Is Tourism a Low-Income Industry? Evidence from Three Coastal Regions

Abstract

Tourism has been a popular development strategy based on its potential of economic impacts and job opportunities, but the quality of jobs created by tourism has received little empirical investigation. This article uses data from IMPLAN and the United States Bureau of Labor Statistics to examine the income distribution of jobs created by tourism and compares the income distribution from tourism expenditures to the income distribution in three different coastal regions with large tourism industries. To examine the distribution of income from tourism, an eight-step procedure was developed by modifying previous procedures for estimating income distribution. Study findings indicate that the jobs generated by tourism expenditures had a lower income distribution than the overall income distribution in the three regions, with the difference being most apparent in more urban areas. The various implications of the low-income distribution of the tourism industry are discussed.

Keywords

economic impact tourism development job creation income distribution coastal tourism

Introduction

Traditionally, job generation has been often considered the primary concern of regional developers, and creating high-paying jobs with good benefits is a secondary or tertiary concern (Courant 1994). Politicians and regional planners enjoy highlighting large manufacturing plants employing numerous residents as evidence of successful economic growth. However, job creation as a simple metric for measuring economic growth is increasingly drawing criticism as being too narrow in focus (Courant 1994; Mack and Grubesic, forthcoming). Waits (2000) summarizes the problem with the conventional philosophy of emphasis on job creation, stating that “the jobs created during the past two decades are not good jobs. They do not pay enough to support a family, and they offer few benefits” (p. 36). Recent empirical research suggests that skilled and educated individuals (and the “creative class”) with high-paying jobs are the key to regional economic development, and low-income industries likely have a negative effect on economic development metrics (Florida, Mellander, and Stolarick 2008). Regions with highly productive workers typically have highly compensated citizens and produce more high-value-added goods and services. These differences may become more pronounced over time and hold important implications for regional development. Mack and Grubesic (forthcoming) conclude that traditional evaluations (i.e., job creation) of economic growth may “grossly misrepresent the true trajectory of regional economies” (p. 13).

Tourism occupies a tenuous position in this move toward a more nuanced view of economic development. Several researchers have argued that while the tourism industry (the status of tourism as an industry has been questioned [see Leiper 2008] but, for brevity, we refer to the assortment of sectors that heavily contribute to the tourism phenomenon as the “tourism industry” in this study) often creates numerous employment opportunities, the job quality is typically below average, featuring too many low-paying jobs and a high percentage of part-time and seasonal employment (e.g., Choy 1995; Lee and Kang 1998; Seckelmann 2002). Tourism has been a popular development strategy because of its potential to create large economic impacts and numerous new job opportunities with what is often perceived to be relatively little economic, environmental, and social costs (Deller, Lledo, and Marcouiller 2001). However, if regional developers become more aware of the type of jobs they create with subsidies or other incentives, tourism may slowly fall out of favor as an instrument for economic development because of concerns over the quality of jobs it creates.

Despite the importance of this issue, the quality of jobs created by tourism has received little empirical investigation. While the general consensus surrounding the issue is that tourism is a low-income industry (Lee and Kang 1998; Marcouiller, Kim, and Deller 2004), this hypothesis has rarely been systematically evaluated. Consequently, this article intends to explore the income distribution of jobs created by tourism using a modified methodology originally proposed by Daniels, Norman, and Henry (2004) and to compare the income distribution in three different coastal regions with large tourism industries. The results will examine both whether or not tourism is a low-income industry and, if so, the degree to which it is a low-income industry.

Tourism’s Income Distribution

Tourism is a frequently prescribed solution for regions that have stagnated economically (Liu and Wall 2006; Ramjee et al. 2010; Seetanah 2010). Besides providing industrial diversity and a new avenue for economic growth, tourism offers the promise of creating numerous employment opportunities, and these new jobs may spur accelerated economic development. Some studies note that regions with large tourism and recreation sectors have experienced faster economic growth (Johnson and Stewart 2005; Reeder and Brown 2005). In addition to the economic benefits of the tourism industry, several nonmarket benefits such as boosting an area’s image to the outside world (Richards and Wilson 2004) and creating a scenic backdrop (Fleischer and Tchetchik 2005) are also brought by tourism, making it an especially attractive development strategy. Marcouiller (2007) asserts that tourism promotion is a “no-brainer” (p. 28) for politicians, who can easily cite benefits of tourism promotion to the community.

