Printer Friendly

Ownership and the regulation of wildlife.


The use of wildlife resources is governed by a combination of private contracts and public regulations. Most often, private landowners control access rights, and government agencies regulate hunting and other uses. This paper shows that these institutions depend on wildlife values and the ability of private landowners to control access to species that inhabit their land. Logit regressions and literary sources are used to test implications about private hunting rights and state regulations. The data support the view that private, legal, and political forces have led to institutions that vary in ways consistent with wealth maximization.


Most resources are not exclusively private or public, but are governed by a mixture of private and public institutions. In the United States wildlife populations are no exception. State governments usually regulate hunting, trapping, and fishing; federal agencies protect endangered species; and private landowners largely control access rights to habitat. Although it is true that owning deer or elk is different from owning cattle or sheep, the facts refute the common assertion that wildlife are unowned or have no value. Existing laws and regulations define property rights to many dimensions of wildlife, and in many areas access rights to wildlife are routinely bought and sold.

Economists typically view private contracts as arrangements that maximize the value of resources governed by the contract. On the other hand, public institutions are given less credit. Becker [1983], Peltzman [1976], Stigler [1971], and others have argued that interest groups can influence political decisions and ultimately distort outcomes. From an efficiency viewpoint, private and public institutions are distinct.

Barzel [1989], Becker and Murphy [1988], McManus [1975], and Wittman [1989] have advanced an alternative view that institutions, including government, not only reflect the optimizing activity of individuals but also maximize the net value of resources. Alchian [1950] argued that persistent and surviving institutions are likely efficient. Because most features of American wildlife institutions have persisted for two centuries, the likelihood of efficiency seems high relative to the likelihood of pervasive inefficiency. In part, I test this efficient organization hypothesis by treating both private and public wildlife institutions as the result of efforts to increase the value of wildlife. Specifically, I argue that the content of private wildlife controls and government regulations depends on the value of wildlife stocks as well as on the ability of landowners to contract with each other and with wildlife users to control wildlife stocks and access to them.


It is costly to establish ownership of wildlife stocks because the ownership patterns of land do not always coincide with the populations' territories. If a stock of wildlife is the only valuable attribute of a parcel of land, then the value of the land is maximized when land ownership coincides with the population's territory. Under these conditions, wildlife would be quite similar economically to domestic animals. The habitat would still be "natural," however, and the species would still be considered "wild." Below, I consider the usual case in which wildlife is not the only valued use of land.

Optimal Landownership and Wildlife Territory

Assume that a homogeneous tract of land has two potentially valuable attributes: agriculture and wildlife. Next, assume that the size of the wildlife stock is not influenced by agriculture. Finally, assume that the stock of wildlife is composed of homogeneous units (that is, individual animals), and that the value of wildlife is tied to the value of the animals and not to the value inherent in the existence of a species.(1)

If the land is used for agricultural production, then the optimal size of the land tract in acres is [S.sub.a].(2) If agriculture is the only valuable attribute of the land, then the pattern of landownership (and the production of farm products) depends on the relative value (net of property rights enforcement costs) of the various agricultural uses. Thus, the optimal plot size depends only on the value of land in agriculture and equals [S.sub.a].

A valuable wildlife attribute on the land, such as a deer herd, creates a different situation. The wildlife stock coexists with the agricultural use of the land and has some optimal plot size, [S.sub.w], that depends on its territorial requirements. [S.sub.w] may be larger or smaller than, or equal to [S.sub.a]. If deer require 10,000 acres and the optimal plot size for agriculture is at least 10,000 acres, then the landowner would essentially own the deer population. But if the optimal plot size is only 1,000 acres, then the landowner's (or another agent's) ability to own the "deer population rights" will depend on the cost of transacting an agreement among ten 1,000-acre landowners.(3) These contracting costs among landowners may eliminate the potential gains they could acquire from specifying their rights to the wildlife.

Landowner Incentives for Owning Wildlife

The ability of landowners to establish rights to wildlife on their property depends on the incentive they have to resolve the conflict between the territorial requirements of a wild population and the optimal tract size of land used for other purposes. A landowner's net value of a wildlife stock is (1) where P is the market price or shadow value of a "unit" of wildlife stock; W(L) is the production function for wildlife stock where land (in optimal size tracts), L, is the only input; and t([S.sub.a],[S.sub.w]) is a simple contracting cost function that depends on the relationship between optimal plot size, [S.sub.a], and wildlife territory, [S.sub.w]. When the gains from owning wildlife are overwhelmed by the costs of contracting among landowners, no wildlife is produced, implying that [V.sup.*] = 0. When the total value of the wildlife stock outweighs the contracting costs, however, landowners will seek ways to establish rights to the wildlife, implying [V.sup.*] > 0.

