes in real wage rates in our simulation.

Table 4.

Top 10 positively impacted industries through tourism shocks.

Industry	Percentage change in output from the base case
Unmanufactured tobacco	10.60
Forage products, fibres, living plants, cut flowers, and flower buds	10.26
Leather and leather products	9.50
Other professional, technical, and business	7.55
Medical appliances, precision and optical instruments, watches, and clocks	7.52
Prepared and preserved vegetables, pulses, and potatoes	7.39
Manufacture of refractory products	7.00
Growing of Rubber	6.64
Manufacture of transport equipment	6.48
Manufacture of pharmaceuticals, medicinal chemical and botanical products	6.33

Note: Values indicate percentage changes from base scenario.

As explained by Copeland (1991), the “Dutch disease” effect in tourism differs from other commodity exports, as tourism consumption involves non-tradable goods consumption within the economy. Thus, a tourism boom can have a direct impact on non-tradable goods. Inchausti-Sintes (2015) and Romao et al. (2016) discuss the different impacts of Dutch disease effect in tourism in various contexts. The evidence from Portugal suggests that the tourism boom following the global financial crisis in 2007 has caused a deindustrialization process like the Dutch disease effect, with positive impacts concentrated on non-tradable sectors and the construction sector (Romao et al., 2016). Further, several other studies focus on the impact of the mining boom on tourism (Dwyer et al., 2016; Moyle et al., 2021; Pham et al., 2015). The literature indicates that the specific process of Dutch disease impacts on tourism varies in different contexts. The net welfare impacts of a tourism boom in the presence of Dutch disease effects could either be positive (Zhang & Yang, 2019) or negative (Nowak & Sahli, 2007). This paper indicates positive effects in the manufacturing sector due to a decline in tourism activity. As we focus on decline in international tourism in our first simulation experiment, the direction of change is different of what we see in other contexts (as explained above) during tourism booms.

The decrease in tourism expenditure in the economy has created negative impacts on a set of industries that are closely connected to tourism. The negatively impacted industries represent non-tradable industries (Table 5 ). The accommodation and the food and beverage industry experiences a 20 percent decline in output due to the reduction in international tourism.

Table 5.

Percentage of industry-level recovery through domestic tourism promotion.

Industry	Change in output due to decline in international tourism	Change in output due to promotion in domestic tourism	Percentage recovery by domestic tourism promotion
Accommodation and food and beverage services	−20.433	2.837	13.88
Building of ships and floating structures	−6.573	0.117	1.78
Natural water	−5.781	0.574	9.93
Sewage and waste collection, treatment and disposal and other environmental protection services	−5.685	0.589	10.36
Education	−5.672	0.025	0.43
Public administration and other services provided to community as a whole	−4.568	0.132	2.89
Creative, arts and entertainment activities, etc.	−3.499	0.035	1.00
Fish and other fishing products	−2.822	0.092	3.27
Manufacture of bakery products	−1.948	0.004	0.23
Meat and meat products	−1.368	0.170	12.39
Electric power generation, transmission, and distribution	−1.448	0.069	4.80
Manufacture of soft drinks; production of mineral waters and other bottled waters	−1.334	0.018	1.37
Construction	−1.156	0.052	4.51
Egg Production	−1.064	0.032	2.96
Live animals and other animal products	−0.506	0.148	29.21
Warehousing and support activities for transportation	−0.484	0.009	1.83
Wholesale and retail trade and repair of motor vehicles and motorcycles	−0.461	0.014	3.03
Manufacture of articles of concrete, cement, and plaster	−0.304	0.034	11.05
Distilling, rectifying, and blending of spirits & manufacture of wines	−0.062	0.016	−25.06

Note: Values indicate percentage changes from base scenario.

Accommodation and the food and beverage industry cover short-term accommodation activities, camping grounds, recreational vehicles, parks and trailer parks, other accommodation, food, and beverage service activities, event catering, and other food and beverage activities. Blake et al. (2003) finds the hotels, catering and pubs industry in the United Kingdom showed a significant negative impact following the Foot and Mouth Disease outbreak. Many of these services provide goods and services to both international and domestic tourism. Based on data from the US, Abdelsalam et al. (2021) find that certain sub-sectors of tourism such as restaurants are heavily influenced by the pandemic, and this depends on the connectedness of these industries and resultant financial stress faced by the sub-sectors following the pandemic. Industries with backward linkages to the affected sectors (water, sewage, recreation, fish, other services) also show negative impacts, compared to the base scenario.

