This study contributes to the discussion of managing tourism to a protected area in a developing country (Galapagos, Ecuador). The first part of the analysis provides quantitative data about preferences of tourists and potential impacts on park revenues from price discrimination. It uses the data from a choice experiment survey conducted in the summer of 2009 in which these four attributes of a tour of the Galapagos were described: tour length, depth of naturalist experience, level of protection of Galapagos from invasive species, and price of the tour. On average the Galapagos tourist would be willing to pay slightly more than 2.5 times for a trip with a high-level of environmental protection than for a trip that is equivalent on all other characteristics but has a lower level of environmental protection. The mean marginal willingness to pay (MWTP) for a trip with an in-depth naturalist experience is 1.8 times more than that for a trip with a less detailed naturalist experience but equivalent on other characteristics. The relatively inelastic demand for travel to the islands would allow managers to adjust access fees to shift the distribution of length of trips while not affecting the revenues.

The second part of the analysis evaluates the influence on travel to the islands by depicting Galapagos as a standard market commodity as well as depicting it as an environmental commodity. This analysis compares the results obtained from two different choice experiment surveys given to tourists finishing their trip to Galapagos. One survey design portrays the archipelago as a standard holiday island destination while the other design highlights the uniqueness and vulnerability of the islands' biodiversity and the challenges that tourism poses to the islands' conservation. Results suggest that additional information modified an individual's decision-making process. In the first design case (which excludes environmental information), the influence of attributes such as length and depth of natural experience is attenuated. The MWTPs estimated for these attributes are smaller in absolute terms although differences on the MWTP are not statistically significant.