[Mathematical Expression Omitted]

Four propositions emerge: First, as the relative value of wildlife increases, the greater will be the gains from transacting an agreement among landowners ([partial derivative][V.sup.*]/[partial derivative]P = W(L) > 0) and the more likely it is that rights to the wildlife stock will be established. Second, as the land's wildlife productivity increases, the greater will be the gains from transacting an agreement among landowners ([partial derivative][V.sup.*]/[partial derivative]W(L) = P > 0) and the more likely it is that rights to the wildlife stock will be established. Third, as the size of landholdings increases, the resulting decrease in the costs of contracting among landowners will increase the gains from transacting an agreement among landowners and the more likely it is that rights to the wildlife stock will be established. Fourth, as the territory of the wildlife stock increases in size, the resulting increase in the contracting costs among landowners will decrease the gains from transacting an agreement among landowners and the less likely it is that rights to the wildlife stock will be established.

[Mathematical Expression Omitted]

Contracting Costs and Government Wildlife Agencies

As the propositions indicate, wildlife ownership will depend on wildlife value, wildlife productivity, land ownership patterns, and the size of wildlife territory. These factors will be important for both private and public ownership rules. In the United States, federal and state governments have been regulating wildlife since the seventeenth century [Tober 1981]. Landholdings are usually relatively small compared to the territories of most valued species, and a large fraction of land is not privately owned. As a result, most rights to wildlife are owned by government agencies, reflecting the comparative disadvantage of explicit private ownership.(4) Government wildlife departments can be thought of as agents that have contracted with landowners for control of the wildlife attributes of their land where the costs of contracting with other private landowners are relatively high. Having shown that government wildlife agencies are consistent with the model, I now examine private hunting rights and state regulations.


Even though landowners in the United States do not own the wildlife on their land, they can sometimes gain from enhancing wildlife habitat by enforcing and selling the access rights to their land for hunting, fishing, or trapping. A key implication of the model is that as the size of a landholding increases, it is more likely that hunting rights will be sold, because larger landowners face lower costs of establishing rights to wildlife stocks. This implies that owners of larger holdings will control access to more species and, therefore, will lease access rights more frequently than small landholders will.

In 1900, the average size of a farm in the United States was 146 acres [Department of Commerce 1970, 162]; by 1986 the average farm had grown to 455 acres [Department of Agriculture 1987, 370]. Because the size of agricultural holdings has increased, I expect, ceteris paribus, that the amount of fee hunting on private lands has increased as well. Outdoor writers and wildlife biologists have presented abundant evidence that fee hunting has increased dramatically.(5) The 1986 Montana Fee Hunting Survey conducted by Montana State University found that only three of the forty-eight ranchers surveyed had been selling hunting rights for more than twenty-five years. Furthermore, the mean size for Montana ranchers selling hunting rights in this sample was 12,970 acres, compared to a statewide average of 2,546 acres. Similarly, the Texas Parks and Wildlife Department [1986] surveyed more than 700 landowners who sold deer hunting rights and found that they had holdings that averaged 4,780 acres, compared to a statewide average of 732 acres per landholding.

The Montana State University survey can be used to further test the model's implications. Each landowner surveyed sold hunting rights for at least one of several big game species (mule deer, whitetail deer, elk, and pronghorn antelope) in 1986. The propositions of the model offer insight into the factors that will determine whether or not a rancher will sell hunting rights. For a given species, landowners are more likely to sell or lease hunting rights as the amount of habitat they own increases, as wildlife populations become more dense, as wildlife value increases, and as a population's territory shrinks.(6) The logit equations in Table I estimate the influence of several variables on a landowner's decision to enforce hunting rights by leasing the right to hunt. The logit specification is used because the dependent variables are dichotomous choices; that is, for each of the four species, hunting rights are either leased or not leased.

