CHAPTER 9 Supply and Demand In Political Markets
159
CHAPTER 9 Supply and Demand In Political MarketsIn chapter 8, we
concluded that the demand for public, sector output approximates
the demand of the median voter. This does not necessarily imply
that the output produced in the public sector corresponds to what
the median voter most prefers. In private markets, consumers can
sometimes find their Options limited by cartels that conspire to
reduce output and raise prices or by firms with monopoly power that
push buyers into tie-in sales. Firms with market power might be
able to price discriminate so that some customers can buy at more
favorable terms than others can. Similar problems can confront
demanders for public sector output. Chapter 8 noted the power of
special interests in the political process that produces public
sector output, so in the public sector it is also likely that some
"customers" are treated differently from others. Public sector
demanders who spend time and money lobbying legislators for
beneficial legislation show that they are politically aware and
that they are high demanders for certain types of legislation.
Demanders who lobby the legislature hope to demonstrate that they
control campaign contributions, votes, and other political support
so that if legislators help the special interests, those, interests
can in turn help the legislators. The legislature acts as a
political market where different interests can demonstrate their
demand for legislation. Legislatures then weigh the supply and
demand for legislation and pass the legislation that has the most
political support. This chapter examines how that political market
works by combining the, model, of public sector demand from the
previous chapter with models of public sector supply. Suppliers and
demanders interacts in the political arena to determine the
composition and level of public sector output, just as in private
markets the level and composition of output is determined by the
interaction of supply and demand. Suggesting that political
exchange has similarities to market exchange actually does not go
far enough, for both politics and markets function through exchange
and gains from: trade, and there actually is a political market
where public sector decisions are made. Political markets differ
from economic markets in important ways, of course, but both are
true markets where exchange and gains from trade determine output.
Even in market economies such as the United States, the public
sector produces well over a third of the nation's GDP, so
understanding how political markets work is very important in
gaining an understanding about bow economic resources are allocated
throughout the economy. SPECIAL INTERESTS AND GOVERNMENT PROGRAMS
Chapter 8 introduced the idea that special interests have undue
influence over political decisions because they have an incentive
to be informed about political decisions in areas that directly
affect them, while the general public is rationally ignorant of
most of the governments activities. A special interest program
provides concentrated benefits to a small group by spreading the
cost over a large group; each person in the large group pays
relatively little. A few dollars paid by everyone can turn into
thousands of dollars for each member of the special interest group.
The general public has little incentive to become informed about
the cost of a few dollars, whereas the special interest has a big
incentive to become informed about the possibility of a much larger
gain. If a legislator decides to vote in favor of the general
public interest and against a special
CHAPTER 9 Supply and Demand In Political Markets
160
interest, few in the general public will even know enough to be
grateful, but such a decision will cost the legislator the
political support of the special interest. The government budget is
not determined all at once in a forum in which Congress decides
what the median voter would most like. Rather, individual policies
and programs are decided one at a time in a forum in which special
interests are able to influence the outcome of legislation. The
surgeon general of the United States speaks out against cigarette
smoking and looks forward to the day when all public areas will be
smoke free, but, at the same time, Congress votes subsidies to
tobacco farmers, helping them to grow the crop the surgeon general
opposes. How can one make sense of these policies? The policies are
not determined together but rather in separate forums in which
special interests are able to influence legislation. There is no
mystery to seemingly contradictory policies when one understands
that the policies are determined separately and that special
interests have a substantial influence over government policies.
Special Interests and Political Exchange Special interests do have
substantial influence over legislative decisions, but they do not
typically get everything they want. The reason is that there are
also interests on the other side of any issue, and legislators must
weigh the demands of the special interests against the demands of
those interests on the other side of the issue. Tobacco interests
do have strong representation in Washington, as do the National
Rifle Association, the American Medical Association, the Airplane
Owners and Pilots Association, and many other organizations. These
groups interact with the legislature to try to get benefits for
themselves, but these benefits come at a cost to others, and those
others have some representation, too. Sometimes there will be one
wellorganized interest group opposing another. Other times
interests on the other side of the issue are simply members of the
general public who might notice that certain legislators tend to be
the pawns of special interests on many issues, so they vow not to
vote for them in the next election, The legislature is a political
market in which legislators must weigh interests on one side of an
issue against interests on the other. This political market can be
examined with the help of figure 9.1, which shows the political
support and opposition generated by providing special interest
benefits. Consider an example of an industry that wants a tariff
placed on foreign competitors to protect domestic manufacturers and
allow the domestic industry to charge higher prices. Such an issue
not only provides clearly defined special interest benefits to the
industry protected by the tariff but also produces costs on those
who would sell the imported goods, on consumers who would have to
pay more for the domestic goods, and on other domestic industries
that might find more demand for their products and more labor
available if other industries did not have tariff protection. In
addition, foreign manufacturers will be hurt by tariffs on their
products (and foreign firms do hire lobbyists to try to influence
domestic legislation). The benefits are concentrated on the
protected industry, but the costs are diffused among many producers
and consumers. How much tariff protection should the legislature
give the industry demanding it? Figure 9.1 shows that as the amount
of protection rises, the marginal political support to legislators
behind the tariff will decline. For example, an increase on a
tariff on shoes from 5 percent to 10 percent will benefit the
domestic shoe industry, and the industry will provide political
support to legislators who back the tariff. An increase from 10
percent to 15 percent would provide fewer benefits to the industry,
and an increase from 15 percent to 20 percent still fewer, so as
the quantity of special interest benefits increases, the marginal
political support to
CHAPTER 9 Supply and Demand In Political Markets
161
legislators from providing those benefits declines. Meanwhile, a
higher tariff would do increasing harm to those who they would be
hurt by the tariff, so the marginal political costs of increasing
the special interest benefit would rise as the benefit is
increased. If the only thing that mattered was what the special
interest wanted, and legislatures gave special interests everything
they asked for, the amount of special interest benefits produced
would be Q', out to the point where there would be no additional
benefit to the special interest. However, because there are
opponents to special interest benefits, legislators must also
consider the political costs of special interest programs and will
provide special interest benefits only to the point where the
marginal political benefits from the special interests out weigh
the marginal political costs from those who are opposed. In figure
9.1, Q* will be the level of special interest benefits
produced.1
Note that in the weighing of various interests, no interest gets
everything it wants. The special interest wanting benefits prefers
to get benefits of Q', but the political marketplace provides only
Q*, while those opposing the special interests prefer zero, and
again get Q*. Interests on both sides of the political market have
their demands weighed against each other to result in an
equilibrium amount of special interest benefits. Those who are
dissatisfied with the outcome in the political marketplace can
devote more resources to trying to change the outcome. If the
special interest asking for the benefit increases the resources it
devotes to trying to get political benefits, the Marginal Political
Support curve will shift upward and Q* will rise. If opponents
devote more resources, the Marginal Political Opposition curve will
rise and Q* will fall. In this way, the legislature acts as a
political marketplace, balancing the interests on all sides of an
issue.
