|Home > Draft Report > SOCIO-5|
As in our analysis of wood products, we first examined the direct tourism-related jobs over time and by State. We then estimated the direct, indirect, and induced effects of forest-based recreation in 1997. The analysis of the role of forests in recreation- and tourism-based employment and income is hampered by the lack of information on exactly how much of the local economy derives from recreation and tourism (Kass and Okuba, 2000). Unlike the wood products sectors, where data are collected in categories that relate closely to forests and forestry, expenditures by visitors to forests are lumped together with expenditures by residents and other travelers for such items as eating and lodging.
As noted above, in the State-level analysis, we used lodging and eating places to proxy for tourism-related industries. For 1996, we developed a measure of outdoor-recreation-based tourism at the county level, and compared this to the totals from the lodging and eating places. The correlation was quite high (>.98) and significant, and the rankings were similar. Thus, we concluded that the time-series of overall tourism was an adequate proxy for the actual, but unobtainable time series of outdoor-recreation-based tourism.
In contrast, the indirect effects were more precisely modeled using three different techniques (see below for a complete discussion of these methods). Thus, the discussion of the time-series direct jobs and income in tourism-related sectors is not directly comparable to the estimates of direct, indirect and induced effects of forest-based and outdoor recreation-based tourism.
Few forest-based recreation activities generate direct income for public land, but some forest-based recreation on private land generates income for the landowner. For example, hunting clubs often lease hunting rights from private landowners. For recreation on both public and private land, the major economic impact is the money spent in local communities by recreationists. As a result, the recreation analysis is very different from the timber analysis. Rather than tracking a physical commodity through several processing steps, we trace the impact of a nonmaterial forest output--the opportunity to recreate--to the final consumer. There are no forward linkages, in the market sense, from the forest to the final consumer. There are only backward linkages from the recreation consumer to the producers of the supplies the consumer buys.
Recreation output from the forest is nonmaterial--it is the setting that is provided. As this output is not being processed in any way, we have no sales value for secondary processing industries as we did for timber. Therefore, to measure the economic impact of recreation activities, we estimated what recreationists purchased in local economies.
The percentages of all southern jobs that are in the hotel and lodging and the eating and drinking place sectors have increased in all 13 southern States (Figure 10). Percentages for Mississippi and Louisiana reflect significant increases in the early 1990s, likely due to changes in State gambling laws. Similar increases occurred in wages and salaries (Figure 11) between 1969 and 1998. Florida had the largest concentration of tourism-related jobs and income, exceeding 7 percent in 1998. There is much less variation by State in tourism-related jobs and income than in wood products jobs and income. Tourism-related jobs are 5 to 6 percent of all jobs, and 3 to 6 percent of income is in tourism-related sectors. Unlike in the wood products sector, these sectors represent a larger share of jobs than of income. This result implies a lower-than-average income per job for tourism-related compared to a higher-than-average income per job for wood products. It is generally accepted that tourism jobs have lower hourly wages than manufacturing jobs. However, because the jobs in this dataset do not represent full time equivalents (40 hour weeks), the lower income per job could also be due to part-time work, rather than lower hourly wages.
We used three different methods to estimate total outdoor or forest-based recreation impacts in the South. These methods give us a range of impacts, with a low of 317 million forest visitor-days and a high of 1,268 million visitor-days. The first method is based on the most recent National Forest System (NFS) estimates of visits to Region 8 in 2000 (USDA Forest Service 2001). This method includes expenditures made for durables, nondurables, and services within 50 miles of the recreation site. We allocated NFS recreation visits using two different methods, participation from NSRE (NFS-P) and land area in National Forests (NFS-L). The second method uses the national Travel and Tourism Satellite Accounts (Kass and Okuba, 2000) (TTSA) to attribute output to travel and tourism, then estimates forest-based proportions using recent study results that outdoor recreation comprises 19 percent of all leisure tourism visits and 33 percent of all leisure tourism expenditures (Pennsylvania Department of Conservation and Natural Resources 1999). This method does not include durables expenditures, but includes all other expenditures for outdoor recreation-related tourism. The third method also uses the 19 and 33 percentages, but applies them to State level estimates of total travel and tourism outputs from the Travel Industry Association of America (TIA) (1999). Similar to the TTSA method, this method does not include durables purchases, accounts for all other purchases regardless of where made, and includes expenditures from all outdoor recreation, not just forest-based outdoor recreation. These are further discussed below.
NFS-methods: National Forest visits in the southern region were estimated at 24,869,000 for 2000 (Personal Communications Don English, USDA Forest Service, Athens, GA.). The NFS-P method assumed that the proportion of visits to public land was equal to the proportion of activity-days occurring on public land in the NSRE (56 percent). Further, the percent of visits to national forest land was equal to the proportion of public land managed by the National Forests (30 percent). This approach resulted in an estimated 17 percent of the recreation visits occurring on NFS land, and thus the remaining 83 percent occurred on private and other public lands (148,115,474 visits). (Table 5) For the NFS-L method we assumed that all forests were visited in proportion to their acreage in the South, so we divided the NFS visits by the percent of all forest land in national forests (6 percent), resulting in 410,043,596 total forest visits.