Nevertheless, the idea of tourism as a strong engine for economic growth is periodically challenged on a number of grounds. Critics including Crompton (2006) and Leiper (2008) have asserted that employment figures are often exaggerated by improper methodologies and a poor understanding of the key issues related to tourism and employment generation. According to Marcouiller, Kim, and Deller (2004), several studies that view tourism as an economic development strategy have also failed to uncover strong relationships between tourism and poverty reduction. More importantly, there is an intense concern about the quality of jobs that tourism creates (Blake et al. 2008; Daniels, Norman, and Henry 2004; Lee and Kang 1998). Tourism frequently generates numerous jobs in the hospitality business sector such as hotels and restaurants that are not generally considered high quality (Choy 1995). These sectors are likely to hire numerous low-wage, seasonal, and/or part-time employees without substantial benefits (Lee and Kang 1998; Marcouiller, Kim, and Deller 2004). Marcouiller and Xia (2008) summarize the debate on tourism employment, stating:

An underlying tension exists within the conventional wisdom related to tourism which can be characterized by extremes that reflect two ill-conceived positions. On the one hand, proponents of tourism argue that broad-based employment benefits are substantial and clearly justify large marketing subsidies to increase the incidence of tourist travel. On the other, opponents of tourism argue that the jobs created by tourism firms tend to be low wage, seasonal and lacking substantial benefits; thus, public resources targeting the creation of jobs are best placed with industrial sectors characterized by higher wage/benefit and year-round employment opportunities. (pp. 447-48)

While acknowledging that tourism creates predominately low-income jobs, other scholars argue that these jobs are still extremely valuable. Szivas, Riley, and Aireya (2003) assert that tourism is an excellent industry for generating employment as it creates a wide variety of jobs. These jobs may require different working hours and various skill sets and provide different wages and working conditions. This diversity enables job seekers to more easily find jobs that meet their needs and match their skill sets. Additionally, it means that locals are likely to fill most of the available positions as opposed to a more specialized industry, which might attract more nonlocals with unique skills; these numerous entry-level jobs can help a region or nation reduce poverty (Blake et al. 2008). Furthermore, tourism jobs are also popular with individuals interested in jobs with nonmonetary benefits, such as providing high quality of life. Lee and Kang (1998) note that while tourism generates primarily entry-level-type jobs, these can be important stepping stones in careers, which can help in building stronger resumes.

While there is a general consensus that tourism is a low-income industry, there is little empirical evidence to support this proposition. Marcouiller (2007) suggests that more “policy-relevant studies” (p. 30) are needed to contribute to the understanding of tourism’s effect on regional economic development. A better understanding of the income distribution of tourism will allow for a more judicious application of tourism promotion as a development strategy. For instance, if tourism is indeed a low-income industry with numerous entry-level jobs, it may not be an adequate development strategy for an area with low unemployment; at the same time, it would be perfectly amenable to an area with high unemployment.

The research gap on the income distribution from the tourism industry is peculiar given the field’s substantial set of economists and the large volume of research on tourism’s regional economic impacts. Perhaps, the major factor behind this lack of research is the field’s widespread reliance on Input–Output (IO) models. Despite their frequently cited flawed assumptions and shortcomings when compared to other types of economic models such as computable general equilibrium modeling (Dwyer, Forsyth, and Spurr 2005; Miller and Blair 2009), IO modeling remains a useful and well-recognized tool for analyzing economic impacts in the tourism field (Bonn and Harrington 2008). While the most basic result from IO modeling estimates the increase in gross regional and/or national products created by a specified tourism impact, figures such as employment and income generated are often considered more useful as they can indicate an effect on the region’s quality of life and are more easily understandable by nonpractitioners (Crompton 1999).

Additionally, researchers have asserted that understanding how these benefits (jobs and income) are spread through a community is even more important than the amount of jobs and income created (e.g., Daniels, Norman, and Henry 2004; Samdahl 1999). Of particular interest is the question of what income segments the economic benefits of tourism accrues to. In other words, is the economic benefit evenly spread throughout the community? Does tourism mainly benefit high-income groups or low-income groups? This type of question is often addressed through a social accounting matrix (SAM)–based modeling system. SAMs are an extension of IO analysis that further disaggregates how economic impacts are circulated throughout an economy and how income is distributed across various income categories. However, when performed using most IO models including the ever popular Impact Analysis for Planning (IMPLAN), this disaggregation is of limited practical use. Daniels, Norman, and Henry explain (2004, p. 182):

The methods used in IMPLAN to distribute income across households assume a constant share accruing in proportion to the existing distribution in the region (county). Data constraints do not allow the income impacts by industry sector to be allocated to household groups. As such, use of standard SAM accounts from IMPLAN may give unwarranted weight to average personal income shares in assessing distributional consequences of an event. Accordingly, it may be necessary to supplement the default SAM sector data in IMPLAN with specifics regarding tourism-related occupations.