CHAPTER 9 Supply and Demand In Political Markets
162
As the quantity of benefits to a special interest increases, the
marginal benefit to the interest declines, while the marginal
political opposition to the benefit increases. The legislature must
weigh the marginal political support from helping the special
interest against the marginal political opposition that helping the
interest generates to decide on the optimal amount of benefit, Q*,
to provide to the special interest. Representation of Interests and
the Organization of Congress The orientation of representative
democracy toward special interests is in part designed into the
political system because of the geographic nature of
representation. By design, representatives are elected to represent
the interests of constituents in their districts. Representatives
make no secret of trying to generate more benefits for their
constituents, and not infrequently those constituents will be
employed in a particular industry that might be looking for
benefits. Representatives whose districts contain navigable rivers
help constituents get waterway maintenance, locks and dams, and
flood control. In drier areas in the West, representatives get
federal money for irrigation and other water projects.
Representatives try to get military contracts for firms in their
districts, try to obtain new military installations (or prevent old
ones from closing), and try to get federal offices to open in their
districts. In urban districts, many industries will be represented
but no one industry will be dominant, with some exceptions like
Detroit, where the auto industry is a major force. In rural
districts, farming will be the main industry, and because there are
many rural districts, farming as an industry tends to have more
representatives than other industries. This overrepresentation of
farming as an industry is why there are many federal programs that
support agriculture when compared with other industries of equal
size. But because of the rational ignorance of voters,
representatives often will push interests that represent only a few
voters in a district. If the interest can provide political support
in the form of campaign contributions and favorable publicity,
there do not need to be a large number of constituents represented
by an interest for a legislator to be willing to help that
interest. Because represent on is geographically based, special
interests will tend to look for support from the representatives in
their geographical area. This does raise an important question.
Although a special interest can influence its representative to
support a program for its benefit, it may be able to influence only
a few representatives out of the many in Congress. How is it
possible for only a few legislators to gather enough support for a
special interest program that it can gain a majority of the
legislatures votes? LOGROLLING AND POLITICAL EXCHANGE Logrolling is
the exchange of political support on one issue for political
support on another issue. For voters in general elections, such as
those for seats in Congress, or when voting for president,
logrolling is not feasible because high transactions costs impede
political exchange. When there are large numbers of voters, it is
difficult for voters to exchange votes in a political market where
those who place little value on their votes can sell them to those
who value votes more. Furthermore, the secret ballot impedes the
trading of votes because one cannot guarantee that one voted as
promised. You could, for example, offer to vote for a certain
candidate in exchange for a $50 payment, but because of the secret
ballot, you could not prove that you voted as you were paid to. For
representatives in a legislature, logrolling is feasible. There are
a relatively small number of legislators, all of whom will know one
another, which lowers transactions costs.
CHAPTER 9 Supply and Demand In Political Markets
163
Furthermore, the votes of legislators are a matter of public
record, so, if a legislator offers to trade his vote, there is a
record that can be checked to verify that the legislator voted as
promised. In this type of situation, legislators can engage in
political exchanges to gain support on particular issues they favor
by trading their votes on issues they do not care much about for
the votes of others on issues in which they have intense
preferences.2 Logrolling and Special Interests If everyone were
perfectly informed about all political issues and transactions
costs among voters and legislators were low, the Coase theorem
would apply and the political marketplace would allocate resources
efficiently. In reality, most voters are rationally ignorant about
most issues and find access to the political process costly in any
event. Most citizens see little point in trying to influence
legislators on most issues, so there are high transactions costs
between the legislature and its constituents that prevent
bargaining. Special interests, because they follow issues closely
and invest resources to hire lobbyists, supply campaign
contributions, and otherwise try to influence issues, have lower
transactions costs and can bargain directly with legislators over
issues that have a significant effect on them. And because there
are relatively few legislators and they all know one another,
transactions costs among legislators are low, and it is relatively
easy to trade votes on issues. Legislators want to favor interests
that can offer them support, and this is how interests express
their demands. Legislators can obtain votes for interests they
support by trading votes with other legislators, though the
logrolling process may be more complex than simply trading votes on
one issue for votes on another. A legislator may receive needed
votes on a particular issue in exchange for an IOU to vote for an
unspecified future program for another legislator. In a relatively
small group, legislators have an incentive to honor those IOUs to
retain a reputation for cooperation. If a legislator does not repay
IOUs, he will find it impossible to trade votes on any issues in
the future. Thus, to retain effectiveness, legislators must trade
votes with one another in the political marketplace and must keep
track of and honor IOUs. In this way, logrolling facilitates
political exchange to produce the outcomes that are most highly
valued by the legislators. Logrolling creates a market and votes
are the medium of exchange in the legislative market. Because
legislators benefit from special interest programs, logrolling is
especially helpful in getting special interest programs passed.
This is the way that programs that benefit a minority can get the
approval of a majority. Logrolling and Economic Efficiency To this
point, logrolling has been presented as a mechanism whereby special
interests can get their programs approved by the legislature. This
can occur when no one represents the general public interest in the
political marketplace. As was suggested earlier in the chapter,
however, all sides tend to be represented to some degree as
legislators balance the interests of those in favor of specific
legislation against those opposed. The better-organized interests
can be, the better they will be represented, but all sides tend to
be represented to some degree. Because legislative bargaining
involves small numbers, some of the problems inherent in majority
rule voting can be overcome by application of the Coase theorem.
With relatively few legislators, the outcome most highly valued to
the legislators will tend to be chosen regardless of the political
process used to choose the outcome. An example can illustrate.
CHAPTER 9 Supply and Demand In Political Markets
164
Table 9.1 lists the hypothetical preferences of three voters
numbered 1, 2, and 3 for three projects labeled A, B, and C. The
amount that each option is worth to each voter is given in the
table. For example, voter I places a value of $1,000 on option A,
$2 on option B, and $1 on option C. Assume that the three voters
must select one of the options by majority rule. The preferences
listed in the table will produce a cyclical majority if each voter
can express a preference by casting only one vote. Under simple
majority rule, A can defeat B, C can defeat A, and B can defeat C.
