Full text of DOI:10.1177/109114210203000507

PUBLICFaFitIhN/ARNECNETRSEEVEIKEIWNG AND FIXED-SHARE POOLS
A fixed-share pool is a given amount of resources to be divided among a given number of
claimants in which a claimant’s share of the pool is fixed exogenously. Consequently,
rent seeking to increase one’s share of the pool is effectively eliminated. However, if
claimants have sharing rights to more than one pool, rent seeking in the form of transfer-
ring resources from a pool where one has a small share to a pool where one has a large
share will generally occur. A game-theoretic analysis of this type of rent seeking is dis-
cussed and then applied to a number of governmental settings.
RENT SEEKING AND FIXED-SHARE
POOLS IN GOVERNMENT
ROGER L. FAITH
Arizona State University
This article is about rent seeking in government. Scholars in public
choice and public finance have long been aware of the incentives cre-
ated by the political system that induce individuals and organizations
to seek rents. Much of the literature on rent seeking has focused on
special interest groups seeking legislation that results in transfers of
wealth from taxpayers, or consumers in general, to the special interest.
The intention in this article is to distinguish and discuss another source
of rent seeking, the fixed-share pool, which has not, to our knowledge,
been discussed in the literature.1 A fixed-share pool is simply an
amount of resources that is to be divided among a given set of claim-
ants according to an exogenous sharing rule. Sharing pools differ from
the more familiar common-property pool in that the number of claim-
ants is set by an exogenous pool “manager,” and howmuch a valid
claimant can drawfrom the pool is determined by the sharing rule,
which is taken as exogenous by all claimants. The exogenous sharing
rule may allowfor variable shares in which a claimant’s share of the
pool can be affected by the claimant’s behavior, or the sharing rule
may establish fixed shares that are unaffected by claimant behavior.
An example of a fixed-share pool, taken from the academic world, is
an across-the-board pay increase in which each faculty member re-
PUBLIC FINANCE REVIEW, Vol. 30 No. 5, September 2002 442-455
© 2002 Sage Publications
442
Downloaded from pfr.sagepub.com at East Carolina University on April 23, 2015

Faith / RENT SEEKING AND FIXED-SHARE POOLS 443
ceives an equal percentage increase in salary. A faculty member’s
share of the pool of salary adjustment funds, then, is equal to the ratio
of the faculty member’s salary to the total salaries of all faculty mem-
bers in the academic unit. In this article, we will focus attention on
fixed-share pools.
Because the fixed-share rule is exogenous to claimants, there is no
incentive to seek a larger share of the pool by lobbying the pool man-
ager, department chairman, legislature, or whomever. However, if
there are two or more fixed-share pools, a claimant may be able to in-
crease his or her total take if rent seeking in the form of transferring re-
sources from one pool to another is possible. The equilibrium amount
of rent seeking and the distribution of resources among pools can be
modeled as the outcome of a noncooperative game played by claim-
ants seeking to have resources moved from a pool in which their share
is relatively small to a pool in which their share is relatively large.
The equilibrium aggregate expenditure on rent seeking is a func-
tion of the sharing rule. Thus, this article is also about rules. Sharing
rules showup in a variety of government settings such as revenue shar-
ing, general fund financing, and government pay practices of the sort
described in the introductory paragraph. We believe it would be a use-
ful exercise to investigate the positive and normative properties of the
various fixed-share sharing rules found in government.
From a normative perspective, there are good reasons for adopting
fixed shares, if controlling wasteful rent seeking is a goal of the rule-
choosing authority. In pools with no sharing rules or variable-share
rules, rent seeking often proves to be very costly socially. Under plau-
sible assumptions about competitor payoffs, the equilibrium aggre-
gate expenditures on rent seeking equal the transfer itself.2 A common
example in public choice is the competition among interest groups for
shares of the government’s tax revenues. In these and many other situ-
ations, it would seem that rent seeking would be brought to a halt by
simply making a group’s or person’s share of a particular pool of re-
sources independent of the group’s or person’s behavior. From the
viewpoint of simply preventing costly attempts to increase shares, a
sharing rule that fixes the shares received by each person in the pool
would seem to do the trick.3 However, as Buchanan (1980, 1983)
pointed out two decades ago, stifling rent seeking at one level may
Downloaded from pfr.sagepub.com at East Carolina University on April 23, 2015