For each of the four species, the probability of leasing hunting rights (measured by ELK, MULEDEER, WHITETAIL, and PRONGHORN) is predicted to be higher for ranches that have more acres of habitat, more densely populated (productive) habitat, and more valuable populations. HABITAT measures the acres of habitat for each species, and DENSITY measures the population density for each species. One way to distinguish variation in the value of wildlife populations across ranches is to consider whether the species damaged ranch assets, such as grain or hay crops. A population that causes damage yields greater gains from the exchange of hunting rights than a population that causes no damage.(7) For this reason, a dummy variable, DAMAGE, is included as a measure of the value of the population. When DAMAGE = 1, this indicates that the animals damaged property and that, ceteris paribus, leasing is more likely than if DAMAGE = 0. The coefficients on HABITAT, DENSITY, and DAMAGE are all expected to be positive; all but one match this prediction.

When individual landowners do not act to establish rights to game, brokers may use contracts to bind separate landholdings together, placing hunting rights to a large area under unified ownership. These third parties might be for-profit businesses (such as guides or hunting companies), hunting clubs, state agencies, or local farm cooperatives. Although data on these organizations is too scarce to allow tests of specific propositions, it is clear that wildlife values exceed the costs of landowner contracting in many cases. For example, Barclay and Bednarik [1968] found that over 5,000 waterfowl hunting clubs leased 2.5 million acres of privately owned habitat in the Mississippi River basin. And in Utah, the United Sportsman Club (a private hunting business) owns hunting rights on over 500,000 acres of private land [Howard 1990].


State wildlife agencies are a rational economic outcome of the high costs of establishing rights to wildlife by private landowners. And just as the existence of governmental intervention in wildlife responds to wildlife values and the costs of landowner contracting, so should the details of state wildlife regulations. In this section I use cross-sectional analysis of state regulations to further test the model's propositions. In all cases data on regulations were directly obtained from the wildlife agencies in each state or from King and Schrock [1985]. I expect, for example, that a state game department would expend more effort to protect valuable species like deer and elk than undesirable species like skunks and woodchucks. Similarly, I expect that in cases where landholdings are large enough so that landowners can more easily control wildlife, the state's role will be reduced.

State Hunting Regulations

Excluding habitat, state wildlife departments own most attributes of wildlife, including the right to kill or "take" the animals. Game departments trade this right by selling licenses for hunting, fishing, and trapping. The rights associated with these licenses are restricted by season closures, bag limits, and weapon requirements. Hunters can cause third-party effects by shooting other hunters, shooting livestock or protected wildlife, poaching, and damaging roads or other property. For hunting regulations to increase the value of wildlife, they must reduce the losses that result from these externalities. States have responded to the problem by placing many restrictions on hunters. For example, hunters are prohibited from using fire, explosives, or bait; they may not destroy nests or dens; they may not shoot from vehicles or from roads; they may not hunt while intoxicated; and they are restricted to certain types of weapons.

The contracting model implies that state regulations will vary with the landowners' ability to establish hunting rights to wildlife and the relative value of wildlife as an attribute of the land. Generally, I expect that regulations will be more restrictive where both landowner contracting costs and wildlife values are high. The average size of landholdings can be used to measure contracting costs. Ceteris paribus, states with larger landholdings are more likely to have longer hunting seasons and less restrictive rules on bag limits and weapons; states with highly valued wildlife stocks are more likely to have shorter seasons and more restrictive rules. For example, in Colorado and New Mexico, large landowners are subject to less restrictive rules on the length of the hunting season and the bag limits for deer and elk. Unfortunately, data on the length of hunting seasons and bag limits are either unavailable or unsuitable for econometric analysis. But, data on shotgun requirements for big game hunters and Sunday hunting prohibitions can be used to test these general implications.

Compared to rifle ammunition, shotgun slugs travel considerably less distance and have less energy beyond immediate range, thus reducing the chance of damage from errant shots. Requiring the use of shotguns is consistent with maximizing the net value of the wildlife by reducing externalities that result from wayward rifle shots. The problem addressed by prohibiting hunting on Sundays is different from the problem addressed by shotgun requirements. Deer hunters using rifles can impose costs on other deer hunters and landowners, but all hunters can impose costs on those who use the land for nonhunting activities. Prohibiting hunting on Sundays is consistent with attempts to maximize the value of land with alternative uses; the value of the land can be increased by setting aside a day for uses other than hunting.

