Paper on THE STRUCTURE AND LEVEL OF ENTRANCE FEE IN SERENGETI NATIONAL PARK IN TANZANIA. Gerald Kibira 1

Transcription

1 Paper on THE STRUCTURE AND LEVEL OF ENTRANCE FEE IN SERENGETI NATIONAL PARK IN TANZANIA Gerald Kbra 1 Abstract Whereas most valuaton studes have provded useful nsghts of protected areas and ther beneft n developng countres, they have typcally focused more on beneft sharng than on how the adjacent local communty wll beneft from wldlfe conservaton and also they have typcally focused more on estmatng consumer surplus rather than on evaluatng user fees as a gude toward desgnng mproved park prcng strateges whch s the focus of ths paper. Many of the vstors to protected areas, such as natonal parks, are foregn toursts who ncur few of the costs but enjoy many of the benefts stemmng from resource conservaton efforts. Toursm revenues, rather than beng earmarked for park mantenance or resource conservaton efforts, are frequently merged wth other sources of general revenues. Fundng conservaton actvtes n the Serenget s a bg problem whch requres revenue management by natonal park that wll ensure sustanablty of the socal ecologcal system. Serenget Natonal Park s regarded as one of the major attracton to the foregn toursts. We appled a twst n the Contngent Behavor (CB) methodology n the context of a developng country, whch has never been appled n lterature. We fnd out that as the entrance fee ncreases toursts tend to swtch to substtute park whch s the Maasa Mara Game Reserve n Kenya and the demand s elastc. In realty, the park agency s nept to charge the revenue maxmzng prce because of the competton from other parks, both locally and regonally. Nevertheless, the fact that we found that the fees could be ncreased sgnfcantly over and above the current fees to maxmze the revenue collecton s mportant. Keywords: Serenget, foregn, toursm, entrance fee, revenue, Contngent Behavor. JEL Codes: Q 51, Q57 1 PhD Student, School of Economcs, Unversty of Cape Town, South Afrca. Moble: Emals: geraldkbra2000@yahoo.co.uk / geraldkbra@gmal.com,

2 1. Introducton Conservaton of nature s mportant for both utltaran and ntrnsc values. Every year around $6 bllon s spent on more than 100,000 protected areas around the world (James et al. 1999, 2001; Pearce 2005). Conservaton n many of the Afrcan countres dffers n terms of approaches but t s generally accepted that no matter what system s adopted all am at protectng the resources n an optmum condton, as would be practcably possble through applcaton of the most contemporary acceptable methodologes. All am to balance development that assures acceptable levels of resource mpacts whle takng nto consderaton beneft to local communtes. Proactve communtes and the prvate sector are key dmensons to ensure ths achevement n the real sense. Presently, conservaton ams at enhancng satsfacton of toursts through ncreasngly dversfed actvtes at a hgh qualty wth very mnmum negatve mpact to the resources. The largest challenge, however, s to balance resources utlzaton wth development of dfferent facltes n lne wth communty needs. A growng body of lterature has emphaszed the role of user fees n the management of natonal parks and protected areas, prmarly n developed countres (Chase et al 1998, Alpzar 2006, Krug et al 2002). In developng countres seekng to balance envronmental and economc growth objectves, the challenges facng polcymakers are partcularly great. Government funds are typcally n short supply and enforcement of envronmental regulatons s neglgent or nonexstent. Many of the vstors to protected areas, such as natonal parks, are foregn toursts who ncur few of the costs but enjoy many of the benefts stemmng from resource conservaton efforts. Toursm revenues, rather than beng earmarked for park mantenance or resource conservaton efforts, are frequently merged wth other sources of general revenues. Yet, wthout user fees to effectvely capture eco-toursm revenues, alternatve land uses that provde greater short-run returns-such as loggng, agrculture, and cattle g razng-wll often be pursued on publc as well as prvate lands (Chase et al 1998). The result s often deforestaton, sol eroson, watershed degradaton, and rreversble loss of bodversty (Southgate and Whtaker 1994).

3 Snce the market fals to reflect these envronmental costs, fewer wldernesses are preserved than s optmal (Dxon and Sherman 1990). Under these crcumstances, the potental benefts from chargng user fees to natonal parks are sgnfcant. User fees are one tool to capture for the publc the benefts of ecotoursm whch often accrue prmarly to the prvate sector. They also can be used to reduce vstaton n areas that suffer from overuse and accompanyng ecologcal damage. Some expermentaton has occurred (Bamford et al. 1988), but these opportuntes have been lmted because of the practcal dffcultes n convncng polcy makers to allow for such experment. In countres where the majorty of the populaton lves n poverty, peoples wllngness to pay (WTP) for conservaton s generally low, so alternatve strateges to fnance conservaton are urgently needed to reduce or stop the current rate of bodversty loss (Krug et al 2002). Studes show that socetes n rch natons are concerned about wld speces and habtats n Afrca and are wllng to pay for ther protecton. The challenge s to create mechansms and nsttutons to capture ths form of nternatonal WTP and to channel these values nto Afrcan conservaton (Krug et al 2002). Achevng conservaton requres mtgatng problems such as natural resource consumpton, and human populaton growth whch requres solutons that address these problems smultaneously. In fndng solutons for conservaton, there are often dffcult trade-offs between conservaton and other socal goals such as development, and between anthropocentrc and bo centrc goals. One example of an area where such trade-off are thought to exst s the Serenget 2 ecosystem. Serenget Natonal Park s the part of the larger Serenget ecosystem. It s managed by Tanzana Natonal Parks and has the area of 14,763 km 2. 2 The name "Serenget" s an approxmaton of the word used by the Maasa to descrbe the area, srnget, whch means "the place where the land runs on forever."