444
PUBLIC FINANCE REVIEW
simply shift it to another, and such shifting will continue as long as
there is the possibility that the private expenditure of resources will
positively affect the spender’s expected wealth. Buchanan gave the
example of licenses. Individuals will seek to obtain licenses that grant
the holder monopoly power over some service. In so doing, they can
capture some of the consumer surplus that would otherwise accrue to
buyers if market conditions remained competitive. Competition for li-
censes, though, will reduce the net rents of the license holder. Once the
net return to holding a monopoly license is driven to zero, rent seeking
shifts to a higher level—competition to gain control of the license-
bestowing authority. If rent seeking is stymied at this level—say, by
legislation that directs all licensing revenue to the government treasury—
rent seeking by interest groups for the privilege to spend this revenue
on public projects or expenditures beneficial to the interest group
should be expected to emerge. The message from all this is that rent
seeking of one form or another is probably inescapable in governmen-
tal settings. The best the economist can do is to describe the activity
and propose institutions or rules that will minimize the social cost of
rent seeking. Our model of rent seeking in fixed-share pools, which
we present below, reflects this sentiment.
POOL GAMES
The competition for transfers in fixed-share pools will be called a
pool game. The idea behind these games is that the players compete to
have resources allocated from a sharing pool in which they have a less
favorable share to a pool where they have a more favorable share. As
an introductory example from a different arena, consider the problem
of making intergenerational transfers in the form of intervivos gifts
and inheritance.4 There is a family of n children. The parents, in an at-
tempt to eliminate wasteful competition among their children for the
family estate, choose and communicate to their children fixed
rules for dividing the parents’estate when the parents die. The es-
tate could be divided up according to primogeniture (the eldest
child receives everything) or divided up equally among all children
or any other fixed-share sharing rule. Furthermore, the amount of inter-
Downloaded from pfr.sagepub.com at East Carolina University on April 23, 2015

Faith / RENT SEEKING AND FIXED-SHARE POOLS 445
generational transfers a child receives from his or her parents depends
not only on his or her share but also on howmuch the parents leave in
their estate, the inheritance “pool,” and other transfers made to the
child while the parents are still alive. Each child then has an incentive
to get his or her parents to make intervivos transfers to the child rather
than waiting until the parents die to get a payoff. Of course, the more
that is transferred in the form of intervivos gifts, the less that is avail-
able in aggregate for inheritance. Those children who are more highly
favored in the inheritance pool relative to the gift pool will seek to have
a smaller gift pool (even though that means less gifts for himself or
herself) and a larger inheritance pool. Those more highly favored in
the gift pool will do the opposite. For the altruistic parent, the problem
is to choose a set of shares that will minimize aggregate rent seeking.
One can showunder some minimal assumptions that equal shares will
accomplish the parents’ goal (see Faith and Tollison 2001).
Formally, suppose that there is a set of n players competing for
shares of two fixed pools of resources, which contain the amounts A
and B, respectively. Each player is entitled to a fixed share of pool B
equal to αi, i = 1 to n, where Σαi = 1 and some amount gi taken from
pool A (where Σgi = A). Each player then has a wealth of αiB + gi. A
player can augment his or her wealth by participating in a winner-
take-all game in which an amount t is transferred from pool A to pool
B, or vice versa, depending on the winning player’s initial shares in
pools A and B. If player k wins and elects to transfer t from A to B, ev-
ery player i receives an additional αit from pool B, and each non-
winning player receives [1/(n – 1)]t less from pool A. If player k wins
and elects to transfer t from pool B to pool A, every player loses αit
from pool B, and the all of the transfer t is captured by player k. Players
compete for the right to transfer t from one pool to another by making
an expenditure on influence ei. The probability pi that player i will win
the right to make a transfer is equal to i’s expenditure as a fraction of
total expenditures across all players, or pi = ei/Σei.5 Finally, we will as-
sume that one of the n players is favored over all of the others in the
sense that player n’s share of pool B is greater than any other player’s
share. A novel feature of this game is that the winning participant’s
costs comprise his or her own expenditure and the loss due to shrink-
ing his or her claim from the pool the amount t is taken from. Likewise,
Downloaded from pfr.sagepub.com at East Carolina University on April 23, 2015

