Conference Agenda

This study examines the impact of recall bias on reported recreation trip counts using data from a representative sample of Great Britain residents, focusing on their most recent visit to a coastal space. By systematically varying the reporting period from one to twelve months and collecting data throughout the year, we uniquely control for both the timing of data collection and the reporting period. We make use of this data to uncover patterns in participant recall accuracy and assess how temporal framing influences reported trip frequency. Employing a novel approach, we use local count models with kernel smoothing to investigate recall bias in trip reporting. Our findings reveal that support for different count model assumptions varies with the reporting period, with longer periods favouring a hurdle negative binomial model due to excess zeros. Additionally, we show that reporting period length influences key model outputs, including consumer surplus estimates. These findings provide new insights into the role of recall bias in recreation demand studies and its implications for economic valuation.

While there is abundant empirical evidence of individuals switching between selfish and prosocial behavior, few economic models formalize these findings. This paper presents a novel model that jointly analyzes three key concepts for understanding prosocial behavior: moral balancing, self-signaling, and motivated reasoning. Individuals maximize material utility under the constraint of maintaining a minimum level of self-image (moral balancing), where self-image depends on signals extracted from past behavior (self-signaling). The processing of these signals is biased toward arriving at a positive self-image (motivated reasoning). The time horizon for which the self-image constraint is active depends on individuals' intrinsic motivation and their awareness of self-image relevant choices. Selfish behavior tends to be higher when the constraint is only active in the long term and it increases with the tendency toward motivated reasoning. These results are partially consistent with experimental results, although the experimental design slightly deviates from the model assumptions. The model suggests that voluntary prosocial behavior is unlikely to be sustained and thus calls for adequate policy measures.

In May 2019, we surveyed persons aged 45-60 in the Czech Republic, asking them to report information about their Willingness to Pay (WTP) for reductions in the risk of dying from cancer. One important feature of that survey is that study participants had to infer the size of such risk reductions from changes in two probabilities: The chance of getting cancer, and the chance of surviving it, assuming that they got cancer in the first place. Our respondents appeared to be willing and able to perform such calculations and form a WTP for the resulting mortality risk reductions, as their WTP was perfectly proportional to the size of the risk reduction and met other internal validity criteria. Over five years later, in November 2024, we conducted the same survey, using the same questionnaire and mode of administration, to a new sample of members of the general public aged 45-60 in the Czech Republic. The new sample matches almost perfectly the old one in terms of education, and was wealthier in nominal terms, but not in real terms. We find that in this latter wave respondents are slightly more responsive to the size of the mortality risk reduction, but the overall VSL is lower than that estimated from the 2019 survey responses. Respondents also indicate that the VSL is lower when cancer brings stronger limitations to everyday activities.

This study quantitatively estimates the impacts of Protected Area Downgrading (PAD) events on housing values across the U.S. We leverage changes in housing values of more than 23 million of properties close to PAs across the United States to quantify the external economic impacts of 1,246 PAD events occurred between 1905 and 2018. A triple difference analysis shows that PAD events signiffcantly decrease housing values within up to 2.5 km by about 9%, which are robust across different speciffcations. The negative impacts on housing values are mainly in areas where PAD were established due to forestry and subsistence reasons. On the contrary, PAD events generate a price premium on nearby houses in areas with infrastructure development. The loss of the reduced PAs are estimated by about $203.563 billion across the United States. Our results contribute to a better understanding of the economic consequences of shifts in land protection strategies and promote a holistic assessment in decision-making processes pertaining to protected areas.