Part 1: Predicting Quantity

This is the first part of a series of four blog posts on the potential uses of GIS in cost-benefit analysis.

The field of economics has been become more data intensive with the increase in computing power over the past several decades. This has allowed economists to test old assumptions empirically and to move away from being a deductive science to an inductive one. The same can be said about the field of geography. The rise of geographic information systems (GIS) and spatially referenced data allows for economists to leverage new datasets and devise entirely new methods to approach a problem.

In cost-benefit analysis, an economist places value on a projects benefits that often don’t have a market price, as they are for goods or services often not bought or sold in a traditional market. For instance, clean air is neither bought nor sold but that does not imply it has no value; however, it does increase the difficulty of calculating its value. To place a value on a traditional market good we take its price and multiply it by the quantity consumed. For non-market goods, like clean air, we use an individual’s willingness to pay (WTP) for that good as the measure of price. This is the first component where GIS can improve the valuation of those hard to quantify non-market goods and services provided by the economy, society, the environment, and the greater global ecological system as a whole. A discussion of how GIS can be used to improve the price component of value, through distance decay functions, can be found here.

Spatially referenced data not only helps improve the calculation of the prices – quantifying the quantity consumed can be framed as a spatial problem as well. For instance when valuing a recreational experience at a new park a person’s distance from the park affects both their willingness to pay for the amenities provided by the park and the likelihood that they will visit the park. GIS models allow economists to replace the assumptions they use when calculating the quantity consumed for non-market goods and services with data driven solutions. In a simplified model we can state that the number of people that will visit a park is a function of the number of people that live around the park, their distance to the park and the number of other parks available to them. To do this we can take zip code level data on population density and calculate the number for people within a 1, 5, and 10 minute walk, or drive, by creating buffer zones. Next we can reduce the likelihood that each person will go to the new park if they are also located nearby other parks with similar or better amenities. Putting this all together we can create an algorithm that predicts the number of expected park visits. This is depicted visually in the image below.

This simple example only highlights the power of GIS in one dimension. The potential for GIS to change the way economists model problems, communicate the results, and aid in the decision making process is immense. Integrating economics with GIS methods and data is a new frontier of research in the field and the applications of this research will improve and enrich infrastructure evaluation.