4 However, governments n poor countres such as Tanzana cannot afford to fully fnance conservaton due to budgetary constrants. There s an ncreasng pressure on park managers to fnd alternatve fnancng mechansms. They need fnancal resources for general park management. They need fnancal resources for sharng wth adjacent communtes as compensaton for ther role n supportng conservaton. There are sgnfcant opportuntes for generatng more revenues by reformng the tarff structures for amentes produced by natonal parks. However lttle experence exsts, partcularly n these developng countres, to gude park managers n desgnng effectve prcng strateges for protected areas. In our context there s a need to help the Serenget Natonal Park to desgn an approprate structure and level of entrance fee that they should mplement hence the objectve of ths paper. Ths paper uses a contngent behavor (CB) methodology to generate expermental data to assess the effects of dfferental prcng of entrance fees to Serenget natonal parks n Tanzana.

5 2. Ecotoursm n Tanzana Ecotoursm s becomng more popular among the toursts. Ecotoursm presents developng countres wth ncreasng opportuntes for mprovng resource conservaton and economc growth, but also rases management and polcy challenges. Afrca hosts some of the world s most bodversty rch areas, but many Afrcan countres fnd t dffcult to mantan ther bodversty due to ncreasng land use pressures and nsuffcent conservaton funds and Tanzana s no exceptonal. The contnung converson of natural habtat to agrculture and other uses s the man reason for the dramatc loss of bologcal dversty. Faced wth many socal and economc problems, governments are forced to cut ther envronmental budgets n favour of other prortes (Mnstry of Natural Resources and Toursm 2002). Tanzana s rchly endowed n natural beauty, but most of ts nhabtants are poor that s 28% of the populaton lve below the natonal basc needs poverty lne (Tanzana household budget survey 2011/12). Toursm presents an opportunty to captalze on the world-class concentratons of wldlfe. Toursm offers the potental to earn substantal ncome from these resources. Whle developng countres lke Tanzana may lack ndustral and technologcal captal that other countres have used to develop ther economes, ts natural captal can be used as a comparatve advantage. The Tanzanan government recognzes the potental the country holds; the Mnstry of Natural Resources and Toursm has stated, Tanzana s exceptonally well endowed wth the type of natural resources that provde the raw materal from whch toursm attractons and products are fashoned (Mnstry of Natural Resources and Toursm 2002). The global toursm ndustry s sze alone makes t mpossble for any country to gnore as a possble ncome generatng actvty; clams dffer, but toursm s generally agreed to be n the top fve or sx sectors n terms of overall value worldwde. Conservaton of eco-systems and toursm development n all areas desgnated as natonal parks s the core busness of the Tanzana Natonal Parks. Nature-based or wldlfe toursm s the man source of ncome that s ploughed back for management, regulaton and fulflment of all organzatonal mandates n the natonal parks. The prmary role of Tanzana Natonal Parks s conservaton.

6 The 16 natonal parks whch are n Tanzana, many of whch form the core of a much larger protected ecosystem, have been set asde to preserve the country s rch natural hertage and to provde secure breedng grounds where ts fauna and flora can thrve, safe from the conflctng nterests of a growng human populaton. TANAPA s partcularly charged wth functons of: Protecton of natural resources, park facltes and toursts vstng the parks, Park management and development, Ecologcal and wldlfe health montorng, Toursm development and Communty nvolvement n conservaton efforts. It s the responsblty of TANAPA to ensure securty and safety of vstors, park nhabtants, wldlfe and park nfrastructures. Although securty s the responsblty of every park nhabtant, the Organzaton has rangers who are drectly responsble for ths. To ensure that rangers are equpped wth the rght sklls to execute ther dutes, dfferent knds of tranng are conducted and equpment procured to ease ther tasks. Tranng focuses on wldlfe protecton, strategc feld patrol methods, ntellgence gatherng operatons and use of modern ntellgence and securty equpment. Average annual growth of tourst numbers n Tanzana for the fve years from 2008/2009 to 2012/2013 was 55,712 whch s 8.1 %. Annual revenue growth for the same perod was TZS bllon, whch s 10.9 %. Ths has been made possble through efforts made by the Organzaton to market the unque attractons the country s endowed wthn ts natonal parks as well as ever mprovng customer servces offered by TANAPA employees and other stakeholders n the hosptalty ndustry. In 2012/2013, the number of toursts recorded n our Natonal Parks was 901,892, of whch 537,675 were foregn toursts and 364,217 were locals. Lkewse, revenue generated was TZS bllon n the year 2012/2013 whch s an ncrease of 0.04% compared to the prevous year s revenue whch was TZS bllon.

7 Parks arrvals hghlghts Table : Number of tourst arrvals to parks for the year 2012/2013 S/N PARKS FOREIGN DOMESTIC TOTAL 1 ARUSHA GOMBE KATAVI KILIMANJARO KITULO LAKE MANYARA MAHALE MOUNTAINS MIKUMI MKOMAZI RUAHA RUBONDO ISLAND SAADANI SERENGETI TARANGIRE UDZUNGWA MOUNTAINS SAANANE TOTAL 537, , ,892 Source; Tanzana Natonal Parks Authorty (2014) The Serenget s one of the most famous natonal parks and tourst destnaton n the world, wth 150,000 vstors n the fnancal year Ths toursm contrbutes consderable economc beneft to the naton, wth revenues from SENAPA amountng to Tsh 5,500 mllon for 2003/4. In terms of revenue generaton, the Serenget Natonal Park s second to only Klmanjaro Natonal Park, and together the two parks account for about 77% of TANAPA s total revenue (Gereta et al., 2003). The neghbourng dstrct authortes also beneft through tax revenue and leves; for example, the Serenget Dstrct Authorty collects about Tsh 1,100 mllon per annum from fve hotels/ camps operatng wthn the Park. Tourst shops n the Park generate a further Tshs 4 mllon per annum n terms of taxes (Gereta et al. 2003).