There is no clear majority winner, so the winning option will be
chosen based on the way in which the alternatives are considered.
This example is identical to the cyclical majority examples in
chapter 8, and under simple majority rule, with no more information
on the selection process, each of the outcomes is equally
likely.
If logrolling is possible, however, project A is almost certain
to win. Option A is clearly worth more to voter 1 than any of the
other options are worth to the other voters, so with logrolling,
voter 1 can offer to trade for the votes of 2 or 3. Because their
votes are worth more to voter 1 than to themselves, a mutually
advantageous trade can certainly be made, and option A will be
chosen. The total value to the voters of option A is $1003, whereas
B and C are worth $6 each, so the highest valued option will be
chosen when logrolling is possible. In the absence of logrolling, a
cyclical majority occurs, but with logrolling, the highest valued
option is selected. This simple example shows how logrolling can
lead to economic efficiency when all parties are able to trade.
When small numbers of voters are involved and transactions costs
are low, the Coase theorem applies, and the most highly valued
option is selected. Special interest legislation tends to be
inefficient not because of the logrolling that produces it, but
rather because the general public finds it too costly to
participate in the political exchange process. Special interests
are party to the exchange but the general public is not, so the
legislature tends to produce outcomes that are most highly valued
to those who are party to the exchange, which are the special
interests and the legislators themselves. If everyone could
participate in the logrolling process at low cost, rather than just
legislators and special interests, logrolling would allocate
resources efficiently. AGENDA CONTROL In the previous chapter on
public sector demand, the suppliers of legislation were depicted as
engaging in competitive behavior to try to satisfy the demands of
the median voter. This competitive political behavior makes a at
deal of sense when voters are equally informed about
CHAPTER 9 Supply and Demand In Political Markets
165
political issues and when candidates must satisfy the demands of
the median voter to maximize their chances of reelection. In cases
in which some voters are not well informed or in which there is
little direct competition for reelection, legislators may behave
monopolistically. Monopolistic behavior can be exhibited through
control of the agenda. Agenda control can occur any time some group
has the power to determine what options a larger group is able to
choose from. The political agenda is simply the collection of
issues that are to be decided. In all settings other than a pure
direct democracy, some individuals will have more control over the
agenda than others will. In a competitive setting, the agenda
setter has an incentive to set the agenda most preferred by the
median voter, but sometimes the agenda setter is insulated from
political competition and will be able to manipulate the agenda to
produce an outcome closer to the setters preference than the
outcome most preferred by the median voter. A Referendum Example
Consider an example in which a referendum will be held to determine
the level of output of a public good, much like the referendum
process described in chapter 8. To give the agenda setter some
power, this section will assume that only one referendum will be
held, and that if the referendum issue fails, the level of the
public good produced will be much less than would be preferred by
the median voter. Further, assume that the voters have no ready way
of replacing the agenda setter if they do not like the agenda that
is proposed and that the agenda setter prefers a larger level of
expenditures than does the median voter. The actions of the agenda
setter can be illustrated in figure 9.2.
Assume that if the referendum issue fails, amount QR of the
public good will be produced. This level is the reversion level,
which the public good reverts to in the event of a
CHAPTER 9 Supply and Demand In Political Markets
166
failure. The median voter would most prefer QM, which would
provide the median voter with consumer surplus equal to triangle
abc when compared with QR. If the agenda setter wants to maximize
the amount of output to be produced, the setter can propose QA as
the alternative to QR, where the consumer surplus gain in triangle
abc is slightly larger than the consumer surplus loss in triangle
cde. Thus the choice between QR and QA can be viewed as the choice
between the consumer surplus gain of abcBcde. As long as there is a
light gain, the median voter benefits from voting for QA, and the
agenda controller will have succeeded in getting the median voter
to select an option that is closer to the agenda setters preference
than to the median voters. The power of the agenda setter comes
because the agenda setter can manipulate the options available to
the voters. Most studies of agenda control in referenda like the
one just described conclude that agenda setters actually choose
options close to the median voters preference and that the outcome
approximates what the median voter most prefers.3 However, a
referendum setting may be the one most friendly to the median voter
because a single issue is considered on which there is a direct
vote and often the same voters who vote in referenda also directly
vote for the agenda setters. Furthermore, referenda are most
frequently held at the local level where voters may have more
direct control over their representatives. In other settings, and
at the state or national level, agenda control may be more likely.
Agenda Control in the Real World The referendum case clearly
illustrates the principle of agenda control and the power that an
agenda setter can have by controlling the options that a group can
vote on. Few decisions are made in a referendum, so this example
may have relatively limited direct applicability. However, the
agenda can be controlled in other types of political settings. A
good example is the power of committees and committee chairs in the
U.S. Congress. Committees are responsible for delivering bills in
their particular areas to the floor of Congress, which gives them a
great deal of influence over the contents of those bills. Special
interests can concentrate their lobbying efforts on the committee
members who determine the provisions of the bills, and, although
the resulting bill may not be exactly what the median member of
Congress would prefer, the member may vote in favor anyway,
preferring the bill as it stands to no bill at all. The principle
is exactly the same as in the referendum case, where the outcome is
moved from that preferred by the median voter toward the outcome
preferred by the person (or group) who controls the agenda. Because
congressional committee members tend to be relatively high
demanders for the types of public goods their committees oversee,
the result of the agenda control of committees is likely to be that
legislation produces more than even the median member of Congress
would most prefer. Another form of agenda control occurs when a
committee can decide who can run for political office. Many
dictatorships hold elections for their leaders, but only one
candidate appears on the ballot. This gives the illusion of popular
elections and support for the dictator, but because the agenda is
controlled, the voters really have no choice but to elect the
candidate chosen by the agenda controllers. This form of agenda
control occurs in many organizations. The Southern Economic
Association holds an annual election to select a president, but,
again, only one name appears on the ballot. Needless to say, the
nominating committees choice is always selected. In many
organizations such as clubs and fraternities, a group of officers
often determines the agenda of a meeting ahead of time, which
enables it to control which issues are considered and enables it to
achieve the results it wants.