It is likely that both of these regulations will be in force where landowners have few rights to wildlife stocks and where the value of alternative land uses is high. The logit equations in Table II estimate the influence of several variables on the choice of these two regulations, measured by the dichotomous variables SHOTGUN and SUNDAY. Several specifications were used in the equations, and all but one of the coefficients have the predicted sign.(8) Because their contracting costs are lower, landowners with relatively large holdings will be able to establish more rights to wildlife and are likely to gain from controlling hunters' behavior. Thus, states with larger private landholdings (measured by FARM SIZE) are less likely to restrict big game hunters to shotguns or to ban hunting on Sundays. The estimated coefficients for FARM SIZE are predicted to be negative in both cases.

The value of land, which measures the potential for damage from hunters, should also be inversely related to shotgun requirements. NONFARM FRACTION, the fraction of non-farm land in a state, measures the amount of property that could be damaged by unskilled or careless hunters.(9) For the SHOTGUN equations, the coefficients for NONFARM FRACTION are expected to be negative; an increase in NONFARM FRACTION means that property damage is less likely. The effect of NONFARM FRACTION on Sunday bans is predicted to be the opposite of the effect on shotgun requirements.(10) As the fraction of land used for non-farm activities increases, non-hunting activities will become more valuable. Farmland is valuable for hunting, but it is less valuable for other activities, such as picnicking, sightseeing, or hiking.

A densely populated state is more likely than a state with few residents to require shotguns, because there is a greater opportunity for errant rifle shots to cause injuries. Greater population densities indicate that the relative value of the land for wildlife is low compared to other uses. Similarly, as population density increases, so does the likelihood that states will outlaw Sunday hunting. Thus, DENSITY and RURAL DENSITY are expected to have positive coefficients in both the SHOTGUN and SUNDAY equations. FARM VALUE and STOCK VALUE measure the value of farm property. As these values rise, the more likely it is that shotgun restrictions will be imposed in order to reduce potential damage to farms and stock. FARM VALUE and STOCK VALUE are also predicted to have positive coefficients in the SHOTGUN equations.

Intrastate variation in the regulations led me to omit Michigan and Minnesota from the SHOTGUN equations and New York from the SUNDAY equations. However, this variation also supports the model. In New York, Sunday hunting is allowed in areas (Upstate) where landholdings are larger and where alternative uses of the land are lower. Similarly, Michigan and Minnesota allow rifles for hunting big game in regions (Upper Peninsula and North Woods respectively) where landholdings are larger and where the likelihood of hunter accidents and property damage is lower.

Legal Classification of Animals

State laws classify species into such categories as big game, small game, migratory game birds, upland game birds, furbearers, predators, nongame animals, and endangered species. Most of these categories mean that killing is restricted, but for nongame animals and predators restrictions are few or nonexistent. All states consider deer, elk, geese, and pheasants to be game, but bobcats, coyotes, foxes, and porcupines are not treated uniformly.(11)

When a state classifies wildlife it assigns rights to private individuals and public agencies. Like other wildlife regulations, I expect classification to depend on the net valuation of the wildlife as well as the costs of landowner contracting. For instance, the more cheaply a landholding pattern allows landowners to own rights to wildlife stocks, the less interest the state has in protecting the wildlife. In addition, it is more likely that a wildlife stock will be protected when its value is high.

The legal status of coyotes offers a chance to test these implications. Coyotes are valued for their pelts, but they also impose costs, mostly by killing domestic sheep. For coyotes, the net value of the population is reduced when animals with valuable winter pelts are killed during summer months when fur values are low. The value of these stocks can be enhanced by restricting harvest to those months when fur values are highest.(12) At the same time, damage to sheep flocks can be reduced by allowing coyotes to be killed year around.

The value of coyote pelts and the amount of property they damage vary geographically, making cross-sectional analysis possible. In 1986-87, coyotes brought from $10 to $100, depending on pelt color, quality, and size. Pelt characteristics are correlated with the subspecies which are correlated with geographic regions. In general, the fur from coyotes taken in northern states brings higher prices because of a more developed winter pelage. Fur prices do not vary because of transportation costs, but only because of quality determined by "natural" factors. Price is truly exogenous with respect to the legal classification.

COYOTE measures the animals' legal status; a value of one indicates that states protected coyotes as game or furbearers and there was a restricted season on killing them. Because higher pelt prices indicate that coyote stocks are more valuable, higher prices are expected to make it more likely that coyotes receive protection from the state. Thus, COYOTE is predicted to be positively related to the price of the pelts; PELT is the mean price for pelts taken in each state.