446
PUBLIC FINANCE REVIEW
the gross transfer t is shared with all players according to each player’s
preassigned share of the pool the transfer goes to.
To start with the simplest case, consider a two-person pool game.
Assume that Player 2 is favored in that α2 > α1. Player 1, if he or she
wins the right to transfer t from one pool to another, receives t addi-
tional dollars for his or her private “account” in pool A. At the same
time, because the transfer t was financed by taking t dollars out of pool
B, of which Player 1 has a share α1, Player 1’s gain if he or she wins is
(t + α1(B – t)). If Player 1 loses, t dollars are transferred away from
his or her private account in pool A to pool B for a net payoff of (–t +
α1(B + t)). The expected payoffs, net of expenditures, to nonfavored
Player 1 is
(1)
P1(t + α1(B – t)) + P2(–t + α1(B + t)) – e1.
The favored player has t dollars subtracted from or added to pool B de-
pending on whether Player 2 loses or wins. So Player 2’s net expected
payoffs are
(2)
P1(α2(B – t)) + P2(α2(B + t)) – e2.
The first-order conditions (assuming an interior solution) for a Nash
solution to this game are
2t α2 e2 – (e1 + e2)2 = 0,
(3)
2t α2 e1 – (e1 + e2)2 = 0,
(4)
for Players 1 and 2, respectively, implying thate1* = e*2 = t α 2 / 2and a
total expenditure of tα2.
Because the game involves a pure transfer, the value of the game is
B – tα2, implying a social loss of tα2.
The more interesting case is when the number of players is greater
than two. To illustrate the solution in this case, let there be three play-
ers, with Player 3 being favored with a larger share of pool B than the
other players. In the three-person game, if the favored Player 3 wins,
the transfer t is financed equally by the two losing unfavored players.
Downloaded from pfr.sagepub.com at East Carolina University on April 23, 2015

Faith / RENT SEEKING AND FIXED-SHARE POOLS 447
The expected net payoffs to each of the three players when Player 3 is
the favored player are as follows:
(5)
(6)
(7)
P1(t + α1(B – t)) + P2(α1(B– t)) + P3(–t/2 + α1(B + t)) – e1,
P1(α2(B – t)) + P2(t + α2(B – t)) + P3(–t/2 + α2(B + t)) – e2,
P1(α3(B – t)) + P2(α3(B – t)) + P3(α3(B + t)) – e3,
where Pi = ei/(e1 + e2 + e3 ) and α1 + α2 + α3 = 1.
Thefirst-order conditions for a Nash equilibriumto thisgame are as
follows:
t(e2 + 3e3/2 – 2α1 e3) = (e1 + e2 + e3)2,
t(e1 + 3e3/2 – 2α2 e3) = (e1 + e2 + e3)2,
t(2α3 (e1 + e2)) = (e1 + e2 + e3)2.
(8)
(9)
(10)
Solving for the interior solution, equilibrium rent-seeking expendi-
tures are
e*i =[t(2α3 +1)][2α3 (2α3 +1) −(3 2 − 2α j )(4α3 −1)] /18α3; i, j = 1, 2; i ≠ j,
(11)
and
e*3 = t(2α3 +1)(4α3 −1) /18α3.6
(12)
Aggregate rent-seeking expenditures are
E = e1* + e*2 + e*3 = t(2α3 +1)/ 3.7
(13)
From (13), we see that aggregate rent-seeking expenditures depend
only on the size of the favored player’s share. The larger the favored
player’s share, the more that is spent on trying to transfer funds from
one pool to another. Notice also that aggregate expenditures are inde-
pendent of the individual shares of the unfavored players. All that mat-
Downloaded from pfr.sagepub.com at East Carolina University on April 23, 2015