8 The communtes neghbourng the Park receve drect economc beneft predomnantly through employment wth SENAPA departments and tourst facltes; both wthn the Park and the neghbourng communty areas. Another area of drect economc beneft comes from the SENAPA Revenue Sharng Programme, whch provdes a percentage of Park fees through the Support to Communty Intated Projects (SCIP). In lne wth TANAPA s Support for Communty Intated Projects (SCIP), the communtes adjacent to the Park are gven techncal, materal and fnancal assstance to mplement communty based development projects, such as the constructon of school, health and water facltes and feeder roads to vllages. The total value of the SCIP projects contrbuted to by the Park snce the start of the scheme n 1992 s about US$ 1,082,000(Document on Serenget Natonal Park general Management Plan ). The SENAPA Communty Outreach Department also has a very strong and actve conservaton educaton programme, whch encourages and supports conservaton clubs n over 70 prmary schools surroundng the Park. In partcular, SENAPA provdes educatonal materals for students ncludng maps, posters and actvty books and holds regular teacher tranng workshops to ncrease teachers ablty to convey conservaton deas to ther students.

9 Map of Serenget Natonal Park showng delmtaton of the study area. Source; Kdeghesho, JR, (2006)

10 3. Methodology Motvatng the Contngent Behavour Chase et al (1998) argued that; there s an emergng, though growng body of lterature that focuses on valung ecotoursm and wlderness areas n developng countres, the prmary approaches used n these studes were travel cost (TC) and contngent valuaton (CV). Where data constrants are typcally greater and elctng consumers valuatons of envronmental amentes has proven more problematc (Schultz et al. 1997). Chase et al. (1998) further argued that, the travel cost approach derves a demand curve for recreatonal use values n a specfc protected area based on travel expenses for a vacaton n that park. It s expected that there wll be more toursts vstng from nearby and fewer comng from greater dstances where travel costs are hgher. Thus a demand curve for a park can be derved based on the relatonshp between travel costs and the correspondng amounts of vstaton. Ths method, though long used n developed countres, has lmtatons, partcularly n applcatons to multple destnaton trps (Pearse 1968). In addton, the assumpton that vstors from each orgn are homogeneous n margnal costs and preferences and other assumptons of the method are questonable (Wennergen 1964). To crcumvent such lmtatons studes that have estmated use values of protected areas n developng countres have often excluded non-resdents (Durojaye and Ipk 1988; Tobas and Mendelsohn 1991) or f foregn vstors are ncluded, restrctve smplfyng assumptons have been mposed (Mungatana and Navrud 1994). On the other hand, contngent valuaton reles on surveys contanng hypothetcal scenaros n order to place values on goods that cannot be prced drectly through a market (Cummngs et al. 1986). Thus, CV has more flexblty than TC n that a survey can be desgned to elct many dfferent types of values, not only the use value of a specfc area such as a natonal park.

11 Usng conventonal CV survey technques, t has often not been possble to collect the data necessary to estmate an unrestrcted system of demand equatons ncludng cross-prce elastctes - whch can then to be used n desgnng effectve dfferental prcng polces (Brown 1994). To accomplsh ths, a contngent behavour approach s employed n ths study to generate expermental data to assess the structure and level of entrance fees to Serenget natonal park n Tanzana. The CB approach has been appled n several prevous studes, ncludng those by Ward (1987), Looms (1993), Adamowcz et al. (1994), Layman et al. (1996) and Chase et al. (1998). A stated preference approach to the estmaton of vstaton demand can be used that presents respondents wth an array of stes and characterstcs and asks them whch ste they would vst (Adamowcz et al. 1994). The CB approach used n our study specfes a change n entrance fees at one park and asks how vstaton patterns would change at a substtute park. The responses provde data that allow for the estmaton of prce elastcty and also n the estmaton of demand functon. Ths s necessary n order to fully understand the effects of change n entrance fee on park vstaton patterns. Model A commonly used economc model of resource valuaton suggests that a representatve consumer maxmzes a drect utlty functon ( U ) U ( X, Q), s. t M P X X P Q Q Where: X s a vector of commodtes; Q s the quantty of an envronmental amenty (n ths case, vsts to natonal parks n Serenget); M s consumer ncome; Px s a vector of commodtes' prces; and PQ s a prce vector of entrance fees for vsts to natonal parks. Maxmzaton of the above constraned utlty functon yelds ndvdual demand curves, and aggregaton across the market yelds the aggregate demand curve for Q: Q = Q (M, Px, PQ).

12 Based on the theory and past emprcal studes, aggregate demand curve for natonal park vstaton n Serenget s expected to be a functon park's entrance fee as well as the entrance fees of other park and attractons, park vstors' ncomes, demographc characterstcs, and trp-related factors. The demand functon can be wrtten n general form as: Q= Q (P, M, Z) Where; Q = vstaton at Serenget Natonal Park (n days); P = vector of entrance fees at Serenget and competng parks (n dollars); M = park vstors' ncome (1,000 dollars); and Z = demographc and trp-related characterstcs. The demographc and trp-related characterstcs that wll be ncorporated n the demand equatons nclude vstor age, educaton, natonalty, and whether the vstor was part of a tour. We wsh to look on the utlty one derves from vstng the Serenget Natonal Park. Ths s the utlty of gong. Therefore the logt model s used n our estmaton. Generc Model We derve the utlty of an ndvdual gong to Serenget Natonal Park compared to that of not gong. It s assumed that the utlty of gong s lnear n explanatory varables; thus the model s specfed as Y (1) X Where Y s the utlty of vstng Serenget Natonal Park, X s a vector of explanatory varables, and s the random error term.

