TRAVEL COST METHOD:

BACK TO OUTLINE

Travel Cost Method

McKean, John R.; Johnson, Donn M.; Walsh, Richard G., "Valuing Time in Travel Cost Demand Analysis: An Empirical Investigation," Land Economics, 71(1), February 1995, pp. 96-105.

1. What does the ordinary travel cost model assume about people's ability to substitute time for money, and why is this probably unrealistic for many recreationists?
2. What are two options with respect to the treatment of money and time costs in empirical travel cost models? Are marginal or average opportunity costs of time appropriate?
3. How did Ted McConnell and Ivar Strand (1981, 1983) estimate the fraction of hourly income to use as opportunity time costs?
4. How is it that disequilibrium in labor market can render wage rates irrelevant as a measure of opportunity time costs?
5. How did Frank Ward (1983, 1989) avoid the assumption of equilibrium in labor markets? What restriction was necessary for this model, and what is assumed implicitly about the relationship between opportunity time costs and travel time per trip?
6. Nancy Bockstael, Ivar Strand, and Michael Hanemann (1987) showed that time and money constraints cannot be collapsed into one when individuals cannot marginally substitute work time for leisure. What does this imply for the specification of a travel cost model?
7. What strategy do the current authors employ to produce a model that subsumes both equilibrium and disequilibrium specifications? Do they test which specification is preferred?
8. What type of econometric model is used to estimate the parameters of their specification?
9. What fraction of income is implied to be the opportunity time cost?
10. Does time value appear to be relatively constant across individuals?

Englin, Jeffrey; Shonkwiler, J. S., "Modeling Recreation Demand in the Presence of Unobservable Travel Costs: Toward a Travel Price Model," Journal of Environmental Economics and Management, 29(3), Part 1 Nov. 1995, pp. 368-77.

1. From what other research traditions have researchers usually transferred the value of time used in travel cost studies?
2. What are the econometric implications of errors-in-variables for estimates of consumer surplus in travel cost models?
3. What basic econometric model for recreational trips is employed in this paper and why?
4. What modification to this model is necessitated by the fact that the sample consists only of users? What is the standard reference for these models?
5. The paper posits that one cannot observe true latent travel costs, only a vector of variables that are closely associated with these costs. What statistical technique is used to convert these indicator variables to an estimate of unobserved travel costs?
6. When this estimated latent variable is used in place of travel costs computing as an accounting identity, what happens to the estimated coefficient on the price variable in this log-linear model?
7. What does this paper imply for the opportunity cost of time spent travellilng? for the value of time as a proportion of the wage rate? for estimated per-trip consumer surplus?

Hellerstein, Daniel, "Welfare Estimation Using Aggregate and Individual-Observation Models: A Comparison Using Monte Carlo Techniques," American Journal of Agricultural Economics, 77(3), August 1995, pp. 620-30.

1. Why have zonal travel cost models been large abandoned in favor of individual travel cost models?
2. What shortcomings afflict individual-data models?
3. Are analytic comparisons of aggregation bias vs. specification bias tractable? What strategy does Hellerstein use?
4. Describe the stylized model that is the basis for Hellerstein's simulations.
5. What dimensions of the range of common empirical problems are simulated in this study?
6. Under what types of circumstances do aggregate models tend to outperform individual-observation-based models? When do they not?
7. If aggregate data are to be used, what additional information does Hellerstein recommend be collected?

Randall, Alan, "A Difficulty with the Travel Cost Method," Land Economics, 70(1), February 1994, pp. 88-96.

1. Why does Randall distinguish between Travel Price and Travel Cost?
2. Randall inventories six "stubborn methodological problems" that persist for travel cost models. What are these?
3. What is meant by the typical assumption that cost is "observable"? (i.e. what are the implied properties of costs for them to be observable?) What is meant by the argument that opportunity costs are subjective?
4. Why is ordinal-scale measurement of household welfare change bad news?
5. How does Randall argue that most researcher mask the problem of ordinal measurability of travel cost welfare measures?
6. How does the household production function story improve upon conventional travel cost models? How does the fundamental unobservability of household production technology mean that we do not get much further with this?
7. Do Random Utility Models (RUM) solve these problems?
8. What two strategies does Randall recommend as holding promise for obtaining valid absolute-valued welfare measures from the Travel Cost Method?

Hellerstein, Daniel; Mendelsohn, Robert, "A Theoretical Foundation for Count Data Models," American Journal of Agricultural Economics, 75(3), August 1993, pp. 604-11.

1. What two properties of individual travel cost demand data necessitate the use of "count data" econometric models?
2. What two types of consumer demand models do Hellerstein and Mendelsohn develop, both of which are consistent with count data specifications?
3. In count data models, how should one compute expected consumer surplus? Contrast this with the methods used in continuous models.
4. How does the repeated discrete choice story end up being a Poisson model?
5. What open questions do these authors identify in their conclusions section?
6. How do the consumer surplus formulas for these count models compare to the standard semilog consumer surplus formulas in the continuous case?

Smith, V. Kerry, "Welfare Effects, Omitted Variables, and the Extent of the Market," Land Economics, 69(2), May 1993, pp. 121-31.

1. For what reasons do economists typically seek to delineate the extent of a market?
2. How does Smith explain Kling's results concerning degrees of correlation between the prices of substitutes and degree of bias in welfare measures for single- or multiple-price changes?