This means that IMPLAN equally distributes the income generated by any industry in the same ratio as what currently exists in the region. Thus, while IMPLAN-based SAMs may seem like an attractive tool initially to analyze the distribution of income from tourism, it does not help analyze the underlying impacts of income distribution specifically generated from tourism. Heavy reliance on IMPLAN and similar IO modeling systems has contributed to a large gap in the field understanding of the income distribution from tourism.

Daniels, Norman, and Henry (2004) addressed this shortcoming and designed a more suitable method of analysis. In brief, after IMPLAN generates estimates of employment in different industries created by tourist spending (as is traditionally done to create estimates of jobs creation), these are then converted via a bridge table into employment estimates across the North American Industrial Classification System (NAICS). Bureau of Labor Statistics (BLS) data are then used to disaggregate jobs generated in different industries into various job categories in the standard occupation classification (SOC) system. The BLS states that “the SOC is designed to reflect the current occupational structure of the United States; it classifies all occupations in which work is performed for pay or profit” (p. ii). After converting jobs created in IMPLAN sectors into different NAICS categories and then into SOC job categories, they used the average income from each classification in the SOC to estimate the distribution of income effects across different income categories. While both novel and practically relevant, this methodology has received little attention since its creation. Specifically, lacking from the research is a comparison of the income distribution of tourism-related jobs to the overall income distribution in the study regions of interest. Thus, this article will be unique in its use of empirical investigation using macro-level data to compare the income distribution of jobs created by tourism to the overall income distribution currently occurring in three coastal regions and will provide valuable policy-relevant information that researchers such as Marcouiller (2007) find lacking in the literature.

Three specific research objectives include (1) modifying the methodology proposed by Daniels, Norman, and Henry (2004) to calculate the incomes distribution of tourism in three regions, (2) examining the types of jobs that tourism creates, and (3) comparing the income distribution of tourism-generated employment to that of the region in order to determine whether tourism incomes are disproportionally distributed across different income groups in the region.

Method

Study Area

The South Carolina coast serves as a useful study area because of the diverse nature of the region. Three coastal regions were included for this study: Myrtle Beach, Charleston, and Beaufort–Hilton Head Island. The city of Myrtle Beach dominates the northern portion of the coast and is known as a low-cost family-friendly destination that has a predominance of amusement park–like activities. Charleston is located in the central region and is known worldwide for showcasing culture and history. Beaufort and Hilton Head and are on the southern coast, near the Georgia border, and are primarily upscale resort destinations and have less development along the coastal areas. The different regions vary from the fairly urban Charleston to the rural Beaufort–Hilton Head region. This variety will assist in examining how income effects from tourism may vary across different levels of development. The results of this study are practically important because of the large economic impact of tourism in the region and the ongoing changes in the region’s economy. Tourism is a major driver of the South Carolina Coast’s economy. The three coastal counties of Horry, Charleston, and Beaufort generated more than $5.7 billion in domestic travel expenditures in 2008; this represents 58.2% of South Carolina’s total tourism expenditures (U.S. Travel Association 2009). Given the importance of tourism to South Carolina’s coast, it is important to understand how income generated by tourism is distributed across different income groups.

Estimating Income Distribution

This study measures the distribution of income from tourism in an eight-step process. These steps are as follows: (1) determining the impact region, (2) collecting data, (3) estimating total expenditures, (4) applying margins, (5) estimating total employment generated by IMPLAN category, (6) converting employment by IMPLAN categories into employment in NAICS categories, (7) using BLS data to calculate employment by SIC, and (8) arranging jobs created by SIC categories into wage categories (i.e., $10,000-$19,999; $20,000-$29,999, etc.). The first five categories are typically employed in tourism impact analysis, and the last three are more unique and follow,with one deviation, the methodology proposed by Daniels, Norman, and Henry (2004).

Step one

Appropriate study areas to conduct economic analysis were selected. Three separate areas are under examination: The Hilton Head–Beaufort area consisting of Beaufort and Jasper counties; the Charleston area consisting of Charleston, Berkeley, and Dorchester counties; and the Myrtle Beach area consisting of Horry county. Beaufort and Jasper counties are considered a micropolitan statistical area, the three Charleston area counties are a Metropolitan statistical area, and the Myrtle Beach area is a Metropolitan statistical area. Micropolitan statistical areas and metropolitan statistical areas are noted for having a “high degree of social and economic integration” (Spotila 2000, p. 82238); this integration made them an ideal spatial scale to use when measuring regional economic impact.

