Expenditure and price elasticities for tourism sub-industries from the Fourier flexible form

Abstract

Tourism studies holding expenditure and price elasticities constant can produce misleading results. The Fourier flexible form provides estimates of expenditure and price elasticities over the business cycle. Results typically show considerable evidence of increased variation in expenditure and price elasticities over the business cycle and during the decline in overall tourism expenditure from 2001 to 2003 and from 2009 to 2011. Estimated own-price elasticities show that air transportation has the most elastic demand while food and beverage have an inelastic demand. Air transportation, shopping, and accommodation often have expenditure elasticities exceeding unity making them luxury goods during those periods. Results show that food and beverages are necessary goods. Estimated Morishima elasticities find air transportation and other transportation-related commodities are substitutes with the degree of substitution changing over time but are complementary in use with the remaining sub-industries. Marketing strategies from tourism agencies and governments should be flexible and respond to how consumers change expenditure over the business cycle.

Keywords

elasticities of substitution Fourier flexible form Morishima elasticities tourism sub-industries variable substitution over time

Introduction

Recent studies find variations in income and price elasticities which are often attributed to structural changes in tourism demand, economic and political shocks to the economy, liquidity constraints, and precautionary savings. There is ample evidence of asymmetric consumer behavior over the business cycle as in Carruth and Dickerson (2003), Süssmuth and Woitek (2013), Smeral and Song (2015), Smeral (2014), Alegre and Pou (2016), Gunter and Smeral (2016), Smeral (2017), Croes et al. (2018), and Aratuo et al. (2019). Tourism studies that hold income and price elasticities of demand constant may provide misleading results about substitution (Gunter and Smeral, 2016; Rudež, 2018; Song and Lin, 2010; Song and Witt, 2003). Using models with a constant elasticity of substitution may lead to forecasting errors as in Smeral (2012), Smeral and Song (2015), and Smeral (2017). However, if the degree of substitution in a tourist industry is relatively constant over time, then there is likely to be relatively little impact on forecasting errors and the estimate of substitution.

Instead of examining a homogeneous indicator of the tourism industry, Chen (2007), Tang and Jang (2009), and Aratuo and Etienne (2019) use sub-tourist industries to provide more insight into the tourism industry. Aratuo and Etienne (2019) and Tang and Jang (2009) find a weak linkage between the tourism industries and economic growth in the long run and that accommodation has the most causal links with the remaining tourist industries. Aratuo and Etienne (2019) expand the four sub-industries used by Tang and Jang (2009) and show that gross domestic product (GDP) co-moves with accommodation and food and beverage but does not cointegrate with the remaining four sub-industries. They only find a long-run relationship between other transportation and air transportation and short-run evidence of unidirectional causality from GDP to the six sub-industries. Overall, Aratuo and Etienne (2019) find support for the economy-driven tourism growth hypothesis. A meta-regression analysis of Nunkoo et al. (2020) also finds support for the tourism-lead growth hypothesis. Evidence of a bidirectional relationship between tourism and economic growth is found in Bilen et al. (2017), Brida et al. (2016), and Song and Liu (2017).

Often a time-varying parameter approach is used to allow for varying income and price elasticities. An alternative methodology to allow for variation in estimates of income and price elasticities is to use a globally flexible functional form such as the Fourier flexible form of Gallant (1981) or the asymptotically ideal model of Barnett and Yue (1988). These functional forms have been used in money and consumer demand studies as in Ewis and Fisher (1985), Barnett and Yue (1988), Fisher (1992), Fleissig and Swofford (1996), Fisher and Fleissig (1997), Fisher et al. (2001), Fleissig (1997), Serletis and Shahmoradi (2005), Fleissig (2016), and Assaf et al. (2019). The Fourier flexible form and asymptotically ideal model can provide global estimates of income and price elasticities from a system of demand equations at each data point over the sample and thus variations over time. This research uses the Fourier flexible functional form because it can incorporate more goods compared to the asymptotically ideal model.

This research contributes to the literature by estimating expenditure and price elasticities that can vary over the sample, for the six US sub-tourism industries as in Aratuo and Etienne (2019). The six US sub-tourism industries are air transportation, food and beverage, recreation and entertainment, shopping, travelers’ accommodations, and other transportation-related commodities. This is the first study to use the globally flexible Fourier functional form to provide estimates over the business cycle for expenditure elasticities, own-price elasticities, and Morishima elasticities of substitution for six tourist sub-industries. The Morishima elasticity of substitution is used to capture the substitution relationship between two sub-industries and can be non-symmetric. The estimated own-price elasticities, expenditure elasticities, and Morishima elasticities for all six sub-industries vary over the business cycle. The own-price elasticities of substitution show that air transportation is the most elastic and, as expected, the food and beverage industry has an inelastic demand. Air transportation is often a luxury industry with expenditure elasticities exceeding unity. The shopping and accommodation industries have fewer periods of expenditure elasticities exceeding unity but food and beverage are inelastic and are necessary goods. The estimated Morishima elasticities show that air transportation and other transportation-related commodities are substitutes for each other and the degree of substitution changes over time. The remaining relationships across sub-tourist industries show complementary in use.