Visits are multiday trips, so we adjusted the visit estimates using trip lengths from the CUSTOMER survey (available from Ken Cordell, USDA Forest Service, Athens, GA) and activity allocation from NSRE to get total days of forest visits. To get days, we used a weighted average length of trip for nonresidents and assumed a single-day visit for residents. The weights were based on the proportion of total trips that were a single type, such as camping, using the NSRE data on participation. The average of 2.14 days per trip resulted in an estimate of the total number of forest-based recreation days of 317,123,332 for the NFS-P method and 878,448,994 for the NFS-L method.
These activity estimates were multiplied by the response coefficients for direct and total impacts derived from IMPLAN. We used expenditure profiles detailing what people spent on various activities from two previously developed surveys, the Public Area Recreation Visitor Survey (PARVS, available from Ken Cordell, USDA Forest Service, Athens, GA) for recreation, and the US Fish and Wildlife Service (FWS) surveys for hunting and fishing, both in dollars of expenditures per person per day (U.S. Fish and Wildlife Service). The response coefficients for the recreational activities (developed camping, mechanized travel, other recreation, trail use and winter activities) were developed using PARVS expenditures bridged to IMPLAN sectors. Hunting and fishing response coefficients were developed by bridging FWS survey data to IMPLAN sectors. These profiles include expenditures within 50 miles for PARVs and within the State for FWS, for both residents and nonresidents. Separate coefficients were estimated for residents and nonresidents. Impacts for residents are substantially lower than for nonresidents.
For both scenarios, allocations to individual forest-related activities were based on the percentages from NSRE. Table 6 shows the number of forest visitor-days for both NFS-L and NFS-P. Mechanical travel (resident and nonresident), other (resident), trail use (resident and nonresident), fresh water fishing (nonresident, and other (nonresident), are the largest in number of visitor-days.
The direct and total impacts by activity are shown in Table 7 for NFS-P and Table 8 for NFS-L. Direct jobs range from 136,944 to 379,116 and total jobs (direct plus indirect plus induced) range from 254,591 to 704,812 jobs. Direct contribution to GRP ranges from $3,805 to 10,533 million, while total GRP from recreation ranges from $9,350 to $25,886 million for this method.
TTSA methods: The second method relies on the Travel and Tourism Satellite Accounts (TTSA) for most data (Kass and Okuba, 2000) supplemented with IMPLAN data. IMPLAN response coefficients for each of the affected sectors were used. The TTSA uses national-level data on consumer expenditures and the national input-output tables to attribute demand to tourism. Only travel further than 50 miles from home is represented, so the data were adjusted using the percentages of resident and nonresident travel from the NSRE. The TTSAs estimate foreign and domestic nonresident leisure tourism, as well as business tourism. Table 9 lists the sectors that are assumed to be influenced by tourism. We used the percentage of each sector that was attributed to leisure tourism and applied that percentage to total southern output (from IMPLAN) from those sectors to estimate southern leisure tourism. Leisure tourism is determined from the proportion of industry output that is purchased by tourists more than 50 miles from home.
We then used two different levels to represent the proportion of outdoor recreation expenditures, 19 and 33 percent. These percentages were derived from a study of outdoor recreation tourism in Pennsylvania (Pennsylvania Department of Conservation and Natural Resources 1999). If the primary purpose of the vacation was outdoor recreation and involved overnight travel or travel further than 50 miles from home, then the vacation was considered an outdoor recreation vacation. The study estimated that 59 percent of all travel included some form of outdoor recreation, but that only 19 percent had outdoor recreation as the primary purpose. The study also found that outdoor recreation travel is increasing faster than other forms of travel, and that outdoor recreation travelers spend more per person per trip than the average leisure traveler. While this study was conducted for a different ecoregion and a single State, other similar research was not found. We therefore used two of the numbers from this study: 19 percent of all travelers are outdoor recreation travelers and 33 percent of all expenditures are made by outdoor recreation travelers. Those numbers represent the high and low bounds of the TTSA and TIA methods.
Table 10 and Table 11 show the direct and total effects by sector assuming either the 19 or 33 percent in outdoor recreation. Direct employment ranges from 212,193 jobs to 427,317 jobs and direct GRP ranges from $6,145 to $11,555 million. Total employment (direct plus indirect plus induced) ranges from 379,373 to 748,094 and total GRP from $13,492 to $25,624 million dollars. The largest impacts are from the airline, eating and drinking, hotel and lodging and recreation and entertainment sectors.