Step two

A survey was conducted during fall 2008 and summer 2009. Tourists, who were not residents of the destination counties, were intercepted along the South Carolina Coast, in a variety of venues including beaches, visitor centers, state parks, and downtown shopping areas. If they agreed to participate in the study, tourists were asked for their name and mailing address for a follow-up mail survey. The survey was delivered on a rolling basis using a modified Dillman (2000) survey method. The first mailing occurred one week after the tourists were intercepted and was followed by a reminder postcard; a second and third questionnaire was mailed to nonrespondents. By mailing the surveys shortly after the tourists were intercepted, recall bias was minimized and tourists did not have to estimate future planned expenditures on their vacation. A total of 1,735 tourists agreed to provide their names and addresses for follow-up mail surveys. We received 797 completed surveys (44.2% raw response rate) and after eliminating surveys that had not filled out the expenditure section of the survey, and tourists that were not from outside the statistical area (e.g., tourists to Beaufort county from Jasper county were eliminated), 657 surveys were deemed usable for this analysis (37.8% useable response rate).

Step three

The survey asked respondents to disaggregate their spending at the destination based on their last trip into seven different spending categories. These spending categories, which have been used in typical tourism economic impact studies, are reported in Table 1. These numbers were then divided by total days of their trip and the number of individuals on the trip. This provided us with expenditures in seven categories on a per person per day basis. An eighth category, “Anything else for this trip,” was disregarded as it would be difficult to assign proper sectors to this expenditure and because 88% of respondents indicated no expenditure in this category.

Table 1.

Per Person per Day Tourist Expenditures (IN U.S. DOLLARS) in Different Categories

Spending Category	Myrtle Beach	Charleston	Hilton Head
Hotel/motel/other lodging	21.09	43.20	29.33
Grocery and retail stores	9.68	11.85	10.68
Restaurants and drinking places	13.01	25.85	13.97
Outdoor recreational activities	2.15	4.29	2.85
Entertainment	2.11	3.01	1.64
Automobile transportation	6.70	10.57	5.66
Other transportation	2.08	9.47	4.08

Step four

Once direct expenditures were determined, relevant margins were applied to the “Grocery and retail stores” and “Automobile transportation” categories. Margining refers to eliminating the costs of a good that immediately leaves the community in order to more accurately measure the expenditure’s impact on final demand for a region. Applying margins is considered necessary to better approximate true economic impact (Crompton 2006). This was done using the margining coefficients for household spending that exist in IMPLAN version 2.

Step five

After margining, the impact analysis was performed. Estimating the number of tourists is often a critical part of performing an economic impact analysis. However for this study’s purpose, the overall size of the economic impact is not important since we were interested in examining ratios (i.e., the distribution of income effects across different income groups). Therefore, an arbitrary number can be chosen based on the fact that the size of the impact does affect the ratios in an IO model. For illustrative purposes, we used 1,000,000 person-days as it is large enough to create employment impacts that are more than 1 across all industrial sectors and are therefore easier to display and analyze. After performing the impact in IMPLAN version 2, employment estimates were obtained for each region in all 440 sectors in IMPLAN.

Step six

A bridge table provided by IMPLAN (2009) was used to convert employment in IMPLAN sectors into NAICS sectors. This was predominately done on the two-digit level with the exception of “Gasoline Stations,” “Accommodations,” and “Food Services and Drinking Places” which are especially relevant to tourist spending. Results of this conversion are shown in Table 2.

Table 2.

Employment by NAICS Categories in Charleston

Industry	NAICS Number	Total Employment
Agricultural, forestry, fish and hunting	11	3.3
Mining	21	2.9
Utilities	22	3.1
Construction	23	10.5
Manufacturing	31–33	6.5
Wholesale trade	42	10.1
Transportation and warehousing	48–49	154.4
Retail trade	44–45	54.7
Gasoline stations	447	30.0
Information	51	9.9
Finance and insurance	52	16.8
Real estate and rental	53	32.0
Professional—scientific and technologic services	54	25.6
Management of companies	55	3.9
Administrative and waste services	56	25.0
Educational services	61	5.7
Health and social services	62	36.9
Arts—entertainment and recreation	71	147.0
Accommodations	721	493.8
Food services and drinking places	722	529.6
Other services	81	31.7
Government and non-NAICs	92	14.5

Step seven

Once the employment estimates for each NAICS category were determined, these estimates were converted into employment estimates by occupation type. This conversion was accomplished by using BLS data that estimates employment by SOC categories for every NAICS sector (U.S. Bureau of Labor Statistics 2009). Employment for each SOC in a given NAICS was first calculated, and then total employment for each SOC was determined by summing the results for that SOC category across every NAICS sector. The SOC covers all jobs in the national economy, including occupations in the public, private, and military sectors” (U.S. Bureau of Labor Statistics 2010, p. ii). At the highest level of detail, 840 different occupations were grouped into 461 broad occupations, 97 minor groups, and 23 major groups. Each industry has customized estimates of the jobs it creates in a large number of occupation categories. This analysis used the broad occupations as its level of detail. This level was chosen as it ensures a high level of accuracy, and several occupations at the higher level of detail (840 categories) had missing data at the MSA level, which would further complicate analysis. As an example, the top 10 jobs generated in Charleston by 100,000 tourist days are displayed in Table 3.