On the other hand, states that have valuable sheep stocks are less likely to protect coyotes and more likely to treat them as predators. SHEEP, the total market value of sheep in a state, represents the value of private livestock that may be damaged by coyote predation. As the potential for damage rises, the likelihood of state protection decreases. COYOTE is predicted to be negatively related to the level of damage to livestock as measured by SHEEP.

FARM SIZE represents the ability of landowners to establish rights to attributes of the species; as landholdings increase in size, the establishment of private rights is more likely. Because contracting costs decline as landholdings become larger, states with large landholdings are less likely to be involved in protecting coyotes. Thus, COYOTE is predicted to be negatively related to FARM SIZE.

Using data from forty-nine states for the 1986-87 season, equation (2) is a logit estimate of the effect that FARM SIZE, PELT, and SHEEP have on the probability that a state will classify coyotes as a valuable game animal.(13) (2) In(Pr COYOTE = 1/Pr COYOTE = 0)
 = -2.6960 - [0.0008.sup.*]FARM SIZE

 (2.564) (1.0317)

 + [0.1564.sup.*]PELT - [0.0004.sup.*]SHEEP,

(2.7523) (2.8782) where the values in parentheses are asymptotic t-statistics. For each variable the estimated coefficients support the predictions. The chi-squared statistic is 23.0137 (with three degrees of freedom); the Chow [R.sup.2] is 0.7579; and 85 percent of the dependent variables are correctly predicted.(14)

Laws regulating the hunting of coyotes might alternatively be explained as the outcome of interest group competition between woolgrowers and fur trappers for political favors. One might hypothesize that as woolgrowers (trappers) become more powerful it is more likely that coyotes will be classified as predators (furbearers). The analysis above is consistent with the interest group hypothesis and, at the same time, is more general. My approach is consistent with interest group theory because the relative "power" of woolgrowers to trappers depends on the relative value of sheep to coyote pelts. An influential trapping lobby is unlikely to emerge in an area where fur-bearing animals have pelts of no value. The model is more general because I also include the costs of owning land for various uses. I rely on market data rather than on measures of interest group strength (such as size of the group) that do not consider the source of interest group strength. In this case a theory that stresses efficient institutions generates the same predictions as a theory that stresses interest group rent seeking.


Property rights to wildlife, although imperfect, exist in the form of private and public constraints. Most apparent, private landowners control access rights, and government agencies regulate hunting and other uses. These specific property rights exist because they economically mitigate the wealth dissipation that results from incomplete ownership. Evidence from logit regressions and literary sources indicate that private hunting rights, variation in state hunting regulations, and legal classification of species vary according to differences in landownership patterns and wildlife values. It can never be proven if the mix of private and public wildlife institutions is efficient, but my evidence shows that given exogenous changes both private contracts and government regulations shift in ways consistent with wealth maximization. [Tabular Data 1 and 2 Omitted]

(1)These assumptions focus on the crucial problem that arises when wildlife space requirements differ from the area owned by a single landowner. (2)In reality, there is no fixed acreage for a particular agricultural use, but this construction distinguishes wildlife from other attributes of land. (3)There are additional contracting costs if the agricultural land does not exactly coincide with the location of the wildlife territory. (4)In Great Britain, however, landowners face lower private contracting costs because nearly all land is privately owned in relatively large tracts and wildlife populations have small territories. As a result, wildlife is privately owned and government plays a trivial role in wildlife regulation [Lueck 1989]. (5)See Reiger [1986-90]. (6)The absence of data on the size of wildlife territories by ranch prevents a test of the last implication. (7)An animal that is worth $50 to a hunter but imposes no costs on the landowner yields $50 in gains from trading the right to hunt. But an animal that is worth $50 to the hunter and causes $10 damage to the landowner yields $60 in gains from trade. Some landowners have actually refunded money to hunters that shot does (female deer); they paid for damage control because females are less valued net of damage costs than bucks (males). (8)Per capita income and the fraction of federal land were included as control variables in all equations shown in Table II; the other coefficient estimates and significance levels were unchanged. NONFARM FRACTION and STOCK VALUE were not used in the same equations because of nearly perfect negative correlation. A variable measuring statewide Christian church adherents was added to the SUNDAY equations to control for religious pressures. This addition refuted the idea that religious pressures determine hunting rules and did not alter the estimates in Table II. (9)NONFARM FRACTION is all non-farm and non-urban land, including timber, park, desert, and other land with lower densities of people and property than farmland or urban land. In other specifications, I replaced NONFARM FRACTION with a variable that included urban land as non-farmland; the estimated equations were almost identical to those presented. (10)There are shotgun requirements in areas where agricultural property values are high, such as Iowa and Indiana; there are no Sunday bans in those states. Neither restriction is found in western states, where landholdings are large and wildlife values are relatively high. (11)States also recognize that game species may cause damage and routinely grant landowners the right to kill specific animals that damage property. (12)This prediction might seem at odds with the model because it says that an increase in wildlife value will increase the state's role. But states actually own nearly all rights to wildlife (except some associated with the habitat) so that an increase in the value of their asset is likely to cause them to increase their enforcement of their rights. This does not preclude landowners from increasing their enforcement of rights to the habitat which, in this case, would mean trapping rights. (13)PELT is from Miranda [1986-87]. FARM SIZE and SHEEP are from U.S. Department of Agriculture [1987]. (14)I omitted Hawaii because it has no coyote population. Per capita income was also included in other equations as a control variable; the coefficient estimates and significance levels were virtually unchanged. Similar analyses conducted for beaver and bobcat classifications offer additional, although somewhat weaker, support.