TIA method: The third method used the TIA report for 1997 in combination with inputs from the Pennsylvania DCNR and the TTSAs for 1997. The Travel Industry Association of America developed impacts for travel by State for 1997 (Travel Industry Association of America 1999) using an input-output model. The results include total impacts for expenditures, payroll and employment. TIA travel includes only travel further than 50 miles from home, so the data were adjusted using the percentages of resident and nonresident recreators from the NSRE. We applied the percentage of tourism that is leisure tourism (from the TTSA) and the percentage attributable to outdoor recreation (19 and 33 percent, from the PA DCNR study). These percentages were also adjusted by the proportion of the State in forest to account for the differences in largely unforested States such as Texas and Oklahoma. These latter two States had the lowest percent of tourism in outdoor recreation related tourism (Table 12), while Alabama, Georgia and Mississippi had the highest rates. Table 12 also has the TIA data for all tourism direct expenditures, payroll and employment.
Table 13 shows the direct and total effects from applying both the 19 and 33 percent of tourism as outdoor recreation related. Direct effects jobs range from 276,000 to 480,000 and expenditures from $16 to $28 billion. Total jobs range from 579,000 to 1,006,000, and total expenditures range from S38.5 to $66.9 billion. Total values were derived by using the multipliers developed at the national level for the TIA report.
Summary: Table 14 compares the six estimates and also estimates the number of visitor-days associated with the TTSA and TIA methods. The relationship between jobs and visitor-days in the NFS methods was used to calculate the visitor days associated with the TTSA and TIA methods. Estimated visitor-days, and the other economic measures of jobs, income, etc., are ordered similarly, with the NFS-P method generating the lowest economic contributions, followed by the TTSA-19 and TIA-19 methods, and NFS-L, TTSA-33 and TIA-33 generating the highest contributions. Direct effects are 0.2 to 1.2 percent of total southern employment and 0.13 to 0.61 percent of total southern GRP (GRP is not available from the TIA methods). Total effects range from 0.38 to 2.62 percent of employment and 0.32 to 1.35 percent of GRP.
The USDA Forest Service recently released revised estimates of national forest visits based on a survey (USDA Forest Service 2001). These estimates will be prepared each year for all national forests. Estimates for the Southern Region, which were used in the NFS methods above, were 24.9 million visits in 2000. Table 15 shows the estimated visits and land area for each of the regions and for the United States as well as the visits per acre. This rate of visitation is highest in the Eastern Region, followed by the Southern Region at 1.89 visits per acre. These numbers are an indication of relative resource scarcity of national forest land for recreation. At this time, the bulk of the national forest land is located distant from most of the population, thus limiting its usefulness in alleviating this scarcity.
Participation in recreational activities has been projected to increase in the South (see Chapter SOCIO-6). It is likely that this recreation will be concentrated on Federal and State parks, forests, and coastlines. As such, these increases in participation will likely lead to increased jobs in areas with public recreation lands. Increases in labor productivity will occur in the leisure service sectors, but they are likely to be small relative to total output. Thus, labor will continue to be a major input into production of these services. One aspect of recreational services that could change in the South is a potential increase in manufacturers of recreation products, leading to an increase in retention of backward linkages within the region, involving both returns to capital and to labor. We expect the proportion of the southern economy in outdoor-recreation enterprises to continue to increase, comparable to increases in the national economy.
Discussions of the forest-based economy often center around the relationship between the wood products and the recreation and tourism sectors because both depend on the existence of forests (see Morton 1994, Schallau 1994). While the relationship between the two uses may be obvious at an individual site, the landscape level effects of these activities on the economy are not clear. The substitution of one site for another in both recreation and wood products will lead to geographic shifts in economic costs and benefits, but may or may not represent an economic loss. To our knowledge no systematic study of the joint production aspects of the forest landscape in supporting both the wood products and recreation/tourism industries has been conducted.
While much of the past controversy centers around public land, the management of private forests is becoming more controversial. Landowners and recreationists have similar perceptions about general forest management, but differing perceptions about harvesting activities (Theodori and others 2000, Marcouiller and Mace 1999). These differences also occur when comparing second homeowners with local residents (Marcouiller and others 1999).
Another source of discussion regarding the two forest uses is the disparity between the average annual incomes from the two sectors (Table 16). The wood products average is higher than the southwide economy average, which is higher than the average of the three recreation methods used. Income per job (not a wage rate) ranges from less than $5,000 per year for timber to over $52,000 per year for pulp and paper. The low level for timber production may reflect the part-time nature of this work. GRP per job, also shown in table 16, is highest for pulp and paper (over $82,000 per year) and lowest for wood furniture and recreation (about $33,000 per year).
Recreation and wood products contribute to the local community by providing jobs and income. These are generally viewed as positive effects of development. However, both recreation and wood products development, on either public or private forest land, have the potential for negative effects on the local community. Murdy and others list some of these as host:tourist conflicts, crime, overcrowding, migration, and loss of family traditions. These and other quality-of-life issues must be balanced with the more traditional standard-of-living issues (jobs and income) when considering any changes in forest management.
|Glossary | Sci.Names | Process | Comments | Final Report|