Table 3.

The Top Ten Jobs Created in Charleston

SOC Category	Jobs
Waiters and waitresses	155
Combined food preparation and serving workers, including fast food	131
Maids and housekeeping cleaners	120
Cooks, restaurant	63
Hotel, motel, and resort desk clerks	59
Cashiers	46
First-line supervisors/managers of food preparation and serving workers	41
Bartenders	34
Dishwashers	33
Food preparation workers	31

Step eight

Lastly, the jobs created in each SOC were organized into income brackets (see Figure 1 for methods summary) according to the BLS mean wage for full-time equivalent employment. The BLS extrapolates part-time and seasonal jobs to their full-time equivalent for this estimate. The full-time estimate is useful as individuals may hold multiple part-time and or seasonal jobs during the year, so data sets that do not use full-time equivalents would underestimate the amount of income that goes to the lower-income brackets. Additionally, Daniels, Norman, and Henry (2004) suggest that data using full-time equivalents will create more accurate assessment of the income distribution. This analysis uses the BLS estimate of the full-time equivalent average annual wage of each SOC category in the three regions in question. This is a deviation from the Daniels, Norman, and Henry (2004) approach, which used wage by industry estimates instead of wage by region estimates. Wage by region may be more accurate as it reflects that higher wages are typically paid in more urban areas and better allows us to compare the distribution of incomes from tourist expenditure to the normally occurring distribution. One spatial issue was encountered as the BLS does not provide data for the Hilton Head micropolitan statistical area. Instead, data from the South Carolina Low Country nonmetropolitan area were used. This region includes the Hilton Head micropolitan statistical area (Beaufort and Jasper counties) along with four other rural counties in the area. Although not ideal, this likely had a minimal effect on the results as we are mainly concerned with comparing the ratio of wage categories created versus currently existing but not overall economic impact. Adding the four similar counties should not have a major effect on the income distribution. After matching average wages to occupation types, employment estimates were arranged in $10,000 increments (Note: this should not be considered an accurate depiction of the actual income distribution in the three regions as the data only used the mean wage of each occupation, and did not account for the variation of wages within each SOC). This method is the most apt to make comparisons to the analysis of wage distribution from tourism as it makes the same assumptions as the BLS data used to generate estimates of the income distribution generated by tourism (e.g., using the mean of all occupations and assuming that all jobs are full-time).

Figure 1.

Summary of the methods

Results

An examination of the wage distribution in the three regions (Table 4) showed that Charleston income distribution was the highest followed by the Low Country, and then by Myrtle Beach. After calculating the distribution of jobs by different wage categories generated by tourism expenditures, results indicate that income generated from tourism went predominately to lower-income brackets (Table 5). A comparison of the income distribution from tourism generated jobs to the overall distribution of income suggests that tourism provided more jobs with low wages than the overall economy of the regions. The difference was greatest in Charleston followed by Myrtle Beach and then the Low Country. Perhaps, the most dramatic result is that roughly half of the jobs created by tourism in Charleston and Myrtle Beach earned less than $20,000 a year. All three destinations displayed a higher percentage of jobs in the lowest-income category. The discrepancy was especially apparent in Charleston. The income distribution from tourism in the Low Country was fairly similar to that of what naturally occurs.

Table 4.

Distribution of Jobs by Mean Income of the Occupation from SOC in Currently Existing in Each Region

	Charleston		Myrtle Beach		Low Country (Hilton Head)
	Jobs	Percentage	Jobs	Percentage	Jobs	Percentage
$10,000–$19,999	43,850	16	27,750	29	22,040	21
$20,000–$29,999	70,330	26	41,040	42	34,780	33
$30,000–$39,999	66,490	25	15,190	16	20,620	20
$40,000–$49,999	30,400	11	3,640	4	11,120	11
$50,000–$59,999	18,250	7	4,170	4	6,330	6
$60,000–$69,999	17,240	6	1,120	1	4,210	4
$70,000–$79,999	7,790	3	2,680	3	1,160	1
$80,000–$89,999	2,300	1	440	0	540	1
$90,000–$99,999	6,900	3	70	0	1,990	2
$100,000+	3,940	1	620	1	1,280	1
Total	267,490	100	96,720	100	104,070	100

Table 5.