Alchian, Armen A. "Uncertainty, Evolution, and Economic Theory." Journal of Political Economy, June 1950, 211-21.

Barclay, John and Karl E. Bednarik. "Private Waterfowl Shooting Clubs in the Mississippi Flyway." Transactions of the 33rd North American Wildlife and Natural Resources Conference, 1968, 133-42.

Barzel, Yoram. Economic Analysis of Property Rights. Cambridge: Cambridge University Press, 1989.

Becker, Gary S. "A Theory of Competition Among Interest Groups for Political Influence." Quarterly Journal of Economics, August 1983, 371-400.

Becker, Gary S. and Kevin M. Murphy. "The Family and the State." Journal of Law and Economics, April 1988, 1-19.

Howard, Reese. Manager for United Sportsman Club, Salt Lake City, Utah. Personal Interview, September 1990.

King, Steven T. and J. R. Schrock. "State Wildlife Regulations." Controlled Wildlife. Lawrence, Kansas: Association of Systematics Collections, 1985.

Lueck, Dean. "The Economic Nature of Wildlife Law." Journal of Legal Studies, June 1989, 291-325.

McManus, John. "The Costs of Alternative Economic Organization." Canadian Journal of Economics, August 1975, 334-50.

Miranda, Tom. "Fur Markets." Fur-Fish-Game: Harding's Magazine, various issues 1986-87. Montana State University. 1986 Montana Fee Hunting Survey, Bozeman, 1986.

Peltzman, Sam. "Toward a More General Theory of Regulation." Journal of Law and Economics, August 1976, 211-40.

Reiger, George. "Conservation." Field and Stream, various issues, 1986-90.

Stigler, George J. "The Theory of Economic Regulation." Bell Journal of Economics, Spring 1971, 3-21.

Texas Parks & Wildlife Department. Deer Hunting Lease Register. Austin, 1986.

Tober, James. Who Owns the Wildlife? Westport, Connecticut: Greenwood Press, 1981. United States Department of Agriculture. Agricultural Statistics, 1986.

Washington, D.C.: U.S. Government Printing Office, 1987.

United States Department of Commerce. Statistical History of the United States. Washington, D.C.: U.S. Government Printing Office, 1970.

United States Department of Commerce. Statistical Abstract of the United States, 1985.

Washington, D.C.: U.S. Government Printing Office, 1986. Wittman, Donald. "Why Democracies Produce Efficient Results." Journal of Political Economy, December 1989, 1395-424.

DEAN LUECK Department of Economics, Brigham Young University. Thanks to Doug Allen, John Antle, Yoram Barzel, Ron Johnson, Sumner LaCroix, Wally Thurman, Doug Young, and especially Terry Anderson for their valuable comments, and to Marianne Keddington for her editing. I also benefited from the comments of two anonymous referees as well as participants in seminars at Brigham Young University and Clemson University. Research support was provided by the Earhart Foundation through the Political Economy Research Center.
COPYRIGHT 1991 Western Economic Association International
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1991 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Leuck, Dean
Publication:Economic Inquiry
Date:Apr 1, 1991
Previous Article:Bank panics, suspensions, and geography: some notes on the "contagion of fear" in banking.
Next Article:The Hayek hypothesis in experimental auctions: institutional effects and market power.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters