Elephants: Reply

by Michael Kremer, Charles Morcom
Elephants: Reply
Michael Kremer, Charles Morcom
The American Economic Review
Start Page: 
End Page: 
Select license: 
Select License

Elephants: Reply

Erwin H. Bulte et al. (2003) note that the rules of the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES) ban trade in products from endangered but not extinct species. They argue that govern- ments may therefore have incentives to strate- gically kill off all their elephants so as to secure the end of the ban on ivory trading. This incen- tive will be stronger if governments hold stock- piles of ivory. Bulte et al. therefore advocate that outside conservation organizations rather than national governments hold any ivory stocks and that CITES be modified either so that trade could not be resumed if African govern- ments deliberately destroyed their remaining populations of elephants or so that trade in ivory could be resumed at some sustainable elephant population. The Comment is interesting and we agree with the policy suggestions. Nonetheless, we feel that the scenario sketched by the authors of deliberate extinction is unlikely.

It is useful to start by reviewing the differ- ence between our approach and that of Bulte et al. In "Elephants" (Kremer and Morcom, 2000), we argue that the economics of storable natural resources, such as ivory, differ fundamentally from those of nonstorable open access resources, such as fish. Future overharvesting of storable resources will reduce long-run yield, or, in extreme cases, lead to extinction. Prices of the resource will therefore be high in the future, which, by arbitrage, implies they will also be high in the short run, stimulating increased har- vesting or poaching. Thus, expectations of high harvesting rates can be self-fulfilling and there may be multiple equilibria: one in which the species is driven to extinction and one in which

* Kremer: Department of Economics, 207 Littauer Center, Harvard University, 1875 Cambridge Street, Cambridge, MA 02138; Morcom: J.P. Morgan Securities, Inc., 270 Park Avenue, New York NY 10017.The views expressed in this Comment are those of the authors and should not be interpreted as those of J.P. Morgan Chase.

it survives. A government which seeks to pre- serve the resource can eliminate the extinction equilibrium if it can ex ante credibly commit to endangered species laws that mandate that if a species nears extinction the government will spend enough to protect the species even if this would not be justified on cost-benefit grounds

ex post.

We then consider the case of a government that values conservation and would be willing to incur the expenditures necessary to preserve the species if it started out with a large stock of the resource, but would not find preservation opti- mal if it starts out with only a small stock and cannot credibly commit to following a strategy that is not subgame perfect. We argue that such a government could eliminate the extinction equilibrium by building up stockpiles of the storable good and threatening to release them on to the market if the species became extinct. This would eliminate the extinction equilibrium by depressing the price in the event of extinction, which would make poaching less attractive now.

While we consider the globally optimal pol- icy for a government that values conservation in each of these cases, Bulte et al. take the insti- tutions embodied in CITES as given, and con- sider the behavior of governments that do not greatly value survival of the species. They argue that under the provisions of CITES, trade in ivory is illegal now and would remain illegal unless the population returns to levels much higher than the current stock, but that the trade would become legal if elephants were driven to extinction. They then argue that given their es- timates of the tourism value of elephants, the crop damage elephants cause, and the value of ivory, countries with elephants would be better off strategically killing all their elephants so as to secure the end of the ban on ivory trading and then selling any initial ivory stockpile as well as the ivory collected by killing off elephants. For their preferred parameters, this would be opti- mal even if countries started with zero ivory


stockpiles, but building up ivory stockpiles would increase incentives to drive elephants to extinction.

We think deliberate strategic extinction is unlikely for several reasons:


Twenty-five African countries have at least 1,000 elephants (African Elephant Database, 1995) and many other countries preserve elephants in zoos. Under the au- thors' assumptions, any individual coun- try would be better off preserving a small stock of elephants if it expects at least one other country to preserve some elephants since in this case the ban on ivory trade would be sustained. This is because with relatively small populations of elephants the tourism benefits of elephants outweigh the losses from destruction of crops. Hence there is an equilibrium in which all coun- tries preserve their elephants. It seems ex- tremely unlikely that all countries could coordinate deliberately killing off elephants to secure the end of the CITES ban on ivory trade. If a country expects even a single holdout, or expects elephants to be pre- served in zoos abroad, then it would be self-defeating for it to wipe out its own elephant population. Not only is survival of elephants the risk-dominant equilibrium, it may be the only equilibrium. African coun- tries are heterogeneous in their tourism ben- efits from elephants, in the damage to crops elephants cause, and even in the value lead- ers attach to preserving elephants. If even a single country would prefer to preserve some elephants, there is no equilibrium in whlch other countries drive elephants to extinction.