The remainder of the article is organized as follows: the second section provides a brief literature review, with the data in the third section, and the Fourier flexible form methodology in the fourth section. The fifth section provides the empirical results, and the last section concludes the research.

Literature review

The semi-nonparametric Fourier flexible form of Gallant (1981) and asymptotically ideal model of Barnett and Yue (1988) are both dense in a Sobolev norm and can globally approximate the indirect utility function and partial derivatives of the indirect utility function. Approximating partial derivatives of the indirect utility function is important because the precision of elasticities of substitution depends on partial derivatives of the indirect utility function. A semi-nonparametric function refers to the use of a truncated series expansion that is dense in a Sobolev norm (see El Badawi et al., 1983). An alternative approach to estimating elasticities of substitution is to use a locally flexible functional form, which provides a local approximation to the data generating function in a delta neighborhood of an unknown and possibly small size (see Gallant, 1981). Early monetary studies using locally flexible forms include the translog for US data by Ewis and Fisher (1984), Serletis (1987, 1988) and an Almost Ideal Demand System by Barr and Cuthbertson (1991). Locally flexible functional forms are frequently used in consumer demand studies. White (1980) shows that locally flexible forms may provide biased and inconsistent estimates of elasticities of substitution. The Fourier flexible form was shown to provide more precise estimates compared to locally flexible functional forms as in Chalfant and Gallant (1985) and Fisher et al. (2001). The Fourier flexible form can include more variables and is used instead of the asymptotically ideal model. The Fourier flexible form has been used to analyze how substitution between monetary assets and across consumer goods changes over time by Ewis and Fisher (1985), Fisher (1992), Fisher and Fleissig (1997), Fleissig (1997), Jones et al. (2008), Fleissig and Jones (2015), Fleissig (2016), and Anderson et al. (2019). The Fourier flexible form has not been used to provide estimates in the tourism industry.

Many estimates of income and price elasticities for tourism are held constant as discussed in Crouch (1995), Smeral and Weber (2000), Song and Witt (2003), Song and Lin (2010), Smeral (2012), Smeral and Song (2015), Gunter and Smeral (2016), Smeral (2017), and Rudež (2018). However, research shows that income elasticities change over time as in Peng et al. (2015), Smeral and Song (2015), Bronner and Hoog (2016), and Gunter and Smeral (2016). During economic slowdowns, tourists substitute toward nearby places that can be accessed by relatively less expensive forms of transportation compared to generally higher price international travel. Also, Dwyer et al. (2006) find that tourists will substitute away from international travel to local travel during times of economic distress so international travel is no longer a luxury good.

Tourism data

The quarterly real tourism data are from the Bureau of Economic Analysis and have been used by Tang and Jang (2009) and Aratuo and Etienne (2019). The six tourism industries used by Aratuo and Etienne (2019) are air transportation, food and beverage, recreation and entertainment, shopping, travelers’ accommodations, and other transportation-related commodities. Food and beverage are transactions in restaurants and places that sell food and beverages. Recreation and entertainment covers leisure time activities like gambling, amusement parks and arcades, museums, historical site, skating rinks, ski lifts, day camps, sporting goods, and so on. Shopping are expenditures by tourists of nondurable commodities except gasoline. Travelers’ accommodation includes hotels, motels, and all other forms of lodging used by tourists. Rail, water transport, intercity bus, local bus, taxi, car rental, travel arrangement and reservation services, gasoline, and so on are part of other transportation-related commodities. The real tourism output encompasses estimates of domestically produced goods and services sold to travelers, and the seasonally adjusted quarterly real tourism data cover the period 1998.Q1 through 2017.Q3. Tourist expenditures across all six industries declined from 2001 to 2003 and from 2009 to 2011 with the largest decreases for accommodation and air transportation industries. Aratuo and Etienne (2019) provide a detailed explanation for each sub-industry. Since domestic tourism is about 80% of total US tourism (OECD, 2018), the estimates may be more representative of local travel.

