Fig. 3. Simulated effort levels. Note: E1 is derived from a simulation where hunting technology is assumed to be con-
stant over the period, while E2 is derived from a simulation where hunting technology is allowed to advance. See text for
details. For the purposes of this figure, both indices of effort are normalised to equal one in 1950.
pose of the ban on trade was to reduce incentives to poach by restricting final demand. The price of
ivory has to fall by some 90 if the simulated 1987 population is to be maintained in the con-
stant and variable technology models if the latter is held at 1987 technology levels: an even larger
decline is needed in the variable technology model if the catchability coefficient is increased to a
value of 10. This is not very encouraging, in that it implies that the effectiveness of the ban at the
level of final demand has to be high if it is to have the required effect on the poachers.
5. Discussion
Any analysis of the economics of ivory poach- ing is going to be constrained by the paucity of
good data available, and in particular, time series data. This paper has looked at various ap-
proaches to the question of the relationship be- tween poaching and economic incentives. With
few direct measures of poaching effort one is reduced to indirect measurement of the links be-
tween economic incentives, enforcement activity and levels of poaching, and then inferring the
responsiveness of the underlying poaching effort.
Those studies that look directly at the determi- nants of poaching activity i.e. those in Section 2
all conclude that there is little evidence of any impact of the ivory ban in 1989, given the data
they have to work with. The simulation models Sections 3 and 4, by construction, contain a
positive relationship between poaching effort and prices, and hence one has a primea facia case for
the ban having a positive effect on elephant con- servation. However, what those models do reveal
is that significant poaching effort will remain even if the ivory price is reduced substantially, and
hence suggest that if the ban is to be effective, the link between final consumer and poacher has to
be practically absolute: any slippage may be suffi- cient to give sufficient incentives to poachers to
prevent recovery in elephant populations. In this context, the comment reported in Dublin et al.
1995 that ‘‘..donor countries in the north seem to feel that the problem of elephant poaching has
been ‘taken care of’ by the ban and that there is duced by some 75 as a result of using the
variable technology model Fig. 4. These simulations take no account of the po-
tential impact of reduced supply of ivory on prices. Barbier and Burgess 1989 provide esti-
mates of elasticities of demand for ivory which could be used to generate flexibilities, but Milner-
Gulland 1993 has raised doubts about these and suggests that the elasticity of demand may be
zero. Furthermore the simulated harvests near equilibrium are so small they may not provide
useful information on price responses. Of perhaps more relevance is the ivory price that is required
to maintain certain population levels, as the pur-
Fig. 4. Actual and simulated ivory yields. Note: Y1 is derived from a simulation where hunting technology is assumed to be
constant over the period, while Y2 is derived from a simula- tion where hunting technology is allowed to advance. See text
for details.
Table 3 Prices for ivory from hunter to first middleman USkg
Pre-ban Post-ban
Cameroon 4
Savanna 2
34–62 Forest
86 Tanzania
10–26 6–12
Malawi 21–42
52–87 36–40
46–56 Ivory Coast
Zaire 16
6 21–42
52–87 Malawi
18–24 Zambia
14–26 Source: Dublin and Jachmann 1992.
Although the structure of the ivory market may be known in general terms there is relatively little
work on quantifying the structure, and this infor- mation is needed if the economic activity of
poaching is to be controlled. The determinants of poaching effort is perhaps the most developed
area, but as we have seen, the limitation of poor data places extreme restrictions on what can be
achieved. Without that information, identifying the effects of the trade ban both ‘ex ante’ and ‘ex
post’ is fraught with difficulties and runs the risk of ‘‘..equating association and correlation with
causality, failing to identify and cut through con- founding factors, failing to replicate, failing to
balance, failing to control’’ Caughley, op cit p. 239. Given that the African elephant listing is
often viewed as one of the more successful exam- ples of the CITES listing mechanism, the difficulty
in showing its effect is of concern. Without estab- lishing this link the objective development of pol-
icy is severely hampered, and evaluating even incremental management, such as the recent
CITES downlisting of local elephant populations, and the associated requirement to assess any neg-
ative impacts of the resumption in trade on other range states will be problematic. The adoption by
the parties in 1997 of a Resolution to establish ‘‘a comprehensive, international monitoring system...
with the objective of measuring and recording current levels and trends of poaching and illegal
trade in ivory in African and Asian range states, and in trade entrepots’’ and assessment as to
‘‘whether, and to what extent, observed trends are the result of changes in the listing of elephant
populations in the CITES Appendices andor the resumption of legal international trade in ivory’’
Conf.10.10, cited in Gray, 1997, may go some way to filling this gap, but without an understand-
ing of the historical patterns of exploitation and the associated data such an analysis will be
incomplete.
Given that the management of elephants, and the ivory trade ban, have probably received more
attention from the social scientists than any other species, the implications of this paper for the
management of other species which are under threat are not hopeful.
no need to continue supporting law-enforcement efforts in the field’’ is particularly worrying. Con-
sistent data on prices is difficult to come by, but those reported by Dublin and Jachmann 1992
summarised in Table 3 suggests that prices have not fallen everywhere, implying a residual demand
for ivory, either held speculatively within Africa or smuggled illegally out of the continent. Even
where prices have fallen, the results outlined above would suggest that changes in prices of this
magnitude are not sufficient to significantly re- duce the level of poaching effort.
6. Conclusions