13 Y s consdered to be a dscrete (dchotomous) varable, takng a value of 1 or zero f the answer to the queston regardng whether an ndvdual s wllng to go s YES or NO respectvely. Y s motvated by a latent varable, y*. Thus, Y 1 f y * 0, and Y 0 f otherwse (2) The latent varable * y s assumed to be a lnear functon of the explanatory varables X and unobserved error varable y * X (3) If we arbtrarly assume a logstc dstrbuton for, then we can defne the probablty of a person to answer YES as and NO as1. Ths probablty s derved as follows: Pr( Y 1) Pr( y * 0) Pr( X 0) Pr( X ) (4) Where the dstrbuton of Y s a Bernoull dstrbuton 3 wth parameter represented by equaton (4), wth a mean= and a varance= ( 1 ). 3 In probablty theory and statstcs, the Bernoull dstrbuton, s a dscrete probablty dstrbuton, whch takes value 1 wth success probablty p and value 0 wth falure probablty q = 1 p. So f X s a random varable wth ths dstrbuton, we have: The probablty mass functon f of ths dstrbuton s Ths can also be expressed as. The expected value of a Bernoull random varable X s, and ts varance s

14 * Pr( Y y ) * * y y 1 (1 ). (5) For the logstc model, both the mean and varance depend on the underlyng probablty. Hence, any factor that affects the probablty wll automatcally alter the mean and varance of the observaton. Ths lmtaton mples that a lnear model that allows the predctons to affect the mean whle the varance remans constant s nadequate for analyss of bnary data. Thus, the logt model (Gujarat, 1988; Hosmer and Lemesher, 1989) s specfed as n equaton (6) below and used for estmaton: logt(π ) π ln( ) α β X ε. (6) 1 π Specfcally we wsh do the followng; * exp( y ) prob ( Gong 1) (7) * (1 exp( y )) Where y * * * X ; * and * are estmated coeffcents Thus, the emprcal model for estmaton s specfed as Ugo 0 1TotalCost 2 Income gong..... (8) Where, Ugo s a bnary dependent varable (whch takes the value 1 for the yes answer and 0 for the no answer.

15 How much can a person wllng to Vst Serenget Natonal Park before one s ndfferent s when f Ugo Unogo 0 1 TotalCost 2 Income gong (9) Utlty of not gong s 0 nogo Thus takng expectatons we wll have, TotalCost Income 0.. (10) If we suppress the role of ncome we wll have the followng; TotalCost 1 TotalCost...(11) Hence the person s ndfferent, and t s ths equaton that wll be fundamental n explanng the swtchng behavor of the toursts.

16 4. Data, Samplng and Survey To get the data for estmatng the vstaton demand functon for Serenget natonal park, we made use of the Contngent Behavour (CB) approach. We have adopted the CB model smlar to the one used by Chase et al (1998). In our context, the CB approach generates expermental data to assess the effects of changes n entrance fees to the Serenget natonal park takng nto account the substtutablty wth the Maasa Mara Game Reserve across the border. We have ntervewed toursts n both Serenget and Maasa Mara. Ths has been done as part of an on-gong project by the Envronment for Development Intatve n Kenya, South Afrca and Tanzana. A random samplng technque of foregn toursts was surveyed n Serenget natonal park n Tanzana. Prmary data were collected through n-person ntervews conducted n 2013 and 2014 durng Serenget peak tourst season. The study reported here s based on a total of 539 usable surveys of foregn vstors. It was decded that n-person ntervews should be conducted n order to ensure full understandng and completon of the questons. The ntervew began wth questons about the toursts' vews about protecton of wldlfe and experences n Afrca and ts natonal parks. Informaton on actual park fees pad, opnons regardng "approprate" fee levels, and wllngness to pay for hgher fees was collected. In the last case, the respondent was asked, "If the entrance fee were ncreased only at ths park, how hgh the daly entrance fee per person would have to be so that you would choose not to vst ths park?" Many respondents had vsted or were otherwse famlar wth the natonal parks n queston and had lttle dffculty statng f and how ther vstaton plans would change. Those that dd have problems were able to understand the queston after dscussng t wth the ntervewer. Whle tme consumng, t was possble to elct responses to these hypothetcal questons through a combnaton of n person ntervews and a chart vsble to both the ntervewer and respondent. Fnally, standard demographc nformaton (household ncome, natonalty, age, educaton, etc.) was also collected.

17 5. Emprcal Fndngs We present a bref dscusson of the emprcs n Serenget Natonal Park n Tanzana. Descrptve Analyss Characterstcs of the Sample The data collected from 539 observatons (nternatonal toursts) were coded and analyzed usng STATA 12. These smple statstcal analyses produced means and other statstcs, crosstabulatons, and percentage frequences that are presented n ths secton. The descrptve statstcs of the varables used n the study are summarzed n Table 1. Of the 539 respondents surveyed from the data 286 (53.06 %) were male and 253 (46.94%) were female. Table 1 shows that the average age of the respondents s 43 years, mplyng a mddle-aged respondents, whch was more lkely to be better placed at understandng the questons than t would have been the case wth a younger or older ndvduals. The educaton status suggests that a majorty of the respondents had acheved at least a mnmum level of schoolng;.e., the majorty attaned, at the hghest, prmary educaton. The ncome from all sources of the majorty of respondents s between US$ 50,001 to US$ 100,000 and the average sze of the number of households accompanyng the tourst s about 2 people (1.6)

18 Table 1; Descrptve Statstcs of some of the varables used n the Study Varable Mean Sd mn max N Age Household(accompanyng the tourst) Income(US$ 0-50,000) Income(US$ 50, ,000) Income(US$100, ,000) Income(over US$ 200,000) Gender(male dummy) Nghts spent so far n Serenget total nghts to be spent n Serenget Prce(cost of trp)1(choce of Serenget) prce(cost of trp)2(choce of Serenget) prce(cost of trp)3(choce of Serenget) Vst(recommendaton by frend)1(choce of Serenget) Vst(recommendaton by frend)2(choce of Serenget) Vst(recommendaton by frend)3(choce of Serenget) Reason(recommendaton by travel agent)1(choce of Serenget) Reason(recommendaton by travel agent)2(choce of Serenget) Reason(recommendaton by travel agent)3(choce of Serenget) unqueness1(choce of Serenget) unqueness2(choce of Serenget) unqueness3(choce of Serenget) Package(other elements of tour package)1(choce of Serenget) Package(other elements of tour package)2(choce of Serenget) Package(other elements of tour package)3(choce of Serenget) Outsde Serenget(%spent) Travel cost Indvdual cost(total cost) New total cost Hghest cost ncrement