Methodology

The Fourier flexible form of Gallant (1981) is

f (υ, θ) = u_{0} + b^{'} υ + \frac{1}{2} υ^{'} C υ + \sum_{α = 1}^{A} (u_{0 α} + 2 \sum_{j = 1}^{J} [u_{j α} cos (j {k^{'}}_{α} υ) - w_{j α} sin (j {k^{'}}_{α} υ)])

The vector of parameters to be estimated, $θ = \{b, u_{0 α}, u_{j α}, w_{j α} : j = 1, 2, \dots, J; α = 1, 2, \dots, A\},$ have no economic interpretation and are used to estimate the elasticities of substitution. Here, υ is a vector of expenditure of normalized prices for the six sub-industries. A multi-index, k_α, denotes partial differentiation of the indirect utility function and for notational purposes, $C = - \sum_{α = 1}^{A} u_{0 α} k_{α} {k^{'}}_{α}$ . The parameters A and J determine the number of terms and degree of the Fourier polynomials and are estimated by empirical testing. The number of terms used and the degree of the polynomials typically depend on the nonlinearity of the data generating function. The leading term of the Fourier flexible form, $u_{0} + b^{'} υ + \frac{1}{2} υ^{'} C υ,$ is the translog function of Christensen et al. (1971) and the remaining terms determine departures from the translog approximation. Applying Roy’s identity gives the Fourier demand share equations which are

s_{i} (υ, θ) = \frac{b_{i} υ_{i} - \sum_{α = 1}^{A} (u_{0 α} υ^{'} k_{α} + 2 \sum_{j = 1}^{J} j [u_{j α} sin (j {k^{'}}_{α} υ) + w_{j α} cos (j {k^{'}}_{α} υ)]) k_{i α} υ}{b^{'} υ - \sum_{α = 1}^{A} (u_{0 α} υ^{'} k_{α} + 2 \sum_{j = 1}^{J} j [u_{j α} sin (j {k^{'}}_{α} υ) + w_{j α} cos (j {k^{'}}_{α} υ)]) {k^{'}}_{α} υ}

where $s_{i} (υ, θ)$ are the share equations. Since the share equations add up to unity, any share equation can be omitted for estimating purposes. Share equations are estimated using the International TSP 4.5 seemingly unrelated regression procedure. The across-equation restrictions are imposed to ensure adding up as in Gallant (1981), Ewis and Fisher (1985), and Fisher (1992). The parameters are homogenous of degree zero and the normalization of b₆ = −1 is imposed in order to estimate the system of share equations. Convergence is set at 0.00001. There was evidence of autocorrelation in the demand system and a first-order vector autoregressive process, as in Berndt and Savin (1975), is applied to correct for serial correlation. The parameters A and J determine the degree of the Fourier polynomials and are determined by the upward F-test procedure of Eastwood (1991) with estimates A = 4 and J = 1. The Fourier parameter estimates are in Online Appendix 1. The share equations provide an accurate approximation to the data in terms of the R², root mean square error, and Q-statistic with the Box–Pierce test for autocorrelation for each share indicating white noise at the 5% level.

An important empirical issue is to estimate the degree of substitution or complementarity in use between commodities. The cross-price Allen elasticity of substitution has frequently been used to estimate substitution between commodities. However, Blackorby and Russell (1989) show that the cross-price Allen elasticity of substitution is quantitatively and qualitatively uninformative and may also provide incorrect estimates of substitution. The own-price Allen elasticities of substitution, in contrast to the cross-price Allen elasticities, do provide appropriate estimates of substitution. When there are three or more variables, Blackorby and Russell (1989) show that the elasticity of substitution can be measured by the Morishima elasticity, ME _ij = s_i(σ_ji − σ_ii), where s_i is share equation i, σ_ii are the own-price Allen elasticities, and σ_ij are the cross-price Allen elasticities of substitution. As Blackorby and Russell (1989) note, the partial differentiation used to derive ME _ij assumes that only p_i changes, since changes in p_j would cause changes in all other relative price ratios and can be non-symmetric. The Morishima elasticity of substitution measures how the ratio of the ith to the jth industry, holding utility constant, responds to a change in its relative price (p_i/p_j) because of a change in the price p_i Tourist sub-industries will be Morishima substitutes when ME _ij > 0 and complements when ME _ij < 0 Both the own-price Allen elasticities and Morishima elasticities are estimated from the parameters of the Fourier flexible form. The Morishima measure is often used in consumer demand studies (Davis and Gauger, 1996; Fisher and Fleissig, 1997; Fisher et al., 2001; Jones et al., 2008).