CHAPTER 9 Supply and Demand In Political Markets
167
Agenda control may be slightly more subtle. There is a story
about Arthur Burns, former chairman of the Board of Governors of
the Federal Reserve, controlling the agenda of the governors
meetings by taking straw votes on issues. During the course of a
meeting, Bums would ask the governors how they would vote if a vote
were to be taken on an issue right then. If the governors were
going to vote against Burnss preferences, he would talk to them
some more about the issue and then ask them again how they would
vote. Bums would continue doing this until the governors said they
would vote his way, at which time he would say, We will now vote on
the issue, thus controlling the agenda to get the outcome he
wanted. That example might not be so subtle, but it shows how the
chair of any meeting can control the outcome of a meeting by
controlling the agenda. Control over the agenda of a collective
decision-making group has the potential to give substantial control
over the outcome of the groups decisions. Even if the entire group
votes on the outcome of an issue, the agenda controller can
manipulate the outcome to reflect the preferences of the
controller. RENT SEEKING If one views the political decision-making
process as a market, it might seem that an efficient outcome would
result from the weighing of the various demands of competing
interest groups. Indeed, we have noted that this would be the case
if everyone had access to the public sector bargaining process.
However, special interests tend to register their demands more
strongly because they have a concentrated interest and have much to
gain, albeit at the expense of the general public who pays a small
amount to support each special interest program. If a simple
transfer of resources was all that occurred, the process might be
redistributive but not inefficient. However, interest groups expend
resources to obtain their special interest benefits, and this
expenditure is a waste of resources from a social standpoint.
Consider an example in which an industry association lobbies
Congress to try to get a tariff put on competing imported goods.
This restriction raises the domestic price and creates profits for
domestic producers. The lobbyists, lawyers, and employees of the
industry association use valuable resourcesCincluding their
laborCwhich could be used for productive activities. These people
could be producing goods and services, which not only would make
them better off but also would add to the total output of the
economy. Instead, they are using valuable resources to try to get a
private gain, but at the expense of others in the market. They are
producing nothing of social value and nothing that adds to the
total output of the economy. Thus, resources that could have been
used productively to increase the economy's output are instead used
to try to produce a private gain at a cost to others. Thus, from a
social standpoint, these resources that are used to try to capture
transfers through the political process are wasted. Economists call
this type of activity rent seeking. The use of the term rent in
this way goes back to David Ricardo, an economist who early in the
nineteenth century argued that the payment of rent for the use of
land was unproductive because the land would be there to use
regardless of how much rent was paid. Labor required wages (because
people would not work if they were not paid), and capital
investment could only be attracted if it earned interest (because
people would consume rather than invest unless investment was
profitable), but land had no other alternative. Thus, Ricardos view
was that payment of rent for the use of land was unproductive.
Ricardos ideas on rent payments to land might be called into
question because rents do serve to ration land and allocate it to
its highest valued use, but the term has stuck and has been applied
to payments that are made to factors of production for their
private benefit
CHAPTER 9 Supply and Demand In Political Markets
168
when those factors do not add anything to social output. This
clearly applies to lobbyists seeking special interest benefits.
They could produce something adding to social output, but instead
they use resources to try to get transfers from others to the
interests for which they lobby. Sometimes firms try to get tariffs
or quotas to give them an advantage over foreign competitors but at
the expense of those foreign competitors and consumers who prefer
less expensive goods. Some firms seek protection from domestic
competitors. Businesses would like for government to create
regulations and other barriers to entry to keep new firms from
competing with established firms. Professional organizations want
government to prevent others from competing in their professions
without meeting stringent government regulations. For example,
medical doctors, real estate agents, barbers, and certified public
accountants all have government-imposed barriers to entry that
prevent people from entering those professions without incurring
large costs and securing government approval. Government-enforced
barriers to entry of any type create rents for those who are in the
industry and who are sheltered from competition. Another source of
rents would be government-guaranteed price supports. Farm price
supports, for example, increase the profits from farming because
farmers are guaranteed a price no lower than the government support
price. Farmers have an incentive to seek higher support prices, or,
if Congress is debating lowering or eliminating supports, they have
an incentive to lobby to keep them. This lobbying is rent seeking
and has the associated rent-seeking losses because while the price
supports provide private benefits to farmers, they provide no
social benefits and add nothing to the total output of the economy.
(They may actually lead to more wasted resources if farmers produce
output that is not consumed.) Businesses might also seek rents by
trying to get the government to undertake expenditures to lower a
firms costs, such as by undertaking government-financed research
and development, building a road or an airport to benefit a
business, or giving a business a tax break if it locates in a
certain area. These examples give an idea of the types of
activities included in rent seeking, but the list could be longer.
It includes any activity in which resources are expended to try to
secure a private gain that adds nothing to social output. In
contrast, when firms produce goods and services for consumers, the
firms benefit by selling their output, and they also add to the
total amount of output that everyone can consume. The Welfare Costs
of Rent Seeking What are the social costs of rent-seeking activity?
This question can be analyzed with the assistance of figure 9.3,
which shows the potential results of rent seeking. The figure
starts with the assumption of a competitive market with a constant
marginal cost, so that MC is also the industry supply curve. If a
barrier to entry, price support, or other factor creates a price
increase and restricts output, the conventional analysis on the
welfare loss because of monopoly would be that the darkly shaded
triangle abc would represent the welfare loss caused by the
reduction in output in the market. In addition, monopoly profit
equal to the lightly shared rectangle PMabP* would also be
produced. However, this analysis ignores the resources devoted to
rent seeking. If a firm realized that there was the potential for
monopoly profits equal to the lightly shaded rectangle if the
government would create a barrier to entry in the industry, how
much would the firm be willing to pay to secure that profit? The
answer is that the firm would be willing to spend up to the amount
of the expected profits to secure the profits. Assume, for example,
that a firm estimates that a tariff could produce about. The waste
from rent $10 million in benefits for the firm. Then the firm would
be willing to spend up to $10 million to get the
CHAPTER 9 Supply and Demand In Political Markets
169
benefits, and the entire monopoly profit from rent seeking might
be expended. The social loss from the barrier to entry is not just
triangle abc, then, but abc plus PMabP*, or the entire shaded
triangle and rectangle.4
In political competition, like market competition, some firms do
better than others. Thus, some firms may get rent-seeking benefits
far in excess of the amount they expend. Other firms may spend a
great deal to try to secure benefits but not get anything. In those
cases, the rentseeking losses far exceed the benefits to the firm.
How much, on average, would one expect these rent-seeking costs to
be? If most firms that engaged in rent-seeking activities received
above-normal profits from their rent-seeking expenditures, this
would encourage entry into rent seeking, just as profits encourage
entry into any activity. Thus, more rent-seeking expenditures will
be made until there is just a normal rate of return to be earned
from rent seeking. Similarly, if rent seeking was, on average, a
losing activity, that would encourage exit from rent seeking until
a normal rate of return was earned. Thus, just like in any market,
entry and exit of firms in rent seeking should occur to make the
rate of return to rent seeking the same as the rate of return to
other activities in the economy. Therefore, one would expect that
on average, rent-seeking expenditures would equal the entire amount
of monopoly profit that interests are competing for, which means
that the welfare loss from rent seeking equals the total shaded
area, PMacP*, in figure 9.3. The social costs of government
transfer activity and associated special interest benefits are thus
larger than at first they appear. Resources used up in trying to
influence the legislature to provide benefits to interests are all
a part of these social costs, and the competitive political process
suggests that those welfare losses will approximate the value of
the rent-seeking benefits that the legislature grants.
CHAPTER 9 Supply and Demand In Political Markets
170
PUBLIC SECTOR SUPPLIERS This section looks directly at the
supply side of the public sector by examining the structure and
incentives implied in government bureaucracy. In terms of supply
conditions in the private sector of the economy, firms are assumed
to be profit maximizers. Of course, the individuals who run firms
have other goals as well, but the assumption of profit maximization
in the private sector has produced simple models that have
significant explanatory power. Public sector suppliers differ in
important ways from private sector suppliers, and one of the most
significant differences is that suppliers in the public sector do
not have the incentive to maximize profits. In fact, most public
sector output-public education, highways, national defense-is not
sold directly on the market but is allocated to the user at little
or no charge when compared with the costs of production. If public
sector suppliers do not maximize profits, then how do they decide
how much output to produce? The profit maximization assumption for
private firms can be explained in terms of the incentives faced by
suppliers. The individuals in the firm benefit most when they
maximize the profits of the firm, and, in many situations, the only
way that a firm can remain in business is to act like a profit
maximizer. Likewise, the behavior of public sector suppliers can be
explained in terms of the incentives faced by those suppliers who
are motivated to maximize their budgets in the same way that
private sector suppliers are motivated to maximize profits. In
examining the institutions within which public sector suppliers
work, we can see what it means to be a budget maximizer in the
public sector and what the implications are when public sector
supply and demand interact. Incentives in Bureaucracy The public
sector counterpart of firms that produce output in the private
sector is the government bureaucracy. The Department of Defense
(DOD) oversees military production, the Department of Energy (DOE)
runs energy programs, and so on. In the private sector of the
economy, the market evaluates the worth of a firm's output, and a
firms profits represent the difference between the value of the
resources the firm uses in production and the value of the firm's
final output. By contrast, public sector output is not usually sold
in the market, so market tests cannot be used to determine whether
the value of the governments output is greater than its resource
cost. Nor can the success of a government bureau be measured by its
profits, unlike the success of a firm in the private sector.
Utility and Budget Maximization in the Public Sector The goal of
profit maximization in the private sector is based on the concept
of utility maximization. The managers of firms have an incentive to
maximize profits because this provides them with the most income.
The people who work for the government are not necessarily any
better or worse than those who work in the private sector of the
economy, and they probably are motivated by the same types of
things. Certainly, they want to further the public interest, just
as citizens in the private sector want to, but they also are
interested in pursuing their own self-interests. To fully
understand their behavior, we must understand the incentives they
face and the actions they must take to achieve those goals. In the
private sector, managers are rewarded based on the profits that
they bring to the firm, which obviously gives them an incentive to
maximize profits. But in the public sector,
CHAPTER 9 Supply and Demand In Political Markets
171
managers do not seek, and they cannot be rewarded for, the
profits they generate. Rather, they typically are rewarded for the
budgets they oversee. Bureaucrats who oversee larger bureaus with
bigger budgets tend to earn higher incomes. Because a bureaucrats
income will tend to rise as the budget the bureaucrat oversees gets
larger, bureaucrats have an incentive to be budget maximizers.5
Undoubtedly, bureaucrats have other motivations besides increasing
their incomes, but these other incentives tend to be positively
related to the size of the budgets they oversee as well. For
example, a bureaucrat may be interested in the power and prestige
associated with the job, but the larger the bureaucrats budget, the
more power and prestige the bureaucrat will have. Bureaucrats will
also be interested in a congenial working environment, which is
also more likely to exist in a growing bureau than in one that is
stagnant or shrinking. In a growing bureau, there will be the
opportunity to promote current employees to make room for more
employees underneath, which enlarges the budget and gives everyone
more income, power, and prestige. In a stagnant or shrinking
bureau, opportunities will be limited and the work environment will
be less pleasant. For a number of reasons, bureaucrats benefit from
large and growing budgets, which gives them an incentive to be
budget maximizers. The Public Interest and Budget Maximization The
reasons just given for bureaucrats to be budget maximizers all have
to do with the personal self-interests of bureaucrats, but it also
likely to be true that bureaucrats will tend to be employed in
agencies that undertake missions the bureaucrats believe in. Those
who work in the Department of Defense are more likely to believe in
the benefits of a strong military force than is the general public,
while those who work for the Consumer Products Safety Commission
are likely to believe in the benefits of government regulation to
enhance public safety. Similarly, those who work for the
Environmental Protection Agency (EPA) are more likely to believe in
the benefits of government regulation to protect the environment.
Thus, bureaucrats who believe in the mission of their bureaus will
want to maximize their bureaus budgets to further their perception
of the public interest, in addition to the private incentives they
have for budget maximization. This public interest argument can
only be taken so far, however. The individual working for the EPA
may believe in the mission of the agency, but the individual still
has an incentive to pursue her own self-interest. After all, if
individuals can be counted on to pursue the public interest when it
conflicts with their narrow self-interests, then managers of
polluting firms could be expected to reduce pollution of their own
accord, without the government requiring it. In reality, we expect
self-interest to be a powerful motivating factor of those who work
in business, but we should not expect any different from people who
work for government. People who work for the EPA must be just as
cognizant of their own interests as those who work in the private
sector. In the private sector, we expect to see firms act to
maximize their profits. By the same reasoning, in the public sector
we expect to see bureaus act to maximize their budgets. The
Bureaucrats Maximand The hypothesis that bureaucrats are budget
maximizers is persuasive, but that is only one hypothesis among
many put forth by students of bureaucracy. In a more complex
formulation of the budget-maximizing hypothesis, bureaucrats might
try to maximize their discretionary budgets rather than their total
budgets.6 Following this line of reasoning, funds used to
produce
CHAPTER 9 Supply and Demand In Political Markets
172
the bureaus output are not as valuable to the bureaucrat as
discretionary funds that can be used for such benefits as vacations
disguised as fact-finding trips and funds for conferences to which
bureaucrats can invite all their out-of-town friends for expensive
meals and luxury accommodations. There is undoubtedly much truth to
this view, but the hypothesis of maximization of discretionary
funds does not consider the benefits, such as power and prestige,
inherent in overseeing a large budget. In reality, the bureaucrat
probably wants both a large total budget and a large amount of
discretionary funds within the budget. Just as profit maximization
is an oversimplification of the activities of individuals who are
running firms, so budget maximization is an oversimplification of
the behavior of bureaucrats. But recognizing the complex and
sometimes conflicting incentives facing bureaucrats, the budget
maximization hypothesis is a good first approximation. Many useful
implications can be drawn from this hypothesis even while
recognizing its limitations. THE PUBLIC SECTOR BARGAINING PROCESS
Having established that bureaus are budget maximizers in the same
way that firms are profit maximizers, we still do not have enough
information about the public sector supply process to understand
how much output will be produced by public sector suppliers. For
example, although all firms in the private sector have an incentive
to maximize profits, a monopolized industry will produce less
output and sell at a higher price than a competitive industry under
similar circumstances. This section will examine the public sector
bargaining process to see what type of bargain is likely to emerge
between a bureau and the legislature that funds the bureau. The
Bureau and the Legislature For most government bureaucracies, the
legislature determines the bureaus budget. For example, the U.S.
Congress determines the budget of the EPA, the Department of
Defense, and other federal agencies. Likewise, state legislatures,
city and county commissions, school boards, and so forth determine
the budgets of state and local bureaucracies, For simplicity, this
section will consider the bargaining process between a bureau and
the legislature, recognizing that the model applies more generally
to the bureau and whatever organization determines the bureaus
budget. An important difference between the bargaining process that
determines a bureaus output and the process that determines the
output of a firm in the private sector is that the private sector
firm sells output at a certain price per unit and customers decide
how much to buy at that price. In contrast, a bureaus total output
is determined at the same time as its budget. Thus, the legislature
decides to purchase a certain amount of output from the bureau at a
certain price, simultaneously negotiating both the price and the
quantity of output. This puts the bureau in the position of being
able to try to make an all-or-nothing sale to the legislature,
attempting to extract all the legislatures consumer surplus from
the bureaus output. The Bureaus Bargaining Advantage In essence,
the bureau and the legislature are engaged in a bilateral
bargaining situation, and for several reasons the bureau will have
a superior bargaining position relative to the legislature. This
superior bargaining position will aid the bureau in capturing all
the gains from trade in the exchange of a budget for the bureau's
output. The first advantage that the bureau has is that the
CHAPTER 9 Supply and Demand In Political Markets
173
bureau has only one legislature to deal with, so it can
specialize in trying to work out a favorable bargaining agreement.
The legislature, in contrast, will oversee many bureaus. For
example, the Department of Defense deals only with the U.S.
Congress in negotiating its budget, but Congress also oversees the
EPA, DOE, DOT (Department of Transportation), and an alphabet soup
of other agencies. Thus, the bureau is able to devote more time and
effort to the bargaining process than is the legislature. Second,
the legislature typically gets all its information about the
bureaus characteristics from the bureau itself, so the bureau can
reveal to the legislature only the information that is in the
bureaus interest. This means that information on the cost of
various programs and how much could be produced if the budget were
a little larger or a little smaller must be obtained from the
bureau. The legislature, on the other hand, typically reveals its
demand curve by votes on issues, speeches, campaign statements, and
so forth. In short, while the bureau will have good information
about the legislatures demand for its output, the legislature will
have only the information about the bureaus supply conditions that
the bureau chooses to reveal. Third, much work is done in
committees before the full legislature votes on an issue, and
members of committees tend to be high demanders for the services of
bureaus they oversee. Members of the Agriculture Committee, for
example, will be more sympathetic than the typical legislator to
the interests of farmers, while members of the Armed Services
Committee will tend to be high demanders of military services. This
makes it easier for bureaus to present good cases for larger than
optimal budgets, which puts bureaus in a good bargaining position
in their quest to maximize their budgets. The Bureaucratic
Bargaining Process The challenge to the budget-maximizing bureau is
to produce the largest budget that the legislature will accept. The
bargaining process is illustrated in figure 9.4, where the
legislatures demand curve for the bureaus output is labeled D. The
bureaucracy is assumed to have supply conditions like a constant
cost industry, so the marginal cost curve, MC, is horizontal, and
MC would be the supply curve if this output were produced by a
competitive industry. A competitive industry would produce output
level Q*, where supply equals demand, and a monopolist, setting
marginal cost equal to marginal revenue, would produce QM. If a
government bureau were to produce Q*, like a competitive industry,
the bureau would produce consumer surplus for the legislature equal
to the area of the shaded triangle above the MC curve. In an effort
to maximize its budget, the bureau will propose to the legislature
to produce a level of output larger than Q*, and, if the
legislature sees no better alternative, it will be willing to
accept the bureaus proposal as long as there is some consumer
surplus produced by the bureau. The budget maximizing strategy is
for the bureau to propose to produce output level QB at price per
unit P*. At QB, the consumer surplus gained from the production of
units below Q* (the shaded triangle above MC) is slightly greater
than the consumer surplus lost on units above Q* (the shaded
triangle below MC). The shaded triangle above MC is only slightly
larger than the triangle below MC, meaning that the legislature
gets very little consumer surplus from approving the bureaus budget
to produce QB at price P* per unit. The legislature might try a
counteroffer, saying that it wants less output at price P*, but
because the bureau bargains for a total level of output in exchange
for a total budget, it can follow the strategy of telling the
legislature that if the level of output is reduced, the price per
unit will rise dramatically. For example, if the legislature wants
output Q*, the bureau could tell the legislature that it would cost
price P' per unit to produce that output. This might be done,
for
CHAPTER 9 Supply and Demand In Political Markets
174
example, if the Department of Defense says that it will be very
expensive to produce ten aircraft because of the development costs,
but once the research and development is done, if many aircraft are
built on an assembly line, the cost per unit will fall
substantially. The legislature would get more consumer surplus from
buying QB at price P* than Q* at price P', so because of the
bargaining advantage of the bureau and the bureau's ability to
control the flow of information to the legislature, the legislature
is likely to decide that its best alternative is to choose output
level QB.
Curve A in figure 9.4 represents the legislatures all-or-nothing
demand curve for the bureaus output. An all-or-nothing demand curve
shows the price-quantity combinations where the purchaser receives
no consumer surplus from the purchase. For example, at quantity QB
and price P* the legislature is indifferent between that amount at
that price and no output at all if the shaded triangle of consumer
surplus gained from the output below Q* is equal to the shaded area
for the consumer surplus lost for output between Q* and QB. The
budget for the bureau will be maximized when the bureau is able to
get a budget to produce QB which is where the all-ornothing demand
curve intersects the bureaus marginal cost curve. By contrast, a
competitive industry will produce the optimal output, Q*, and a
monopolistic industry will produce less than the optimal output at
QM. The conclusion of this analysis is that bureaus have an
incentive to be budget maximizers, and because they have a
bargaining advantage in negotiating their budgets with the
legislature, they are able to obtain budgets that are larger than
optimal. Thus, while a monopolist misallocates resources by
restricting output and producing too little, government bureaus
misallocate resources by producing too much.
CHAPTER 9 Supply and Demand In Political Markets
175
THE GOVERNMENT AS AN INTEREST GROUP The role of special
interests in government has been discussed at some length. One
significant factor in the perpetuation of government programs once
they are instituted is that government agencies themselves can act
as interest groups. Before a particular program is started, there
will be a demand from some constituency, but the program may also
have some opponents. Once the program is passed and put into place,
a government bureaucracy is created (or an existing one expanded)
to administer the program, and the people in the bureaucracy will
become another interest group favoring the program. This means that
in addition to the interest groups that wanted the program passed
in the first place, the government employees who run the program
will form yet another interest group. Government employees exert
significant influence because they are the governments source of
expertise on the program. For example, in determining whether a
program should be continued, the government can, check with its
experts-the individuals in the administering agency-whose jobs
depend on the continuation of the program and who have an incentive
to maximize the programs budget. Furthermore, if the individuals
who work for the agency believe in the agency's mission, they have
yet another reason to defend the programs they work for and argue
that they need even larger budgets. The result is that once
government programs are instituted, they are very difficult to
eliminate. The interest groups that originally wanted the programs
will still want them, and a new and powerful interest group created
by the program will help ensure its survival. This helps to explain
why a government program, once started, will tend to perpetuate
itself. FISCAL FEDERALISM One way to limit the influence of special
interests and control the power of government bureaucracies is to
produce government goods and services at the lowest possible level
of government. Fiscal federalism is a system in which government
programs are undertaken at different governmental levels, such as
local, state, and national. Because chapter 25 focuses on
federalism, this section just introduces the topic and suggests
some implications within the context of the supply and demand for
public sector output. Special interests operate at each level of
government, but the problems posed by special interests can be
reduced by producing government output at lower levels, for three
main reasons. Federalism and Special Interests First, the influence
of special interests can be limited if the government producing the
program does not extend beyond the special interests that will
benefit from the program. Consider, for example, waterway shippers
in Alabama, who benefit from the federal governments maintenance of
navigable waterways in the state. A study of the costs and benefits
of these waterways estimated that the total cost of maintaining
them was greater then the total benefits produced by them. However,
Alabama was a net beneficiary of the program because most of the
benefits accrued to shippers in that state while most of the costs
were paid by taxpayers in other states. It makes sense that people
in Alabama would favor maintenance of the waterways by the national
government because the cost to the state was less than the
benefits, even though the cost to the nation was greater than the
benefits.
CHAPTER 9 Supply and Demand In Political Markets
176
Because the benefits of the program accrue to the people in
Alabama, it would make more sense for Alabama to finance the
waterway program through taxes levied on people in that state. In
this case, the people of Alabama would be much less likely to favor
the special interest expenditures because they would bear the total
cost. This would be even more true if the taxes were levied on the
waterway shippers themselves rather than on all the taxpayers in a
state. The same logic would apply to all government programs. If
the people of North Carolina paid for their own tobacco price
support programs and the people of Wisconsin paid for their own
dairy price support programs, special interests would have a lot
more trouble getting these programs passed. At the federal level,
the total cost to each taxpayer is very diluted, but, if such
programs were financed at the state level, the costs would be, on
average, fifty times higher, giving taxpayers more incentive to
actively oppose them or at least to become more aware of the
relative benefits and costs. Federalism and Consumer Choice A
second reason that special interest expenditures can be mitigated
by placing government programs at the lowest level of government
possible is that voters will have more choices over what types of
programs they want. For example, if public education were produced
by the federal government, citizens who wanted public education
would have to accept the type of education provided by that one
government. However, when public education is produced by local
governments, people can choose to live in areas that provide the
kind of education they want at a price they are willing to pay. By
moving to jurisdictions that are more in line with their
preferences, voters can, in a sense, vote with their feet. They can
move to a location with a local government that will provide them
with the types of government services they want.7 To attract
residents and a tax base, local governments have an incentive to
produce the type of output their residents want, making government
more responsive to the general public and less responsive to
special interests. Comparisons with Other Governments The third,
and perhaps most important, reason for producing government goods
and services at the lowest possible level is that it enables voters
to compare what they are getting for their tax dollars with what
other people are getting for their tax dollars. If voters are not
satisfied with how their local government compares with other
nearby governments, either they can move or they can vote their
local officials out of office. This provides local government
leaders with an incentive to look good in comparison with other
governments. This type of intergovernmental competition need not
require that people continually move from one jurisdiction to
another. It only requires that voters replace their existing
elected officials if their government does not compare favorably
with others. Production at Lower Levels of Government By producing
government output at the lowest level of government possible, the
influence of special interests can be reduced, consumers of
government output can be given a wider range of choices, and
citizens will have other governments to compare theirs with when
evaluating their government's performance. These are good arguments
for maintaining a federal system and for trying to shift production
to lower levels of government when possible, but this does not
mean
CHAPTER 9 Supply and Demand In Political Markets
177
that the federal government should do nothing. Some government
programs are truly national in scope. National defense is a good
example, and the interstate highway system provides another
example. Local traffic uses interstate highways, of course, but
people who are passing through a state on a limited access highway
also benefit. For programs that are national in scope, the national
government is the appropriate level of government to produce the
good. But when the benefits of a program are geographically
concentrated, it makes more sense to produce the program at a lower
level of government, thereby limiting the ability of special
interests to foster programs that are not cost-effective.
CONCLUSION Part 2 of the text illustrated that the market may not
always allocate resources in the most efficient manner possible
because of the problems of externalities, public goods, and
incomplete assignments of property rights. However, this does not
necessarily mean that the government can do any better. The
government can also encounter problems in allocating resources
efficiently. Some of these problems were touched on in the previous
chapter. For example, individual preferences may be such that
majority rule produces a cyclical majority, and, even if there is a
median voter outcome, it is unlikely to be completely efficient.