448
PUBLIC FINANCE REVIEW
ters is the size of the favored player’s share. The obvious normative
implication is to set the favored player’s share to its lowest possible
value—one third in the limiting case. That is, equal division of the
pool minimizes rent-seeking costs.
If we retain the basic structure of the pool game—one favored
player and all players seeking to transfer a fixed amount t from one
pool to another—then we can generalize our results to n players. In
this case, the equilibrium aggregate expenditure of the set of n – 1
unfavored players is
n −1
n −1
(14)
e* = t[(n − 2) + 2α ][n − 2
α ] / 2 n2α
∑
∑
i
n
i
n
i =1
i =1
and for the favored player is
e*n = t[(n − 2) + 2αn ][2(n −1)αn −(n − 2)] / 2 n2αn.
(15)
(16)
The equilibrium aggregate expenditure is
n
E =
e* = t[(n − 2) + 2α ] / n.
i n
∑
i =1
Differentiating E with respect to n yields ∂E/∂n = 2(1 – αn)/n2 > 0.
Aggregate rent-seeking expenditures increase with the number of
players, and as n gets large, total expenditures approach t.
Although the game described above is simplistic in both its explicit
form and the characterization of strategies and payoffs, the game and
some of its variants can be a useful for modeling behavior in a number
of governmental settings.
APPLICATIONS
Pool games may be representative of a number of government ac-
tivities such as budget determination, pay practices, and perhaps
more. We will look at each in more detail.
Downloaded from pfr.sagepub.com at East Carolina University on April 23, 2015

Faith / RENT SEEKING AND FIXED-SHARE POOLS 449
MERITOCRACY IN THE BUREAUCRACY
A common feature of governmental salary practices is the equal
treatment of employees in terms of salary shares. Anyone who has
worked for a state-run university, for example, has likely experienced
“across-the-board” pay increases. The government—the legislature
or the Board of Regents—allocates a certain sum of money to the uni-
versity for salary adjustments that represents some percentage in-
crease x over the total wage bill in the university. Each eligible person
then receives a salary increase equal to x percentage of the person’s
current salary. That is, all faculty (I will ignore staff and academic pro-
fessionals in this discussion) receive a fixed share of the salary adjust-
ment pool. Individual raises in this setting are independent of marginal
productivity. The “meritocracy,” as such systems are sometimes
called, is frequently denounced by those faculty members who re-
cently have had “better” years than the average faculty member in
their academic unit but is welcomed by those who have had “poorer”
years than their colleagues. Our model of pool games offers a possible
rationalization for these pay schemes. Suppose that faculty rewards
come in both pecuniary and nonpecuniary forms. Perquisites such as
money for travel to conferences, newcomputer equipment, greater of-
fice space, and favorable teaching schedules are examples. Nowsup-
pose that salary comes out of one pool and perquisites out of another,
and both pools are fixed in size. To reduce the amount of unproductive
faculty rent-seeking activity such as complaining and/or cajoling, a
department chair who is not interested in flattery but in productivity
maychoosetofix the shares an individual faculty member receives out
of each pool. If the share received from one pool is not equal to the
share received from the other pool, a faculty member may still seek
rents by trying to have resources moved from one pool to the other.
The difference in this game as compared to our general game de-
scribed above is that a dollar of salary is not likely valued the same as a
dollar’s worth of perquisites. Suppose that a transfer of t dollars from
the salary pool, pool B, is worth kt dollars, k < 1, when put in the per-
quisite pool, pool A. Salary dollars are preferred to travel dollars, for
example. So if the unfavored faculty member i wins the rent-seeking
game and is able to move t from pool B to his or her personal allocation
in pool A, he or she loses the share αit but gains kt. If the favored mem-
Downloaded from pfr.sagepub.com at East Carolina University on April 23, 2015