19 From the table above we can see that most of the toursts go to Serenget Natonal Park because of ts unqueness and ths s the motvaton behnd the vst to Serenget Natonal park. Unqueness s ranked wth the hghest mportance. Moreover other elements of the Safar package also plays a role on the motvaton of an ndvdual to choose Serenget. Though they rank t as a moderate reason. Park prcng for Serenget s dfferent from other parks and we fnd that prce provdes a reason as to why people choose Serenget though the reason behnd the prce s ranked as moderate about. The recommendatons by travel agent and frends do play part n one s decson but not to a large extent. Moreover on the answerng the queston f an ndvdual wll stll choose Serenget f there s entrance fee ncrease. The table shows that f there s an ncrease n entrance fee the tourst wll stll choose Serenget. Almost 71% wll stll choose to vst Serenget. Ths can be attrbuted by ts unqueness. Also we looked on toursts and where they are comng from so as to see whch regons represent more tourst n vstng the natonal Parks. Here we compared the Serenget Natonal Park and ts substtute whch s Maasa Mara Game Reserve. The followng tables llustrate; Table 2; Toursts vstng Serenget Natonal Park Regons where tourst come from Freq. Percent Cum. Australa Europe North Amerca Asa South Amerca Afrca New Zealand Total As we can see the majorty of the toursts vstng the Serenget Natonal Park come from Europe (56.22%) followed by the North Amerca (25.79%). It s therefore mportant for the park authortes to envsage the means of attractng more toursts from other regons and also get more of them from these leadng regons.

20 For the substtute Park whch s Maasa Mara we have also the large proporton of toursts comng from Europe (48.5%) followed by North Amerca (26.43%) as shown n table 3 below. Table 3; Toursts vstng Maasa Mara Game Reserve Regons where tourst come from Frequency Percent Cumulatve Australa Europe North Amerca Asa South Amerca Afrca New Zealand Total Upon comparng the choce of the tour package accordng to the regons where toursts come from f there s an ncrease n entrance fee n both natonal parks we have the followng results. Table 4; Choce of the same tour Package due to an ncrease n entrance fee n Serenget Choce of the same package (for Serenget) Australa Europe North Amerca Asa South Amerca Afrca New Zealand No Yes Total Total From the above table we can see that the large proporton of the toursts would stll choose the same tour package f there s an ncrease n entrance fee Europeans and Amercans beng the leadng toursts optng that. For Maasa Mara we see that though t s a substtute park the tourst would stll choose the same tour package to reman n game reserve as the table below ndcates.

21 Table 5; Choce of the same tour Package due to an ncrease n entrance fee n Maasa Mara Choce of the same package (for Masa Mara) Australa Europe North Amerca Asa South Amerca Afrca New Zealand Total No Yes Total Furthermore, f there s an ncrease n cost n Serenget Natonal Park the tourst would stll choose Serenget over the substtute park whch s Maasa Mara. Almost 66% of the respondents sad they would stll choose Serenget. Table 6 below provdes a summary. Table 6; Choce between Serenget and Maasa Mara Choce Frequency Percent Cumulatve Mara Serenget Total If we look on the choce accordng to the regons the table below shows that most of the tourst from Europe and Amerca would stll choose Serenget over the substtute park whch s Maasa Mara. Ths could be attrbuted by the unqueness nature of Serenget Natonal park. Table 7; Choce accordng to the Regons vstng Serenget Choce Australa Europe North Amerca Asa South Amerca Afrca New Zealand Total Mara Serenget Total For substtute Park whch s Maasa Mara Game Reserve the dstrbuton s as follows;

22 Table 8; Choce between Serenget and Maasa Mara Choce Frequency Percent Cumulatve Mara Serenget Total Table 9; Choce accordng to the Regons vstng Maasa Mara Choce Australa Europe North Asa South Afrca New Total Amerca Amerca Zealand Mara Serenget Total From the above tables we can see that, the toursts tend to stck on the same natonal park they frst vsted. In other words we can say that the senstvty to ncrease n the entrance fee s low and toursts can afford to pay hgher as they derve hgh utlty n wldlfe amenty. Therefore the park authorty can slghtly ncrease the entrance fee so as to generate more revenues. Furthermore f we look on the behavor of the tourst to swtch as the entrance fee ncreases (focusng on Serenget Natonal Park) we fnd out that hgher the entrance fee prompts tourst to swtch to substtute park. The followng table llustrates. Entrance Fee No Yes Total % of Tourst who wll go to Serenget Total

23 Moreover we wanted to know that f the behavor of the toursts follow the ratonalty of the consumer behavor f there s an ncrease n prce of the commodty. In our case we looked on the behavor of the toursts f there s an ncrease n entrance fee. The followng graph show the demand functon when you ncrease the entrance fee n Serenget Natonal Park Graph 1; Here t shows that as you ncrease the entrance fee the demand decreases as we expected. That s the hgher the prce the lower should be the demand.

24 To conform our result on Serenget Natonal Park we calculated the elastcty of demand to see f the demand s elastc. Elastc demand means that demand for a product s senstve to prce changes. The reason of calculatng the elastcty comes from the noton of Monopoly. In the theory of demand the monopoly ams at maxmzng revenue. In order to maxmze revenue/proft the monopoly frm operates at the elastc part of the demand curve. Therefore at the pont where demand s elastc s where the monopoly maxmzes revenue. Thus we calculated elastcty at varous ponts and we pcked a pont where demand s elastc. The formula for the coeffcent of prce elastcty of demand for a good s e Q / Q R P / P [ ]/ 0.83 At entry fee equals to 50 the elastcty s; [50 75]/ 50 [ ]/ 0.78 At an entry fee equals to 75 elastcty s; [75 100]/ 75 [ ]/ 0.71 At entry fee equals 100 elastcty s; [ ]/100 [ ]/ 0.65 And at entry fee equals 125 the elastcty s; 1 [ ]/125 Thus at 125 s where revenue s maxmzed snce elastcty equals to 1. Therefore Entrance fee equals to 125 s where the park wll maxmze revenue. Ths s because the demand s elastc. It s at ths pont where the Serenget Natonal Park should charge the entrance fee f they want to look at the qualty and maxmze revenue. Thus for polcy perspectve we recommend the entrance fee of US$125. The revenue s gven by the followng; Revunue P * Q Currently the entry fee n Serenget Natonal Park s $60. Gven the sample of 539 of toursts who were ntervewed n Serenget Natonal Park we can calculate the revenue at current entry fee and our recommended entry fee that Serenget Natonal Park should charge so as to maxmze revenue.