Distribution of Jobs by Mean Income of the Occupation from SOC in Different Income Brackets from Tourism vs. Overall Distribution in the Three Regions

	Charleston		Myrtle Beach		Low Country (Hilton Head)
	Tourism%	Overall %	Tourism%	Overall %	Tourism %	Overall %
$10,000–$19,999	51	16	47	29	27	21
$20,000–$29,999	19	26	32	42	31	33
$30,000–$39,999	15	25	11	16	23	20
$40,000–$49,999	7	11	3	4	7	11
$50,000–$59,999	4	7	3	4	5	6
$60,000–$69,999	2	6	0	1	3	4
$70,000–$79,999	1	3	2	3	0	1
$80,000–$89,999	0	1	0	0	0	1
$90,000–$99,999	0	3	0	0	2	2
$100,000+	0	1	0	1	1	1

An examination of the tourists’ expenditures and the jobs created by the different NASICS categories demonstrates why the income distribution was so low. The two most common expenditure categories were on hotels and restaurants. The most common jobs for the NASIC “accommodations” category was “maids and housekeeping cleaners,” which made up 31% of total employment in that sector and fall into the lowest-income category. Similarly the most common jobs generated by the NASIC category “Food Services and Drinking Places” were “combined food preparation and serving workers, including fast food,” which make up 42% of total employment in that sector and also fell into the lowest-income category.

Conclusion

This study examined the income distribution of tourism expenditures by converting employment estimates using industrial sectors from IMPLAN into employment estimates by occupational type in the SOC. The average wages of the jobs generated by tourism were then compared to the average wages of the jobs currently existing in the region. Study results indicate that the jobs generated by tourism expenditures had a lower income distribution than the naturally existing income distribution. While using a new methodology, the results are consistent with previous study findings into the issue (Blake et al. 2008; Daniels, Norman, and Henry 2004; Marcouiller, Kim, and Deller 2004). These results may be viewed as a negative for tourism by regional planners and developers.

This is an especially pressing concern in light of the growing gap between the rich and poor in the United States (Autor and Dorn 2009). The results support the hypothesis proposed by Marcouiller, Kim, and Deller (2004) that tourism may be a contributor to the reduction of the middle class in America. Tourism-related jobs may contribute to what economists refer to as “skill-biased technological change” (Machin 2008). This concept holds that recent technological changes have increased the demand for skilled labor more than unskilled labor, and therefore caused an increase in income inequality. Since tourism is often reliant on unskilled labor such as hotel maids and waiters, the jobs created by tourism will likely continue to be low wage in the future. Goos and Manning (2003) refer to this as the creation of “lovely and lousy” (p. 1) jobs, and in fact cite “low-paying service occupations” (p. 3) as the primary source of lousy jobs. If the hollowing out of the income distribution in the United States continues, tourism may even lose its appeal as a means of creating entry-level jobs—as these will be less likely to lead to medium-income jobs. Goos and Manning (2003), however, do offer reasons for supporters of tourism developers to be optimistic. They believe that jobs most at risk of being lost are those that involve repetitive tasks such as clerical and manufacturing jobs. As the demand for these low-skilled laborers may even increase in the future, at least these occupations will still remain viable even if wages for these entry-level tourism jobs remain low.

If more evidence of the low wages of the tourism industry continues to arise, proponents of tourism development may have to change their selling points in the future. Instead of emphasizing only the impact of tourism on job creation, developers may point to the positive externalities that tourism development can create, such as more local recreation and leisure opportunities and cities that are more enjoyable to live in. This in turn may support the rise of a “creative class” that will generate more high-wage jobs in the future (Florida, Mellander, and Stolarick 2008; McGranahan and Wojan 2007). Alternatively, tourism supporters could claim that the low-income distribution is advantageous for the local community as this means that numerous entry-level jobs are created for residents (Szivas, Riley, and Aireya 2003). These jobs rarely require specialized skill sets that could necessitate importing workers from outside the region. As a result, this job creation can reduce unemployment and poverty in the area. Furthermore, regions or countries with high levels of unemployment among the younger segment of the labor market may be especially interested in the type of jobs tourism creates. Tourism may not only provide needed jobs but also the experience necessary to further young individuals’ careers. In contrast, regions that need to generate employment opportunities in the older segments of the population may not find tourism to be suitable to their needs.

The study faces several limitations. First, while tourism is noted as having a high percentage of part-time and seasonal jobs, the BLS data only assumes full-time and year-round employment. Tourism has a high percentage of jobs that are seasonal and/or part time, and this imbalance may contribute to this study likely underestimating the amount of income that goes to the lower-income brackets as the tourism industry has a higher percentage of individuals who may be unable to find full-time year-round employment. Second, the BLS keeps some job and income data confidential in cases in which one firm comprises the vast majority of a particular sector; this results in data (number of jobs) being lost in the conversion process. Finally, it should be also noted that the implications of this analysis may not be relevant for developing countries. Tourism jobs may actually require relatively high levels of education and be high paying in these economies. Accordingly, this type of analysis may need to be adapted to apply in the developing world because of the lack of available data as well as different economic conditions.