450
PUBLIC FINANCE REVIEW
ber wins and transfers money away from the perquisite pool, each of
the n – 1 unfavored faculty members loses kt/(n – 1). The expected
payoffs from playing this pool game in a three-person department, as-
suming that both unfavored players have the same k, are as follows:
P1(kt + α1(B – t)) + P2(α1(B – t)) + P3(–kt/2 + α1(B + t)) – e1,
P1(α2(B – t)) + P2(kt + α2(B – t)) + P3(–kt/2 + α2(B + t)) – e2,
P1(α3(B – t)) + P2(α3(B – t)) + P3(α3(B + t)) – e3.
Aggregate rent-seeking expenditures are E* = 2tα3[3k + 2α3 – 2]/[2k +
6α3 – 2]. The derivative of E* with respect to k is positive, implying
that as an unfavored faculty member’s valuation of perquisites rises,
total rent-seeking cost rises, as one would expect because a transfer is
more valuable.
Although the assumption of one favored player is restrictive when
the number of players is larger than three, our model would apply to
any bureaucratic setting where employees are paid in some lock-step
fashion. The insight from our analysis of fixed-share pool games is
that when there is more than one pool that an employee has claim to,
and if rent seeking is costly to the administrator, equal treatment of
employees can be rationalized for reasons other than bureaucratic
preferences for equity or administrative simplicity.
LOCATION OF PROGRAMS
A different type of rent seeking is that undertaken by special inter-
est groups seeking wealth transfers from the government. The behav-
ior of special interest groups within a representative democracy has
been studied, modeled, and empirically tested extensively.8 An un-
stated assumption in this literature is that there is a single source, or
pool in our terminology, of special interest benefits. Suppose a special
interest group is successful in getting the legislature to create and fund
a program the group valued. Suppose further that two bureaus could
administer the program and that each bureau has several other inter-
ests that have pet programs the bureau is administering. The program’s
funds, if not earmarked, become part of the bureau’s pool of funds,
Downloaded from pfr.sagepub.com at East Carolina University on April 23, 2015

Faith / RENT SEEKING AND FIXED-SHARE POOLS 451
which are to be distributed across all of the bureau’s programs. If the
newprogram is placed in bureau B and thereby receives a certain share
of bureau B’s budget, the special interest group may seek to be moved
to bureau A if it expects to receive greater total benefits due to either
having a larger share in A or bureau A having a larger pool of funds.
In terms of our basic pool game, the amount t that players are seek-
ing to move from one pool to another is the program’s initial entitle-
ment. Programs that are “favored” in pool B will attempt to prevent
other programs from leaving the pool, just as programs that are
“unfavored” will attempt to move to a pool where they receive a higher
payoff.9
EARMARKING VERSUS GENERAL FUND FINANCING
A program, bureau, or agency may fare better under one financing
plan than another and thus engage in wasteful rent seeking in an at-
tempt to get a better sharing arrangement. Consider a bureau whose
programs are funded from two different sources. Each source is a pool
from which the bureau can draw. In this application, pool B is a pool of
tax dollars that constitute a state government’s general fund, for exam-
ple (cf. Buchanan 1963). A certain share goes to public safety, a cer-
tain share goes to education, and so on. Shares are fixed, perhaps
based on historical precedent or past legislative commitments that
have somehowbecome institutionalized. Pool A consists of tax dollars
that are earmarked for specific programs. A bureau, of course, would
always want to have sole access to funds rather to share funds with
other bureaus, unless the general-fund pool was sufficiently large or if
the bureau was favored over others in the shared-fund pool. Applying
our model of pool games, a bureau that is unfavored in the general-
fund pool would seek to move money from the general fund to its ear-
marked fund. A bureau that is favored in the general fund would seek
to have dollars going to earmarked programs moved into the general
fund. For example, if pool A were earmarked for education and public
safety were financed and favored in the general fund, the public safety
bureau would lobby the legislature to move education’s earmarked
funds to the general-fund pool. Of course, if education were unfavored
inthegeneral fund, itwould seek the opposite movement of monies.
Downloaded from pfr.sagepub.com at East Carolina University on April 23, 2015