Empirical results

Given the overwhelming evidence of linkages between tourism sub-industries and changing patterns of tourism spending over the business cycle, the own-price elasticities, expenditure elasticities, and Morishima elasticities are likely to show considerable variation and non-symmetry over the business cycle. The full estimates are extensive and available upon request with summary statistics in Online Appendices 2 through 4. While it is not feasible to discuss all of the estimates, I now provide some highlights.

Own-price elasticities of substitution

The estimated own-price elasticities of substitution are all negative, statistically significantly different from zero at the 5% level, and show considerable variation of the sample (see Figure 1). Summary statistics are provided in Online Appendix 2. The evidence of variability provides further evidence that studies that impose constant own-price elasticity of substitution may be misleading as discussed in Song and Lin (2010), Song and Witt (2003), Song et al. (2010, 2011), Smeral and Song (2015), Gunter and Smeral (2016), and Rudež (2018).

Figure 1.

Own-price elasticities.

Food and beverage is the only sub-industry that has an inelastic demand over the entire sample. The estimated food and beverage own-price elasticities become less inelastic around the Great Recession and typically during the period of declining tourism expenditures from 2001 to 2003. Over the period 2010–2011, the own-price elasticity of demand for the food and beverage industry remained relatively constant even though tourism expenditure declined over this period.

The own-price elasticity of demand for air transportation is generally elastic and shows the most variation of all six sub-industries. These estimates suggest that consumers are very responsive to the price changes for air transportation. The estimated own-price elasticity of demand for air transportation becomes elastic during the Great Recession. During the periods of declines in overall tourism expenditure of 2001–2003 and 2009–2011, the own-price elasticity of demand for air transportation shows much variability. There are also other periods outside of these two periods that show much variability in the own-price elasticity estimates for airline transportation. Toward the end of the 1990s, there was relatively strong US economic growth. Also, the structure of the US airline industry changed in the early 2000s with two major airline mergers and four bankruptcies. For air transportation, the own-price elasticity of demand is highly variable and elastic from 1998 to 1999 but becomes inelastic during the first two quarters of 2000. The own-price elasticity of demand for air transportation became more elastic following the September 2001 terrorist attack when air transportation declined. According to Berry and Jia (2010), low-cost carriers increased their market share from about one-fifth in 1999 to about one-third by 2006. The variability in estimated own-price elasticities for air transportation from 2012 through 2015 coincides with a significant decline in RGDP growth in 2014.Q1, relatively slow 0.1% growth in 2015.Q4, and during the 2015 investigation by the US Department of Justice into industry price collusion and subsequent settlement by some airlines. The average estimate for international air transportation for the meta-analysis of Peng et al. (2015) was −0.920, whereas the mean for this study is −1.178 (see Online Appendix 2). Except for food and beverage, the remaining tourist sub-industries have similar trends to air transportation in terms of variability for the estimated own-price elasticities, including the periods outside of the two recessions. Estimated own-price elasticities for other transportation-related commodities also vary over the sample but much less compared to air transportation. Recreation, shopping, and accommodation are typically more elastic in demand compared to food and beverage and other transportation-related commodities. Changing own-price elasticities of demand during a recession were also found by Song and Witt (2003), Crouch (1995), Song and Lin (2010), Smeral and Song (2015), Gunter and Smeral (2016), and Rudež (2018). Peng et al. (2015) calculate the average price elasticity for accommodation at −0.727 which is lower in absolute value than the −0.857 for the estimates for accommodation (see Online Appendix 2). While the average own-price elasticity of demand for accommodation, shopping, and recreation and entertainment is similar, the variation of the estimated own-price elasticities for these three sub-industries differs.

Expenditure elasticities of demand

Expenditure in a tourism industry responds more to changes in economic growth if it has an elastic expenditure elasticity of demand. If a sub-tourist industry has an inelastic expenditure elasticity of demand, then purchases may change relatively little over the business cycle. A highly inelastic expenditure of demand may suggest that the tourist industry has become saturated.

The estimated expenditure elasticities of demand are statistically significant from zero at the 5% level and also show considerable variation over the sample as shown in Figure 2. Summary statistics are provided in Online Appendix 3. Once again, the evidence of variability highlights the problem of providing constant estimates of the expenditure elasticity which masks the changes over time. The most elastic estimates of expenditure elasticities of demand are generally for air transportation. In particular, air transportation demand becomes increasingly more elastic during the Great Recession and a luxury industry with estimates exceeding unity. The expenditure elasticity for the air transportation industry became more variable over the periods of declines in overall tourism expenditure from 2001 to 2003 and from 2009 to 2011. There is also evidence of considerable variability in the expenditure elasticities outside of these two periods. The maximum expenditure elasticity for air transportation of 1.477 occurs in 2008.Q3. The meta-analysis of Crouch (1995) found that international income elasticities for outbound tourism exceeded unity for over 70% of the studies so that international outbound tourism is a luxury good as in Smeral (2004), Garin-Munoz (2007), and Dogru et al. (2017). The average international air transportation income elasticity of 1.605 from the meta-analysis of Peng et al. (2015) exceeds the average estimate for this study of 0.950, probably because international travel is generally considerably more expensive compared to local travel in the United States for which the data represent about 80% for this study.