Our analysis then focused on how the active promotion of domestic tourism may help to create stimulus at the industry level. The third column in Table 5 illustrates the percentage gain in output due to domestic tourism promotion in the industries which are negatively affected by the decline in international tourism. The highest magnitude of positive impact through domestic tourism can be seen in the accommodation and food and beverage services industry. When the gain through domestic tourism promotion is taken as a percentage of loss due to international tourism (Column 4 in Table 5), the gains are highest among industries that supply goods and services to tourism-related industries. These industries include live animals and other animal products; distilling, rectifying, and blending of spirits & manufacture of wines; meat and meat products, etc.

Our findings are in line with Mahadevan et al. (2017) that finds that the promotion of domestic tourism in Indonesia led to greater demand for restaurant food and services than international tourism. This was due to the comparatively higher proportion of domestic tourism expenditure on restaurants and food services in Indonesia. However, the contribution of domestic tourism does not fully compensate for the loss incurred by these industries when international tourism is severely affected. This may be because a significant proportion of domestic tourists stay in informal accommodations, which is not captured by official tourism statistics.

Apart from the industries which are directly linked with tourism, several other non-tradable sectors such as sewage and waste collection, natural water, public administration, electricity, and construction show positive impacts due to domestic tourism promotion. Increased domestic tourism has led to the consumption of these non-tradable goods and services.

5. Conclusion and policy implications

The lessons learned from the pandemic experience can guide how to effectively deal with future crises (Sigala, 2021). The paper examined if domestic tourism promotion could play a role in compensating for the negative economic impacts of the loss of international tourism due to the COVID-19 pandemic. While the paper makes an contribution in assessing the role of domestic tourism in crisis management, it also provides important empirical evidence regarding an emerging economy context. The simulation experiments indicate potentially positive signs of promoting domestic tourism regarding a shock such as COVID-19. However, due to heavy reliance on international tourism such economy-wide benefits are not sufficient to offset the negative economic impacts of reduced international tourism.

The second major contribution of the paper lies in its discussion on the Dutch disease effect. We illustrate that decline in international tourism has led to gains in manufacturing sectors and losses in non-tradable sectors. Our findings indicate a Dutch disease mechanism. Further, domestic tourism promotion has caused beneficial impacts on some of the negatively affected non-tradable sectors due to loss in international tourism.

This paper focused on Sri Lanka, a tourism destination that has substantial dependence on international tourism. As a result, most of the developments, investments, statistical systems, and marketing and promotion activities have been oriented towards international tourism. Further, being an emerging economy, the demand for domestic tourism and leisure is low in Sri Lanka, compared to developed economies. This may be the case for other small economies with a high dependence on international tourism. Our results point out that the promotion of domestic tourism is likely to create a positive impact on the recovery path in the short run. Future developments in the domestic tourism segment may potentially help in further reducing the negative impacts of tourism shocks. As indicated in Section 1, empirical evidence produced through the paper may have helpful implications concerning the ongoing economic crisis in Sri Lanka.

Small economies with a heavy reliance on international tourism could develop domestic tourism segments, with the view of retaining economic leakages and enhancing the substitution impact between domestic tourism versus international tourism. These developments would call for the need for research-based inputs and statistical systems. As a future research direction, we suggest more research on the economic drivers and benefits of domestic tourism, especially in underdeveloped and emerging economies. Further, the development of statistical systems, preferably tourism satellite accounts may guide the formulation of sound policies. However, developing domestic tourism to retain expenditure within the economy remains a strategy for the medium-to long-term. Volgger et al. (2021) find that retentive advertising, price discounts, and flexible cancellation policies increase the booking intentions of domestic tourists during the COVID-19 pandemic based on a randomized experiment carried out in Australia. However, not much research has been conducted on the potential strategies for promoting domestic tourism during crisis periods in small developing and emerging economies. This remains an important area for future research.

A long-run scenario was not considered due to absence of required estimates to design robust simulation experiments. This may be a fertile research area. Further, the simulations reveal that the economic impact of domestic tourism is mostly seen in employment impact, as tourism is a labour-intensive industry. Further research may shed light on the specific labour market implications of promoting domestic tourism.

The role of domestic tourism may depend on the specific characteristics of a tourism destination. Empirical research is required to understand the ability of domestic tourism to compensate for the losses in international tourism in other contexts. Moreover, Dutch disease warrants more empirical evidence from developing tourism economies. Further, future research can also focus on assessing the effectiveness of various government interventions relevant to domestic tourism. Further, the learnings from the pandemic may have direct implications in dealing with unabated impacts of climate change as highlighted by Gössling and Schweiggart (2022). Future research may also focus on the important aspect of domestic tourism promotion in reducing the carbon footprint in tourism.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.