In conclusion, we would suggest that more data be collected and analyzed to examine an existing trend in our data. Our results indicate that the more urban the destinations, the lower the wage distribution from tourism will be. Results from this study show that income distribution from tourism is fairly similar to the naturally occurring income distribution in sparsely populated Hilton Head, and very different in the urban Charleston. With only three cases, it is difficult to determine whether this pattern is happenstance or if there is actual correlation. Additionally, we suggest that the method outlined by Daniels, Norman, and Henry (2004) remains underutilized. Understanding the types and quality of jobs created by tourism is an important issue that has not received enough attention and more macro-level data and analysis is needed to fully understand this issue.

Footnotes

Acknowledgements

We would like to thank the entire staff of Clemson’s International Institute for Tourism Research and Development for their immense support throughout the entire project. Additionally, we would like to thank our reviewers for their helpful comments. This report was prepared by Clemson University under NA06OAR4170015, Am. 9. The statements, findings, conclusions, and recommendations are those of the author(s) and do not necessarily reflect the views of the South Carolina Sea Grant Consortium or the National Oceanic Atmospheric Administration.

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.

Bios

R. Geoffrey Lacher is a Ph.D candidate in the Parks, Tourism, and Recreation Management Department at Clemson University. His research interests include the economics of tourism, consumer demand for sustainable tourism certifications, tourism in the developing world, and the economics of souvenirs.

Chi-Ok Oh is an assistant professor in the Department of Community, Agriculture, Recreation and Resource Studies at Michigan State University. His research interests are in tourism and recreation economics and human dimensions of natural resources, with a particular interest in estimating social and economic benefits of tourism activities and related natural resources.

References

Autor

David H.

Dorn

David

. (2009). “The Growth of Low Skill Service Jobs and the Polarization of the U.S. Labor Market.” NBER Working Paper No. 15150.

Blake

Adam

Arbache

Jorge Saba

Sinclairc

M. Thea

Teles

Vladimir K.

(2008). “Tourism and Poverty Relief.” Annals of Tourism Research, 35 (1): 107-26.

Bonn

Mark A.

Harrington

Julie

. (2008). “A Comparison of Three Economic Impact Models for Applied Hospitality and Tourism Research.” Tourism Economics, 14 (4): 769-89.

Choy

Dexter J. L.

(1995). “The Quality of Tourism Employment.” Tourism Management, 16 (2): 129-37.

Courant

Paul N.

(1984). “How Would You Know a Good Economic Development Policy If You Tripped over One? Hint: Don t Just Count Jobs.” In Tax Policy in the Real World, edited by Slemrod

Joel

. Cambridge: Cambridge University Press, pp. 249-68.

Crompton

John L.

(2006). “Economic Impact Studies: Instruments for Political Shenanigans?” Journal of Travel Research, 45 (1): 67-82.

Crompton

John L.

(1999). Measuring the Economic Impact of Visitors to Sports Tournaments and Special Events. Ashburn, VA: National Recreation and Park Association.

Daniels

Marget J.

Norman

William C.

Henry

Mark S.

(2004). “Estimating Income Effects of a Sport Tourism Event.” Annals of Tourism Research, 31 (1): 180-99.

Deller

Steven C.

Lledo

Victor

Marcouiller

David W.

(2008). “Modeling Regional Economic Growth with a Focus on Amenities.” Review of Urban Regional Development Studies, 20 (1): 1-21.

10.

Dillman

Don A.

(2000). Mail and Internet Surveys: The Tailored Design Method. New York: John Wiley.

11.

Dwyer

Larry

Forsyth

Peter

Spurr

Ray

. (2005). “Estimating the Impacts of Special Events on an Economy.” Journal of Travel Research, 43 (4): 351-59.

12.

Fleischer

Aliza

Tchetchik

Anat

. (2005). “Does Rural Tourism Benefit from Agriculture?” Tourism Management, 26 (4): 493-501.

13.

Florida

Richard

Mellander

Charlotta

Stolarick

Kevin

. (2008). “Inside the Black Box of Regional Development—Human Capital, the Creative Class and Tolerance.” Journal of Economic Geography, 8 (5): 615-49.

14.

Goos

Maarten

Manning

Alan

. (2003). “Lousy and Lovely Jobs: The Rising Polarization of Work in Britain.” Labour, 22 (1): 7-26.

15.

IMPLAN. (2009). Bridge Implan 509 To BEA NAICS 1997 NAICS 2002. IMPLAN.com (accessed January 10, 2011).

16.

Johnson

Kenneth M.

Stewart

Susan. I.