25 Therefore gven the current entry fee the revenue s; Revenue $60 * 539 $32, 340 and at the entry fee equals to 125 where the demand s elastc the revenue Revenue $125*539 $67, 375. The revenue foregone s $35,035. However there wll be a declne n the percentage of tourst who wll be gong to Serenget. At entry fee of $125 only 65% wll go to Serenget compared to 81% f the entry fee was $60. On the other hand f the Serenget Natonal Park wants to maxmze revenue as well as mantanng the qualty of the park they should charge an entry fee equals to $125.

26 6. Results and Dscusson We want to look on the key questons n the decson of the tourst on Serenget Natonal Park. Frst we look on the choce of the tourst f there s an ncrease n the entrance fee what would be the reacton of the tourst. As we have seen from the table the tourst would stll choose Serenget. Of mportance here was the total travel cost and ts nfluence on choosng the Serenget Natonal Park. As table 10 below ndcates f there s an ncrease n entrance fee the total cost s sgnfcant. Our expectaton s that the travel cost nfluences one s decson thus t concurs. Table 10; Estmaton Results from the Logstc Regresson of the effect of Travel Cost on the the choce of package to Serenget Varables Regresson Standard Error Total travel cost 0.545*** Constant ** Observatons 539 *** p<0.01, ** p<0.05, * p<0.1 If we look on the nfluence of other explanatory varables on the choce of the tour package we fnd that along wth the total travel cost among the varables hypotheszed to nfluence the ndvdual choce of the same tour package, ncome, the tour package to Serenget, satsfacton of safar experence and day s package were statstcally sgnfcant. The statstcal sgnfcance of ncome ndcates ts mportance n nfluencng people s choce of tour package. The varable day s package s sgnfcant because t plays an mportant role n determnng ones decson of choosng the safar package to Serenget. Of a surprse s the educaton whch s statstcally nsgnfcant and ths can be due to the fact that educaton level has a lttle mportance n ones decson of choosng the same tour package or not. The followng table llustrates.

27 Table 11; Estmaton Results of the varables nfluencng the choce of package to Serenget Varables Regresson Standard Error Age Age Square Household member accompanyng the tourst Income category * Income category ** Income category ** Educaton category Educaton category Educaton category Educaton category Tour Package to Serenget 0.954** Satsfacton of safar experence 0.477** Days Package to Serenget * Gender Total travel cost 0.393* Constant ** Observatons 524 *** p<0.01, ** p<0.05, * p<0.1 We also wanted to see f the cost change had any nfluence on the choce of the tour package. As table 12 below ndcates f there s an ncrease n entrance fee the cost change s sgnfcant. And ths confrms our expectaton that the cost change nfluences one s decson Table 12; Estmaton Result on the choce of same tour package due to changes n cost Varables Regresson Standard Error Cost Change *** Constant 4.987*** Observatons 539 *** p<0.01, ** p<0.05, * p<0.1

28 If we look on the nfluence of other explanatory varables on the choce of the tour package f we nclude the cost change we now fnd t beng sgnfcant along wth educaton ths tme whch we thought t s very mportant n one s choce. The other varables whch are sgnfcant n nfluencng the ndvdual choce of the same tour package, ncome, the tour package to Serenget, satsfacton of safar experence and travel cost. The statstcal sgnfcance of ncome ndcates ts mportance n nfluencng people s choce of tour package. The followng table llustrates. Table 13; Estmaton Results on the choce of same tour package when all varables are ncluded Varables Regresson Standard Error Age Age Square Household member accompanyng the tourst Income category Income category ** Income category ** Educaton category Educaton category Educaton category * Educaton category Tour Package to Serenget 0.948** Satsfacton of safar 0.489** experence Cost Change *** 0184 Gender Travel cost 0.549*** Constant Observatons 524 *** p<0.01, ** p<0.05, * p<0.1

29 Moreover, we wanted to know that what wll be the choce of the tourst between Serenget and the substtute ste whch s Maasa Mara f there s cost ncrease and the added cost n whch the tourst can tolerate. As ndcated from above people would stll choose Serenget. Thus we regressed choce on added cost usng logt and see f the added cost had sgnfcant mpact on the choce of the tourst. The table below shows that Added cost s statstcally sgnfcant at 5% n one s choce. Table 14; Results for Serenget Natonal Park on the added cost n whch the tourst can tolerate. (1) -(b0/b1) EQUATION VARIABLES Regresson Choce of the tourst Added cost ** 7.34 (0.156) Constant 2.709*** (0.868) Observatons 539 *** p<0.01, ** p<0.05, * p<0.1 As we can see from the above table. The added cost n whch the toursts can tolerate s $7. Besdes we wanted to see the added cost n whch the toursts from Europe and Amerca can tolerate and reman n the same Natonal Park. We chose Europe and Amerca because they brng a large number of toursts to our Natonal Parks.