Figure 2.

Expenditure elasticities.

Shopping also has periods of elastic expenditure elasticities of demand that exceed unity, and estimates become increasingly variable during the Great Recession. The estimated expenditure elasticities for accommodation become elastic and exceed unity during the Great Recession when consumer spending on tourism sharply decreased. The estimated expenditure elasticity for accommodation became relatively inelastic when tourism expenditure declined in 2010–2011. Peng et al. (2015) find that international accommodation has an income elasticity of 1.166 which is higher than the estimated mean of 0.699 (see Online Appendix 3). Divisekera (2010) found accommodation to be a luxury good and shopping a necessity which is consistent for the estimates, but only over some periods of the business cycle. Recreation, other transportation-related commodities, and food and beverage all have inelastic estimated expenditure elasticities of demand. Costa (1997) finds the estimated income elasticities for recreational goods have decreased over time and by 1991 were above unity. The inelastic estimates for recreation may reflect the downward trend found by Costa (1997). As expected, the estimated food and beverage expenditure elasticities are generally the most inelastic compared to the other sub-industries as these are necessary expenditures. Smeral (2017) finds that tourism products were elastic in demand in slow growth periods and thus luxury goods but income inelastic and necessary goods during fast growth periods, which is largely consistent with the estimated expenditure elasticities. Changing income elasticities over the business cycle were attributed to liquidity constraints, loss aversion, and precautionary savings. Smeral (2017) and Divisekera (2010) also find that income elasticities vary across travel-related products.

Morishima elasticities

The Morishima elasticity (ME _ij ) shows substitution between two sub-industries for a change in the price of sub-industry i. A positive (negative) ME _ij indicates that the sub-industries are substitutes (complements). The estimated Morishima elasticities are statistically significantly different from zero at the 5% level with much evidence of variation over the sample. Summary statistics are provided in Online Appendix 4.

The Morishima estimates of substitution between tourism sub-industries for a change in the price of air transportation (ME_4j) show variation over the sample and are in Figure 3. This is further evidence that substitution between sub-industries changes over the business cycle and is not constant. For price changes in air transportation, all sub-industries are complements except for other transportation-related commodities (ME₄₆) which are substitutes. A price increase in air transportation will reduce the demand for air transportation from the estimated own-price elasticities and induce substitution into other transportation-related commodities. In contrast, a price increase in air transportation will reduce expenditure on the remaining sub-industries, which are complements in use, with the least impact on food and beverage (ME₄₂). Other transportation-related commodities (ME₄₆) have the most elastic response to a price change in air transportation compared to the remaining sub-industries. For changes in the price of air transportation, there is much variation in substitution from 2001 to 2003 and from 2009 to 2011 which coincides with the declines in tourism expenditure.

Figure 3.

Morishima elasticity of substitution changes in price of air transportation.

Shopping and the remaining sub-industries are complements in use for a change in the price of shopping (ME_3j) and the estimates also vary over the sample (see Figure 4). An increase in the price of shopping will have relatively little impact on reducing food and beverage purchases (ME₃₂) which are necessary sub-industries but have the largest impact on accommodation (ME₃₁) and air transportation (ME₃₄). The estimated Morishima elasticities of substitution across sub-industries fluctuate in response to changes in the price of shopping over the periods of declines in tourism expenditure from 2001 to 2003 and from 2009 to 2011, as well as outside of these periods.

Figure 4.

Morishima elasticity of substitution changes in price of shopping.

The estimated Morishima elasticities can be non-symmetric. For example, substitution between air transportation and other transportation-related commodities shows greater substitution for changes in the price of air transportation (ME₄₆) compared to other transportation-related commodities (ME₆₄) as in Figure 5. The estimated Morishima elasticities of substitution between air transportation and other transportation-related commodities (ME₄₆) show many periods where substitution becomes elastic particularly from 2001 to 2003 and from 2009 to 2011 as well as other periods in the sample.

Figure 5.

Morishima elasticity of substitution air transportation and other transportation.