(2005). “Amenity Migration to Urban Proximate Counties.” In Amenities and Rural Development: Theory, Methods and Public Policy, edited by Green

Gary P.

Deller

Steven C.

Marcouiller

David

. Cheltenham, UK: Edward, pp. 177-196.

17.

Lee

Choong-Ki

Kang

Seyoung

. (1998). “Measuring Earnings Inequality and Median Earnings in the Tourism Industry.” Tourism Management, 19 (4): 341-48.

18.

Leiper

Neil R.

(2008). “Why ‘the Tourism Industry’ Is Misleading as a Generic Expression: The Case for the Plural Variation, ‘Tourism Industries.’” Tourism Management, 29 (2): 237-51.

19.

Liu

Abby

Wall

Geoffrey

. (2006). “Planning Tourism Employment: A Developing Country Perspective.” Tourism Management, 27 (1): 159-70.

20.

Machin

Stephen

. (2008). “An Appraisal of Economic Research on Changes in Wage Inequality.” Labour, 22 (1): 7-26.

21.

Mack

Elizabeth A.

Grubesic

Tony H.

(Forthcoming). “All Jobs Are Not Created Equal: Divergent Indicators in the Knowledge Economy.” Applied Geography.

22.

Marcouiller

David W.

(2007). “Rural Tourism Promotion as Public Policy: Panacea or Pandora’s Box?” Journal of Regional Analysis and Policy, 37 (1): 28-31.

23.

Marcouiller

David W.

Kim

Kwang-Koo

Deller

Steven C.

(2004). “Natural Amenities, Tourism and Income Distribution.” Annals of Tourism Research, 31 (4): 1031-50.

24.

Marcouiller

David W.

Xia

Xianli

. (2008). “Distribution of Income from Tourism-Sensitive Employment.” Tourism Economics, 14 (3): 545-65.

25.

McGranahan

David A.

Wojan

Timothy R.

(2007). “The Creative Class: A Key to Rural Growth.” Amber Waves, 5 (2): 16-21.

26.

Miller

Ronald E.

Blair

Peter D.

(2009). Input-Output Analysis: Foundations and Extensions. 2nd edition. Englewood Cliffs, NJ: Prentice Hall.

27.

Ramjee Singh

Wright

Hayle

Craigwell

(2010). “Is the Tourism-Led Growth Thesis Valid? The Case of the Bahamas, Barbados, and Jamaica.” Tourism Analysis, 15 (4): 435-45.

28.

Reeder Richard

Brown

Dennis M.

(2005). “Recreation, Tourism, and Rural Well-Being.” ERR-7, US Department of Agriculture, Economic Research Service. Washington, DC: US Government Planning Office.

29.

Richards

Greg

Wilson

Julie

. (2004). “The Impact of Cultural Events on City Image: Rotterdam Cultural Capital of Europe 2001.” Urban Studies, 41 (10): 1931-51.

30.

Samdahl

Daine M.

(1999). “Epistemological and Methodological Issues in Leisure Research.” In Leisure Studies: Prospects for the Twenty-First Century, edited by Burton

Edgar L.

Jackson

Thomas L.

State College, PA: Venture, pp. 119-33.

31.

Seckelmann

Astrid

. (2002). “Domestic Tourism—A Chance for Regional Development in Turkey?” Tourism Management, 23 (1): 85-92.

32.

Seetanah

Boopen

. (2011). “Assessing the Dynamic Economic Impact of Tourism for Island Economies.” Annals of Tourism Research, 38 (1): 291-308.

33.

Spotila

John T.

(2000). “Standards for Defining Metropolitan and Micropolitan Statistical Areas.” Federal Register, 65 (245): 82228-38.

34.

Szivas

Edith

Rileya

Michael

Aireya

David

. (2003). “Labor Mobility into Tourism Attraction and Satisfaction.” Annals of Tourism Research, 30 (1): 64-76.

35.

U.S. Bureau of Labor Statistics. (2009). “May 2009 National Occupational Employment and Wage Estimates.” BLS.gov (accessed January 10, 2011).

36.

U.S. Bureau of Labor Statistics. (2010). “2010 SOC User Guide.” http://www.BLS.gov (accessed January 10, 2011).

37.

U.S. Travel Association. (2009). “The Economic Contribution of Tourism to South Carolina.” http://www.hrsm.sc.edu (accessed January 10, 2011).

38.

U.S. Travel Association. (2010). “The Economic Contribution of Tourism to South Carolina.” http://www.BLS.gov (accessed January 10, 2011).

39.

Waits

Mary Jo

. (2000). “The Added Value of the Industry Cluster Approach to Economic Analysis, Strategy Development, and Service Delivery.” Economic Development Quarterly, 14 (1): 35-50.