While the model allows for tourism benefits from wildlife, it does not allow for the possibility that aid from the rich countries would depend in part on the policies of African countries toward their wildlife. Given that aid represents more than 10 per- cent of GDP in some African countries, that African countries would likely do great harm to their image in the West by delib- erately eradicating elephants, and that any country that tried to preserve elephants in the face of an eradication campaign would

be seen favorably, it seems unlikely that such a cartel could succeed.

(3) Note also that the attractiveness of collud-


ing to secure the extinction of elephants depends on the assumption that the ban on ivory trade would be lifted following this extinction. The authors point out that the convention covers only endangered species and not extinct species. We believe that this is a somewhat overly legalistic interpreta- tion of the treaty. Few experts would con- sider elephants actually endangered right now; yet the ivory ban is in place because powerful countries wanted it in place. There would certainly be strong pressure from environmental groups in the West not to certify extinction and lift the ban following a deliberate campaign to drive the species to extinction, and under the rules of the treaty the ban could not be lifted without a two- thirds vote of the parties. So if enough countries refused to certify that the animal is extinct, the ban would remain in place. It seems unlikely that Western countries would certify the extinction and remove the ban.

(4) Moreover, the authors' analysis assumes that without extinction the ban would only be lifted if the elephant population were restored to its level of 1.2 million (or in one simulation 750,000) prior to the imposition of the ban. However. if the ban were lifted at a lower level, countries would not have incentives to poach so as to secure the end of the ban. Lifting the ban at this lower level would allow a stable population to survive. It is unclear why the iuthors as- sume that the ban would only be lifted if the population returns to 1.2 million or 750,000. The imposition of the ivory ban may have been driven as much by the rapid fall of elephant populations as by their level, so there is little reason to assume either population level is critical. In fact, the ban has already been loosened to allow countries that have demonstrated that their elephant populations have increased, in par- ticular Namibia, Botswana, and Zimbabwe, to sell ivory from culled elephants. If coun- tries that maintain stable populations will be able to trade in ivory, then the motive for


strategic extinction under their analysis is no longer present.

(5) Finally, the analysis of ivory prices in the comment is static and thus does not take into account a key point of our paper-that prices for storable goods will be influenced by expectations of future prices and sup- plies. One indication that the empirical analysis may be misspecified is that it coun- terfactually predicts that African countries would have already had incentives to delib- erately drive the animals to extinction.

The authors of the Comment note that many species are endangered because they are used to produce storable goods. Since many of these species will be spread among several countries, the same issues that arose regarding collusion would arise in those contexts as well. It is not clear that there are in fact any cases where few countries have a species and all of them would find it more profitable to drive the animal to extinction rather than to preserve it.

While we do not think that the danger of deliberate extinction is severe, we do agree with the policy implications of the paper that it may be preferable for conservation organizations rather than governments in poor countries to hold ivory stockpiles that would be used to flood the market in the event of extinction. It also makes sense for CITES to specify that the ban on trade will not be lifted if elephants become extinct and that countries judged to be adequately protecting elephants will be able to export ivory.


African Elephant Database. Occasional Paper Series of the IUCN Species Survival Com- mission, 1995.

Bulte, Erwin H.; Horan, Richard D. and Shogren, Jason F. "Elephants: Comment." American Economic Review, September 2003, 93(4), pp. 1437-45.

Kremer, Michael and Morcom, Charles. "Elephants." American Economic Review, March 2000, 90(1), pp. 212-34.

  • Recommend Us