The Morishima elasticities of substitution show that accommodation and food and beverage are complements in use. As expected, the Morishima elasticities between accommodation and a necessary good like food and beverage are more responsive to price changes in accommodation (ME₁₂) compared to changes in food and beverage prices (ME₂₁), as shown in Figure 6. There is considerably more variation in the complementary relationship between accommodation and food and beverages for price changes in accommodation.

Figure 6.

Morishima elasticity of substitution accommodation and food.

These examples of non-symmetry in the estimated Morishima elasticities demonstrate the importance of allowing for substitution to change over the sample for a change in prices of a sub-industry. Not only is it important to allow substitution to vary over time but also to analyze the impact on substitution between sub-industries with respect to which tourist industry undergoes a price change.

Conclusion

The globally Fourier flexible functional form was used to estimate substitution across six US sub-tourism industries of air transportation, food and beverage, recreation and entertainment, shopping, travelers’ accommodations, and other transportation-related commodities. A major result is that the own-price, expenditure, and Morishima elasticities vary over the business cycle. The estimated own-price elasticities show that the food and beverage industry was very inelastic in demand, whereas air transportation was often elastic in demand over parts of the sample. Estimated expenditure elasticities of demand for the food and beverage industry were typically the lowest out of all six industries and are necessary purchases. Air transportation, shopping, and accommodation had expenditure elasticities that exceed unity for periods over the business cycle and are considered luxuries over those time frames.

The estimated Morishima elasticities of substitution had air transportation and other transportation-related commodities as substitutes for each other but all the remaining pairwise relationships for the tourist industries show complementarity in use. There was much evidence of non-symmetric Morishima elasticities over the business cycle. For example, an increase in the price of air transportation would induce more substitution into other transportation-related commodities than an increase in the price of other transportation-related commodities would into air transportation. An increase in the price of accommodation would reduce food expenditure. In contrast, a rise in food prices would have little impact on the accommodation industry.

The own-price, expenditure, and Morishima elasticities typically vary considerably over the periods when tourism expenditure declined from 2001 to 2003 and from 2009 to 2011 as well as other parts of the sample. Given the considerable variations in the estimated own-price, expenditure, and Morishima elasticities over the business cycle, the marketing strategies of tourism agencies and governments should be flexible and respond to changes in the economy which is consistent with the conclusions of Song and Witt (2003), Song et al. (2010, 2011), Smeral and Song (2015), Gunter and Smeral (2016), and Rudež (2018). Investing in the tourism industry needs to consider various sub-industries. Breaking down elasticity estimates across sub-industries provides important information. During an economic slowdown or crisis, tourism marketing may be better off steering away from the elastic air transportation and shopping industries and to the less elastic sub-industries like other transportation-related commodities and food and beverage. Tourism expenditure on pairs of sub-industries is often non-symmetric and any substitution or complementarity relationship may be sensitive to which industry experiences a price change. In times of increases in the price of both air transportation and other transportation-related commodities, the focus should be on other transportation-related commodities as demand is likely to decrease by a smaller amount compared to air transportation. The estimates are consistent with results that show that consumers change their tourism strategies over the business cycle including length of stay, less expensive accommodation, cheaper travel, and close destinations as in Eugenio-Martin and Campos-Soria (2014) and Campos-Soria et al. (2015).

About 80% of the tourism data used in this study represent local US travel and future research could focus on international travel across industries. International tourism demand is generally regarded as income elastic and a luxury good, so estimates of sub-tourist industries will provide further insight for tourism planning. It would also be useful to analyze the impact of policy uncertainty on the own-price elasticities, expenditure, and Morishima elasticities.

Supplemental material

Appendix - Expenditure and price elasticities for tourism sub-industries from the Fourier flexible form

Appendix for Expenditure and price elasticities for tourism sub-industries from the Fourier flexible form by Adrian R Fleissig in Tourism Economics

Footnotes

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) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Adrian R Fleissig

Supplemental material

Supplemental material for this article is available online.

Author biography

Adrian R Fleissig is a professor of Economics at California State University. He has a joint PhD in Economics and Statistics from North Carolina State University. His research focuses on developing and estimating consumer and money demand systems from semi-nonparametric functional forms and developing nonparametric revealed preference tests. Another area of research is developing and testing Divisia monetary aggregates. Much of his applied research focuses on consumer demand, money demand, and monetary policy.

References

Alegre

Pou

(2016) US household tourism expenditure and the Great Recession: an analysis with the Consumer Expenditure Survey. Tourism Economics 22(3): 608–620.

Anderson

Duca

Fleissig

, et al. (2019) New monetary services (Divisia) indexes for the post-war U.S. Journal of Financial Stability 42: 3–17.

Aratuo

Etienne

(2019) Industry level analysis of tourism-economic growth in the United States. Tourism Management 70: 333–340.

Aratuo

Etienne

Gebremedhin

, et al. (2019) Revisiting the tourism-economic growth nexus: evidence from the United States. International Journal of Contemporary Hospitality Management 31: 1–15.

Assaf

Berger

Roman

, et al. (2019) Does efficiency help banks survive and thrive during financial crises? Journal of Banking and Finance 106: 445–470.

Barnett

Yue

(1988) Seminonparametric estimation of the asymptotically ideal model: the AIM demand system. In: Rhodes

Fomby

(eds) Nonparametric and Robust Inference Advances in Econometrics. Greenwich, CT: JAI Press, pp. 127–174.

Barr

Cuthbertson

(1991) Neoclassical consumer demand theory and the demand for money. The Economic Journal 101(407): 855–876.

Berndt

Savin

(1975) Estimation and hypothesis testing in singular equation systems with autoregressive disturbances. Econometrica 43: 937–957.

Berry

Jia

(2010) Tracing the woes: an empirical analysis of the airline industry. American Economic Journal: Microeconomics 2(3): 1–43.

10.

Bilen

Yilanci

Eryüzlü

(2017) Tourism development and economic growth: a panel Granger causality analysis in the frequency domain. Current Issues in Tourism 20(1): 27–32.

11.

Blackorby

Russell

(1989) Will the real elasticity of substitution please stand up? (A comparison of Allen/Uzawa and Morishima elasticities). American Economic Review 79(4): 882–888.

12.

Brida

Cortes-Jimenez

Pulina

(2016) Has the tourism-led growth hypothesis been validated? A literature review. Current Issues in Tourism 19(5): 394–430.

13.

Bronner

de Hoog

(2016) Crisis resistance of tourist demand: the importance of quality of life. Journal of Travel Research 55(2): 190–204.

14.

Bureau of Economic Analysis (2017) Tourism satellite accounts 1998–2017. Available at: https://www.bea.gov/newsreleases/industry/tourism/tour_glance.htmhttp://www.bea.gov/industry/index.htm (accessed September 2017).

15.

Campos-Soria

Inchausti-Sintes

Eugenio-Martin

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

16.

Carruth

Dickerson

(2003) An asymmetric error correction model of UK consumer spending. Applied Economics 35(6): 619–630.

17.

Chalfant

Gallant

(1985) Estimating substitution elasticities with the Fourier cost function: some Monte Carlo results. Journal of Econometrics 28(2): 205–222.

18.

Chen

(2007) Interactions between business conditions and financial performance of tourism firms: evidence from China and Taiwan. Tourism Management 28(1): 188–203.

19.

Christensen

Jorgenson

Lau

(1971) Conjugate duality and the transcendental logarithmic function. Econometrica 39: 255–256.

20.

Costa

(1997) Less of a luxury: the rise of recreation since 1888. No. w6054, National Bureau of Economic Research, pp. 1–31.

21.

Croes

Ridderstaat

Rivera

(2018) Asymmetric business cycle effects and tourism demand cycles. Journal of Travel Research 57(4): 419–436.

22.

Crouch

(1995) A meta-analysis of tourism demand. Annals of Tourism Research 22(1): 103–118.

23.

Davis

Gauger

(1996) Measuring substitution in monetary-asset demand systems. Journal of Business and Economic Statistics 14(2): 203–208.

24.

Divisekera

(2010) Economics of tourist’s consumption behaviour: some evidence from Australia. Tourism Management 31(5): 629–636.

25.

Dogru

Sirakaya-Turk

Crouch

(2017) Remodeling international tourism demand: old theory and new evidence. Tourism Management 60: 47–55.

26.

Dwyer

Forsyth

Spurr

, et al. (2006) Economic effects of the world tourism crisis on Australia. Tourism Economics 12(2): 171–186.

27.

Eastwood

(1991) Asymptotic normality and consistency of semi-nonparametric regression estimators using an upwards F test truncation rule. Journal of Econometrics 48(1–2): 151–181.

28.

El Badawi

Gallant

Souza

(1983) An elasticity can be estimated consistently without a priori knowledge of functional form. Econometrica 51(6): 1731–1751.

29.

Eugenio-Martin

Campos-Soria

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

30.

Ewis

Fisher

(1984) The translog utility function and the demand for money in the United States. Journal of Money, Credit and Banking 16(1): 34–52.

31.

Ewis

Fisher

(1985) Toward a consistent estimate of the substitutability between money and near monies. Journal of Macroeconomics 7(2): 151–174.

32.