There are other incentive problems in the public sector, however,
so it is likely that the public sector will not produce the output
that allocate resources the median voter most prefers. Government
output typically is produced by government bureaucrats who have an
incentive to maximize the budgets of their agencies in the same way
that managers of private firms have an incentive to maximize their
profits. This tendency toward budget maximization means the
government produces greater-than-optimal output. Another problem is
that most voters are rationally ignorant about most of the
activities of the government, whereas special interests have an
incentive to become informed and to influence government policies
to favor their interests. Elected representatives, who need
political support to keep their jobs, are more inclined to favor
special interests than the general public interest because that
increases the political support they can obtain. In small settings
such as legislatures, logrolling and vote trading can lead to the
passage of special interest programs that could not be passed if
they were all voted on individually by the general public. Also,
the fact that some individuals can control the agenda of a
representative body for their own benefit contributes to a level of
production of public sector output that is not completely
reflective of the demands of the median voter. Placed in proper
perspective, this picture is not as grim as it first appears. Just
as the discussion of externalities, public goods, and so on implied
that market allocation of resources is inefficient, much of the
discussion of this chapter centered on reasons why public sector
resource allocation is likely to be inefficient. But in the real
world, things rarely work perfectly, and, by understanding the
limitations of government activity, we can better use, the
government. Frequently resources are allocated more efficiently
through the public sector. Few people would want to rely on the
private sector to produce national defense, for example, and it is
unlikely that any organization other than the federal
governmentCincluding state governmentsCcould have developed the
interstate highway system. But although the government can most
effectively allocate resources for some purposes, it is clearly not
the answer to allCor even most resourceCallocation problems. Where
externalities and public goods exist, or where property rights are
not well defined, there is good reason to regard the government as
a possible solution to the resource-allocation
CHAPTER 9 Supply and Demand In Political Markets
178
problem. However, the successes of the government in some areas
should not obscure the fact that in many other areas, the incentive
structure of the government is inappropriate for resource
allocation. Moreover, understanding the strengths and limitations
of government resource allocation is of more than just academic
interest because ultimately a democratic society chooses the
activities that it wants its government to perform. That most
people will be ignorant of most of what the government does is good
reason for choosing carefully when the opportunity arises. It is
ironic that in these matters that collectively are so important,
individuals have a minimal incentive to become informed. QUESTIONS
FOR REVIEW AND DISCUSSION 1. Because most voters are rationally
ignorant about most government activities, but special interests
have an incentive to become informed, special interests tend to
have their demands better represented in the political marketplace.
Does this mean that special interests get what they want from the
legislature? Explain how the political marketplace weighs interests
on all sides of a political issue. Use a graph to help illustrate
your explanation. 2. In general elections, votes are cast by secret
ballot, but the votes of legislators are a matter of public record.
Explain how this difference influences the ability of voters to
engage in logrolling. 3. The Coase theorem would suggest that if
transactions costs are low, political exchange should lead to the
efficient allocation of resources. Are transactions costs higher
for some groups than for others? Explain how differences in
transactions costs for political exchange among different groups of
citizens can affect the allocation of resources in the public
sector. 4. If others tend to be rationally ignorant, why are
special interests well informed about issues? Are they well
informed about all issues or just some issues? 5. Explain why the
typical college student knows more about the differences between
various brands of pizza than about the debate on national health
care, even though health care is more important than pizza. 6. How
can control over a group's agenda give an individual control over
the decisions the entire group will vote on? Illustrate in a graph
how the agenda can be controlled using the example of a referendum.
7. What is meant by the term logrolling? What are the effects of
logrolling on legislative outcomes? When will logrolling be
efficient or inefficient? 8. Why is budget maximization the goal of
a government bureau in the same way that profit maximization is the
goal of a private sector firm? 9. Explain the bargaining
arrangements that lead to the level of output that will be produced
by a government bureaucracy. Compare the level of output produced
by a bureaucracy with the levels that would be produced by a
competitive market and by a monopolist.
CHAPTER 9 Supply and Demand In Political Markets
179
10. What is fiscal federalism? Why is it desirable? List and
explain several reasons why programs produced at lower levels of
government tend to be more efficient when it is feasible to produce
them at lower levels. 11. Explain how the government acts as an
interest group. What are the implications of government agencies
acting like special interests? 12. How does the use of a secret
ballot in elections aid in preventing the outright buying and
selling of votes? NOTES FOR CHAPTER 9 1. This discussion is based
on Gary Becker, A Theory of Competition among Pressure Groups for
Political Influence, Quarterly Journal of Economics 98 (August
1983): 371B400. 2. James M. Buchanan and Gordon Tullock, The
Calculus of Consent (Ann Arbor: University of Michigan Press,
1962), gives a classic discussion of the economic implications of
log rolling. 3. The possibility of agenda control in a referendum
is discussed by Thomas Romer and Howard Rosenthal, Political
Resource Allocation, Controlled Agendas, and the Status Quo, Public
Choice 33, no. 4 (1978): 27B43. Empirical studies concluding that
referenda tend to produce what the median voter wants include
Randall G. Holcombe, An Empirical Test of the Median Voter Model,
Economic Inquiry 18, no. 2 (April 1980): 260B74, and Rodney D.
Fort, The Median Voter, Setters, and Non-Repeated Construction Bond
Issues, Public Choice 56, no. 3 (March 1988): 213B31. 4. The
concept of rent seeking and the model of welfare losses from rent
seeking was introduced by Gordon Tullock, The Welfare Costs of
Tariffs, Monopolies, and Theft, Western Economic Journal 5 (June
1967): 224B32. The name rent seeking came from Anne 0. Krueger, The
Political Economy of the Rent-Seeking Society, American Economic
Review 64, no. 3 (June 1974): 291B303, who described the
rent-seeking activity and associated welfare losses in the economy
of India. 5. The budget-maximizing hypothesis with respect to
bureaucracy was popularized by William A. Niskanen, Bureaucracy and
Representative Government Chicago and New York: AldineAtherton,
1971).6.
This idea was developed by William A. Niskanen, Bureaucrats and
Politicians, Journal of Law & Economics 18 (December 1975):
617B43. Niskanen was extending his own model from his book,
Bureaucracy and Representative Government.
7. The classic article on this subject is Charles M. Tiebout, A
Pure Theory of Local Expenditures, Journal of Political Economy 64,
no. 5 (October 1956): 416B24.