Sunday, September 14, 2014

The Economics of Fisheries, a Common Property Resource

In a modern society of high product demand for a multitude of products, one company’s inventory strategy can give them a competitive advantage over similar companies. However, an inherent challenge associated with inventory management is accounting for a product’s life cycle. A product’s lifecycle will dictate how often it needs to be replenished by vendors, and how much of it should be in stock at a given time. Errors in predicting the product lifecycle can lead to either overstocking or under-stocking a product, both of which will result in a net profit.

            To me, the fisheries industry presents a uniquely interesting inventory management situation. Historically, fisheries are chronically overfished. Overfishing results in depleted stocks that can support only a fraction of what be harvested if stocks were maintained at higher levels. Typically this phenomenon is modeled with an open access model that depicts what happens when there are unregulated access and harvesting of a common property resource. Additionally, fisheries have a considerably short life cycle, so procurement lead times are extremely short to allow for replenishment.  By extension, I imagine that this prevents wholesale fishing companies from performing adequate order quantity analysis. Marginal product analysis would prove very difficult, because suppliers cannot predict from one week to the next how many fish will be caught. In fact, the fishery industry is one in which price per product is constantly changing. This implies that there is significant inefficiency in the management of fisheries inventory.


             From a policy perspective, regulating the stock of fisheries is particularly challenging. Fisheries are considered a common property resource. Fish is not recognized as “private property” until it is captured. There are four policies that have been used to avoid stock depletion under open access: closed seasons, gear restrictions, total allowable catch, and limited entry. Although helpful, none of these have succeeded in facilitating efficiency of fishery inventories. For example, gear restrictions are imposed deliberately to reduce the efficiency of fishers and/or to prevent adverse impacts on the supporting ecosystem. Specifically, off the coast of New England, the regional management council has imposed a minimum mesh size of 5.25 inches for nets to be pulled by trawlers. Gear restrictions might be viewed as the raising of the unit cost of effort to increase fish inventory from c to c’:





This causes the cost equation to rotate upward, establishing a new steady state equilibrium. As such, gear restrictions reduce effort, but do not really address the underlying problem of inventory inefficiency and a resource being harvested to the point where is has a zero marginal value.

The inability of traditional management policies to stop the phenomenon of over-fishing has led to a discussion of the potential to apply “incentive-based” policies, such as Individual Transferable Quotas. This is essentially a cap-and-trade system, applied to fisheries. At this point, I wonder if such a system has the potential to alter the nature of fisheries inventory management, or are the major fisheries players to set in their ways to be open to such a change.

References:
Resource Economics, Jon M. Conrad, Cambridge University Press, 1999.
“Managing Inventory,” Hammond, Harvard Business School.
















1 comment:

Note: Only a member of this blog may post a comment.