Fisher

(1992) Money-demand variability: a demand-systems approach. Journal of Business and Economic Statistics 10(2): 143–151.

33.

Fisher

Fleissig

(1997) Monetary aggregation and the demand for assets. Journal of Money, Credit and Banking 29(4): 458–475.

34.

Fisher

Fleissig

Serletis

(2001) An empirical comparison of flexible demand system functional forms. Journal of Applied Econometrics 16(1): 59–80.

35.

Fleissig

Swofford

(1996) A dynamic asymptotically ideal model of money demand. Journal of Monetary Economics 37(2): 371–380.

36.

Fleissig

(1997) The consumer consumption conundrum: an explanation. Journal of Money, Credit and Banking 29(2): 177–192.

37.

Fleissig

(2016) Changing trends in US alcohol demand. Atlantic Economic Journal 44(3): 263–276.

38.

Fleissig

Jones

(2015) The impact of commercial sweeping on the demand for monetary assets during the Great Recession. Journal of Macroeconomics 45: 412–422.

39.

Gallant

(1981) On the bias in flexible functional forms and an essentially unbiased form: the Fourier flexible form. Journal of Econometrics 15(2): 211–245.

40.

Garin-Munoz

(2007) German demand for tourism in Spain. Tourism Management 28(1): 12–22.

41.

Gunter

Smeral

(2016) The decline of tourism elasticities in a global context. Tourism Economics 22(3): 466–483.

42.

Jones

Fleissig

Dutkowsky

, et al. (2008) Monetary policy and monetary asset substitution. Economics Letters 99(1): 18–22.

43.

Nunkoo

Seetanah

Jaffur

, et al. (2020) Tourism and economic growth: a meta-regression analysis. Journal of Travel Research 59(3): 404–423.

44.

OECD (2018) OECD Tourism Trends and Policies. Available at: https://www.oecd.org/cfe/tourism/2018-Tourism-Trends-Policies (accessed March 2018).

45.

Peng

Song

Crouch

, et al. (2015) A meta-analysis of international tourism demand elasticities. Journal of Travel Research 54(5): 611–633.

46.

Rudež

(2018) Relationship between income and tourism demand: old findings and new research. Academica Turistica-Tourism and Innovation Journal 11(1): 1–6.

47.

Serletis

(1987) On the demand for money in the United States. Empirical Economics 12(4): 249–255.

48.

Serletis

(1988) Translog flexible functional forms and substitutability of monetary assets. Journal of Business and Economic Statistics 6(1): 59–67.

49.

Serletis

Shahmoradi

(2005) Semi-nonparametric estimates of the demand for money in the United States. Macroeconomic Dynamics 9(4): 542–559.

50.

Smeral

(2004) Long-term forecasts for international tourism. Tourism Economics 10: 145–166.

51.

Smeral

(2012) International tourism demand and the business cycle. Annals of Tourism Research 39(1): 379–400.

52.

Smeral

(2014) Forecasting international tourism with due regard to asymmetric income. Tourism Economics 20(1): 61–72.

53.

Smeral

(2017) Tourism forecasting performance considering the instability of demand elasticities. Journal of Travel Research 56(7): 913–926.

54.

Smeral

Song

(2015) Varying elasticities and forecasting performance. International Journal of Tourism Research 17(2): 140–150.

55.

Smeral

Weber

(2000) Forecasting international tourism trends to 2010. Annals of Tourism Research 27: 982–1006.

56.

Song

Kim

Yang

(2010) Confidence intervals for tourism demand elasticities. Annals of Tourism Research 37(2): 377–396.

57.

Song

Lin

(2010) Impacts of the financial and economic crisis on tourism in Asia. Journal of Travel Research 49(1): 16–30.

58.

Song

Lin

Witt

, et al. (2011) Impact of financial and economic crisis on demand for hotel rooms in Hong Kong. Tourism Management 32(1): 172–186.

59.

Song

Liu

(2017) Predicting tourist demand using big data. In Xiang

Fesenmaier

(eds) Analytics in Smart Tourism Design. Cham: Springer, pp. 13–29.

60.

Song

Witt

(2003) Tourism forecasting: the general-to-specific approach. Journal of Travel Research 42(1): 65–83.

61.

Süssmuth

Woitek

(2013) Estimating dynamic asymmetries in demand at the Munich Oktoberfest. Tourism Economics 19(3): 653–674.

62.

Tang

CHH

Jang

(2009) The tourism–economy causality in the United States: a sub-industry level examination. Tourism Management 30(4): 553–558.

63.

White

(1980) Using least squares to approximate unknown regression functions. International Economic Review 21(1): 149–170.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.15 MB