452
PUBLIC FINANCE REVIEW
DIFFERENTIAL RENT-SEEKING ABILITIES
To this point, we have assumed that all players in the pool game are
equally effective. The only thing that determines a player’s probability
of winning is the amount of dollars spent on rent seeking. However, it
is likely that one bureau’s rent-seeking efforts may be more effective
in influencing the legislature than another bureau’s. Differences in
players’ abilities can be incorporated into our model by weighting
each player’s lobbying effort ei by a cost factor wi. We will now think
of the variable e as representing units of “influence.” Less able players
would have a higher unit cost (a larger w) of producing a unit of influ-
ence. The expected net payoff for unfavored Player 1, with a cost fac-
tor of w1, would equal (in a three-person pool game) the following:
P1(t + α1(B – t)) + P2(α1(B – t)) + P3(–t/2 + α1(B + t)) – w1e1.
Players 2 and 3 would have their effort expenditures weighted by w2
and w3, respectively. At the solution to this game, the number of units
of influence purchased by the favored player is
e*3 = t(2α3 +1)(2α3(w1 + w2 ) − w3 ) / 2α3(w1 + w2 + w3 )2.
Notice that if the favored player’s cost factor were high enough, e*3
would be nonpositive, and Player 3 would refrain from participating in
the game. The aggregate amount of influence generated is equal to
E* = t(1 + 2α3)/(w1 + w2 + w3). The greater the weight wi, the more ex-
pensive and thus less effective rent seeking is for player i. It is clear
from our expression for E* that if there is a reduction in any player i’s
effectiveness (an increase in wi), total rent-seeking effort falls.
NORMATIVE CONSIDERATIONS
The central normative implication of our model of fixed-share pool
games is that equalizing shares minimizes aggregate rent seeking. The
concept of equal treatment of fiscal participants to reduce or eliminate
inefficient behavior or costly rent seeking can be found elsewhere in
the public finance literature. Buchanan (1994), for instance, has ar-
Downloaded from pfr.sagepub.com at East Carolina University on April 23, 2015

Faith / RENT SEEKING AND FIXED-SHARE POOLS 453
gued for the application of a “generality principle” as a constitutional
constraint against nongeneral (i.e., discriminatory) taxation to reduce
the wasteful competition of rents that would take place in an otherwise
unconstrained majority-rule political system. The argument for gen-
eral taxation, defined to mean uniform treatment of all members of the
inclusive voting constituency, is that it will reduce the incentive to
seek rents through the democratic process. Under majority rule and
multiple issues, one can always find a majority coalition of voters that
would prefer some alternative set of discriminatory taxes to any status
quo. If the status quo were equal tax rates on all individuals, a new al-
ternative could be found with differentially low tax rates on the major-
ity and high tax rates on the minority. However, even if the generality
principle is not violated, pool games can still arise. For example, let a
certain amount of tax revenue raised by a set of nondiscriminatory
taxes be placed into different funds, or pools. If claimants to these
funds—bureaus, coalitions of special interests, or combinations of the
two—can shift funds from one pool to another by appeal to the legisla-
ture, apoolgame, alongwithitsattendantrent-seekingcosts, willarise.
CONCLUSION
In this article, we have constructed a simple game that represents
howagents in a governmental setting might behave under certain shar-
ing rules. Instead of concentrating on agents competing to increase
their share of government expenditures, we have concentrated on the
behavior of agents competing to expand the size of the resource pools
to which their fixed shares apply.
In the introductory paragraph, we said that this article was also
about rules. Our analysis of rent-seeking pools and pool games has fo-
cused primarily on the positive aspects and possible applications of
these games. We also noted one normative implication regarding the
setting of sharing rules. Any fixed-share rule will eliminate wasteful
competition to establish rights to bigger shares of the pool’s resources.
However, given that fixed-share rules may result in rent-seeking pool
games, the question of what the fixed shares should be set to becomes
relevant. If we assume that the goal of the government is to minimize
Downloaded from pfr.sagepub.com at East Carolina University on April 23, 2015