30 Table 15; Regresson when the tourst are from Europe the added cost n whch the tourst can tolerate. (for Serenget Data) (1) -(b0/b1) EQUATION VARIABLES Regresson Choce of the tourst Added cost ** 7.05 (0.214) Constant 3.220*** (1.205) Observatons 303 *** p<0.01, ** p<0.05, * p<0.1 We can see that the added cost n whch the toursts from Europe vstng Serenget Natonal Park would be ndfferent s approxmately $7. Lkewse t s the same for the toursts from Amerca. Table 16; Regresson when the tourst are from North Amerca the added cost n whch the tourst can tolerate. (for Serenget Data) (1) -(b0/b1) EQUATION VARIABLES Regresson Choce of the tourst Added cost (0.311) Constant 3.304* (1.710) Observatons 139 *** p<0.01, ** p<0.05, * p<0.1

31 Table 17; Results for Maasa Mara the added cost n whch the tourst can tolerate. (1) -(b0/b1) EQUATION VARIABLES Regresson Choce of the tourst Added cost 0.382* (0.227) Constant ** (1.309) Observatons 366 *** p<0.01, ** p<0.05, * p<0.1 As we can see from the above table. The added cost n whch the toursts can tolerate s almost $7 for Maasa Mara game reserve. And t s statstcally sgnfcant, thus t has a sgnfcant mpact on the choce of the tourst. On the other hand the added cost n whch the toursts from Europe and Amerca can tolerate s $11 and $6 respectvely as the tables below llustrate. Table 18; Regresson when the tourst are from Europe the added cost n whch the tourst can tolerate. (for Maasa Mara Data) (1) -(b0/b1) EQUATION VARIABLES Regresson Choce of the tourst Added cost (0.308) Constant (1.765) Observatons 177 *** p<0.01, ** p<0.05, * p<0.1

32 Table 19; Regresson when the tourst are from North Amerca the added cost n whch the tourst can tolerate. (for Maasa Mara Data) (1) -(b0/b1) EQUATION VARIABLES Regresson Choce of the tourst Added cost (0.419) Constant (2.396) Observatons 97 *** p<0.01, ** p<0.05, * p<0.1 We can therefore conclude that the added cost n whch the toursts from Europe and Amerca can tolerate so as to vst Serenget Natonal Park s almost the same as the added cost for the all toursts across the globe combned. Whle for Maasa Mara Game reserve s $11 for Europe and $6 for Amerca and $7 for all the tourst across the globe. Our am of calculatng the prce at whch the tourst can tolerate was to see at whch pont we can ncrease our entry fee over and above the current entry fee n Serenget Natonal Park whch s our prmary focus. We found out that the tolerance level across the globe for toursts vstng Serenget Natonal Park s $7. Thus regardless of the qualty the Park authortes can ncrease the entry fee by $7 makng t to be $67 and ths wll slghtly mprove revenue as well as mantanng all the toursts. We saw earler at $60 entry fee gven our sample the revenue s $ but when the park ncreases revenue by $7 so that the entry fee s $67 the revenue wll be; Revenue $67 * 539 $36, 113 thus the park wll make $3,773 extra by chargng $67 as well as mantanng all the toursts because they wll tolerate that ncrease n entry fee.

33 Further stll, we estmate the equaton (by dong regresson analyss) whch resembles the demand functon we used earler. We frst estmate a smple equaton where the choce of package s only the functon of the entry fee. We fnd out that n our regresson result that the entry fee s sgnfcant n nfluencng ones decson and t s also negatve. The mplcaton here s that the hgher the entrance fee the fewer number of toursts who wll choose to vst Serenget Natonal Park. The followng table llustrates, Table 20; estmated demand of Entrance fee effect Varables Regresson Standard Error Entrance Fee *** Constant 2.408*** Observatons 539 *** p<0.01, ** p<0.05, * p<0.1 If we nclude other varables that affect the choce of gong to Serenget Natonal Park or not we fnd out that along wth the entry fee, ncome, educaton, total travel cost, satsfacton of safar experence as well as f the tourst s a part of the tour package are sgnfcant n nfluencng one s decson. The followng table llustrates;

34 Table 21; Estmaton Results of estmated demand that nclude other varables along wth the entry fee Varables Regresson Standard Error Age Age Square Household member accompanyng the tourst Income category Income category * Income category * Educaton category Educaton category Educaton category * Educaton category Tour Package to Serenget 1.007*** Satsfacton of safar 0.424** experence New entrance fee *** Gender Total travel cost 0.585** Constant ** Observatons 524 *** p<0.01, ** p<0.05, * p<0.1

35 7. Concluson From the Analyss we have seen that the behavour of the tourst does not change much when the prce changes. Ths can be attrbuted by the unqueness of the Serenget Natonal Park. Regardng the optmal entrance fee on whch Serenget Natonal Park should charge, we fnd out that people are wllng to pay above what s currently charged whch s $60. Thus n order to mantan the tourst and not takng nto consderaton f the park s crowded the Serenget Natonal Park should charge an entry fee equals to $67. However our prmary am was to gve the recommendaton on the entrance fee at whch Serenget Natonal Park should charge where t maxmzes revenue. The revenue s maxmsed at elastc part of the demand. And at that pont the entry fee s $125. Therefore we recommend that n order to maxmze revenue and mprovng the qualty of Serenget Natonal Park, the park authortes should charge an entry fee of $125.

36 References Adamowcz, W., Louvere, J. & Wllams, M. (1994). Combnng revealed and stated preference methods for valung envronmental amentes. Journal of Envronmental Economcs and Management 26, Alpízar, F. (2006) The Prcng of Protected Areas n Nature-Based Toursm: A Local Perspectve. Ecologcal Economcs, Andam K., Ferraro, P., Pfaff, A. Sanchez, A. and Robalno, J Measurng the Effectveness of Protected Area Networks n Reducng Deforestaton : A Rgorous Impact Evaluaton Approach, Proceedngs of Natonal Academy of Scence. 105 (42) Ben-Akva. M. and Lerman, S., Dscrete Choce Analyss: Theory and Applcaton to Travel Demand. MIT Press, Cambrdge, MA, 390 pp. Bromley, Danel W. (1991), Envronment and Economy, Property Rghts and Publc Polcy, Blackwell Chamber, R. (1987). Rural Development: Puttng the Last Frst. London: Longman. Davd Pearce, "Paradoxes n Bodversty Conservaton," World Economcs, World Economcs, Economc & Fnancal Publshng, PO Box 69, Henley-on-Thames, Oxfordshre, Unted Kngdom, RG9 1GB, vol. 6(3), pages 57-69, July. Dasgupta, P. and K.-G. Mäler (1995), "Poverty, Insttutons, and the Envronmental Resource-Base," n J. Behrman and T.N. Srnvasan eds., Handbook of Development Economcs, Vol. IIIA (Amsterdam: North Holland, 1995): Dxon, J.A., and P.B. Sherman Economcs of Protected Areas: A New Look at Benefts and Costs. Washngton: Island Press. Ferraro PJ (2001) Global habtat protecton: lmtatons of development nterventons and a role for conservaton performance payments. Conserv Bo 15(4):