454
PUBLIC FINANCE REVIEW
aggregate rent seeking, then the implication is clear. Equal shares are
best. In some cases, shares may have to adjust slightly to account for
differences in the rent-seeking abilities of the players, but the principle
is the same. If one player receives a larger than pro rata share of the
pool’s resources, rent seeking will be more prevalent than if shares are
equal. Of course, another welfare-enhancing practice would be sim-
ply not to have any favored players. In a political environment, how-
ever, that may be difficult to achieve.
We should also point out that it may not politically feasible to main-
tain, let alone initially set, equal fixed shares of tax revenues if for
no other reason than there is little interest from government to do
so. In fact, if government authorities were the beneficiaries of any rent-
seeking expenditures that may arise from a pool game, then we would
expect them to adopt highly unequal fixed shares to encourage rent
seeking.10 And, of course, we have ignored in our analysis any possi-
ble efficiency rationale for unequal shares.
That our model of rent-seeking pools can be considered descriptive
of modern bureaucratic arrangements or other governmental institu-
tions and practices is certainly debatable. We have only suggested
some areas where our model of pool games would seem to apply. Our
analysis, both positive and normative, is intended as another step in
furthering our understanding of rent-seeking behavior within the gov-
ernment sector.
NOTES
1. The literature on rent seeking, however, is voluminous. The probability that others have
discussed rent seeking in fixed-share pools is probably not zero.
2. See Hillman and Riley (1989) for a thorough discussion of political rent seeking and dis-
sipation of net rents.
3. SeeBuchanan (1983) foranapplicationofthis principleto rules for dividingupestates.
4. See Faith and Tollison (2001) for a more complete discussion of inheritance as a pool
game.
5. Thus, our pool game is a Tullock type of imperfectly discriminating contest (see Hill-
man and Riley 1989).
6. Details for this and all other derivations and calculations are available from the author on
request.
Downloaded from pfr.sagepub.com at East Carolina University on April 23, 2015

Faith / RENT SEEKING AND FIXED-SHARE POOLS 455
7. Equation (12) implies that the equilibrium expenditure for Player 3 is positive if and
1
1
only ifα3
> >
4, which is always true because Player 3 is favored,α3 3. It is easy to showfrom
Equation (11) that because αj must be less than α3 and α3 > 1 , that e*i > 0. Thus, an interior solu-
tion to the game exists.
3
8. Although much work on these topics has been done in the past decade, Mueller (1989)
remains an excellent survey to this literature.
9. The pool game where player i has the fixed-share αi in pool A and a different fixed-share
βi in pool B is more complex than the simple games modeled here and thus is not included in this
article.
10. On the other hand, if we follow the argument that restricting rent seeking at a lower level
will induce more rent seeking at a higher level, a wise bureaucrat may choose not to set shares too
unequal for fear of attracting other bureaucrats to compete for the rent-seeking expenditures of
the players.
REFERENCES
Buchanan, James M. 1963. The economics of earmarked taxes. Journal of Political Economy
71:457-69.
. 1980. Rent seeking and profit seeking. In Toward a theory of the rent-seeking society,
edited by J. M. Buchanan, R. D. Tollison, and G. Tullock, 3-15. College Station: Texas A&M
University Press.
. 1983. Rent seeking, noncompensated transfers, and laws of succession. Journal of Law
and Economics 26:71-85.
. 1994. The political efficiencyofgeneral taxation. National TaxJournal46(4): 401-10.
Faith, Roger L., and Robert D. Tollison. 2001. Inheritance, equal division and rent seeking.
Working paper, Department of Economics, Arizona State University.
Hillman, Arye L., and John G. Riley. 1989. Politically contestable rents and transfers. Econom-
ics and Politics 1:17-40.
Mueller, Dennis. 1989. Public choice II. Cambridge, UK: Cambridge University Press.
Roger L. Faithis a professor of economics at Arizona State University. He received his
Ph.D. in economics from the University of California, Los Angeles in 1973 and has
taught at the University of California, Santa Barbara and Virginia Polytechnic Institute
prior to Arizona State. He is the author or coauthor of numerous journal articles and
book chapters in the area of public choice and public finance.
Downloaded from pfr.sagepub.com at East Carolina University on April 23, 2015