37 Ferraro PJ, Smpson D (2002). The cost-effectveness of conservaton payments. Land Econ 78(3): Fscher C., E. Muchapondwa and T. Sterner A Bo-Economc Model of Communty Incentves for Wldlfe Management under CAMPFIRE. Envronmental and Resource Economcs 48(2): Gandwa, E., (2012): Local knowledge and perceptons of anmal populaton abundance by communtes adjacent to the northern Gonarezhou Natonal Park. Journal of Tropcal Conservaton Scence Vol.5 (3): Gujarat, D. N. (1988). Basc Econometrcs. New York, NY: McGraw-Hll. Hogan, K. (2011) The Expanson of Ecotoursm n Tanzana and ts Implcatons for the Maasa Homewood, K. W. & Rodgers, W. A. (1991), Maasaland Ecology: Pastoralst Development and Wldlfe Conservaton n Ngorongoro, Tanzana, New York: Cambrdge Unversty Press, ISBN Horvtz, D. G., and Thompson, M. E. (1952). A generalzaton of samplng wthout replacement from a fnte unverse. J. Atrmr. Statst. Assoc. 47, Hosmer, D. W., and S. Lemeshow. (1989). Appled Logstc Regresson: Wley Seres n Probablty and Mathematcal Statstcs. New York, NY: John Wley Sons. Imbens, G., (2004) Nonparametrc Estmaton of Average Treatment Effects under Exogenety: A Revew, Revew of Economcs and Statstcs, 86, James, A.N., Green, M.J.B. and Pane, J.R. 1999a. Global Revew of Protected Area Budgets and Staff. WCMC: Cambrdge, UK. Kss, A Lvng wth wldlfe: wldlfe resource management wth local partcpaton n Afrca. World Bank Techncal Paper 1990 No. 130 pp. 31 pp.

38 Karanth, K., and N. Sanjay, (2011): Local Resdents percepton of benefts and losses from protected areas n Inda and Nepal. Elsever Kdeghesho, J.R., Nyahongo, J.W., Hassan, S.N., Tarmo, T.C. and Mbje N.E Factors and ecologcal mpacts of wldlfe habtat destructon n the Serenget ecosystem n northern Tanzana. AJEAM-RAGEE 11: Kdeghesho, JR, (2006) Wldlfe conservaton and local land use conflcts n Western Serenget Corrdor, Tanzana Kurz, M., The general nstablty of a class of compettve growth processes, Revew of Economc Studes 35, LaSalle, J. and S. Lefschetz, 1961 Krutlla J.V Conservaton Reconsdered. Amercan Economc Revew 1967; 57: Lsa C. Chase, Davd R. Lee, Wllam 1998 Ecotoursm Demand and Dfferental Prcng of Natonal Park Access n Costa Rca. Looms, John Integrated Publc Lands Management: Prncples and Applcatons to Natonal Forests, Parks and Wldlfe Refuges, and BLM Lands. Columba Unversty Press, New York, NY nd Edton McFadden, D., Condtonal logt analyss of qualtatve choce behavour. In: P. Zarembka (Edtor), Fronters n Econometrcs. Academc Press, New York, NY, pp Marks, S.A., The Imperal Lon: Human Dmensons of Wldlfe Management n Westvew Press, Boulder. Afrca. Muchapondwa, E., F. Carlsson, and G. Kohln, (2008): Wldlfe Management n Zmbabwe Evdence from a Contngent Valuaton Study. South Afrcan Journal of Economcs, Vol. 76:4

39 Pfaff, A., J.A. Robalno, G.A. Sanchez-Azofefa, K. Andam and P. Ferraro (2009). Park Locaton Affects Forest Protecton: Land Characterstcs Cause Dfferences n Park Impacts across Costa Rca, The B.E. Journal of Economc Analyss & Polcy:Vol.9:Iss.2(Contrbutons),Artcle5(avalableat: bejeap/vol9/ss2/art5 ) Pfaff, A., Robalno, J., and Sánchez, G.A. 2007a. Evaluatng the Impacts of Payments for Envronmental Servces. Presentaton at NBER Summer Insttute, July. ( R.A.Arragada, P.J. Ferraro, E.O. Slls, S.K. Pattanayak and S. Cordero (2012) Do payments for envronmental servces reduce deforestaton? A farm level evaluaton from Costa Rca R. C. Layman, J. R. Boyce, and K. R. Crddle (1996), An Economc Analyss of the Recreatonal Kng Salmon Fshery on the Gulkana Rver, Alaska. Land Economcs, 72 (February): Skonhoft, Anders, On the prncples of conservaton and utlzaton of wld lvng resources : Envronment and Development Economcs, Cambrdge Unversty Press, vol. 2(01), pages , February Skonhoft, Anders and Solstad, Jan Tore, Investng n Wldlfe: Can Wldlfe Pay Its Way? Journal of Afrcan Economes, Centre for the Study of Afrcan Economes (CSAE), vol. 7(2), pages , July Skonhoft, Anders and Solstad, Jan Tore, Wldlfe management, llegal huntng and conflcts. A bo economc analyss, Envronment and Development Economcs, Cambrdge Unversty Press, vol. 1(02), pages , May. Schulz, C. E. and A. Skonhoft Wldlfe Management, Land-Use and Conflcts. Envronment and Development Economcs 1: Southgate, D. and M. Whtaker Economc Progress and the Envronment: One Developng Country's Polcy Crss. New York: Oxford Unversty Press.